TECHSYS 2017 Proceedings

TECHNICAL UNIVERSITY – SOFIA
PLOVDIV BRANCH
SIXTH INTERNATIONAL SCIENTIFIC CONFERENCE

“ENGINEERING, TECHNOLOGIES AND SYSTEMS”
TECHSYS 2017
18-20 May, Plovdiv
PROCEEDINGS СБОРНИК ДОКЛАДИ
ТЕХНИЧЕСКИ УНИВЕРСИТЕТ – СОФИЯ

ФИЛИАЛ ПЛОВДИВ
ШЕСТА МЕЖДУНАРОДНА НАУЧНА КОНФЕРЕНЦИЯ

“ТЕХНИКА, ТЕХНОЛОГИИ И СИСТЕМИ”
ТЕХСИС 2017
18 - 20 май, Пловдив
The conference will be held within the Конференцията ще се проведe в рамките
framework of the Science Days of Technical на Дни на науката на Технически
University of Sofia under the patronage of университет - София под патронажа
Prof DSc Georgi Mihov - Rector of the Technical на проф. дтн. Георги Михов - Ректор на
University of Sofia. Технически Унивеверситет - София.
TECHNICAL UNIVERSITY – SOFIA, PLOVDIV BRANCH
SIXTH INTERNATIONAL SCIENTIFIC CONFERENCE

“ENGINEERING, TECHNOLOGIES AND SYSTEMS”
TECHSYS 2017
18-20 May, Plovdiv
ТЕХСИС 2017
18 -20 май, Пловдив
ТЕХНИЧЕСКИ УНИВЕРСИТЕТ – СОФИЯ, ФИЛИАЛ ПЛОВДИВ
ШЕСТА МЕЖДУНАРОДНА НАУЧНА КОНФЕРЕНЦИЯ

“ТЕХНИКА, ТЕХНОЛОГИИ И СИСТЕМИ”
This conference is technically co-sponsored by: Tази конференция е технически

IEEE Bulgaria Section подпомогната от:
IEEE Young Professionals Affinity Group of Българска секция на IEEE
Bulgaria група IEEE Млади Професионалисти,
България
TECHNICAL UNIVERSITY - SOFIA, PLOVDIV BRANCH
Organizing Committee:
Honorary Chairman: Prof. DSc Georgi Mihov
Chairman: Prof. Dr. Valyo Nikolov
Vice Chairman: Prof. Dr. Michail Petrov
Members:
Prof. Dr. Grisha Spasov Assoc. Prof. Dr. Pepo Yordanov
Assoc. Prof. Dr. Krum Kutryansky, Assoc. Prof. Dr. Hrisitan Panayotov
Assoc. Prof. Dr. Nikolai Kakanakov Assoc. Prof. Dr. Toni Mihova
Assoc. Prof. Dr. Ivan Kostov Assoc. Prof. Dr. Angel Lengerov
Assoc. Prof. Dr. Anton Lechkov Prof. Dr. Dobrin Seizinski
Assoc. Prof. Dr. Borislav Penev Assoc. Prof. Dr. Valentina Proycheva
Assoc. Prof. Dr. Nikola Georgiev Assoc. Prof. Dr. Nikolaj Petrishki
Assoc. Prof. Dr. Dechko Ruschev Assoc. Prof. Dr. Valentin Vladimirov
International Programme Committee:
Chairmen: Prof. Dr. Galidia Petrova
Members:
Acad. Prof. DSc Yuri Kuznetsov, Ukraine Prof. DSc Radi Romanski, Bulgaria
Acad. Prof. DSc Nikolaj Mishkin, Belarus Prof. DSc Todor Stoilov, Bulgaria
Prof. Dr., Dr.h.c. Nikolai Ganev, Czech Republic Prof. DSc Marin Nenchev, Bulgaria
Prof. Dr. Marsel Popa, Romania Prof. DSc. Vesko Panov, Bulgaria
Prof. Dr. Ernst Wintner, Austria Prof. Dr. Ognyan Nakov, Bulgaria
Prof. DSc Mark Himbert, France Prof. Dr. Vladimir Pulkov, Bulgaraia
Prof. DSc Okyay Kaynak, Turkey Prof. Dr. Petar Gecov, Bulgaria
Prof. DSc Andre Barraco, France Prof. Dr. Georgi Todorov, Bulgaria
Prof. DSc Edmunds Teirumnieks, Latvia Prof. Dr. Lubomir Dimitrov, Bulgaria
Prof. Dr. Frantisek Zezulka, Czech Republic Prof. Dr. Mladen Velev, Bulgaria
Prof. Dr. Petr Louda, Czech Republic Prof. Dr. Boncho Bonev, Bulgaria
Prof. Dr. Nikola Kasabov, New Zealand Prof. Dr. Ilia Iliev, Bulgaria
Prof. Dr. Ahmed Hafaifa, Algeria Prof. Dr. Daniela Goceva, Bulgaria
Assoc. Prof. Dr. Abdellah Kouzou, Algeria Prof. Dr. Ivan Kralov, Bulgaria
Prof. DSc Ivan Jachev, Bulgaria Prof. Dr. Valeri Mladenov, Bulgaria
Prof. DSc Emil Nikolov, Bulgaria Prof. Dr. Andon Topalov, Bulgaria
Scientific Secretary: Dr. Sevil Ahmed

Technical Secretary: Eng. Tsvetan Petrov, Eng. Christo Christev, Eng. Lalka Boteva
ISSN Online: 2535-0048

3
ТЕХНИЧЕСКИ УНИВЕРСИТЕТ – СОФИЯ, ФИЛИАЛ ПЛОВДИВ
Организационен комитет:
Почетен председател: проф. дтн. Георги Михов
Председател: проф. д-р Въльо Николов
Зам. Председател: проф. д-р Михаил Петров
Членове:
проф. д-р Гриша Спасов доц. д-р Пепо Йорданов
доц. д-р Крум Кутрянски доц. д-р Христиaн Панайотов
доц. д-р Никола Георгиев доц. д-р Николай Каканаков
доц. д-р Иван Костов доц. д-р Ангел Ленгеров
доц. д-р Антон Лечков проф. д-р Добрин Сеизински
доц. д-р Борислав Пенев доц. д-р Валентина Пройчева
доц. д-р Тони Михова доц. д-р Николай Петришки
доц. д-р Дечко Русчев доц. д-р Валентин Владимиров
Програмен комитет:
Председатели: проф. д-р Галидия Петрова
Членове:
акад. проф. дтн. Юрий Кузнецов, Украйна проф. дтн. Тодор Стоилов, БАН, България
акад. проф. дтн. Николай Мишкин, Беларус проф. дтн. Марин Ненчев, ТУ – София, филиал
проф. д-р, д-р х.к. Николай Ганев, Чехия Пловдив
проф. д-р Марсел Попа, Румъния проф. дтн. Веско Панов, ТУ – София, България
проф. дтн. Ернст Винтнер, Австрия проф. д-р Бончо Бонев, ТУ – София, България
проф. дтн. Марк Химберт, Франция проф. д-р Илия Илиев, ТУ – София, България
проф. д-р Окияй Кайнак, Турция проф. д-р Даниела Гоцева, ТУ – София, България
проф. дтн. Андре Барако, Франция проф. д-р Владимир Пулков, ТУ – София, България
проф. дтн. Едмундс Теирумниекс, Латвия проф. д-р Огнян Наков, ТУ – София , България
проф. д-р Франтишек Зезулка, Чехия проф. д-р Петър Гецов, ИКИ, БАН, България
проф. д-р Петр Лоуда, Чехия проф. д-р Георги Тодоров, ТУ – София, България
проф. д-р Никола Касабов, Нова Зеландия проф. д-р Любомир Димитров, ТУ – София,
проф. д-р Ахмед Хафайфа, Алжир България
доц. д-р Абделлах Кузу , Алжир проф. д-р Младен Велев, ТУ – София, България
проф. дтн. Иван Ячев, ФНТС, ТУ – София, България проф. д-р Иван Кралов, ТУ – София, България
проф. дтн. Емил Николов, ТУ – София, България проф. д-р Валери Младенов, ТУ – София, България
проф. дтн. Ради Романски, ТУ – София, България проф. д-р Андон Топалов, ТУ – София, ф-л
Пловдив, България
Научен секретар: д-р инж. Севил Ахмед

Технически секретариат: инж. Цветан Петров, инж. Христо Христев, инж. Лалка Ботева
4
I-5
CONTENTS
PLENARY PAPERS • ПЛЕНАРНИ ДОКЛАДИ
1. Smart Grid and Power Electronics Integration Challenges
Abdellah Kouzou, PhD
The Applied Automation and Industrial Diagnostics Laboratory LAADI University
of DJELFA, Algeria
2. Control Theory and Its Application in Informatics and Communication Systems
Prof. DSc Todor Stoilov
Institute of Information and Communication Technologies - Bulgarian Academy of Sciences
3. SMC - INDUSTRY 4.0 - Reality and Education
Rosen Kosturkov
Sales Manager, SMC Industrial Automation Bulgaria
SECTION 1 • СЕКЦИЯ 1
AUTOMATION AND CONTROL SYSTEMS
Автоматика и системи за управление
1. ABDELHAFID BENYOUNES, AHMED HAFAIFA, ABDELLAH KOUZOU,

MOULOUD GUEMANATЕ ............................................................................................................ II-2
FUZZY MODELING AND SIMULATION OF GAS TURBINE USING FUZZY CLUSTERING
ALGORITHM
2. BACHIR NAIL, ABDELLAH KOUZOU, AHMED HAFAIFA, AHMED CHAIBET................................ II-9
MULTIVARIABLE OUTPUT-ERROR MODEL IDENTIFICATION OF ENERGY PROCESS BASED
ON EXTENDED LEAST SQUARE KRONECKER OPERATOR USING EXPERIMENTAL DATA
3. ANGEL DANEV, ATANASKA BOSAKOVA-ARDENSKA, ILIYAN DOBREV,
TODOR STAMENOV....................................................................................................................II-15
BILATERAL FILTER INTEGRATION INTO THE OPEN SOURCE SOFTWARE GELJ
4. PENKO MITEV, GEORGI SHTEREV...............................................................................................II-21
AUTOMATIC CONTROL AND CONTROL SYSTEM FOR AUTOMATIC MACHINE BUILDING
OF COUPLES AND CROPS FOR WINDOWS
5. CAGRI CANDAN, EMRE CAN DIKMEN, T. CETIN AKINCI, SERHAT SEKER................................II-24
MULTI-FUNCTION AND RECONFIGURABLE ROBUST ANALOG I/O MODULE DESIGN
METHODOLOGY FOR INDUSTRIAL CONTROLLER SYSTEMS
6. BORISLAV RUSENOV, ALBENA TANEVA, IVAN GANCHEV, MICHAIL PETROV........................II-30
MACHINE VISION SYSTEMS FOR INTELLIGENT QUALITY CONTROL OF MANUFACTURING
PROCESSES
7. STOYCHO PENKOV, ALBENA TANEVA, VASIL KALKOV.............................................................II-36
INDUSTRIAL NETWORK DESIGN USING LOW ENERGY PROTOCOLS
I-6
8. DIMITAR SPIROV, NIKOLAY KOMITOV.......................................................................................II-42

DETERMINATION OF THE PARAMETERS OF QUASI RESONANT DC LINK CONVERTER
WITH TWO PARALLEL SWITCHES
9. DIMITAR SPIROV, NIKOLAY KOMITOV.......................................................................................II-46
DESIGN OF INDUCTION MOTOR DRIVE WITH QUASI-RESONANT CONVERTER
10. NIKOLINKA YАNKOVA, EBRU ADEM, IBRYAM AHMEDOV.......................................................II-50
MODERN DEVICES FOR ACCESS CONTROL
Electrical Engineering and Energetics Electronics;
Electronic Technologies and Optoelectronics
Електротехника и електроенергетика; Електроника и оптоелектроника
1. EMEL ONAL ..................................................................................................................................II-55

SHORT TERM ELECTRICITY CONSUMPTION FORECASTING: A CASE STUDY OF ELECTRIC
UTILITY AT TURKEY
2. EMEL ONAL...................................................................................................................................II-59
ANALYSIS OF TIME TO BREAKDOWN VOLTAGE MEASUREMENT
3. ALEXANDER SARAFOV ...............................................................................................................II-63
HYBRID NETWORKS USING DENSE WAVELENGTH DIVISION MULTIPLEXING (DWDM)
TECHNOLOGY
4. ANGEL SLAVOV, PETER Z. PETKOV, STOYAN ILIEV...................................................................II-66
DIPOLE ANTENNA OVER EBG STRUCTURE FOR UHF RFID APPLICATIONS
5. SARANG M. PATIL, PETER Z. PETKOV, BONCHO G. BONEV.....................................................II-71
A REVIEW ON RECENT ANTENNA DESIGNING TECHNIQUES FOR ELECTROMAGNETIC
COMPATIBILITY (EMC) TEST
6. LILYANA KOLEVA, ELENA KOLEVA, VENTZISLAV TZOTCHEV ..................................................II-77
OPTIMIZATION OF ELECTRON BEAM LITHOGRAPHY PROCESS THROUGH ROBUST
ENGINEERING APPROACH
7. IVAN HADZHIEV, DIAN MALAMOV, VASIL SPASOV..................................................................II-83
DEFINING THE COEFFICIENT OF ADDITIONAL LOSSES BETWEEN BUSBARS
EXPERIMENTALLY AND NUMERICALLY
8. NIKOLA GEORGIEV......................................................................................................................II-89
STUDY OF THE INFLUENCE OF THE PIEZOELECTRIC CERAMICS PARAMETERS ON
THE OUTPUT CURRENT OF A PIEZOELECTRIC TRANSFORMER, POWERING A SUPER-
CAPACITOR
9. NIKOLA GEORGIEV.......................................................................................................................II-95
STUDY OF TWO TYPES OF SENSORS OF STATIC FORCES - A PIEZOELECTRIC SENSOR
AND A PIEZOELECTRIC ELASTOMER SENSOR
I-7
10. ANGEL LICHEV, ALEKSANDAR VUCHEV, NIKOLAY BANKOV, YASEN MADANKOV ...............II-99
LOAD AND CONTROL CHARACTERISTICS OF A SERIES RESONANT DC-DC CONVERTER
WITH A SYMMETRICAL CONTROLLED RECTIFIER
11. ANGEL LICHEV, ALEKSANDAR VUCHEV, NIKOLAY BANKOV, YASEN MADANKOV.............II-105
LOAD AND CONTROL CHARACTERISTICS OF A PHASE-SHIFT CONTROLLED
BIDIRECTIONAL SERIES RESONANT DC/DC CONVERTER
12. IVAN KOSTOV, BOJIL MIHAYLOV, VASIL SPASOV....................................................................II-111
EXPERIMENTAL ANALYSIS OF THE SUPPLY VOLTAGE QUALITY OF INDUCTION MOTORS
WITH PWM CONVERTERS
13. YANKA IVANOVA, SVETOSLAV IVANOV, ROSSEN BOJILOV...................................................II-115
INVESTIGATION OF ONE LOOP ELECTRIC CIRCUIT FOR CONTROL OF FLASH LAMP
14. SVETOSLAV IVANOV, YANKA IVANOVA, ROSSEN BOJILOV...................................................II-119
ULTRASONIC SENSOR FOR MEASURING WATER SPEED
15. VANIA RANGELOVA, VENCESLAV PEEV, NIKOLAI PAUNKOV................................................II-125
VIRTUAL MEASUREMENT TOOL AND DATA PROCESSING OF DIRECT REPEATED
MEASUREMENTS
16. MARGARITA DENEVA, MARIN NENCHEV................................................................................II-129
HIGH SENSITIVE LONG DISTANCE SCANNING FIBER-OPTICS LASER SENSOR SYSTEM
17. VALKO KAZAKOV ......................................................................................................................II-134
THERMO-SENSITIVE PAPER APPLICATION AS METHOD FOR LASER BEAM SPOT STUDY –
CASE OF Q-SWITCHED LASER PULSES
18. METIN SALTIK.............................................................................................................................II-139
THE OBSERVATION OF ATMOSPHERIC ELECTRICITY AND LIGHTNING EFFECTS
INFORMATICS AND COMPUTER SYSTEMS AND TECHNOLOGIES
Информатика и компютърни системи и технологии
1. TARIK YERLİKAYA, EMRECAN ADLI ........................................................................................II-144

RSA ENCRYPTION USING MORE THAN TWO PRIME NUMBERS
2. SENOL SEN, TARIK YERLIKAYA.................................................................................................II-147
PASSWORD CRACKING ATTACKS
3. CEM TASKIN, ALTUG YIGIT, DENIZ TASKIN, AYSE CELIK TASKIN ...........................................II-151
IMAGE PROCESSING ON ANDROID DEVICES WITH OPENCV
4. OĞUZ KIRAT, TARIK YERLİKAYA ...............................................................................................II-156
COMMON ATTACKS TYPES AND PENETRATION TESTING ON WI-FI NETWORKS
5. ATANAS KOSTADINOV...............................................................................................................II-159
ALTERA (PART OF INTEL) AND XILINX TOOLS IN EDUCATION
6. DIMITAR GARNEVSKI ..............................................................................................................II-163

IMPLEMENTATION OF SOFTWARE ARCHITECTURE FOR COMPLEX ANALYSIS OF SOLAR
CORONA IMAGES
I-8
7. DIMITRE KORMICHEV................................................................................................................II-167
ADVANCED ALGORITHM FOR CANNY EDGE DETECTION GRADIENT MAGNITUDE
COMPUTATION TO BE IMPLEMENTED ON FPGA
8. DIMITRE KORMICHEV................................................................................................................II-171
APPROXIMATION IN THE FPGA BASED CANNY EDGE DETECTION COMPUTATIONS
9. KOLYO RAYCHINOV, HRISTO VALCHANOV..............................................................................II-177
INTRUSION DETECTION AND PREVENTIG SYSTEM
10. VENETA ALEKSIEVA, AYDAN HAKA..........................................................................................II-181
SIMULATION FRAMEWORK FOR REALIZATION OF PRIORITY-BASED LTE SCHEDULER
11. IBRYAM AHMEDOV, EBRU ADEM, NIKOLINKA YАNKOVA ....................................................II-186
DIAGNOSTICS USING DELPHI DS150E
12. SVETOSLAV ENKOV, TONY KARAVASILEV...............................................................................II-190
BENCHMARKING HASH FUNCTIONS
13. MILENA ANGELOVA...................................................................................................................II-196
CLUSTERING TECHNIQUES FOR ANALYSIS OF LARGE DATASETS
14. T. HRISTEVA, M. MARINOVA.....................................................................................................II-202
ALGORITHMS IN DEEP LEARNING
15. JORDAN KYRPAROV, DILYANA BUDAKOVA, LYUDMIL DAKOVSKI ....................................II-206
A BAYESIAN NETWORK MODEL FOR FOOTBALL GAMES OUTCOMES PREDICTION
16. DIMITAR GROZEV, GRISHA SPASOV, NIKOLAY KAKANAKOV, GALIDIYA PETROVA........................ II-210
DEVELOPMENT OF CLOUD COMPUTING BASED SCADA IN ELECTRICAL POWER SYSTEMS
Mechanical, Transport and Aviation Engineering
Mашиностроителна техника и технологии; Машиностроене и уредостроене;
Транспортна и авиационна техника
1. CENK MISIRLI, ISIK CETINTAV, YILMAZ CAN, CEM MISIRLI....................................................II-216

NUMERICAL ANALYSIS OF BI-METALLIC PARTS USING A FINITE ELEMENT SOFTWARE
2. ZHIVKO ILIEV, GEORGI DINEV...................................................................................................II-221
COMPUTER AIDED RESEARCH OF THE SHAFT OF A ROTOR CRUSHER OF AN ECCENTRIC TYPE
3. VASILIY CHITANOV, LILYANA KOLAKLIEVA, ROUMEN KAKANAKOV
TETIANA CHOLAKOVA.............................................................................................................II-227
COMPARATIVE STUDY OF THE MECHANICAL PROPERTIES OF Ti AND Cr BASED HARD
COATINGS DEPOSITED BY CLOSE FIELD UNBALANCED MAGNETRON SPUTTERING
4. STEPHAN KOZHUKHAROV, CHRISTIAN GIRGINOV................................................................II-232
TOPOLOGICAL FEATURES OF PHOSPHATE TREATED SPECIMENS OF PRELIMINARY
CERIUM OXIDE COATED AIRCRAFT ALLOY
5. SVETLIN STOYANOV..................................................................................................................II-236
SENSORS MASS INFLUENCE ON THE NATURAL FREQUENCY OF A CANTILEVER BEAM
I-9
6. ILIA POPOV, SABI SABEV .........................................................................................................II-240

THE APPLICATION OF THE MODAL ANALYSIS IN DETERMINING THE DYNAMIC MODULE
OF ELASTICITY OF POLYMERBETON COMPOSITES
7. IVO DRAGANOV, YULIAN ANGELOV........................................................................................II-240
NUMERICAL MODELING AND PARAMETRICAL ANALYSIS OF CALIBRATED DOUBLE
LAYERED ROLL
8. ZDRAVKO VITLAROV.................................................................................................................II-248
DYNAMIC SYNTHESIS OF THE AGGREGATE OF ELECTROMECHANICAL PRESS FOR
EXPANDED METAL AT A SET LEVEL OF INEQUALITY
9. NIKOLAY GUEORGUIEV, ALEKSANDAR KOLAROV, VENCISLAV PEHIVANSKI......................II-252
AN APPROACH FOR SELECTION CRITERIA FOR EVALUATION OF REMEDIES OF PROJECTS
10. STEFAN VODENICHAROV, ALEKSANDAR KOLAROV, NIKOLAY
GUEORGUIEV, VENCISLAV PEHLIVANSKI .............................................................................II-258
NEW APPROACHES TO INCREASE ANTI-MINING RESISTANCE TO SEA BOTTOM MINES
11. HRISTIAN PANAYOTOV, STANIMIR PENCHEV.........................................................................II-264
DETERMINATION OF PROPELLER CHARACTERISTICS FOR MULTIROTOR DRONE DESIGN
12. KRASIMIR AMBAREV, VALYO NIKOLOV...................................................................................II-269
SYSTEM FOR MEASURING THE PRESSURE AT WORK AND CONSTRUCTION OF
INDICATOR DIAGRAM OF DIESEL ENGINE
13. RADOSLAV KOSTOV...................................................................................................................II-275
INDICATORS EVALUATING GAS EXCHANGE PROCESS IN COMBUSTION ENGINES
14. RADOSLAV KOSTOV...................................................................................................................II-279
INFLUENCE OF CONSTRUCTION AND OPERATING FACTORS GAS EXCHANGE ON THE
COMBUSTION TURBOCHARGER
15. DANCHO KOLIBAROV................................................................................................................II-284
CIVIL-MILITARY PROCEDURES IN AIR AT THE ACTION OF THE MILITARY AIRCRAFT TO
ASSURING AIR SURVEY. PART 1
16. DANCHO KOLIBAROV................................................................................................................II-290
CIVIL-MILITARY PROCEDURES IN AIR AT THE ACTION OF THE MILITARY AIRCRAFT TO
ASSURING AIR SURVEY. PART 2
17. ILKO TARPOV..............................................................................................................................II-296
HIGHLIGHTS IN THE TRAINING FOR OPTIMAL CONTROL OF ELECTRICAL RAILWAY
SMOOTH VARIABLE TRACTION
18. ILKO TARPOV, SILVIYA SALAPATEVA .......................................................................................II-300
INCREASING THE EFFICIENCY OF THE RECUPERATIVE BRAKING BY ALTERING THE
TRAIN SCHEDULES
19. IVAJLO GEORGIEV, DOBRIN SEJZINSKI ...................................................................................II-304
METHODOLOGICAL ASPECTS OF SAFE AND EFFECTIVE FIRE FIGHTING BY MEANS
OF A FIRE FIGHTING SYSTEM ATTACHED TO A HELICOPTER
20. DOBRIN SEYZINSKI, STOYAN AVRAMOV.................................................................................II-310
MILITARY AVIATION EQUIPMENT LIFESPAN PROLONGATION IN ECONOMICAL
CRISIS SITUATION
I-10
INDUSTRIAL MANAGEMENT • Индустриален мениджмънт
1. TONI MIHOVA, VALENTINA NIKOLOVA - ALEXIEVA .............................................................II-316

PROBLEMS IN RECRUITING HUMAN RESOURCES FROM BULGARIAN INDUSTRIAL
ENTERPRISES
2. MINA ANGELOVA, DANIELA PASTARMADZHIEVA ...............................................................II-321
CHALLENGES AND OPPORTUNITIES FOR FLEXIBLE CREDITING OF SMALL AND
MEDIUM-SIZED ENTERPRISES IN BULGARIA
3. ANTONIA LAZAROVA ................................................................................................................II-325
APPLICATIONS IMITATION MODELING IN INVESTIGATION OF A LENDER LEVERAGE IN
BULGARIA
4. OTMAN AHMED, SVETOSLAV DIMKOW..................................................................................II-331
A MODEL TO SOLVE THE SUPPLIER SELECTION PROBLEM
5. MAYA DILKOVA-PAVLOVA ........................................................................................................II-337
MARKETING MIX AND ITS COMPATIBILITY WITH RELATIONSHIP MARKETING IN
BANKING SERVICES
6. TANYA GIGOVA...........................................................................................................................II-341
STATE OF THE LABOR MARKET IN BULGARIA
7. ELENA ZLATANOVA-PAZHEVA..................................................................................................II-347
SURVEY OF FOOD BEHAVIOUR OF CHILDREN AND TEENAGERS IN DEPENDANCE OF THE
TYPE OF POPULATED PLACE THEY STUDY
8. KRISTIAN CVETKOV, VLADISLAV GERASIMOV, ELENA KOLEVA, LILYANA KOLEVA............II-353
QUALITY AND SATISFACTION FOR PORTABLE COMPUTERS
9. STOYANKA KASABADZAKOVA, EKATERINA PERPERIEVA.....................................................II-359
PRINCIPLES OF AUTOMATIC LINES OPTIMAL DESIGN
10. MARIYA PETKOVA-KOZOVSKA ................................................................................................II-363
PROBLEMS AND SOLUTIONS OF WASTE MANAGEMENT IN GLOBAL AND NATIONAL
ASPECT
11. SVETLANA VASILEVA.................................................................................................................II-369
ONE APPROACH FOR MODELLING HOTEL REVENUE MANAGEMENT SYSTEMS
12. RADOSLAV RADEV.....................................................................................................................II-375
USE OF PECTIN COATINGS FOR FRESH FRUIT AND VEGETABLES
13. RADOSLAV RADEV.....................................................................................................................II-379
APPLICATION OF LIPID COATINGS IN FOOD TECHNOLOGY
NATURAL SCIENCES • Природни науки
1. ALBENA PAVLOVA .....................................................................................................................II-386

QUADRATURE ITERATIVE METHOD FOR A NUMERICAL SOLUTION OF A NONLINEAR
HAMMERSTEIN FUZZY FUNCTIONAL INTEGRAL EQUATION
I-11
2. DOBRINKA GRIBACHEVA, DIMITAR RAZPOPOV ...................................................................II-392

RIEMANNIAN ALMOST PRODUCT MANIFOLDS GENERATED BY A CIRCULANT
STRUCTURE
3. GEORGI PASKALEV.....................................................................................................................II-396
VARIATIONAL METHOD FOR A CLASS OF HIGHER ORDER HYPERBOLIC EQUATIONS”
4. NELI KERANOVA.........................................................................................................................II-399
MATHEMATICAL METHODS FOR ESTIMATING THE AVERAGE YIELD OF SPINACH IN
MEMBER STATES OF THE EUROPEAN UNION FOR THE PERIOD 1961-2014
5. NELI KERANOVA.........................................................................................................................II-403
DATA MODELING IN THE AGRICULTURAL SECTOR IN BULGARIA, RELATED TO THE
BREEDING OF SOME FARM ANIMALS FOR THE PERIOD 2000-2014
6. LILIYA PETROVA.........................................................................................................................II-407
RECTANGULAR PLATES – SINGLE TRIGONOMETRIC SERIES, EXPONENTIAL FUNCTION
I-12
© International Scientific Conference on Engineering, Technologies and Systems

TECHSYS 2017, Technical University – Sofia, Plovdiv branch
18 – 20 May 2017, Plovdiv, Bulgaria
PLENARY PAPERS • ПЛЕНАРНИ ДОКЛАДИ

I-13
Abdellah Kouzou
PhD, HDR, Assoc. Prof
Applied Automation and Industrial Diagnostic Laboratory,
University of Djelfa 17000 DZ, Algeria.
URL: http://www.univ-djelfa.dz/labo/laadi
Mr Kouzou Abdellah (IEEE Senior memebr & IACSIT Senior member, IFAC,IAENG & IISRO member,
IEEE-HKN Alumni Member) was born in Djelfa, Algeria in 1964. He is a collaborator researcher at
Texas A&M University at Qatar. He has been the president of the Scientific council of the faculty
since June 2014. He has participated in several research projects and has led several research
projects. He is the founder of the Power Electronics and Power Quality research group at the Applied
Automation and Industrial Diagnostic Laboratory, University of Djelfa in Algeria. He is the supervisor
of many PhD Students in Algeria. He is a member of the Smart Grid Center at Qatar SGC-Q . He is a
member of many editorial boards for several scientific journals and a member of the scientific and
steering committees in several national and international conferences. He is the Editor-In-Chief of
two journals. He is the coordinator of the Algerian IEEE Power Electronics Chapter and the chair of
the sub-committee on FACTs and HVDC under the international committee PETC/IEEE-IES. He was a
plenary and an invited keynote speaker and session chair in several national and international
conferences and an experts in several national and international scientific activities and project
evaluations.
He has published more than 160 papers, his main research interests include Active Power Filtering
techniques, Power Quality issues, Power Electronics Devices, Application of Power electronics in
Renewable Energies and Application of meta-heuristics optimization algorithms, Smart Grid and
Smart Buildings, reliability and diagnostics in power electronics converters.
© International Scientific Conference
I-14 on Engineering, Technologies and Systems
Smart Grid and Power Electronics Integration

Challenges
Abdellah Kouzou
IEEE Senior Member
Head of Power Electronics and Power Quality Team, Applied Automation and Industrial Diagnostic Laboratory,
Djelfa University, Algeria
Member of the Smart Grid Center at Qatar SGC-Q, Texas A &M
kouzouabdellah@ieee.org
Abstract
Nowadays the idea of smart grid is becoming of great interest for the researchers, designers, industrial partners,
the electrical energy producer, the electrical energy transporters and distributors, the consumer and the
governments. Where the main objective is to provide within the existing structures of urban, electrical network
and power sources, new improved quality of electrical power that can ensure several performance indicators
such as the reliability, the comfort of use, the security, the safety, the privacy, the accuracy of the used
technologies, the self controlling, the friendly environment interaction, the energy efficiency, the sharing of
energy information within the city and the network power system for the people inside and a precise interaction
among the energy producer, transporter and consumer of all levels.
Indeed, it is actually well known that the existing grid structures that are considered to be smart grid are
estimated to be a complex concatenations of structures, systems and technology. Whereas; the different
components used inside the considered grid have been designed, developed and improved to fulfill the
requirement of the modern-day grid control. However the level reached of the grid smartness with the ensured
aforementioned systems, the smart grid owners today are still having their aims passing beyond the contour of
their own region and grid, indeed they want that their grid will share and interact smartly and positively with
the surrounding systems, such as; the electrical power system network, the renewable energy sources, the other
distributed energy systems, the communication network, the group of the buildings in the neighbor region, the
city or even more, the global environment consideration etc. It can be said that to ensure the approach of
interoperability between the inside systems and the outside systems many different parties are required to
collaborate dynamically, accurately and seamlessly under some primordial aspects that would be achieved by
smart grids, such as; the energy efficiency, the energy minimizing cost, and the energy management. On the
other side, it is well obvious that the smart grids are designed initially to support many tasks based on new and
advanced technologies of several domains, starting from the construction materials that can be called smart
materials such as : Self-Shape-Shifting metals and Self-Healing-Materials, and other domains, especially the
smart communication systems, smart meters and smart sensors.
In this context the present plenary will focus mainly on the presentation of an overview on smart grids from the
traditional grid to the modernized grid, the integration of renewable energy sources and the integration of the
power electronics systems, and finally the different challenges and future perspectives. The overall idea of the
integration of different smart systems in smart grid will be presented in this plenary from point of view of the
general research perspectives for the next generation of the smart grids based on some future perspectives,
especially based on the used technologies such as smart metering, smart energy monitoring, smart power
electronics devices for energy conversion and interfaces. where the main aim to give an idea to the practical
researchers to go ahead for different parts of research that are giving a promising ways to obtain significant
results, experimental achievement, industrial implementation and more theoretical basis.
Keywords: Smart Grid, Renewable energy, Smart meters, Smart Materials, Smart Sensors, Power electronics
devices.
I-15 on Engineering, Technologies and Systems
© International Scientific Conference
CONTROL THEORY AND IT APPLICATIONS

IN INFORMATION AND COMMUNICATION
SYSTEMS
Todor Stoilov
Institute of information and communication technologies
1113 Sofia, Acad.G.Bonchev str, Bl.2,
e-mail: todor@hsi.iccs.bas.bg
Bulgaria
18-19.May 2017 TechSys, Plovdiv
Content
Control Theory
Optimization calculations
in information services
Automatic control in information

systems/program systems
Automatic control
in communication systems
18-19.May 2017 TechSys, Plovdiv

I-16
SMC Industrial Automation Bulgaria

„INDUSTRY 4.0 - reality and education"
SMC International Training is the educational division of SMC Corporation, the world leader in innovation
and the sales of pneumatic and electro-pneumatic components for industrial automation.
SMC International Training has a clear commitment to offer Training adapted to the needs of industry;
training orientated to the development of the professional expertise required for the most diverse sectors
(automotive, pharmaceutical, semi-conductors, foodstuffs, etc.)
With a global market share of 33% we are worldwide leading experts in pneumatics.
We are a strong partner to the industry and develop state-of-the-art automation solutions
- customized solutions
- innovative products
- tailor-made services
Being the world leader in pneumatics, we take pride in offering an extensive range of high quality pneumatic
products; however, automation requires non-pneumatic solutions as well.
This has resulted in the development of products such as electric actuators, thermo-chillers, ionizers, ultra
high purity gas and fluid control equipment, allowing SMC to offer a total solutions package
Our overall objective is to offer the highest level of support to our customers and become their sole
automation supplier
SMC has a clear strategy when it comes to Industry 4.0.

We are both a leading provider and user of Industry 4.0 technologies.
We see a major need for optimization in both productivity and work schedules.
INDUSTRY 4.0 Examples
1. MALFUNCTIONS
When it occurs, Malfunctions should be quickly forwarded. The system reports automatically to these
responsible. The components affected are displayed in real time.
The software assists diagnoses and provides more detailed information including details of
malfunction and relevant CAD drawings, so it becomes easy to find exact faulty element and to be
replaced as quick as possible
2. PREDICTIVE MAINTENANCE
One of the components of Industry 4.0 is predictive maintenance with which unscheduled production
stoppages can be avoided or even entirely excluded. The predictive maintenance system is capable
for seeing expensive repairs or serious component failure and initiating preventive activities before
greater damage occurs
3. RAPID CHANGEOVER
In flexible production facilities product changeover is a rapid process. Flexible product changeover of
a machine affected without setup delays by means of higher level control system such a MES or
ERP.
This means that the machine can produce various products in batches of any size.
In the domain of Industry 4.0, SMC consistently expends its portfolio in the direction of distributed
intelligence and Rapid networking. The goal here is to connect the system horizontally or vertically.
We’re thereby enable users to optimize their production processes constantly
I-17
The content-related drivers for the advance of

Industry 4.0 solutions require interlinking of
different disciplines that, even though known,
have so far been viewed in isolation.
In order to be able to fulfill increasing
requirements for production systems and
machines, numerous generally known
catchwords are known:
AUGMENTED REALITY
OPTIMIZATION,
DIGITAL COMUNICATIONS,
BIG DATA HANDLING,
SMART SERVISE,
INDIVIDUALIZATION,
COST REDUCTION,
REAL TIME REPORTING,
MALFUNCTIONS, PREDICTIVE
MAINTENANCE, RAPID CHANGEOVER….
INDUSTRY 4.0 is a term generated by German

Federal Government in 2011 as a High-Tech
Strategy for increasing Manufacturing of German
companies with focus of highly customized
products.
Current usage of the term as the 4-th Industrial
Revolution has been criticized as meaningless
.
What is happened in the industry in a past 50 year
is constant EVOLUTION and INOVATION, but still
is not REVOLUTION.
Nevertheless, the term become popular worldwide
and have been used as key target for many
production companies.
Nevertheless- Industry 4.0, Machine-to-Machine

(M2M), SMART FACTORY, Internet of Things
(IoT), Internet of Service, Cyber-physical systems, Supply-Chain-Real-Time or whatever you like to call
it, It’s not just about generating business value, it’s not about saving or earning money in the first place! It’s
much more about customization, innovation and finding (or creating) new needs!
It’s about CREATIVITY. And let not forget – Creativity still is exceptional Human property.
And there is a BIG Question – How the education system should be adapted to modern requirements of the
industry?
Can we create Creative People in high schools and universities?
The direct answer of this question is “NO” as the people a creative as they are by origin. But what the
education can do with no doubt is to create an appropriate environment to develop student’s creativity.
Of course, first basic knowledge of technologies and particular skills should be created and developed
I-18
Training solutions, adapted to the modern industry

demands
Based on our leading position, experience and knowledge, SMC developed series of designed for Technical
High Schools and Technical Universities Training solutions, adapted to the modern industry demands
eLEARNING-200 - The perfect theory companion

eLEARNING
eLEARNING 200 is a flexible learning system where knowledge can
eLEARNING-200
be acquired dependant on the user’s available time and requirements.
Students have the option to take actual classes and additional Internet
courses when this suits them best.
To be able to develop different technology skills, theoretical knowledge

must be acquired first. SMC International Training presents the
eLEARNING
eLEARNING-200 course programs the perfect complement to our
didactic equipment.
eLEARNING-200
200 offers a total of 13 courses based on the automation pyramid, covering technologies in
automated systems.
• SMC-100 – Introduction to industrial automation

• SMC-101 – Principles of pneumatics
• SMC-102 – Introduction to electricity
• SMC-103 – DC electricity
• SMC-104 – AC electricity
• SMC-105 – Solid state
• SMC-106 – Introduction to wiring
• SMC-107 – Introduction to electric motors
• SMC-108 – Sensors technology
• SMC-109 – Programmable controllers
• SMC-110 – Process controls
hydraulics
• SMC-111 – Hydraulics / electrohydraulics
• SMC-113 – Robotics
A. Training systems in different technologies:
PNEUMATICS AND ELECTROP

ELECTROPNEUMATICS and VACUUM
PNEUMATE-200, PNEUTRAINER 200, VAC 200

PNEUMATE
Small sized, light and portable training equipment that includes extensive
Small-sized,
features which provide an enjoyable and rapid way of acquiring the
fundamental concepts of pneumatics and electro-pneumatics.
electro pneumatics.
HYDRA
HYDRAULICS
HYDROTRAINER
HYDROTRAINER, HYDROMODEL
Fully modular systems designed for the development of professional
skills
I-19
ELECTRICAL DRIVES, SENSORS, LASER, OPTICS, ARTIFICAL VISION

ELECTRICAL PANELS, IDENTIFICATION SYSTEMS, PROGRAMMING COBTROLLERS
B. Training systems with INTERACTIONS OF TECHNOLOGYES:
AUTOMATION, MANIPULATION, SCADA

FMS 200
ROBOTICS, HMI, MOTION CONTROL

FAS 200
ROBOTICS, HMI, MOTION CONTROL, ERP, MES

HAS 200
C. INDUSTRY 4.0 in education = CREATIVITY ENVIRONMENT
We know what is Creativity – Knowledge-of-Ideas, Freedom of Thinking, Open Space in Mind,

Connectivity, Smart and variable tools, Suitable environment…
Everything is providing by
The Connected Enterprise Reality makes all this possible. It converges plant-level and enterprise
networks, and securely connects people, processes, and technologies
SiF400 has been designed to emulate Industry 4.0’s automated smart factory
that includes technologies, advanced manufacturing concepts and connected
enterprise reality. The training system is optimized to unpack Industry 4.0,
enabling organizations to – with minimal disruption – amend their factory
operations.
Industry 4.0’s smart factory includes cyber-physical systems that control the
physical processes of the factory and make decentralized decisions.
Essentially, these physical systems become Internet of Things (IoT),
communicating and cooperating both with each other and with humans in real time via the wireless web
The modular features of all training equipments enable the introduction of variations in its stations so that
they adapt to the different requirements of companies and training centers.
I-20
From a simple configuration of one station only (working fully autonomously) to a complex configuration with
eight, ten or more stations, the possibilities are endless.
In addition, it facilitates a staggered investment, i.e. starting with an initial simple configuration which can be
easily enhanced by adding workstations.
All the components in the systems are used in industry, so that the students can work with real elements at
all times making the learning process more meaningful.
The system includes a whole series of feeding, handling, verification and loading operations etc. carried out
using components from different technologies (pneumatics, hydraulics, sensors, robotics,
communications, control and HMI).They includes the breakdown simulation system which generates up to
16 different breakdowns to be diagnosed by the students.
The combination of all these possibilities means that a lot of different assemblies can be obtained enabling
the use of production management strategies.
The control panels are completely modular and can be rapidly disassembled so the students can design and
integrate a new control.
Aspects such as aesthetics, user motivation and the development of transversal skills (such as teamwork
etc.) have also been taken into account in the conception and design process.
At university level, these systems represent a powerful development platform for research projects.
Every Training System includes complete documentation, oriented to the development of the professional
skills. The documentation includes:
USER’s MANUAL, THEORY MANUAL, EXERCISES MANUAL (STUDENT and PROFESSOR)
The acquired skills in different technologies when students use the SMC training equipments are
represented in SKILLS/TECHNOLOGIES table.
18.05.2017, SMC Industrial Automation Bulgaria EOOD

Sofia, Bulgaria Business Park Sofia, Building 8C - 6th floor
1766 Sofia
www.smc.bg
www.smctraining.com
II-1

AUTOMATION AND CONTROL SYSTEMS
Автоматика и системи за управление


FUZZY MODELING AND SIMULATION OF GAS

TURBINE USING FUZZY CLUSTERING
ALGORITHM
ABDELHAFID BENYOUNES (1), AHMED HAFAIFA (1*), ABDELLAH KOUZOU (1) ,
MOULOUD GUEMANA (2)
Abstract: Gas turbines are one of the major parts of modern industry. They have played very
important role in aeronautical industry, power generation and main mechanical drivers for large
pumps and compressors. This study addressed the modeling and the simulation of the Industrial
Gas Turbine GE ms5001P, located in the electrcal production station of M'sila in Algeria. The
used method for modeling of this gas turbine is based on fuzzy inference system with the use of
Gustafson-Kessel clustering (GK) algorithm.
Key words: Gas turbine modeling; fuzzy modeling; fuzzyinference system; GK clustering
algorithm
1. INTRODUCTION fuzzy model). Since then, the fuzzy model of Takagi

and Sugeno TS has proved its effectiveness in the
Currently, the development of mathematical
study of many other nonlinear dynamic systems.
models for the representation and approximation of
Conversely, conventional approaches use a single
complex nonlinear systems is an essential subject in
model to describe the overall behavior of a
several engineering disciplines. The need for a
nonlinear system, whereas fuzzy Takagi and Sugeno
strong understanding of physical phenomena in
TS models essentially use a multi-model approach,
industrial systems is a great restriction at the
in which the simple sub-models are linear and are
practical level when dealing with complex non-
combined between them, for the purpose of
linear systems. Indeed, the equation of the laws
describing the overall non-linear system behavior.
governing such systems generally leads to a too
complex model of knowledge and its This strong property of fuzzy models of type
implementation is delicate. In this case, the use of Takagi and Sugeno can be applied in several
modeling techniques developed from the input / dynamic systems modeling applications, that can be
output measures collected on the system is required. described by differential equations. In this paper,
fuzzy identification and modeling from
For this reason, this work relies on fuzzy logic
experimental data will be presented to approximate
and its tools to present a complete and integrated
nonlinear systems. In this framework, this work is
approach to solve all the problems encountered in
intended for the design of an original method for the
the use of classical modeling and control methods.
identification and modeling of industrial systems,
The problem of exploiting fuzzy models based on
with a view to its application to a gas turbine
Takagi-Sugeno approach, using input / output data
system. This model was trained by use of real
collected on the system being tested, for a gas
operational data of a GE MS5001P gas turbine used
turbine system application.
for electrical energy production.
Indeed, Takagi-Sugeno fuzzy modeling is a
universal approximation of real systems, which has
shown these efficiencies in several applications in 2. PROCESS DESCRIPTION
the literature [17, 24, 28, 32, 38, 81, 86]. In 1985
The GE MS 5001P gas turbine engine is made of
Takagi and Sugeno proposed the use of properties of
three main sections [1]: the axial compressor, the
fuzzy sets and the use of rules in a fuzzy model (TS
Copyright © 2017 by Technical University of Sofia, Plovdiv branch, Bulgaria ISSN Online: 2535-0048
II-3
combustion chamber and the turbine. There is a
variable inlet guide vane (IGV) in the inlet of the
axial compressor and a variable nozzle guide vane Table 1. Examined gas turbine specifications
(NGV) in the turbine section [3]. In other part, two parameters
valves are used as the main controlling devices of Quantity Value
the gas turbine located on the inlet fuel line of the
combustion chamber, the first one is for the control Compressor stage 16
of the inlet fuel pressure which called the stop ratio Firing temperature 1.730 (F°)
valve (SRV) and the second, the most important
one, is used for governing the speed of the load Exhaust
898 (F°)
shaft this one called the gas control valve (GCV). temperature
The figure (figure.1) represents the schematic of the Air flow 928.5 ( Lb/hr)
gas turbine system.
Output 24,700 (kw)
12,950 (kJ/kW-
Heat rate
h)
A. Nonlinear System Modeling

The fuzzy modeling from experimental data are
effective tools to approximate a non-linear system.
Among the models are widely used in the modeling
techniques we find those of Takagi-Sugeno (TS)
[12, 15].
Fig. 1. Signal shaft gas turbine GE MS 5001P
A TS model uses the idea of linearization of
The gas turbine under investigation in this work the fuzzy regions in the state space. Based on these
is the GE MS 5001P industrial gas turbine regions fuzzy, a non-linear system can be
specifically, this turbine is designed for mechanical decomposed into a multi structure models consisting
drive applications with a wide operating speed range of several linear models which are not necessarily
to meet operating conditions of the most common independent [2]. The fuzzy sets premises partition
driven equipment In our case it is used as an turbo the input space in a number of fuzzy regions, while
generator in the M’sila electrical production station the functions consequences describe the behavior of
located in M’sila –Algeria (figure 2). The design the system in these regions. The fuzzy model TS is
point specifications of this gas turbine are presented generally built in two steps:
in Table I. The detailed information can be found in
Step 1: Determine the membership functions
the product manuals.
(MFs) antecedent of the rules;
Step 2: Estimate the parameters of the functions
consequences.
One of the techniques used to achieve the first
step is the fuzzy clustering, in this paper apply the
Gustafson-Kessel (GK) algorithm.
B. Gustafson-Kessel algorithm
The GK algorithm constitutes a reference among
the different among different methods of fuzzy
classification based on minimizing the objective
function of the form:
i 1 
c N
J FCM (Z ;U ;V )  (ik )m DikA
2
 K 1 
Where Z is the data set, U   ik  is the matrix

Fig. 2. M’sila electrical production station
of fuzzy partition (size C  N ) and V   1 , 2 ,..., c 
II-4
is the vector of the center of classes to be 1
determined with  i   n the center of the i th class  ik  2
1  i  c,1  k  N
1  i  c , m  1, is a factor that denotes the c D  m 1
degree of fuzziness of the partition.
 j 1 D ikA 
 

jkA 
N
The standard quadratic distance in space in
vi 
K 1 ( ik ) z k m
question, which defines the distance measure N

between the Z k observation and the  i center K 1 ( ik ) m
within the meaning of the metric induced by A. This leads to the GK algorithm given in three
Gustafson and Kessel in employing a standard steps:
of adaptive distance in the purpose of detecting of Step 1: Calculate the cluster centers
classes of different geometrical shapes in a set of

N
data. In this case, each class has its own matrix of ( ik (l 1) )m zk
v l
 K 1
standard, which leads to: i

N
( ik (l 1) )m
K 1 1 i  c
2 2 Step 3: Calculate the covariance matrix

DikA  z k  vi  ( z k  v i ) T Ai ( z k  v i )
A  N
1  i  c,1  k  N Fi 
K 1 (  ik (l 1) ) m ( z k  vi (l ) )( z k  vi (l ) ) T
N
K 1 (  ik ) m
It is assumed that the matrix Ai tested the
Step 3: Calculate the distances
hypothesis:
2
DikA ( zk  vil )T A( zk  vil )
 1  i  c,1  k  N
 Ai   i   i  0  Step 3: Update the partition matrix
If DikA  0 for 1  i  c,1  k  N

2
Where  i is fixed for each class.
In this case, the optimization of (1) gives us the 1
 ik(1)  2
following expression for Ai : D  m1
 j 1 ikA 
c

1  D jkA 
 Ai  [  i det( Fi )] n Fi 1 
Otherwise
ik(1)  0 if DikA  0
and
ik(1) [0,1]
Where Fi is the covariance matrix blurred the i-

c (1)
 ik  1 U (l )  U (l 1)  
th class given by: with i too .
4
Cluster 1
N
Fi 
K 1 ( ik ) m ( z k  vi )( z k  vi ) T

3 Cluster 2
Cluster 3
N
K 1 ( ik ) m 2 Centre
In the equation (1), the measurement of non- 0
similarity is expressed by of the sum of squares of -1

distances between each data vector and the center of
the corresponding class. The effect of this distance -2
is weighted by the degree of activation ik m -3
-4
corresponding to the Z k data vector, the value of the
-5
cost function FCM

J ( Z ;U ;V )
can be seen as a
-4 -3 -2 -1 0 1 2 3
measure of the total variance of Z k with respect to Fig. 3. Gustafson Kessel clustering algorithm
the  i centers. The minimization of the objective
function (1) is given as flows:
II-5
C. Construction of Takagi-Sugeno models
Consider a system described by the equation (7) :
 y k  f NL ( x k )  
Our objective is to approximate the nonlinear

function fNL of equation (7) by Takagi-Sugeno
model (TS):
R i : if x 1 is A i1 and x 2 is A i2 and … x p is A ip
Then y i  ai x  d i i  1,...,r
aiT  a1 , a1 ,..ain  With R i represents the i iéme

rule
x  x1 , x1 ,..x n  Vector of observations Fig. 4. Examined gas turbine inputs and outputs
used in modeling
A i1 , Ai2 ,…, A in : Represents the fuzzy sets,
y i : represent the output of i iéme rule ai  R p Gas control valve opening
45
is the vector of parameters and d i is scalar .
40
The determination of fNL is done in two steps:
position[%]
35
Step1. It starts first of all by apply the
clustering algorithm Gk, in order to calculate the 30
matrix of fuzzy partition U.
25
Step 2. We then estimate the parameters a and b 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
. In effect, the method of defuzzification used in the Time[S]

x 10
4
model of Takagi-Segeno, is linear in relation to the Axial Compressor Discharge pressure

110
parameters. However, these parameters can be
estimated by using the techniques of least-squares. 105
Pressure [bar]
100
3. GAS TURBINE SYSTEM 95

MODELING
90
In this study, the identification of gas turbine 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
system is carried out in closed loop, or the control Time [S] 4
x 10
type is isochronous, the use of an open-loop Axial Compressor Discharge Temperature
procedure for identification is not preferred. The 600
outputs inputs data have to be correlated together

and generated in normal operating conditions during 580
Temperature [F°]
5 hours and half, the inputs parameters are axial

compressor discharge pressure and temperature
560
(PCD)(TCD), Gas control valve opening (GCV),
and the outputs are the speed of the turbine and
Exhaust temperature (ET) Figure(4). 540
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Time [S] 4
x 10
Fig. 5. Validation data of the examined gas turbine

inputs
4
II-6
Speed Rotor
5140
Degree of membership
1
5130
Speed [rpm]
0.5
5120
0
5110 26 28 30 32 34 36 38 40 42
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 GCV
Time [S] 4
1
x 10
Exhausted Temperature
650
0.5
600
Temperature [F°]
550
0
500 94 96 98 100 102 104 106 108
Pressure
450
1
400
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Time [S] 4
x 10 0.5
Fig. 6. Validation data of the examined gas turbine 0

545 550 555 560 565 570 575 580 585 590 595
outputs Temperature
The obtained result of this simulation is shown Fig. 9. Membership functions used in fuzzy models
in Figures 7 and 8 representing respectively the two inputs
main outputs of the fuzzy models ; speed turbine
and exhausted temperature. Figure 9 and 10 shown
the membership functions used in fuzzy models. 1
0.8
0.6
5140
0.4
Real Data 0.2

5135 fuzzy Model
0
5120 5125 5130 5135 5140 5145
Speed Rotor
5130
vitesse de rotor [RPM]
1
0.8
5125
0.6
0.4
5120 0.2
0
440 460 480 500 520 540 560 580 600 620
Exhausted Temperature
5115
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Time[S] 4
x 10
Fig. 10. Membership functions used in fuzzy models
outputs
Fig. 7. Variation of obtenied model of speed rotor
The parameter used for the numeric validation is
620
the root-mean-square error (RMSE). and is shown in
600
Table II.
580
560 Real Data D. Root mean square error (RMSE)

The RMSE is a frequently used measure of the
Fuzzy Model
Temperature [F°]
540
520
difference between values predicted by a model and
500
the values observed from the environment that is
480
being modeled. These individual differences are
460
also called residuals.The RMSE of a model
440
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 prediction with respect to the estimated variable X
Time [s]
model is defined as the square root of the mean
4
x 10
squared error, given by:

Fig. 8. Variation of obtenied model of the exhaust
temperature
II-7
utilization in gas turbine plants via the

n
( X obs,i - X mo del ,i ) 2
RMSE  i 1
integration of power to gas. Petroleum Journal,
n  March 2017,
5. Ahmed Hafaifa, Guemana Mouloud, and
Belhadef Rachid, Fuzzy Modeling and Control
Table 2. Table I: RMSE comparison between the of Centrifugal Compressor Used in Gas
two model Pipelines Systems. Multiphysics Modelling and
Simulation for Systems Design and
Gas turbine parameter Monitoring, Book Chapter, Applied Condition
Monitoring vol. 2, 2015, pp.379-389.
Model type LP shaft ET system 6. Babuška R., Verbruggen H.B., Identification of
speed composite linear models via fuzzy clustering.
Fuzzy model 0,0076 0,011846 Proceedings of the European Control
Conference, Rome, Italy, 1995, pp. 1207–1212.
7. Babuška Robert, Fuzzy modeling for control.
International Series in Intelligent Technologies,
4. CONCLUSION
Volume 12, Kluwer Academic, Publishers
This work has addressed one of the major Norwell, MA, USA, ISBN: 0792381548,
problems when looking for a reliable mathematical 9780792381549, 1998.
representation; the proposed Fuzzy model provides 8. Balazs Feil, Janos Abonyi, Ferenc Szeifert,
a good improvement in performance during its Model order selection of nonlinear input–
operation in the examined gas turbine. The use of a output models: A clustering based approach.
GK fuzzy clustering algorithm has the important Journal of Process Control, 2004, vol. 14, no. 6,
advantage to allow the automatic generation of pp. 593-602.
membership functions of fuzzy regions from studied 9. Berchtold M., Riedel T., Decker C. and Van
data. This work, confirm that the development of Laerhoven K., Gath-Geva specification and
digital approach is obviously More flexible to genetic generalization of Takagi-Sugeno-Kang
implement. The obtained results from data fuzzy models. IEEE International Conference
classification with the associated models on Systems, Man and Cybernetics. 2008, vol.
construction offer advantageous performance in SMC 2008, pp. 595 – 600.
modeling of the examined gas turbine system. This 10. Bezdek J.C. and Dunn, J.C., “Optimal Fuzzy
approach can provide reliable models for controlling Partitions: A Heuristic for Estimating the
the gas turbine and system Fault diagnosis. Parameters in a Mixture of Normal
Distributions”. IEEE Transactions on
REFERENCES Computers, 1975, vol. C-24, no. 8, pp. 835 –
1. Abdelhafid Benyounes, Ahmed Hafaifa and 838.
Mouloud Guemana, Fuzzy logic addresses 11. Bezdek J.C. and Dunn, J.C., Optimal fuzzy
turbine vibration on Algerian gas line. Oil & partitions: A heuristic for estimating the
Gas Journal, 2016, vol. 114, no. 1, pp. 22-28. parameters in a mixture of normal distributions.
2. Abdelhafid Benyounes, Ahmed Hafaifa and IEEE Transactions on Computers, 1975, vol. C-
Mouloud Guemana, Gas turbine modelling 24, no. 8, pp. 835 – 838.
based on fuzzy clustering algorithm using 12. Buckley J.J., Universal fuzzy controllers.
experimental data. Journal of Applied Artificial Automatica, 1992, vol. 28, no. 6, pp. 1245–
Intelligence, Taylor & Francis, 2016, vol. 30, 1248.
no. 1, pp. 29-51. 13. Enrique H. Ruspini, A new approach to
3. Abdelhafid Benyounes, Ahmed Hafaifa, clustering. Information and Control, 1969, vol.
Abdellah Kouzou1 and Mouloud Guemana, 15, no. 1, pp. 22–32.
Gas turbine modeling using adaptive fuzzy 14. Ford C.L., Carrotte J.F., Walker A.D., The
neural network approach based on measured application of porous media to simulate the
data classification. Mathematics-in-Industry, upstream effects of gas turbine injector swirl
Case Studies, vol.7, no.4, Springer 2016, DOI vanes. Computers & Fluids, 2013, vol. 77, pp.
10.1186/s40929-016-0006-3. 143-151.
4. Ahmed Boubenia, Ahmed Hafaifa, Abdellah 15. Günyaz Ablay, A modeling and control
Kouzou, Kamal Mohammedi and Mohamed approach to advanced nuclear power plants
Becherif, Carbone dioxide capture and with gas turbines. Energy Conversion and
Management, 2013, vol. 76, pp. 899-909.
6
II-8
16. Gustafson D.E. and Kessel W.C., Fuzzy ASME J. Eng. Power, 1983, vol. 105, pp. 865–
clustering with a fuzzy covariance matrix. 869.
Proceedings of IEEE Conference on Decision 22. Ruano, A. E., Fleming, P. J., Teixeira, C., Rodrı
and Control including the 17th Symposium on Guez-V azquez, K., and Fonseca, C. M.,
Adaptive Processes, 1978, vol. 17, Part 1, pp. Nonlinear identification of aircraft gas turbine
761 – 766. dynamics. Neurocomputing, 2003, vol. 55, no.
17. Kulikov G.G. and Thompson H.A., Dynamic 3–4, pp. 551–579.
modeling of gas turbines. Book, 1st edition 23. Setnes M., Babuška R., Verbruggen
2004, London, Springer. H.B.,Complexity reduction in fuzzy modeling.
18. Mohamed Ben Rahmoune, Ahmed Hafaifa, Mathematics and Computers in Simulation,
Mouloud Guemana, Fault Diagnosis in Gas 1998, vol. 46, no. 5–6, pp. 507-516.
Turbine Based on Neural Networks: Vibrations 24. Takagi T. and Sugeno M., Fuzzy identification
Speed Application. Book Chapter in Advances of systems and its applications to modelling
in Acoustics and Vibration, Volume 5 of the and control. IEEE Transactions on Systems
series Applied Condition Monitoring pp 1-11, Man and Cybernetics, 1985, vol. SMC-15, no.
on line. 02 September 2016, ISBN: 978-3-319- 1, pp. 116 – 132.
41458-4, 2017.
19. Nadji Hadroug, Ahmed Hafaifa, Abdellah
Kouzou and Ahmed Chaibet, Faults detection Authors’ contacts
in gas turbine using hybrid adaptive network Abdelhafid Benyounes is with the Applied
based fuzzy inference systems to controlling Automation and Industrial Diagnostics
there dynamic behavior. DIAGNOSTYKA the Laboratory, Djelfa University, Algeria (Email:
Journal of Polish Society of Technical ab_benyounes@univ-djelfa.dz).
Diagnostics (PSTD), 2016, vol. 17, no. 4, pp. 3- Abdellah Kouzou is with the Applied
17. Automation and Industrial Diagnostics
20. Nadji Hadroug, Ahmed Hafaifa, Kouzou Laboratory, Djelfa University, Algeria (e-mail:
Abdellah and Ahmed Chaibet, Dynamic model kouzouabdellah@ieee.org).
linearization of two shafts gas turbine via their
input / output data around the equilibrium Mouloud Guemana is with the Science and
points. Energy Elsevier Journal, 2016, DOI: Technology Faculty, Médéa University, Algeria.
http://dx.doi.org/10.1016/j.energy.2016.11.099 (e-mail: guemana.mouloud@univ-medea.dz)
21. Rowen W.I., Simplified mathematical Ahmed Hafaifa is with the Applied Automation
representations of heavy duty gas turbines. and Industrial Diagnostics Laboratory, Djelfa
University, Algeria. (e-mail:
hafaifa.ahmed.dz@ieee.org).
Multivariable output-error model identification of

energy process based on extended least square
kronecker operator using experimental data
Bachir Nail1,* , Abdellah Kouzou1 , Ahmed Hafaifa1 , and Ahmed Chaibet2
1
The Applied Automation and Industrial Diagnostics Laboratory LAADI, University of DJELFA, Algeria.
2
Aeronautical Aerospace Automotive Railway Engineering school, ESTACA Paris, France.
*
b.nail@univ-djelfa.dz, kouzouabdellah@ieee.org, hafaifa@hotmail.com, Ahmed.CHAIBET@estaca.fr
Abstract—This paper focuses on finding an approximate Mul- chamber and power turbine , and alternator drived by this gas
tivariable Output-Error (MIMO OE) model of energy process turbine (GE MS 5001P).
(power plant) based on experimental information acquired on
site. In order to obtain the parametric model, the Extended MANY results have been published about the modelling
Least Square with the help of the Kronecker Operator (ELSK) of gas turbine and energetic plant using largely the artificial
and the Left Matrix Fraction Description (LMFD) theory were intelligence theory (ANFIS, Fuzzy Logic and Neural Network
used. There are many validations criteria to select the best
model order, but under restriction, only two criteria were used
models), for the following grounds easy to apply and un-
(AIC, RMSE and VAF), while taking into account that the conditioned constraints in the quality and the characterisation
model must be left coprime (no (pole/zero) cancellation). With of the data (the convergence, the smoothing and the number
the objective to prove the reliability of this model and the of samples is not required), but there are disadvantages on
estimator, a comparative study with recent estimation algorithm these models, the difficulty of controlling, and the problem
was conducted, such as: Multivariable Output Error State Space
(MOESP) and with artificial intelligent model, Adaptive Neuro-
of instability, also the limitations of applying the theory of
fuzzy Inference System (ANFIS). advanced control for the reason of the lack of the mathematical
model, majority of the researches they have been done in
Index Terms—Multivariable model, Output-error, ELSK, LMFD, recent years in this area : Gas turbine modeling based on
Validations criteria, MOESP, ANFIS, energy process, power
plant. fuzzy clustering algorithm [1], Fuzzy Modeling of Centrifugal
Compressor [2], Gas turbine modelling using fuzzy logic and
artificial neural networks[3], and others.
I. I NTRODUCTION
Therefore, the parametric system identification theory is
The implementation of a strategy of control and regulation
used to solve the problem of approximate mathematical model
of an industrial engineering systems requires the use of reliable
which has the same characterization of the empirical model,
models of it which can more exploitable in control. The main
this paper dealing with the implement the theory of left
aime in this work is the modelling of the dynamics behavior of
matrix fraction description (MFD) to identify the behavior
a energetic plant, based on system identification theory using
of the turbo-alternator plant multivariable output-error model.
experimental data. Due to the increasing complexity of this
In the literature many paper have been done in field of the
equipment and their severe operating constraints increase with
(MFD) system identification in the left and the right among
their use of added values on these supervisory strategies.
them : Identification of Turbo-compressor using (LMFD) [4],
In this framework, the examined system is a Turbo- instrumental variable identification and extending the SRIV
alternator (gas turbine+alternator) power plant installed in unit algorithm methods for LMFD models[5],[6] , ARMAX models
of production of electricity in Msila Algeria, this plant is an identification using MFD [7], and MIMO least squares using
internal combustion engine that uses the gaseous energy of the MFD [8], and others [9],[10].
air, converting the chemical energy of the fuel into mechanical
In our study, the MIMO OE model is considered to
energy, it is designed to extract, as far as possible, the energy
represent the dynamic behaviour of energetic plant (Turbo-
of the fuel. Indeed, gas turbines are also called combustion
alternator) by the extended least square kronecker operator
turbines, are used in a wide range of applications, including
estimator based on LMFD theory, using inputs/outputs real
power generation, natural gas transmission, as well as various
data, the inputs signals, GCV: gas control valve, PCD: Pressure
process applications..etc. However, a gas turbine is an internal
compressor discharge and TCD: Temperature compressor dis-
combustion engine, which operates with rotary motion and
charge, and the outputs signals, ET: Exhaust Temperature, RS:
reciprocation. These gas turbines are essentially composed of
Rotor Speed, and PG: Power Generation, the experimental data
three main elements: the axial compressor, the combustion
II-10   
used in this identification acquired on site of length=10000  Om Im ··· Om

  Om

 Om ··· Om 
samples in time of 167 minutes, the selection of the final 
 −1

 .. ..



 Ac = Tc ATc =  Om 
model needed to the validations criteria decision, we have   . . ... 
using three criteria (AICs, RMSE, and VAF) [11], in order  Om Om ... Im 

 −Al −Al−1 ··· −A1
to test the ratability of the obtained model and implemented 


 T
estimator a comparative study has been done with intelligent 
 Bc = Tc B = Om Om · · · Im


artificial model ANFIS and MOESP algorithm introduced by Cc = CTc −1 = Cl Cl−1 · · · C1
Katayama [12].
With
The paper is organized as follow: first an introduction Xc ∈ Rn , Ai ∈ Rm×m , Ci ∈ Rp×m , i = 1, ..., l, Im and Om
discussed the historical and the recent development of system are m × m identity and null matrices respectively, and the
identification, then it is followed by some preliminaries of superscript T denotes the transpose.
MFD transformation and the third section the main procedure c) Definition: The system (1) is Block observable of
of ELSK with application on energetic plant and a discussion index l if the
matrix T
of the obtained results. Finally comments, perspectives and a l−1
has full rank
• Ωo = C T AT C T · · · (AT ) C T
conclusion will finish the paper. n
• l= is an integer
II. P RELIMINARIES p
d) Theorem: The multivariable control system described
Matrix Faction Description (MFD) is a representation of in (10) can be transformed into a Block observable form if
a matrix transfer function of a multivariable system as a two conditions are satisfied [13],[14],[15]:
ratio of two polynomial matrices. An introduction to matrix • l =
n
is an integer.
polynomials and MFD properties are given in this papers, but p
• The system is Block observable of index l .
restricted in this research to show the important properties
which needed to clarify the contribution [13],[14],[15]. If the system is block observable then it can be transformed
in a block observer form. So if n/p = l is an integer, and
A. Transformation To Block Canonical Forms if the block observability matrix Ωo is full rank, then we can
Consider a linear time-invariant system described by a state convert the state equation into block observer form using the
equation in general coordinates: following similarity transformation xo = To x

X(k + 1) = AX(k) + Bu(k) Xo (k + 1) = Ao Xo (k) + Bo u(k)
(1) (3)
Y (k) = CX(k) + Du(k) Y (k) = Co Xo (k) + Do u(k)

where: To = Tol ATol ... Al−1 Tol , Tol = Ω−1o Co
T
X ∈ Rn , Y ∈ Rp ,u ∈ Rm , A ∈ Rn×n , B ∈ Rn×m , and   

 O p · · · Op −Al
C ∈ Rp×n . 
  Ip · · · Op −Al−1 

  
 . .. .. ..


 Ao = To −1 ATo =  .. . . .


a) Definition: The system (1) is Block controllable of 
  

  Op · · · Op −A2 
index l if the matrix Op · · · Ip −A1
2
• Ωc = [B, AB, A B, ..., A
l−1
B] has full rank     T

 Bl Op
n 

• l = is an integer 
  Bl−1   Op 
m 

 Bo = To −1 B = 

 
.. , Co = CTo =  .. 

b) Theorem 1: The multivariable control system de- 
  .   . 

scribed in (1) can be transformed into a Block controller form B1 Ip
if two conditions are satisfied [13],[14],[15] :
n III. MIMO OE MODEL IDENTIFICATION USING EKLS
• l = is an integer. ESTIMATOR BASED ON LEFT MFD
m
A MIMO OE (output-error) model given as:
• The system is Block controllable of index l .
If both conditions are satisfied, then the change of coordinates A(q −1 )y[k] = B(q −1 )u[k] + e[k] (4)
Xc (k) = Tc X(k) transforms the system into the following Can be written in LMFD form as:
Block controller form −1 −1
y[k] = A(q −1 ) B(q −1 )u[k] + A(q −1 ) e[k] (5)
Xc (k + 1) = Ac Xc (k) + Bc u(k)
(2)
Y (k) = Cc Xc (k) + Dc u(k) Where u[k] ∈ Rm and y[k] ∈ Rp are inputs and outputs
vectors of the system respectively, while e[k] ∈ Rp is a white-
where:
  noise signal and the polynomial matrices A(q −1 ), B(q −1 )
Tc1 have the following structure:
 Tc1 A 
 
Tc =  .. , Tc1 = Bc T Ωc −1 A(q −1 ) = Ip + A1 q −1 + · · · + Ana q −na (6)
 .  −1 −1 −nb
B(q ) = B1 q + · · · + Bnb q (7)
Tc1 Al−1
II-11
The objective is to identify the matrix coefficients Ai ∈ Rp×p and Φuf and Φyf are constructed as follows :
and Bi ∈ Rp×m of the matrix polynomials A(q −1 ) and 
−yf [1 + p × (na − 1) , :]
 
−yf [1, :]

B(q −1 ) .  ..   .. 
Expanding equation (4) yields Φyf = 


.
···
  .


−yf [p × (N − 1) , :] −yf [p × (N − na) , :]
y [k] = − A1 [k − 1] · · · − Ana y [k − na] + B1 u [K − 1] · · · + (18)
Bnb u [k − nb] + e [k] (8) N is the number of I/O data.
Then using the Kronecker operator we get IV. MOESP AND ANFIS A PPROACHES

T T
IP ⊗ y[k] col (IP ) = − IP ⊗ y[k − 1] col AT1 · · · − In this paper, we have conducted a comparison between
our algorithm EKLS and MOESP and ANFIS approaches, and
T T
IP ⊗ y[k − na] col ATna + IP ⊗ u[k − 1] col B1T due to the lack of ability to display all the theories related to
T them, cite only the sources for those wishing to learn more, for
T
· · · + IP ⊗ u[k − nb] col Bnb + e [k] (9) MOESP algorithm introduced by Katayama [12], ANFIS, in
Here, the col{.} operator is one which forms a vector from a matlab there is a interface dedicated to identify and modelling
matrix by stacking its columns on top of one another. any inputs/outputs data with ANFIS, but it is limited to give
Moreover, equation 2.2.7 can be written as a model with single output therefore we are obliged to obtain
the numbers of outputs of sub-models in our case is three [3].
T T
e [k] = IP ⊗ y[k] col (IP ) + IP ⊗ y[k − 1] · · · +
V. A PPLICATION
T T
IP ⊗ y[k − na] − IP ⊗ u[k − 1] · · · − The system under examination is power plant used in the
  generation of electricity in unit of Sonelgaz M’sila Algeria.
col AT1
 ..  It consists of two main parts, first alternator which generate
 .  the electricity and trained by the the second part is the gas

 col AT 


T  na  turbine (GE MS5001P), the figure 1 present real show of gas
IP ⊗ u[k − nb]  col B1T  (10)
  turbine, and the figure 2 show the schematic block diagram
 ..  of Turbo-alternator with inputs/piputs position of the Turbo-
 
. T alternator model, the length of the measured inputs/outputs
col Bnb
real data used in this identification is 104 samples taken in
Or simply, period of 104 seconds.

T
e [k] = IP ⊗ y[k] col (IP ) − φTf [k] θ (11)
Let’s define the new signals

T
yf f [k] = IP ⊗ y[k] col (IP ) (12)

T
yf [k] = IP ⊗ y[k] (13)

T
uf [k] = IP ⊗ u[k] (14)
−1 T
θ̂ = ΦTf Φf Φf Y f (15)
These signals yf f [k], yf [k] and uf [k] belong to Rp×1 ,
Rp×pp and Rp×pm .respectively
 
yf f [1 + p × na]
 .. 
Yf =   Fig. 1: GE MS 5001P Gas Turbine
 . 
yf f [p × N ] A. Validation of the Model
Then θ can be estimated using least squares The model that have been select must have order n of the

Φf = Φyf Φuf (16) multiplier of 3, numbers (3,6,9...) because each l (l = np )
matrices blocks (Ai , or Bi , i = 1...l ) have a dimension of
where 3×3, because we have m=3 inputs and p=3 outputs . The table
   
uf [1 + p × (na − 1) , :] uf [1 + p × (na − nb) , :] shown below clarify the change of the orders n until we stop
 ..   .. 
Φuf =  
.
···
.
 at the best one.
   
From the table 1, the best model having both minimums
uf [p × (N − 1) , :] uf [p × (N − nb) , :]
(17) AICs/FPEs and RMSE and maximum VAF with no Pole/Zero
II-12
cancellation is the one having order n=12 with l=4 blocks.
The system write in linear discrete-time state space Block 5500
Observable canonical Form as follow :
5000 REAL
x (k + 1) = Ax (k) + Bu (k)
(19) ELSK
y (k) = Cx (k) + Du (k) 4500 ANFIS
Speed (rpm)
      MOESP
O O O −A4 B4 O 4000
 I O O −A3     O  4905
A= , B =  B3 , C T =   3500
 O I O −A2   B2   O 
O O I −A1 B1 I
3000 4895
     
0 0 0 1 0 0 0 0 0
D =  0 0 0 , I =  0 1 0 , O =  0 0 0  2500
4885
0 0 0 0 0 1 0 0 0 7950 7975 8000
2000
Where, 0 2000 4000 6000 8000 10000
  Time (s)
0 0 0 0 0 0 0 0 0 −0.3216 0.0066 −0.0000
 0 0 0 0 0 0 0 0 0 −4.6996 −0.0321 −0.0014 
 

 0 0 0 0 0 0 0 0 0 6.3228 −0.2936 −0.1102 
 Fig. 4: Rotor Speed signal
 1 0 0 0 0 0 0 0 0 0.0207 −0.0097 0.0004 
 
 0 1 0 0 0 0 0 0 0 −5.3544 −0.1124 −0.0007 
  10000
 0 0 1 0 0 0 0 0 0 33.057 0.9956 0.0379 
A=  
 0 0 0 1 0 0 0 0 0 0.5639 0.0028 −0.0006  REAL
 
 0 0 0 0 1 0 0 0 0 3.3842 0.1703 −0.0063 



 9500 ELSK
 0 0 0 0 0 1 0 0 0 −10.600 0.4401 −0.0453 
 0 0 0 0 0 0 1 0 0 0.7361 0.0003 0.0003  ANFIS
 
 0 0 0 0 0 0 0 1 0 6.7027 0.9739 0.0052  MOESP
Power (KW)
0 0 0 0 0 0 0 0 1 −28.890 −1.1427 1.1217 9000

    9865
−0.0000 −0.0014 0.0021 0 0 0
 −0.0000 −0.0180 −0.2198   0 0 0 
    8500 9860
 −0.0004 −0.2406 1.1031   0 0 0 
   
 0.0000 −0.0010 −0.0139   0 0 0 
    9855
 −0.0000 −0.0690 0.3445   0 0 0 
   
 0.0002 0.0474 0.7124   0 0 0  8000
B= 
 −0.0000 0.0006 −0.0276
,
 CT = 
 0 0 0

 7210 7220 7230
   
 −0.0000 −0.0357 0.4108   0 0 0 
   
 −0.0002 0.0539 0.9463   0 0 0 
    7500
 0.0000 0.0042 0.0221   1 0 0  0 2000 4000 6000 8000 10000
   
 0.0000 0.0266 0.1108   0 1 0  Time (s)
0.0045 −0.1648 −0.7925 0 0 1
Fig. 5: Power Generation signal
4
Exhausted Temperature (F°)
620
ELSK
3
600 REAL ANFIS
ELSK 2 MOESP
580 ANFIS
1
Error
MOESP
560 0
616.5
540 616 −1
520 615.5
−2
615
500 614.5 −3
7020 7030 7040
480 −4
0 2000 4000 6000 8000 10000
0 2000 4000 6000 8000 10000
Time (s)
Time (s)
Fig. 3: Exhausted Temperature signal Fig. 6: Error validation of Exhausted Temperature signal
II-13
GCV y3
y1
Fuel Gas Combustor Power
u1 Exhaust Generation
Temperature
Air Inlet
Coupling
PCD Turb/Alter
Alternator
Turbine
Lancer
u2
Axial Compressor
u3
Coupling Rotor Speed
TCD
Lner/Comp y2
Fig. 2: Schematic Block Diagram of Turbo-alternator power plant

1
TABLE I: Turbo-alternator power plant model orders n and blocks l with validations criteria values
ELSK MOESP ANFIS
l n AIC FPE RMSE VAF(%) AIC FPE RMSE VAF(%) AIC FPE RMSE VAF(%)
1 3 2.5483 12.7855 9.65e+27 -5.59e+51 2.5769 13.1561 9.44e+28 6.89e+52 ··· ··· ··· ···
2 6 0.0453 1.0463 1.54e+5 -1.20e+6 0.0260 1.0263 2.44e+5 2.50e+6 ··· ··· ··· ···
3 9 0.2630 1.3008 358.1150 92.4540 0.3732 1.4524 388.1250 90.3245 -2.4348 0.0885 186.2565 98.1658
4 12 0.1152 1.1221 239.6157 96.6654 0.1451 1.1561 248.5187 94.9854 ··· ··· ··· ···
5 15 0.1997 1.2212 245.3060 95.4452 0.4326 1.5412 254.4860 93.0548 ··· ··· ··· ···
42
80
ELSK
Valve Position (%)
60 40
ANFIS
40
MOESP 38
Error
20 36 39.92
39.91
0
34 39.9
6220 6240 6260
−20
32
−40
30
0 2000 4000 6000 8000 10000 0 2000 4000 6000 8000 10000
Time (s) Time(s)
Fig. 7: Error validation of Rotor Speed signal Fig. 9: Gas Control Valve signal
60 110
40
108
Pressure (Psia)
20
106
Error
0
104 107.8
−20 107.75
102
−40 ELSK 107.7
ANFIS 107.65
100
−60 MOESP 107.6
6800 6850
−80 98
0 2000 4000 6000 8000 10000 0 2000 4000 6000 8000 10000
Time (s) Time(s)
Fig. 8: Error validation of Power Generation Fig. 10: Axial Compressor Discharge Pressure signal
II-14
is off-line, but in our case in this work for more precision.
595 As further works, can make this EKLS estimator in on-line
system identification (recursive loop) and done a comparative
590 study, also can applied many advanced control theory on this
Temperature (F°)
obtained dynamical model.

585 VII. A PPENDIX
591
580 TABLE II: The main performance of the turbo-alternator
590.5
575 590 Quantity Value

Exhaust temperature 898 (F◦ )
589.5 Output power 24,700 (kW)
570 7020 7030 7040
Rotor Speed 5355 (rpm) (105%)
565
0 2000 4000 6000 8000 10000 R EFERENCES
Time(s) [1] Abdelhafid Benyounes, Ahmed Hafaifa & Mouloud Guemana, Gas tur-
bine modelling based on fuzzy clustering algorithm using experimental
Fig. 11: Axial Compressor Discharge Temperature signal data. Journal of Applied Artificial Intelligence, Taylor & Francis, 2016,
vol. no., pp.
[2] Ahmed Hafaifa, Guemana Mouloud, and Belhadef Rachid, Fuzzy Mod-
B. Discussion of the obtained results eling and Control of Centrifugal Compressor Used in Gas Pipelines
From the obtained results shown in the previous figures Systems. Multiphysics Modelling and Simulation for Systems Design and
Monitoring, Book Chapter, Applied Condition Monitoring, 2015, vol. 2,
and due to the values numbers of the validations criteria in pp.379-389.
table 1, can we say the obtained model is written in the block [3] Abdelhafid Benyounes, Ahmed Hafaifa and Mouloud Guemana, Gas
observable canonical form of index l=4 blocks, on the other turbine parameters modelling based on fuzzy logic and artificial neural
networks: Solar TITAN 130N investigation. 5me Symposium de lAs-
hand by compared ELSK model with MOESP and ANFIS sociation Algrienne de lIndustrie du Gaz AIG Sonatrach, 16-17 Fvrier
models, it is clear that obtained model is better than the 2016, Alger.
MOESP model, and more suitable with ANFIS model, and [4] Bachir Nail, Abdellah Kouzou, Ahmed Hafaifa, and Choayb Djeddi,
Parametric Identification of Multivariable Industrial System Using Left
through the figures of the outputs signals, the figures 3-5, Matrix Fraction Description. J. Automation & Systems Engineering ,
the real signals of the exhausted temperature, rotor speed 10-4 (2016): 221-230.
and power generation are identical with the estimated signals [5] Akroum, M., Hariche K., An optimal instrumental variable identification
method for LMFD models, Studies in informatics and control, Vol 17,
in all the time (0:1000) seconds and no statistical error and No.4, December 2008.
peak , the figures 6-8 demonstrated the convergence of the [6] Akroum, M., and Hariche K., Extending the SRIV algorithm to LMFD
this identification (unbiased estimator), where the error con- models, Journal of electrical engineering and technology, Korean Insti-
tute of electrical engineers, Vol. 4, No.1, march 2009.
verge asymptotically, the outputs signals, the figures 9-11, gas [7] Akroum, M., and Hariche K., A matrix fraction description-based
control valve, axial compressor discharge pressure, and axial identification algorithm for MIMO ARMAX models, submitted to Asian
compressor discharge temperature are riches with frequency journal of control, 2008
[8] R.A. de Callafon, D. de Roover, P.M.J. Van den Hof Multivariable
which covered all the behavior of the Turbo-alternator. least squares frequency domain identification using polynomial matrix
fraction descriptions. Decision and Control, 1996., Proceedings of the
VI. C ONCLUSION 36th IEEE Conference on Decision and Control,13-13 Dec. 1996.
[9] Nehorai, A., and Morf, M., Recursive identification algorithms for Right
To the best of our knowledge, this is a new contribution Matrix Fraction Description models . IEEE Transactions on Automatic
combined the (LMFD) theory in identification of class of Control, 29, 1103-1106, 1984.
MIMO OE model with the help of EKLS estimator and [10] R. Johansson, Multivariable system identificationvia continued-fraction
approximation, IEEE Transactions on Automatic Control, 40(3), 507-
the first application on the energetic domain Turbo-alternator 512, 1995.
investigation, using inputs/outputs real data, based on the [11] Ljung, L.,System Identification : Theory for the user (Prentice Hall ),
richest of the acquired informations with frequency about 1999.
[12] Katayama, T., Subspace methods for System Identification, Springer-
the black box (Turbo-alternator) plant, and the efficient of Verlag, 2005.
the kronicker product in the extended least square estimator [13] Bachir Nail, Abdellah Kouzou, Ahmed Hafaifa, Belkacem
and the flexibility of the LMFD theory all of this advantages Bekhiti,Parametric output feedback stabilization in MIMO systems:
Application to gas turbine power plant, 8th International Conference
allowed us to obtain an accurate identification and a valid on Modelling, Identification and Control (ICMIC), 971-976, November
and reliable model with no factors common (coprime) , this 2016.
is according to the results of the validations criteria (AIC, [14] L. S. Shieh and Y. T. Tsay, Transformation of a class of multivariable
control systems to Block companion forms,IEEE Truns. Autonwt. Con-
FPE, RMSE and VAF) and the comparative study with recent trol 27: 199-203 (1982).
MOESP algorithm which generate subspace model and with [15] L. S. Shieh and Y. T. Tsay, Transformation of solvent and spectral factors
artificial intelligence model ANFIS, and recognized that they of matrix polynomial, and their applications,Internat. J. Control 34:813-
823 (1981).
give high-fidelity models, and can say that EKLS is unbiased [16] I. Gohberg, P. Lancaster, L. Rodman, Matrix Polynomials,Academic
estimator, despite this is prejudge, due to this identification Press, 1982.
BILATERAL FILTER INTEGRATION INTO THE

OPEN SOURCE SOFTWARE GELJ
ANGEL DANEV, ATANASKA BOSAKOVA-ARDENSKA, ILIYAN DOBREV, TODOR STAMENOV
Abstract: This paper presents the application of the computer program GelJ with additional
bilateral filter implementation in the context of DNA analyses of lactic acid bacteria
recombinants. The results show that GelJ could be applied for lactic acid bacteria
recombinants study.
Key words: open source software, GelJ, DNA analyses, bilateral filter, lactic acid bacteria
recombinants.
1. Introduction and difficult to use in some cases. On the other hand

The open source software began in the late free tools support the basic functionality for DNA
1970s and early 1980s when the shearing of source analysis. So, the main advantage of GelJ is that it is
code from a computer program and the proprietary free and open source software. This program is user-
software began to really come into conflict [1]. friendly, platform-independent and feather-weight.
The term “open source software” means GelJ provides a lot of outstanding features such as
computer software that is made available for people digital image processing including different types of
to study, modify and distribute the source code to filters, mechanisms for accurate lane- and band-
anyone and for any purpose. The copyright holder detection, a number of band- and curve-based
provides these rights and there is no problems for similarity methods, different techniques for
software developers to inspect, change and enhance generating dendrograms, comparison of banding
the source code. The difference between open source patterns from different experiments, and database
software and other types of software is that the support.
authors of open source software make its source code
2. Bilateral filter
available to others who would like to view that code,
copy it, learn from it or share it. On the other hand The process of filtering in terms of digital
some software has source code that only the person image processing is fundamental operation in the
or organization who created it can modify or share. science of computer vision and image processing.
Only the original authors can legally copy, inspect The term “filtering” means that the value of the
and alter that software. filtered image at the given location (pixel of the
There are different applications of open image) is a function of the values of the input image
source software. In this paper is presented a in a small neighborhood of the same pixel or entire
computer open source program that is used for image.
specific analyses of Lactobacillus recombinants in Bilateral filter is firstly presented by Tomasi
gel electrophoresis. and Manduchi in 1998 [4]. The concept of the
GelJ version 2.0 is developed by Jόnathan bilateral filter has been also presented in [5] as the
Heras (joheras@gmail.com), César Domínguez, SUSAN filter and in [6] as the neighborhood filter.
Eloy Mata, César Larrea, and Vico Pascual at the In brief, bilateral filtering smooths images
Department of Mathematics and Computer Science while preserving edges, by means of a non-linear
of University of La Rioja (Spain) and Mysic [2]. The combination of nearby image values. Each pixel is
program is an open source software for analyzing replaced by an average of its neighbors.
DNA fingerprint gel images. DNA fingerprinting is This type of filter is an effective image
a technique for comparing DNA patterns that has denoising technique and it can be used for blurring
applications in a wide variety of contexts [3]. There an image while respective strong edges too [4, 7].
are several freely-available and commercial tools The bilateral filter is also defined as a weighted
that can be used for analyzing DNA fingerprint average of nearby pixels in a manner very similar to
images, but the commercial software is expensive Gaussian convolution. The difference is that the
Copyright  by Technical University - Sofia, Plovdiv branch, Bulgaria

II-16
bilateral filter takes into consideration the difference fairing and e.t.c. Denoising of an image is the
in value with the neighbors to preserve edges while primary goal of the bilateral filter and it is used in
smoothing. The bilateral filter is defined by the applications for medical image processing, tracking,
following formula: movie restoration and so on [7].
𝑩𝑩𝑩𝑩[𝑰𝑰]𝒑𝒑 =
𝟏𝟏
∑𝒒𝒒𝒒𝒒𝒒𝒒 𝑮𝑮𝝈𝝈𝒔𝒔 (‖𝒑𝒑 − 𝒒𝒒‖) 𝑮𝑮𝝈𝝈𝒓𝒓 (|𝑰𝑰𝒑𝒑 − 𝑰𝑰𝒒𝒒 |)𝑰𝑰𝒒𝒒 (1) 3. GelJ structure and bilateral filter
𝑾𝑾𝒑𝒑
integration
where 𝑰𝑰𝒑𝒑 is notation for the gray-scale image GelJ is a software product for analyzing
DNA fingerprint gel images. It is developed using
value at pixel position p; the weight of pixel q is
the objected oriented language Java. GelJ is platform
defined by the Gaussian 𝐺𝐺𝜎𝜎 (‖𝑝𝑝 − 𝑞𝑞‖) where σ is a
independent program because Java is platform
parameter defining the neighborhood size; s is the set
independent language. The product provides
of all possible image locations; r is the set of all
different features for image processing using the
possible pixel values. Normalization factor 𝑊𝑊𝑝𝑝
ImageJ library [8]. On the other hand a library called
ensures pixel weights sum to 1.0 [7]: Weka [9] provides variety of machine learning
algorithms for data mining tasks [9, 10]. GelJ
𝑾𝑾𝒑𝒑 = ∑𝒒𝒒𝒒𝒒𝒒𝒒 𝑮𝑮𝝈𝝈𝒔𝒔 (‖𝒑𝒑 − 𝒒𝒒‖) 𝑮𝑮𝝈𝝈𝒓𝒓 (|𝑰𝑰𝒑𝒑 − 𝑰𝑰𝒒𝒒 |) (2)
contains a special embedded database that is
integrated using a library called JavaDB. GelJ has
where the parameters 𝜎𝜎𝑠𝑠 and 𝜎𝜎𝑟𝑟 specify the
three common concepts – experiment, comparison
level of filtering for the image I [7].
and research. The process of making an experiment
Median and bilateral filter are applied to an
refers to analyzing the DNA fingerprints gel image
image in figure 1 (A) to yield the image in figure 1
provided from the biological experiment. Than the
(C).
process of comparison evaluates the similarities
between the samples from one or more experiments.
At the end researching process collect all data from
the experiments and comparisons. The main window
of the GelJ program is shown in figure 2.
Fig. 1. A picture before (A) and after (B)

median and (C) bilateral filtering. Fig. 2. GelJ main window
The original image has salt-and-pepper The process of making experiments consists
noise. It can be seen that the most of the fine texture of several steps that the user have to go through –
has been filtered away, and yet all contours are as preprocessing of the DNA fingerprint gel image, lane
crisp as in the original image. Figure 1 (d) shows a detection, normalization and bands detection.
detail of figures 1 (a), 1 (b), and 1 (c). Figure 3 illustrates all steps needed for gel
The bilateral filter can be applied to either image processing in details. On the first step the user
grayscale images and to colored images. can crop the image to cut out a specific important
There are variety of applications that the region. On the other hand the user can adjust
bilateral filter can be used. According to the specific brightness and contrast of the image; invert the colors
needs this type of filter is used for image denoising, and so on. The second step is lanes detection. The
texture and illumination separation, data fusion, 3D user have two options- automatically or manual
II-17
detection. He can choose to detect the lanes

automatically but after that manually editing is
required. It is important to choose and mark the
reference lanes. Bands detection is the third step of
every analysis. First the user can automatically detect
all of the bands in the entire image. For greater
precision manually picking and editing the bands is
required. The program provides features for adding
and removing bands. The step of normalization is
important for creating accurate analysis. There is a
feature for adding or loading reference marker.
Fig. 4. Dropdown menu for filter type selection
Value Channels – This parameter specify

the number of channels. Every image has color
channels for example some colored image has 3
channels for Red, Green and Blue (RGB)
components.
Mode Iteration – The bilateral filter can be
iterated. The value of this field specify the iteration
schema: 0 –None; 1 –Range Iteration; 2 – Space
Iteration; 3 –Overall Iteration;
Number of iterations – in this field the
number of iterations can be specified. Larger number
of iterations leads to achieve cartoon-like renditions
of images.
For integration of bilateral filter in GelJ as an
Fig. 3. Steps needed for agarose gel image additional feature BilateralFilter_.jar file is
processing downloaded from:
bigwww.epfl.ch/algorithms/bilateral-filter/
Figure 4 illustrates the user interface for
[11, 12].
selecting the bilateral filter and all of the necessary
After that the file is included in a directory
parameters.
called lib where the external libraries are situated. In
Below are listed all of the available
package ij.plugin a class named
parameters that the user can change in order to use
Bilateral_Filter_Instant.java was created. All of the
the bilateral filter.
necessary variables are initialized in it. They are
Sigma Range – as the Sigma Range
passed to the method named process where the
parameter increases the bilateral filter becomes
source code for implementation of bilateral filter is
closer to Gaussian blur.
placed. All of the classes used in the bilateral filter
Sigma X, Y, Z – As the spatial parameter of
library are described below.
bilateral filter (σd) increases, the filter smooth larger
Allocation – memory allocation for storage
features. In this parameter x – is the amount of pixel
all intermediate result from bilateral filter.
neighborhood in x – axis of the coordinate system;
Data – contains the input data needed for
y – is the amount of pixel neighborhood in y – axis
image processing.
of the coordinate system; z – It is a coordinate value
Filter – contains the logic for image
for smoothing 3D meshes while preserving their
processing.
most prominent features [7].
Gaussian – contains the program logic
Method Channels – The possible values
needed for the Gaussian case.
are: 0 – Best channels; 1 – All channels; 2 – By
Initialization – a class responsible for
Truncation; 3 – By Tolerance;
initialization process of tables for sin and cos.
II-18
All of the methods that are part of the Filter

class are listed below.
approximate() – returns three indicators on
the quality of the estimate of the Gaussian function,
discretized on bin items;
computeOrder() – establishing the execution
order of the filter;
execute() – processing the image
selectChanels() – determines the choice of
channels;
setIterativeScheme ()– determines the
iterative scheme;
setLog() – specifies whether to generate a Fig. 6. Automatically detected bands and
log file after treatment; manually processed
setMultithread() – determines whether the
image will be processed multithreaded; As an important step in procedure of DNA
The Gaussian class implements the interface analyses of lactic acid bacteria recombinants is
java.lang.Runnable. It is used following methods: creating a special dendrogram which is designed to
isRunning() - to check whether it is running image illustrate the grade of similarity between couples
processing; run() – for running. parental strains and the recombinant clones. Such a
dendrogram is shown below in figure 7. It is
4. Analyses of lactic acid bacteria
generated by the GelJ software using the Similarity
recombinants
method called Dice [16]; UPGMA linkage and the
The method of agarose gel electrophoresis output is selected to be only dendrogram. The Dice
allows the creation of images containing valuable method is also known as Dice’s coefficient. It is used
information for DNA analysis. Today’s modern for comparing the similarity of two samples. The
computer technologies provides variety of open output can be dendrogram, dendrogram + images,
source software products for creating DNA analyses dendrogram + images + bands. The window for
easier in comparison to manual analyses. creating a dendrogram is shown in figure 8.
In this research, one modern computer
program for DNA analyses was used to analyze lactic
acid bacteria recombinants. Randomly Amplified
Polymorphic DNA (RAPD) analysis was conducted.
It was used an image obtained from biological
experiment. Two main primers are selected – OPP-7
and j-8 [13, 14, 15] that can work with the genomes
of the couples parental strains - L. rhamnosus 1D and
L. acidophilus 2; L. rhamnosus 1D and L.
acidophilus Ar; L. acidophilus 2 and B. bifidum L1.
The image is shown in figure 6.
After a few steps of the image pre-processing
the user can continue his DNA analysis with either Fig. 7. A dendrogram illustrating the grade of
automatically or manually lane detection. The similarity between couples parental strains and
process of automatically lane detection often the recombinant clones
requires manually editing the height, width and
position of the lanes. As a next step in the DNA
analysis using GelJ takes place the procedure of Creating a dendrogram follows the following
bands detection. Again it can be done either iterative process: on each step the nearest two groups
manually or automatically. In the most cases the user are combining one another in a group of higher level.
have to do this manually for greater accuracy. Every single generated dendrogram can additionally
All of the bands automatically detected and contains information about the lanes, the original
after that manually processed using the software image or the image with detected bands.
product GelJ are shown in figure 6. A band marker
should be placed on every important bright peak. The
groups of red colored markers determines which are the
reference lanes.
II-19
5. Smith, S. M. and Brady, J. M. (1997), Susan

- a new approach to low level image
processing, Int. Journal of Computer Vision,
volume (23), pp.45–78.
6. Yaroslavsky, L. (1985). Digital Picture
Processing - An Introduction, Springer
Verlag.
7. Paris, S., Kornprobst, P., Tumblin, J. and
Durand, F. (2008). Bilateral Filtering:
Theory and Applications, Foundations and
Trends in Computer Graphics and Vision,
Vol. 4, No. 1, pp.1–73.
8. Schneider, CA, Rasband, WS, Eliceiri, KW
Fig. 8. Comparison window for generating a (2012). NIH Image to ImageJ. Nat Methods;
dendrogram 9:671–5.
9. Markov, Z., Russell, I. (2006). An
Introduction to the WEKA Data Mining
5. Conclusion System, pp. 367–368
The computer program GelJ as an open
10. Amara,l K., Netbeans IDE Tutorial for using
source software for creating DNA analyses could be
the Weka API, University of Massachusetts
applied for lactic acid bacteria recombinants study.
Boston
Using GelJ by lactic acid bacteria recombinants
analysis sufficiently accurate results were received in 11. Sage, D., Chaudhury, K.N. (2011).
comparison with commercial software products [13, Constant-time bilateral filter, Biomedical
14, 15]. GelJ provides a lot of features thanks to Imaging Group (BIG) Ecole Polytechnique
which the users can easily create their analyses. In Federale de Lausanne (EPFL) Lausanne,
addition the open source conception of the product Switzerland.
allows computer programmers to develop better
12. Chaudhury, K.N., Sage, D., Unser, M.
tools and features.
(2010). Fast O(1) bilateral filtering using
trigonometric range kernels, IEEE
REFERENCE
Transactions on Image Processing,
1. Wong, K., Sayo, P. (2004). Free/Open submitted Dec.
Source Software. A General Introduction. 13. Dobrev, I.N. (2011). Application of
United Nations Development Programme’s lactobacilli and bifidobacteria in the
Asia-Pacific Development Information manufacture of healthy foods from goat
Programme (UNDP-APDIP) Kuala milk, PhD thesis, University of food
Lumpur, Malaysia, ISBN: 983-3094-00-7 technology, Plovdiv.
2. Heras, J., Domínguez, C., Mata, E., Larrea, 14. Denkova, R., Dobrev, I., Denkova, Z.,
C., and Pascual, V. (2016). GelJ manual Yanakieva, V., Urshev, Z., Yordanova, M.,
v.2.0, Department of Mathematics and Ilieva, S. (2012). Genetic, biochemical and
Computer Science of University of La Rioja physiological characteristics of the
(Spain) and Mysic. recombinant strain Lactobacillus RL15,
3. Heras, J., Domínguez C., Mata, E., Pascual, obtained by intergeneric hybridization
V., Lozano, C., Torres, C. (2015). GelJ – a between Lactobacillus acidophilus 2 and
tool for analyzing DNA fingerprint gel Bifidobacterium bifidum L1. Food and
images, BMC Bioinformatics Environmental Safety – Journal of Faculty
of Food Engineering, Stefan cel Mare
4. Zhang, M. (2006), Bilateral filter in image University – Suceava, Volume XI, Issue I –
processing, B.S. Department of Electrical 2012, pp.5-14.
and Computer Engineering, Beijing 15. Goranov, B., Denkova, R., Teneva, D.,
University of Posts and Denkova, Z., Popova, P. (2015). Molecular-
Telecommunications. genetic identification of Lactobacillus
II-20
strains, isolated from homemade yoghurt. Атанаска Босакова-Арденска

Ukrainian Food Journal 4 (1): pp.67-76 УХТ – гр. Пловдив
бул. „Марица“ 26
16. Dice, Lee R. (1945), Measures of the
тел. 032 603 860
Amount of Ecologic Association Between
a_bosakova@uft-plovdiv.bg
Species, University of Michigon,
Ecology, 26 (3): pp.297–302
Илиян Добрев
УХТ – гр. Пловдив
бул. „Марица“ 26
тел. 032 603 696
Контакти: indobrev@abv.bg
Ангел Милчев Данев
УХТ – гр. Пловдив Тодор Стаменов
бул. „Марица“ 26 УХТ – гр. Пловдив
тел. 032 603 860 бул. „Марица“ 26
angel_danev_bg@abv.bg тел. 032 603 860
todor.stamenov93@gmail.com
AUTOMATED SYSTEM FOR CONTROL AND

MANAGEMENT OF A MACHINE FOR AUTOMATIC
ASSEMBLY OF CURTAIN BRACKETS AND HOOKS
PENKO MITEV, GEORGI SHTEREV
Abstract: The automated system for control and management is intended to assemble
curtain brackets and hooks by a predefined movement sequence, registering each of the
details “roll” or “bracket” with their subsequent assembly. In the work is pointed out the
way of choosing of a technological scheme for control and management of the executive
devices for a real automated system built and integrated in the production. The invention is
implemented by "KMS Engineering" LTD - Plovdiv in "UYUT" LTD - Sankt Petersburg.
Key Words: fiber optic converters, logic controller, pneumatic distributors, ejector nozzle.
1. Introduction demonstrating the various process activities.[1,

The target of automation is an automatic 2, 3, 4]. The machine control system if based on
assembly machine which joins together two a programmable logic controller which accepts
plastic parts - "hook" and roller". This features signals from fiber optics sensors for detection
the following processes: of parts presence in assembly position and
 Initial check of all default positions of controls four pneumatic valves through relay
actuators outputs.
 Presence of all start conditions related to the
automatic mode; 2. Networks of possible variants
 Start of automatic mode;
 Stop of automatic mode;
 Control of all mechanisms in manual mode;
 Control of each of the two vibratory bowl
feeders depending on the signals from the
sensors for minimal and maximal parts
quantity;
 Control of the signal lamp tower with red
and green lights;
 Control of the sequence of actions, which
take part in the assembly process;
 Tracking of statistical parameters - Fig. 1. Network of the possible variants for project
"productivity", "number of errors for a realization
period of time", etc.
The main tasks are related to the design Fig. 1 shows a network of the possible
and assembling of the described system, based variants for project realization based on the
on standard components, writing of control main functions which are required from the
algorithm for the technological process and control system. An optimal solution is to be
creation of a simulative laboratory unit found among these variants. Some devices are
II-22
able to do more than one operation or two fulfill Table 1

two functions simultaneously. The process of
decision taking is connected to a complex
analysis of many factors and for each solution
there are various advantages and disadvantages.
Green dots mark the taken decisions for this
specific task.
3. Cyclogram of the machine process
The assembly process is cycle-based
where each cycle takes 1,5 seconds. Fig.2
demonstrates the time for the working stroke of
every of the four pneumatic cylinders(lifting,
assembly, opposing cylinders as well as a
blowout nozzle). In the event of anomaly in the Based on the results, the average feeder
working cycle or in case of other emergency productivity is 3190 parts / hour. However, the
situation, the control system activates light and assembly mechanism is able to work faster.
sound signals to attract attention of the site In case the bowl feeder could not feed
personnel. enough parts, its own feeder rate is the
maximum possible productivity value for the
machine.
5.2. Experimental analysis on the

productivity of the vibratory bowl
feeder for rollers
- after 10 trials the results show that the
rollers feeder has higher feed rate than the
hooks feeder. This is easily explained with
the smaller roller dimensions and the easier
Fig. 2. Cyclogram of the machine process orientation in general.
The results are shown in Table 2:
4. Working algorithm
It is based on several modules - preparatory, Table 2
subroutines calling module, module for
delaying fiber-optic sensors' signals, control
module for vibratory bowl feeders and alarms
handling module. All modules are written in
LADDER language and each module does not
surpass 30 program steps.
5. Experimental analysis
5.1. Experimental analysis on the
productivity of the vibratory bowl
feeder for hooks
The vibratory bowl feeder for hooks is a 5.3. Experimental analysis on the PLC scan
critical part of the whole system. Depending on cycle - a "scan cycle" by PLCs is the time
the number of arriving and properly oriented needed to execute the following sequence
parts, the machine will either have minimal of operations:
presence of parts for normal work cycles, or  check of states on every input - logic "0"
will need to wait until this quantity is attained. or logic "1"
The following statistical information was  program execution
acquired based on 10 trials.  update of output signals based on the
(table 1): changes in the current PLC cycle.
II-23
The PLCs from the XINJE XC3 series process, a detailed research activity related
have a scan cycle from 1 ms to 10 ms, to industrial sensors used in industrial
depending on the program complexity and the environment, automation components and
number of additional expansion modules, innovations in IT and communication
connected to the base PLC. During the phase of technologies is to be done. Additionally,
PLC program creation, the following statistical analysis on the application of programmable
information was acquired: logic controllers for industrial automation
(table 3): projects and other areas, based on the end
user's inquiry are to be researched;
2. The design of the electrical and pneumatic
diagrams is connected with the necessary
Table 3
number of input/output signals and
Number of steps PLC scan cycle (ms) parameters of the assembly process;
inside the PLC
program 3. Economical parameters are to be taken into
100 0 account when making decisions on the
200 1 choice of automation components.
300 1
400 1
500 2 7. REFERENCES
600 2 1. Гановски В., Нешков Т., Бояджиев И.,
700 2 Ликов Ц., Механизация и автоматизация на
800 3 монтажнитепроцеси в машиностроенето“,
900 4 изд. „Техника“, София, 1986 год.
2. KrauseWerner, Geratekonstrution, VEB
The following additional research on the scan VerlagTechnik, Berlin, 1986.
cycle has been conducted independently from 3. Лебедовский М.С., А.И. Федотов,
the system with a resulting scan cycle of 20 ms: Автоматизация сборочныхработ, Л.,
 Base PLC XC3-60PRT-C with 36 inputs Лениздат, 1970.
Таблица 2 and 24 outputs 4. Замятин В.К., Технология и автоматиза-
 7 additional expansion modules (a total of циясборки, Москва, Машиностроение, 1993.
140 inputs / 100 outputs)
 5000 program steps inside the program 8. About the authors
Penko Valkov Mitev: engineer, Msc,
6. Conslusion: "Machine building" (2017), TU-Sofia, branch
Plovdiv;
1. The experience gathered from the project for Georgi Panayotov Shterev: electrical
automatic assembly of hooks and rollers engineer, Msc „Electrical measurement
could be applied in many other projects equipment” (1978), Ph.D (2007), Assoc. Prof.
having similar parts. During the (2009), TU-Sofia, branch Plovdiv
development of a specific manufacturing
MULTI-FUNCTION AND RECONFIGURABLE

ROBUST ANALOG I/O MODULE DESIGN
METHODOLOGY FOR INDUSTRIAL
CONTROLLER SYSTEMS
CAGRI CANDAN, EMRE CAN DIKMEN, T. CETIN AKINCI, SERHAT SEKER
Abstract: In this paper, applied design methodology of robust analog I/O (input and
output) module of industrial controller is explained. Analog I/O modules are one of the
essential part of the controller. Main aspects of the design methodology are reducing
microprocessor processing complexity, selecting proper ICs (Integrated Circuits), analog
signal correlation, transducer reading performance improvements, ADC (analog to digital
converter) and DAC (digital to analog converter) implementation techniques. Additionally,
test results and serial communication protocols also mentioned.
Key words: Analog module design, transducer performance improvements, noise filtering,
IC selection, signal converter, signal generator.
1. Introduction on the process capacity, time consumption might

Industrial Control Systems (ICS) is a increase exponentially. In order to decrease
general term covering various types of control complexity, a modular approach has been used. To
systems and associated instruments used in solve problem extra microprocessors has been used.
industrial production, process control, machinery Implementing microprocessor increased the speed
and critical infrastructures [1]. High capabilities and of device furthermore it implements several benefits
multiple functions of industrial controllers have on device as well such as well-known
caused widely usage and therefore have become a communication tools (RS232-485), local algorithms
crucial part of the industry. As systems became which can embedded on its registers or OLED
more and more complex, demands of more screens for notify users the status of process and
advanced controller has gained momentum [2]. device [4].
This system provides more flexible and In the following level several components
effective measurement solutions to industry this and their purpose has been explained. Overall
utilizes the efficiency of overall process. System has system requirements process levels of methods
efficient processing power and connectivity explained.
capabilities on the industrial computers thus
2. Design Methodology
systems. In this decade the industry which has
depended on electronic design has becoming more From the analysis of literature review, most
advanced because of unique designs products. of conversion systems has not capable of fulfilling
Because of these the requirement of conversion the system requirements of industrial processes.
systems increased in terms of flexibility, accuracy Most of the systems are not supply current
and reliability. The previous version of the technologies demand. Therefore, our design offers
conversion systems has seen that because of their low cost, low power consumption, portability, high
design issues the data accuracy and capacity of speed and high accuracy in terms of both academic
system is quite low. perspective and industrial demands. Our design is a
To obtain better performance over the compact size device. It uses 32 bit microprocessor
industrial systems several methods has been which has the speed of up to 48 MHz Beside of this
researched. In early 2000’s technology a central 256 Kb EEPROM. 16 Mb Internal RAM. It includes
process unit has to be connected to the all of the 4 channel 24 Bits A/D converters for connecting
sensors and devices via reading or transmitting several sensors. For dominating the sensors and
voltage [3]. The experiments showed that regarding actuators it includes 16 bit D/A converter with both
Copyright  by Technical University - Sofia, Plovdiv branch, Bulgaria
II-25
current and voltage options. Device has capable of In this decade the design of the Analog
supplying all industry demands such as variety of input/output modules has been changed. All of the
voltages (+-10V, +-5V) and for current (0-20, 4-20, analog/output modules separated with the host
4-24 etc.). microprocessors via some separators and
controllers. There were been several methods for
2.1 Central Process Unit (CPU) Selection separating the modules from the processor. In this
The demand of the industrial projects has design we used galvanic isolation because of
high microprocessors speed. Due to the natural several advantages over the other well-known
environment of microprocessors this issue solved separation methods [7]. The Digital to analog
with high clock cycle, long resolution bit and smart converter chip has specialized for industrial
algorithms which can able to solve big cluster of projects. Selected chip especially capable of
data packets and communication protocols providing to 20 mA, 0 to 20mA or 0 to 24 mA
simultaneously [5]. Almost all of the industrial current outputs or 0-5 V, 0-10 V, ±5 V or ±10
processes count as nonstop processes generally the voltage outputs with a10% false supply accuracy
systems work on 24 hours and 7 days because of range (0-5.5 V, 0- 11 V, ±5.5 V, or ±11 V). Because
this the selected microcontroller of the systems of these properties in this design our team select TI-
should consume less energy from that perspective DAC8760 model IC chip.
energy consumption is equal enough with
processing speed. In our design we used one well 2.2.2. Voltage Output
known microcontrollers company’s products. The DAC chip has property of
To accomplish complex industrial projects a configurability. When chip is configured for voltage
smart communication and sensor analyze algorithm output, max resistivity is 1 K Ohm at 10 mA. The
required. The easiest way to solve this problem is voltage level of output the modules output can
using a fast, powerful microprocessor. In this extend up from +15V to -15V. The voltage source
design ARM Core based microprocessor has been of the chip is supplied by power rail; therefore, the
used. ARM based microprocessors have supreme rail must be at least +- 10 V supply range in order to
flexibility peripherals and cortex layers which supply the IC chips.
increase our systems compatibility. The required
microprocessor has to have high frequency which is 2.3 Digital to Analog Converter (DAC)
up to 40 MHz or higher in order to achieve As known as to accomplish an industrial
acceptable computation speed. To balance power project the designer of systems needs to control
consumption of the design, a low power consumed several sensors and peripherals. Besides of shut
microprocessor required. In order to achieve of/on devices any peripherals or sensors requires a
process in required power level and speed the well- wide range of input streams such as 0-10 V or 4-20
known company the ST is selected. Consider to mA or 0-20mA or so. To accomplish these
GPIO requirements of the system 48 pin requirements a high precision digital to analog
STM32F0xxx microprocessors series is selected. converter (DAC) has required. The DAC has to be
work with either 0-10 V or 4-20mA or both. To
2.2 Analog to Digital Converter (ADC)
accomplish this standards 16 bit dual (both V and
Selection and Implementation
mA output) integrated chip has been selected [8].
Due to the accomplished a successful The DAC chip has 16 bit resolution. This
industrial project sensor communication required. range has been chosen because of the requirement
Industrial devices, peripherals have different level of high level integration. The IC chip has max 0.1
of scale however each of which most commonly full scale range (FSR). This range includes offset
uses well known standards such as 4-20 mA or 0-10 error, gain error, calibration error, scale error etc.
V output. This design has capability of both 4-20 This FSR scale range is acceptable for all of the
mA standards and 0-10 V standards. Design has voltage and current supply range. The maximum
internal jumper sets each of which for making nonlinearity is +-1 V or +-2 mA regarding to our
decision of whether current or voltage uses. experiments in real-time environment. The IC chip
Receiving analog input requires a high precession offers a programmable interface for determining the
and speed. To solve this, issue a four channel -24 bit voltage and current ranges, with using calibration
delta sigma analog to digital converter has been register user able to program the gain error and zero
used. The converter can able to work with 50Hz or error and other calibration parameters. The slew rate
60 Hz analog signals. The noise resistance of of output is also programmable. All of the
integrated circuits has up to 600nB RMS [6]. communication between The IC chip and host
processor have achieved through SPI interface. SPI
2.2.1. Isolation
is a well-known fast communication protocol which
II-26
is increased the speed of the overall design. For levels. 7 tests have been made with this device 3 of
each DAC module one microprocessor GPIO has them for voltage and current test one of them for
been assigned and with full-duplex SPI temperature test. The first test has been made 100
communication protocol all of devices programmed ms period of time. In between two following 100
separately. The calibration parameters and design ms time period the microcontroller adjusted to send
information have stored in an EEPROM. The a new voltage level starting from -10 V to +10 V
storage unit also uses the SPI buss therefore the
and the devices output level measured by voltage
communication speed is sufficient for obtaining all
source. As seen in Figure 1 the integrated chip
of the information from converter peripherals [9].
. In this system there has 8 blocks which cannot reach the speed level of microprocessor
represent the most important functions. The continuously therefore the level of voltage cannot
sequence of device starts with the input signal from increase sequentially. We observed that the voltage
sensors. The input determined from user side at level increased like a saw-tooth wave.
early system execution. The user could either 12
control the analog output by DAC or receive 10
information from analog input by ADC peripheral. 8

100 ms Period Voltage
Test over 100 comment
All of the communication ways handled through 6
serial communication line [10]. Both of the

information of analog input and output peripherals
4
Voltage [V]
values stored at micro controller. After each 2
sequence the memory part which contains the 0
information of devices peripheral is stored on the -2
EEPROM for providing secure information and -4
safety to user in case of power failure. The CPU -6

handles all of these events with using high clock
frequency which is up to 42MHz. Microcontroller -8
0 2
0 4
0 6
0 8
0 10
0 12
0
Time [Sec.]
gather input data form ADC and generate output
data which has the same parameter with input data Fig. 1. Accuracy test +-10V level 100 ms Period
to DAC. Thanks to that the input data remaining for 100 samples
scaled and secured without losing information.
After these processes all of the information sends to Second test has been made with 250 ms
viewing module display for showing the final time interval. To measure accuracy of the device the
results. time interval increased by 150 ms. The results get
better however instead of overall linear graph we
3. Device Working Principle
have been seen that ripples between sequential
Designed to comply with IEC61000-4 voltages. As shown in figure two the voltage
standards become more stable than previous test however we
•Selectable Voltage Output ranges from± 10V, 0 can clearly assume that the time period is not long
-10V, ± 5V, 0 -5V with output enough. Third test has been made with 500 time
•Selectable Current Outputs ranges from: 0 - 20 interval. To get optimum result 500 ms have been
mA, 4 -20 mA, 0 -24mA selected. The results are get optimum as expected.
•Output filtering, protection up to 15kV for As seen in figure 3 the voltage and the time period
Electrostatic Discharge protection gets completely linear.
•SPI interface applied with isolation which has
communication speed up to 20MHz.
•High temperature protection, short circuit
protection.
4. Test Results
The chip has capability of working the view
of ADC on integrated circuit on 2.7 to 5 V periods.
The device has been tested interns of current and
voltage level and the temperature over the voltage
and/or current consumption.
All of the accuracy and speed test have made with
100 sample rate with different voltage/current

II-27
20
10
8 18
250 ms Period
6 16
Voltage Test over 100 samples 250 ms Period Ampere
Test over 100 samples
4 14
Current [mA]
Voltage [V]
2
12
0
10
-2
8
-4
6
-6
4
0 5
0 10
0 15
0 20
0 25
0 30
0
-8
0 5
0 10
0 15
0 20
0 25
0 30
0 Time [sec.]
Time [sec.]
Fig. 5. Accuracy test 4-20 mA level 250 ms Period
Fig. 2. Accuracy test +-10V level 250 ms Period for for 100 samples
100 samples As seen in figure 4 we obtained a nonlinear
10 graph in our first 30 trial we observed almost
8 correct output however when the current level
6
500 ms Period Voltage increases the chip cannot reach the level of
microcontroller and some additional delays
occurred. Second test has been made with 250 ms
4
time interval. As it seen with the voltage test the

Voltage [V]
0 result becomes sharper however as seen in figure 5

-2 additional ripples are still disrupt the overall current
-4
signals.
Third test has been made with 500 time
interval. As seen in previous test set we observed
-6
-8
0 10
0 20
0 30
0 40
0 50
0 60
0 that for the optimum result the time interval must be
Time [sec.]
higher than 250 in order to compere between two
Fig. 3. Accuracy test +-10V level 500 ms Period for
test sets the level adjusted to 500 ms time interval.
100 samples As seen in figure 6 after adjusting the right time
After linearization of voltage test the second interval the current level become completely linear.
phase began. In second phase the level of current 24
accuracy has been tested in the same condition with 22
voltage tests. The first test has been made 100 ms 20
period of time. In between two following 100 ms 18

500 ms Period Ampere
time period the microcontroller adjusted to send a Test over 100 samples
Current [mA]
new current level starting from 4 mA to 20mA and

16
the devices output level measured by current source.

14
12
20
10
18 8
100 ms Period Ampere 6

16
4
0 10
0 20
0 30
0 40
0 50
0 60
0
14
Time [sec.]
Current [mA]
12
Fig. 6. Accuracy test 4-20 mA level 50 ms Period

10
for 100 samples
8
After finishing the accuracy test a
6 temperature test has been started. The device
4
applied with various level of current and voltage
0 2
0 4
0 6
0
Time [sec.]
8
0 10
0 12
0
output and the temperature of chip measured
externally. As seen in the figure 7 the level of
Fig. 4. Accuracy test 4-20 mA level 100 ms Period temperature increase regarding to level of output up
for 100 samples to a point. We observed that the device becomes
stable after level of the current or voltage has been
II-28
produced by the chips. The stabilization may cost
because of the space between the chips on PCB as
shown figure 8. The vain spaces could absorb the
temperature of device which can reach over a
certain range
70
65
60
Temperature over 16 bit Code
sample(0x0000-0xFFFF)
Temperature [C]
55
Fig. 9. Module simulation overview

50
45 REFERENCES
40
1. Hosny, A.A., Mohammed, H.A. (2015). On the
35
Adoption of Multi-Agent Systems for the
30
0 10
00 20
00 30
00 40
00 50
00 60
00 70
00 Development of Industrial Control Networks:
Code value[Hex.]
A Case Study, ICAS 2015, Rome, Italy.
2. Balasubramanian, S., Brennan, R.W., Douglas,
Fig. 7. Temperature test 20 samples with the range
H., Norrie, H. (2001) An architecture for
of 16 bit data from 0 to 65535
metamorphic control of holonic manufacturing
systems, Computers in Industry, vol.46, no.1.
pp.13-31.
3. Al-Zobi, Q., Al-Tawil, I., Gharaibeh, K., Al-
Kofahi, I.S. (2008). Design of Power-Line
Communication System (PLC) Using a PIC
Microcontroller, Journal of Active and Passive
Electronic Devices, vol. 3, pp. 331–340.
4. Almeida, E.E., Luntz, J.E., Tilbury, D.M.
(2007). Event-Condition-Action Systems for
Reconfigurable Logic Control, IEEE
Transactions on Automation Science and
Engineering, vol.4, no. 2. Pp.167-180.
5. Myers, G.J., Yu, A.Y.C. House, D.L. (1987).
Microprocessor Technology Trends”
Fig. 8. the DAC on PCB frontal seen Microelectronics Reliability, Proceedings of
the IEEE, vol.74, no.12, pp.1605-1622.
5. Conclusion 6. Walden, R.H. (1999). Analog-to-digital
In this paper an industrial sensor peripheral converter survey and analysis, IEEE Journal
reader controller has been explained. The device on Selected Areas in Communication, vol.17,
currently in is working on with several different no.4, pp.539-550.
applications in throughout several different places 7. Inoue, S., Akagi, H. (2007). A Bidirectional
and facilities. The combination between Isolated DC–DC Converter as a Core Circuit of
DAC&ADC enhanced the device usage thus several the Next-Generation Medium-Voltage Power
industrial processes has speeded up and get some Conversion System, IEEE Transactions on
extra qualifications such as combinations of analog Power Electronics, vol.22, no.2, pp.535-542.
input outputs, smart control over the process thanks 8. Ching-Wen Hsue, Chih-Jen Lin (1993). Built-
to microprocessor etc. Improvement of smart in Current Sensor for IDDQ Test in CMOS,
devices will increase the process speed and reduce IEEE International Test Conference, pp.635-
the errors. 641.
The IC chip has capability of producing 9. Leens F. (2009). An Introduction to I2C and
vide range of voltage and currents. The IC can also SPI Protocols, IEEE Instrumentation &
be programmed by designer in order to change the Measurement Magazine, pp. 8-13.
level of voltage and current. The general design of 10. Machacek, J., Drapela, J. (2008). Control of
device can be seen in Figure 9. Serial Port (RS-232) Communications in
LabVIEV, Modern Technique and

II-29
Technologies MTT 2008 Conference, pp.36-
40.
Authors’ contacts
Cagri Candan,
Istanbul Technical University, Department
of Electrical Engineering, Istanbul-Turkey.
E-mail: cagricandan91@gmail.com
Emre Can Dikmen,

E-mail: emrecan@uestco.com
Tahir Cetin Akinci,

E-mail: akincitc@itu.edu.tr
Serhat Seker,
E-mail: sekers@itu.edu.tr
INDUSTRIAL NETWORK DESIGN USING LOW

ENERGY PROTOCOLS
STOITCHO PENKOV, ALBENA TANEVA, MICHAIL PETROV, VASIL KALKOV
Abstract: In this work a network development for industrial application is presented.

Solutions related to the network protocols and standards for industrial applications are
summarized. The main goal is focused on net operation by combining low energy protocol
and transmission method. In the developed network project for industrial data exchange was
use MQTT protocol and LoRaWAN. An application for security purpose is obtained. The
real test and verification with sending and receiving data between the connected nodes are
made. The advantages of the developed network with MQTT and LoRa are given.
Key words: Industrial Network, MQTT protocol, LoRa
1. Introduction This paper is focused on configuration and

Using network in security applications is development of the networked sensors for
usual, complex and sometimes dangerous tasks. monitoring and control the environment and
There are many different ways to solve, depending security system. Many articles show variety of
of cases. An example a security in places without protocols and communication networks [2, 3, 4, 5].
electricity in order to be organized, is not easy task. Generally efforts are focused on problems with
How to secure buildings or value staff against tiffs network possibility and delays, packet dropouts,
and there no electricity? These tasks over a long address channel limitations related to the packet-
period of time are performed. It is case of highly rates, but there is no universal solution. In the paper
dynamic phenomena, such as change of security problems definition, development suggestions and
status because of fire alarm or intruder, means that maintenance for networked sensors are made. This
the results do not reflect the real value. A solution work summarizes solutions related to the network
of such problem is to use low cost nodes equipped protocols and standards for predefined application.
with relevant networked sensors for data collection. Proposition of node based sensor network with
Several nodes can be organized and formed a grid specified low energy protocols are also presented.
which will bring more complexity, and we will
gather all needed information. If the goal is to
2. Layer protocols
secure object, or to gather info under the sky, it
could be use nodes with GPS to know exact Regarding to the OSI model is important to
location, but considering power plan that is not discuss the way of data transmission through layers.
energy efficient. Where is needed can be used In this point of view the transmission method will
mobiles and flying robots for deploying these nodes, be summarized.
and at moment when nodes are deployed, it can be 2.1 LoRa™ is a proprietary spread
marked GPS position. Thus are covered large areas spectrum modulation scheme that is derivative of
and different surfaces. One of the main factors for Chirp Spread Spectrum modulation (CSS). It trades
such system development is the implementation data rate for sensitivity within a fixed channel
cost. In order to achieve it is appropriate to use bandwidth. It implements a variable data rate
small, low-power nodes, which are controlled and utilizing orthogonal spreading factors, which allows
supervised by a main network grid organized by the system designer to trade data rate for range or
LoRaWAN. power. Furthermore the network performance is
The paper presents a WSN extension to optimized in a constant bandwidth. LoRa™ is a
provide communication between the nodes and the PHY layer implementation and is agnostic with to
gateway over LoRaWAN, instead of IEEE 802.11 higher-layer implementations. This allows LoRa™
(WiFi), instead of 3,4G. to coexist and interoperate with existing network
architectures. This application note explains some
of the basic concepts of LoRa™. Modulation and
II-31
the advantages of the scheme can provide when modern computer systems use a data exchange
deploying both fixed and mobile low-power real- model referred to as Request-Response. However,
world communications networks [6]. when you try to use such a model for data exchange
in the sensor network you can encounter some
` difficulties. A possible solution is using the Publish-
Subscribe method. In this method, the data
publishing modules send it to a server called the
broker, which then sends the data to clients
subscribed to certain information. Using these
methods of data exchange allows the clients to
receive not all the information sent by the node, but
only the data that interest. There is also no need to
constantly calling the modules that generate
information about the data.
2.3 MQTT protocol
Among different options described one of
the most appropriate is MQTT protocol (Message
Queue Telemetry Transport), details in [5]. It was
Fig. 1. Representation of the relationship between
designed in 1999 for transferring data from
Range and bandwidth of the signal
telemetry devices. The main goal of the designers
Fig.1 presents relations between signal was to create an efficient protocol to transfer data
range for successful connection establishment and from devices with limited hardware resources,
frequency bandwidth. In case of frequency which is equipped with a low-performance
decreasing regarding to the modulation the signal microprocessors and a small amount of memory.
path increases with LoRa. Also expected to work in networks with severely
limited bandwidth for data transmission. The
2.2 LoRaWAN™ /layer2/ is a Low Power protocol uses a publish-subscribe method and
Wide Area Network (LPWAN) specification transmits the data over TCP/IP or UDP. In its
intended for wireless battery operated Things in a implementation requires a special computer called a
regional, national or global network LoRaWAN messages broker. The task of the broker is to collect
would correspond to the Media access control (MAC) messages and sending them to devices interested in
layer. LoRaWAN targets key requirements of specific information. Fig. 2 shows the organization
Internet of Things such as secure bi-directional diagram for data exchanging between Publishers
communication, mobility and localization services. and Subscribers by MQTT Broker.
It specification provides seamless interoperability
among smart Things without the need of complex
local installations and gives back the freedom to the MQTT Broker Subscriber 1
Publisher 1
user, developer, businesses enabling the roll out of Topic 1
Internet of Things. Subscriber 2
The choice of transport protocol, when Topic 2
internet connectivity is need basically, is reduced to Publisher 2
Subscriber 3
two options TCP and UDP. The Protocols allow
multiple devices to communicate effectively using
the Internet. However, the determination way for Fig. 2. Organization of data exchange in MQTT.
different data types, how are divided and stored in
frames, are required. In case of a system design for MQTT protocol messages are assigned to
collecting data in order to reduce the workload, names that are topics. In context of the client and
related to the data exchange organization, it is the broker, there is no need to configure the topic.
possible to use application-layer protocols. There is The client sends a message to a specific topic. If
variety of options. One is to use industrial there is a particular topic the broker will update its
automation protocols. In problem definition arises data, in the absence a new topic will be created
high cost of implementation. automatically, to which will be assigned the
information transmitted in the message. Topics may
By analyzing network protocols, extra be organized in a hierarchical manner using the
attention should be pay to the model they use for separator in the form of a forward slash (/). This
data exchange. Many of the technologies used in allows us to organize data in a broker in a manner
II-32
similar to the file system. Example topic for
networked grid nodes may have the following form:
Network22/NODE11/sensor33/DATA
An important feature of the MQTT protocol
is the ability to manage the quality of service by
implementing QoS (Quality of Service). It allows
you to manage the way to deliver a message and
confirmation of its receipt.
3. System design and configuration
The developed solution for industrial
network combines LoRa and MQTT protocol.
An important step in designing
communication layer using MQTT protocol is to
Fig. 4. Designed and Assembled Gateway included
determine the structure of topics and related
in LPWAN
messages. In current system only test messages were
send, which allows reading data from nodes sensors. The usage of these bands is submitted to
Therefore is achieved LPWAN based limitations: The output power (EIRP) of the
LoRaWAN, presented on fig.3. It is evident there is transmitter shall not exceed 14dBm or 25mW, and
variety of tasks: many sensor types, devices through the duty cycle imposed in Europe by ETSI is
Gateway to the many user defined applications. limited to 1% (for devices) or 10% (for gateways)
depending on the used sub-band. In this project the
Gateway is placed on the laboratory roof. As an
option in a startup it was assumed. The achieved
signal coverage was good enough. In future work it
will be improved. On the next fig. 5 are shown the
used and connected LoRa nodes based on
Microchip ®. In this way is developed and obtained
network based on LoRAWAN.
Fig. 3. Overview of the developed network with

industrial applications
The network can consist of thousands nodes.

In case of new LoRaWAL project is need a
Gateway in case with no LoRa coverage. The
proposed solution includes Gateway. This can be
viewed as an advantage of the application, fig.4.
LoRaWAN uses licence-free spectrum, usually ISM Fig. 5. Numbers of the used nodes based on
(Industrial, Scientific, Medical) bands to Microchip in the developed network.
communicate over the air. In Europe, ETSI The architecture including three different
regulates the ISM band access on the 868MHz and classes (A,B,C), of communication profiles are
433MHz bands. available in LoRa networks between devices and
applications, are presented on fig.6.

II-33
small nodes to collect data. Among the standards
for devices with low power the most interesting
solution is using LoRaWAN network
IEEE standards are for the physical layer of
the OSI model. They allow transferring data. But do
not provide a convenient way to control
communications and meeting the requirements of
QoS (Quality of Service). Accordingly, the next
step is the selection of a transport layer protocol.
Fig. 6. Typical system architecture of a LoRaWAN
Level2 Sections
node
For second level of headings use
„SectionL2” mark-up style.
Level3 Sections
For second level of headings use
„SectionL3” mark-up style.
4. Application in security area

There are many places around cities (levers,
fields, car morgues, etc.) without electricity and
220V power supply. From security point of view
Fig. 7. Description of A,B and C profiles this is a big problem and complex task. To establish
and maintain the communication with such plant.
Each class serves different application The options are very less. Generally is used security
needs and has optimised requirements for specific system equipped with transmitters and detectors
purposes. The key difference between A,B and C using 220V power supply. They are not applicable
profiles is the trade-off made between latency and in such case.
power consumption, fig.7. It is a difficult task to It is possible to use GSM operators. We are
gather information from many and different points. facing with need to use special equipment with
In a stage of problem definition the task is related to batteries. These batteries should therefore be
a network design and configuration of nodes charged. Even in this way the operability depends
equipped with sensors. The proposed solution is of the power supply. The lifecycle is not so long.
based on a grid of networked nodes. If there is a lot This weakness is other come with the developed
of equipment, it is important to find a low-cost solution combining LoRaWAN and MQTT. In this
solution with low energy consumption. In order to way can be achieved promising security without
reduce costs it is necessary to minimize the tasks of 220V power supply.
networked nodes. Therefore cable connections are
not relevant or suitable. In such case of many
distributed nodes cooperation, the option is to use
radio communication. The solution performs the
data transmission from the sensors to the gateway
/one or many/. This allows using of nodes with low-
cost microcontrollers. The power consumption.
Easy operation and reconfiguration of the system
are also given and discussed in [6].
Nowadays there are many radio
communication standards, but not all are well suited
to the task. The ability and widely used popular, not
expensive modules with easy connection to the
Internet is very important.
First one solution that comes to mind is the
standard IEEE 802.11 (WiFi). It allows connecting
multiple devices to the network, and easy
integration with the Internet. However, it has the
disadvantage of relatively high demand for energy,
Fig. 8. The coverage measurement with TTN
which is particularly important in the case of using
mapper
II-34
.LoRa uses low energy hence the workability is networked sensors for robots to perform various
expected to be 5-10 years. To one node could be tasks. To industrial environment for monitoring
connected more than one detector and passive detectors process values like temperature, air conditioning
as well. systems and etc. Each “node” could present sensor
The radio coverage of developed network, one or many, or robot sensors, connected the main
fig.8 can be viewed as an industrial solution in case node through the Gatrway with LoRa. It is possible
with IoT implementation. The signal covers the to use robots to spread sensors in case of hostile
laboratory location and connected nodes. To environment.
investigate and conduct the preliminary experiments The paper shows an example of the use of
is used software named TTN /The Things Network/ open network protocols in networked mobile robot
mapper, fig.9. The signal coverage and connected sensor system. MQTT protocol allowed for the use
node was monitored. The results demonstrate the of standardized methods for data exchange in the
good enough signals. It shows the achieved passing sensor network. Also greatly simplified the
of the signal at a large distance with low power. integration of new nodes, the use of nodes
ТХ- is the emitted power by the information from other systems and integrating with
transmitter. The signal is gained by the Gateway. the Internet.
The signal above the dashed red line confirms the The use of open protocols simplifies
established and “good” connection. software development work, especially when it
consists of a large number of independent nodes.
Also simplifies the maintenance process, since it is
possible to read information about the industrial
values and the status of individual sensors without
having to use special tools. Future work will be
focused on industrial network implementation.
REFERENCES
1. Souza R., L. Agostinho, F.Teixeira,

D.Rodrigues, L. Olivi, E. Guimaraes,
E.Cardozo. Control of Mobile Robots
Through Wireless Sensor Networks. XXIX
Simpósio Brasileiro de Redes de
Computadores e Sistemas Distribuídos,
pp.805-818.
Fig. 9. The TX and RX signals 2. GreenPeak Technologies (2015), Wireless
Communication Standards for the Internet of
Things, White paper
5. Conclusion 3. Silicon Laboratories, Inc., The Evolution of
This work is devoted on developing, Wireless Sensor Networks, 2015.
configuration and investigation of the network for 4. ITU-T, Ubiquitous Sensor Networks (USN),
security purpose in the case of missing 220V supply. ITU-T Technology Watch Briefing Report
A solution is found out by combination between the Series, No. 4, February 2008.
LoRaWAN network and MQTT protocol. All 5. MQTT V3.1 Protocol Specification.
developed nodes covered IP 67 standard. International Business Machines Corporation,
Some advantages of the system are: easy Eurotech, IBM, 2010
maintenance; low cost; low energy consumption; 6. Beser, Nurettin Burcak. "Operating cable
reliability and security; easy connection to the modems in a low power mode." U.S. Patent
Internet, long lifecycle. The system nodes (the grid) No. 7,389,528. 17 June 2008. – тва не го знам
can be enlarged with up to more than 1000 nodes какво е
connected with one gate. 7. SEMTECH, AN1200.22 LoRa™ Modulation
In the particular test application for the Basics
industrial network development it is obtained 8. LoRa Device Developer Guide, Orange
promising results. There are eight connected nodes Connected Objects & Partnerships
in this case study. Furthermore it will be possible
and very easy to extend the node numbers if
necessary. The idea could be associated with
II-35
ACKNOWLEGEMENTS Department of Control Systems
The authors would like to acknowledge Ministry of Technical University–Sofia, Branch Plovdiv
Education and Science of Bulgaria, Research Fund 25 Tsanko Diustabanov St.
Project No: DFNI-I 02/6/2014. 4000 Plovdiv
Phone: +359 032 659 585

E-mail: stoitcho@abv.bg
E-mail: altaneva@tu-plovdiv.bg
E-mail: mpetrov@tu-plovdiv.bg
E-mail: vassilkalkov@gmail.com
MACHINE VISION SYSTEMS FOR

INTELLIGENT QUALITY CONTROL OF
MANUFACTURING PROCESSES
BORISLAV RUSENOV, ALBENA TANEVA, IVAN GANCHEV, MICHAIL PETROV
Abstract: An intelligent system for automated quality control of manufacturing process

applications, based on machine vision is presented in this paper. The quality of many
produced parts in manufacturing processes depends on dimensions and surface features.
The presented automated machine vision system analyzes those geometric and surface
features and decides about tile quality by utilizing statistical analysis. Refined methods for
geometric and surface features extraction are presented also. The efficiency of processing
algorithms and the usage of an advanced analysis as a substitution of human visual quality
control are investigated and confirmed.
Key words: quality control, machine vision, intelligent systems
Introduction becomes one of the weakest and unreliable links.

By replacing the human with machine, the whole
process should have better production yield and
Many industrial processes use or require could be more efficient [3].
visual inspection in quality control as an integrated
Industrial control system (ICS) is a general
part of their production stages. Such processes are
term that encompasses several types of control
based on visual perception principles to successfully
systems used in industrial production, including
determine levels of product quality by quantifying
supervisory control and data acquisition (SCADA)
its visual appearance in general and some specific
systems, distributed control systems (DCS), and
visual features, respectively [1]. A visual inspection
other smaller control system configurations such as
system is based on machine vision principles by
programmable logic controllers (PLC) often found
using acquisition cameras and also, one or more
in the industrial sectors and critical infrastructures.
industrial computers. The main motivation for
Industrial automation is a discipline that includes
machine vision implementation is economic factors,
knowledge and expertise from various branches of
which constantly require less production costs.
engineering including electrical, electronics,
One of processes that use machine vision chemical, mechanical, communications and more
for product quality control is the production in recently computer and software engineering.
mechanical manufacturing processes [2]. The
In order to stay on top of a competitive
production phases are more or less automated. The
market, companies have to keep their production
exception is quality control stage with mostly
costs as low as possible. One element of their
human vision inspection. Some production lines
strategy is to collect production and control data,
still use human vision in quality control. The main
analyze it to find improvements, and incorporate
reason lies in complexity of this task. Human
those improvements in each new plant.
resources are used because the visual quality control
process is very complex and highly demanding and The role of including quality control in
often should be on-line adaptive on changeable other control systems may lie in control and
quality requests in classification stage of information flow of plants, in integrating processing
production. Because of human features limitations machines, in the Manufacturing Execution Systems
as controlling element in production line, man (MES) [7] that monitor the processes, and in the
II-37
data-based Enterprise Resource Planning (ERP) [8] characteristics. Process product quality testing is
system that provides decision support. achieved by testing products’ quality feature in the
machining processes or machining process interval.
Process quality control system is a novel Common processing measurement parameters for
computer-aided process quality control system, example could include cutting force, temperature,
which integrates hardware and software. The spindle motor current changes, vibration and noise
system could realize quality data collection, signals. Process quality control should establish the
transmission, storage, quality monitoring and correlation between process parameters and the
quality statistical analysis for spare parts production final product quality characteristics, and ensure the
process. It could accomplish the collection and quality of the final product by the adjustment to
monitoring of quality data automatically in field. parameters.
Once the production process has problems, it can
give an alarm and begin to analyze, providing a
basis for process quality control. Furthermore, it can
Role of SCADA in Quality Management
also carry out offline quality statistical analysis of
the quality data derived from the machining field, A production process includes the quality
guaranteeing after-process control of processing assurance testing of samples from each product lot.
quality. The test data is used to produce a certificate of
conformance report for such a lot. The data is
collected from test equipment, then sometimes
manually entered into a customized database form,
Quality Control System and then formatted to produce the certificate of
conformance. Manual data entry is time-consuming,
error-prone, and repetitive. Here is the challenge to
The complete system block diagram in
introduce a supervisory control system to automate
Figure 1 presents the role of the SCADA system;
that process and integrate data collection with its
the quality measurement data is collected from the
other manufacturing systems. With the automation
final product and stored in special registers inside
of the quality assurance process by electronically
the controller/ Remote Terminal Unit (RTU) which
collecting data from the measurement tools, the
in this case is a programmable logic controller
operator does not need to write test measurements
(PLC). This data is transferred to the SCADA node
into a form and then into a computerized
using industrial network which could be a local or
spreadsheet. Dozens of samples with up to hundreds
remote network, this data is analyzed and a control
of measurements are displayed; with manual work,
decision to tune the controller if necessary to ensure
the system can only handle a limited number of
that the product is within the bounds of required
measurements. In being automated, it can be
quality.
upgraded to manage much more. After data
collection, a system can retrieve the expected
measurement values from database for the samples
of that lot.
A summary screen might he immediately
displayed to the operator indicating, for each
sample, whether all measurements were within
control limits, within specification limits, or outside
specification limits. The operator can also view
details about each sample and adjust the data
manually. Any data modifications are stored in an
Quality control system block diagram audit trail, and the initial raw data is kept for
historical records. When the operator is satisfied
Process quality data acquisition and
that the data is correct, the system sends this
controlling contents include the monitoring of
validated data to its database, and it is possible to
parameters process product quality testing. Process
produce the certificate of conformance. Time saving
parameters monitoring is realized by measuring the
for these procedures is significant: it could easily
relevant parameters on product quality
II-38
take an operator longer to record the measurement protectors with another trigger temperature. In
data than it takes the tool to create it. A supervisory addition, there are thermo-protectors with missing
control system brings the time for the whole or incomplete alphanumeric codes. The task of
process, from measurement to a report, down to a identifying and separating the inappropriate
matter of seconds. incoming product from the incoming good one is
accomplished by the machine vision system shown
in Fig. 3.
Description of an application system.
A typical machine vision and quality

control system has the structure shown on fig.2.
Structure of a typical machine vison system
The system includes: camera, light sources,

a transport system for moving the inspected
products, control modules and actuators. Typically,
the transport system is a part of the production
process. The cameras are positioned so that the
inspected parts or finished article falls into the
camera's work area. In general, the control module
Factory machine vision system
serves to process the camera signal and to
communicate with the executive actuator. If it is The system of machine vision and quality
provided, it serves with the quality measurement control of components for temperature protection of
system and the SCADA system of the factory as compressors for refrigerators and freezers with
well, where the inspected product is manufactured. thermo-protectors is constructed schematically
The light source allows accurately determine the according to Fig. 4.
right amount of light flux to properly capture the
scene.
1.1. Basic machine vision system of modules
for temperature protection
In the manufacture of modules for
temperature protection of electric motors of
compressors in refrigerators and freezers a main
component which is embedded is a thermo-
protector. The operating temperature of these
thermal protectors is indicated by an alphanumeric
code. Thermo-protectors are produced in several Structure of machine vision and quality control
factories around the world. Upon entrance, each system
type of batch of thermo-protectors with a specified
trigger temperature is "contaminated" with thermo-
II-39
The smart camera model FQ [4] of the intervention to be processed, which means it can not
manufacturer for industrial automation company be included in the enterprise's SCADA system.
Omron was used. The equipment includes a visual
inspection camera, a setup controller with This problem is solved in the new
predefined different quality parameters, a light generation of intelligent camera model FQ2. This
source, a parallel interface for direct digital output camera already has had an EtherNet/IPTM interface.
control. A machine vision system with camera model FQ2 is
shown on Fig.5.
The controller of the visual inspection
camera allows operation in 5 different recognition
modes: in case of difference in form, difference in
position, difference in width, difference in size, and
color difference. For inspection of thermo-
protectors, the difference in form recognition mode
is used. The embedded digital outputs of the
intelligent camera are used to send signal to the
programmable logic controller PLC of the thermal
protection module manufacturing machine. When
an unsecured thermal protector is detected, the cycle
stops and the actuator removes the product.
Structure of machine vision system with camera

model FQ2
When using such a camera included in a

SCADA system it is no longer a problem. The high-
speed interface allows to be achieved the necessary
data exchange. Significant improvements are also in
recognition algorithms. This generation of
intelligent cameras have the high-tech Shape Search
III algorithm [3], [6]. It allows work with complex
objects in light interference conditions and poor
background. Several objects entering the camera's
The described intelligent camera system FQ work area can be inspected simultaneously, even if
can also be seen as an intelligent sensor. Unlike objects are poorly or partially illuminated or rotated
conventional sensors, through camera inspection, at and overlapped. It is possible to measure distances
every stage of the production cycle of a product the between the outline of an object, detection of
quality control could be achieved. Quality control defects such as cracks, for example. There is
consists of separating the unsuitable product. compensation for displacement and rotation of the
object. The ability to recognize text, different types
1.2. Smart machine vision system of of bar code and 2D code is built in. All of these
modules for temperature protection features make it possible to expand the range of
possible smart camera application areas. In addition
For quality management this is not enough.
to the enhanced recognition algorithms, a
In order to be able to derive full information about
significant difference is also noted in the camera
the quantity of inferior production which is evident
setup. The setting of inspection areas is dramatically
and what is deviation in relation to the set
simplified compared to the previous FQ equipment.
parameters, it is necessary to use more digital
outputs. Intelligent camera FQ has a datalog In the FQ equipment, as in previous models
function. This feature allows keeping a file with a of Omron's cameras, an underlying search algorithm
table of values for each criterion measurement. For based on image outlines (dense edge) is built. It can
a modern production system, this data exchange is easily be influenced by slight changes in image
not fast and effective enough. It requires human contours due to unevenness in conveyor systems,

II-40
slight geometric differences in the inspected targets,

all of which lead to a drop in measurement accuracy
and sometimes resulting in unexplained behavior of
the systems. One possible solution to this problem
is to be performed iteratively, comparing to multiple
templates, which of course leads to a drastic drop in
productivity.
Experimental results and quality analysis

A solution to the machine vision system
with camera model FQ2 is an algorithm using
variation-absorbing templates. This method
generates tens of thousands of possible variations
for each local region to absorb the changes caused
by these variations (Fig. 6)
Pareto chart of inspection of thermal protectors
Three or five most pronounced defects/

problems are ispected. Each of them is subjected to
thorough investigation using the Root cause
analysis (RCA) method. RCA is a technique used to
identify the conditions that initiate the occurrence of
an undesired activity or state. The process of
problem solving used to identify the underlying or
initiating source of a nonconformance. In order to
reach the cause of the defects, the so-called
"Fishbone” diagram is built and shown in Fig. 8.
In the major bones or contributing to the
The variation-absorbing templates method occurrence of defects are:
Man Power (Personnel)
Along with this, the generated patterns are Machines (Equipment)
clustered and pooled into groups based on their Materials (Reagents and Supplies)
similarity, resulting in a reduction in the Methods
consumption of memory from 1/100 to 1/1000
compared to the amount of the generated templates.
This does not lead to a drop in productivity and
performance, only leads to increased accuracy and
enables highly accurate recognition at high speeds.
Some results of the data represent the observations
of the different variations.
The results of the quality control of the described
thermo-protector product are systematized for
further processing. They are arranged in a table
describing in detail: the type of defect that has
occurred and number of defects of the type, change
in which the defect occurred, date of occurrence,
batch, operator, etc. To analyze the defects of Fishbone diagram
production, first a Pareto diagram is built. The
Pareto diagram for manufacturing of thermo-
protectors is shown in Fig.7.
II-41
Besides the described techniques for analysis The presented machine vision system has
of defects and measures for their removal, a quick purpose to replace a human vision quality controller
analysis of the condition of a defect also gives Time in the manufacture of modules for temperature
series plots. Time series plot can be used to protection of electric motors of compressors in
visualise things that vary over time to also present refrigerators and freezers. The presented system
in which order they occurred. A time series plot consists of at least two cameras for image
diagram can be constructed without advanced registration and one computer. After the image
statistical software. Another advantage is that it is acquisition and trimming basic optic parameters, the
easy to interpret without deeper statistic knowledge geometric and surface analysis is performed.
and gives the possibility to see trends and process Geometric analysis relies on contour tracing method
variation. A graph of the development over time of where several geometric inspection methods are
the most pronounced defect of a Pareto diagram of united into one. Using this method, the complete
Fig. 7 is shown on Fig.9. time for analysis is reduced to acceptable limits
suited for real time operations.
REFERENCES
[1] Batchelor B.G., Whealan P.F. (2002). Intelligent
vision systems for industry.
[2] Forsyth D.A., Ponce J. (2003)., Computer
vision. A modern approach. Whiliams
publishing house.
[3] Gonzalez R., Woods R. (2002)., Digital image
processing. Whiliams publishing house
[4]http://www.omron.com/technology/core/thinkAn
dSee/
Time series plots [5] Said Ibrahim Abu Al-Roos SCADA Online
Product Quality Control.(2013)
Thus, the accumulated and evaluated data
from the quality control of the thermoprotection [6] OMRON Corporation, Technology overview
component can be used not only to enhance and Pattern Matching Algorithm Shape search Ⅲ
permanently stabilize the quality. They are also to [7] Jürgen Kletti (Ed.) Manufacturing Execution
take preventative measures in servicing the Systems – MES. Springer-Verlag Berlin
machinery and equipment, involved in the Heidelberg 2007. p.271
production of the materials and elements involved n [8] Monk, Ellen; Wagner, Bret (2006). Concepts in
the final product. Enterprise Resource Planning (Second ed.).
Boston: Thomson Course Technology.
Conclusion
An intelligent system for automated quality Authors’ contacts
control of manufacturing process applications based
on machine vision is presented in this paper. The
Organization:
quality of many produced parts in manufacturing
Technical University Sofia, Plovdiv branch
processes depends on dimensions and surface
features. The presented automated machine vision Address: 25, Tsanko Dyustabanov Str.,
system analyzes those geometric and surface 4000 Plovdiv, Bulgaria
features and decides about tile quality by utilizing
E-mail: borislav.rusenov@abv.bg
statistical analysis. Refined methods for geometric
E-mail: altaneva@tu-plovdiv.bg
and surface features extraction are presented also.
E-mail: ganchev@tu-plovdiv.bg
The efficiency of processing algorithms and the
E-mail: mpetrov@tu-plovdiv.bg
usage of an advanced analysis as a substitution of
human visual quality control are investigated and
confirmed.

ПРОЕКТИРАНЕ НА АСИНХРОННО
ЕЛЕКТРОЗАДВИЖВАНЕ С
КВАЗИРЕЗОНАНСЕН ПРЕОБРАЗУВАТЕЛ
ДИМИТЪР СПИРОВ, НИКОЛАЙ КОМИТОВ
Резюме: Разработен е изчислителен алгоритъм за проектиране на асинхронно

електрозадвижване с квазирезонансен преобразувател и управление с широчинно-
импулсна модулация. Разгледана е работата на схемата по интервали и са
определени напрежението и тока в постояннотоковото звено при синусоидална и
простарнствено-векторна модулация. Представените сумулационни резултати
потвърждават работоспособността на предложената методика.
Ключови думи: асинхронно електрозадвижване, квазирезонансен преобразувател
DESIGN OF INDUCTION MOTOR DRIVE WITH

QUASI-RESONANT CONVERTER
DIMITAR SPIROV, NIKOLAY KOMITOV
Abstract: A methodology for design of induction motor drive with quasi-resonant

converter and pulse width modulation control is developed. It examined the operation of
the scheme in intervals and voltage and current in the DC link are calculated for sine-
PWM and space vector-PWM. The simulation results confirm the performance of the
proposed methodology.
Key words: induction motor drive, quasi-resonant converter
1. Основни положения притежават голяма точност, бързодействие,

Бързо намаляващите енергийни ресурси надеждност и устойчивост при стационарни и
поставят въпроса за тяхното по-рационално и динамични режими в целия обхват на
по-ефективно използване. Това довежда до регулиране [2, 3].
необходимост от разработване и внедряване на Повишаването на честотата на
енергоспестяващи производства. Като основен системите, захранвани от инвертор с
елемент на производствените системи се явява широчинно-импулсна модулация (ШИМ)
електрообзавеждането и електрозадвижването довежда до повишаване на загубите в силовите
им. Основните елементи на ключове и комутационни ударни натоварвания.
електрозадвижването са двигателите и Тези недостатъци могат да бъдат
системите им за управление. Асинхронните избегнати чрез използването на резонансни
електродвигатели са проектирани да работят с преобразуватели, осигуряващи намаляване на
най-висока ефективност при пълно натоварване, комутационните загуби за сметка на добавяне на
но на практика това рядко се случва [1]. пасивни, а в някои случаи и на активни
При това положение системите за компоненти. Извесни са различни видове
управление са тези, които трябва да осигурят резонансни преобразуватели, но основната идея
високоефективен режим при широк диапазон на при всеки от тях е, че превключването на
изменение на изходните параметри и на силовите ключове трябва да стане при нулево
смущаващите въздействия. Съвременните напрежение върху тях или на нулев ток през тях,
регулируеми електрозадвижвания трябва да в резултат на което се намаляват значително
комутационните загуби [2, 4].
II-43
Целта на настоящия доклад е тока io се определя от сигнала на ШИМ-
проектиране на асинхронно електрозадвижване генератора и от режима на работа на товара
с квазирезонансен преобразувател. Необходимо (асинхронната машина).
е да се изследва работоспособността на Ефективната стойност на основния
предложения алгоритъм. хармоник на изходното линейно напрежение на
инвертора при синусоидална ШИМ е:
2. Анализ на преобразувателя
Параметрите при проектирането на 1 3 3
преобразувателя и на системата за управление се UЛ  mU s  mU мр , (1)
2 2 2
определят от асинхронната машина и
задвижвания от нея механизъм, от техните А при пространствено-векторна ШИМ е:
режими на работа и товаров график. На фиг. 1. е
дадена схема на регулириемо асихронно 1
електрозадвижване с квазирезонансен UЛ  mU s  mU мр . (2)
преобразувател. 2
Модулирания синусоидален сигнал се

дефинира с изразите:
АД
Механичен u*мА  U m* sins t   u ;
Променливо- Асинхронна товар
токов източник Мостов Постояннотокова
на ел. енергия изправител квазирезонансна
Трифазен машина u*мB  U m* sins t   u  2 / 3; (3)
мостов
u*мC  U m* sins t   u  4 / 3,

и филтър верига инвертор
Фиг. 1. Регулириемо асихронно

електрозадвижване с квазирезонансен където:
преобразувател. ωs=2πfм – кръгова честота на
модулирания синусоидален сигнал;
Захранването се осъществява от Um* – амплитудна стойност на
променливотоковата мрежа. Входното напрежението в относителни единици:
синусоидално напрежение се изправя от Um*=Um/Umb, Umb=Umн;
диодния мостов изправител и се изглажда от ψu – начална фаза на напрежението.
филтровия кондензатор. Трифазния мостов Модулираните токове на трите фази се
инвертор осигурява трифазното напрежение за дефинират от изразите:
захранване на асинхронния двигател с
напрежение и честота, определени от режима на i*мА  S Ai*A  S A I m* sins t  ;
работа. За управлението на силовите ключове се i*мB  S B i*B  S B I m* sins t    2 / 3; (4)
използва генератор на широчинно модулиран
сигнал (ШИМ генератор). Амплитудата на i*мC  SC iC*  SC I m* sins t    4 / 3,
изходното напрежение се определя от
модулационния индекс m=Um/Uтр, където Uтр е където:
амплитудата на модулиращия трионообразен Im* – амплитудна стойност на фазовия
сигнал, а Um е амплитудата на модулирания ток в относителни единици: Im*=Im/Imb, Imb=Imн;
синусоидален сигнал. Честотата на изходното φ=ψu−ψi – фазова разлика.
напрежение се определя от честотата на Токът в постояннотоковото звено io се
модулирания синусоидален сигнал fм. определя от израза [4]:
ШИМ-генератора формира входния
управляващ сигнал на системата за управление с io  S Ai A  S B iB  SC iC , (5)
честота fpwm. Честотата на превключване на
резонансната верига се определя от броя рамена където SА, SB и SC са състоянията на ключовете
на инверторa, за три фази се получава fc=3fpwm. S1, S3 и S5, определени от ШИМ генератора, а iA,
Периода на превключване на резонансната iB и iC са моментните стойности на тока в трите
верига е Tc=1/fc. фази на двигателя. Ефективната стойност на
Режимът на работа на тока в постояннотоковото звено може да бъде
постояннотоковото квазирезонансно звено изчислена по израза за синусоидална ШИМ [5]:
зависи от входното постояннотоково
напрежение Us и от товарния ток в
постояннотоковото звено io. Формирането на Io  Im
m 3
4

1  4 cos  2 ,  (6)
II-44
където Im е максималната стойност на фазовия
ток на двигателя, а cosφ е фактора на
мощността.
Постояннотоковото квазирезонансно
звено осигурява нулиране на постоянното
напрежение за превключване на силовите
ключове на трифазния инвертор при нулево
напрежение. При паралелните схеми това звено
се състои от последователен ключ Sa1 с обратен
диод Da1, един или няколко паралелни ключа и Фиг. 3. Зависимост на i*мA=f(t) при СШИМ
диода и пасивни елементи: резонансна бобина Lr
(резонансен трансформатор с Lr1 и Lr2) и
резонансен кондензатор Cr.
3. Получени резултати
Изследвана е работата на
преобразувателя при захранване на асинхронна
машина 4AO 80B-4D с номинални данни,
дадени в Приложение.
Захранването на схемата е трифазно с
Фиг. 4. Зависимост на u*мA, SA=f(t) при
Uмр=380V. Входното напрежение на
ПВШИМ
постояннотоковото звено се получава
Us=√2Uмр=537V. Номиналния режим на
асинхронната машина се получва при
захранване с UЛ=380V и fм=50Hz и номинален
товар. При синусоидална модулация
номиналното захранващо напрежение се
получва при модулационен индекс
m=2/√3≈1,155, определен по израз (1). Тъй като
в режим на надмодулация (m>1) ефективната
стойност на основния хармоник зависи
нелинейно от m [2], то точната стойност на m се Фиг. 5. Зависимост на u*мA, SA=f(t) при
опрделя симулационно m=1,395. При ПВШИМ
пространствено-векторна модулация
номиналното захранващо напрежение се
получва при модулационен индекс m=1
определен по израз (2).
Графиките на сигнала SА и модулираните
сигнали на напрежението u*мA и тока i*мA за фаза
А при cosφ=cosφH (φ=43,95°), за синусоидална
(СШИМ) и за пространствено-векторна
(ПВШИМ) модулация са показани съответно на
фиг. 2. - фиг. 5.
Фиг. 6. Зависимост на i*o=f(t) при СШИМ
SPWM
1 Токът в постояннотоковото звено io,
uMA*, SA определен по израз (5), при синусоидална и
uMA* пространствено-векторна ШИМ е представен на
0,5
фиг. 6 и фиг. 7.
t, s
0
0 0,01 0,02
Фиг. 2. Зависимост на u*мA, SA=f(t) при СШИМ

II-45
В Таблица 2.3 са систематизирани
резултатите от изчисленията на Io, Io1 и Io2 за
първия интервал от работата на схемата, като
превключтането става в момента t0.
Таблица 1. Резултати от
изчисленията
Io Io1 Io2
Модулация
Фиг. 7. Зависимост на i*o=f(t) при ПВШИМ A A A
СШИМ 2,182 1,061 3,17
Анализът на работата на ПВШИМ 2,217 1,073 3,172
преобразувателя за първия интервал при Tи=Tpwm
се прави на базата на графиките, представени на
фиг. 8 за синосуидална ШИМ и на фиг. 9 за 4. Заключение
пространствено-векторна ШИМ. Разработен е изчислителен алгоритъм за
проектиране на асинхронно електрозадвижване
с квазирезонансен преобразувател и управление
с широчинно-импулсна модулация. Разгледана е
работата на схемата по интервали и са
определени напрежението и тока в
постояннотоковото звено. Представените
сумулационни резултати потвърждават
работоспособността на предложената методика.
ЛИТЕРАТУРА
1. Chao, K., Liaw, C. (2001). Three-phase soft-

switching inverter for induction motor drives.
IEE Proc. Elect. Power Appl, Vol 148, No I,
Фиг. 8. Работа на преобразувателя за първия pp. 8-20.
интервал при СШИМ 2. Юдов, Д., и др. (1999). Повишаващ DC/DC
преобразувател с подобрени енергетични
показатели. Annual Journal of Electronics,
TU-Sofia, стр. 120-125.
3. Dochev M. (2016). A test study stand for an
adjustable asynchronous electric drive, ХХV
МНК „АДП 2016“, Созопол, стр.275-280.
4. Cho, J.G., Kim, H.S., and Cho, G.H. (1991).
Novel soft switching PWM converter using a
new parallel resonant dc-link, Proc. IEEE
PESC, pp. 241-247.
5. Implementing Space Vector Modulation with
the ADMCF32X, (2000). Analog Devices Inc.
ПРИЛОЖЕНИЕ
Фиг. 9. Работа на преобразувателя за първия Технически данни на асинхронен
интервал при ПВШИМ електродвигател тип 4AO-80B-4D: PN=0,75kW;
UN=380V; IN=2,25A; nN=1390min-1; ηN=72%;
Ефективната стойност на тока за x-тия cosφN=0,72; Jm=0,0013kgm2.
интервал Iox от работата на схемата може да бъде
определен с израза:
Димитър Спиров
t1 x Университет по хранителни технологии
1
 i t  dt . Адрес: гр. Пловдив, бул. Марица 26
2
I ox  o (7)
t1x  t0 x t0 x Телефон: 032/ 603 738
E-mail: dimitar_spirov@abv.bg
ОПРЕДЕЛЯНЕ НА ПАРАМЕТРИТЕ НА
КВАЗИРЕЗОНАНСЕН ПРЕОБРАЗУВАТЕЛ С
ДВА ПАРАЛЕЛНИ КЛЮЧА
ДИМИТЪР СПИРОВ, НИКОЛАЙ КОМИТОВ
Резюме: Разработен е алгоритъл за определяне на параметрите на квазирезонансен

преобразувател с една бобина и два паралелни ключа за захранване на асинхронна
машина. Получени са аналитични изрази за работа на схемата по интервали.
Формулирана е задача за оптимално определяне на параметрите на резонансните
елементи. Представеното елетрозадвижване е моделирано в средата на PSpice.
Симулационните резултати потвърждават достоверността на теоретичния
анализ.
Ключови думи: асинхронно електрозадвижване, квазирезонансен преобразувател
DETERMINATION OF THE PARAMETERS OF

QUASI RESONANT DC LINK CONVERTER
WITH TWO PARALLEL SWITCHES
DIMITAR SPIROV, NIKOLAY KOMITOV
Abstract: A parallel quasi resonant DC link converter for induction motor drive
application is analyzed and designed. The analytical equations and operating modes of the
presented inverter are explained in details. The proposed drive system is modeled and its
performance is simulated in PSpice. The design considerations are presented and the
simulation results verify the theoretical analysis.
Key words: induction motor drive, quasi-resonant converter
1. Основни положения комутационните загуби. Различните видове

При инверторните електрозадвижвания с резонансни преобразуватели си приличат по
„твърда комутация”, силовите прибори се това, че при всеки от тях превключването на
включват когато напрежението върху тях е силовите ключове трябва да стане при нулево
ненулево и се изключват когато токът през тях напрежение върху тях или на нулев ток през тях,
има ненулева стойност [1, 2]. Всяко в резултат на което се намаляват значително
превключване е свързано със загуба на енергия. комутационните загуби [1, 2, 3, 4].
Използването на инвертори с „твърда Инверторите с квазирезонансно
комутация” за захранване на двигатели води до: постояннотоково звено са получили най-широко
електромагнитни смущения към разпространение при трифазните схеми и в
електродвигателите и по цялата електрическа частност при асинхронните задвижвания,
верига; при използването на дълги захранващи поради редица причини [1]: възможност за
линии се получават пренапрежения върху „мека комутация” както при включване, така и
клемите на двигателя; скъсяване на живота на при изключване на всички прибори на схемата;
лагетите; появат на шум и вибрации и др. загубите в резонансните елементи извън времето
Тези недостатъци могат да бъдат за комутация са нулеви; товарът не участва в
избегнати чрез използването на резонансни резонансния процес и др.
преобразуватели, осигуряващи намаляване на
II-47
Целта на настоящия доклад е На фиг. 2 е представена работата на
разработване на алгоритъм за определяне на преобразувателя по интервали.
параметрите на квазирезонансен преобразувател От фигурата се вижда, че когато е
с една бобина и два паралелни ключа за необходимо превключване на някой от
захранване на асинхронна машина. Необходимо основните ключове на инвертора (например SA),
е да се изследва работоспособността на то се осъществява с известно закъснение tз=Δt1+
предложената методика. Δt2 след като напрежението в постояннотоковото
звено спадне до нула чрез сигнала S’A. Първо се
2. Анализ на преобразувателя включва резонансната бобина и нейния ток
Принципната схема на асинхронно започва да нараства до стойност IW. Тази
електрозадвижване с квазирезонансен стойност трябва да бъде достатъчно голяма, така
преобразувател с една бобина и два паралелни че запасената енергия в бобината да гарантира
ключа в постояннотоково звено е представена на резонансното зареждане на кондензатора.
фиг.1. От равенството на електромагнитната и
електростатичната енергия при пълно
преобразуване се получава:
1 2 1 2
Wем  Lr I Lr  C r U Cr (1)
2 2
От тук може да се определи граничната

стойност на тока IW, съответстващ на
необходимата енергия, запасена в бобината:
Фиг. 1. Принципна схема IW  U Cr Z r (2)
където Zr=√(Lr/Cr) – характеристичен импеданс

на резонансния кръг.
След заместване на изразите за UCr и Zr
се получава [1]:
2 2
U  U 
IW   s  I o1  I o 2    s   I o1 , (3)
 Zr   Zr 
където Io1 и Io2 са ефективните стойности на тока

в постояннотоковото звено в интервала преди и
след превключване на основните ключове.
Времето за зареждане на резонансната
бобина Δt1, времето за разреждане на
кондензатора Δt2 и общото време за преходния
Фиг. 2. Зависимости SA, S’A, Sa1, Sa2, uCr, iLr =f(t) процес на резоанасните елементи TW се
получават [1]:
Постояннотоковото квазирезонансно
звено съдържа три силови ключа Sа1, Sа2 и Sа3, Lr IW
три диода Dа1, Dа2 и Dа3 и резонансните бобина и t1  ;
Us
кондензатор Lr и Cr.
Работата на постояннотоковото 1  Us 
квазирезонансно звено може да бъде разделена t2  a tan ; (4)
на шест интервала (фиг. 2): установено
r  Z r IW  I o1  
състояние (до t0); включване на бобината (Δt1); TW  t1  t2 .
резонансно разреждане на кондензатора (Δt2);
нулево напрежение на постояннотоковото звено Входните изисквания при проектиране
(Δt3); резонансно зареждане на кондензатора на преобразувателите с квазирезонансно звено
(Δt4); връщане на енергия в захранващия са дадени в [1]:
източник (Δt5). - Входното напрежение на инвертора
трябва да бъде намалено до нула за
II-48
превключване при нулево напрежение и отново на TW се получват при минимални стойности на
да се повиши до напрежението на захранващия Lr и Cr, а минималната стойност на IW се
източник в рамките на резонансния цикъл; получава при при минимална стойност на Cr и
- Токът при превключването трябва да максимална стойност на Lr.
бъде възможно най-малък, за да се намалят Зададчата за оптималното определяне на
загубите в електрическата верига; параметрите на резоанасните елементи можем
- Пиковите стойности на резонансните да формулираме в следния вид: да се намерят
напрежение и ток трябва да се сведат до стойностите на Lr и Cr при които времето TW има
минимум, за да се намали претоварването на минимална стойност, а тока IW не надвишава
елементите във веригата; максималния установен ток на товара Io.max.
- Скоростта на нарастване и намаляване Уравнението на TW е целевата функция, а
на изходното напрежение на инвертора трябва уравнвнието на IW е ограничително условие.
да е ниска за електрозадвижвания със Следователно на функцията трябва да бъде
сравнително дълги кабели; намерен локален минимум. Целевата функция се
- Резонансният цикъл трябва да бъде задава с израз (6), а ограничителното условие се
много по-кратък от цикъла, определен от получава от израз (3) във вида: IW–kзIo.max=0,
носещата честота на инвертора. където kЗ=1,1 е коефициент на запаса. Като
второ ограничително условие се въвежда
изискването резонансния кондензатор да има
стойност не по-малка от реалната, която може да
се използва, т.е. не по-малка от Cr.min. За Cr.min се
приема стойността 1nF. Функцията на Лагранж
приема следния вид:
LLr ,Cr , 1   TW Lr ,Cr  

(8)
 1 IW Lr ,Cr   k З I o . max    2 Cr  Cr . min ,
където λ1 и λ2 са коефициенти.
Условията на Kuhn-Tucker се поучават с
изразите:
Фиг. 3. Зависимости на TW=f(Lr) и TW=f(Cr) LLr ,Cr ,1 , 2 

 0;
Lr
LLr ,Cr ,1 , 2 
 0;
Cr
(9)
LLr ,Cr ,1 , 2 
 0;
1
LLr ,Cr ,1 , 2 
 0.
 2
Получената система е от четири

уравнения с четири неизвестни – Lr, Cr, λ1 и λ2.
Оптималните стойности на резонансните
елементи зависят само напрежението и токовете
в постояннотоковото звено – Us, Io1 и Io2.
Фиг. 4. Зависимости на IW=f(Lr) и IW=f(Cr)
3. Получени резултати
За удовлетворяване на входните В средата на Pspice е разработен
изисквания, е необходимо времето за преходния симулационен модел на асинхронно
процес на резоанасните елементи TW и токът, електрозадвижване с квазирезоанасен
осигуряващ запасяване на енергия в бобината IW преобразувател с една бобина и два паралелни
да имат минимални стойности. На фиг. 4 и фиг. ключа. С помощта на разработените модели е
5 са представени графиките на TW и IW в изследван трифазен асинхронен двигател тип
зависимост от изменението на Lr и Cr. От 4AO-80B-4D с номинална мощност PN=0,75kW.
графиките се вижда, че минималната стойност
II-49
Проектирането на квазирезонансния инвертора става, когато захранващото
преобразувател включва определянето на напрежение се установи на нула.
параметрите на Lr и Cr. На база на входните Фиг. 3 и фиг. 4 представя част от
данни: Us=537V; Io1=1,073A; Io2=3,172A, са получените резулатаи за един цикъл на работа
изчислени оптимални стойности по система на схемата. В таблица 2 са ставнени резултатите,
уравнениа (9). Резултатите са систематизирани в получени от изчислителната методика с тези,
таблица 1. В същата таблица са дадени и получени от симулационния модел.
минималните стойности на времето Tw.min и тока
Iw.min, получени по израз (6) и израз (3) Таблица 2. Резултати от изчисленията
съответно.
Изч. методика Сим. модел
Таблица 1. Оптимални стойности Δt1, μs 17,326 17,22
на резонансните елементи Δt2, μs 0,118 0,21
Δt3, μs 0,5 0,5
Lrо mH 2,67
Crо nF 1
Δt4, μs 2,567 0,57
λ1 - 1,08.10-4 Δt5, μs 15,751 18,0
IW, A 3,3285 3,45
λ2 - -1,74.104
ILr.max, A 3,499 3,46
Tw.min μs 0,6873
Iw.min A 3,3285
От таблицата се вижда, че резулатаите,
20 получени по изчислителната методика са близки
10 до резултатите, получени от симулационния
SA
0
8.31 8.315 8.32 8.325 8.33 8.335 8.34 8.345 8.35 8.355 8.36
-3
модел.
x 10
20
10 4. Заключение
S'A
0
8.31 8.315 8.32 8.325 8.33 8.335 8.34 8.345 8.35 8.355 8.36
-3
Разработен е алгоритъм за определяне на
20
x 10
параметрите на квазирезонансен преобразувател
10 с една бобина и два паралелни ключа за
Sa1
0
8.31 8.315 8.32 8.325 8.33 8.335 8.34 8.345 8.35 8.355 8.36
-3
захранване на асинхронна машина.
20
x 10
Симулационно е изследвана работата на
10 алгоритъма с асинхронно електрозадвижване,
Sa2
0
8.31 8.315 8.32 8.325 8.33 8.335 8.34 8.345 8.35 8.355 8.36 моделирано в средата на PSpice. От получените
t, s
резултати потвърждават достоверността и
-3
x 10
Фиг. 5. Зависимости SA, ,S’A, Sa1, Sa2 =f(t) работоспособността на алгоритъма.
600
400
1. Kedarisetti J. (2012). A Motor friendly Quasi-

uCr, V
200
0 resonant DC-link Inverter. Ph.D. Thesis, TU-

-200
Darmstadt.
8.31 8.315 8.32 8.325 8.33 8.335 8.34 8.345 8.35 8.355
x 10
8.36
-3
2. Юдов, Д., и др. (1999). Повишаващ DC/DC
4
преобразувател с подобрени енергетични
показатели. Annual Journal of Electronics,
2
TU-Sofia, стр. 120-125.
iLr, A
0
3. Dochev M., G. Rashev (2013). Method and
device for coil spring characteristics
-2
8.31 8.315 8.32 8.325 8.33 8.335 8.34 8.345 8.35 8.355 8.36 measurement, RaDMI 2013, pp. 777 – 782.
t, s
x 10
-3
4. Bose B. (2001). Modern Power Electronics and
AC Drives. Prentice Hall PTR, Upper Saddle
Фиг. 6. Зависимости uCr, iLr =f(t) River, NJ.
При генерирането на сигнал за
превключване на основните ключове, Димитър Спиров
допълнителните вериги трябва да се включат, Университет по хранителни технологии
така че да предизвикат резонансен преходен Адрес: гр. Пловдив, бул. Марица 26
процес. Превключването на ключовете на Телефон: 032/ 603 738
E-mail: dimitar_spirov@abv.bg
СЪВРЕМЕННИ УСТРОЙСТВА ЗА КОНТРОЛ

НА ДОСТЪПА
НИКОЛИНКА ЯНКОВА, ЕБРУ АДЕМ, ИБРЯМ АХМЕДОВ
Резюме: Системите за контрол на достъп позволяват физическия достъп до сгради

и помещения, като позволяват достъпа само на оторизираните лица с помощта на
пълен набор от електронни методи за идентификация и електронни заключващи
механизми. Функционирането им се базира на три признака на идентификация –
„Какво знам?“, нещо което знаем като ПИН код, „Какво имам?“, нещо което
притежаваме, като смарт карта, „Какво съм?, нещо което сме, като биометрични
белези или комбинация от тях.
Ключови думи: биометрични данни, контрол на достъпа
MODERN DEVICES FOR ACCESS CONTROL

NIKOLINKA YАNKOVA, EBRU ADEM, IBRYAM AHMEDOV
Abstract: Access control systems allow physical access to buildings and premises by
allowing access to authorized persons only by means of a full range of electronic
identification methods and electronic locking mechanisms. Their operation is based on
three identifying features – “What do I know?”, Something we know as a PIN, “What do I
have?”, Something we own as a smart card, “What am I?” , Which we are, like biometric
marks or a combination of them.
Key words: access control, biometrics
1. Увод • защита на неприкосновеността на

Съществуват много опасности, които с личния живот;
времето и напредването на технологиите, • оценка на контрол на потока;
човечеството успява да се справи. • контрол на влизане и излизане на
Защитата на обект се състои от няколко транспорта.
точки. Това зависи от нивото на чувствителни В допълнение, ACS е бариера за
места. Във всички случаи е важно т. к. системата „любопитни“.
ще контролира контрол на достъпа (ACS) на С изпълнението на специфични системи
обекта. за сигурност се използват различни методи за
Добрата организация е използването на прилагането на устройства за идентификация и
модерни технологии. ACS включват следното: удостоверяване на самоличността.
• противодействие на промишлен
шпионаж; Най-често се използва като ACS:
• борба с кражба; • въртящи се врати;
• противодействие на саботаж; • турникет за преминаване в коридорите;
• борба с умишлено повреждане на • шлюз кабина;
оборудване ; • автоматични врати;
• записи на работното време; • бариери;
• мониторинг на навременността на • системи за паркиране;
пристигането и заминаването на персонала; • кръгли плъзгащи врати.

II-51
2. Динамични методи Биометричните идентификатори работят
Ръкопис и подпис динамика, гласови и добре само когато операторът може да провери
речеви модели, темпо на работа на клапана. две неща:
Един от известните методи е следният: Първо – че биометричните данни
Интелигентен микропроцесорен получени от даден човек са верни
контролер за контрол на достъп до две врати и Второ – че тези данни съвпадат с модела,
работно време с RS485 комуникация. Този който се съхранява в досиетата.
модел е подходящ за изграждане на обекти с Биометрични характеристики са
множество контролери от серията iCON 1XX уникални идентификатори, но въпросът за
свързани в шина или самостоятелно ползване тяхната надеждно съхраняване и защита срещу
със сериен или LAN конвертор. подслушване остава отворен.
Известно разработване ACS, въз основа
на показанията и сравнението СОИ решетъчни
конфигурации вените на китката, миризмата на
пробите превърнати в цифров вид.
Всяка биометрична технология се
прилага в следните етапи:
• сканиране обект;
• премахване на индивидуална
информация;
• създаване на шаблон;
• сравняване на текущата база данни
Фиг. 1. iCON110
шаблон.
Контролерът iCON110 от Фиг. 1 има Потребителят адресира искане до ACS за
възможност за управление на една врата достъп предимно идентичен, като се ратифицира
двустранно или две врати едностранно. с лична карта, ключ или личен
Разполага със стандартизирани входове за два идентификационен номер.
четеца, които имат възможност за работа в 3. Идентификация с ирис
режим с ПИН код, карта с работен код,
биометрични четци или комбинация от тях. Пионер в областта на идентификацията
Във връзка с обслужването на системата от ирис на окото е д-р Джон Даугман. През 1994
контролерът работи в два режима: г. той е патентовал в САЩ. Биометрично
1. самостоятелен описване на структурата на ириса е подобно на
2. серийна комуникация. баркод.
При самостоятелния режим на работа
добавянето и изтриването на картите се
осъществява посредством Мастер карта, която
се активира първоначално чрез потребителски
софтуер. В режим комуникация имаме
възможност да се възползваме от удобството и
бързината на серийната RS485 комуникация.
Това ни дава възможност да ползваме iCON110
в самостоятелен режим с LAN или сериен
конвертор или да го свържем в шина с до 253 Фиг. 2. Разпознаване по ирис
устройства посредством RS485.
Всички настройки се извършват с Предимството на скенери на ириса е, че
помощта на платен (Andromeda Pro) или те не изискват потребителят да се концентрира
безплатен (Andromeda Tool) софтуер, които върху целта, тъй като извадката пет-петна в
допринасят за ползване на допълнителни ириса е на повърхността на окото. В
функции, описани във функционалността на действителност окото може да се сканира на
контролера. разстояние по-малко от 1 m, което прави
Трябва да се използват такива уникални възможно използването на скенери на ириса за
статични методи, като идентификация на слой цели например в банкоматите. Разработване на
на кожата, по отношение на сканиране на технологии за идентификация.
пръста, динамични методи – идентификация на Въз основа на сканиране на ириса по
движението на устните при възпроизвеждане на настоящем се използва в повече от 20 компании,
кодова дума. включително British Telecom.
II-52
Има активна и пасивна система за съхранявате голямо количество записи и бързо
разпознаване. В система от първият тип трябва търсене. Личните идентификационни устройства
да конфигурираме самата камера. за ръчна геометрия така са широко използвани
Пасивни системи – лесни за използване, устройства в САЩ.
тъй като камерата се настройва автоматично. Най-популярното устройство сканира
Имат висока надеждност, оборудването има както вътрешната така и страничната част на
голямо приложение на всякакви обекти. дланта, с помощта на вградената видеокамера и
алгоритми за компресия. Това възлиза на повече
4. Идентификация на капилярите на от 90 различни характеристики. Представител на
ретината тази тенденция е развитието на ACS American.
При определянето на ъгловото Фирмата Steller Systems, произвежда
разпределение, измерено от ретината, Identimat терминал. За отчитане на
кръвоносните съдове в повърхността на геометричните характеристики на ръката й с
ретината по отношение на мъртвата зона на длан надолу в специалния панел. През
очите и други симптоми. Моделът на капилярна процепите в повърхността на оптичния сензор
система се различава от ретината дори при клетка сканира четири пръстена. Тези клетки
близнаци, и може да се използва много успешно определят началната точка за два чифта пръсти
за идентифициране на индивида. – показалеца и средния, пръстен и малкия пръст.
Основното устройство за този тип Всеки пръст е сканиран по цялата дължина,
система е бинокулярно. При провеждане на където дължината се измерва в огъване и
процедура за удостоверяване на потребителя разстоянието до „съсед“.
трябва да се придържа окото към окуляра. През Резултатът на удостоверяване е
това време системата има време за осветяване на положителен за потребителя. Цифровият
ретината и получава отразения сигнал. Не всеки стандарт се съхранява в база данни или в
човек ще се осмели да погледне в неизвестно паметта на идентификационна карта. По този
нещо, което свети в очите му. начин, за защита данните са криптирани.
F702 от Фиг. 4 е самостоятелен
5. Идентификация по геометрията на биометричен терминал за контрол на достъп и
ръката отчитане на работно време с пръстов отпечатък
Метод за удостоверяване на потребителя и парола.
чрез формата на ръката в технологичната си
структура и нивото на надеждност е сравнима с
метода идентификация чрез пръстови
отпечатъци. Статистическата вероятност за
съществуването на две ръце е изключително
малка със същата геометрия. На Фиг. 3 е
примерен скенер.
Фиг. 4. F702
Внедрената операционна система

базирана на Linux гарантира надеждността при
непрекъсната работа на терминала. Отговаря на
всички световни индустриални стандарти.
Може да се използва в учреждения,
където се изисква висок контрол на достъп като
банки, корпоративни офиси, помещения със
сървъри, помещения със складирана
класифицирана информация, дори и за нужди на
армията.
Фиг. 3. Скенер за геометрията на ръката Биометричен терминал – контрол на
достъп и работно време, базиран на лицево
Математически модел на идентификация разпознаване, 125 kHz, ASK RFID и ПИН
за този параметър изисква някаква информация идентификация VF380. Общ вид на VF380 е
– само 9 байта, която ви позволява да даден на Фиг. 5.
II-53
3. Изводи:
Системата трябва да има възможност за
интеграция на високо ниво с останалите системи
за сигурност на един обект (основно контрол на
достъп и видеонаблюдение) и графично
представяне на информацията за актуалното
състояние на системата и алармените събития на
РС дисплей или на мобилен телефон.
Периферните устройства на системата -
детекторите, трябва да могат надеждно и
Фиг. 5. VF380 своевременно да реагират на всички видове
атаки към обекта и същевременно процентът на
Може да се използва в учреждения, фалшиви сигнали, независимо от пораждащите
където се изисква високо ниво на сигурност: ги причини, трябва да е близък до нула.
банки, корпоративни офиси, помещения със Не на последно място, от значение за
сървъри, помещения за достъп до крайния потребител е простотата на управление
класифицирана информация. Терминалът може на системата – колкото по-опростено е
да се използва за отчитане на работно време, за управлението, толкова по-малко са
контрол на достъп или в комбинация от двете. възможностите за човешки грешки по време на
Високо ниво на сигурност гарантирано експлоатацията.
от различни комбинации за идентификация:
лицево разпознаване, RFID карта и парола, като ЛИТЕРАТУРА
всички методи могат да се комбинират.
TeraAccess – Представлява система за 1. Ворона В., Тихонов. В. (2010). Системы
контрол на достъпа и отчитане на работно контроля и управления доступом. М.:
време, която се състои от следните компоненти: Горячая линия – Телеком, 272 с.
• компютър, на който е инсталиран 2. http://www.sectron.com/bg/grid/62/kontrol-na-
управляващият софтуер TeraAccess; dostyp-62/
• контролери за достъп ACT120-O/O2;
• интерфейсен контролер, осъществяващ
връзка между компютъра и контролерите Николинка Спасова Янкова
(RS232/RS485); Студент в Шуменски университет
• входни/изходни терминали за достъп „Епископ Константин Преславски“,
(ACA1XX); гр. Шумен 9712, „Университетска“ 115
• безконтактни карти. Телефон: +359 877 884 519
Новите устройства за достъп с карти са E-mail: niki_qnkova@abv.bg
стилни и надеждни, поради което вече навлязоха
и в големите хипермаркети например. Външен Ебру Гюнай Адем
вид на устройството е даден на Фиг. 6. Студент в Шуменски университет
Те са Ethernet мрежови устройства, което „Епископ Константин Преславски“,
позволява да са много в една локална мрежа, не гр. Шумен 9712, „Университетска“ 115
е необходимо да се довежда до тях Телефон: +359 896 554 297
допълнителен кабел, а само кабел с четири E-mail: ebru.9500@abv.bg
усукани двойки (например UTP или FTP).
Захранващото напрежение се подава по същия Ибрям Сюлейманов Ахмедов
кабел чрез PoE (Power over Ethernet). Студент в Шуменски университет
„Епископ Константин Преславски“,
гр. Шумен 9712, „Университетска“ 115
Телефон: +359 899 377 696
E-mail: ibrqm_96@abv.bg
Фиг. 6. Устройство за достъп с карти

II-54

Electrical Engineering and Electronics
Електротехника И Електроника
ANALYSIS OF TIME TO BREAKDOWN

VOLTAGE MEASUREMENT
EMEL ONAL
Abstract: In this paper, regression analysis of the experimental results for different gas
mixtures at different pressures is done by a general purpose statistical software package
called STATA. Experimental results contain breakdown times of varied gas mixtures under
several pressures as nanosecond depends applied different voltage magnitudes at different
polarities. Through the statistical software regression analysis done between three
parameters, these are breakdown time as nanosecond, voltage magnitude as kilovolt and
gas pressure as bar. After several regression analyses, gas pressure and voltage magnitude
dependent breakdown time equations are obtained.
Key words: Impulse voltage measurement, breakdown, regression, statistics.
1. Introduction experimental results. Results are compatible with

In practice high voltage insulation systems theoretical predictions.
consist of different types of insulators such as gases,
2. Experimental results and regression
liquids, solids or any combination of them. When
analysis
electric field applied to the insulator, breakdown is
occurred. If insulation strength is higher than In many engineering problems, two or more
applied electric field breakdown is not occurred. random variables are not statistically independent of
This electrical breakdown phenomenon is not each other during the same observation, so there is a
deterministic process, it is random. For this reason, relationship between these variables. The
properties of the insulator examined by statistically. relationship between two variables is that one of
To examine the electrical insulating properties of a them is influenced by the other, or that both
system statistically, some experimental data are variables are affected together by other variables.
required. Therefore, to get experimental data about For example, the relationship between flow and
insulation system, under different voltage rainfall in a river basin is arisen by flow is emerged
magnitudes, different insulators, different voltage after rainfall. The relationship between the flows in
polarities, some different any combination of them, the two neighboring basins depends on whether
measurements are obtained. Because the volt-time they are affected by the precipitation in that area.
characteristics of the electrical breakdown of However, the relations are not deterministic
compressed gases is of practical importance in (ie, functional), meaning that when one of the
designing protection and insulation coordination for variables takes a certain value, the other will not
gas insulated systems, it has been studied always get the same value. This value may be
theoretically [1-2] and experimentally [3-4]. slightly different in various observations due to the
After this data obtaining process, some influence of other variables that we do not consider
statistical formulas can be obtained. In this paper in a related way. For example, when one of the two
using some experimental data, such as breakdown adjacent river basins has a certain amount of flow,
time depending on applied lightning impulse the flow in the other is not always has the same
voltage magnitude and polarity and various amount. Nevertheless, it is of great importance in
insulating gases under different pressures, via a practice to identify and determine the existence of
statistical software called STATA for each gas nonfunctional relations between variables. Using
mixture regression analysis are done between three this relation, it is possible to estimate the value of a
parameters, these are breakdown time, gas pressure variable depending on the known values of the other
and applied voltage magnitude. The theoretical (or more) variables. This estimation is closest
predictions are examined in the light of extensive estimation of subjected variable although it is not
the actual value. It can be said that the difference
between the estimated value and the actual value
II-56
(error) will also located within certain limits with a at much higher voltages than the latter one and it
certain probability. The mathematical expression usually occurs at small electrode separations. In the
that represents a relation of the type mentioned first regime the time to breakdown decreases more
above is called the regression equation. The purpose or less regularly with increasing crest voltages
of regression analysis is to determine if there is a while in the second one the time to breakdown
meaningful relationship between the variables shows at wide range distributions. SF6 and SF6
considered, to obtain the regression equation mixtures are generally used at switchgear and
expressing such a relation and to calculate the insulation equipment’s because the breakdown
confidence intervals of the predictions to be made voltage of this gas is very higher than that of other
using this equation[5-8]. insulation gases.
The results of the regression analysis and
experimental analysis with different gas mixtures at 350
different pressures are shown as follows. 300
The lightning impulse voltages used in this 250
study are produced by a 1 MV, 50 kJ, Marx type 200
U (kV)
impulse generator. The voltages are measured by 150
means of a capacitive divider and a HIAS 743 100
digital oscilloscope with 12 bit real vertical

50
resolution at 120 Mega sample / sec. All

0
0 500 1000 1500 2000 2500
t (ns)
measurements of the experimental study are given

in IEC standard. The present paper describes a Fig. 2. Experimental results of relationship
study of breakdown strength signal analysis of between applied voltage magnitude U(kV) and
Sulphurhexafluoride (SF6) and SF6 gas mixtures in breakdown time t(ns) for %100 SF6 gas at the
positive impulse voltage. Sulphurhexafluoride gas pressure of 1 bar
due to its exceptional insulating and arc-
extinguishing properties has been widely employed If these values are transferred to the
as insulation of high voltage power apparatus. STATA environment, the values at following table
Experimental test set-up is shown at Fig.1. 1 are obtained.
In engineering field, STATA is used to find
out the breakdown time of machines. This analysis
is also called reliability or failure time analysis[7-
13].
Table 1. Regression analysis of %100 SF6 gas
Fig. 1. Experimental set-up
2.1. Analysis of %100SF6 Under Pressured

Positive Polarity Lightning Impulse
Voltage
First of all, breakdown time as nanosecond
depended applied positively polarized lightning
From here, breakdown time equation is
impulse voltage under 9 bar, 7 bar, 5 bar, 3 bar, and
obtained as follows;
1 bar pressure for 100%SF6 gas is obtained
experimentally. As an example graph of 100%SF6 t(ns)=3442.428-9.095287*U(kV)-4.98655*bar (1)
and 1 bar pressure is shown as Fig 2. As seen from
Fig 2, when the breakdown time increases, the In equation (1), t shows the breakdown time
breakdown voltage decreases. Finding the time to in nanoseconds. U kV represents the peak value of
breakdown curve is important for all equipment applied impulse voltage. The bar in the equation
especially for designing insulation coordination. takes the gas pressure between 8 and 0 bar as a bar
Basically two regimes can be distinguished, unit.
depending on whether breakdown occurs on the rise Statistically, the value of P>|t| is 0, 543
of the impulse or on its tail. The first regime occurs which is inconsistent. Here is a conclusion, the
II-57
effect of the gas pressures in this range is less breakdown time t (ns) for %1 SF6+% 99 N2 gas
effective than impulse voltage peak magnitude. As mixture at the pressure of 1 bar
seen figures 2,3,4, breakdown time and voltage
values are varied in large range. For this reason, it is Regression equation for %1 SF6+% 99 N2
important to know time to breakdown values. gas mixture is shown at equation 3.
2.2. Analysis of %50 SF6 +%50 N2 Under

Pressured Positive Polarity Lightning t(ns)=1962-9.530341*U(kV)+34.51702*bar (3)
Impulse Voltage
In our second model, the following
regression analysis is obtained from our data, which
consist of a mixture of 50% SF6 + 50% N2 gases
and have different breakdown time due to different
voltage amplitudes with positive polarity at
different pressures as 9, 5, 3, 1 bar. As an example
time to breakdown voltage graph of 100%SF6 +%50
N2 gas mixtures for the pressure of 1 bar is shown as
Fig 3.
350
300
250
200
U (kV)
150
100
50
0
0 500 1000 1500 2000 2500 3000 3500
Fig. 5. Regression analysis results of relationship
t (ns)
between applied voltage magnitude U(kV) and
Fig. 3. Experimental results of relationship breakdown time t (ns) for %1 SF6+% 99 N2 gas
between applied voltage magnitude U(kV) and mixture at the pressure of 1 bar
breakdown time t(ns) for %50SF6+%50 N2 gas
mixture at the pressure of 1 bar
Table 2. Regression analysis of different percentage
Regression equation for %50SF6+%50 N2 of SF6 and SF6 mixtures
gas mixture is shown at equation (2). Test Standard Skewness Kurtosis
environment deviation
t (ns)=3328.191-10.59744*U (kV)+4.73382*bar (2) %100 SF6 gas at 684.7446 0.1774 1.6092
pressure of 1 bar
2.3. Analysis of %1 SF6 +%99 N2 Under
%50 SF6+%50 839.2957 0.8939 2.6595
Pressured Positive Polarity Lightning
N2 gas mixture
Impulse Voltage at pressure of 1
Experimental result is shown at Fig. 4 bar
for %1 SF6+%99 N2 gas mixture. %1 SF6+%50 N2 426.1696 0.3600 2.1473
gas mixture at
160 pressure of 1 bar
140
120
100
U (kV)
80
3. Conclusion
60
40
The volt-time curves predicted by
20 regression method with stata. The third case is
0
0 200 400 600 800 1000 1200 1400 1600 1800 2000
presented in Fig. 5, together with the experimental
t (ns)
results.
Fig. 4. Experimental results of relationship It can be seen that, for the three impulse
waveshapes considered, the method that best fits the
between applied voltage magnitude U(kV) and
experimental data, which leads to a very good
II-58
agreement between predicted and observed results. 6. Lawless, J.F., (2003), Statistical Models and
The greatest difference between measured and Methods for Lifetime Data, University of
calculated times to breakdown is at %50 SF6+%50 Waterloo, Wiley Interscience a john wiley &
N2, this case is seen at table 2 clearly. This values Sons Inc publication.
are insufficient to reach a definite result. It is 7. Wilson, M.P., Given, M.J., Timoshkin, V.I.,
needed to obtain more data for exact result. MacGregor, S.J., Wang, T., Sinclair, M.A.,
Thomas, K.J., Lehr J.M, (2011), Weibull
For tail chopped impulses with the same statistical analysis of impulse driven surface
time to breakdown, in principle the voltage value on breakdown data, 2011 IEEE Pulsed Power
the volt-time curve should be higher for waves with Conference, Pages: 218 - 222
shorter tails. As such impulses have faster decays, 8. Wang, Y., Wen, H., Jian, Z., Wu, Z, (2014).
higher voltage levels should be required for the Steganalysis on positive and negative echo
breakdown to occur. The longer the time to hiding based on skewness and kurtosis, 9th
breakdown in relation to the time to crest, the more IEEE Conference on Industrial Electronics and
pronounced should the differences between the Applications, pages: 1235–1238.
curves be. As seen in fig. 5, breakdown values of 9. Hayakawa, N., Yoshitake, Y., Koshino, N.,
experimental values are lower than that of predicted Ueda, T., Okubo, H. (2005). Impulse partial
values. Because predict values are more results. discharge characteristics and their mechanisms
It is carried out experiments and simulation under non-uniform electric field in N2/SF6 gas
to investigate the relationship between breakdown mixtures, IEEE Transactions on Dielectrics
characteristics and time parameter of applied and Electrical Insulation, Vol.(12), Issue:5,
lightning impulse voltage in this paper. pages:1035-1042.
These relationships are also evident from 10. Ushakov, V.Y.(2007). Statistical Investigations
Tables 1, 2 and fig.5. Since the tables depict that of the Electrical Breakdown, Springer Berlin
regression model among the three variables are very Heidelberg, pages: 239-262.
good fit for the considered analysis. 11. Yuan, Z., Yu, L., Chun, D., Yi-chao, Y., Jin-
liang, H., Xi, W. (2014). Experimental Study
and Analysis of Insulator Breakdown
REFERENCES Characteristics with Short-tail Lightning
Impulse, Journal of International Council on
1. Sima, W., Sun, P., Yang, M., Wu, J., Hua, J. Electrical Engineering, pages:199-203.
Smith, S.E. (2016). Impact of time parameters 12. Kadir, M., Cotton, I. (2010).Application of the
of lightning impulse on the breakdown Insulator Coordination Gap Models and Effect
characteristics of oil paper insulation. IET High of Line Design to Back flashover Studies,
Voltage, Vol. (1), pages: 18–24. Journal of Electrical Power and Energy
2. Mazzetti, C., Pompili, M., (1990). Study of the Systems, Vol. (32), no. 5, pages: 443-449.
Time to Breakdown in Transformer Oil under 13. Carrus, A., Funes, L.E. (1984). Very Short
Impulse Conditions. Transactions on Electrical Tailed Lightning Double Exponential Wave
Insulation, Vol. (25), No. 6, pages: 12-26. Generation Techniques Based on Marx Circuit
3. Joshi, V., Solanki, J., Solanki, S.K. (2013). Standard Configurations, IEEE Transactions
Statistical method to improve voltage profile on Power Apparatus and Systems, Vol. (103),
under incomplete distribution system no. 4, pages: 782-787.
information. 2013 North American Power
Symposium (NAPS), pages:1-5.
4. Maluckov, C.A., Mladenovic, (2016), S.A. AUTHOR: EMEL ONAL
Brakdown in low pressure Ne gas: mechanism
and statistical analysis of time delay, IEEE
Transactions on Dielectrics and Electrical Istanbul Technical University, Electrical
Insulations, Volume (23), Issue: 1, pages: 202- Electronics Engineering Department,
210. Maslak, Istanbul, 34469, Turkey.
5. Nyanteh, Y., Srivastava, S., Edrington, S., E-mail: eonal@itu.edu.tr
Cartes, D., Rodrigo, H. (2015), A novel
approach towards the determination of the time
to breakdown of electrical machine insulating
materials, IEEE Transactions on Dielectrics
and Electrical Insulation, Vol. (22), Issue:1,
pages: 232 – 240.
SHORT TERM ELECTRICITY CONSUMPTION

FORECASTING: A CASE STUDY OF ELECTRIC
UTILITY AT TURKEY
EMEL ONAL
Abstract: Energy need has been increasing day by day, and this situation makes energy
management is requirement. That’s why, significant rise can be seen about tending to
energy management applications and different modelling in last years. This concept has a
wide definition and also many subtitle. This study focus on consumption side, demand side
management is explained by detail. Demand side management is the reaction of electricity
price change given by consumers ranged from industrial to home. Demand switch occurs
when the reaction effects the market. Demand response is entegrated via programs to
market. This research is analyzed through consumption. Also demand management is
defined by some resources as a whole applications for longer process. Demand response is
kind of subtitle of demand management, which both of them aim the increasing of energy
efficiency. At this study, the prediction is done by using daily peak electric consumption
values for June 2016. The results are compared with the results of Turkey Electric
Transmission Corporation. Data of Turkey Electric Transmission Corporation at maximum
is % 5.94 bigger than that of found prediction values.
Key words: Demand side management, Energy management, Electric consumption
1. Introduction world and problems of this distribution process of

Purpose of Demand Side Participation is energy.
using more efficient and economic energy. This
2. Turkey’s and Other Countries Positions
system has great benefits. That’s why a lot of
on Demand Side Management
country has started to try that system many years
ago. Still, there are studies about the development Beginning of the 2000’s, USA, some
and using effective of this program. Even some European countries and Australia started to operate
countries issued regulations to consumers involve this system successfully. Energy Efficiency
the system. Turkey is behind from these countries Directive of European Commission put some
when it is compared. Fundamental of Demand articles to support the demand side participation for
Participation based on compatibility of supply and regulators in 2012. USA could be seen as more
demand. Demand Participation becomes development country regarding demand side. One
requirement in case of the harmony failures. Also, of the supplier companies, called ENERNOC, has
satisfying studies are not done in Turkey to come out with new projects with aggregators and
entegrate demand participation into the market. conducted a survey country-wide including 4 other
Load flexibility must be provided and let the corporation. With this survey, 11.8 Billion Dollars
suppliers have opportunity to sale in market [1-4]. energy saving is announced between 2013-2014
Therefore, these all prove that active demand side years. Additionally, after demand side participation
participation and works are needed in Turkey. is done and operated correctly, 500 Million Dollars
While surveying why Turkey needs demand side income made is seen and announced by another
participation, examples of this subject on foreign research on USA since 2010. In Australia, this
countries must be examined and achievements of system is advanced level and consumers are willing
countries whose implement that program very to attend. Also, Canada, Japan and South Korea are
successfully should be analyzed. Demand side the other countries whose attend in a significant
participation is important to provide flexibility to level. It is possible to create more secure and clean
the market [5-9]. Because according to produced sources by increasing the efficiency of this method.
and purchased energy amount on all countries on

II-60
In our country, demand management When compared 2013 and 2014 years, Belgium,
projects are not developed and well planned. Energy England and Swiss are the countries whose are
is produced by natural gas on highest rate. Hence, active in this program in 2013, remain their activity
energy interruption can be seen when the occasion in 2014. Besides those countries, Ireland, France
that demand goes to peak point. For this reason, and Finland become the active countries in 2014.
increasing the rate of renewable energy sources on These three countries studied partially about this
production is very serious subject both our country program in 2013and by increasing their
and foreign countries[10-12] . developments, they became an active situation in
Demand participation is the system focused 2014. Austria, Norway and Sweden are three
on consumer. When capital and small consumers countries whose are in partial study for both 2013
entegrate into the system, providing the demand and 2014 years as seen in fig.1. When looked at
flexibility shows that how important the roles of 2015 mapping, there is no country not active before
every kind of consumers. According to research 2015 and become an active after 2015. However, It
which is done by Smart Energy Demand Coalition, can be seen that France and Swiss step forward as
2.2 Billion Dollars income was made in USA in more active compared to countries whose status are
2013. Therefore, it was indicated that this situation in active in 2015. The reason for this is that
prevented a large amount of investments to the wide participants and free aggregators, who are in
grids and power plants just by using system. And market, roles and responsibilities are defined well in
this means huge benefit for consumers. So, all kind both country. Italy and Spain are the only the
of consumer can take an advantage of this process countries whose status are not in active for three
and direct source can be made up through that, also years [13-14].
productivity of this method can be measured. The
direct source which is made up can launch to the
market of countries local economies. In this way
market could be relieved and consumers and
suppliers might be encouraged for the market
competition. Despite that, operating this process
efficiently, whole barriers must be defined and
looking for solutions to eliminate them. Likewise,
entire tools, infrastructures, technological
developments and projects must be examined.
3. Mapping Demand Response in Europe

Development of countries’ demand
response and their progress would be evaluated
according to some criteria’s. Also, mapping demand
side response in Europe would be done based on
these criteria’s. These are;
*Consumer access & aggregation

*Programme requirements
*Measurement & verification
*Finance & penalties Fig. 1. Demand Response Activity in Europe
Development of demand response of 4. A Case Study: Prediction the

countries could be compared with each other consumption of next four years from
according to these criteria’s. As specified SEDC’s daily peak load in June 2016
(Smart Energy Demand Coalition) research,
countries using this method efficiently do not have Peak load analyze is done in the fig. 2 at
to make large scale investments to build power below for June 2016. The figure shows the highest
plants to produce energy. So, this source can be and the lowest peak load values in one day in June.
used their local economies. Reducing of Load values can be seen in the figure for what time
investments for peak load back-up generators can they reach highest or lowest point and how much
be pointed as an example. Mapping and comparing they are. Because of the season, it could be
are done in Europe according to reports which is predicted that high temperatures may lead to this
published by SEDC for 2013, 2014 and 2015 years. situation. Also this proves that, those values are not
II-61
only up to temperatures, but also terms, season or Table 1. Difference of predicted values of TEIAS
other different conditions can effect to peak load and this study according to years
hours and their values. At here, certain days of one
month are taken into the consideration. years difference of predicted
values as percentage
(%)
2016 1.43
2017 2.25
2018 3.42
2019 4.67
2020 5.94
Fig. 2. Daily highest and lowest hourly peak load

Of course, all independent factors,
Fig. 2 is analyzed by choosing random day
occasions, financial parameters cannot known as
in one month and examining the highest and lowest
certain on these predictions. It can be said that those
peak load in that day. This is also practiced to other
differences are reasonable. The reason that the
random days in other months. As showed in the
TEIAS data is high is increasing energy need.
figure, according to values and hours when peak
loads are highest and lowest level, it is possible to
make some predictions. Although, these predictions
could be made just based on normal circumstances,
some other variables in certain time period can
effect those predictions.
When considering 2016 monthly and yearly
peak load analyze, the conclusion proves that why
demand side management needs to be done, why
focus on those projects and new modelling must be
produced. Short-term demand forecasting also plays
a role in the process of regulation. A precise
estimate of demand is important for the purpose of
setting tariffs. A detailed consumer category-wise Fig. 3. Real consumption and prediction values
consumption forecast helps in the determination of between 2016-2020
a just and reasonable tariff structure wherein no
consumer pays less than the cost incurred by the 5. Conclusion
utility for supplying the power. Also, the utility can
then plan the power purchase requirements so as to In conclusion, Demand Side Response is
meet the demand while maintaining the merit order defined, how it is operated on other countries is
dispatch to achieve optimization in the use of their shown and benefits of this system are indicated.
resources. Peak load is analyzed by hourly, weekly, monthly
According to reports published by TEIAS and yearly for 2016. With all these data’s, according
(Turkey Electric Transmission Corporation) real to previous months, terms and years, prediction is
consumption amounts since 1970 can be seen on the practiced with consumption values. It all summarize
fig.3. Using 3rd degree curve method, on the basis of how important the demand side management is.
the real consumption values between 1970-2015 Because otherwise, at negative scenario different
years, demand prediction is made until 2020. kind of problems which my lived can be estimated.
Herewith, values found by using prediction is As a result, nowadays, considering expanding
compared with other values which was made by energy needs of countries, more significant and
TEIAS and the difference between both of them are effective planning and new modelling can point the
shown in the fig. 3. Hereby, when compared with benefits of this system.
TEIAS predictions, the values at table 1 are seen.

II-62
The nature of the forecasts has also changed markets, Electric Power Systems Research,
over the years. It is not enough to just predict the Vol. (78), pages: 1989–1996.
peak demand and the total energy use, say on an 8. Ikäheimo, J., Evens, C., Kärkkäinen, S. (2010).
annual basis. Since the whole objective of demand- DER Aggregator business: the Finnish case,
side management is to alter the system load shape, a Research Report: Technical Research Centre
load shape forecasting capability within the system of Finland (VTT): Finland, pages 39-41.
is most desirable. The various methods of 9. Türkyılmaz, O., (2015). Turkey's Energy
implementing demand-side management are time of Outlook Report. Journal of Turkish Chamber
use pricing, use of curtailable/interruptible rates, of Mechanical Engineers, Vol. (200), pages:
and imposition of penalties for usage beyond a 20-22.
predetermined level, and real time pricing. A time- 10. Turkısh Electrical Energy 10- Year Generation
of-day tariff structure to manage peaks and troughs Capacity Projection (2009-2018). (2009).
in electricity demand, an hour-by-hour load shape Turkısh Electricity Transmission Corporation
forecast has become an essential prerequisite. Research Planning and Coordination
Further, the end-use components of the load Department, pages:56-67.
shape must also be known in order to plan the other 11. International Energy Agency, (2016), Energy
demand-side management activities to achieve Policies of IEA Countries 2016 Review Turkey,
maximum conservation, while avoiding undue pages:177-198.
demand restrictions. Another use for demand 12. Yuksel, I. (2015). South-eastern Anatolia
forecasting models is the assessment of the impact Project (GAP) factor and energy management
that a new technology might have on the energy in Turkey, Energy Reports, Vol. (1),
consumption. This helps planners to evaluate the pages:151-155.
cost effectiveness of investing in the new 13. Yumurtaci, Z., Asmaz, E. (2004). Electric
technology and the strategy for its propagation. energy demand of Turkey for the year 2050,
Energy Sources, Volume (26), Issue 12,
pages:1157-1164.
REFERENCES 14. Boiron, A.,Lo, S.,Marot, A. (2012). Predicting
Future Energy Consumption, CS229 Project
1. Capehart, B. L., Turner W. C., Kennedy W. J. Report, pages:1-5.
(2012). Guide to Energy Management, 7th ed.
Bookshub publication, pages:120-130.
2. Wang, Y.,Yang, W., Liu, T. (2017).
Appliances considered demand response AUTHOR: EMEL ONAL
optimisation for smart grid, IET Generation,
Transmission & Distribution, pages: 856-864.
3. Huang, Y., Tian, H., Wang, L. (2015). Demand Istanbul Technical University, Electrical
response for home energy management system, Electronics Engineering Department,
International Journal of Electrical Power Maslak, Istanbul, 34469, Turkey.
Energy Systems, Vol. (73), pages: 448–455. E-mail: eonal@itu.edu.tr
4. Samadi, P., Mohsenian-Rad, A.H., Schober,
R., Wong, V. W. S., Jatskevich, J. (2010).
Optimal real-time pricing algorithm based on
utility maximization for smart grid, IEEE
International Conference on Smart Grid
Communications, pages: 415–420.
5. Bertoldi, P., Zancanella, P., Boza-Kiss, B.
(2016). Demand Response status in EU
Member States, European Commision JRC
Science for policy report, Pages:14-35.
6. Ishik, M.Y., Göze, T., Özcan, İ., Güngör, V.Ç.,
Aydın, Z. (2015). Short term Electricity Load
Forecasting:A case study of electric utility
market in Turkey, Smart Grid Congress and
Fair (ICSG), 2015 3rd International Istanbul.
7. Albadi, M.H., El-Saadany, E.F., (2008). A
summary of demand response in electricity
ХИБРИДНИ МРЕЖИ ИЗГРАДЕНИ ПО

ТЕХНОЛОГИЯ ЗА ПЛЪТНО УПЛЪТНЕНИЕ НА
КАНАЛИТЕ ЧРЕЗ ДЕЛЕНИЕ НА ДЪЛЖИНАТА
НА ВЪЛНАТА
АЛЕКСАНДЪР САРАФОВ
Резюме: Документът представя работата и принципът на действие на хибридни

мрежи изградени по технология за плътно уплътнение на каналите чрез деление на
дължината на вълната. Разглеждат се потенциалните проблеми при използване на
метода за плътно уплътнение на каналите чрез деление на дължината на вълната,
като загуби по влакното, прислушване, нелинейни ефекти, капацитет на
възможните канали, които могат да бъдат уплътнени, скорост за предаване на
данни и общ капацитет на хибридната мрежа. Засяга се тяхното настоящо и
бъдещо приложение в съвременият бързо развиващ се свят на компютърните
мрежи.
Ключови думи: оптични, мрежи, плътно, уплътнение, канали, светлина,
дължината на вълната, капацитет, скорост, кохерентност.
HYBRID NETWORKS USING DENSE

WAVELENGTH DIVISION MULTIPLEXING
(DWDM) TECHNOLOGY
ALEXANDER SARAFOV
Abstract: The document reviews the work of the hybrid networks using dense wavelength
division multiplexing (DWDM) technology. The potential problems in using the dense
wavelength division multiplexing (DWDM) technology are considered. A review of their
present and future usage in the fast changing technology world in the field of computer
communications or data communications is given.
Key words: optics, networks, dense, multiplexing, channels, light, wavelength, capacity,
velocity, coherent
1. Въведение като той също влиза в това число с частичен принос).

Като начален момент в съвременната Още един значително важен извод от теорията на
история на предаване на информацията в електромагнитното поле е, че светлината е също
пространствено-времевия контур се дава периода електромагнитна вълна и може да се разглежда като
1887-1891 г. в който германския физик Хайнрих функция от пространствено-времевия контур.
Рудолф Херц, успява експериментално да докаже Експерименталното доказателство на Хайнрих Херц
съществуването на електромагнитната вълна се състои в долавянето на излъчена
предсказана теоретично от шотландския физик електромагнитната енергия разпространяваща се в
Джеймс Кларк Максуел през 1873 г. на базата на пространствено-времевия контур под формата на
съставената от него динамична теория на електромагнитна вълна или с други думи
електромагнитното поле (която от своя страна е предаването и приемането на електромагнитото поле
селектирана съвкупност от емпирични и теоретични между два токови контура, които нямат обща
зависимости от формули и постулати на физици галванична връзка помежду си. Няколко десетилетия
изследвали природата на електромагнитното поле, по-късно след откритието на Херц става и догатката
за използването на свойствата (параметрите) на
II-64
електромагнитната вълна чрез тяхната промяна за точката, където има разположено DWDM оборудване
пренасянето на информация, като информацията се е прието и наименованието точка на присъствие PoP
смята за отделна трета съставка зависима от (Point of Presence). Кръговите мрежи изградени на
пространствено-времевия контур, което възприемане принципна за плътно уплътнение на каналите чрез
и анализ допълва знанието/незнанието на човешки разделяне на дължината на вълната се налагат поради
индивид за конкретно събитие в пространствено- огромното количество от трафик, което мрежата
времевия контур, като всичко случило се до момента трябва да поеме. Също така с тяхното използване се
на предаване на информация се счита за такова. След изпълнява едно от важните за клиентите изискване, а
доказателството на Хайнрих Херц на труда на именно резервираността на услугата. Т.к.
Джеймс Кларк Максуел динамична теория на топологията е кръгова, ако възникне повреда между
електромагнитното поле, започва и нейното по- две точки на присъствие от мрежата е възможно
нататашно развитие, като от една страна параметрите пренасочване на трафика в обратна посока, така че
на теорията на модерната електродинамика са той отново да стигне до дестинацията. Това
преобразувани в по-прости дименсиални такива за неименуемо води до повишаване на трафика, който
улеснение на инженерни цели и се съсдава науката се носи от кръговата топология, като реално трафика
електротехника. Тя придобива широк интерес и който е пренасочен се наслагва сумира с трафика в
масовост поради леснотата си и връзката с обратна посока т.к. техните посоки съвпадат след
практическите си приложения, като измервания, пренасочването. По този начин общият сумарен
изследвания, анализ и др. От друга страна теорията за капацитет на мрежата нарастра драстично и е важно
електромагнитното поле продължава да се развива с тя да е така проектирана да няма възможност за
непроменени параметри и така се създават науките, надхвърляне на нейния капацитет. Мрежи по
като микровълнова техника още позната и като технологията DWDM се използват още и за връзка
свръхвисокочестотна техника, поради познатата между държави, а именно между две точки на
връзка между дължината на вълната и честотата, присъствие разположени в две страни, почти винаги
както и науки свързани с теоретични познания за управлявани от два различни доставчика на услуга.
разпространението на електромагнитните вълни в
различни веществени материални среди и ефектите, 3. Симулации и експериментални
които настъпват в тях. Тази наука поражда резултати
теоретичния анализ на предавателните линии, като Развитието на мрежите изградени по DWDM
различни видове вълноводи, обемни резонатори и технологията отива още по-далеч, като започва
микровълнови елементи и устройства, но добива изпозлването на хибридни мрежи изградени от
приложение една след 1940 г.. Теорията на DWDM оборудване, включвайки кръгови
връзковите уравнения в предавателните линии, като топологични мрежи, мрежи от точка до точка, както
телеграфните уравнения, коаксиалния кабел и други и смесени мрежи. Една такава хибридна мрежа е
видове проводници се основава на комплексни показана на Фиг. 1, като същата може да обхваща
принципи между електродинамиката и територии на цели континенти. Важен аспект е да се
електротехниката. Използването на тази теоретична отбележи, че такива мрежи имат възможност за
база, както и на множество експериментални тестове свръзване на информационни центрове познати под
на редица учени в периода от 1920-1970 г. води и до английското наименование, като Data Centers.
създаването на модерното оптичното влакно познато
днес. Информацията е основен градивен елемент
подържащ общуването или комуникацията, като
обменен процес в съвремените комуникации между
индивиди с разнообразни цели.
2. Хибридни мрежи изградени по

технология за плътно уплътнение на
каналите чрез деление на дължината
на вълната
Технологията за плътно уплътнение на
каналите чрез разделяне на дължината на вълната
позната с английската абривиатюра DWDM (Dense
Wavelength Division Multiplexing) се използва за
изграждане на основни кръгово топологични мрежи
познати и като (token ring) в английската литература
при съвременните оператори и доставчици на
Фиг. 1. Топология на хибридна мрежа
интернет. Именно от тези мрежи обхващащи големи
територии доставчика на услугата успява да снабди
изградена от оборудване за плътно уплътнение
крайният потребител директно или използвайки друг на каналите чрез разделяне на дължината на
поддоставчик по известен като (Tier 3 provider). вълната [2]
Основната кръгова топология минава през големи
градове или населени райони, като съответно за
II-65
Потенциални проблеми на мрежите изградени от Data Centers в които има колокирано оборудване на
оборудване за плътно уплътнение на каналите чрез корпоративни клиенти било то фирми или частни
разделяне на дължината на вълната: лица и точки на обмен на мрежови префикси по
известни с английското наименование Internet
- неефективно при малък брой мултиплексирани exchange point (IXP или IX).
канали
- загуби по влакното; освен това спектрална ЛИТЕРАТУРА
зависимост на загубите
- прислушване 1. J. G. Proakis, Digital Communications, 4th
- нелинейни ефекти
Ed., McGraw-Hill, 2000.
- с увеличаването на броя на каналите предавани по
едно влакно се увеличава и общата (сумарната) 2. Cisco Systems, Inc. Introduction to DWDM
енергия на разпространяващото се ОЛ-е. Достига се Technology.
до ниво, при което се проявяват нелинейни ефекти. 3. Ultimate Spectral Efficiency Limits in DWDM
Те зависят и от вълновото разстояние ∆λ между Systems Joseph M. Kahn and Keang-Po Ho.
отделните канали 4. Лекции по Оптоелектроника и оптични
- четиривълново смесване – появяват се комуникации на Техническия Университет-
допълнителни компонентиo София
- стимулирано раманово разсейване – появяват се
допълнителни компоненти
- изисква използването на оптични усилватели [4]
4. Заключение Контакти
Хибридните мрежи изградени по технология за Александър Сарафов
плътно уплътнение на каналите чрез деление на Адрес: България, гр. София, Връбница-
дължината на вълната са бъдещето за свързване на
1 526 вх.Б ет. 4 ап.33
най-големите доставчици на интернет в света (Tier-1)
с мрежите на техните корпоративни клиенти, а
Телефон:+359885125401
именно по-малки разпространители на услугата E-mail: alexander.sarafov@abv.bg,
интернет (Tier-2) и т.н. по топологията до достигане alexander.sarafov@gmail.com
на крайните потребители с информационни центрове
DIPOLE ANTENNA OVER EBG STRUCTURE

FOR UHF RFID APPLICATIONS
ANGEL SLAVOV, PETER Z. PETKOV, STOYAN ILIEV
Abstract—This paper examines the possibility of designing a low profile antenna with
high gain for RFID identification in the frequency range 865 to 868MHz. The advantages
of the suggested structure are high gain and low profile compared to traditional
antennas.
Key Words: EBG, antenna, dipole, gain, RFID
DIPOLE ANTENNA OVER EBG STRUCTURE

FOR UHF RFID APPLICATIONS
ANGEL SLAVOV, PETER Z. PETKOV, STOYAN ILIEV
Abstract—This paper examines the possibility of designing a low profile antenna with
high gain for RFID identification in the frequency range 865 to 868MHz. The advantages
of the suggested structure are high gain and low profile compared to traditional
antennas.
Key Words: EBG, antenna, dipole, gain, RFID
1. Introduction surface varies from 180 ◦ to−180 ◦ versus

Modern antenna technologies have evolved frequency, not only 180 ◦ for the PEC surface or 0 ◦
extremely and rapidly last years. One of the driving for the PMC surface. This reflection phase feature
reason for that evolution is so-called metamaterials. makes EBG unique. For example one possible
In this paper we will focus on a specific part of application of that surface it is usage as a ground
metamaterials namely Electromag- netic Band Gap plane of wire or patch antennas for low profile
structures (EBG). These metamatirials exhibit design which is desirable in many wireless
electromagnetic features which may not exist in communications systems. [3]
nature. These materials have simultaneously < 0 A basic question here is: How effective is
and µ < 0. [1] This type of structure is applicable to the EBG structures for low profile antenna
a wide range of applications in antenna and applications? Mushroom-like is known to have
propagation fields. For instance it can be used as a effective bandgap for a surface-wave propagation.
spiral and curl antenna to achieve low profile It can be used to improve antenna radiation patterns.
design. [2] One of the most interesting properties of When the incident wave is a plane wave (kx 2 + ky
metamaterials is the reflection phase. It is defined as 2 ≤ k0 2 , kz has a real value), the reflection phase
the phase of the reflected electric field at the of the EBG structures varies with frequency. At
reflecting surface compared to the incident one. [3] certain frequency the reflection phase is zero
It is known that a perfect electric conductor (PEC) degree, which resemble the perfect magnetic
has 180 ◦ reflection phase for a normal incident conductor that does not exist in nature. [1]
plane wave. A perfect magnetic conductor (PMC) To understand how the EBG improves
has the reflection phase 0 ◦ but does not exist in antenna parameters, a vector dipole antenna is used.
nature. [1], [3] The antenna consists of two orthogonal dipoles. The
However, EBG structures are more than the advantage of the suggested construction is that the
PMC surface. The reflection phase of the EBG antenna can realize circular polarization which is

II-67
L  0 r h (1)
important for the RFID communications if
both dipoles are used. Here only one dipole is used
 0 (1   r ) g
as radiator, for the sake of construction C cosh 1 ( ) (2)
simplification.  g
The paper focuses on mushroom-like EBG
structures invented by Sievenpiper et al. [4]. The The fractional bandwidth (BW) ±90 ◦ of
structure studied here has a feature of compactness the reflection phase is received by the following
and can be integrated into printed circuit boards, formulas [5].
which is very critical in handheld devices. To prove
that it is necessary to be used the vector dipole 1
antenna, the conventional ground plane and the 0  (1)
LC
EBG structure. Consequently three comparisons on
antenna gain are presented: the vector dipole
antenna, the dipole antenna with metal reflector and  1 L
BW   (2)
the dipole antenna with EBG structure. 0 Z0 C
2. The EBG Structure
3. Results
The mushroom-like EBG structure
displayed here consisted of four elements: metal
patches, a substrate, a ground plane and vias, which 3.1. EBG Structure
connected metal patches and the ground plane. This
For the RFID frequency range 865 MHz to
is illustrated in Fig. 1.
868 MHz the EBG structure is made with finite
dimensions 300×300mm. The substrate material is
FR-4 with thickness 2, 4mm. Accord- ing to
formulas 1, 2 and 3 patch width is approximately w
= 70 mm. Consequently the structure consists of 16
patches - 4×4. To simulate radiation structure
described above commercial software CST
Microwave Studio is used. The final dimensions
after fine tuning are: patch width - 74, 24 mm; gap
width - 0, 76 mm and via radius - 0, 25 mm. The
Фиг. 1. Schematic diagram of the EBG structure EBG surface is shown in Fig. 3.
The reflection phase of the structure is
The parameters in Fig. 1 are: w - width of determined by CST and presented in Fig. 4. As one
the patch; g - patch distance; h - substrate height; r - can see from Fig. 4 the surface has reflection phase
radius via; r = 4, 3. The functionality of the EBG 0 ◦ at 866 MHz which is in desire bandwidth
structure can be explained by a LC equivalent between 865 -868 MHz. In ±90 ◦ the construction
circuit model [1], shown in Fig. 2. has 30 MHz bandwidth from 850 to 880 MHz.
Фиг. 2. Equivalent scheme of the EBG structure
The inductance in that model comes

from the current flowing along adjacent patches
and connecting vias. The capacitance result comes
from the fringing electric field between adjacent
metallic patches. The equivalent circuit model is
able to predict the reflection phase as well as some
surface wave properties. The capacitance and
inductance in the equivalent circuit can be Фиг. 3. The finite EBG structure
approximated by following formulas [5].
II-68
exact thickness of 1 mm. The measured radiation
pattern of the dipole reflector structure is displayed in
Fig.7.
Фиг. 4. The Reflection phase of the EBG structure

Фиг. 6. Return loss of the dipole antenna (no
reflector)
Фиг. 5. The Radiation pattern of the dipole

antenna (no reflector is employed)
Фиг. 7. The Radiation pattern of the dipole
3.2. T h e Dipole antenna antenna over a metal reflector
In this subsection the measurement results of the
dipole antenna are presented. The antenna
dimensions are: width – 104 mm, length – 104
mm, height – 2, 4 mm and substrate – FR-4. For
feeding of the antenna a coaxial cable (50Ω) and
balun mounted on a small PCB circuit in the
middle of the antenna are used. The radiation
pattern of proposed dipole is displayed in Fig. 5,
resulting in gain of 2 dBi. The input return loss is
displayed in Fig. 6.
3.3. The Dipole antenna with metal

Фиг. 8. The Return loss of the dipole antenna over
reflector
ground plain sheet metal reflector
In this subsection the results from the dipole
antenna over a plain sheet reflector are presented. 3.4. D. Dipole antenna with EBG reflector
The distance between the antenna and metal A driving reason for researching the low-
reflector is 1 mm. Later will be shown why a gap is profile (closely spaced reflector) antenna is the
chosen exactly 1 mm. For feeding of the antenna is necessity of compact systems for handheld
selected again balun and coaxial cable (50Ω). applications. To achieve high gain and low profile at
There is a small hole on the metal reflector used to the same time, a parametric analysis is
pass thru the coax feed cable. The dipole is fixed to performed. The results are shown in table I.
the reflector with double sided adhesive tape with

II-69
From table I it can be concluded that a Figure 10 shows the radiation pattern of the
reasonable compromise between high gain and antenna with EBG structure. Also figure 10 shows
low profile antenna can be achieved at 1 mm gain of the antenna is G = 5, 7 dBi. The width of the
reflector distance from the radiating element. When main lobe is 70 ◦ which is a typical angle for RFID-
the distance becomes over 15 mm the efficiency of application antennas. The return loss of the antenna
the EBG surface significantly decreases. is displayed in Fig. 11.
The real construction is shown in Fig. 9. According to Fig. 11 the antenna has good
The final dimensions are: width – 300 mm, impedance matching with reasonable bandwidth
length – 300 mm and height – 5, 8 mm. For which makes antenna insensitive to manufacturing
feeding of the antenna here also are used coaxial tolerances.
cable (50Ω) and balun which is mounted on a small
PCB circuit in the middle of the antenna. Also on
the EBG reflector there is a small pass-thru hole.
The dipole and the EBG reflector are connected
with double sided adhesive tape which has 1 mm
thickness such as on the previous model with plain
reflector.
Фиг. 11. The Impedance matching of the dipole

antenna over EBG structure
4. Conclusion
This paper introduced possible design of
dipole antenna with EBG surface. EBG surfaces
are used to increase antenna gain and decrease the
antenna thickness compared with traditional air
Фиг. 9. The Real model of the antenna gap antennas. All this features give the antenna
possibility for mass production. It is intended
mainly for RFID communications.
Фиг. 10. The Measured radiation pattern of the

dipole antenna over EBG structure
II-70
REFERENCES 5. T.T. Nguyen, D. Kim, S. Kim and J. Jang,
”Design of a wide- band mushroom-like
1. F Yang and Y Rahmat-Sami, Electromagnetic electromagnetic bandgap structure with
Band Gap Structures in Antenna Engineering, magneto- dielectric substrate”, ICITA, 2000.
Cambrigde University Press, 2009.
2. T.H. Liu, W.X. Zhang, M. Zhang and
K.F.Tsang, ”Low profile spiral antenna with Contacts
PBG substrate”, Electron Lett., vol.36, no.9, Angel Slavov
2000, pp. 779 − 780 Technical University of Sofia
3. F Yang and Y Rahmat-Sami, ”Reflection Sofia, Bulgaria
Phase Characterizations of the EBG Ground email: angel.slavov.13@gmail.com
Plane for Low Profile Wire Antenna
Applications”, IEEE Transactions on Antennas Peter Z Petkov
and Propagation, Vol. 51. 10, October 2003, Technical University of Sofia
pp. 2691 - 2703 Sofia, Bulgaria
4. D Sievenpiper, L. Zhang, R. F. J Broas, email: pjpetkov@tu-sofia.bg
G. Alexopolus and E. Yablonovich ”High-
impedance electromagnetic surface with a Stoyan Iliev
forbidden frequency band”, IEEE Trans. Kathrein RFID
Microwave Theory Tech., vol. 47., November Stephanskirchen, Germany
1999, pp. 2059 - 2074 email: s.iliev@kathrein-rfid.de

A REVIEW ON RECENT ANTENNA DESIGNING TECHNIQUES

FOR ELECTROMAGNETIC COMPATIBILITY(EMC) TEST
MR.SARANG M. PATIL, PROF.PETER Z. PETKOV, PROF.BONCHO G. BONEV
Abstract: All devices from electronics category must meet EMC requirement. In this
article presents a review summary of various antenna structure designed for EM signal
radiated emission and susceptibility measurements currently used for evaluating the
electromagnetic compatibility/interference (EMC/EMI) characteristics of electronics
systems and devices. For accurate EMC/EMI test require to capture of unknown
electromagnetic radiation from an equipment using suitably calibrated EMI sensor. The
response of sensors is analyzed in terms of the antenna factor-which is the ratio of the
incident electric field on the antenna surface to the detected voltage at load. At the end
comparison is done of all antennas specifications and new idea is proposed for
improvement in EMC measurement
Key words: Antenna Factor, Electromagnetic compatibility, Electromagnetic Interference
sensor, Immunity.
1. Introduction present in surrounding like computers, ideally

All the electronics devices and appliances computer performs task and during the computation
are important part of our day today life, like kitchen time its generate some signals internally because of
tools to satellite, because of the more demands for electronic circuits within the system will be
wireless devices, the electromagnetic territory has contained and not be radiated, however some
becomes polluted. Electromagnetic interference signals are radiated as EMI because of internal
(EMI) both inter- and intra- device is the well- signals are not contained totally [5].
known "pollutant". Electromagnetic pre-compliance measurement
Electromagnetic compatibility is the applications antenna is not used widely used
member of electrical science which deals with because of large in size and are sensitive to nearby
unintentional propagation, reception and generation reflections and interact with surrounding metal
of electromagnetic energy with reference to the objects .If EMC testing is carried out during the
unwanted reaction that such energy may induce. design phase of product it provides following
Electromagnetic compatibility testing of electronics advantages,
equipment Necessitate to measurement of field  Passing rate of final compliance test is
strength radiated form devices [1].Radiated increased.
emission from equipment are measured and  Retest count is decreased at EMC test
calibrate the field level for immunity test. laboratory.
Ordinarily, antennas with large bandwidth are used  Eliminates surprises late in the design.
for emission measurement and diode based field  Conform that EMC considerations are part of
sensors are widely used for calibrating field level the original.
for immunity test [2].The definition of EMC/EMI as As per the data received for EMC test
per IEEE dictionary as follows [3]. laboratory, 50% submitted products are fail in
Efficient EMC measurement, observes and compliance testing at first attempt. To reduce this
determines the behaviour of electrical apparatus statistic and improve the percentage of passing form
from the prospect of EM radiation with electronic 50% to 90% need to use EMC measurement
circuits [4].In communication system various antenna /Probe at designers/Manufacturers end [6].
electronics equipments are used, at side of A variety of antennas have been designed
transmitter are designed to radiate a specific radio- to help the EMC engineer test quickly over the
frequency, power at selected frequencies to antenna, range from 30MHz to 1GHz, and they can have
antenna radiated signal intentionally but along with quite interesting shapes [7]. A very significant
that another uninitiated signal also transmitted, it is concept about all antennas is that they required
form of EMI. Another non-radiators devices is also calibrating, and their calibration factors must be
taken into consideration in any measurement of
Copyright
Copyright©2017 by Technical
by Technical University-of
University Sofia,Plovdiv
Sofia, Plovdiv branch,
branch,Bulgaria
Bulgaria ISSN Online: 2535-0048
II-72
radiated emissions [8]. Test field calibration is N.Narang et el.[10]. Present A Coplanar
described and analyzed in [9]. Microstrip Antenna as a Dosimetric E-field Probe
for GSM /UMTS Applications. Co-planar
2. Common Antennas Used for EMC Microstrip antenna used as E-field sensor. This
Testing probe is specifically used for Specific Absorption
Now a day various antenna types and Rate (SAR) measurement in mobile
geometry are present, some of them are commonly communication. In this probe used antenna is dual
used for EMC test, they are as below: band Microstrip, its support for the frequency band
i.e. 915MHz and 2.10GHz in Global System for
2.1. Tunable Dipole Antenna Mobile communication frequency. Over all
Electrical field measurement purpose dimensions of antenna is 30mm*100mm in size.
tunable antenna is widely used, this antenna support The Microstrip design is made on a low dielectric
for the range of frequency over 25-1000 MHz substrate, FR4 epoxy of er = 4.4 and thickness t =
Tunable dipole having two types- the resonant 1.6 mm. structure of antenna is see in figure 1.
dipoles and half-wavelength dipoles.
2.2. Biconical Antenna
This antenna is called broadband dipoles
that consist of two conical conductors having a
common axis and vertex. The receiver is connected
at vertex. Biconical antenna operates in the range of
30-300 MHz, this antenna is best for vertical
polarization measurement because of smaller in
size.
2.3. Log-Periodic Antenna
This antenna is called log-periodic because
of structural geometry that its impedance and Fig. 1. Proposed E-field probe [10].
radiation characteristics repeat periodically as
logarithm of frequency. It’s operating range of Mohammed Siddeqet el. [11], A New
frequency from 200-1000MHz Printed Monopole Antenna as EMI Sensor is
proposed, structure of proposed antenna is shown in
2.4. Bilog Antenna
figure 2, Tthis antenna operates in triple band
This antenna is called log-periodic because resonating frequency i.e. 2.6, 3.5, 5.7 GHz.
of structural geometry that its impedance and Proposed antenna overall size of
radiation characteristics repeat periodically as 33.06mm*25mm*1.6, etched on FR-4 epoxy
logarithm of frequency. It’s operating range of substrate, testing result observed like the percentage
frequency from 200-1000MHz bandwidth of 6.9% and 12.2% is observed in the
2.5. Loop Antenna structure at 2.6GHz and 5.7GHz.
The shape of this antenna is like coil and is
highly sensitive for magnetic field but shielded
against electric field. Measurement of
electromagnetic field takes place with electrically
small loop for the range of frequency approximately
20Hz-30MHz.
2.6. Horn Antenna
This antenna is specially used for high
range frequency measurement of
electromagnetic signals. This measure frequency Fig. 2. Proposed E-field probe [10].
above the 1GHz. Range of frequency is 1-
40MHz. Zdenek Kubík et el [12], see in fig.3, has developed
Optimization of electrical properties of parallel
3. literature survey plate antenna for EMC testing, this parallel plate
A number of researchers work on antenna antenna is also called as Strip line antenna or TEM
designing for electromagnetic compatibility test. In cell antenna, for the measurement device under test
this session consider some recants important and is kept within the uniform EM field region.
efficient antenna designs with specifications.
II-73
5GHz.loop dimensions are 3*3mil and it is very
small. Measurement results agree very well with the
expectations from 0. 1 GHz to 3 GHz. Figure 6 is
small loop field probe.
Fig. 3. Proposed parallel plate antenna [12]
Parallel plate antenna designed specifications are:

W = 0.3m, H = 0.4 m, and thickness of plate’s t =
0.0012m, this design support for the range of
frequency from 9KHz to 30MHz. Fig. 6. Proposed Design of Flex Circuit Probe [15]
Amit Kumar Srivastava et el. [13], , first M. F. P. Tartaglia1 et el. [16], discussed the
proposed design of patch antenna is simulated on Metamaterial-based Probe for EMC Measurements,
HFSS. After the simulation simulated module is in this antenna author try to improve the gain,
fabricated, with substrate ‘FR4’ having electric bandwidth and resonance with the supports of
permittivity, εr=4.3 with the height, h= 1.588mm is metamaterials. In this work planer probe used with
used. This antenna operates the 2.45GHz to square format along with 5mm of track width with
3.00GHz.so circular patch is act as good EMI 35 mm of thickness on the o ROGERS R3003
sensor is concluded. Observe dimensions of substrate with 1.5mm of thickness with a full
proposed antenna in figure 4. ground plane shown in figure 7. EMC measurement
can be performed with low cost planer probe is
concluded.
Fig. 4. Proposed circular patch antenna [13]

Fig. 7. Proposed Design of Flex Circuit Probe [15]
S. Capdevila, J. Romeu et el. [14], design of
an Modulated Scatterer Technique (MST) probe H. Lv et el.[17] , see Fig. 8 is the Indoor
antenna to be used in sensing applications, antenna Electromagnetic Testing Antenna Design with
designed with substrate RO4003C with a thickness Better Standing Wave Ratio and Gain is proposed,
of 1.52mm, has been scaled down to accommodate this covers frequency range between 0.14GHz and
the new frequency range (from 868MHz to 6.3GHz so relatively bandwidth is 191.3%
2.45GHz). Figure 5 indicate proposed structure of proposed. The antenna is constructed with dielectric
MST antenna with dimensions in mm. constant of 2.55 and 0. 8 mm thickness of the
dielectric substrate of polytetrafluoroethylene.
Fig. 5. Proposed MST Probe antenna [14]

Fig. 8. Prototype antenna module. [17]
Surbhi Mittal et el. [15], approach to
electrically small loop electrically short field probe Xingyu Zhang et el. [18],this article present
to obtain a flat frequency response using the highly sensitive integrated photonic electromagnetic
combination with active oscilloscope probe. field sensor based on a silicon-organic hybrid
Proposed design supports to operate up to modulator driven by a bowtie antenna shown in
II-74
figure 9, fabrication and characterization of this
compact antenna is analyze. Device capable to
measure electromagnetic power density of
8.4mW/m2, corresponding to a minimum detectable
electric field of 2.5V/m. at 8.4GHZ.
Fig. 11. Proposed antenna Top view [20]
The constructed of two-layer printed circuit board,

Fig. 9. (a) Schematic top view (b) Magnified
substrate material is Rogers RT6002, relative
image of the feed gap region in (a). [18]
dielectric constant εr of 2.94 and a thickness t of
F. H. Wee et el.[19] ,electromagnetic wave 0.762 mm (overall 1.524 mm), After result analysis
detection based on multiband antenna design of DBTSA, concluded that it is a good candidate for
proposed, antenna design is shown in figure 10 to a dual-polarized sensor antenna for applications in
solve the hazard electromagnetic wave detection wireless EMC measurement.
problem a multi-band Microstrip is designed.
Antenna is simulated and analyzed the results like,
efficiency, VSWR, return loss, radiation pattern etc.
Antenna radiates for multiple frequency like ,2.3,
4.5, 4.68, and 5.2GHz.overall dimensions of
antenna is 80mm*70mm with patch
30mm*60.32mm.For mentions dimensions antenna
return loss below -10dB and input impedance equal
to 50ᾩ .
Fig. 12. Proposed BTSA antenna [21]
4. Important parameters for EMC/EMI

Antenna Selection
The most common fundamental parameters
of technical tools, the EMC antenna is used. The
review of the basic parameters and related antenna
parameters are discussed below:
Fig. 10. Proposed Design of Detector antenna [19] 4.1. Antenna Factor
Nisha Gupta et el. [20] has Design of In Electromagnetic, AF is the ratio of the
Printed Log Periodic EMI Sensor, this printed log- electric field strength to the voltage V induced across the
terminal of an antenna. This is linear definition expressed
periodic dipoles called array with linear polarization in volt per meter (V/m)
with desired frequency range. Antenna is printed on
FR-4 substrate with dielectric constant 4.4 and 4.2. Frequency Response
thickness 1.6mm, printed pattern separated by Frequency response is most crucial
dielectric and named as element1 to 5. This 5 parameter in probe characteristics, is defined as the
element dipole array supports the frequencies of range the probe will respond to. Probe should
720 MHz, 1.344 GHz. and 2.64 GHz. Proposed provide flat frequency response over all frequency
dipole array is shown in figure 11. range
Fenghan Lin et el.[21], invented 0.7–20-
GHz Dual-Polarized Bilateral Tapered Slot Antenna 4.3. Sensitivity
for EMC Measurements. DBTSA in constructed Sensitivity is defined as the how small and
with bilateral tapered slot antenna, this maintain RF signal a probe can respond to accurately.
wide operating frequency range from 700MHz to Sensitivity is most important when small RF field
20GHz. Figure 12 indicate Antenna structure, should be measure. It is measured in V/m.
II-75
Table 1. Comparative analysis of 3 -vector Probe
Frequency Range. Antenna

Ref. Simulator Over all
Sensitivity Factor Applications
Paper Used antenna size.
From To (AF)dB/m
No.
Ref.10 915MHz 2.10GHz HFSS 13.0 30mm*100mm 1mV/m 50 dB/m SAR Measurement
33.06mm ×
2.5 to 5.3 to -
Ref.11 HFSS 13.0 25mm ×1. EMI sensor
2.68 ,3.1 6GHz -
6mmm
to 4.1
EMC,EMI Testing of
W=0.3m, GSM cell phones,
Ref.12 9KHz 30MHz - - -
H=0.4m radio and TV
Transmitters
CST
-1.8930 to 41.7375 to
Ref.13 2.45GHz 3GHz Microwave 130*130mm EMI sensor
2.4728V/m 19.6dB/m
Studio
Ref.14 868MHz 2.45GHz - 22*6mm - - EMC measurement
Ref.15 0.1GHz 3GHz - 3x3 mil 3.3mV - EMC measurement
Ref.16 - - - 6cm*5cm - - EMC measurement
electromagnetic
Ref.17 0.14GHz 6.3GHz HFSS 220*111mm - -
detection
Ref.18 8.4GHz 10GHz 12*6mm - electromagnetic field

2.5V/m.
sensor
CST
electromagnetic Wave
Ref.19 2.3GHz 5.2GHz microwave 80*70mm - -
detection
studio 2014.
Array 13.08-
Ref.20 625MHz 2.6GHz IE3D EMI Sensor
Length=206mm 24.24dB/m
CST
wireless
Ref.21 700MHz 20GHz microwave 320*230mm - -
EMC measurements
studio
4.4. Dynamic Range
Dynamic range is the total range of RF field different problems like, Improvement in antenna
coverage a probe will respond to. The greater the factor, complexity of structure, gain, sensitivity etc,
Dynamic range the better a probe is suited to whatever antenna’s currently used as EMI sensor
address test applications that span the gamut from are only detect one or two axial components, but
low to high field strengths accurate measurement of electromagnetic
compatibility the three vector antenna must.
5. Proposed Module
In this proposed solution for EMC/EMI
measurement system, antenna constructed for three
axial field measurements, designed antenna has
wide bandwidth, Good antenna Factor, sensitivity
should be measured in V/m ,small in size along with
multiple polarization capability. Features of
proposed module shown in Fig.21.
1. Conclusion
In this survey paper describes and reviews
the various antenna designing techniques/ methods
with different antenna geometries, Nevertheless,
useful solution are still less and suffer from Fig. 13. Proposed Antenna Features
II-76
Hence, we have proposed three axial Journal of omputational and Applied
antenna geometry for calculating the average 3- Mathematics 270, May 2014.
axisal vector field for EMC measurement. 13. Amit Kumar Srivastava,” Analysis of Circular
Patch Antenna as an Electromagnetic
REFERENCES Interference Sensor,” Conference on Electrical,
Electronics and Computer Science, March
1. Suk-Hwan Choi and Dong-Seok Kim, “Three 2012 IEEE Students.
Axis Isotropic Field Strength Measuring 14. S. Capdevila, “Design of a Small MST Probe
Antenna,”the journal of korean institute of for EM-Field Measurements and Sensing
electromagnetic engineering and science. 2014 Applications,” Antennas and Propagation
sep.; 25(9), 879-885 Society International Symposium (APSURSI),
2. Benjamin Loader and Martin Alexander, 2012 IEEE, 8-14 July 2012.
“development of optical electric field sensor 15. Surbhi Mittal, “Active Probes for Creating H-
for EMC measurement,”www.ieice.org/ Field Probes for Flat Frequency Response,”
proceedings/EMC14. Electromagnetic Compatibility, 2009. EMC
3. IEEE standard dictionary of electrical and 2009. IEEE International Symposium , 17-21
electronics terms,ANS/IEEE Std 100-1977, Aug. 2009.
2nd ed.,Jay, ed, New yorl:wiley 1977. 16. M. F. P. Tartaglia1 et el, “A Metamaterial-
4. J. Leutcher, V. Stekly, E. Blasch, based Probe for EMC Measurements,” PIERS
“Invetsigation of Avionics Power Switch Proceedings, Guangzhou, China, August
Loading versus Aircraft Electromagnetic 25{28, 2014.
Compatibility,” IEEE AESS Magazine, Vol. 17. H. Lv, X. S. Xia, Y. L. Yu, Z. X. Hua, “An
30, Issue 9, pp. 24-34, Sept. 2015 Indoor Electromagnetic Testing Antenna
5. Mark T.MA,and Motohisa Kanda,” A review of Design with Better Standing Wave Ratio and
electromagnetic compatibility/interference Gain,” International Conference on Artificial
measurement methodologies,” procedding of Intelligence and Industrial Engineering (AIIE
the IEEE,Volume:73,Issue:3, page 388- 2015), Published by Atlantis Press,May 2015.
411,yeat-1985. 18. Xingyu Zhang, Amir Hosseini,, “Integrated
6. Mark T.MA, Motohisa Kanda and Myron Photonic Electromagnetic Field Sensor Based
L.Crawford, “A review of Electromagnetic on Broadband Bowtie Antenna Coupled Silicon
compatibility/Interference Measurement Organic Hybrid Modulator,” Journal of
Methodologies,” Proceeding of the IEEE, Vol Lightwave Technology, Volume: 32, Issue: 20,
73, No3, March1985. oct.15, 15 2014.
7. Tim Williams, “What to look for in an EMC 19. F. H.Wee, F. Maleke et el., “electromagnetic
antenna,” Compliance Engineering European. wave detection based on multiband antenna
8. M J Alexander, “Calibration and use of EMC design, ARPN Journal of Engineering and
antennas, “Measurement good practice guide Applied Sciences, ISSN 1819-6608, VOL. 11,
No. 4 from the National Physical Laboratory NO. 8, APRIL 2016.
9. E. L. Bronaugh, J. D. M. Osburn, “ Whole- 20. Nisha Gupta and Md. Anjarul Haque, “Design
vehicle radiated EMI immunity tests in of Printed Log Periodic EMI Sensor,”
automotive EMC: establishing and calibrating international journal of microwave and optical
the test field,” Electromagnetic Compatibility, technology,vol.4 no.4 july 2009.
1992., Eighth International Conference on, 21. Fenghan Lin et el ,” 0.7–20-GHz Dual-
21-24 Sep 1992, Edinburgh, pp. 39-42, 1992. Polarized Bilateral Tapered Slot Antenna for
10. Narang, N., Dubey, S.K., Negi, P.S. et al., “A EMC Measurements,” IEEE transactions on
Coplanar Microstrip ntenna as a Dosimetric electromagnetic compatibility, Page -1271 -
E-field Probe for GSM Frequencies,”MAPAN 1275, Volume: 56 Issue: 6,Dec 2014.
journal of metrology society of india,pp 1-5,jan
2017.
11. Mohammed Siddeq and Dr M. Satyanarayana, Authors’ contacts
“ A new printed monopole antenna as EMI Organization: Faculty of Telecommunication,
sensor,” International Journal of Engineering Technical University of Sofia,
Research & Technology (IJERT), ISSN: 2278- Address: Post Code 1000, boulevard Kliment
0181, Vol. 2 Issue 10, October – 2013. hridski 8,Sofia, Bulgaria
12. Zdeněk Kubík , Denys Nikolayev et el, E-mail: Sarang.p86@gmail.com,
“Optimization of electrical properties of pjpetkov@tu-sofia.bg,
parallel plate antenna for EMC testing,” bbonev@tu-sofia.bg.
КАЧЕСТВО И УДОВЛЕТВОРЕНОСТ ПРИ

ПРЕНОСИМИ КОМПЮТРИ
КРИСТИАН ЦВЕТКОВ1, ВЛАДИСЛАВ ГЕРАСИМОВ1, ЕЛЕНА КОЛЕВА1,2, ЛИЛЯНА КОЛЕВА1
Резюме: Проведено е изследване за удовлетвореността на потребтелите от

преносимите компютри. За характеризиране на гореупоменатата са използвани
индикатори, като ефективност, използваемост, качество и ефикасност.
Анкетираните са разделени в две групи, до и над 36 годишна възраст.
Ключови думи: удовлетвореност, ефикасност, ефективност, качество,
използваемост, потребители, преносими компютри.
QUALITY AND SATISFACTION FOR

PORTABLE COMPUTERS
KRISTIAN CVETKOV1, VLADISLAV GERASIMOV1, ELENA KOLEVA1,2, LILYANA KOLEVA1
Abstract: A study about complacency of portable computers. For characterization on

complacency were used indicators witch efficiency, usability, effectiveness and quality.
Respondents were split into two groups, up to and over 36 years of age.
Key words: satisfaction, efficiency, usability, effectiveness, quality, user, portable
computers.
са с плоски LCD-TFT монитори, вградени

1. Въведение оптични устройства, мрежова карта,
Преносим компютър (портативен безжична комуникация, сензорни екранни,
компютър, ноутбук, нетбук) или често възможност за включване на външни
наричан лаптоп, е компютър с малки размери устройства, които разширяват
и същата функционалност, като персоналния функционалността на компютъра [1].
компютър. Думата лаптоп идва от lap- скут. В днешно време поради своята
Идеята за мобилните устройства от този тип функционалност те се използват широко по
възниква около 70те години на 20 век. През целия свят от милиарди потребители, което
1973 година компанията IBM демонстрира поражда интерес към спецификата, коя от
проекта SCAMP (Special Computer APL техните характеристики е най- важна за
Machine Portable). След две години задоволяване нуждите на клиента.
компанията предлага IBM 5100, който е Удовлетвореността от мобилното устройство
първия преносим компютър. Първоначално е функция, която зависи от ефективността,
те били с 8 битови процесори, работели с ефикасността, използваемостта и не на
5.25 инчови (13.335 см.) флопидискови последно място от качеството на продукта.
устройства и с малки CRT дисплей. С Ефективността характеризира само
развитието на технологиите търсенето на отношението на постигнатия резултат спрямо
тези устройства става все по- голямо, заради поставената цел, тя не отчита разхода бил то
своите известни предимства, едно от които пари, време и други, а само постигнатата цел.
безспорно е тяхната мобилност. Използваемостта изразява честотата
Съвременните модели преносими устройства на ползване на даден продукт.
II-78
Ефективност
1.При покупка на лаптоп от какво

значение
са параметрите и марката на
процесора му?
2.При покупка на 3.При покупка на лаптоп от
лаптоп от какво какво значение е Ефикасност
значение е с колко RAM големината на HARD- 41.До каква степен за вас
- памет разполага? диска?
качеството на продукта се
4.При покупка на лаптоп от 5.При покупка на лаптоп от определя от това платената
какво значение е с каква какво значение е марката 39.За да сте удовлетворен цена да отговаря на
видеокарта разполага? на дъното? от направената покупка полученото качество?
6.При покупка на лаптоп от 7.При покупка на лаптоп от на устройство? 42.До каква степен за вас
какво значение е какво значение е да има 40.До каква степен за вас качеството на продукта се
издръжливостта на инсталирана операционна качеството на продукта се определя от съотношението
батерията му? система? определя от цената? цена-качество?
24.При покупка на лаптоп 31.До каква степен за вас Удовлетвореност

8.Използвате ли Интернет 16.Използването на
от какво значение за Вас е качеството на продукта се за обучение? Интернет разширява ли
цената му? определя от големината на 9. Използвате ли Интернет кръга от хора, с които
25.При покупка на лаптоп за комуникация? редовно общувате?
от какво значение е марката гаранционното време?
32.До каква степен за вас 10.Използвате ли Интернет 17.Използвам Интернет
му?
26.При покупка на лаптоп качеството на продукта се за търсене на информация? основно по работа.
от какво значение е размера определя от това гаранцията 11.Използвате ли Интернет 18.Използвам Интернет за
на монитора? да е валидна и за чужбина? за работа? получаване на информация
27.При покупка на лаптоп 33.До каква степен за вас 12.Използвате ли Интернет за лични въпроси.
от какво значение е теглото качеството на продукта се за общуване чрез
социалните мрежи? 19.Използвам Интернет за
му? определя от материалите, с
13.Използването на забавление ( филми, книги,
28.При покупка на лаптоп които е изработен?
Интернет пречи на онлайн игри и др.)
от какво значение е да се 34.До каква степен за вас
обичайния ми 20.Използвам често
предлага с аксесоари качеството на продукта се
ритъм на живот. електронната си поща.
(чанта, мишка, слушалки и определя от външния му
14.Използването на 21.Когато съм в Интернет,
др.)? вид?
Интернет редуцира е основно, за да съм в
29.При покупка на лаптоп от 35.До каква степен за вас
времето, прекарано на живо социалните мрежи.
какво значение е да има качеството на продукта се
с приятели. 22.Използвам интернет, за
водоустойчива клавиатура? определя от наличието на 15.Изпитвате ли нужда
30.До каква степен за вас допълнителни функции? непрекъснато да имате да пазарувам онлайн дрехи,
качеството на продукта се 36.До каква степен за вас връзка с Интернет? продукти и услуги.
23Поне веднъж в
определя от дългосрочното качеството на продукта се
седмицата поръчвам храна
използване, без повреди? определя от високото ниво
38.До каква степен за вас през Интернет.
на изпълнение на
качеството на продукта се стандартни функции?
определя от спазването на 37.До каква степен за вас Използваемост
международни стандарти качеството на продукта се
(ИСО-стандарт и др.)? определя от страната, в
която е произведен?
Качество
Фиг. 1. Диаграма на Ишикава
Качеството в нашия случай е показател разпространен инструмент за решаване на

определящ изследваната удовлетвореност, управленски задачи, свързани с
който е свързан с характеристиките на управлението. Причинно-следствената
разглежданото устройство, като например диаграма помага не само да елиминира даден
размер, марка, материал на изработка и проблем, но и да се предприемат действия за
други. неговото отстраняване. Причинно-следствена
Ефикасност- показва отношението на диаграма на Ишикава тип „рибена кост“ е
вложените разходи спрямо постигнатия наречена така, заради структурата си, която
резултат. наподобява рибена кост. За целите на
изследването е използвана диаграма на
2. Методика Ишикава, която е разклонена и разделя
Диаграмата на Ишикава е един от причините, които характеризират проблема
класическите примери и най-широко
II-79
на 4 категории: ефективност, ефикасност, 3. Определяне на удовлетвореност-

използваемост и качество [2, 3]. та на потребителите
На Фиг. 1 е представена диаграма на Всеки от отговорите е разпределен в
Ишикава тип рибена кост, на която са петстепенна скала. На всяка оценка се
представени разглежданите показатели по присвоява съответния коефициент, който е в
които ще се съди за удовлетвореността от граница от 0 до 1. В Таблица 1 са
преносимите компютри. Основните представени възможните отговори и
разклонения са: съответните им коефициенти.
- ефективност – свързана е с
хардуерните компоненти изграждащи Таблица 1. Коефициенти на
разглеждания продукт; удовлетвореност
- използваемост- показва, колко често Оценки Да По- Нито Да, По- Не
и по каква причина потребителите използват скоро нито Не скоро
интернет; Да Не
- качество – според ISO то е Коефициент 1 0.75 0.50 0.25 0
съвкупност от свойства и характеристики на
обектите, които определят пригодността им В Таблица 2 са показани отговорите
да удовлетворяват установени и
групирани с помощта на допълнителен
предполагаеми потребности;
критерий, възраст. Потребителите са
- ефикасност – показва колко е
определяща цената, като вложен ресурс за разделени под и над тридесет и шест
удовлетворението на потребителите. годишна възраст с цел по-голяма
На базата на определените чрез информативност. В тях са дадени
диаграмата на Ишикава основни причини за произведенията на съответните коефициенти
удовлетвореността е съставена анкетна карта по броя на анкетираните отговорили със
с 42 въпроса. На диаграмата е показан, като съответния отговор. В последната колона е
подразклонение всеки един от въпросите със дадено нивото на удовлетвореност според
съответния му номер. Анкетираните са хора съответния критерии в проценти. За да се
от всяка възраст. За попълването на анкетата получи тази удовлетвореност са сумирани
анкетираните са използвали петстепенна положителните отговори за даден критерий,
скала на оценяване: Да; По-скоро Да; Нито след което този сбор е разделен на броя на
Да, нито Не; По-скоро Не; Не. При
анкетираните и е умножен по 100 за бъде в
проведеното запитване са събрани 138
валидни анкети, които са обработени за проценти.
целите на статията. Анкетата е проведен
онлайн.
Таблица 2. Отговори от анкетата, групирани по възраст: до 36 г. / над 36 г.

Да По- Нито По- Не Удовлет-
Въпроси скоро Да, скоро вореност
Да нито Не Не (%)
1.При покупка на лаптоп от какво значение са параметрите и
73/33 13/14 3/0 1/0 0/1 95.55/97.91
марката на процесора му?
2.При покупка на лаптоп от какво значение е с колко RAM - памет
68/32 17/14 3/1 2/0 0/1 94.44/95.83
разполага?
3.При покупка на лаптоп от какво значение е големината на
43/23 28/20 14/4 5/0 0/1 78.88/89.58
HARD-диска?
4.При покупка на лаптоп от какво значение е с каква видеокарта
45/23 29/21 13/4 3/0 0/0 82.22/91.66
разполага?
5.При покупка на лаптоп от какво значение е марката на дъното? 28/18 24/20 19/8 15/2 4/0 57.77/79.16
6.При покупка на лаптоп от какво значение е издръжливостта на
39/29 30/10 11/3 7/2 3/4 76.66/81.25
батерията му?
7.При покупка на лаптоп от какво значение е да има инсталирана
11/12 23/17 6/7 13/2 37/10 37.77/60.41
операционна система?
II-80
8.Използвате ли Интернет за обучение? 60/19 21/16 5/10 4/2 0/1 90/72.92

9.Използвате ли Интернет за комуникация? 85/47 4/0 0/0 1/0 0/1 98.88/97.92
10.Използвате ли Интернет за търсене на информация? 90/46 0/1 0/0 0/0 0/1 100/97.92
11.Използвате ли Интернет за работа? 68/31 8/7 5/2 4/3 5/5 84.44/79.16
12.Използвате ли Интернет за общуване чрез социалните мрежи? 76/38 6/4 3/2 3/1 2/3 91.11/87.5
13.Използването на Интернет пречи на обичайния ми ритъм на
7/4 10/2 20/13 25/6 28/23 18.88/12.5
живот.
14.Използването на Интернет редуцира времето, прекарано на
20/5 22/7 11/7 16/12 21/17 46.66/25
живо с приятели.
15.Изпитвате ли нужда непрекъснато да имате връзка с Интернет? 16/15 20/9 14/8 16/7 24/9 40/50
16.Използването на Интернет разширява ли кръга от хора, с които
28/20 20/10 12/7 19/5 11/6 53.33/62.5
редовно общувате?
17.Използвам Интернет основно по работа. 12/7 18/8 24/19 23/5 13/9 33.33/31.25
18.Използвам Интернет за получаване на информация за лични
43/22 25/11 13/9 5/3 4/3 75.55/68.75
въпроси.
19.Използвам Интернет за забавление ( филми, книги, онлайн игри
70/25 13/11 3/4 3/4 1/4 92.2/75
и др.)
20.Използвам често електронната си поща. 54/32 14/7 7/4 4/4 1/1 75.55/81.25
21.Когато съм в Интернет, е основно, за да съм в социалните
12/13 18/1 18/11 23/6 19/7 33.33/50
мрежи ( Facebook, Twitter и др.).
22.Използвам интернет, за да пазарувам онлайн дрехи, продукти и
24/11 14/9 10/7 20/11 22/10 42.22/41.66
услуги.
23.Поне веднъж в седмицата поръчвам храна през Интернет. 5/3 2/1 4/3 12/7 67/34 7.77/8.33
24.При покупка на лаптоп от какво значение за Вас е цената му? 33/11 25/20 25/15 4/0 3/2 64.44/64.58
25.При покупка на лаптоп от какво значение е марката му? 27/15 23/19 29/9 4/2 7/3 55.55/70.83
26.При покупка на лаптоп от какво значение е размера на
33/16 29/18 19/12 6/2 3/0 68.88/70.83
монитора?
27.При покупка на лаптоп от какво значение е теглото му? 10/8 20/13 22/10 22/10 16/7 33.33/43.75
28.При покупка на лаптоп от какво значение е да се предлага с
7/5 20/7 10/10 19/14 34/12 30/25
аксесоари (чанта, мишка, слушалки и др.)?
29.При покупка на лаптоп от какво значение е да има
4/3 18/17 16/3 27/13 25/12 24.44/41.66
водоустойчива клавиатура?
30.До каква степен за вас качеството на продукта се определя от
63/32 21/15 5/0 0/0 1/1 93.33/97.92
дългосрочното използване, без повреди?
20/20 34/17 21/8 7/2 8/1 60/77.08
големината на гаранционното време?
13/10 15/13 17/11 20/5 25/9 31.11/47.92
това гаранцията да е валидна и за чужбина?
38/17 21/18 17/10 10/1 4/2 65.55/72.92
материалите, с които е изработен?
8/8 25/10 22/19 23/4 12/7 36.66/37.5
външния му вид?
19/10 34/19 22/12 9/4 6/3 58.88/60.42
наличието на допълнителни функции?
48/22 27/18 11/6 4/0 0/2 83.33/83.33
високото ниво на изпълнение на стандартни функции?
15/14 24/14 18/12 14/3 19/5 43.33/58.33
страната, в която е произведен?
27/22 33/15 14/6 8/2 8/3 66.66/78.08
спазването на международни стандарти (ИСО-стандарт и др.)?
39.За да сте удовлетворен от направената покупка на устройство? 12/10 26/12 37/17 15/7 0/2 42.22/45.83
16/16 58/16 21/14 10/0 5/2 60/66.66
цената?
48/25 26/16 12/5 2/0 2/2 82.22/85.41
това платената цена да отговаря на полученото качество?
39/20 32/18 11/8 5/1 3/1 78.88/79.16
съотношението цена-качество?
Следващата стъпка е определянето на За всеки от посочените критерии уi се

важността на параметрите, което става по изчислява сума от техните оценки:
метода на Стоянов за определяне на тегловни
коефициенти при многокритериално вземане
на решения, чрез ранжиране на целевите
параметри [4]. и сумата на теглата за всеки целеви
параметър .
II-81
Таблица 3. Тегловни коефициенти, групирани по възраст: до 36 г. / над 36 г.
Vj i Vj i
1 84.5/43.5 1.010/1.015 0.1386/0.1406 22 44.5/24 1.033/1.033 0.0631/0.0630
2 82.75/43 1.013/1.017 0.1391/0.1408 23 11.5/7 1.058/1.056 0.0646/0.0645
3 72.25/40 1.032/1.027 0.1417/0.1423 24 65.25/33.5 1.019/1.021 0.0661/0.0664
4 74/40.75 1.029/1.025 0.1413/0.1419 25 59.75/34.25 1.024/1.020 0.0664/0.0664
5 59.25/37.5 1.056/1.036 0.1450/0.1435 26 65.75/36 1.019/1.017 0.0660/0.0662
6 68.75/38.5 1.039/1.032 0.1426/0.1430 27 41.5/25.25 1.038/1.033 0.0673/0.0672
7 34.5/28.75 1.102/1.066 0.1513/0.1477 28 31.75/18.75 1.046/1.043 0.0678/0.0679
8 79.25/36.5 1.007/1.015 0.0616/0.0620 29 32.25/20.5 1.045/1.040 0.0678/0.0677
9 88.25/47 1.001/1.001 0.0611/0.0611 30 81.25/43.25 1.006/1.007 0.0652/0.0655
10 90/46.75 1/1.001 0.0611/0.0611 31 57.75/37.25 1.025/1.015 0.0665/0.0661
11 77.5/38 1.009/1.013 0.0616/0.0619 32 37.75/26.5 1.041/1.031 0.0673/0.0671
12 82.75/42.25 1.005/1.007 0.0614/0.0615 33 64.75/35.75 1.020/1.018 0.0661/0.0662
13 30.75/13.5 1.043/1.047 0.0637/0.0639 34 43.5/26 1.036/1.032 0.0672/0.0672
14 46/16.75 1.032/1.043 0.0631/0.0637 35 57.75/31.25 1.025/1.024 0.0665/0.0667
15 42/27.5 1.035/1.028 0.0632/0.0627 36 74.75/38.5 1.012/1.014 0.0656/0.0660
16 53.75/32.25 1.026/1.021 0.0627/0.0623 37 45.5/31.25 1.035/1.024 0.0671/0.0667
17 43.25/23.75 1.034/1.033 0.0632/0.0631 38 60.75/36.75 1.023/1.016 0.0663/0.0661
18 69.5/35.5 1.015/1.017 0.0620/0.0621 39 53.75/29.25 1.134/1.130 0.2584/0.2597
19 82/36.25 1.005/1.016 0.0614/0.0620 40 57.5/35 1.120/1.090 0.2552/0.2505
20 69/40.25 1.015/1.010 0.0620/0.0617 41 74/39.5 1.059/1.059 0.2413/0.2434
21 40.25/28.25 1.036/1.027 0.0633/0.0627 42 69.75/37.75 1.075/1.071 0.2449/0.2462
Определянето на тегловните но се вижда също и, че той носи най-малка

коефициенти j става по следната формула: удовлетвореност и на двете групи.
където Vj се определя като:
,
където R е броят на анкетираните хора, а m е
броят на оценяваните критерии.
Получените резултати за сумата на
оценките, тегловните коефициенти и
коефициента Vj , които също са разделени на две
групи от критерия възраст, са дадени в Таблица
3.
4. Анализ на резултатите
Фиг. 2. Ефективност на преносимите
Фиг. 2-5 показват зависимостта между
компютри: „‟ – възраст до 36 г.; „o‟ – възраст
удовлетвореноста във проценти и тегловните
над 36 г.
коефициенти на всеки от основните четири
критерии. Всяка от тях е разделена на четири
квадранта, в които са групирани въпросите.
Квадраните са означение с латинските букви: A,
B, C, D:
- А – висока удовлетвореност и тегловни
коефициенти;
- B – висока удовлетвореност и ниски
тегловни коефициенти;
- C – ниска удовлетвореност и ниски
- D – ниска удовлетвореност и високи
тегловни коефициенти.
Според ефективноста, за потребителите
с най-голямо значение е продуктът да има Фиг. 3. Използваемост на преносимите
инсталирана операционна система, въпрос № 7, компютри: „‟ – възраст до 36 г.; „o‟ – възраст
над 36 г.
II-82
По въпроса за използваемостта с най- 5. Заключение

голямо „тегло“, но с най-малка удовлетвореност Няма съществена разлика между
е въпрос № 23, поне веднъж поръчвам храна параметрите носещи удовлетворениет на
през интернет, а с най-голяма удовлетвореност потребителите под и над 36 годишна възраст. С
са въпроси № 9 и 10 използвате ли интернет за изключение на използваемостта, може да се
търсене на информация и за комуникация. обобщи, че потребителите над разделящата
възраст са по-удовлетворени от изследваните
параметри.
Единствено според основния критерий
качество тези под 36 години са по-
удовлетворени отколкото тези над тази възраст.
1. Тончев Милен, (2015)

http://www.digital.bg/koga-e-sazdaden-
parviqt-laptop-article435090.html
Фиг. 4. Качество на преносимите компютри: 2. Todorova Maria, Koleva Lilyana, Koleva
„‟ – възраст до 36 г.; „o‟ – възраст над 36 г. Elena, Iliev Veselin, (2015). Organization and
optimization of e-learning at UCTM,
При качеството се наблюдава разлика Proceedings international scientific conference
между тегловните коефициенти на различните “Post - crisis management in business”
възрастови групи, като за тези под 36 години с November 27, 2015 Sofia, Bulgaria; Volume
най-голям тегловен коефициент е въпрос № 29, ІІ; 438-446.
важно ли е при покупка на лаптоп, той да има 3. Koleva Lilyana, Koleva Elena, Dzharov
водоустойчива клаватура, а за потребителите Volodya, Todorova Maria, (2015).
над 36 въпрос № 28, устройството да се предлага Investigation of the satisfaction from the
с допълнителни аксесоари. И двата въпроса education in UCTM - SOFIA, Proceedings
носят най-малка удовлетвореност на international scientific conference “Post -
съответните потребители. Най-важно според crisis management in business” November 27,
този критерий е дългосрочното използване без 2015 Sofia, Bulgaria; Volume ІІ; 399-406.
повреди, въпрос № 30. 4. Стоянов С. (2005), Интелигентни методи
за оптимизация. Въведение в
оптимизацията, Химико-технологичен и
металургичен унверситет – София.
Контакти:
Кристиан Цветков1, Владислав
Герасимов1, Елена Колева1,2, Лиляна
Колева1
1
Химико-Технологичен и Металургичен
Университет – София 1756, бул.
“Климент Охридски” № 8
2
Институт по Електроника - БАН -
София 1784, бул. “Цариградско шоссе”
Фиг. 5. Ефикасност на преносимите №72
компютри: триъгълник – възраст до 36 г.; кръг E-mails: kpcvd@mail.bg,
– възраст над 36 г. vladivd123@abv.bg, eligeorg@abv.bg,
sura@abv.bg
Според критерия ефикасност въпрос №
39 е с най-голямо тегло и най-малка
удовлетвореност, а въпрос № 41 е негова
противоположност.
ЕКСПЕРИМЕНТАЛНОИЧИСЛЕНО
ОПРЕДЕЛЯНЕНАКОЕФИЦИЕНТАНА
ДОПЪЛНИТЕЛНИЗАГУБИМЕЖДУШИНИ
ИВАНХАДЖИЕВ,ДИАНМАЛАМОВ,ВАСИЛСПАСОВ
Резюме: В тази работа са проведени експериментални и числени изследвания на

влияниетонаскинефектаиефектаблизостнапаралелноразположенитоководещи
шини. Експерименталните изследвания са направени чрез разработената за тази
целексперименталнауредба.Численитеизследваниясаизвършеничрезразработени
компютърни модели в средата на програмния продукт Comsol. Направено е
сравнениемеждуполученитеексперименталниичислениданнизакоефициента на
допълнителнизагубимеждутоководещишини.
Ключовидуми:коефициентнадопълнителнизагуби,тоководещишини,МКЕ
DEFININGTHECOEFFICIENTOFADDITIONAL
LOSSESBETWEENBUSBARS
EXPERIMENTALLYANDNUMERICALLY
IVANHADZHIEV,DIANMALAMOV,VASILSPASOV
Abstract: Experimental tests and numerical study of the influence of both skin and
proximityeffectsonparallelcurrentcarryingbusbarshavebeencarriedoutandtheresults
havebeendescribedinthispaper.Theexperimentshavebeenconductedbymeansofan
experimental installation, specially developed for the purpose. The numerical study has
been performed with the help of computer models, developed in the software product
Comsol. A comparison has been made between the obtained experimental and numerical
dataofthecoefficientofadditionallossesbetweencurrentcarryingbusbars.
Keywords:coefficientofadditionallosses,currentcarryingbusbars,FEM
1.Introduction magneticfieldintheconductorshasbeencalculated
It is known that when alternative current bythefiniteelementmethod.Thenumericalresults
flowsthroughcurrentcarryingelements,additional have been presented in a graphical format and
losses occur due to skin and proximity effects [1]. comparedwithanalyticalones.Paper[7]presentsa
These losses depend in a complex way on the study of the skin and proximity effects in case of
dimensionsofthecurrentcarryingelements,onthe using Ushaped busbars in terms of their
frequency of the current, on the properties of the dimensions and location, as well as vs. the
material,etc.Itmakestheirdirectcalculationquite frequency.
difficult.Theskineffectisdescribedbydifferential Thispaperdescribesbothexperimentaltests
equations [2], [3], [4]. Analytical solution is only and a numerical study of the coefficient of
possible for simpler cases, such as the ones, additional losses between currentcarrying busbars.
presentedin[5].Papers[3]and[4]considertheskin A comparison between the obtained experimental
effectinamassivebusbarandgiveanapproximate andnumericalresultshasbeenmade.
analytical solution. An electric device has been 
shown in [2] for measuring the flux density, while 
in [6] the skin effect has been discussed and the 
 
Copyright
 © 2017 by Technical University of Sofia, Plovdiv branch, Bulgaria ISSN Online: 2535-0048
II-84
2.Busbarlocationvariants
Table1presentsthevariantsoflocationof
the busbars, for whichthe coefficient of additional
losseshasbeenstudied.
 Typesoflocationofthe
investigatedbusbars
Variant1

a)

Variant2
 
 b)
3.Studyingthecoefficientofadditional Electriccircuitoftheexperimental
lossesbetweencurrentcarryingbusbars installationforstudyingofthecoefficientof
experimentally additionallossesat:а)–currentsinthesame
Aspecialexperimentalinstallationhasbeen direction;b)–currentsinoppositedirections;
developed for the purpose of studying the Qs–circuitbreaker;АТ–autotransformer;LT–
coefficient of additional losses experimentally, as loadtransformer;A–ammeter;V–voltmeter.
showninFig.1.
The coefficient of additional losses is
definedaccordingtotheformula:
P∼ I 2 .R∼ R∼
 k
= = 2 = , (1)
=P I= .R =R
where: P~ are the losses per unit of length at
alternating current with r.m.s. value I; P= are the
losses per unit of length at direct current I of the
same magnitude; R~ and R= are the resistances of
the busbars at alternating and direct current
 respectively.
TheresistancesofthebusbarsR~andR=are
Experimentalinstallationforstudyingthe definedaccordingtothefollowingformulae:
coefficientofadditionallosses:stationarypart;
movablepart;aluminiumbusbarswith U
dimensions[60x6]mm;terminalsformeasuring  R∼ = ; (2)
I∼
thevoltagedrop;Ktypethermocouple;–fixer.
The electrical circuit of the experimental U

 R= = . (3)
installationisshowninFig.2. I=
HereUisthevoltagedropofthebusbars.


II-85
3.1.Experimentalresultsforthecoefficient
ofadditionallosses
Thecoefficientofadditionallosseshasbeen
defined by means of the experimental installation
described above. Fig.3 and Fig.4 illustrate the
dependence of the coefficient of additional losses
onthedistancebetweenthebusbarsforvariant1in
case of currents in the same direction and in
oppositedirections.

Dependenceofthecoefficientofadditional
lossesonthedistancebetweenthebusbarsfor
variant2andcurrentsinthesamedirection.

Dependenceofthecoefficientofadditional

Dependenceofthecoefficientofadditional
variant2andcurrentsinoppositedirections.
4.Numericalstudyofthecoefficientof
additionallossesbetweencurrent
carryingbusbars
4.1.Mathematicalmodel
 The mathematical model of a quasi
Dependenceofthecoefficientofadditional stationaryelectromagneticfieldisdescribedbythe
lossesonthedistancebetweenthebusbarsfor Helmholtzequationinacomplextype[8]:
variant1andcurrentsinoppositedirections. . . .
 ∇ 2 A  jωσ A =
− Je ,  (4)
Fig.5 and Fig.6 show the dependence of
the coefficient of additional losses on the distance where: A isthemagneticvectorpotential; ω  isthe
betweenthebusbarsforvariant2incaseofcurrents
angular frequency;  is the magnetic permeability;
inthesamedirectionandoppositedirections.
 σ  is the specific electrical conductivity; Je  is the
current density of external sources. The second
memberoftheequationreflectstheeddycurrents.
The solution of the electromagnetic
problem is given at a predetermined boundary

 

II-86
conditionalongtheborderofthebufferzoneofthe
type:
 nxA = 0.  (5)
The losses per unit of length at alternating

currentareobtainedfromtheelectromagneticfield 
problemsolutionaccordingtotheformula: Distributionofthemagneticfluxdensity
B[T]forvariant1andcurrentsinthesame
 P∼ = ∫∫ J.EdS. 
S
(6) direction.
The losses per unit of length at direct

currentarecalculatedbytheformula:
I2
 P= = , (7)
σS

where I is the current through the busbar; S is the Distributionofthemagneticfluxdensity
crosssectionofthebusbar. B[T]forvariant1andcurrentsinopposite
The coefficient of additional losses is directions.
definedbytheformula:
P∼
 k= . (8)
P=
4.2.Analysisbythefiniteelementmethod
Numerical models for defining the
characteristics of the electromagnetic field for the
variantspresentedinTable1atfrequencyof50Hz
have been created by the software product
COMSOL[9].
Fig.7showsthemeshoffiniteelements.

Distributionofthemagneticfluxdensity
B[T]forvariant2andcurrentsinthesame
direction.
 
Meshoffiniteelements. Distributionofthemagneticfluxdensity
B[T]forvariant2andcurrentsinopposite
Fig.8, Fig.9, Fig.10 and Fig.11 illustrate directions.
the distribution of the magnetic flux density for
variant1andvariant2atdimensionsx=60mm,y= The obtained distribution of the current
6mmandz=20mm. densityinthebusbarsforvariant1andvariant2at
II-87
dimensionsx=60mm,y =6mmandz=20mmis
showninFig.12andFig.13.

a)
 
a)   b)
 Distributionofthespecificlosses[W/m3]
b) forvariant2at:a)currentsinthesamedirection;
b)currentsinoppositedirections.
Distributionofcurrentdensity[A/m2]for
variant1at:a)currentsinthesamedirection;b) 5.Comparisonbetweentheobtained
currentsinoppositedirections. numericalandexperimentalresults
Fig.16, Fig.17, Fig.18, and Fig.19 show
the graphic dependencies of the coefficient of
additional losses on the distance between the
busbars obtained from both the experimental tests
(curve1)andthenumericalstudy(curve2).
 
a)   b)
Distributionofcurrentdensity[A/m2]for
variant2at:a)currentsinthesamedirection;b)
currentsinoppositedirections.
The obtained distribution of the specific 

losses in the busbars for the different variants at
Dependenceofthecoefficientofadditional
dimensionsx=60mm,y =6mmandz=20mmis
showninFig.14andFig.15.

a)

b)
Distributionofthespecificlosses[W/m3] 
forvariant1at:a)currentsinthesamedirection;
Dependenceofthecoefficientofadditional
b)currentsinoppositedirections.
variant1andcurrentsinoppositedirections.

 

II-88
for studying the coefficient of additional losses
numericallyandobtainingarraysofdatainorderto
beabletodefinethiscoefficientdirectly.

REFERENCES

1. Alexandrov,A.(2002).Electricalapparatuses
 part I (Electromagnetic and thermal
problems),Heronpress,Sofia(inBulgarian).
2. Kosek, M., Truhlar, M., Richter, A. (2011).
Skineffect in massive conductors at technical
frequencies,Przeglad Elektrotechniczny
 (Electrical Review), ISSN 00332097, R. 87
Dependenceofthecoefficientofadditional NR5.
lossesonthedistancebetweenthebusbarsfor 3. Kosek, M., Truhlar, M., Richter, A. (2012).
variant2andcurrentsinthesamedirection. Skineffect in conductor of rectangular cross
section – approximate solution, Przeglad
Elektrotechniczny (Electrical Review), ISSN
00332097,R.88NR7a.
4. Gerling, D., Approximate analytical
calculation of the skin effect in rectangular
wires, University of Federal Defense Munich,
85579Neubiberg,Germany.
5. Hanka, L. (1975). Theory of electromagnetic
field,SNTL,Prague.(inCzech).
6. Greconici, M., Madescu, G., Mot, M. (2010).
Skin effect analysisina free space conductor,
Elec. Energ. vol. 23, No. 2, pp. 207215,
Serbia.
 7. Lažetic, К., Prša, M., Ðuric, N. (2009).
Proximity effect against skin effect in two
Dependenceofthecoefficientofadditional coupled U–shaped busbars, Elec. Energ. vol.
lossesonthedistancebetweenthebusbarsfor 22,no.3,pp.293303,Serbia.
variant2andcurrentsinoppositedirections. 8. Yatchev I., Marinova, I. (2011). Numerical
methods and modeling of circuits and fields,
6.Conclusion
Partone,Sofia.(inBulgarian).
Based on the results from the numerical 9. COMSOLVersion4.2User’guide.(2011).
study and the experimental tests carried out, the
followingconclusionscanbemade:
thepatternofchangeofthecoefficientof AssistantProf.IvanHadzhiev,Ph.D.
additional losses considerably depends on the DepartmentofElectricalEngineering
distancebetweenthebusbars; Email:hadzhiev_tu@abv.bg
the coefficient of additional losses for 
variant1decreaseswiththeincreaseinthedistance Assoc.Prof.DianMalamov,Ph.D.
between the busbars, regardless of the direction of DepartmentofElectricalEngineering
thecurrent; Email:deanmalamov@abv.bg
the coefficient of additional losses for 
variant2decreaseswiththeincreaseinthedistance Assoc.Prof.VasilSpasov,Ph.D.
between the busbars at currents in the same DepartmentofElectricalEngineering
direction, while at currents in different directions Email:vasilspasov@yahoo.com
thesituationisjusttheopposite; 
thecoefficientofadditionallosseshasthe TechnicalUniversitySofia,
highestvaluewhenthebusbarsarepositionedasin BranchPlovdiv
variant 1 and in case of currents in opposite 25TsankoDyustabanovStr.
directions. 4000Plovdiv,Bulgaria
The difference between the obtained Telephonenumber:+35932659686
numericalandexperimentalresultsis7%.Itallows 
STUDY OF TWO TYPES OF SENSORS OF

STATIC FORCES - A PIEZOELECTRIC SENSOR
AND A PIEZOELECTRIC ELASTOMER SENSOR
NIKOLA GEORGIEV
Abstract: The present paper studies a piezoelectric sensor (PS) and a combined
piezoelectric elastomer sensor (CPES), consisting of a piezoelectric sensor and an electro-
conductive elastomer sensor, both sensors measuring static and slowly changing forces.
The piezoelectric sensor works in a resonance mode, since higher level of sensitivity is thus
achieved. Both sensors have been modeled and the influence of the measured static force
on the mechanical resistance of the piezoelectric sensor has been defined, as well as on its
output voltage. Thence the influence of the measured static force on the resistance of the
electro-conductive elastomer sensor has also been determined. Models have been obtained
for both sensors in Simulink in MATLAB environment. The models have been
experimentally tested.
Key words: combined sensors, piezoelectric, elastomer, Matlab, Simulink model, static
force
1. Introduction piezoelectric sensor and a sensor with an electro-

conductive elastomer. The considered sensors
measure static and slowly changing forces, and the
A great number of mechanical quantities such as piezoelectric sensor works in a resonant mode. The
mass, force, acceleration, momentum, deformation electro-conductive elastomer sensor is represented
and angular velocity are measured by means of by a resistor Rr,, by means of which its volume
piezoelectric transformers [1, 2, 3]. They work both resistance is expressed. Modeling of the PS and
in static and dynamic modes. Usually by means of CPES is carried out and the influence of the
this type of sensors static or slowly changing forces measured static force on the mechanical resistance
are measured and the measured force is applied to of the metal rod of the piezoelectric sensor and
the piezoelectric transformer itself or to a metal thence on its output voltage is defined. The
element, to which piezoelectric elements are influence of the measured static forces on the
attached [2]. resistance of the electro-conductive elastomer
Piezoelectric resonant sensors are also used in sensor is registered.
robotics to measure distances, touch, and static or
dynamic forces in resonance mode [2]. 2. Exposition
Electrically conductive elastomer sensors are a
special class of tactile sensors of force, marked by
The piezoelectric sensor (PS) of force, studied
their compactness and low cost.
here, is composed of two piezoelectric plates 2 and
The most frequently used construction of tactile
3, stuck on a metal (brass) rod 1, fig 1. The
sensors with electro-conductive rubbers (elasto- measured static force is applied to the metal rod and
mers) is the sandwich type (fig.1). The elastomer
this allows for increasing the range of the measured
material in it 5 is placed between two metal (brass)
forces considerably.
electrodes 1 and 6. The measured static force is
applied perpendicular to the sensor, and the volume
resistance of the electro-conductive rubber changes
in result [4].
This paper studies a piezoelectric sensor and a
combined sensor, which, in turn, consists of a
II-90
V0(t). Thus the measured static force influences both

4 Vpt(t) sensors, first reducing the output voltage of the
8 2 F 3 7 piezoelectric senor, and then reducing the voltage
VG(t)
Vi(t)
P G P
on the electro-conductive sensor. This allows for
R
increasing the level of sensitivity of the combined
V
Rr 1 Vo(t) 9
piezoelectric elastomer sensor significantly.
5
6 ii(t) Cp Rm Lm Cm n2:1 io(t)

1:n1 Rp Lp
Fig.1. Experimental installation of the CPES R
im(t) i2(t)
Ce1 Tr2 Ce2 Vpt(t)
Fi(t) Rr
Vi(t) Tr1 ic(t)
The applied static force F can be expressed by VG(t) Vo(t) V
the concentrated mass M-4. The sinusoidal
generator 5 supplies electrical energy to the input of
the piezoelectric plate of the PS-2, where it is Fig.2. Equivalent electro-mechanical circuit of
the CPES
converted into mechanical oscillations due to the
reverse piezoelectric effect. Since the generator is
Fig. 2 presents an electro-mechanical circuit of
set up for resonant frequency ω0 for РS, the
the CPES in which the equivalent electrical and
mechanical oscillations reach their maximum and
mechanical quantities are denoted as follows:
are in the form of standing waves in the metal rod 1.
Vi(t),Vpt(t) - input and output electrical voltages on
The mechanical oscillations are converted again
the PS;
into electrical signal due to the forward
Vo(t) - output electrical voltage for the CPES;
piezoelectric effect in the output piezoelectric plate
iо(t) - output electric current of the PS;
3. The sensor with the electro-conductive elastomer
n1, n2 - coefficients of electro-mechanical
consists of two brass electrodes 1 and 6 and an
conversion at the input and output of
electro-conductive rubber 5 between them. The
the PS;
measured static force changes the volume resistance
F1(t) - input mechanical force for the PS;
of the rubber Rr, and this is registered by means of
im(t) - mechanical current (oscillatory speed).
the voltmeter 9.
3.Modeling The piezoelectric plates, used both for the

purposes of modeling and for the experimental tests,
Fig. 2 shows the electro-mechanical circuit of are rectangular in shape, with dimensions: length
the PS, containing electrical input and output parts, Lp=8.10-3 m, width Wp=15.10-3 m and thickness
presented by its capacitors Ce1, Ce2. A mechanical Tp=1.10-3 m and are made of piezoelectric ceramic
part is located between them, which comprises the material PZT4m having the following parameters:
mechanical resistances of the piezoelectric plates piezoelectric module d31=123.10-12 C.N-1, Young’s
Rp, Lp, Cp, as well as those of the metal rod Rm , Lm, modulus YE1p =0,83.1011 N.m-2, density p=7600
Cm. Energy transformation from electrical to kg.m-3, permittivity εT33=11,5.10-9 F.m-1, elastic
mechanical and vice versa is presented by the ideal constant SE11=11,7.10-12 m2.N-1, longitudinal
transformers Tr1 и Tr2. Similar electro-mechanical frequency constant N=1600 Hz.m and quality factor
circuit for a PS is considered in [2]. The active of the piezoelectric ceramics QMp=500. The metal
resistance R together with the resistance, presenting rod is made of brass with Young’s modulus YE1m
the sensor with the electro-conductive elastomer Rr =0,98.1011 N.m-2, density m=8500 kg.m-3 and the
are connected to the output of the piezoelectric following dimensions: Lm=7.10-2 m, Wm=15.10-3 m
transformer. Thus the measured static force и Tm=1.10-3 m.
changes: the mechanical resistance of the metal The sinusoidal generator V(t) is with amplitude
resonator and thence the output resistance of the V m=7 V and circular frequency ω0=660991 s-1 equal
piezoelectric transformer Vpt(t); the electrical to the frequency of the PS.
resistance of the electro-conductive sensor Rr and Since the sinusoidal generator is assumed ideal,
thence the voltage of the output voltage divider it can be presented in Fig. 3 by means of the input
voltage of the PS
II-91
VG  t   V1  t  F
. (1) M
g
. (6)
Thus the input voltage V1(t) is also sinusoidal Thus the density of the metal rod m1 can be
and, due to the reverse piezoelectric effect, expressed by the measured concentrated force
expressed by the coefficient of electro-mechanical
transformation n1, it is converted into a sinusoidally  F 
m1 m 1  
changing mechanical force F1(t) in the simplified  gG  (7)
equivalent circuit of the CPES, fig. 3
After substituting (7) into (4) for the mechanical
F1  t   n1 V1  t  resistance, presenting the losses in the metal rod, it
. (2)
is obtained
Rp Lp Cp Rm Lm Cm n2:1 io(t)  F 
 TmWm Y1Em m 1  
R  gG 
im(t) i2(t) Rm 
Tr2 Ce2 Vpt(t) 4QM 0
Fi(t) (8)
Rr
ic(t) Vo(t) V
By means of the derived expression the
influence of the measured static force on the
mechanical loss resistance is taken into account for
Fig. 3. Simplified equivalent electro-mechanical the PS.
circuit of the CPES The mechanical quality factor QM of the metal
rod of the PS can be obtained from the damping
The mechanical resistances and coefficients of
ratio of the mechanical waves in α, and the circular
ideal transformations for the piezoelectric ceramic
resonant frequency ω0 of the РS
plates are equal to [3]
0
 TpW p  p Y1Ep QM 
n
1 n
2 W p d31 Y1Ep , R
p , 2 . (9)
2QMp
The dependence between the volume resistance
Lp
 TpW p L p  p , C p
L p 2 . of the electro-conductive elastomer and the meas-
2
2 TpW pY1Ep ured force Rr =f(F) is non-linear. The theory of the
(3) elastic properties of bodies [5] is used for its analyt-
The mechanical resistances of the metal rod are ical derivation.
found in the same way With the help of the Hooke’s law the measured
force can be presented as follows
h
 TmWm m1 Y1Em F  S Yr  S Yr S
Rm  , Lm TmWm Lm m , h , (10)
4QM
Lm
Cm  . where:
 2TmWmY1Em S is the cross-section of the electro-conductive
(4)
elastomer;
Metal rod density m1 can be expressed by means Yr - Young’s modulus of the electrically
of the concentrated mass М, as well as by the mass conductive elastomer;
G and the density m of the brass metal rod ε - relative deformation ;
 M h - thickness of the electrically conductive
m1 m 1 
 elastomer.
 G 
. (5) From (10) for the relative deformation of the
electro-conductive elastomer it is obtained
The concentrated mass can be presented by the
concentrated force F and the acceleration of gravity

II-92
F

, (11)Yr S
Taking into account the relative deformation ε
and introducing a corrective coefficient β for the
resistance of the electro-conductive elastomer it is
obtained

 F
Yr S
Rr  Rro e , (12)
By Rro the resistance of the electro-conductive
elastomer is denoted at zero static force F.
With the help of the second law of Kirchhoff for Fig. 4.
the mechanical part of the PS, fig.3, the following
expression is obtained Fig. 5 presents the change of the simulated r.m.s.
dim  t   1 1  output voltage Vpt of the piezoelectric sensor in case
F1  t    R p  Rm  im  t    Lp  Lm      im  t  dt
 
dt  Cp Cm  of changing the measured concentrated force F
(13) within the range from 0 to 10N.
After integrating both members of the equation

and certain transformations, for the mechanical Vpt,
current of the PS it is obtained V
 1 1 
  
1  Rp  Rm   C p Cm
im  t  dt  


    m
im  t   F  t  dt  i  t dtdt
 L p  Lm
1
L p  Lm  
L p  Lm
(14)
The output electric current through the capacitor
can be obtained from the mechanical current and the
electro-mechanical transformation for the output of
the PS
1 F, N
ic  t   im  t 
n2 Fig. 5.
. (15)
Thus the output voltage of the PS can be found With the help of the expressions (8), (14), (16)
and (17) a model of the combined piezoelectric
1 elastomer sensor can be developed in Simulink
 t  V
V pt c t 
. Cе 2
(16)
 i t  dt
c (MATLAB environment), fig. 6.
This output voltage is supplied to the resistive
divider, consisting of R and Rr, whereas for the
output voltage of the combined piezoelectric
elastomer sensor it is obtained
V pt  t 
Vo  t  
R
1
Rr
. (17)
With the help of the expressions (8), (14) and
(16) a Simulink model of the piezoelectric sensor in
MATLAB environment is developed, fig. 4.
Fig. 6.

II-93
Fig. 7 presents the change of the simulated r.m.s. 1,6

output voltage Vо of the Vpt, Vmоd
combined piezoelectric elastomer sensor for the V 1,4
measured concentrated force F within the studied 1,2 Vmeas
range. 1
0,8
0,9 0,6
Vo, Vmоd 0,4
0,8
V Vmeas
0,7 0,2
0,6 0
0,5 0 1 2 3 4 5 6 7 8 9 10
0,4 F, N
0,3
0,2
Fig. 9.
0,1
0 The model in Simulink of the combined piezoe-
0 1 2 3 4 5 6 7 8 9 10 lectric elastomer sensor is also experimentally test-
F, N ed by means of the circuit in fig. 1. The change of
the effective values of the output voltages Vо ob-
Fig. 7. tained both by the model and from experiments as a
function of the measured concentrated force F is
shown in fig. 9.
The voltages obtained in the simulation are
higher both for the piezoelectric sensor and for the
V0, combined piezoelectric elastomer sensor, since the
V sinusoidal generator is assumed ideal, fig. 1.
The maximum relative error for the piezoelectric
sensor is δmax=8,53%, and this shows that the Sim-
ulink model well describes the behavior of the stud-
ied sensor for static forces.
For the combined piezoelectric elastomer sensor
the maximum relative error is δmax=9,33%, which
means that the Simulink model well describes its
behavior as well.
The relative levels of sensitivity measured
within the studied range of the concentrated force F,
F, N are: SF=51% for the piezoelectric sensor; SF=77%
for the piezoelectric elastomer sensor.
Fig. 8. Consequently the relative level of sensitivity of the
combined piezoelectric elastomer sensor
The presented model of the piezoelectric sensor considerably exceeds the one, of the standard
in Simulink is experimentally tested by means of piezoelectric sensor. This is due to the fact that the
the circuit in fig. 1. The change of the effective measured force acts on both parts of this sensor –
values of the output voltages Vpt obtained by the the piezoelectric and the elastomer.
model and from the experiment in function from the
measured concentrated force F is shown in fig. 8. 5.Conclusion
Two types of sensors have been studied: a pie-
zoelectric sensor and a combined piezoelectric elas-
tomer sensor, consisting of a piezoelectric sensor
and an electro-conductive elastomer sensor, both
II-94
sensors measuring static and slowly changing forc-

es.
Both sensors have been modeled and the in-
fluence of the measured static force over their out-
put voltages has been defined.
The models have been tested experimentally and
it has been established that they define well the
behavior of the sensors while measuring static forc-
es.
The relative level of sensitivity of the combined
piezoelectric elastomer sensor considerably exceeds
the one, of the standard piezoelectric sensor and it
suggests the possibility to use it for measuring static
and slowly changing forces.
References
[1] Lavrinenko, V. (1975). Piezoelektricheskie

transformatori. Energia, Moscow.
[2] Trofimov A. I. (1979). Measuring transducer
static loads. 32-45. “Machine building” Moscow.
[3] Kartashev, I. and Marchenko, N. (1978).
Piezoelektricheskie transformatori toka, Tehnika,
Kiev.
[4] Kostova, M. (1993). Electrically Conductive
Elastomeric Composition for Rubber Electrodes.
Copyright Certificate, №2603, RB.
[5] Harper, C. A. and Edward M. P. (2003). Plastics
Materials and Processes. John Wiley & Sons.
Department of Electrical Engineering

Technical University–Sofia, Branch Plovdiv
25 Tsanko Dystabanov St.
4000 Plovdiv
BULGARIA
E-mail: geotek@abv.bg

STUDY OF THE INFLUENCE OF THE PIEZOELECTRIC

CERAMICS PARAMETERS ON THE OUTPUT CURRENT OF A
PIEZOELECTRIC TRANSFORMER, POWERING A SUPER-
CAPACITOR
NIKOLA GEORGIEV
Abstract: A piezoelectric transformer, working in three-resonance mode with two parallel

electric resonances at the input and the output and one mechanical resonance in series has
been considered in this paper. The influence of the parameters of the piezoelectric ceramics
on the magnitude of the output current of a piezoelectric transformer of current in short-
circuit mode has been studied. The obtained expression can serve as informational output
parameter of this type of transformers. The basic output electrical parameters of a number
of piezoelectric ceramic materials (used in practice) have been compared. The times for
charging the super-capacitor, connected to the output of the piezoelectric transformers
have been calculated. Experimental tests have been conducted, confirming the validity of
the obtained expression for the output current in short-circuit mode.
Key words: piezoelectric transformer of current, three-resonance mode, short-circuit
current, super-capacitor
1. Introduction is the output current, obtained from the generator

Traditionally, on-off control of semiconductor section.
devices (thyristors and triacs) is accomplished by The principle of operation of a transverse-
means of optrons and pulse inductive transformers. transverse type of a PTC is explained by fig. 1:
There is a need for additional power supply with the I2 S1
I SC
optrons, and the pulse transformers are 2 3 +
characterized by low workability and a high extent L1
of dissipation. One of the promising directions for V1 V2

CS
V3
eliminating these shortcomings is the use of Е T P P L2
W
piezoelectric transformers of current (PTC). L
[1].They have the following basic advantages: 1 -
compactness, high workability, temporal and

temperature stability, resistance both to electrical Fig.1. Equivalent electro-mechanical circuit of
and mechanical overloads, and capability to work in a PTC
short-circuit mode.
The aim of this paper is to study the influence of Two pairs of silver electrodes 2 and 3 have
the parameters of the piezoelectric ceramic been laid on a piezoelectric ceramic plate 1, i.e., a
materials over the output short-circuit current, PTC has been constructed [2]. In the excitation
which is one of the main electrical parameters of the section 2 of the PTC electrical energy is converted
piezoelectric transformers of current. into mechanical, while in the generation section 3,
mechanical energy is converted into electrical. The
2. Exposition electric resistance of the piezoelectric ceramic plate
is of the order of tens of G and determines the
excellent separation of the two electrical parts of the
It follows from the name of the piezoelectric
PTC.
transformers of current that the most essential
The voltage of the sinusoidal generator is
parameter for this type of piezoelectric transformers
supplied to the excitation section and the electrical
II-96
energy is converted into mechanical oscillations. Consequently, the total mechanical resistance
They propagate into the piezoelectric ceramic plate of the PTC is equal to the mechanical loss
and when they reach the generator section, they are resistance
converted into an electrical signal, which, after Z M  RM . (2)
being rectified, charges the super-capacitor CS.
Since the input and the output electrical parts
If the sinusoidal generator is set at some of the
of the PTC have two parallel resonance circles, then
resonant frequencies of the PTC, then standing
the two equivalent reactive conductivities are equal
mechanical waves arise in it and the output electric
to
voltage has a maximum value.
Inductances L1 and L2 are connected in parallel  1   1 
B1     C1   0 B2     C2   0
both to the input and the output with values,   L    L 
1
, 2
. (3)
allowing for obtaining two parallel (current)
resonance circles with the electrical capacitances Се After reducing the mechanical part of the
of the excitation and the generator sections – fig. 2. circuit to the electric part for a three-resonance
These parallel resonance circles are with frequency, mode [3] it has the form (fig.3):
coinciding with the resonant frequency of the I cs
S
mechanical part of the PTC. Thus three resonance
circles are obtained: two parallel electrical and one Rm
in series (mechanical), with the same resonant 2n 2 CS
frequency – fig. 2. Е
V3
Both sections – the excitation and the generator
one – are expressed by their electrical capacitances
Се, while the electro-mechanical conversion is
expressed by the ideal transformers Тr.1 and Тr.2. Fig.3. Reduced electrical circuit of the PTC for three-
The mechanical part of the PTC is shown by its resonance mode
connected in parallel concentrated mechanical
The mechanical loss resistance RM can be
parameters: active resistance RM – taking into
expressed in the following way:
account the losses in the PTC; inductance LM –
 Z0
defining the mass; and capacitance CM - taking into  RM  , Z0 T W  Y
consideration the elasticity of the PTC [3]. 4QM
, (4)
Before the switch S1 is turned on, the PTC
works in idle-run mode; when it is turned on, the n  n1  n2 Wd31Y
PTC works in a mode, close to a short-circuit mode, ,
since the super-capacitor is not charged. where:
W, T are the width and the thickness of the PTC;
1:n1 Rm Lm Cm n2:1 I
2 S I SC d 31 ,Y,  - the piezoelectric module, Young’s
+
D3 modulus and density of the PTC;
V1 L1
V2
CS QM - the mechanical quality factor;
Tr1 Tr2
Е Ce2 D2 V3 Z0 - the mechanical characteristic
Ce1 L2
- resistance of the PTC;
n1 , n2 - the coefficients of ideal
Fig.2. Equivalent electro-mechanical circuit of
the PTC transformation.
When the excitation and generator sections are By using the reduced electric circuit of the
with the same length, the coefficients of ideal PTC for three-resonance mode (fig. 3), for the r.m.s.
transformation are the same n1= n2=n. value of the output short-circuit current of the PTC
Because of the resonance mode in the it is obtained
mechanical part of the PTC 2 n2 E
I SC 
 1  RM
  LM    0 . (5)
 C M  . (1)

II-97
The coefficient of ideal transformation n and depleted super-capacitor.

the mechanical resistance Rм in (5) are expressed by
the parameters of the piezoelectric ceramics and the Based on the obtained expression for the short-
geometric dimensions of the PTC (4) and it is circuit current, the coefficient of transformation in
obtained idle run mode kU0 and the charging time of the
2
8W QM d31 Y E super-capacitor t, several types of piezoelectric
I SC  ceramics have been analyzed and their parameters
T . (6) presented in Table 1. The main purpose in this case
From [2], Young’s modulus and the velocity of is to determine the most appropriate among them
the sound in the piezoelectric ceramics c1 are equal for producing a PTC. Table 2 illustrates both the
to calculated by (8) the output short-circuit current and
2 T
k31 33 Y by (9) the charging time of the super-capacitor t and
 Y  , c1
d312
 the corresponding measured quantities. The
. (7)
calculations, as well as the measurements, are for a
After substituting by (7) into (6), for the r.m.s. super-capacitor with capacity of 300 F and Е =3V,
value of the output short-circuit current is finally W=25 mm, T=0,5mm.
obtained The coefficient of transformation in idle-run
2
8W QM k31 2
c1  33 E mode kU0 is calculated in the following way [3]
I SC  2
4 QM k31
T . (8) kU 0 
The obtained expression (8) shows the 
 2 1  k31
2
 . (11)
dependence of the r.m.s value of the output short-
circuit current on the parameters of the piezoelectric Table 1.
ceramics (Qм, k31, 33, c1), geometric dimensions Piezo QM k31 T
33 c1
(W,T) and the r.m.s. value of the input electromotive ceramics -9
x10 F/m m/s
force (Е). This expression shows in perceivable
form which of the parameters of the piezoelectric PZT-4 500 0,33 11,5 4560
ceramics exert significant influence and this PZT-5A 75 0,34 15 3860
facilitates the right choice of the needed PZT-4H 65 0,38 30 3900
piezoelectric material; it also illustrates how the
output current in short-circuit mode can be PZT-7A 600 0,3 4 4800
increased by changing the geometric dimensions of PZT-8 1000 0,3 9 4600
the PTC.
The coefficient of transformation in idle-run Table 2.
mode kU0 gives additional information about the
Piezo kU0 Isc. Isc. t, s t,s
output electric parameters of the PTC. It has to be ceramics mA mA calc. meas.
taken into account since in some of the electronic calc. meas
control circuits the PTC initially works in idle run PZT-4 25 658 645 2 181 2 245
mode and then it switches into a mode, close to PZT-5A 4 124 121 1 161 1 231
short circuit. The time for charging the super-
capacitor t can be calculated in the following way PZT-4H 4,5 268 257 5 373 5 448
[4] PZT-7A 24 254 241 5 669 5 809
CS V31  V30  PZT-8 40 918 899 1 760 1 817
t
I , (9)
I The maximum relative error for the output
I  CS
2 , (10) short-circuit currents is δmax=2,1%, while for the
charging times of the super-capacitor it is
where:
δmax=3,1%.
V31 is the voltage of the charged super-capacitor; From the results, given in Table 2 the
V30 - the initial voltage value of the super- following conclusions can be drawn:
capacitor before charging;  The most appropriate piezoelectric ceramic
ICS - the short-circuit current in case of a materials for producing PTCs are PZT-4,
II-98
PZT-8, since the highest short-circuit cur- Department of Electrical Engineering

rents are obtained with them. Technical University–Sofia, Branch Plovdiv
 With these two types of material the short- 25 Tsanko Dystabanov St.
est charging times for the super-capacitors 4000 Plovdiv
are obtained, which is crucial for their prac- BULGARIA
tical application. E-mail: geotek@abv.bg
 The calculated short-circuit current values
are always higher than the measured ones,
since the sinusoidal generator is assumed
ideal;
 The lower short-circuit currents in compari-
son to the calculated ones lead to longer
charging times of the super-capacitors than
the calculated times.
3.Conclusion
A piezoelectric transverse-transverse trans-

former, working in three-resonance mode with two
parallel electrical resonances at its input and a me-
chanical resonance in series at its output has been
considered in this paper.
An expression has been obtained, which takes
into account the influence of the parameters of the
piezoelectric ceramics, the dimensions of the PTC
and the input electromotive force on the magnitude
of the output current of the piezoelectric transform-
er of current in a short-circuit mode.
The main output electrical parameters of prac-
tically used piezoelectric ceramic materials have
been compared.
The times for charging the super-capacitor,
connected at the output of the considered piezoelec-
tric transformers have been calculated.
Experimental studies have been carried out,
which confirm the validity of the obtained expres-
sion for the output current in short-circuit mode.
References:
[1] Magori, V. (1982). Piezo-Zndkoppler. Eektronik

N6, 85-90.
[2] Катц, Г.В. (1964). Магнитые и диэлектрчес-
кие приборы. , Energia, Moscow.
[3] Lavrinenko, V. (1975). Piezoelektricheskie
transformatori. Energia, Moscow.
[4] http://www.illinoiscapacitor.com/

LOAD AND CONTROL CHARACTERISTICS OF

A SERIES RESONANT DC-DC CONVERTER
WITH A SYMMETRICAL CONTROLLED
RECTIFIER
ANGEL LICHEV, ALEKSANDAR VUCHEV, NIKOLAY BANKOV, YASEN MADANKOV
Abstract: A series resonant DC-DC converter operating at frequency higher than the
resonant one is examined. A symmetrical controlled rectifier is used, which allows a phase
shift control. On the base of a first harmonic analysis, the processes in the converter are
studied. As a result, expressions for basic parameters are obtained and the load
characteristics are built.
Key words: series resonant dc-dc converter, controlled rectifier, ZVS, phase-shift control
1. Introduction identical bridge inverter stages, resonant tank circuit

The well-known series resonant DC-DC (L, С), matching transformer Tr, capacitive input
converter operating at frequency higher than the and output filters (Cd and C0), snubber capacitors
resonant one has a number of advantages: zero C1÷C8 by which a ZVS is obtained.
voltage switching (ZVS), small size and low
weight, natural short circuit protection and so on [1,
2, 3]. However, it does not allow a bidirectional
power transfer between the power supply source
and the load to be achieved. This disadvantage can
be avoided by using a symmetrical controlled
rectifier [1, 4, 5] combined with a phase shift
control method [2, 3, 6]. Thereby, the output power
varies from zero to the maximum value.
Even though, resonant DC-DC converters
with unidirectional power flow are thoroughly
described in literature, only very few is found on
bidirectional resonant converters. The reasons might
be the higher converter complexity and the
additional components needed [7].
In [8], a first harmonic analysis of a
bidirectional series resonant DC-DC converter
operating above resonant frequency is presented and
controlled by the conduction time of the output
stage transistors. As a result, expressions for basic Fig. 1. Circuit of the Examined Resonant
parameters of the converter are obtained. DC/DC Converter
The current paper presents sequel of the
theoretical examinations achieved in [8]. Its purpose The „input” stage (transistors Q1÷Q4 with
is both the load and control characteristics of the freewheeling diodes D1÷D4) operate as an inverter
bidirectional series resonant DC-DC converter to be and the „output” stage (transistors Q5÷Q8 with
obtained with phase-shift control and operation at freewheeling diodes D5÷D8) – as a rectifier.
constant frequency above the resonant one. Voltages Ud and U0 are applied to the input and the
output terminals, respectively.
2. Principle of the Converter Operation The operation of the converter is illustrated
The circuit of the examined bidirectional by the waveforms, presented in Fig. 2. Two
converter is presented in Fig. 1. It consists of two operating modes are known – DIRECT MODE and
Copyright©2017
Copyright 2016byby Technical
Technical University
University - Sofia,
of Sofia, Plovdiv
Plovdiv branch,
branch, Bulgaria
II-100
REVERCE MODE. In the first mode, it is assumed Q5, Q7 (Q6, Q8) conduct longer than its diodes D5,
that energy flows from the “inverter” to the D7 (D6, D8).
“rectifier” (from the source of voltage Ud to the one
of voltage U0). In the second mode (which is 3. Results of the Converter Operation
possible because of the converter is reversible) – the Analysis
energy flows from the “rectifier” to the “inverter” The examination of the converter is carried
(from U0 to Ud). out with harmonic analysis. The impact of an only
the first harmonic is taken into account.
For the purposes of the analysis, the
following assumptions are made: the matching
transformer is ideal with a transformation ratio k, all
the circuit elements are ideal, the influence of the
snubber capacitors and the pulsation of the supply
voltage Ud and the output voltage U0 are neglected,
i.e. the voltages ua and ub have rectangular shape.
On the basis of the assumptions made, the
resonant frequency, the characteristic impedance
and frequency detuning of the resonant circuit are:
0  1 / LC
;  0  L / C ;   S / 0
In accordance with the chosen method of

analysis, it is assumed that only the first harmonics
of the current iL and the voltages ua and ub have
impact in the examined circuit. This gives the
opportunity the processes in the converter to be
illustrated by means of vector charts. (Fig. 3). The
aforementioned operating modes are realized by the
variation of the angle α. In DIRECT MODE, it is
assumed that α varies in the range 0 ≤ α ≤ π/2,
(Fig.3a). In REVERCE MODE, the control
parameters vary in the range π/2 ≤ α ≤ π (Fig.3b).
Fig. 2. Waveforms of the Basic Voltages and
Currents
U a 1
The converter operates at constant
φ
frequency ωs which is higher than the resonant ω0. α δ
Therefore, the transistors pairs of the “inverter” Q1, I kU b 1
1 jXI1 jXI1
Q3 and Q2, Q4 operate at ZVS. The transistors of the kU b 1  α
“rectifier” also operate at ZVS. When the current iL
passes through zero, Q5 and Q7 or Q6 and Q8 begin δ U a 1
φ
to conduct. This pair switches off after time,
corresponding to an angle α (as described in [8]).
а) б)
As a result, the voltage ub is shifted in time from ua.
In this way, the control of the output power
Fig. 3. Vector Diagrams of the Basic Voltages and
is obtained by the variation of angle α
Currents
When φ > π/2, than the conduction time of
the inverter reverse diodes D1, D3 (D2, D4) is more The normalized values of the output voltage
than that of the transistors Q1, Q3 (Q2, Q4). In this U 0 and the output current I 0 are determined as
case, the energy transferred back to the power
follows:
supply is more than the consumed one and the
I k
average value of the current Id is negative. U 0  kU0 U d и I 0  0 (1)
The output voltage U0 cannot change its U d 0
sign. This is due to the rectifier diodes operation. As
a result, when α > π/2, the energy transferred back The output characteristic U 0  f I 0 ;  of
from the load is more than the consumed for the the converter is:
rectifier and the average value of the current I0 is
negative. This means that the rectifier transistors
II-101
Expression for the normalized average
U 0  cI 0 tg  1  c 2 I 0 2 (2)
current through the “rectifier” transistors is obtained
as follows:
When I 0  0 , the sign is positive and the
energy is transferred to the load. When I 0  0 , the  I 0 
sign is negative and the energy is transferred back  
 2 cos 
to the power supply source.  
I QR (8)
According to (2), the output current can 2 1  cos 
vary in range, defined by a minimum and a
maximum value: The normalized current through the
“rectifiers” diodes is:
 1/ c  I 0  1/ c (3)
 I 0 
 
From the expression it is observed that, the  2 cos 
possible extreme values of the output current  
I DR (9)
2 1  cos 
dependent only on the frequency detuning of the
resonant circuit.
The expression for the normalized capacitor
On the basis of (2) the control
peak voltage is:
characteristics of the converter are obtained. It is
derived that:
I 0
U ' CM   (10)
R0 2 cos
U 0  , (4)
c 2  c  tg  R0 
2
The normalized RMS value of the current
through the inductor is:
where R0  U 0 / I 0 is the normalized value of the
load resistor. I 0
I ' EFF  (11)
The direction of the output voltage U0 is 2 2 cos
uniquely determined. Therefore the above equation
does not give enough information about the possible 4. Load and Control Characteristics
operation modes of the converter. So, it is more For the purposes of the design of the
convenient the following expression to be used: converter, its load and control characteristics have
to be known.
1 Fig. 4 presents the output characteristic of
I 0   (5)
the converter. They are obtained (on the basis of
c  c  tg  R0 
2 2
equation (2)) at frequency detuning ν = 1,15 and
transformation ratio k = 1.
On the basis of the first harmonic analysis
presented above the following expression for the
normalized current through the “inverter”  =1,15
11/20 U'0 9/20
transistors is obtained: k=1
3/5 3 2/5
 I 0  B A
 
 2 cos 
 
I QI (6) 13/20 2 7/20
2 1  cos 
7/10 3/10
For the normalized average values of the /5 /10 1 /10 /5
currents through the “inverter” diodes is derived
=  =0
that:
C
 I 0 
  -3 -2 -1 0 1 2 I'0
 2 cos 

I DI  (7)
2 1  cos( ) 
Fig. 4. Normalized output characteristics at 0≤α≤π
Copyright  2016 by Technical University - Sofia, Plovdiv branch, Bulgaria

II-102
The output characteristics present the results in the conclusion that at DIRECT MODE the
normalized dependencies of the output voltage U0 stress is mainly on the “inverter” transistors, and at
from the output current I0. They are obtained when REVERCE MODE – on its reverse diodes.
the converter operates at a constant frequency with Fig.6 presents the normalized dependencies
different values of the control angle α. (drawn on the basis of (8)) of the average value of
The output characteristics situated in the the current I'QR through the „rectifier” stage
first quadrant correspond to the DIRECT MODE (0 transistors from the output current I'0. This is
≤ α ≤ π/2), and the others in the second quadrant – to another case when the characteristics for the
the REVERCE MODE (π/2 ≤ α ≤ π). The ordinate is average value of the “rectifier” reverse diodes
the boundary between the two modes. The current I'DR (based on (9)) have the same form as
characteristics are arranged symmetrically with those from Fig. 6 but are mirrored with respect to
respect to it. the ordinate. This results in the conclusion that
From Fig. 4 it is observed that the output obtaining higher output voltage leads to an increase
voltage does not change its polarity and, in all the “rectifier” devices stress.
independently from the energy transfer direction, The calculations show that, I'0 and I'QR
can significantly exceed the input one. Moreover, increase to a certain value, then decrease to the
the output current I0 cannot exceed the value at the limit. However, it is not noticeable on the
output short circuit for α = 0 or α = π respectively. characteristics.
The ZVS operating range boundaries are
pointed out in dotted lines. These are the curves А
(for φ = 0), В (for φ = π) and С (for φ = π/2). The
values of the control angle are chosen in a way to be I'QR  =1,15
B k=1
couples symmetrical to π/2.
1 1
The dependencies of the average value of 3
the current I'QI through the „inverter” stage /20
A
transistors from the output current I'0 (based on (6))
3
/5
are shown in Fig. 5. 

/1
13
0 2


/2
/5
0
0
/2

/10
9
1
I'QI  =1,15 C  /5
2 0
k=1
A 7/2 /10
3
4 = /5
20
9/
-3 -2 -1 0 1 2 I'0
/5
3
2
0
/2
7
2
/10
11/20  Fig. 6. Normalized average current through the

3
3/5
“rectifier” transistors - IQR
1 5
C /
7/10 13/20 = Fig. 7 presents the normalized dependencies
/10 B of the peak value of the capacitor voltage U'CM
-3 -2 -1 0 1 2 I'0 (based on (10)) from the output current I'0. Like the
output characteristics, these are also symmetrical
with respect to the ordinate. Therefore,
independently from the energy flow direction, the
resonant tank elements are similarly loaded for the
Fig. 5. Normalized average current through the same energy quantity. Moreover, it can be observed
“inverter” transistors - IQI that for a random characteristic the capacitor
voltage is always greater than the output one.
On this figure the boundary B falls with the
Here, like in Fig. 5, the variations of I'0 and
abscise axis and other boundary C is parallel to that.
U'CM about the limit values are not noticeable on the
The conducted analyses show that the
characteristics.
characteristics of the average value of the “inverter”
The dependencies of the effective value of
reverse diodes current I'DI (built on the basis of (7))
the current through the inductor I'EFF (drawn on the
have the same form as those from Fig. 5, but they
basis of (11)) from the output current are shown in
are mirrored with respect to the ordinate. This
II-103
Fig. 8. It is easy to be seen that the characteristics With increase of the value of the load
are symmetrical with respect to the ordinate and resistor the range of change of the control parameter
again, independently from the energy flow direction, decreases. As a result the characteristics become
the resonant tank elements are similarly loaded for largely nonlinear.
the same energy quantity. When the output current
I'0 = 0 (for α=π/2) the RMS current can reach very
high values. Here, again, the variations of I'0 and 25
U'CM about the limit values are not noticeable on the I'EFF  =1,15
characteristics. 20
k=1
It could be seen, that the characteristic in
Fig. 7 are similar to these in Fig. 8.
0
11
 A
/2
B
/
15
9
20
3 10  /5
 /5 2
B U'CM  =1,15 A 7/1 13/ 20
k=1 0 20 7/ 3/10
/5 5
/5
=  C =0
10 -3 -2 -1 0 1 2 I'0 3
11 
20
9/
/20
Fig. 8. Normalized RMS current - IEFF


/5
3
13
20
/5
 /
/1 20 /
7 /10

0 5 3 4
/5 C /5 R'0 = 0,1 0,3 0,5 1,0 2,0 ν = 1,15

I'0
9 =
2
-3 -2 -1 0 1 2 I'0
π/4 π/2 3π/2 π α
0
Fig. 7. Normalized resonant capacitor peak voltage

- UCM -2
Normalized control characteristics for the

output current I'0 as a function of the control angle
α.are presented in Fig. 9. They are built on the basis
of equation (5), at frequency detuning ν = 1,15 when
Fig. 9. Control Characteristics of the Converter
the converter operates at several different values of
the load resistor R'0 = 0,1; 0,3; 0,5; 1,0; 2,0. 5. Conclusion
In this case, the abscissa appears to be the
boundary between the two modes of energy A series resonant DC-DC converter,
transfer. From the figure, it is observed that operating above the resonant frequency, is
independently from the value of the load resistor examined. A symmetrical controlled rectifier is
these characteristics have two extreme values. The used. As a result of the first harmonic analysis,
output current gains same values for these points, normalized load and control characteristics of the
which can be determined by expression (3). converter are built. Based on the research, it can be
Fig. 9 shows that from the point of view of concluded that the considered converter is
the converter output power control, the most bidirectional and it can operate without violation of
appropriate interval of variation of the angle α is the the ZVS conditions in wide range of variation of the
one limited by the two extreme values. Moreover, control parameter. It is proven that the converter
the control characteristics can be assumed as linear output voltage can exceed the input one
for a comparatively wide range. independently from the energy transfer direction. It

II-104
is found that the converter power devices stress Device for Control of Energy-Exchange
grows up with the output power increase. between Electrical Generating and/or
The obtained results can be used for further Receiving Systems
examination and design of such converters. 6. Grigorova, Tsv. and Bankov, N. (2008)
Investigation of the Series Resonant Converter
REFERENCES with Secondary-Side Transistor Bridge
Rectifier, Elektrotechnika & Elektronica, №3-
1. Cheron, Y., Foch, H. and Roux, J. (1986) 4, pp. 9-14
Power Transfer Control Methods in High 7. Krismer, F. (2010) Modeling and optimization
Frequency Resonant Converters. In: PCI of bidirectional dual active bridge DC-DC
Proceedings, June 1986, Munich, pp. 92-103. converter topologies. PhD thesis, ETH Zurich
2. Liu, Y. and Sen, P. C. (1996) Source 8. Bankov, N., Vuchev, A. (2015) An Analysis of
Reactance Lossless Switch (SRLS) for Soft- a bidirectional resonant DC/DC Converter. :
Switching Converters with Constant Switching National Forum “ELECTRONICS 2015”, 14-
Frequency. In: IEEE Transaction on Circuits 15 May 2015, Sofia, pp. 110 – 115
and Systems, 1. Fundamental Theory and
Applications, 1996, Vol. 43, pp. 301-312. Authors: Angel Lichev,
3. Rossetto, L. and G. Sipazzi, G. (1988) Series Aleksandar Vuchev,
Resonant Converter with Wide Load Range. Nikolay Bankov,
In: Industry Applications Conference, Thirty- Yasen Madankov
Third IAS Annual Meeting, 1998. IEEE. Vol. 2, Organization: University of Food
pp. 1326-1331. Technologies
4. Dixneuf, D. (1988) Etud d’un variateur de Address: 26 Maritsa Blvd,4000 Plovdiv,
vitesse à résonance pour machine asynchrone BULGARIA
triphasée. Thèse. E-mail: angel_lichev@abv.bg
5. Patent № 4717990, U.S. H04M 7/00; H02J
3/38: Cheron Y., P. Jacob, J. Salesse; Static
LOAD AND CONTROL CHARACTERISTICS OF

A PHASE-SHIFT CONTROLED BIDIRECTIONAL
SERIES RESONANT DC/DC CONVERTER
ANGEL LICHEV, ALEKSANDAR VUCHEV, NIKOLAY BANKOV, YASEN MADANKOV
Abstract: A phase-shift-controlled bidirectional series resonant DC-DC converter

operating at frequency higher than the resonant one is examined. On the base of a first
harmonic analysis, the processes in the converter are studied. As a result, expressions for
basic parameters are obtained and both load and control characteristics are drawn.
Key words: series resonant dc-dc converter, controlled rectifier, operation above resonant
frequency
1. Introduction
The trend of increasing the share of green
energy is remains stable [1]. Therefore, the issues
related to the conversion and accumulation of this
energy gain popularity.
For this purpose, bidirectional resonant DC-
DC converters are used, which can operate at zero
voltage switching (ZVS) [2, 3, 4].
In [5], a bidirectional series resonant DC-
DC converter operating above the resonant
frequency is presented with a phase-shift control.
An examination is carried out with harmonic
analysis. Expressions for the basic parameters of the
converter are obtained.
This paper presents sequel of the theoretical
examinations achieved in [5]. Its purpose is load
and control characteristics of the bidirectional series
resonant DC-DC converter to be obtained with
phase-shift control and operation at constant
frequency above the resonant one. Fig. 1. Circuit of the Bidirectional Resonant
DC/DC Converter
2. Principle of the Converter Operation
The circuit of the examined converter is The operation of the converter is illustrated
presented in Fig. 1. It consists of two identical by the waveforms, presented in Fig. 2. Two
bridge inverter stages, resonant tank circuit (L, С), operating modes are possible. The first of them is
matching transformer Tr, capacitive input and DIRECT MODE. In this mode, it is assumed that
output filters (Cd и C0). Fig. 1 also presents the energy is transferred from the source of voltage Ud
snubber capacitors C1÷C8 by which a zero voltage to the one of voltage U0 (from the “inverter” to the
switching is obtained. “rectifier”). In the second mode – REVERCE
A voltage Ud is applied to the DC terminals MODE, (which is possible because of the converter
of the „input” inverter stage (transistors Q1÷Q4 with is reversible), the energy flows in reverse direction
freewheeling diodes D1÷D4), and a voltage U0 – to – from U0 to Ud.
those of the „output” stage (transistors Q5÷Q8 with The “input” stage generates a voltage ua.
freewheeling diodes D5÷D8). The converter operates at constant frequency ωS,
which is higher than the resonant one ω0. For that
reason, the current iL falls behind the voltage ua at
an angle φ.
II-106
snubber capacitors and the pulsation of the supply
voltage Ud and the output voltage U0 are neglected,
i.e. the voltages ua and ub have rectangular shape.
On the basis of the assumptions made, the
resonant frequency, the characteristic impedance
and frequency detuning of the resonant circuit are:
0  1 / LC
;  0  L / C ;   S / 0
In accordance with the chosen method of

analysis, it is assumed that only the first harmonics
of the current iL and the voltages ua and ub have
impact in the examined circuit. This gives the
opportunity the processes in the converter to be
illustrated with vector charts (Fig. 3). The
aforementioned operating modes are realized by the
variation of the angle δ. In DIRECT MODE, it is
assumed that δ varies in the range 0 ≤ δ ≤ π,
(Fig.3a), and in REVERCE MODE - π ≤ δ ≤ 2π
(Fig.3b).
Fig. 2. Waveforms of the Basic Voltages and U a 1

Currents jXI1
δ
I1 kU b1
The “output” stage generates a voltage ub.
This voltage is shifted from ua at an angle δ (δ= jXI1
α+φ). kU b1 δ

U a 1
Angle φ corresponds to the conduction time I1
of the „input” stage freewheeling diodes, and angle
а) б)
α – to the conduction time of the „output” stage
Fig. 3. Vector Diagrams of the Basic Voltages and
transistors.
Currents
When φ > π/2, the energy transferred back
to the power supply source is more than the The normalized values of the output voltage
consumed one and the average value of the current U0 and the output current I 0 are determined as
Id is negative. In this case the conduction time of the follows:
inverter reverse diodes D1, D3 (D2, D4) is more than
I k
that of the transistors Q1, Q3 (Q2, Q4). U 0  kU0 U d и I 0  0 (1)
Because of the rectifier diodes operation, U d 0
the output voltage U0 cannot change its sign.
Therefore when α > π/2, the energy transferred back After relevant calculations it is obtained
to the load is more than the consumed for the that:
rectifier and the average value of the current I0
becomes negative. That means that the rectifier sin 
I 0  , (2)
transistors Q5, Q7 (Q6, Q8) conduct longer than its c
diodes D5, D7 (D6, D8).
2
3. Results of the Converter Operation where: c    1   .
8
Analysis
The examination of the converter is carried The normalized output voltage is:
out with harmonic analysis. The impact of only the
first harmonic is taken into account. The results of cos
the analysis can be used for engineering U 0  (3)
cos
calculations because they are accurate enough.
For the purposes of the analysis, the According to the expression the boundaries
following assumptions are made: the matching of the ZVS operating mode are obtained for the
transformer is ideal with a transformation ratio k, all “inverter” transistors (at φ = 0 and φ = π) and for
the circuit elements are ideal, the influence of the the “rectifier” transistors (at α = 0 and α = π).
II-107
For (π/2 ≤ δ ≤ 3π/2), the transistors in both 1,15). The output characteristics situated in the first
stages operate at ZVS. quadrant for 0 ≤ δ ≤ π/2 correspond to DIRECT
On the basis of the first harmonic analysis MODE, and the others in the second quadrant when
presented above the following expression for the –π/2 ≤ δ ≤ 0 – to REVERSE MODE. In the area of
normalized current through the “inverter” the output characteristics, the boundaries of ZVS
transistors is obtained - (4): mode are shown with dotted line. The analysis of
the dependencies shows that in this case the
 I 0  converter operation is very limited and practically
  no-load mode is not possible. Apparently, this range
2 cos 
  
I QI (4) of variation of the control parameter δ is not
2 1  cos  recommended for operation of the converter.
For the normalized average values of the
currents through the “inverter” diodes is derived
4
that:
U'0  =1,15
 I 0 
  3
 2 cos 
 
I DI (5)
2 1  cos( )   = =0
2
The expression for the normalized average
 = /10
current through the “rectifier” transistors is:
- /5
 /5
- /10
9 /20
-3 /10
-9/20
1
-2/5
3 /10
2 /5
 I 0 
 
 2 cos   = -  =
 
I QR (6) 0
2 1  cos 
-3 -2 -1 0 1 2 I'0 3
The normalized current through the
“rectifier” diodes is:
Fig. 4. Normalized output characteristics at
–π/2 ≤ δ ≤ +π/2
 I 0 
 
 2 cos 

I DR  (7) Fig. 5 presents the normalized output
2 1  cos 
characteristics of the converter for value of the
control parameter in the range π/2 ≤ δ ≤ 3π/2. The
The expression for the normalized capacitor
ones situated in the first quadrant for π/2 ≤ δ ≤ π
peak voltage is:
correspond to the energy transfer to the load, and
the others in the second quadrant for π ≤ δ ≤ 3π/2 –
I 0
U ' CM   (8) to the mode of energy return back to the power
2 cos source respectively. The ordinate appears to be the
boundary between the two modes. Referring to it,
The normalized values of the RMS current the characteristics for the equally distanced from π
through the inductor are: control angles are situated symmetrically.
Obviously, limitations for the converter operation at
I 0 ZVS are not observed.
I ' EFF  (9)
2 2 cos The output characteristics (Fig. 4 and Fig 5)
show that, independently from the range of
4. Load and Control Characteristics variation of the control parameter, the output
Detailed information of the examined series voltage does not depend on the output current, i. e.
resonant DC-DC converter design can be obtained the examined converter behaves as an ideal current
from its load characteristics. source. Moreover, the output voltage does not
On the basis of (2), the normalized change its polarity and, independently from the
dependencies of the output voltage U'0 from the energy transfer direction, can significantly exceed
output current I'0 are obtained (Fig. 4) when the the input one.
converter operates at a constant frequency (ν =

II-108
of (3) and (5)) through the “input” stage
freewheeling diodes from the output voltage U0 are
4 shown in Fig. 7.
U'0  =1,15
3 3  =1,15 23/20
I'DI
3/2
2 27/20
9/10
29/20
/20
10
9/
11/10
13/10
7/10
3/5
6/5
7/5
4/5
20
17/
1
1
13/20
-3 -2 -1 0 1 2 I'0 3
/2
0
0 1 2 U'0 3
Fig. 5. Normalized output characteristics at

π/2 ≤ δ ≤ 3π/2
In Fig. 6 the normalized dependencies of Fig. 7. Normalized average current through the
the average value of the current I'QI (drawn on the „input” stage freewheeling diodes – I’DI
basis of (3) and (4)) through the “input” stage
transistors from the output voltage U'0 are The variation of angle δ is in the range π/2 ≤
presented. They are obtained for variation of the δ ≤ 3π/2. The characteristics corresponding to the
control parameter in the range π/2 ≤ δ ≤ 3π/2. DIRECT MODE are presented with thick lines, and
those to the REVERCE MODE – with dotted lines.
Here like in Fig. 6, for a wide range of variation of
angle δ, both the current and the voltage increases.
3  =1,15 17/20
In this range, the converter power devices stress is
I'QI
3/2 bigger for greater values of the output voltage. In
0
/1 DIRECT MODE, this is not valid for all values of
13/20 9
2 the control parameter.
20
The analysis of the characteristics in Fig 6
23/
and Fig. 7 shows that the current through the devices
has significant value in no-load mode, fact that is not
1 of minor importance.
27/20 Fig. 8 presents the normalized dependencies
of the average value of the current I'QR (based on (3)
3/2
and (6)) through the „output” inverter stage
0
0 1 2 U'0 3 transistors from the output voltage U'0. Similar
dependencies of the average value of the current I'DR
(built on the base of (3) and (7)) through the
Fig. 6. Normalized average current through the „output” inverter stage freewheeling diodes are
“input” stage tranzistors - I’QI shown in Fig. 9.
The characteristics corresponding to the
The characteristics corresponding to the
DIRECT MODE are presented with thick lines, and
DIRECT MODE are presented with thick lines, and
They are obtained for variation of the control
For a wide range of variation of angle δ,
parameter in the range π/2 ≤ δ ≤ 3π/2. For both
monotonous rise of the current is observed with the
modes, monotonous rise of the currents I'QR and I'DR
increase of the voltage. In this range, the converter
is observed with the increase of U'0. This shows that,
power devices stress is bigger for greater values of
independently from the energy transfer direction, the
the output voltage. However in REVERCE MODE,
converter power devices stress is bigger for greater
this is not valid for all values of the control
values of the output voltage.
parameter.
The normalized dependencies of the
average value of the current I'DI (drawn on the basis
II-109
RMS current value increases at the output voltage
increasing.
3 27/20
Likewise as the capacitor peak voltage in
I'QR 23/20 no-load mode the inductor current has maximal
value which is significant. Moreover, the
3/2
characteristic in Fig. 11 are similar to these in Fig.
2
10.
/10
9
/20 /20
17 13 10 /10, 13/10
1 /2  =1,15
= U'CM /5, 6/5 2 k=1
5 /
8 /10, 11/10 / /2, 3
,7 
/5
 =1,15 3 
0 6
0 1 2 U'0 3
Fig. 8. Normalized average current through the 2

„output” stage transistors – I'QR
0
The normalized dependencies (based on (3) 0 1 2 U'0 3
and (8)) of the peak value of the capacitor voltage
U'CM from the output voltage U'0 are presented in
Fig. 10. It could be seen that obtaining higher output
voltage leads to an increase of the capacitor peak
Fig. 10. Normalized resonant capacitor peak
voltage. Moreover, for a random characteristic U'CM
voltage – U’CM
is always greater than the output one.
Fig. 10 shows that in no-load mode the
capacitor peak voltage has maximal value which is 10 /10, 13/10
significant. I'EFF /5, 6/5
0
8 /10, 11/10
5 /2
/ 3 
,7 0,
/5 /2
3 13/20 3 
I'DR 6 
17/20
/2 4
2
 =1,15
/10 /20 2
9  k=1
23 
/20
27  3
/2
0
1 = U'0
0 1 2 3
 =1,15 Fig. 11. Normalized RMS current – I’EF

0
0 1 2 U'0 3
Fig. 12 presents normalized control
characteristics of the examined converter for several
values of the frequency detuning (ν = 1,08; 1,10;
Fig. 9. Normalized average current through the 1,15; 1,20; 1,30). According to the fact that the
„output” stage freewheeling diodes – I’DR output voltage does not depend on the output
current, they are easily obtained with the
Fig. 11 presents the dependencies of the substitution φ = δ/2 (U'0 = 0). In the range π/2 ≤ δ ≤
normalized RMS value of the current through the 3π/2, the dependencies variate monotonously, and a
inductor I'EFF (built on the basis of (3) and (9)) from significant linear section can be observed. This is
the output voltage U'0. It is possible to see that the

II-110
the recommended control interval. There are not The obtained results can be used for design
limitations for ZVS in this case. of resonant DC-DC converters.
REFERENCES
6
 =1,08
I'0 1. Kramer, W. et al. (2008) Advanced Power
1,1
4 Electronic Interfaces for Distributed Energy
1,15
1,2 Systems. Colorado: National Renewable
2 1,3 Energy Laboratory
 5/4 3/2 7/4 2  2. Alhaddad, K., Cheron, Y., Foch, H. and
0
Rajagopalan, V. (1986) Static and Dynamic
/4 /2 3/4
Analysis of a Series-Resonant Converter
-2
Operating above its Resonant Frequency, PCI
-4 Proceedings, Boston, October 1986, pp. 55-68
3. Cheron, Y., Foch, H. and Roux, J. (1986) Power
-6 Transfer Control Methods in High Frequency
Resonant Converters. In: PCI Proceedings, June
1986, Munich, pp. 92-103.
4. Sivachandran, P., Gomathy, D. and
Fig. 12. Normalized control characteristics Priyadarshini, G. (2016) A Reviw of
Bidirectional Resonant Converter, Middle-East
5. Conclusion Journal of Scientific Research, 24 (3), pp. 877-
The presented output and control 886
characteristics show that the bidirectional series 5. Bankov, N., Vuchev, A. and Vuchev, S. (2015)
resonant DC-DC converter can operate without An Analysis of a Bidirectional Series Resonant
violation of the ZVS conditions in wide range of DC/DC Converter, Annual journal of
variation of the control parameter. It is proven that electronics, Sofia, pp. 228-231
the converter behaves such as an ideal current
source. The converter output voltage can exceed the Authors: Angel Lichev,
input one independently from the energy transfer Aleksandar Vuchev,
direction. It is found that the converter power Nikolay Bankov,
devices stress grows up with the output power Yasen Madankov
increase. Organization: University of Food
With regard to the “inverter”, this is not Technologies
valid. For some values of the control parameter – Address: 26 Maritsa Blvd,4000 Plovdiv,
with the increase of the output voltage, the BULGARIA
converter power devices stress decreases. E-mail: angel_lichev@abv.bg
EXPERIMENTAL ANALYSIS OF THE SUPPLY

VOLTAGE QUALITY OF INDUCTION MOTORS
WITH PWM CONVERTERS
IVAN KOSTOV, BOJIL MIHAYLOV, VASIL SPASOV
Abstract: This paper presents an experimental analysis of the harmonic content of the
supply voltage in induction motor electric drives. The drives are fed by voltage and
frequency converters with autonomous voltage inverters with sinusoidal Pulse Width
Modulation. For the purpose of analysis a laboratory stand is developed for measuring
non-sinusoidal periodic signals. The stand consists of an induction motor, frequency
converter, digital storage osciloscope and harmonics analyzer. The quantitative integral
evaluations of non-sinusoidal periodic voltages are obtained based on the V/f=const and
V=var; f=const control law. The analysis is carried out in Simulink environment by the fast
Fourier transform Analysis Tool in SimPowerSystems Toolbox for Matlab. Conclusions are
drawn about the supply voltage quality and the possible application of converters for the
supply of induction motors in operation modes.
Key words: electromagnetic compatibility, induction motor electric drives, harmonic
spectrum, fast Fourier transform
1. Introduction Matlab. On obtaining the quantitative indicators of

Presently induction motor electric drives the supply voltage quality, their impact on the
based on non-linear semiconductor converters are electrical drive characteristics is evaluated.
widely used in various industrial areas. While The paper is organized as follows. Section 2
offering enhanced opportunities for the control of outlines the harmonic spectrum indicators of the
technological processes, these drives also create supply voltage of the motor. Section 3 focuses on
problems. They generate current and voltage the mathematical analysis of the modulated periodic
harmonics that can reduce the rated torque and signals. The developed laboratory stand and
efficiency, increase heating and magnetic noise of research methodology are given in Section 4. In
motors [1, 2]. For this reason, it is necessary to Section 5 the results from the experimental analysis
determine the supply voltage quality. are presented. Finally, conclusions are drawn in
The induction motor drives use variable Section 6.
voltage converters (VVC) and variable frequency
2. Harmonic spectrum indicators
converters (VFC) with a DC link. Such converters
have numerous capabilities regarding the forming of The phase voltage of the motor in an
the output voltage and its harmonics toward high induction motor (IM) electric drive controlled by a
carrier frequencies due to the implementation of fast semiconductor converter is analysed. The latter is
switching elements. One of the most commonly set to operate as frequency converter and voltage
used methods for modulating the VFC output converter. To evaluate the IM periodic non-
voltage is through Pulse Width Modulation (PWM). sinusoidal supply voltage, the following coefficients
The aim of this paper is to determine the are used [3, 4]: total harmonic distortion coefficient
supply voltage quality of induction motors fed by (kTHD), coefficient of deformation (kD), form
static frequency and voltage converters using PWM. coefficient (kf) and weighted harmonic voltage
For that purpose a laboratory stand of induction factor (kHVF).
motor electric drive is developed and the voltage In addition, the rms values of the stator
harmonic spectrum indicators of the PWM phase voltage (U) and of its first harmonic (U1) are
equipment are determined. The experimental also evaluated.
analysis is carried out by the fast Fourier transform The above coefficients are determined by
Analysis Tool in SimPowerSystems Toolbox for the formulae [5, 6]:

II-112
∞ 4. Laboratory stand and research
∞
U2
∑U 2
∑ methodology
 2
kTHD  ; k HVF   5
. (1) Based on the analysis, a laboratory stand of
U1 U1 induction motor electric drive is developed. It
consists of a static AC to DC converter capable of
U1 U working as VFC and VVC, digital oscilloscope and
kD  ; kf  . (2)
U U AV personal computer. The stand is shown in Fig. 1.
Here UR stands for uncontrolled rectifier and CF for
All odd harmonics not multiple of 3 are DC-link capacitor. The rms value of the supply
included in kHVF in (1). voltage is U=220 V and the frequency is f=50 Hz.
It should be taken into account that for
induction motors, fed by sinusoidal voltage with Osciloscope
Type:
harmonic components, the standard determines the SIGLENT U(t)
ability for loading within the range from 1 to 0.7 of SHS810
the rated power with kHVF varying from 0.030 to

0.115 [7]. Implementation of static PWM converters U,f
increases the steel losses by about 12% [5]. The Control
total increase in the motor losses is by another 3%,

0.5% of wich is in the stator coil, 2% in the rotor 220V UR
CF Inverter
coil and 0.5% are additional losses. This leads to a U,f=var

IM
decrease in efficiency of the general-purpose motors
by about 0.7% (95.3%-94.6%) at rated speed and
load for 3kHz modulation signal. Fig. 1. Schematic of the laboratory stand
3. Mathematical analysis of modulated A full wave bridge rectifier and DC-link are
periodic signals used. The power inverter operates with sinusoidal
The output voltage formed by the PWM is PWM and carrier frequency fc=5(2.5) kHz. The DC
given by [8, 9]: voltage maximum value is VDC= 220 2 V. The
induction motor model has the folowing rated
ua  0,5mUd sint    F Mc  N  . (3) parameters: active power Pr=180 W; speed
nr=1366 rpm; line voltage Vr=380 V; frequency
Here m is the index of modulation; ω=2πf is fr=50 Hz; number of pole pairs zp=2;
the angular frequency of the modulated signal; ψ is efficiency  =64 % and power factor cosφ=0.64.
its phase angle; ωc=2πfc is the carrier angular The T-shaped equivalent circuit parameters
frequency; Ud is the rectified voltage by the DC-link are: stator resistance Rs=55.8 Ω; stator leakage
unit and the residual function F(Mωc ±Nω) equals reactance Ls=0.089 H; rotor resistance Rr=46.6 Ω;
the sum of all high-frequency harmonics. rotor leakage reactance Lr=0.157 H and
Expression (3) contains harmonics that magnetizing reactance Lm=1.367 H.
depend on the carrier frequency. They are The FFT Analysis Tool from the
concentrated around the multiples of M (both even SimPowerSystems ToolBox in Simulink uses
and odd) of the carrier frequency. As shown in Structure with Time format for the measured values
Table 1, the sum M+N is a positive odd integer. of the voltage V(t). The values are stored in the
The most commonly used techniques for Workspace. The discretization time tk is between
evaluating the measured signals include Fast 2µs and 20µs depending on the modulated signal
Fourier Transform (FFT) and specialized software. frequency. The induction motor operates at no-load.
The supply voltage is varied based on the V/f=const
Table 1. Harmonic spectrum control law for the VFC mode and V=var;
of sinusoidal PWM voltage f=50 Hz=const for the VVC mode. A Siglent 1810
digital oscilloscope with 2GB memory is used.
N(M) М
Odd Even 5. Results from the experimental analysis
1 2 ....
M M
Fig. 2 shows the induction motor phase
0 ±1 1ωc 2ωc ±1ω .....
voltage at 30 Hz when supplied by VFC. The values
±2 ±3 1ωc±2ω 2ωc ±3ω .....
±4 ±5 1ωc ±4ω 2ωc ±5ω .....
of the first 33 harmonics are also displayed as a
±6 ±7 1ωc ±6ω 2ωc ±7ω ..... percentage of the first hamonic peak value (V1max).
±8 ±9 1ωc ±8ω 2ωc ±9ω ..... Here V1max=124.2 V and kTHD=91.22 %.
...... ...... ..... ..... .....
II-113
Fundamental (30Hz) = 124.2 , THD= 91.22% 5 kHz) are analyzed for the VVC mode. Tables 4
FFT window: 1 of 4.915 cycles of selected signal
and 5 show the obtained results for the harmonic
8
spectrum indicators of the voltage at no-load.
Mag (% of Fundamental)
200
6 Fundamental (30Hz) = 124.4 , THD= 91.07%
0
40
4
-200
30
2 0.025 0.03 0.035 0.04 0.045 0.05 0.055
Time (s)
20
0
0 200 400 600 800 1000
Frequency (Hz)
10
Fig. 2. VFC output voltage form and values of
the first 33 voltage harmonics at fc=2500 Hz, 0
f=30 Hz 0 5000 10000 15000
Frequency (Hz)
The shape of the motor phase voltage at Fig. 4. VFC output voltage harmonic spectrum at
50 Hz, minimum voltage amplitude V1max = 27.83 V fc=2500 Hz, f=30 Hz
and VVC converter mode is shown in Fig. 3. The
values of the first 20 harmonics expressed as a Fundamental (50Hz) = 27.92 , THD= 265.90%
percentage of the first harmonic peak value are also 100
displayed. Here kTHD = 266.45%.
80
Fundamental (50Hz) = 27.83 , THD= 266.45%
35 FFT window: 1 of 4.096 cycles of selected 60

signal
30 200
40
25 100
20 0 20
-100
15
-200 0
10 0 5000 10000 15000
0.0240.0260.028 0.03 0.0320.0340.0360.038 0.04 0.042 Frequency (Hz)
Time (s)
5 Fig. 5. VVC output voltage harmonic spectrum
0 at fc=2500 Hz, f=50 Hz
0 200 400 600 800 1000
Frequency (Hz)
Table 2. VFC output voltage harmonic
Fig. 3. VVC output voltage shape and values of the spectrum indicators at fc=2500 Hz, no-load
first 20 voltage harmonics at fc=2500 Hz, f=50 Hz
f U U1 kf kTHD kd tk
The phase voltage harmonic spectrum at [Hz] [V] [V] [μs]
30 Hz and VFC mode is shown in Fig. 4. The 10 71 30 1.92 2.11 0.43 20
analysis is carried out up to fmax=15 kHz. It can be 20 96 58 1.51 1.33 0.60 8
seen that that harmonics tend to group around the 30 118 88 1.32 0.91 0.74 8
multiples of M of the carrier frequency. 40 139 118 1.19 0.62 0.85 8
The phase voltage harmonic spectrum at 50 150 136 1.12 0.46 0.91 8
50 Hz, V1max= 27.92 V and VVC mode is shown in 100 153 140 1.11 0.45 0.91 8
Fig. 5. The analysis is also carried out up to
fmax=15 kHz. There is again grouping of harmonics Table 3. VFC output voltage harmonic
around the carrier frequency multiples of M. spectrum indicators at fc=5000 Hz, no-load
To determine the quantitative indicators by
f U U1 kf kTHD kd tk
means of formulae (1) to (2), a set of experiments is
[Hz] [V] [V] [μs ]
carried out. Six modulated frequencies (10, 20, 30, 10 73 30 1.97 2.18 0.42 20
40, 50 and 100 Hz) are analyzed for the VFC mode.
20 96 58 1.51 1.33 0.60 8
Tables 2 and 3 show the obtained results for the 30 118 88 1.32 0.91 0.74 8
harmonics spectrum indicators of the voltage when 40 139 118 1.19 0.62 0.85 4
the motor follows the V/f=const control law. 50 150 136 1.12 0.46 0.91 4
Five voltages (20, 48, 77, 106 and 133 V) at 100 153 139 1.13 0.46 0.91 2
f=50 Hz and two carrier frequencies (2.5 kHz and

II-114
Table 4. VVC output voltage harmonic of M of the carrier frequency for both the VFC and
spectrum indicators at fc=5000 Hz, no-load VVC mode. Despite shifting of the harmonics
f U U1 kf kTHD kd tk towards high carrier frequencies, the weighted
[Hz] [V] [V] [μs] harmonic voltage factor exceeds the limit, set by the
50 57 20 2.08 2.66 0.35 4 Standard IEC 60034-1. To improve the quality of
50 87 48 1.62 1.50 0.55 4 the supply voltage, sinusoidal filters can be
50 112 77 1.44 1.05 0.69 4 implemented on converters outputs.
50 133 106 1.25 0.74 0.80 4 The obtained results show that the total
50 148 133 1.18 0.49 0.90 4 harmonics distortion coefficient and the distortion
coefficient do not significantly vary with the carrier
Table 5. VVC output voltage harmonic frequency, but strongly depend on the voltage. The
spectrum indicators at fc=2500 Hz, no-load values of kTHD, however, are well above the limit,
recommended by the IEEE Std 519-1992.
f U U1 kf kTHD kd tk The developed laboratory stand and the
[Hz] [V] [V] [μs ]
experiments carried out can be used as a foundation
50 57 20 2.08 2.66 0.35 4
for future research of other types of modulation.
50 88 49 1.62 1.50 0.55 4
50 111 77 1.44 1.03 0.70 4 REFERENCES
50 133 106 1.25 0.74 0.80 4
50 148 133 1.18 0.49 0.90 4 1. Bose B. (2002). Modern power electronics and
a.c. drives, 711 p. Prentice Hall.
Tables 6 and 7 present the kHVF values, 2. Rizov, Pl., Sotirov, D., Batashki, St., and
calculated from the experimental data. The first Spasov, R. (2010). Analysys of the harmonic
value takes into account all harmonics with content of e.m.f. in induction generators by
magnitude greater than 5% of the main harmonic. FEM. Proceedings of the Technical University
The second value is obtained by (1). Table 6 is for of Sofia, Vol. 60, Issue 2, pp. 225-234.
VFC at two carrier frequencies. Table 7 presents the 3. Bose B. (2006). Power electronics and motor
kHVF values for the VVC at one carrier frequency. drives - advances and trends, 915 p. A.Press.
4. Rizov Pl., Spasov, R., et al. (2015). The
Table 6. Weighted kHVF coefficient application of finite element method for
for the VFC converter determining the THD in the electromotive
f [Hz] force in a synchronous machine with
fc 10 20 30 40 50 100 permanent magnets under different loads.
[kHz] Proceedings of the 14-th International
kHVF [%]
Conference ELMA 2015, pp. 171–180.
4.9 6.9 5.4 4.4 3.7 5.1
2.5
8.5 10.0 8.5 6.4 8.0 7.6 5. IEC 60034-17:2002 Cage induction motors when
4.3 4.6 4.3 0.8 2.5 3.1 fed from converters – appl. guide, 44 p., 3Ed.
5 6. IEC 60034-25:2007 Guidance for the design
15.9 7.4 6.1 4.5 4.1 5.3
and performance of a.c. motors specifically
Table 7. Weighted kHVF coefficient designed for converter supply, 74 p., 2Ed.
for the VVC converter 7. NEMA MG 1-2009: Motors and Generators,
U1 [V] Section IV, Part 30, 575 p.
fc 8. Krishnan R. (2003). Electric motor drives -
20 48 77 106 133
[kHz] modeling, analysis, and control, 650 p. Singapore.
kHVF [%] 9. Kazmierkowski, M., et al. (2002). Control in
11.5 5.5 9.0 5.6 3.8 power electronics, 518 p. Elsevier Science.
2.5
17.1 17.5 12.3 8.5 5.8
Assoc. Prof. Ivan Kostov, Ph.D.
6. Conclusion Control Systems Department
This paper evaluates the supply voltage E-mail: ijk@tu-plovdiv.bg
quality and its impact on induction motors, fed by Assistant Prof. Bojil Mihaylov
static frequency and voltage converters using PWM. Control Systems Department
A series of experiments are conducted on a
Assoc. Prof. Vasil Spasov, Ph.D.
specially developed laboratory stand. The voltage
from the converters is examined by FFT-analyser in
SimPowerSystems Toolbox of Simulink. Technical University - Sofia, Branch
Based on the results, it can be concluded, Plovdiv, 25 Tsanko Dyustabanov Str.
that harmonics tend to group around the multiples
INVESTIGATION OF ONE LOOP ELECTRIC CIRCUIT FOR

CONTROL OF FLASH LAMP
YANKA IVANOVA, SVETOSLAV IVANOV, ROSSEN BOJILOV
Abstract: This article provides a transient analysis in discharge flash lamp operated by
one-loop high-voltage electric circuit. Lamp is designed to emit a strong non- coherent
exciting radiation for management of Solid Nd: YAG, ruby, or Er: YAG lasers. It is
designed a scheme for impulse control of Xenon flash lamp. For simulation research in
program environment is proposed replacement scheme. The results of simulation studies
are presented.
Key words: impulse control, xenon gas discharge lamp, one-loop electric circuit
1. Introduction automatically synchronized formation of a

Modern laser technology is a combination discharge in discharge xenon flash lamp. Composed
of the physical principle of the laser, the selection of is a model for transient analysis of the scheme and
suitable active environments in combination with engineering design of this type of driver circuits.
electrical and electronic devices for excitation of
these environments and receiving laser radiation. 2. Driver management scheme for xenon
Basic for practical applications in industry, science, flash lamp
in medicine are lasers with condensed active In common pulsed lasers, solid-state and the
substances (liquid and solid), in particular solid- dye lasers, the optical pumping is carried out with
state Nd: YAG, ruby, Er: YAG and the like lasers, flash lamps, [1] having infinitely large initial
as well as its unique ability to adjustment dye lasers resistance, and negative dynamic resistance. The
and lasers with F - centers. The dye lasers in the power of this type lamps usually consists of a
range of 0,57μm to 0,62μm are currently the main capacitor for accumulating energy and for the
tunable sources widely used in lidar technology for discharge, the coil forming the pulse, high-voltage
remote monitoring of the composition and DC power supply and the ignition circuit for the
atmospheric pollution. They also find important ionization of the lamp. The length of the resulting
applications in systems for the separation of high-energy (~ 100 ÷ 300 J) pulse with a steep edge
isotopes, as well as medical applications for the (~ 1 μs) is determined by the type of the controlled
treatment of serious illnesses. Typical of them is laser.
that to obtain sufficient output power and energy
required highly specialized non - coherent - Lamp 2.1. Principal scheme for control of flash
excitation requiring very steep front of excitation lamp
radiation (̱μs). Obtaining of such exciting Principal electrical control circuit of flash
radiation massively available and widely used lamp is shown in Figure 1. In the researched circuit
discharge lamps is not easy technical task. This is used a gas discharge flash lamp filled with xenon
requires discharge across the lamp to be induced by at a relatively low gas pressure of a few thousand
a capacitor or a capacitor battery with a significant Pascals. To provide the necessary electrical charge
capacity (~ μF) charged to a very high voltage in parallel to the power source is switched storage
(~ 10kV). It is thus possible the formation of fast capacitor C1. In order to trigger discharge between
rising edge (1 ÷ 3μs) pulse with a very high electrodes by an additional ignition device
intensity respectively received in a powerful pulse connected to the lamp is fed to pulse which ionizes
of current through the lamp. The result is a sharp the gas in advance [2].
increasing to several times of the output generated The resistance of the lamp decreases and
power. the accumulating capacitor C1 discharges through it
This article presents a practical [3,4]. As a result of the discharge a powerful pulse
implementation of the scheme with a single loop is emitted with duration determined by the

II-116
capacitance of the storage capacitor C1 and the theoretical analysis of voltamper characteristic of
inductance of the discharge circuit L1. the impulse control of a xenon discharge lamp is
done[5].
Fig. 1. Electrical scheme for control of flash lamp Fig. 2. Substitution scheme for the study of
transients in the programming environment NI
The electrical schematic diagram of an Multisim
ignition device includes the following elements: a
transformer T1, capacitor C3, a resistor R3, thyristor It is established that the variation of current as a
D1, DC power source V3 and a generator of control function of applied voltage on the lamp can be
pulses V4. The source of DC voltage V3 = 400 V is experimentally received from the equivalent electric
charging capacitor C3. Upon receipt of a control circuit including a parallel combination of a
pulse from the control system thyristor D1 turns ona capacitor and an active resistance and inductance in
result of which the capacitor C3 discharges through series included, figure 3. The active resistance R is
the primary winding of the transformer. Thus arises defined as the real part of the total resistance of the
the initial ionization of the gas in the lamp and it is lamp.
ready to be incorporated into the basic power supply The established value of the current in the
from the capacitor C1, charged with a voltage V1 = switching power supply of the lamp has been
5kV. In the scheme in question generator pulses V4 expressed by the inclusion of a consistent
generates pulses with TTL - levels. Duration of inductance L.
generated pulse is 26μs. The repetition period is During the arc discharge in the lamp
2ms, which duration corresponds to the frequency positive and negative electric charges are generated.
500 Hz. Because the scheme under consideration Their amount related to a the applied voltage on the
comprising a gas discharge flash lamp, the lamp determines the value of the capacity C in the
inductance of the discharge circuit L1 and the equivalent circuit of the lamp. This replacement
accumulation capacitor C1 is one-loop, ignoring all scheme can be used for a maximum duration of
active losses, excluding the lamp the transients control pulses 40μs.
processes can be described with the following
integro-differential equation:
diL1 1
U C1 (0) L1  K 0 iL1   iL1dt , (1)
dt C1
where:
iL1 is the current in the lamp as a function of
time;
UC1(0) is the voltage to which the capacitor C1
is charged in the initial moment t = 0;
K0 is the impedance of the lamp, which depends
on the geometry of the lamp, as well as the gas Fig.3. Equivalent substituting scheme of the
pressure in it. lamp
2.2. Replacement circuit for simulation For the present research of transient
research processes in one-loop power control scheme of the
For the study of transients the control flash lamp is sufficiently the lamp to be furnished
scheme of the flash lamp it is made a replacement by the included resistor R2 (Fig. 2)
circuit for simulations in programming environment It is known that the active resistance of the
NI Multisim. Constructed scheme is presented in lamp according to the duration of the control pulses
Figure 2. In the scheme the trigger device is can be determined from the formula:
represented by the impulse generator V4 and switch  t  . l
S2.With the aim of producing a replacement circuit R2  t   (2)
diagram for carrying out simulation research a A
II-117
passive components in the control circuit, is shown
where: A is the cross-section of the lamp (cm2); in Figure 4. The current through the lamp is
l is length of the lamp (cm); variable, reaching a maximum value of 956A for
t is 1/3 of the pulse duration (ms); 7μs. The change of the current represents a fading
ρ(t) is the impedance of the plasma (Ω.cm), transition process.
which depends on the impulse duration in time. Figure 5 shows the shape of the voltage on
With R1 is referred to the internal resistance the capacitor C1, switched in parallel to the power
of the voltage source V1. Pulse duration of the source UC1. The supply voltage V1 is 5kV.
source V4, indispensable to ignite the lamp is 26μs. Figure 6 illustrates the transient process of charge
Key used (relay) S2 is a controlled by voltage the capacitor C1 with the high voltage of the power
switching level 4V, making it possible to control of source, after the excitation of the lamp.
the circuitry of source signals with TTL levels. From the graph of the transient processes in
the one-loop electric circuit it is established that the
process charge has a duration of 2ms, like this time
3. Results interval affixes restriction on the maximum
Simulation researches have been done in the frequency of the control impulses.
Multisim environment. The graph of change of a
current IL2 through the flash lamp obtained by
simulation researches with the set values of the
Fig. 4. Graph of change of the current IL2 through the flash lamp
Fig. 5. Graph of the voltage on the capacitor C1

II-118
Fig. 6. Graph of the transient process of charge the capacitor C1
4. Conclusion IEEE International Conference, Page(s):1615-1618

This paper examines a classical solution for Vol.3, 2-5, Sept. 2001.
impulse control of xenon lamp in programming 5. Yanca Kissova, Svetoslav Ivanov. (2014)
environment Multisim, as the results are Analysis and simulation research of scheme for
approaching with sufficient accuracy to practical obtaining high energy pulse excitation of laser with
researches of the scheme. The substitution scheme a discharge lamp. Машиностроене и
allows optimization of transient processes in the машинознание. Година IX, книга 3. ISSN 1312-
flash lamp by changing the values of passive 8612, Издателство на технически университет-
elements included. The results of the simulation Варна, стр. 50-53.
researches allow the evaluation of energy and
frequency characteristics of the control circuit. The
substitution circuit assembly does not include the
excitation lamps of the lamp, such as the impulse
transformer, the thyristor and the passive elements. Contacts :
This is not necessary because this schematic
solution is often applicable in practice, and ready Yanka Ivanova
impulse transformers for different circuit solutions Svetoslav Ivanov
are available on the market. Rossen Bojilov
LITERATURE: Technical University–Sofia, Branch Plovdiv
25 Tsanko Dystabanov St.
1. Beggar R., Louhibi D., Almabouada F., Noukaz 4000 Plovdiv
A. (2008). “Caractérisation expérimentale d’une BULGARIA
lampe flash utilisée pour le pompage d’un laser à E-mail: yankakiss777@abv.bg
solide ”, the 4 th National Seminary on Laser and E-mail: etehsv@gmail.com
Applications, SENALAP’2008, Tizi-Ouzou from 10 E-mail: rossen_chi@abv.bg
to 11 May 2008.
2. Браун М. Токозахранващи устройства.
Техника, София, 1997.
3. Almabouada F.,Louhibi D., Hamici M.,
Hammoum Y., Haddouche A. (2011). Power supply
for xenon flashlamp. SIPP’2011 / UKM Ouargla /
13 - 15 February/Février 2011.
4. Dondon P., Lequertier J. (2001) An integrated
pedagogical approach for a switching power supply
design: example of fly back. Electronics, the 8th
ULTRASONIC SENSOR FOR MEASURING

WATER SPEED
SVETOSLAV IVANOV, YANKA IVANOVA, ROSSEN BOJILOV
Abstract: In the proposed article, an ultrasonic receiver and an ultrasonic transmitter are
designed and tested to measure the rate of water flow in a cylindrical tube. A device
running on time - impulse method of speed measurement, has been investigated. An
analysis of the frequency and selectivity of the receiver has been made. A methodology has
been developed to select the pulse frequency and the frequency of the low frequency
modulation signal. The proposed schematics of the transmitter and the receiver were
studied.
Key words: Ultrasonic flowmetering, signal processing, transit-time flowmeter
1. Introduction
The phenomena arising from ultrasound
propagation in a moving environment are used in
sensors to measure velocity and fluid flow. The
sound velocity c which is a material property value
is the propagation velocity of a sound wave in a
medium. It changes with the density of the
measuring medium. Therefore it is temperature
dependent in liquids and pressure and temperature
dependent in gases [1, 2]. The time changes when
the sound carrier is also in motion, in fact, it is the
Fig.1. One-channel ultrasonic sensor
sum of the sound velocity in the measuring medium
and the measuring medium velocity. This effect is According to the character of the acoustic channel,
utilized in an ultrasonic flowmeter. The advantages the sensors are without refraction of the ultrasound
of ultrasonic sensors are high performance, beam and by refraction by means of special
allowing determining short-term speed fluctuations, elements. In some ultrasonic sensors, the emitters
and lack of additional impedances along the way of and receivers are located in the stream, and other
supplying the liquid with external placement of ultrasonic sensors are located outside the flow. The
transmitters and receivers of ultrasonic waves, acoustic channels of these sensors may be parallel
possibility of control of reversal streams. The to each other but may also be crossed. Virtually all
disadvantage that makes it difficult to use ultrasonic ultrasonic sensors can operate in time-pulse,
sensors to measure velocity and cost is the frequency-pulse and phase mode [1].
significant difference in ultrasound speed (1000-
2000m/s) and medium (1-10m/s). Differences in
velocity and flow rate sensors are based on
measuring the velocity of the ultrasound wave
going upstream and downstream. The main
attributes for which they are classified are: the
number of channels, the nature of the acoustic
channel, the location of the emitters and the
receivers, the mode of operation, the method of
processing the signals. According to the number of
channels, acoustic sensors can be single channel
Fig. 2. Two-channel ultrasonic sensor
(fig.1) and multichannel (fig. 2) [3].
II-120
When working in time-impulse mode, the
principle of operation of ultrasonic sensors is based
on measuring the propagation time of the ultrasound
pulses along the flow and against the flow in the
pipe (fig. 1). The time for passing the ultrasound
pulse from the transmitter A to the receiver B is
given by the expression:
L
t AB  (1)
 C  V .cos  Fig. 3. Construction of the ultrasonic sensor
The velocity of the fluid as a consequence

The time for passing the ultrasound pulse from the of equation 4 can be determined by the formula:
transmitter B to the receiver A is given by the
expression [4]:
T .2.L
L V (5)
tBA  cos .TAB  TBA 
2
, (2)
 C  V .cos 
The magnitude of the angle θ can be
where: C is the ultrasound rate in the water; determined by the rectangular triangle formed by
 is the angle of propagation of the the diameter of the tube D and the direction of the
ultrasound to the pipe axis; sensors L.
V is the speed of the water;
L is the distance between the ultrasonic D D
transmitter and receiver (transducers A and B).  arcsin
 sin 1 (6)
According to equations (1) and (2), since V << C, it L L
can be shown that
Therefore, the water velocity in the pipe V can be
2.L.V .cos definitively determined by the equation:
t t BA  t AB  (3)
C2
T .2.L
Therefore, the velocity of the fluid V can be
V (7)
 D
cos  sin 1  . TAB  TBA 
2
determined by the formula:
 L
C 2 .t
V (4) When designing the ultrasonic sensor, it is
2.L.cos necessary to determine the length of the ultrasound
wave λ which is distributed in the water as a
The main disadvantage of this method of function of the selected frequency f by the formula:
determining the velocity of the fluid is the
discrepancy between the speed averaged in the C
direction of ultrasound and the velocity averaged  (8)
over the cross section of the pipeline. It is possible f
to introduce an empirical coefficient that allows
error exclusion at the expense of this factor only in where: C is the velocity of the ultrasound wave in
narrow range of medium viscosity variations as well the water [6].
as a slight change in the number of Reynolds Re
[5]. Changing the speed of ultrasound in the The time for transmitting the ultrasound wave from
environment, for example, when changing the the emitter to the receiver at the distance labeled L
temperature, also gives rise to a measurement error. is determined by the formula:
2. Transit Time Flow meter L

The principle schematic of the speed sensor t (9)
c
is shown in fig. 3.
The coefficient k, which shows the relationship
between the distance L and the length of the
ultrasonic wave λ can be determined by the formula:
II-121
L the broadcast packet of high frequency pulses are at

k (10) a frequency of 10 kHz with duty cycle D = 50%.
 With the operational amplifier U2, the pulses are
amplified to an amplitude of 10V. For transmitters
Therefore, the reference frequency with and receivers, double ultrasonic receivers and
which the radiated high frequency oscillations are WC75 - 1 transmitters can be used. Figure 5 shows
repeated is determined by the formula: the high frequency pulses emitted by the emitter f,
modulated by the duration of the Frep reference
f pulse. The duration of the PW packet is 50 μs.
f rep  (11) Figure 5a shows the packet of impulses
k generated by the ultrasonic transmitter. The
duration of this packet is determined by the duration
With a pre-selected duty cycle of the
of the reference pulse shown in figure 5b.
reference pulses D, the number of cycles containing
the ultrasound wave is determined by the formula:
2.2. Ultrasonic Receiver
The scheme of the ultrasonic receiver is
f .D shown in fig. 6. The generated signal from piezo
N cycle  (12)
f rep sensor U5 is amplified sequentially by two
operational amplifiers powered by a single-pole
voltage. The resistor divider R1, R3 creates the so-
The duration of the packet of ultrasonic pulses has a
called virtual zero without changing the amplitude
value that is determined by the formula:
of the input signal. In this case, a virtual zero value
equal to ½ of the supply voltage for the operational
N cycle amplifiers is selected. The scheme is designed to
PW  (13)
f amplify frequencies 1MHz. With the D2 diode and
with the capacitor C5 an amplitude detector is
The aim of the research is to design and test realized for the received high frequency pulses. The
an ultrasonic transmitter and an ultrasonic receiver voltage at the detector output goes to the input of
to measure the water velocity in a pipeline by the the analog comparator with a digital output U3A.
direct measurement of the transition time. The With the logic elements U7C and U6B a R  S
pipeline diameter D is 100mm and the distance trigger is realized. When the impulse with response
between the sensors L is 150mm. Schematics of the frequency Frep reaches, after logging in. „1” at the
designed and investigated ultrasonic receiver and output of the comparator, the digital output of the
transmitter are shown in Figures 4 and 5. trigger RX1 detects a registered ultrasonic signal
from the receiver. Figure 7a shows a graph of the
2.1. Ultrasonic transmitter output voltage of the detector when receiving the
The transmitter includes a rectangular pulse ultrasonic signal with frequency 1MHz.
generator with a frequency of 1MHz, implemented
with a Schmitt trigger (DD U3A), fig. 4. The Frep
repercussion pulses that determine the duration of
Frep
R2 V2
1kΩ U1A 15 V 7 8
U2
U3A 3 C2
74HC08N_4V R4 6
R1 8.2pF
2
74HC14N_4V 10.0kΩ OPA640H U4

4 5
10kΩ
C1
530pF
Fig. 4. Scheme of the ultrasonic transmitter

II-122
Fig. 5. Graphs of the generated high frequency ultrasound pulses (a) and the reference pulse defining the
duration of radiation (b).
V2
15 V R5
R7
20kΩ
U2 20kΩ
C1 C3
4 5
R4 U1 C4 D2
2 82pF R6
4 5
2kΩ 82pF 1N916 Detector

82pF 6 2
3 OPA640H 2kΩ 6
3
U5 7 8
OPA640H C5 R10
R2 7 8 D1 10kΩ
20kΩ R1 1N916 82pF
U4A U7C
20kΩ
R3 R8
1MΩ
20kΩ V1
C2 5V 8 U3A RX1
R9 3
U6B
100nF 47kΩ 2
1
Frep
4
MAX907EPA 74HC00D_4V
Fig. 6. Scheme of the ultrasonic receiver
In Fig. 7b shows the reference frequency pulses Frep that start the emission of ultrasound pulses from
the transmitter, as well as inputs to the logic element of the receiver U4A, which allows switching of the
trigger to the output of the receiver.
In Fig. 7c shows the graph of the output impulses of the receiver RX1, which allow for the
calculation of the delay time TAB.
These calculations can be made from a microprocessor control system. Another fundamental task of
the ultrasound sensor management system is to generate pulses with the reference frequency determining the
start and end of the ultrasound pulses.
II-123
Fig. 7. Graphics of the amplitude detector output signals (a), the reference pulses entering the
receiver (b) and the output pulses of the receiver (c).
3. Results and discussion received signal frequency is 51 (34dB) and is kept

The ultrasonic receiver is designed to almost constant for frequencies up to 9 MHz.
receive 1MHz signals. Figure 8 shows the Analog amplifiers are designed to amplify
amplitude-frequency characteristic of the analogue high-frequency signals, with low frequency
part of the receiver. Research has been done in capability being achieved by the transfer capacitors
“Multisim” software. The gain coefficient for the at their inputs.
Fig. 8. Amplitude-frequency characteristic of the ultrasonic receiver.
Fig. 9. Graphs of the spectral density of the noise input at the input of the amplifier and the output
after amplification of the signal

II-124
It is made a study of the spectral density of REFERENCES

the noise generated by the elements in the amplifier
of the ultrasound receiver (fig. 9). The sources of 1. O’Sullivan, I. and W. Wright. (2002)
this noise are: heat noise, contact noise, lungs Ultrasonic Measurement of Gas Flow Using
(Schottky noise) and pulse noise. The graph shows 2. Jan, G., D. and G. be Boer. (2001) The
that the input noise at 1 MHz has a value 4.10-19 Manufacturing of Ultrasonic Gas Flowmeter,
V2/Hz, and after signal gain there is value 1.10-13 Flow Measurement and Instrumentation 12,
V2/Hz. 89-99.
Therefore, the analog amplifier circuit 3. Cannizo, M., M. Dell’Isola, M. Diritti. (1997).
suppresses signals with a frequency lower than Measurement of high-pressure natural gas flow
1MHz. With the arrival of each reference impulse at using ultrasonic flowmeters, Measurement Vol.
the input of the ultrasonic receiver the output signal 20 No. 2, , 75-89.
4. Gai, H., M.S. Beck and R.S. Flemons. (1989).
of the R  S trigger is reset and it is prepared to
An Integral Ultrasound Transducer/Pipe
read the delay time of the ultrasound signal passing Structure for Flow Imaging, Ultrasonics
through the water TAB. Symposium, 1077-1082.
4. Conclusion 5. Yoder, J., B. Liptak. (2003). Instrument
engineers’ handbook: Process Measurement
The selected mode for measuring the water
and Analysis, Fourth Edition, Volume I., 353-
velocity in a pipeline is the time - impulse method.
361.
This method allows measurement of the velocity of
6. Асенов, А. (2007). Измерване на разход на
fluids in tubes up to 6m in diameter. Designed are
флуиди. София.
ultrasound transmitter and receiver models with
good resistance to low frequency sound
signals.Generation of high frequency pulses and
reference frequency is realized with the appropriate Contacts :
choice of microprocessor control system. The
microprocessor control system can determine the Svetoslav Ivanov
water velocity in the pipeline by solving equation Yanka Ivanova
(5). Rossen Bojilov
Schematics of receivers and transmitters Department of Electrical Engineering
need to be connected to both piezoelectric Technical University–Sofia, Branch Plovdiv
converters (transducers), A and B, figure 3. 25 Tsanko Dystabanov St.
Accordingly, the emission sequence of the 4000 Plovdiv
ultrasound pulses must be against the direction of BULGARIA
fluid movement to determine the time TBA and in the E-mail: etehsv@gmail.com
direction of movement of the fluid TAB. E-mail: yankakiss777@abv.bg
E-mail: rossen_chi@abv.bg
ВИРТУАЛЕН ИНСТРУМЕНТ ЗА ИЗМЕРВАНЕ

И ОБРАБОТКА НА ДАННИ ОТ ПРЕКИ
МНОГОКРАТНИ ИЗМЕРВАНИЯ
ВАНЯ РАНГЕЛОВА, ВЕНЦЕСЛАВ ПЕЕВ, НИКОЛАЙ ПАУНКОВ
РезюмеВ работата се представя разработен виртуалния инструмент за измерване и

обработка на многократни измервания в реално време. Като източник на реалния
сигнал се използва функционален аналогов генератор, системата за събиране на
данни е АЦП - EMANT 300. На предния панел на виртуалния инструмент могат да се
видят всички параметри, свързани с измерването. Също така хистограмата, реалното
наблюдение на измерения сигнал, компонентите на неопределеността и други.
Ключови думи: използва се отново стила Abstract, не се оставят празни редове
преди или след параграфите
VIRTUAL MEASUREMENT TOOL AND DATA

PROCESSING OF DIRECT REPEATED
MEASUREMENTS
VANIA RANGELOVA, VENCESLAV PEEV, NIKOLAI PAUNKOV
Abstract:This work presents the development of the virtual instrument for measuring and
processing of multiple measurements in real time.As a source of the real signal is used
functional analog generator, the data acquisition system is EMANT 300.On the front panel
of the virtual instrumentcan be seen all parameters related to measurement. Also the
hystogram, the real monitoring of the measured signal,uncertainty components and others.
Key words:virtual instrument. repeated measurements, statistics, uncertainty, LabView
1. Въведение генератор тип JUPITER500, f=100Hz , като се

Неопределеност на измерване е направят многократни наблюдения. Време за
параметър [1], свързан с резултата от измерване, измерване на всеки един резултат 10 секунди.
който характеризира дисперсията на 2.1. Съставяне на спецификация на
стойностите, които могат да бъдат приписани на измерването.
измерваната величина. Винаги, когато е
възможно се предпочитат многократните -Анализ на условията на измерване-
измервания пред еднократното, съответно и измерването се извършва в лабораторни условия
достоверността на оценката е по голяма. В при температура на околната среда +35,0°C
нашата разработка е създаден виртуален
инструмент за измерване в реално време и - Анализ на схемата на измерване -време на
обработка на данните в реално. За обект е взет цикъла за измерване с виртуалния уред – 10 s
сигнал от функционален генератор, чиято -Анализ на техническите характеристики на
честота на сигнала ще се измерва многократно. АЦП -- работни условия на използване –
2. Същност на измервателната задача. температура на околната среда от 0°C до 50°C ,
80% R.H, относителна грешка на измерваната
Да се измери 100 пъти честота на
честота на триъгълни сигнали  f .
нискочестотен сигнал зададена с функционален

II-126
2.2. Съставяне на уравнението на От фигура 1 се вижда, че

връзката. разпределението на измерваната величина не е
нормално ( Гаусово ).Затова, се използва
критерия на Ирвин. След проверка на
f  f `  f кв  cft критерия, се установи, че няма груби грешки.
Където: Защото за всяко ηк получаваме, ηк ≤ η
` таблично=1,0 за 100 измервания и ниво на
f - непоправения резултат от измерването. достоверност р = 0,95, където
cf t - поправка за изменението на честотата на ( fi  fi 1 )
k 
опорния генератор от промяна на S( f ) .
температурата на околната среда.
f кв- поправка за грешката от квантуване- 3. Преглед и обсъждане на съставящите
методична. на неопределеността.
Неопределеността на резултата от
Изключване на грубите грешки. измерването съдържа две групи съставки.
Поради естеството на експеримента, е ясно, 1.Неопределеност, получена чрез
че би трябвало да няма груби грешки, но с оглед оценяване тип А, посредством статистическа
спазване на методиката се прави и такава обработка на резултатите от измерването. Към
проверка.Използва се статистическия пакет тази група се отнася неопределеността
SPSS Следвъвеждане на многократните характеризираща разсейването на резултатите от
измервания и последваща статистическа многократни наблюдения.
обработка на измерените данни се получават 2. Неопределеност получена въз основа
следните стойности за средноаритметичното на на априорната информация за измерването, чрез
` оценявания тип Б.
извадката от 100 наблюдения f и Към тази група се отнасят:
средноквадратично отклонение S ( f `) : неопределеността, свързана с избора на броя
отчети от предния панел на виртуалния
инструмент, неопределеност, свързана с
грешката от нелинейност на АЦП и
неопределеност, свързана с изменението на
При N  30 , средноквадратичното отклонение честотата на опорния генератор в АЦП от
на резултата се изчислява по формулата: промяна на температурата на околната среда.
3.1.Документиране и изчисляване на
съставляващите на неопределеността.
U1-неопределеност, свързана с
Проверка на вида на разпределението се разсейването на резултатите от многократните
прави с програмата SPSS. За целта се въвеждат наблюдения ( Оценяване тип А ).При N>100
измерените стойности и се построява може да се приеме, че:
хистограмата дадена на Фиг. 1.
U2- неопределеност, свързана с избора
на броя отчети Nотч от предния панел на
виртуалния инструмент ( Оценяване тип Б ).
Изчислява се чрез израза за абсолютна
грешка:грешка от дискретизация. Стойността на
големината на кванта на преобразуване се
определя от измерената честота разделена на
броя на избраните отчети:
Фиг.1Хистограма на резултатите. q е големината на кванта на преобразуването.

II-127
U3- неопределеност свързана с свързана Изчисляване на комбинираната

с грешката от нелинейност на АЦП. Границите средноквадратична и разширена неопределеност
на абсолютната грешка от нелинейност не на резултата от измерването Uc:
надхвърлят 10ppm[4], тогава при прието
правоъгълно разпределение ще бъде равен на:
= =0,577. Hz
При ниво на достоверност р = 0,95,
коефициентът на припокриване к е 2 и
U4-неопределеност, свързана с разширената неопределеност е равна на:
нестабилността на честотата на опорния
генератор в АЦП.Границите на абсолютната
грешка не надхвърлят 122 ppm [4], тогава при k=2
прието правоъгълно разпределение ще бъде
равен на: Обобщение на бюджета на
неопределеността [2] е даден в Табл. 1.
= =0,71. Hz
Таблица 1.Бюджет на неопределеността
Съста- Тип Тип Брой Коефициент Стойност на
влява- оценя разпределе- измервания на съставляващата на
ща i -ване ние N разпределе- неопределеността
нието b ui , Hz
u1 A Правоъгълно 100 - 0,005
u2 Б Правоъгълно - 0, 6 0,577
u3 Б Правоъгълно - 0, 6 0,577.
u4 Б Правоъгълно - 0, 6 0,71.
4. Разработване на виртуален отчети,зададения брой на многократното

инструмент за многократни измерване на честотата на сигнала,индикатор за
измервания. номера на текуща измервана честота,индикатор
С така направената постановка се на общия брой измерени честотии,измеренана
разработи и виртуален инструмент, който да средна стойност на честотата,измерената
реализира по горната процедура автоматизирано средноквадратична стойност на извадката,
и контролируемо. Като аналого-цифров хистограма на многократните измервания,
преобразувател се използва система за четирите видове съставки на неопределеностите
събиране на данни АЦП – EMANT 300 [3] и и техните стойности, разширената
персонален компютърсъс заредена програмна неопределеност,крайният резултат.
среда LabView8.5. На фиг. 2 е показан един резултат от
На предния панел на разработения направените експерименти. А именно, както се
виртуален инструментпоказан на Фиг. 2, са вижда това са следните стойности на
включени следните възможности:Наблюдение проследяваните параметри :Стойността на
на входния измервателен сигнал, цифрова
индикация на измерената честота,означение на измерената честота
кои входни портове е сигнала,индикация на ID,
време да измерване,зададения брой на цифрови 91,661Hz.,стойността на средно кв. откл

II-128
Фиг.2. Преден панел на разработения виртуален инструмент за обработка на преки

многократни измервания в реално време.
0,0225, избрания брой отчети - 100 , избрали свързаната с тях средноквадратична

сме брой измервани честоти – 100, неопределеност и методите за нейното
хистограмата на разпределение на тези 100 оценяване. Изхождайки от казаното до тук,
разработката може да се използва навсякъде,
измервания на една и съща
където има потребност от такава обработка на
честота,съставящите общата неопреде- резултати, а също и от студентите като обобщен
леност : пример на методиката по изразяване на
=0,0023 =0,26 =7,1 неопределеността на измерванията. Тази
методика за задаване на неопределеността на
=5,8 =0,52 измерването се изисква в много страни и е
задължителен елемент от документите за
5. Заключение сертификация на дадено изделие.
Резултатите от проведени измерванията
никога не са абсолютно еднакви и без
колебания, защото зависи с колко чувствителен ЛИТЕРАТУРА
уред ще се проведе експеримента и при какви
условия, а също и от използваната 1. ISO/IEC GUIDE 99:2007 (E/F)
измервателната техника. Ето защо 2. Радев Х., Метрология и измервателна
„неопределеността на измерванията” е част от техника , том 1, Софттрейд, 2009г
резултата на измерването. Тя е необходима и 3. http://emant.com
допринася за установяване на прецизността в 4. https://www.dataforth.com/signal-
изразяването на измерената величина. conditioning-accuracy-vs-resolution.aspx
Неопределеността на измерването трябва да е
известна за всяко измерване. Всеки резултат от
измерване е известен само в границите на своята Ваня Рангелова
неопределеност. Бюджетът на неопределеността ТУ София, ф-л Пловдив, кат. ЕЛТ
обобщено представя оценяването на E-mail: vaioran@abv.bg
съставящите на неопределеността на резултата
от измерването. Бюджетът включва списък на
всички източници на неопределеност,
HIGH SENSITIVE LONG DISTANCE SCANNING

FIBER-OPTICS LASER SENSOR SYSTEM
MARGARITA DENEVA, MARIN NENCHEV
Abstract: The principle of the proposal is to use laser with long (25 m and more)
resonator, composed by series of short free space parts, coupled by long optical fibers. We
combine the laser with such resonator construction with spectral characteristics of the
generation, which are appropriate for intra-cavity laser spectroscopy registration of
expected atoms and molecules (including atmospheric non-desired pollutions – SO2, NO2,
Na etc.). The registration on the base of the intra-cavity laser spectroscopy (ILS) method
makes the laser developed extremely sensitive for atoms or molecules presence in the
investigated spaces of the resonator free-space region, in principle down to single atoms.
The laser spectral control includes also original solution for producing suitable controlled
multi-bands spectrum for simultaneous monitoring in different spectral bands. The laser
system proposed is very convenient for controlling the air purity in number of rooms in
given laboratory, enterprises, where the presences of the non-desired pollutants are
expected. We treat theoretically the system action and present the experimental laboratory
test on the example of registration of Sodium atoms presence in the air.
Key words: intra-cavity spectroscopy, special laser system, long resonator, multi
registration parts, theoretical treatment, experiment with Sodium atoms
1. Introduction importance for the military production specialized

Specialized laser systems are established at enterprises. We present also the experimental
present as important tools for distant monitoring of laboratory test of developed such system by
air pollutions, in ecology, in industrial enterprises registration of Sodium atoms presence in the air in
producing different type chemical materials, in different places.
military production, etc. [e.g. 1-4]. Here we propose
high-sensitive long-distance scanning fiber optics
2. Details of the principle of the proposed
laser sensor system for such ecological control,
system
especially for monitoring of close working places
(number of separated rooms). The principle of the 2.1. General description
proposal is to use laser with long (50 m and more) The laser source, which is suitable for intra-
resonator, composed by series of short free space cavity spectroscopy, is with homogeneously
parts in each monitored place (room), coupled by broadened amplification and with controlled wide-
long optical fibers and combined by the laser with band generation. As an additional advantage in our
such resonator construction with spectral case, the spectrum is tunable in the laser gain. The
characteristics of the generation that are appropriate general condition for application of the laser source
for intra-cavity laser spectroscopy registration of in intra-cavity laser spectroscopy is that the spectral
expected atoms and molecules (including band of the emission must be smooth, structure-less.
atmospheric non-desired pollutions – SO2, NO2, Na, The principle of the extremely high sensitive intra-
etc.). The registration on the base of the intra-cavity cavity registration consists in this that the spectral
laser spectroscopy method makes the laser absorption lines of the investigated atoms or
developed extremely sensitive for atoms or molecules presented in the laser cavity make holes
molecules presence in investigated free-space in the smooth laser spectrum.
resonator region, in principle down to single atoms. The drastically increasing of the sensitivity
Except the cheapest engineering realization, the in comparison with this one in the case of the extra-
important advantage is the completely elimination cavity registration is result of strong competition
of possibility for spark formation that is of essential between the generation at different lines in the wide
II-130
spectrum in the homogeneous broadened active AR coated ) and at the end of fiber OF4 with retro-
laser medium [5,6]. The introduction of smallest reflected output by the end-concave mirror CM. The
losses for one line leads to strong decreasing or each box is disposed in the corresponding room –
suppressing of the lasing at this line. The additional Roomi. The registration subsystem RS consists of
important factor to increase the competition and spectrum analyzer, which in one variant is based on
respectively - the sensibility is to use temporally diffraction grating DR in near grazing incident
long generation – long pump pulses (flashlamp- angle, where the resolution is highest, and
pumping with long exciting pulses or cw diode photograph-plate R with corresponding construction
pumping) or cw operation. The condition to be arrangement (first inset in Fig.1, details are not
avoided the typical structure in the wide-band laser shown) or realized on the base of CCD matrix. In
spectrum is that in the resonator are precluded any equal manner, the spectrum can be registered after
parallel reflective surfaces (including low expanding telescope (LT1-LT2) and using convenient
reflection), forming the resonance structure of Interference Wedge IW2 [8, 9] second inset in Fig.1.
Fabry-Perot type. As very convenient laser for intra- Below, as a point of the presentation, we give
cavity laser spectroscopy, which is applied here, general detailed description of the characteristics of
following our earlier work [7], is a waveguide type the system, which leads to the reality of its expected
laser that naturally enables to exclude the parallel advantageous practical functioning.
surfaces into the cavity and to obtain smooth The laser is the waveguide flashlamp
spectrum. pumped dye laser (in other realization can also be
The schematic of the system proposed, suitable spectrally semiconductor laser and
taking into account the given general description, is flashlamp pumped Ti:Sapphire laser). The dye
presented in Fig.1. active solution – 3.10-4 mol/l Rh6G in ethanol,
flows through the cell that is cylindrical glass
Room 4 Room 3 Room 2 Room 1 capillary-tube with internal diameter of 1 mm,
BR4OF4 BR3
OF3
BR2
OF2
BR1
external diameter of 10 mm and length of 100 mm.
The pumping is by flashlamp in elliptical type
L1
CM L3 L2 reflector thus assuring the exciting pump energy of
10 J in 15 µs pulse (rise front 2 µs, plato 4 µs and
OM IW1 fall front 7 µs - at half maximum). The focal length
pump
of the lens Lo is 10 cm and for the lens Lc it is 5 cm,
RS respectively; the radius of end-concave mirror CM
OF1 is 5 cm and its reflectivity is 99 %. The IW1 and
L0 WAM LC
IW2 are composed by dielectric layers and both are
with thickness of 3 µm, wedge angle 5x10-5 rad and
R reflectivity of the composed mirrors for IW1 – 0.8
LT2 L and for IW2 - 0.3. The mechanical construction of
T1
R each box (20x20x25 cm) contains a door that can
DR assure its prevention of the registration, i.e. the
IW2
corresponding room to be out of control. The aim of
such isolation is to assure possibility to monitor
Fig. 1. The schematic of the proposed system each room separately. The total length of the high
quality connected optical fiber is with β = 7 dB/km
In the proposed system, the laser is (multimode, 125/250 µm) and the total length of the
composed by a capillary-tubular cuvette filled with fibers is 15 m. Accepting that the loss of any
liquid active medium (WAM) or other type connection (by lenses Li) is (0.2-0.3) dB, the ratio
waveguide realization active medium; pump source, (returned power/incident power) for total fiber
and the special resonator. In one side of the WAM system can be evaluated to be of order of (0.55 –
the resonator reflector is formed by output coupling 0.6), or the fiber part can be accepted as equivalent
using wedged mirror OM (R= 0.96), spectral mirror with reflection of ~0.6.
selective and tuned Interference Wedge IW [8] and
focusing lens Lo. The last transforms the emitted 2.2. Theoretical description of the system
light from the output of the waveguide AM. In the action.
other side, the resonator reflector is formed by A discussion of the strongly increased
series, numbered with i, (i = 1,2,3,4), of boxes for sensitivity obtained by applying intra-cavity
registration BRi connected via long optical fibers spectroscopy technique will be presented.
(OFi) that are coupled each other with the In the consideration we will use the values of
corresponding lens (Li ; all lenses in the system are laser system discussed in the previous point. The
II-131
differential rate equations system [5], adapted for concentration provides absorption losses for single
description of the case under investigation described pass at D2 line of 1% (respectively – 0.5 % for D1
Rh6G dye solution and pumping with the system line). Using the given above differential equation
parameters described above, is: system, we obtain the spectra of laser light for the
two cases.
dN 2   N For the extra-resonator case (simple
 R p (t )    Bi  qi   N 2  2
dt    passage) we obtain practically non-observable (or
 i 
specially detected holes due to of 1 and 0.5 %
dqi q
 Va  Bi  qi  N 2  i absorptions at the lines). The calculated output laser
dt  ci spectrum for this case is shown in Fig.2, in the inset
with Pout (t )   1  c 2L '  h  qi (t ) the formed hole by the absorption at D2 line in
i
expanded scale is given.
Here qi are the generated photon number for the Eout, J
Eout, J
corresponding wavelength in the considered laser
spectrum; Pout is the corresponding output power,
i
which integration in the time (from 0 to the length
pulse) gives the output energy. In the systems, with
N 2 is noted the population of the upper laser level
per unit volume in the WAM. The term
  
Bi   ei  l  c Va  L' [s ], where  ei is the
-1 , nm
emission cross-sections for the given wavelength in Fig.2. The calculated output laser spectrum for the
the laser spectrum; Va is the working volume case of registration by single passage through the
(0.05 cm3); c = 3x1010 cm/s is the light velocity; Sodium D-lines. In the inset is shown the formed
L'  L  (n  1)  l - the optical length of resonator, hole by the absorption at D2 - line (the figure is
expanded).
where l=10 cm is the length of the active medium,
n is the refractive index and L is 25 m. The time The calculations for the intra-cavity case of
term  of 3 ns is the lifetime of the upper laser emission spectrum with the same volume of Sodium
level for the Rh6G WAM. The dumping time of a atoms, gives the picture, shown in Fig.3. The
photon in the resonator is  c  L' c   i  , where drastically increasing of the holes at the D-
i absorption lines can be seen. This confirms our
 i [5] describes the loss into the resonator for the expectations for the possibility for high sensitive
wavelength in the considered laser gain spectrum registration of desired atoms presence by
(depending on the reflectivity of the combined end application of the proposed laser system and for
resonator mirror OM, IW1 and Lo). The system was monitoring of such presence in few rooms.
solved numerically by Runge-Kutta-4 method.
From the solution we obtain qi (t) and the
respective output power for each wavelength in the
spectrum;  1 characterises the output of the laser
resonator. The calculations are prepared for the
spectral range (588 – 591) nm, in which the Sodium
D-lines are also comprised.
In the example considered we will compare
the extra-cavity and intra-cavity registration of
presence of Sodium vapors outside and inside of the
described special resonator. We will study the Fig.3. The calculated output laser spectrum for the
behavior of the formed by the Sodium atoms case of intra-cavity registration by single passage
absorption holes in the spectrum at the Sodium D- through the Sodium D-lines.
lines (doublet D2 =589.0 nm and D1=589.6 nm).
Let’s we accept that the same volume of Sodium Let us evaluate the concentration of the
atoms with the same concentration and length – Sodium atoms that leads to 1% decreasing of the
path of the laser beam through the volumes, is transmission – i.e. the case considered above.
disposed one time outside the resonator and after Following Ref. [9] we can write:
this - inside the resonator. Let’s the Sodium atom
II-132
I I o  exp    N  la  In Fig.5 are shown the corresponding trace-

where I0 is the incident intensity, I – the intensity of curves for the two spectrograms – left for the given
the transmitted beam, both I0 and I for wavelength in Fig.4(a) and right – for the spectrogram in
Fig.4(b).
of the absorption line,  = 6.10-12 cm2 is the
absorption cross-section for the D2 line and la=5 cm
is the path of the beam trough the volume with
Sodium atoms. For these conditions, after a simple
arithmetic, we obtain that N is of order of (107÷108)
cm-3. Thus our system assures reliable registration
of presence of Sodium atoms with such
concentration.
3. Experimental test
In the given in the work practical test- Fig.5. The corresponding trace-curves for the two
realization (example) that clarifies the system spectrograms in Fig.4 – left for case (a) – Sodium
action, we use only two structures - BR1 and BR4, atoms are outside the cavity (practically the
excluding BR2 and BR3. Both optical fibers OF1 and absorption D-lines are not observed ) and right for
OF4 were with length of 1 m each and damping the case (b)– the atoms are inside the cavity (real
factor β = 300 dB/km and four connections with the observation of the D-lines). The scale is given by
lenses (equivalent mirror with reflection ~ 60 %). the holes of D2 and D1 lines marked on the figure.
The other used characteristics of the elements and
of the construction are the given above. The volume
of Sodium atoms is formed by burning of three Following the theoretical considerations,
standard matches with the flame outside the cavity given above, and the obtained spectrograms, we can
near the output mirror OM (extra-cavity conclude that practical laboratory model of the
spectroscopy) or inside the cavity - in the space system proposed is able for real registration of
between the cell output and lens Lc. Sodium atoms with concentration 107 cm-3. This
The typical real spectra of the emission of concentration can be accepted to correspond to the
laser system with spectrum of generation tuned at obtained one in the matches flame.
Sodium D-lines are given in Fig.4. The top
spectrogram (Fig.4a) is for the case when the
Sodium atoms are formed outside the cavity and the 4. Discussions
laser light passes through the space with the Sodium Following the results from the theoretical
atoms (absorption outside the cavity). The bottom analysis and from the carried out experiment we
spectrogram (Fig.4b) is for the case of presence of make in evidence the feasibility of the proposed
the Sodium atoms inside the complex laser cavity. system and its advantages for specific ecologic
monitoring and survey for dangerous pollutions,
accompanied the industrial production. The
sensibility of such intra-cavity spectroscopy based
technique is more than of order of magnitude higher
than the standard extra-cavity laser technique. Very
(a) suitable laser sources are these one that avoid the
flat parallel reflected (as well as partially) surfaces,
what in high degree is natural property of the
waveguide laser sources. With application of
modern low-losses optical fibers and suitably
treated elements (without parallel surfaces, there is
(b) no parallel passage of the light rays) the proposed
system can be organized as very suitable technical
Fig.4. Typical real spectrums of the laser system instrumentation for extremely sensible detection of
emission with spectrum of generation tuned at desired pollutant atoms. Note that in particular case
Sodium D-lines. The top spectrogram is for the of many rooms disposed at both sides of building
absorption outside the cavity and the bottom - for corridor, the optical fiber part of the resonator can
the Sodium atoms inside the complex laser cavity. be adapted to form a ring type laser resonator. This
The strong increasing of the sensitivity is evident. will permit to use more effectively fiber parts of the
resonator.
II-133
5. Conclusion Tech. Eksperimena, (1994) Improvement
In the work is proposed the principle of the analytical …, Burakov V. S. et all); M.Deneva,
special laser system with an optical fiber based long E.Krasteva. P.Bakardjiev, M.Nenchev (2001).
(20 m and more) resonator, composed by series of Tunable laser devices with inverted spectrum
short free space parts, coupled by long optical for application in intracavity spectroscopy and
fibers. We combine the laser with such resonator spectral selective excitation of atoms, Proc. of
construction with spectral characteristics of the LTL Plovdiv'2001, International Symposium
generation that are appropriate for intra-cavity laser (2001) 153-156.
spectroscopy registration of expected atoms and 5. O.Svelto. Principles of lasers, 5th ed. Springer
molecules. As it is shown, the registration on the Science-Business Media, 2008.
base of the intra-cavity laser spectroscopy (ILS)
method makes the developed laser extremely 6. M.Deneva, P.Uzunova, M.Nenchev, Tunable
sensitive for atoms or molecule presence in subnanosecond laser pulse generation using an
investigated spaces of the resonator free-space active mirror concept", Opt. and Quant.
region. The laser spectral control includes also the Electronics, 39 (2007) 193-212, USA
original solution with Interference Wedge for 7. M.Nenchev, V.Stefanov, A. Gizbrekht, Study of
producing suitable controlled multi-bands tunable Self frequency locking in dye laser at the
spectrum (for monitoring in different spectral Sodium D-lines. 3rd Confer. on Luminescence,
bands). The theoretical treatment of the system Szeged, Hungary,1979, Confer. Digest,. Vol.1,
action and the experimental laboratory test for 135-139 (in Russian); M.Nenchev, V.Stefanov,
registration of very low concentration presence of A.Gizbrekht, A.Bokhanov, Self frequency
Sodium atoms in the air confirm the feasibility of locking in lasers at lines of intracavity gases,
the system and its expected advantages. The laser Bulg. J. Phys. 7, (1981) 521-524
system proposed is very convenient for controlling
the air purity in number rooms of given laboratory, 8. E.Stoykova, M. Nenchev, Gaussian Beam
enterprises, where the presence of the non-desired Interaction with Air-gap Fizeau Interferential
pollutants is expected. wedge, J. Opt. Soc. America A, 27 (1), 58-68
(2010)
Acknowledgement: The work is supported by 9. Y.H.Meyer, M.N.Nenchev, On intracavity
DN 08/13 (2016) Contract with NSF-BG. absorption and self-frequency locking in
pulsed dye laser. Opt. Communs., vol.4, No5,
REFERENCES (1982) 292-294 and the literature therein.
1. W. Demtröder Laser spectroscopy: basic
concept and instrumentation, (2003) 3rd ed.
Springer, Germany and the literature therein.
Contacts:
2. P. Sudhakar, P. Kalavathi, D. Ramakrishna M. Deneva, M. Nenchev
Rao, M. Satyanarayana (2014). Design of Technical University-Sofia, Branch Plovdiv
Laser Based Monitoring Systems for Address:
Compliance Management of Odorous and Dept. Optoelectronics & Lasers
Hazardous Air Pollutants in Selected Chemical and Scientific Laboratory QOEL
Industrial Estates at Hyderabad, India. Remote 25 Tsanko Diustabanov St,
Sensing and Spatial Information Sciences, 4000 Plovdiv
volume XL-8, and the literature therein . Bulgaria
3. E. D. Hinkley, P. L. Kelley. Detection of Air E-mail:
Pollutants with Tunable Diode Lasers, mar.deneva@abv.bg
Science 19 Feb 1971: Vol. 171, Issue 3972, pp.
635-639, DOI: 10.1126/science.171.3972.635
4. Бураков В.С., Исаевич А. В., Мисаков,
П.Я., Ненчев М.Н., Патриков Т.В., Пашов
А.Е., Пешев З.И., Разширение аналитических
возможностей внутрирезонаторных лазер-
ных спектрометров, Приб. и техн.
Эксперимента - ПТЭ, No 5, 1994, 150-156,
(RUSSIA, trad. and parallel ed. in USA – Prib.

THERMO-SENSITIVE PAPER APPLICATION AS

METHOD FOR LASER BEAM SPOT STUDY –
CASE OF Q-SWITCHED LASER PULSES
VALKO KAZAKOV
Abstract: Recently, we have developed, as a new technique for laser spot profile
investigation, a combined spot registration on appropriately chosen thermo-paper with
convenient computer treatment [3]. Such technique presents some essential advantages:
spectral insensitivity in very large range (from UV to IR); the registration is not affected by
electromagnetic noise; as addition, it is extremely cheap and accessible. Here, on the base
of our previous experience, obtained mainly with free lasing Nd:YAG laser pulses, we
present results of application for Q-switched pulses of this laser. Some specific behavior of
the registration under condition of more than thousand times shorter pulse influence than
in free lasing are shown and the condition – power density and energy density limits of
correct registration are carried out. The technique under carefully laser illumination
control can give acceptable results also in the considered case.
Key words: laser spot study, thermo-sensitive paper, Q-switching.
1. Introduction. General aim of the work. Taken into account the essential difference between
The aim of the work is extension of our the characteristics of the free lasing and Q-switched
previous results, related with development of the laser pulses, especially temporal (ms against ns) and
practical application of Thermo-Sensitive Paper power (kW against hundreds kW for comparable
Registration Technique (TeSPeRT) for laser spot energies), the application of the developed by us
imaging. This old technique [1, 2] is used in the technique for the Q-switching case needs particular
literature and in the laboratory practice only for treatment of the Q-pulses registration, which is the
visual illustration (including also successfully - for aim of the presented work. Here, we present the
the interesting cases) of the laser spot. The base of results of noted above study and namely concerning
the technique, as a rule, is the whitening in the laser the energetic aspects and conditions for correct
beam incident area on the black thermo-sensitive applications of the TeSPeRT for Q-switched laser
paper. In previously reported applications [1, 2] are pulses treatment. Note that such technique presents
used accidentally taken sensitive materials and some essential advantages in comparison with
without conclusion about energetic characteristics electronic based techniques of this type: spectral
of the spot. As we have shown in Ref. [3], in insensitivity in very large range (from UV to IR);
condition of combining the convenient chosen the registration is not affected by electromagnetic
thermo-sensitive paper for registration and noise; and as addition, it is extremely cheap and
convenient computer treatment, and with correctly accessible. The beam-profile knowledge is of
defined illumination, this technique can be essential importance in the laser manufacturing and
successfully used for laser beam energetic in the scientific and practical application of the
parameter study. In the noted work we demonstrate lasers [1 - 5].
the correct determination of the energy density
2. Experimental investigations and results.
distribution in the beam cross-section, as well as
evaluation of the beam energy. Our previous We start the proposed study using the
investigation is by the use of the free lasing results, given in our previous work [3], about the
Nd:YAG laser light (pulse length 0.1 – 3 ms; power free laser pulse light treatment applying TeSPeRT.
~ 10 - 40 kW). The obtained results confirm the In the work [3] we have presented and discussed the
usefulness and the expected potential of the principle and the convenient condition to obtain
technique developed by us, especially for the correct energy density distribution in the laser beam
condition of typical free lasing light emission. cross section. The given there results are for pulse

II-135
energies in the range of 0.3 J to 10 J with pulse spots by scanner with white cover. Thus, the
duration in the range 100 - 3000 µs – i.e. the typical cleaned/ transparent points of the spots are scanned
free lasing operation of used in the tests Nd:YAG like white points, which gives more accuracy,
lasers (at 1.06 m and 1.36 m generation). Taken because the base of the method is whitening
into account the well-known essential difference proportional to the energy/ power density. In case of
between the characteristics of free lasing and Q- free lasing, there is no notable difference between
switched laser pulses, especially temporal and pulse scanning with white or black scanner cover, because
power characteristics [1, 2], we provide systematic there is only whitening without cleaning of the
study for the case of the Q-pulses registration by the blacked layer.
TeSPeRT. As a laser beam source for the investigation
In the study of free lasing case in Ref. [3], we used passively Q-switched Nd:YAG laser. The
we have shown that a very suitable material for the study was performed at laser parameters: generally
beam spot marking is the tracing paper, blacked at for the energy between 0.1 J to 1.5 J and pulse
one side through the copy machine and illuminated length 1 µs ± 0.2 µs; repetition rate – single pulse
at the non-blacked side by the laser beam. Here, (to 1 Hz); emitting at line 1.06 µm. In most of the
following our works [3] and complementary experiments the energy of the beam, illuminating
investigations, we have used for beam spot the studied material, was equal to the laser output
registration the same type of tracing paper (A4 92 energy, except the case where fine plate filters were
gr/m2 from Sihl Digital Imaging Company), blacked used. The energy was measured with FIELDMAX
one (1X), two (2X) or three (3X) times through the energy meter, Coherent (USA), the pulse duration -
standard xerography type action machine - Konica with two-channel 200 MHz storage oscilloscope
Minolta Dialfa Di 5510. The blacking-out was made RIGOL DS1202CA (USA) and light detectors with
by copying the tracing paper at normal blacked resolution better than 5-ns - R-108 (BG).
position of the machine option. We do not observe At the left side of Fig.1 are shown the
noticeable difference comparing few tracing papers typical marked spot by Q-switched pulse (Fig.1a)
from different producers and also the blacking by and the graphs, obtained by 3D (Fig.1b) and 2D
different copiers of the noted before type (Sharp- (Fig.1c, across the spot diameter of ~ 0.7 cm)
MX-3500, Toshiba 2500c) for normal blacking computer treatment of its whitening. The Q-pulse
operation. The transmission of pure tracing paper is duration is ~ 1 µs and the pulse energy is 0.88 J on
measured to be ≈ 50 % when the paper is placed the 1X type of paper. For comparison, in the same
closely to the entrance of the light power meter picture at the right side are shown - the spot, made
(Thorlabs SN6050506). In this manner the by the free lasing laser light (Fig.1d) and the 3D
measurement with flat and large entrance of 2 cm (Fig.1e) and 2D (Fig.1f) computer treatment of the
diameter accept all the diffused light. We used both white spot. The free laser pulse duration is ~
He-Ne yellow (0.595 m) laser beam with incident 300 µs, pulse energy ~ 1.17 J and spot diameter is ~
diameter on the paper of ~ 2 mm and power of 3.6 0.7 cm. The computer curves are normalized to their
mW, and focalized light from standard electric bulb maximal values.
with tungsten filament. There is no notable
difference of the measured transmissivity. The
corresponding measurements of the transitivity after
blacking are: for 1X - (10 ± 2) %; for 2X - (1 ± 0.1)
% and for 3X ≈ 0.1 %. The transmission through
marked spot by Q-switched pulse is measured to be
≈ 25 %, which is ~ 2 times more than the
transmission through spot illuminated by free lasing
pulse with the same energy – given in Ref. [3].
There are quite differences in 2D and 3D graphics,
when the Q-switched marked spot is scanned with
white cover and with black cover. The graphics of
spots, scanned with black cover are with unusual
form and lower height than is expected to be.
Hence, in case of Q-switched pulses there is not
only whitening of the blacked paper, but also
cleaning at some points of the spot where is a higher
power density and there the blacked paper is Fig.1. Marked spots by Q-switched pulse (a) and
transparent like a pure tracing paper. The computer free lasing (d), and their corresponding 3D (b, c)
treatment in this work is made after scanning the and 2D (e, f) images for 1X type of paper.
II-136
The general difference between the Q- As example of the realized investigations, a
switched and free lasing spots, as it can be seen also collection of Q-switched laser light spots marked
in Fig.1, is that the Q-spots are colored in red. This with different pulse energies on the paper 1X, noted
leads to different colors presentation of the 2D as Q1, Q2, Q3, Q4 and their computer treatment are
image of the spot – red, green and blue curves. This given in Fig.2. For comparison are given also the
is quite different from the 2D image of the spot for two spots F1 and F2 of free lasing pulses.
the free lasing, where the three colors practically
coincide completely. We attribute this difference
with formation of products of burning and/ or
photochemical reaction at the black layer due to the
high power of the illuminating light in the Q-pulses.
The nature of the coloring needs future chemical
investigation. Thus, the first step of the
investigation is to see: i) at which degree the
registration (near constant R, see below) is correct
when we use for beam chosen color and ii) which
color corresponds better with the real energy
density distribution in the spot. The curves of the
three colors that describe the 2D description of the
graph are shown in the figure with notation – Red,
Green, Blue and their composition as RGB.
We accept, as approach to computer
treatment of the spot, the methodology argumented,
developed and experimentally tested in details in
Ref. [3] (our previous work for the free lasing case).
Thus, following the discussion of this work, for
criterion of correctness for the obtained energy
density distribution WE (x, y) J/cm2 in the spots, we
will use the ratio R = E/ V of the pulse energy E and
the calculated volume V under the computer
obtained envelop graph WH (x, y) = WE (x, y)/ K1 of
the whitening. The constant value of K1 = R = E/ V
can serve as a criterion for correctness of energy Fig.2. Marked spots by Q-switched pulses Q1 - Q4
density distribution, obtained by the experimental and free lasing light formed spots F1, F2 on 1X type
measurement of E and computer treatment - the of paper (a); their 3D (b) and 2D (c) graphics.
arbitrary graphs of whitened and computed volumes
under the 3D graphs. The correctness of TeSPeRT applications
Following the formulated above problems, for the free lasing case is shown in Ref. [3]. Thus,
we treated series of stamps with spots on the we treat here the Q-switched pulse spots. The pulse
described blacked tracing papers, illuminated by length is ~ 1 µs and the pulse energies are given in
variety of Q-switched pulses (Nd:YAG, 1.06 m). Table 1. In the same table, the calculated volumes
The passive Q-switched operation of the laser was under the 3D graphs and the ratio Ri for the RGB
with controlled variable energy output between 0.1 registration are also given. With RGB are noted the
J and 1.5 J and pulse length of ~ 1 µs. For the spots averaged curves, giving the average value for each
with diameter of 0.7 cm, the energy (and the power) point from the sum of the corresponding points for
density vary respectively between ~ 0.5 J/cm2 (0.5 the three curves. The computer program
MW/cm2) and 3 J/cm2 (3 MW/cm2). automatically gives the averaged volume of the
Preliminary question is the dependence of volumes for the three colors scanning. We have also
the pulse energy (and power) density and especially study separately the 3D images for each color and
limits of notable marking and saturation. We found the results are shown in the Table 2 and Table 3.
that for the three investigated type of papers - 1X, Processing of each paper type will be
2X, 3X the noted limits are relatively close to each detailing described on the example of 1X, for which
other - minimum energy density for reliable the best results are obtained. After computer
registration was ~ 0.5 J/cm2 (power density ~ treatment of the four (Q1 – Q4) spots shown in
0.5 MW/cm2) and the saturation was observed for ~ Fig.2 and with measured light energy that forms
3 J/cm2 (~ 3 MW/cm2). each of them, we obtained the average value
Rav=(∑Vi/Ei)/N=∑Ri/N, where sum is for i = 1 - 4,

II-137
Vi and Ei are the volume and the energy of the spots 3. Discussion.
numbered as i and N = 4; Rav = 1.29. We defined the The syntheses of the typical most important
deviation (or error) Ri for each spot as Ri = Ri - results of the investigation are summarized in the
Rav and the relative average deviation for the examples in Table 1, Table 2 and Table 3.
considered group spots of 1X sheet was Rav ≈ 0.12 Following the results, we can conclude that the
= 12 %. application of the technique developed gives
Table 1 and Table 2 are illustrations of the acceptable accuracy of the beam profile registration.
analisys - measured energies and calculated For the defined parameters of the Q-switched
characteristics of the computer treatment for the illuminating laser, described in the work, as most
averaged three colors (RGB) and separately – for suitable blacked paper can be considered the 1X
each color – red, green, blue. The sheet is 1X, paper type. The average error for this case is ~
which gives the best results. 12 %, which is acceptable information for the
In the same manner, we have treated the desired energy density distribution. Interesting fact
noted other cases. The results are given in Table 3 - is that the treatment taking the three colors gives not
comparison of the results of treatment for the two very different results - less than 1 %.
types blacked tracing paper - 2X and 3X.
4. Conclusion.
Table 1. Characteristics for 1X type of In the work we present results of developed
paper – energy and results for averaged three by us simple, cheap and accessible technique
colors (RGB) (TeSPeRT) for Q-switched laser beam profile study.
We have shown this study using appropriate, well
ΔRav
E, V(RGB), Ri, Rav, |∆R|, ΔRav, defined and reproducible as a production sensible
№ Rav,
[J] [a.u] [a.u.] [a.u.] [a.u.] [a.u.] materials – xerox blacked tracing papers for laser
%
Q1 0,95 1,09 1,14 0,14 spot registration in combination with suitable
Q2 0,82 1,21 1,48 0,19 computer treatment. This technique can be
1,29 0,12 9 successfully employed as beam-profile registration
Q3 0,76 1,01 1,32 0,04
Q4 0,88 1,06 1,20 0,08 method with well acceptable accuracy. We present
the conditions (our experiments), when the
Table 2.a. Characteristics for 1X type developed technique shows such applicability. Note
of paper – volume and Ri for R, G, B colors that, such technique represents some general
essential advantages in comparison with electronic
V(Red), Ri, V(Green), Ri, V(Blue), Ri, based beam-profile treatment techniques: the
№
[a.u.] [a.u.] [a.u.] [a.u.] [a.u.] [a.u.] registration is not affected by electromagnetic noise
Q1 1,19 1,24 1,06 1,19 0,93 0,98 and has spectral insensitivity in very large range
Q2 1,34 1,63 1,19 1,45 1,02 1,25 (from UV to IR).
Q3 1,01 1,44 0,99 1,30 0,87 1,15
Q4 1,17 1,33 1,03 1,17 0,9 1,02 Acknowledgement:
The author thanks for the support for the
Table 2.b. Characteristics for 1X type present publication to the Contract DN8/13 FSI–
of paper - Rav, ΔRav, ΔRav/Rav for R, G, B colors BG, assoc. prof. M. Deneva and prof. M. Nenchev.
Color Rav, [a.u.] ΔRav, [a.u.] ΔRav/ Rav, %
REFERENCES
Red 1,41 0,13 8,8
Green 1,26 0,12 9,1 1. Ifflander, R. (2001). Solid-State Lasers for
Blue 1,26 0,10 8,8 Materials Processing. Springer-Verlag Berlin.
Heidelberg, New York, series in Optical
Table 3. Characteristics for 2X and Sciences. ISSN0342-4111, ISBN 3-540-66980-
3X type of paper 9 (2001).
Sheet Rav, [a.u.] ΔRav, [a.u.] ΔRav/ Rav, % 2. Carlos B. Roundy, Spiricon, Inc.,
2X (RGB) 1,54 0,17 11,4 http://aries.ucsd.edu/LMI/TUTORIALS/profile
2X (Red) 1,68 0,19 11,2 -tutorial.pdf, Beam Profile Measurements and
2X (Green) 1,51 0,17 11,3 the literature therein.
2X (Blue) 1,36 0,17 12,2 3. V.P. Kazakov, M.A. Deneva, M.N. Nenchev,
3X (RGB) 1,79 0,44 24,8 N.Y. Kaymakanova. Xerox treated tracing
3X (Red) 1,89 0,45 23,8 paper as suitable and accessible material for
3X (Green) 1,77 0,44 25 development of new laser beam–profiler
3X (Blue) 1,55 0,4 25,6 technique. Journal of the Chemical Institutes of
II-138
the Bulgarian Academy of Sciences and of the radiation in human teeth: determination of the
Union of Chemists in Bulgaria 2016 Volume absorbed and scattered parts.
48 / Special issue G, pp 85-90.; 5. Deneva M., Nenchev M., Wintner E., Topcu S.
M. Deneva, M. Nenchev, V. Kazakov, N. (2015). Coaxial-geometry tunable dual-
Kaymakanova. Development of the thermo- wavelength flashlamp-pumped Nd:YAG laser.
sensitive paper application as method for laser Opt. Quant. Electronics, DOI 10.1007/s11082-
beam spot study. Journal of the Technical 015-0205-5 online publ.
University – Sofia, Plovdiv branch, Bulgaria
“Fundamental Sciences and Applications”,
Volume 22, 2016, pp 92-95. Contacts:
4. P. Uzunova, S. Rabadgiiska, M. Deneva, M. V. Kazakov
Nenchev et al. Proc. SPIE, Vol.8770-54, “QOE” Scientific Laboratory, R&D Dept.,
(2013). Study the penetration of IR laser Technical University of Sofia and OELE
Dept., TU-Branch Plovdiv, BG
v_p_k@mail.bg

THE OBSERVATION OF ATMOSPHERIC
ELECTRICITY AND LIGHTNING EFFECTS
METIN SALTIK
Abstract: In this study, using high-precision JFET sensor, the impact of lightning
and electrical changes in the atmosphere have been observed. Observations made
24 hours, electrical changes in the atmosphere was recorded. As a result of the
observations, in the case of the rise and fall of the sun was observed to vary the
electrical load in the atmosphere. On the other hand, is the electrical changes in the
atmosphere has been observed that the maximum level of the lightning event. This
study also showed that the most important changes are lightning current parameters
of atmospheric electricity. Another feature of this study also, taking advantage of the
electrical changes in the atmosphere before the earthquake, is to predict
earthquakes.
Keywords: Atmospheric electricity, lightning, JFET, earthquakes, global

atmospheric electricity.
1.Introduction terrestrial sources produce cluster ions which

make the atmosphere weakly electrically
The Earth’s atmosphere is continually conductive. These ions flow vertically because
electrified. Our understanding of the global of the vertical potential difference, causing the
atmospheric electrical circuit has improved air-Earth conduction current density, Jc, of
since its discovery in the early twentieth order 10-12Am-2.The total electrical resistance
century, although this area of geophysics still for a unit area of the atmospheric column from
provides both theoretical and experimental the surface to the ionosphere is called the
scientific challenges. The variability of the columnar resistance, Rc. The ionosphere-Earth
electric field and current density measured at potential difference (U1), columnar resistance
the surface is attributed to meteorological and conduction current area related by Ohms
sources, both global and local in origin. By law,
investigating the variability in electrical
parameters under different atmospheric U1=jc.Rc
conditions, these global and local sources can
be separated. As a consequence, information
on global thunderstorm and shower cloud
The physical units of RC are
activity can be retrieved, which is of direct 2
Ohm.meter (or more commonly
relevance to research on global climate change. 2
Pico.Ohm.meter due to the inherently large
Thunderstorms, earthquake fault and values).This is because RC is the integral of
shower clouds cause separation of electric resistivity .U1represents the electric potential
charge between the ground and ionosphere, an of a layer at a height above the surface (the
electrically conductive layer about 60 km zero potential reference) so the units are in
above the surface. This charge separation volts(usually quoted in KV).
causes the ionosphere to have a potential (U1)
of approximately +300Kv with respect to the
surface. Ionization from cosmic rays and

II-140
the atmosphere removes small ions by

attachment. An increase in aerosol number
concentration therefore reduces the ion number
concentration and decreases the air
conductivity. A change in aerosol number
concentration subsequently modifies the total
columnar resistance (Rc) between the
ionosphere and the surface, and therefore Jc
directly through and consequently has an
indirect effect on PG through .
Of the quantities which can be measured

at the surface, the air-Earth conduction current
density (JC) presents one of the most
Fig. 1. Basic components of global
fundamental parameters of the global circuit
atmospheric circuit
(Chalmers, 1967). It is therefore of great
importance in this study, with its continuo us
measurement being one of the aims of this
By convention, the electric field (E) is thesis. The conduction current density is one of
referred to as the potential gradient, defined as several components contributing to the total
the negative of electric field: potential gradient current density, JS received by a horizontal
given by conducting electrode at the earth’s surface,
F=-E electrically isolated from the ground. J S
comprises contributions from turbulence JT,
At the surface, the PG arises because of conduction JC, displacement JD and
JC flowing through the electrically conductive precipitation Jp,
air. It is therefore JC that is the parameter that
permits the effect of the global circuit to be
measured at the surface, either directly through JS=JC+JD+JT+JP
measurement of Jc itself, or by PG. However,
PG is also a function of the local air 2. Experimental study
conductivity (ɗT). Away from sources of
charge separation, the air conductivity In this study, the electrical changes the
(ɗT),potential gradient (F) and conduction electrical charge in the atmosphere with a
current density are related by Ohm’s Law: sensitive FET sensor've recorded. When the
records, especially at sundown and sunrise
F=JC/ ɗT varies the electric field in the atmosphere. This
result is evidence that a significant change in
The atmosphere is positively charged the sun's atmosphere electric field. Change
with respect to the ground during fair weather features is the sunrise and sunset.
and fraction to the fault system. This produces
a downward pointing (negative) electric field Another important influence,Lightning effect.
(E) so PG is positive for fair weather Lightning effect, other solar influence is
conditions. In the fair –weather part of the dominant or effective. Lightning events, they
circuit, small ions dominate the charge make a greater impact on global atmospheric
transport since that have a large electrical cycle. Effects of lightning currents are changes
mobility. Therefore an increase in small ion from cloud to cloud. The other effect is from
concentration will increase the air conductivity cloud to cloud and the ground. Until the
by providing more charge carriers. Aerosol in

II-141
changes in atmospheric electrical activity, 24

hours is seen in the records.
Fig. 2. Block diagram of the measurement

system used in the experiment.
3. Results and Discussion

Fig. 4. 24-hour electric field change.( A 10-
This study is provided below the records
minute lightning transitions can clearly be
obtained. Recordings were obtained on
seen.)
uninterrupted for 24 hours. Register in
accordance with the impact of the effects
created by the sun seems to be more dominant
lightning. 4. Conclusion
From this experimental study,very significant

results were obtained. In this study, it was
concluded that the power structure of the
atmosphere has been observed many alternate.
Results obtained from the experimental study
can be expressed as follows.
1. Global atmospheric, other electrically

varies from seasonal variations.
2. In electrical changes, the effect of the
sun is changing the ion density.
3. Lightning currents cause the biggest
change in the atmosphere.
Fig. 3. Changes in the electric field (Heavy 4. Lightning strokes, gives rise to the
rain and lightning two-hour recording) formation of ions in the atmosphere
greatly.
Changing the direction of flow will change the 5. Lightning currents, thus affecting a
direction of the electric field change. Here it is large proportion of the global
seen that the lightning is cloud-to-cloud from atmospheric cycle is changing the
the ground or floor. electrical parameters of the
atmosphere.
6. After the lightning impact, depending
on the severity of impact parameters,
which come to its normal state after a
certain time.

II-142
References 5.Agro meteorology Group-1997

6.Svensmark and Friis-Christensen,1997
1. Doctoral studies, Metin SALTIK 7.By Donald E.olson,The evidence for Auroral
2. electrical-engineering-portal 2014 Effect on Atmospheric Electricity
3.Mike Bertone,protect electrical equipment 8.Nehar.1967 and Indian Institude of Tropical
from solar flare,2014
Meteorology
4.V.Gopalakrishnan,S.D.Pawar,P.Murugavel,K
9.Bolestad and Nashewski; Pearson Publishing
irankumar P.Jhare,Agust 2011

II-143

INFORMATICS AND COMPUTER SYSTEMS AND TECHNOLOGIES
Информатика и компютърни системи и технологии

RSA ENCRYPTION USING MORE THAN TWO

PRIME NUMBERS
TARIK YERLİKAYA, EMRECAN ADLI
Abstract: Fast and easy usable internet services are one of the gaining of 21.th Century.
Online shopping, online messaging even online bank services are most popular ones. Also
new attacking techniques have been evolved. To prevent attack to similar services, we have
to strengthen our encryption systems. One of them is RSA Algorithm. In this paper we try to
increase complexity RSA Algorithm to make it more secure.
Key words: RSA, Encryption, Decryption
1. Introduction encryption/decryption key [2]. To increase security,

With the developments in technology, numbers must be selected very high [3],[4].
computers and the Internet have become an
important part of our lives, which also has increased In this paper tried to use more than two
the number of operations carried out on the Internet. prime numbers instead of two to calculate key
Therefore, it is important to provide security for generation [5].
users to help them with their commercial, official or
1.1. Base of RSA Encryption
private transactions.
The RSA algorithm was publicly described
Cryptology and authentication is very in 1978 by Ron Rivest, Adi Shamir, and Leonard
important for providing security. There are two Adleman at MIT; the letters RSA are the initials of
ways of cryptography, which are symmetrical and their surnames, listed in the same order as on the
asymmetrical. Asymmetrical encrypting is based on paper.
prime numbers. Very large or very long numbers
are needed to be used as “keys” in cryptographic The RSA algorithm involves three steps:
applications key generation, encryption and decryption. Big
prime numbers are selected in the key generation
The most used communication systems are step. Then encryption and decryption rules are
the public ones. In the public communication provided using generated key. In the encryption step
systems there is a risk of reading, recording or the unencrypted message powered by e and mod n
changing of the messages by the unauthorized is calculated, it gives encrypted message (c = me
people.[6] mod n). In the decryption step the encrypted
message powered by d and mod n is calculated, it
If we want to secure a message, we must gives the main message(m = cd mod n ).
make it unreadable or inexplicable. Only receiver
must understand what the message is. So before  M is the message
sending the message to the receiver, it must be  C is the encrypted message.
encrypted. To understand the message by receiver  K is the key.
it must be decrypted [1]. A lot of algorithms exist to  ek is the encryption rule and dk is
provide this process. One of them is RSA the decryption rule which provide k
Algorithm. RSA crypto system, also using for Є K rule.
generating digital signatures.[7]  ek : P  C is an encryption
function and dk : C  P is a
RSA Algorithm developed by Ron Rivest, decryption function.
Adi Shamir and Leonard Adleman in 1978. The
algorithm uses two prime numbers to generate
1.2. Existing RSA Algorithm
II-145
 Key generation : Select two Compute

different primes, p and q c = me mod n
 Calculate n = p * q
 Calculate Q(n) = ( p – 1 )*( q – 1 )
Compute
 Select an integer e which provides 1 m = cd mod n
< q < Q(n) ; gcd(e, Q(n)) = 1
 Calculate private key d = e-1 (mod
Q(n)), then public key is { e , n } END
and private key is { d , n }
 Encryption c = me mod n
 Decryption m = cd mod n
The gcd( e , Q(n)) = 1 function is Euclidean 1.5. Example

Algorithm. Used for e and Q(n) coprime.
 p[1] = 11, p[2] = 7, p[3] = 5, p[4] =
1.3. Designed RSA Algorithm 3
 n = 11 * 7 * 5 * 3 = 1155
 Key generation : Insert different  Q(n) = 10 * 6 * 4 * 2 = 480
prime numbers in array p[l]  e = 419 , 1 < e < 480 ; gcd(e , Q(n)
 Calculate n = p[1] * p[2] * … p[l] )=1
 Calculate Q(n) = (p[1] – 1)*( p[2] –  d = 419-1 (mod 480) = 299 than
1 )*…(p[l] – 1 )  Public Key = { 419 , 1155 }
 Select an integer e which provides 1  Private Key = { 299 , 1155 }
< e < Q(n) ; gcd(e, Q(n)) = 1  Normal message is m = 12
 Calculate private key d = e-1 (mod  c = 12419 mod 1155 = 738 (c is
Q(n)), then public key is { e , n } encrypted message)
and private key is { d , n }  Decrypted message is
 Encryption c = me mod n  m = 738299 mod 1155 = 12 (m is
 Decryption m = cd mod n decrypted message which is normal
message)
1.4. Flow Chart

1.6. Application Program
START
Proposed algorithm is written as a c#
application. Results can easily verified by the
application. Prime numbers can be added as
Input Prime Numbers requested. Then the numbers can processed. n and
p[l] Q(n) values are displayed. After that, e value must
be selected to process private and public keys. Then
message can be encrypted.
n = p[1]*p[2]*…*p[l]
Q(n)=(p[1]-1)*(p[2]-1)*…*(p[l]-1)
Input e Number
1 < e < Q(n)
gcd (e, Q(n)) = 1
Compute
e*d= 1 mod Q(n)
Input Message (m)

II-146
must be considered to cause more process time and
memory space requirements.
REFERENCES
1. D. R. Stinson, Cryptography, Theory and
Practice, 3rd Ed., CRC Press, Boca Raton,FL
(2002).
2. R. Rivest, A. Shamir and L. Adleman, A
Method for Obtaining Digital Signatures and
Public-Key Cryptosystems, Communications
of ACM, 21(2), 120-126 (1978).
3. B. P. U. Ivy, P. Mandiwa and M. Kumar, A
Modified RSA Cryptosystem Based on ‘n’
Prime Numbers, Int. J. Engg. Comput. Sci.,
1(2), 62-66 (2012).
4. S. Pund and C. Desai, Implementation of RSA
Algorithm using Mersenne Prime, Int. J.
Networking & Parallel Computing, 1(2), 33-41
(2013).
5. Ch. J. L. Padmaja, V. S. Bhagavan and B.
Srinivas, RSA Encryption Using Three
Fig. 1. Example 1.5 in program Mersenne Primes 14(4), 2273-2278, (2016)
6. YERLİKAYA, T., GENÇOĞLU, H., EMİR,
2. Conclusion M. K., ÇANKAYA, M., & BULUŞ, E. RSA
Using more than two prime numbers ŞİFRELEME ALGORİTMASI VE
increase n value which can increase algorithm ARİTMETİK MODÜL UYGULAMASI.
complexity and security. Especially if small prime 7. Yerlikaya, T., Buluş, E., & Buluş, N. (2006).
numbers must be used, it gives n and Q(n) value ASİMETRİK ŞİFRELEME
that is higher than existing RSA algorithm. Also ALGORİTMALARINDA ANAHTAR
designed RSA algorithm can easily implement in DEĞİŞİM SİSTEMLERİ.
existing RSA algorithm because they share same
formulas. However, all those big prime numbers
PASSWORD CRACKING ATTACKS

ŞENOL ŞEN, TARIK YERLIKAYA, TRAKYA UNIVERSITY
Abstract: Today, password cracking attacks are one of the most dangerous attacks. We use
passwords to login information systems like our computers, social accounts, etc. Password
is the biggest and mostly only barrier between a hacker and an information system. The
only thing that can be the weakest link in the security chain of information security is weak
passwords. The reason for this is the human factor. To create passwords with the names of
our hobbies or beloved ones that everyone knows are main mistakes which people usually
make. Even the most secured institutions can be exposed to these kind of attacks because
they are too lazy or they ignore. From the expert user, using the most powerful security
tools, to the end user connecting via home computer to internet, the first thing to be learned
is how to create secure and robust passwords. A password cracking example has been
given in this study, by explaining how password attacks are carry out and the popular main
tools used in these attacks. Also things to create a strong password has been explained at
the last part of our study.
Key words: password cracking, password attack
1. Introduction weakest link in the security chain is weak

Close your eyes and think. After a nice day, passwords and codes. Passwords are usually
you came home and it’s time to sleep. You slept by encrypted and stored as hash rather than as plain
leaving all the doors and windows of your house text. When a hash is generated, the text in variable
open. You may have survived that day without length is converted to fixed length, and when this
coming to somebody’s attention, but what about convertion is in process, unique texts consisting of
other day? Will you be able to get the same chance mixed numbers and digits are produced due to the
the next day? Stepping into the world of informatics utilization of a one way mathematical function. For
in an environment where security is not provided is example, the user information stored in the database
similar to leaving your door and windows open with or the user information stored in the operating
the thought of "I hope someone will not notice" systems are not stored directly in plain text.
your house. Considering the possibility that someone can read
Your password is the first obstacle between the information, it is kept in a hashed form, ie a
the attacker and your information system. The first kind of encrypted form. Keeping information in this
attack type that attackers will attempt before way makes things more difficult for attackers,
attacking a system is password cracking attacks. If because the hashes have a one-way mathematical
we see the password as the key of the house door connection with the original text. It is often a
here and it’s under the door mat, there is no need to difficult task to convert a text that is reached to its
strive to open the door. Because the key is in the Hash into its original form.
place we guess, that is under the door mat. Weak Since Hashes are one-way mathematical
and predictable passwords are like keys under the functions, the task to find the original text is to do a
door mat as in our example. These can be found and trial from a list of words. It is to try until right hash
break within the seconds in today's technological matching is ensured by comparison via the
possibilities. To protect against these attacks, we conversion of the text into hash. So if the password
must create strong passwords and securely store our is in our word list, the password can be found at the
passwords. end of this trial. Otherwise, it will not be found[1].
Also, while a hash is being created, a small
2. Password change in the text changes the entire structure of the
Passwords are used in many fields for Hash. For example, if we encode the "trakya" text
identity and authorization checking. The only with md5 algorithm, we get
element in information security that may be the 2c8135467f5423f06d297d4d276456f3 hash. We get

II-148
the hash of d9052fa7adb01862abd6d81be4e66ce7 passwords are stored on the system. User names and
by enlarging the letter T and using the md5 hash of passwords may be kept as text or as encrypted and
"Trakya" again. It is obvious that there is no hashed. While the redirecting is made, the summary
connection between the two hashes when they are value of the password is taken and compared with
compared. Therefore, it is not possible to the summary of the password, entered by the user,
understand that hash is close to being solved. We previously saved on the system. Accordingly, the
can group the methods that are used to capture the user is authorized. There are methods such as MD5,
password into 4 main headings[2]. SHA, NTLM for summary processing[2].
2.1. Active Online Attacks In this type of attack, three different
2.2. Passive Online Attacks methods can be mentioned as brute-force,
2.3. Offline Attacks dictionary and rainbow tables.
2.4. Non Technical Attacks Rainbow Table: Because many system passwords
are stored in hash form, even if this data is reached,
2.1. Active Online Attacks data must be detected by matching. One way to do
Brute-force and dictionary attacks are listed this is to find through comparison by converting the
in this kind of attack. Brute-force attack is a simple words into hash with a wordlist in hand. Also,
but effective method. It is based on the testing of all another way is to use the rainbow table, which has
possibilities. Optionally, all possible combinations hash in it for each word and has tables ready for
of letters, numbers, special symbols that may exist direct comparison.
in the specified length are tried. Weak and
predictable passwords can be broken easily through 2.4. Non Technical Attacks
brute - force. The important thing here is the time. We can group the attacks in this title as
100% success can be achieved within sufficient social engineering and forecasting. Social
time. engineering is to obtain information from victims or
In the case of the dictionary attack, the make them to do the desired processes by using
combinations that are tried are a predetermined list. influence and persuasion methods. It is basically a
This is the main difference with brute force. More type of assault that is carried out with the use of
targeted, customized dictionaries are needed to be trust, fear and compassion which are readily
prepared. In addition, different combinations can be available in human nature. Generally being
tried with numbers and special characters instead of considered as the weakest link of the security chain,
the letters in the password. human factor is one of the biggest threats despite
Brute force can be largely avooided through network filtering devices, antivirus softwares and
policies that detect and block false password all the security methods.
attempts. On the other hand, the estimation method is a
method of finding the password by prediction by
2.2. Passive Online Attacks means of the usage of the information (name,
This attack happens in 2 ways: Man-in-the- surname, telephone number, supported team, etc).
middle and sniffing. In the case of the man-in-the-
middle kind of attack, environment and the traffic 3. Tools that can be used during an attack
on the targeted computers are tapped and the Ophcrack: A free application that works
password is tried to be obtained. In this type of with rainbow table logic. Windows password can be
attack sniffing tools may be used as well. used to break hash.
On the other hand, sniffing attacks are Cain Abel: One of the most popular
effective when being connected on the same password cracking tools developed for Windows. It
network and in the case of using a networking tool can crack many hash such as NTLM, NTLMv2,
such as a hub in the system. Because the hub sends MD5, wireless, Oracle, MySQL, SQL Server,
a package to all ports, by this means, all devices on SHA1, SHA2, Cisco, VoIP.
the network can see the package sent. If such tools Brutus: Supports many authentication
as switches, bridges, etc., are used, since these tools methods only developed for Windows such as
only filter in a way to be sent to the destination port, HTTP, POP3, FTP, SMB, TELNET, IMAP, NNTP.
sniffing will not be effective in this environment. John The Ripper: A popular password
The tools used for such attacks are: Cain & Abel cracking tool developed with C language.
(Windows), ettercap, hash, Dsniff, SSLStrip. Cupp: A powerful tool for creating original
wordlists. It tries to make more relevant words by
2.3. Offline Attacks asking questions about the target user.
In this type of attack attempts, attackers try
to gain access to the spots where usernames and
II-149
Cewl: Wordlist creation tool. It aims to With the Hydra tool, brute force attacks
increase the relevancy of the words by examining against many systems (http, ssh, ftp, telnet, pop3,
the web page provided. smtp etc.) can be done[4].
Crunch: A very successful software used to
create wordlist in order to use brute force attacks.
Hydra: Working with dictionary logic, a
password-cracking tool that can execute remote
attack for more than 50 protocols. It is not only
limited to the forms in the web services but it can
also attack many services such as MAP, SMB,
HTTP, VNC, MS-SQL, MySQL, SMTP and SSH.
It can be run through the Linux terminal as well as
in the user interface.
Medusa: A fast password cracking tool that
can execute parallel scan and is very easy to syntax.
It provides an opportunity to test services such as
HTTP, HTTP Form, IMAP, MS-SQL, MySQL,
NCP, Oracle, POP3, AFP, CVS.
It is important that the words in the wordlist
used for the attack contain the language, the level of
relevancy and the most used passwords. We are Fig. 1. The contents of the created passlist.txt file
more likely to guess the passwords that someone
you know will use. Most people choose either the
most common insecure passwords to remember
easily, or create a password with the information
like animal name, birth date and phone number.
Therefore, it is more logical to create a wordlist in
accordance with the individiual. Thus, the tool
named Cupp which asks questions about the person
and Cewl which creates words by examining the
web sites can be used[1].
4. Brute force attack on SSH service
The most important step to take before
attack is to collect information about the system,
system administrator, or system users to be
attacked. If we want to break a user's password, we
need to get information about that user and write it
into the dictionary list, which we will use on the
attack. In addition to user information, we also need
to create a file of possible password information.
Crunch is one of the most useful tools we can use
when creating password files. This tool allows us to
create various combinations of the characters given
to it[3].
The sample password file we have created Fig. 2. Example of attack against SSH service with
for our example is shown in “Fig. 1”. Hydra
In our example, we only need to enter the
5. Powerful Password Creation
password list because we know the name of the user
in our target to Hydra. Therefore, we enter the name Strong passwords consist of random
of that user and the IP of the computer that is characters and strings. The two most important rules
running ssh service with -l instead of -L. We run the for creating a strong password are the length and
Hydra tool with these parameters and find the root complexity of the passwords. An ideal password is
password of the SSH service as 1qaz2wsx, as long and contains letters, punctuation marks,
shown in “Fig. 2”. symbols and numbers at the same time[5]. The most
important points to note about creating a secure
password are:

II-150
1. It must be longer than eight characters. Short systems. It has become one of the easiest ways for
passwords are easier to resolve than long hackers to get access to our computer or network.
passwords. Along with social networks, password
2. It must consist of letters, numbers and symbols. security has increased extremely in line. As we
Consecutive or repeating combinations (such as write our own information about ourselves,
"12345678," "222222," "abcdefg") or letters that passwords present great importance with the
stand side by side on the keyboard should not be increase of these environments. Of course,
used. There should not be any known words that use password security does not go beyond just creating
numbers instead of letters. (Such as "M1cr0 $ 0ft" your password difficult. Protection and keeping safe
or "P @ ssw0rd"). Unfortunately, hackers know are also important. You should never save your
these tricks too. password on your or another computer and also
3. It must be easy for you to remember, but hard to phones and tablets.
guess for others. Protecting the critical information in
4. Your sign-in name should not be your spouse’s computer systems by recollective, repeatable and
name or your birthday. simple passwords, is increasing the risk of system
5. There should not be words from any languages breakage. When it is thought the next generation
which exist in any dictionary. Hackers use computers will control by voice commands, open
sophisticated tools that can quickly guess passwords and close with visual commands, it’s certain that
using words in various dictionaries and also their stronger passwords will be needed to store high
reverse spellings. importance information. It is a fact that there will be
6. It must be easy to remember. There should be no systems which will support all of these and
random combinations of letters, numbers and biometric-based passwords and these passwords can
symbols that must be taken note down, misspelled be obtained by using features such as fingerprints,
or should be found and use by others. iris or face in the near future[6].
7. Under any circumstances, you should not share
your password with anyone else. You may be REFERENCES
meticulous, but the person who knows your
password may not be careful and may leave you in a 1. Turkoglu, C., Password cracking,
difficult situation. If we have to share, as a method http://blog.btrisk.com
changing the password would be a solution. 2. Onal, H., Akbas, E., Password cracking
8. Update your password on a regular basis. This attacks in penetration tests,
method is a bit differs from the structure of the https://www.slideshare.net/ bgasecurity/.
encrypted account, especially if the account 3. Altuntas, A. (2016). Kali Linux, 275-289,
contains your critical information this method must Kodlab, İstanbul.
be used. For example, most banks, various 4. Hydra, https://www.thc.org/thc-hydra/.
communications companies apply this for their 5. Dogu, K., Keep your password strong,
users and send new passwords to their users. http://www.bilgiguvenligi.gov.tr.
9. Do not use your passwords on computers you do 6. Canbek G., Sagiroglu Ş.(2006), Information
not know. One of the most common methods used and Computer Security Spyware and
by identity thieves is to install spywares such as a safeguarding methods,82-83, Grafiker Ltd.Sti.,
keylogger on target computers and access the Ankara.
information of persons who use the computer.
10. Never send your password by email to any Organization: Trakya University
address. Some malicious people may look like Address: Information Technologies
Internet sites and could ask for your password, Department Balkan Campus
never send your password in this way[5]. Edirne/TURKEY
6. Conclusion Phone: +905367409955
E-mail: tarikyer@trakya.edu.tr
Today, passwords are the only form E-mail: senolsen@trakya.edu.tr
provided for the security of websites and computer
IMAGE PROCESSING ON ANDROID DEVICES

WITH OPENCV
CEM TASKIN, ALTUG YIGIT, DENIZ TASKIN, AYSE CELIK TASKIN
Abstract: Nowadays, computer vision is a popular approach in computer science. Many

applications running on personal computers use computer vision libraries in order to have
object detection capability for many years. Today mobile devices have powerful hardware
like quad core cpus, high density cameras and high speed internet connection. With the
hardware achievements in mobile devices, programmers developed mobile versions of
computer vision libraries. In this paper, we are testing an Android application using
computer vision library for distinguishing colors. This sample application works on real
time camera and distinguish a color from others in the scene that was selected by user.
Key words: image processing, mobile device, android, color, filtering, opencv
1. Introduction Restricting the color information merely to

Object detection is one of the most RGB components is a simple abstraction that
challenging problems in computer vision. It is dismisses the information available within the color
difficult to detect due to the significant amount of object. A color is fully defined by its complete
variation between images belonging to the same wavelength response, whereas the RGB color space
object category. Other factors, such as changes in represents only three wavelengths. The perceived
viewpoint and scale, illumination, partial occlusions color also depends on the illumination condition,
and multiple instances further complicate the viewing angle, and sensor type. As a consequence,
problem of object detection. Most state-of-the-art efficient color image processing requires an
approaches to object detection rely on intensity- adequate color representation. Different color
based features that ignore color information in the spaces can be used to represent various color
image. Color has been shown to yield excellent components, with different degrees of
results in combination with shape features for image interdependency among them. Of the four classic
classification. [1] color spaces, the hue/saturation/value (HSV),
Usually computer vision libraries use YCbCr , L*a*b*, and RGB shown as Figure 1. [6]
captured images as input. But in real time
applications, the input is always a collection of
continuous frames (pictures). It is a main problem
processing these frames in short times. And also it
is very important to give the result of process as
soon as possible. Computer vision techniques like
3D segmentation and modelling is not suitable for
real time processing. In this study, color filtering
method is used for distinguishing colors and tested
on an Android mobile device.
2. Color Filtering
In color filtering method, determining the
correct threshold value is the most important fact on
accuracy. In order to get consistent results,
appropriate color space must be selected before
processing images.

II-152
and is now maintained by Itseez. The library is
cross-platform and free for use under the open-
source BSD license. The image processing
algorithms are implemented using OpenCV.
OpenCV was designed for high computational
efficiency with a strong focus on real time
applications. OpenCV was written with optimized
C. [3,4] OpenCV library is a common library used
in image processing applications.
Table 1. Worldwide Smartphone Sales by

Operating System
Fig. 1. HSV, YCbCr , L*a*b*, and RGB Color

Space
The HSV (Hue-Saturation-Value) theory is

the most common representation of points in an
RGB (Red-Green-Blue) color technical model.
Computer graphics pioneers developed the HSV
model in the 1970s for computer graphics It is possible to use this library on Android
applications (A. R. Smith in 1978, also in the same Studio by adding OpenCV Android Software.
issue, A. Joblove and H. Greenberg). A HSV theory Development Kit (SDK). This SDK can be
is used today in color pickers, in image editing downloaded from the
software, and less commonly in image analysis and http://opencv.org/downloads.html.
computer vision. [7] The HSV color space is more As seen on Table 1, Android OS is the
related to human color perception. [8] common operating system used by mobile devices.
HSV color space separating luminance Android as a development platform use the Java
component with chrominance component. The Standard Edition with additional Android classes.
luminance component lies in V (intensity value) and Applications are developed in the Java language
the chrominance component lies in H (Hue) and S using Android Software Development Kit (SDK).
(Saturation). In HSV model, Hue defines the kind of For a low level interaction with hardware, a
color like red, yellow, green, blue, or the modified Linux kernel is used [2].
combination between those colors. Hue value is
between 0-360. Saturation gives a measure of the
4. OpenCV on Andoroid OS
degree by which a pure color is diluted by white
light, and value show the intensity of light-dark In our study, we are testing an Android
color.[9] application using OpenCV for real time color based
Color is a strong tool for image image segmentation. The flowchart of the
segmentation. Color-based segmentation is better application is shown on Figure 2. OpenCV’s
than edge-based and luminance histogramming onCameraFrame method captures frames in RGBA
techniques because color is computational color space. These frames are converted into 8-bit
inexpensive, and it can give more information than 4-channel pixel matrix. After that step, RGBA to
a luminance-only image or an edge-segmented HSV conversion are applied.
image which need more computational resources, so 5. Application Flow Chart
they make real time application systems hard to
realize.[5] In the RGBA color space R (Red), G
(Green), B (Blue), A (alpha) channels are available.
3. OpenCV and Android For each channel, color values are stored in 8 bits
OpenCV (Open Source Computer Vision) is and has a range between 0 and 255. Alpha channel
a library of programming functions mainly aimed at is ignored on conversion from RGBA to HSV. In
real-time computer vision. Originally developed by the first step of converting we divide each color
Intel's research center in Nizhny Novgorod value with 255 and we obtain a conversion value
(Russia), it was later supported by Willow Garage between 0 and 1.
II-153
Original image
Converted Image
Fig. 3. RGB to HSV Conversion

Fig. 2. Flow Chart of Our Application
5.1. Threshold
In the next step, we are using following
formulas to calculate HSV values [11]. After RGB to HSV conversion, we select a
color for filtering. In our test, we select Red color.
Before the threshold filtering, two color range
𝑉𝑉 ← 𝑚𝑚𝑚𝑚𝑚𝑚(𝑅𝑅, 𝐺𝐺, 𝐵𝐵) values are selected (The low and high color tone).
For threshold filter we use OpenCV’s inRange
min(𝑅𝑅, 𝐺𝐺, 𝐵𝐵) function. The formula used by this function is
𝑉𝑉 − 𝐼𝐼𝐼𝐼 𝑉𝑉 ≠ 0 shown in Formula 2.
𝑆𝑆 ← { 𝑉𝑉 }
0 𝑜𝑜𝑜𝑜ℎ𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒
(1) (2)
60(𝐺𝐺 − 𝐵𝐵) 1 𝐼𝐼𝐼𝐼 𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼 ∈ 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅
𝐼𝐼𝐼𝐼 𝑉𝑉 = 𝑅𝑅 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 ← { }
𝑉𝑉 − min(𝑅𝑅𝑅𝑅𝑅𝑅)
0 𝐼𝐼𝐼𝐼 𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼 ∉ 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅
60(𝐵𝐵 − 𝑅𝑅)
𝐻𝐻 ← 120 + 𝐼𝐼𝐼𝐼 𝑉𝑉 = 𝐺𝐺
𝑉𝑉 − min(𝑅𝑅𝑅𝑅𝑅𝑅) In Figure 4, the output of this function
applied to sample image is shown.
60(𝑅𝑅 − 𝐺𝐺)
240 + 𝐼𝐼𝐼𝐼 𝑉𝑉 = 𝐵𝐵
{ 𝑉𝑉 − min(𝑅𝑅𝑅𝑅𝑅𝑅) }
This conversion can be performed in

OpenCV with function cvtColor. In Figure 3 a
conversion from RGB to HSV is shown
Fig. 4. Output After Threshold Filter

II-154
5.2. Weighted Sum of Two Arrays Fig. 5. Output After Gaussian Blur Filter
In order to increase performance of color 5.4. Contour Marking

filtering, weighted sum of two threshold output
matrix is calculated. In this calculation operation, After applying Gaussian Blur filter we
two matrix has equal weight (50% and 50%). We obtain a smooth image. In this output image, we
use Formula 3 for this operation. In OpenCV, for find the positions of white pixels and the
calculating the weighted sum of two threshold relationship of each other.
output matrix,addWeighted function is used. Contours can be explained simply as a
curve joining all the continuous points (along the
boundary), having same color or intensity. The
(3) contours are a useful tool for shape analysis and
object detection and recognition. In OpenCV that
𝑑𝑑𝑑𝑑𝑑𝑑(𝐼𝐼) = 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠(𝑠𝑠𝑠𝑠𝑠𝑠1(𝐼𝐼) ∗ 𝑎𝑎𝑎𝑎𝑎𝑎ℎ𝑎𝑎 can be applied by calling the findContours function.
+ 𝑠𝑠𝑠𝑠𝑠𝑠2(𝐼𝐼) ∗ 𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 + 𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔) This function has three arguments. First one is
source image. The second is contour retrieval mode.
Parameters: Third parameter is contour approximation method.
src1 – first input array. And it outputs the contours and hierarchy. [13]
alpha – weight of the first array elements Contour retrieval mode selected for this Android
src2 – second input array application is CV RETR TREE. In this mode all of
beta – weight of the second array elements the contours output vector are obtained with
dst – output array hierarchical relationships. [12] The approximation
gamma – scalar added to each sum.[11] method selected for this Android application is CV
CHAIN APPROX SIMPLE.
By using CV CHAIN APPROX
5.3. Gaussian Blur approximation method all redundant points are
removed and that saves memory which is very
We use Gaussian Blur to reduce noise of the important for mobile devices.
weighted sum value obtained. When the Gaussian After this step, the pixel locations of the
blur applied, pixel values apart from the others rectangle are obtained. In OpenCV rectangle
become closer. We use Formula 4 for this filter. In function is used for drawing a rectangle. In Figure
OpenCV, for applying Gaussian Blur filter 6, the result of color filtering is shown. In our test,
GaussianBlur function is used. the application follows the color and draws contours
in real time video.
(4)
2
−(𝑥𝑥−𝜇𝜇𝑥𝑥 )2 −(𝑦𝑦−𝜇𝜇𝑦𝑦 )
+
2𝜎𝜎𝑥𝑥2 2𝜎𝜎𝑦𝑦2
𝐺𝐺0 (𝑥𝑥, 𝑦𝑦) = 𝐴𝐴𝑒𝑒
In Figure 5, the output of Gaussian Blur

filter applied to sample image is shown.
Fig. 6. Result Of Color Filtering
6. Performance Testing
Application performance was tested
with an Intel® Atom™ Z2560 1.6 GHz CPU
equipped mobile device on Android Studio.
The results of CPU usage, Memory usage and
FPS(Frame Per Second) are shown in Figure 7.
II-155
REFERENCES
1. Fahad Shahbaz Khan, Rao Muhammad Anwer,

Joost van de Weijer, Andrew D. Bagdanov,
Maria Vanrell, Antonio M. Lopez. Color
attributes for object detection. In CVPR, 2012.
Threshold Filter performance 2. Damir Demirović, Amira Šerifović-Trbalić,
FPS value is between 17.50-21.0 at 640x480 pixels Naser Prljaˇca, Philippe C. Cattin. Evaluation
of image processing algorithms on ARM
powered mobile devices. MIPRO, 2014.
3. https://en.wikipedia.org/wiki/OpenCV#cite_no
te-2
4. Deepthi.R.S, S.Sankaraiah. Implementation of
mobile platform using qt and open cv for
Gaussian Blur filter performance
image processing applications. IEEE
FPS value is between 15 and 17.50 at 640x480
Conference, 2011.
pixels
5. Houssem Lahiani, Mohamed Elleuch, Monji
Kherallah. Real time hand gesture recognition
system for android devices. ISDA, 2015.
6. Hossein Nejati, Victor Pomponiu,Thanh-Toan
Do, Yiren Zhou,Sahar Iravani, and Ngai-Man
Cheung. Smartphone and mobile image
Contour Marking processing for assisted living. IEEE Signal
FPS value is between 9.0 and 16.50 at 640x480 Processing Magazine, 2016.
pixels 7. Ján Haluska, On fields inspired with the polar
HSV -- RGB theory of colour. In arXiv, 2015.
Fig. 7. Test Results 8. Alberto Albiol, Luis Torres, Edward J. Delp.
Optimum color spaces for skin detection. IEEE
7. Conclusion Image Processing Conference, 2001.
As a result of our mobile application 9. Priyanto Hidayatullah, Miftahuddin Zuhdi.
tests, we can say that mobile devices has Color-texture based object tracking using HSV
capable of color filtering in real time. Our color space and local binary pattern. In
future work is to test real time object detection International Journal on Electrical Engineering
methods that use deep learning techniques. and Informatics, 2015.
10. http://docs.opencv.org/2.4/modules/imgproc/do
c/miscellaneous_transformations.html
11. http://docs.opencv.org/2.4/modules/core/doc/o
perations_on_arrays.html
12. http://docs.opencv.org/2.4/modules/imgproc/do
c/structural_analysis_and_shape_descriptors.ht
ml
13. http://docs.opencv.org/trunk/d4/d73/tutorial_py
_contours_begin.html

COMMON ATTACKS TYPES AND

PENETRATION TESTING ON WI-FI NETWORKS
OĞUZ KIRAT, TARIK YERLİKAYA
Abstract: Wi-Fi is widely used wireless technology but not secure if it is not configured
properly. This paper discusses Wi-Fi networks’ weaknesses and possible attack vectors.
Penetration testing methods and tools to measure security of certain wireless networks are
also discussed.
Key words: Wi-Fi, wireless, networks, penetration testing, network security
1. Introduction network to everyone sniffing the network and

Wi-Fi (Wireless Fidelity) or 802.11 was should be considered non-secure.
firstly introduced back in 1997 [1] as a low cost,
user-friendly and simple wireless local area network
solution. It is widely used as a network solution in
universities, businesses and homes. Many devices
are Wi-Fi enabled including mobile phones, tablets,
computers and IoT devices. Though, Wi-Fi is not
always secure, if not configured properly. We will
discuss common attack vectors in this paper.
2. Network monitoring, sniffing and
package capturing
Sniffing is a technique to capture network Fig. 1. MacOS™ Sniffer Tool

packets and analyze network traffic. Many wireless
networks are open to network sniffing 3. Package analysis on non-secured or
vulnerabilities, including Wi-Fi because packages decrypted Wi-Fi networks
are transferred over-the-air.
Wireless network traffic can easily be
To “sniff” a Wi-Fi network there is no need inspected on non-secured Wi-Fi networks. If users
to join it, that makes sniffing attacks hard to detect. connected networks uses insecure network
If the attacker has a supported Wi-Fi hardware that protocols (such as HTTP, FTP, Telnet) it can be
supports “monitor mode” he/she can capture monitored using tools and changed.
network packages. Wireshark is popular tool and available on
major operating systems to inspect network packets.
Sniffing tools are widely available: On the All the HTTP traffic (including passwords)
Microsoft® Windows® operating system WinPcap, can be monitored using Wireshark if traffic is not
and on the Linux® distribution airodump-ng and encrypted.
available for package capturing.
In MacOS™ operating system, wireless 4. MAC address spoofing and ARP table
network sniffer tool for network diagnostic is attacks
included.
Some networks on hotels and schools The Address Resolution Protocol (ARP) is a
provide no encryption and uses HTTP based user telecommunication protocol used for resolution of
authentication. Wireless networks that are not Internet layer addresses into link layer addresses, a
secured exposes all the traffic passing on the
II-157
critical function in computer networks. ARP was There are various DOS attack types on the
defined by RFC 826 in 1982 [2] WiFi networks such as TCP/SYN floods, TCP
Replay, Land Attack and teardrop.
ARP is used to link local IP addresses to
physical addresses (MAC address). airodump-ng
tool can be used to discover MAC addresses of 8. Password Cracking
access points and clients connected them. MAC 8.1. WEP Password Cracking
spoofing methods can be used to trick MAC based
limitations on the network and combined with ARP WEP’s main goal was to provide data
spoofing can be used for man in the middle and privacy by encapsulating data frames. [6]
DDOS attacks. [3] WEP uses 24 to 48 bit key (Initialization
Vectors) and CRC32 as the control algorithm.
It is also demonstrated that MAC spoofing WEP is considered insecure and can be cracked if
can be used to connect to most insecure Wi-Fi enough IVs captured.
networks even if they use time and MAC based aircrack-ng is tool developed to crack WEP
restrictions. [4] based passwords in minutes.
5. Session Hijacking
Session hijacking attacks can compromise 8.2. WPA/WPA2 Password Cracking
user accounts on Wi-Fi connections. Combined The weakest point of WPA/WPA2 is using
with package sniffing attackers can hijack, for the same pre-shared keys for all users. Which
example HTTP cookies, to steal account means it is still to open brute-force attacks of
information. network sniffers.
One of the popular proof-of-concept tools
about session hijacking is Firesheep which is an WPA2 brings Enterprise mode which can
extension for Mozilla® Firefox® web browser. It is strengthen security.
designed to steal session cookies of popular social-
network sites [5] (which were not using HTTPS
protocol back in the days) 8.3. WPS Based Methods
There possibility of attacking WPS registrar
PINs.
Reawer is one of the proof-of-concept tools
that can crack plain text WPA/WPA2 passwords in
4-10 hours.
Fig. 2. Wireshark can be used to view HTTP 9. Conclusion

session on the network.
Though Wi-Fi is widely used technology,
6. Man In The Middle Attacks there are many vulnerabilities and attack vectors
available to abuse connections.
If attacker can capture network packages Advanced firewall and authentication
and decrypt them, he/she can secretly capture and servers are required with strong encryption methods
relay the wireless communication between the (such as WPA2 Enterprise) required.
access point and clients.
Attackers can force clients to reconnect to
capture handshakes and use ARP poisoning REFERENCES
methods to act as a wireless access point. aireplay
and d-sploit are tools designed for man in the 1. IEEE standard for information technology-
middle attacks and packet injection. telecommunications and information exchange
between systems-local and metropolitan area
7. Denial of Service Attacks networks-specific requirements-part 11:
The purpose of the Denial of Service Wireless LAN medium access control (MAC)
Attacks (DOS) is slowing down or stopping a and physical layer (PHY) specifications. IEEE
running system. Std 802 11-1997. 1997:i-445.
2. David C. Plummer (November 1982). "RFC
826, An Ethernet Address Resolution Protocol

II-158
-- or -- Converting Network Protocol 5. Firesheep, A Firefox extension that
Addresses to 48.bit Ethernet Address for demonstrates HTTP session hijacking
Transmission on Ethernet Hardware". Internet attacks.[Online] Available:
Engineering Task Force, Network Working https://github.com/codebutler/firesheep
Group.[Online] Available: 6. Wang Shunman, Tao Ran, Wang Yue, Zhang
https://tools.ietf.org/html/rfc826 Ji. WLAN and it's security problems. Parallel
3. Ramachandran, Vivek & Nandi, Sukumar and Distributed Computing, Applications and
(2005). "Detecting ARP Spoofing: An Active Technologies, 2003 PDCAT'2003 Proceedings
Technique". In Jajodia, Suchil & Mazumdar, of the Fourth International Conference on.
Chandan. Information systems security: first 2003:241-244.
international conference, ICISS 2005, Kolkata,
India, December 19-21, 2005 : proceedings.
Birkhauser. p. 239. ISBN 978-3-540-30706-8.
4. McDonald, Kyle, Free Wifi, [Online] Authors’ contacts (obligatory) – mark-up
Available: https://github.com/kylemcdonald/Fr style „Contacts”
eeWifi Organization:
Address:
Phone (optional):
E-mail:

ALTERA AND XILINX TOOLS IN EDUCATION

ATANAS KOSTADINOV
Abstract: In this paper, Altera and Xilinx tools used in Technical University - Sofia,
Plovdiv branch are presented. Initially, Altera (now the Intel Programmable Solutions
Group) tools are applied in Reconfigurable Logic and VLSI (Very-large-scale integration)
design courses. Recently, it has been received donation from Xilinx university program
(XUP) in the form of FPGA (Field-programmable gate array) design kits and Vivado
Design Suite: System Edition. Now tools from both reconfigurable logic market leaders are
used as themes in bachelor's and master's degree syllabuses.
Key words: Altera and Xilinx tools, Xilinx university program, Vivado Design Suite:
System Edition, Nexys4 DDR
1. Introduction reconfigurable integrated circuits and tools used

The reconfigurable integrated circuits from the beginning up to now.
especially CPLD (Complex programmable logic
2. Altera (now the Intel Programmable
device) and FPGA (Field-programmable gate array)
Solutions Group) tools
are used in higher education very often [1, 2, 3, 4, 5,
7, 8, 9, 10, 11] in different courses. The subjects can Initially, the current lecturer of
vary from Digital logic design [1, 2, 10, 11], Reconfigurable Logic and VLSI design courses has
Embedded Systems design [7, 8, 9] to graduated from ALaRI (Advanced learning and
Reconfigurable Logic and VLSI (Very-large-scale research institute) [13] in Lugano, Switzerland.
integration) design [4, 5]. In the lectures are During education process, there were two subjects
presented some basic issues regarding related to reconfigurable integrated circuits
reconfigurable integrated circuits, their internal produced by Altera and VHDL. After finishing
organization and the way of their reconfiguring. master degree course and returning to Bulgaria, I
In laboratory exercises are used CPLD and mainly have been involved in teaching of similar material
FPGA boards and students implement some digital to students from Technical University - Sofia,
and microprocessor circuits. This practically Plovdiv branch. Together with it, the education in
oriented higher education helps students to make reconfigurable circuits has been started in the
required steps from using fixed logic to applying former John Atanasoff Technical College. In order
reconfigurable one and corresponding hardware to get licensed version of the design software and to
description languages as Verilog and VHDL (Very obtain boards for laboratory exercises, it has been
high speed integrated circuits description language). applied for a donation through Altera University
Technical University - Sofia, Plovdiv Program [14].
branch has about fifteen years tradition of using 2.1. First CPLD board
contemporary reconfigurable integrated circuits in
the educational process [2, 3, 4, 5]. During these
years some improvements have been done in course
syllabuses as well as in software and hardware
tools. The last change was received donation from
the Xilinx University Program (XUP) [12] in the
form of five FPGA design kits of type Nexys4 DDR
and twenty licenses for Vivado Design Suite:
System Edition. Together with previously used
Altera tools now design software and FPGA boards
from both major reconfigurable logic suppliers are
used in bachelor's and master's degree syllabuses.
In the next sections are presented some details about Fig. 1. UP2 board
II-160
The application was successful. The first sufficient amount required for the laboratory
donation from Altera Corporation has been in form exercises.
of licensed software (Quartus II and ModelSim) and Unfortunately, we had no laboratory
one CPLD board (UP2). The received UP2 equipment and design software by another major
(University Program 2) is presented in Fig. 1. FPGA producer - Xilinx. We have decided to apply
It has been realized a license server in 2202 for Xilinx University Program [12]. The results of
computer hall and the students have been used the this application was again successful. Details about
licensed versions of Quartus II and ModelSim new donation are presented on the next lines.
design software. The obtained single UP2 board
was a good starting point in the lab equipment 3. Xilinx tools
process. It has been used as a demonstration of We have applied for the five FPGA boards
loading a simple design into one of two CPLDs of type Nexys4 DDR. The application was again
located on the board. UP2 connection to personal successful, as it mentioned before, and we have
computer (PC) was parallel interface using printer obtained these boards. The view of one of them is
port (LPT - Line print terminal 1) of the PC. The presented in Fig. 3.
length of this cable was not enough UP2 to be put
on the laboratory desk. It has been decided to be
found another way in order to be increased the
number of the FPGA boards that are needed for the
laboratory exercises.
2.2. Support from the University
With the financial support of the Technical
University - Sofia have been obtained two Cyclone
II FPGA starter development boards. One of them is
presented in Fig. 2.
Fig. 3. Nexys4 DDR board
Additionally, we have received twenty

licenses for the Vivado Design Suite: System
Edition. Now we have FPGA boards and
corresponding design software from both global
players of reconfigurable integrated circuit market.
These tools we use in the educational process in
Technical University, Plovdiv branch as well in the
research activities, too.
Fig. 2. Cyclone II FPGA starter development
board 4. Altera and Xilinx tools used in the
teaching and research activities
4.1. Educational process using both tools
These boards are used in the laboratory At the beginning, we have started using
exercises of the courses connected to the Altera (now the Intel Programmable Solutions
reconfigurable integrated circuits. Group) tools. The corresponding syllabuses were
2.3. Donation from the firm partner presented in the paper [5]. As example, in the
lectures are introduced IPs (Intellectual Properties)
With help of the firm partner (ASIC Depot)
modules. In the Fig. 4 is presented a block diagram
of Computer Systems and Technologies Department
of the PLL (Phase - locked loop) circuit. Then
has been added additionally two Cyclone II FPGA
corresponding IP module called ALTPLL
starter development boards. In this way the number
Megafunction is presented [15].
of existing FPGA boards are very close to the
II-161
In the Fig. 7 a screenshot of a simple VHDL
design (multiplexer) realized using Vivado Design
Suite: System Edition is presented . This a part of
the lectures in Reconfigurable Logic and VLSI
(Very-large-scale integration) design courses.
Fig. 4. PLL block diagram
At the end of Reconfigurable logic course a

small design task is given to the students. They have
to describe in VHDL a simple test system.
This system uses a 4-bit binary counter and a LFSR
(Linear feedback shift register) in order to be
generated input stimuli to device under test. One
solution of this task is presented in Fig. 5 [5].
Fig. 7. Vivado Design Suite: System Edition

project
4.2. Research achievements using both tools

Fig. 5. Synthesized test circuit It has been published many research papers
in which FPGA boards and tools (mainly from
Altera Corporation) have been used [2, 3, 4, 5].
Now, in the lectures' material an Some research results have been included in the
information about Nexys4 DDR and Vivado Design book section [6].
Suite: System Edition has been added. In the Fig. 6 As example, a predicate logic processor
an electronic circuit of Nexys4 DDR basic I/O (PLP) has been designed and tested during ERCIM
(connected Input / Output devices) is presented. (European research consortium for informatics and
mathematics) postdoctoral fellowship in NTNU
(Norwegian University for Science and
Technology) under guidance of Prof. DSc.
Guennadi Kouzaev.
A simplified block diagram of the PLP is
presented in Fig. 8.
Fig. 8. PLP block diagram
Fig. 6. Nexys4 DDR basic I/O electronic circuit

II-162
Additionally, a patent application has been 4. Kostadinov, A. (2009). Using HDL translators
done [16]. Now, in the future research activities will in VLSI design laboratory exercises. Annual
be added some results using Nexys4 DDR FPGA Journal of Electronics, vol. 3, no. 2, pp. 56-59.
board and Vivado Design Suite: System Edition. 5. Kostadinov, A. (2013). Field-Programmable-
We have already started with a bachelor thesis Logic device education, Annual Journal of
connected to design of a master device controlling Electronics, vol. 7, pp. 46-49.
I2C (Inter-Integrated Circuit) bus. After successful 6. Kouzaev, G. (written together with
defense and with an additional work, the material is Kostadinov A.). (2013) Lecture notes in
able to be presented on a scientific conference. electrical engineering 169, Applications of
advanced electromagnetic, Components and
5. Conclusion systems, Predicate logic, gates and processor
In the paper, the organization of the for topologically modulated signals (TMS),
educational process and research activities pp. 451 – 460, Springer, Germany.
connected to the reconfigurable integrated circuits 7. Kumar, A., Fernando, S. and Panicker, R.
have been presented. (2013). Project-Based Learning in Embedded
It has been given information regarding Systems Education Using an FPGA Platform.
using tools from both major suppliers of CPLDs and IEEE Transaction on Education, vol. 56, no. 4,
FPGAs. pp. 407 - 415.
It has been mentioned that the XUP 8. Mahmoodi, H. at al. (2012). Hands-on
donation is already used in the teaching activities. Teaching of Embedded Systems Design Using
The current lecturer plans that this donation will be FPGA-Based tPad Development Kit.
used also in the future research activities. Interdisciplinary Engineering Design
Education Conference, Santa Clara, USA,
6. Acknowledgements pp. 1 - 6.
I gratefully acknowledge the funding 9. Nakano, K. and Ito, Y. (2008). Processor,
received towards my two postdoctoral fellowships Assembler, and Compiler Design Education
from the European Research Consortium for Using an FPGA. 14th IEEE International
Informatics and Mathematics (ERCIM) as well as Conference on Parallel and Distributed
from FP7 (Framework Program 7) of the EU Systems, Melbourne, Victoria, Australia,
(European Union). pp. 723 - 728.
I would like to thanks to Prof. DSc. 10. Petrescu, I., Păvăloiu, I. and Drăgoi, G. (2014).
Guennadi Kouzaev for his excellent supervisory Digital Logic Introduction Using FPGAs. The
role. 6th International Conference Edu World,
I would like to thanks to Prof. PhD Grisha Pitesti , Romania, pp. 1507 - 1513.
Spasov for his support during the last application 11. Zeyad, A., Sulaiman, N. and Hamidon M.
and received donation from Xilinx University (2009). FPGA-based Implementation of Digital
Program (XUP). Logic Design using Altera DE2 Board.
International Journal of Computer Science and
Network Security, vol. 9, no. 8, pp. 186 - 194.
REFERENCES 12. https://www.xilinx.com/support/
1. Kiray, V. and Jambulov, A. (2009). FPGA university.html
based digital electronics education and a 13. http://www.alari.ch/
simulator core design for a/d communication. 14. https://www.altera.com/support/training/
Academic Journals, vol. 4, no. 11, university/overview.html
pp. 1292 - 1301. 15. https://www.altera.com/en_US/pdfs/literature/
2. Kostadinov, A. (2009). Using of a hardware ug/ug_altpll.pdf
description language in the digital design 16. https://www.ipo.gov.uk/
subject. Journal of the Technical University – p-ipsum/Case/PublicationNumber/GB2508162
Sofia, Plovdiv branch “Fundamental Sciences
and Applications”, vol. 14, no. 1, pp. 421-426.
3. Kostadinov, A. (2009). Using of embedded
logic analyzer in laboratory exercises of testing Authors’ contacts
and diagnostics subject. Journal of the Organization: Technical University - Sofia,
Technical University – Sofia, Plovdiv branch, Plovdiv branch
“Fundamental Sciences and Applications”, Address: 25 Tsanko Diustabanov Str.
vol. 14, no 1, pp. 427-432. Phone: +359 32 659 726
E-mail: kostadat@tu-plovdiv.bg
IMPLEMENTATION OF SOFTWARE
ARCHITECTURE FOR COMPLEX ANALYSIS OF
SOLAR CORONA IMAGES
DIMITAR GARNEVSKI
Abstract: A study of the processes in the nearest star - the Sun is important because of its
influence on space weather and life of the planet as a whole. As in any scientific fields
research in the solar corona require specialized software for processing the received data.
This publication will investigate aspects of the architecture and the way of implementation
of software for processing and analyzing images of the solar corona and in particular data
retrieval for CME (Coronal Mass Ejection).
Key words: image processing, filters, parallel computing, solar corona, OpenCL, file
formats, meta-data
1. Introduction Fig.2 presents different stages of processing of

In process of analysis of Solar corona and image and flow of data between steps.
especially research of events and processes in
3.1. Pre-Processing
magnetic field scientists uses special software
which meets their requirements. Common tasks can One of the additional steps which can be
be divided in few groups like initial discover of the implemented and can improve results of image
event (CME - Coronal Mass Ejection), estimation of processing is quality estimation of input images and
parameters of the event, additional classification of their initial pre-processing. Although currently
the CME, tracing of changes in the CME in time. significant amount of solar corona images which
Software must also present information to the captured from Earth surface (capturing Solar corona
scientists in proper format for additional processing. with coronagraph with CCD or CMOS technology
camera) has parameters which in significant amount
2. Requirements cases meet requirements of used algorithms for
We can be define following requirements image analysis and processing. When we use
which must be fulfilled by the implemented images received from cosmic probes cannot always
software guarantee good quality of input images. If we leave
• defining model of processed data (solar aside the parameters such as number of bits per
corona images and meta-data) pixel and levels of compression of images, most
• defining states of images quality estimation modern tools are able to provide color or black and
algorithms white images with dimensions at least 1024 x 1024
• implementation of adaptive model of input pixels, it provides a starting point for further
images pre-processing depending on images processing.
parameters
• implementation of pre-processing of the
images like noise filtering etc.
• implementation of common processing
algorithms stages
• visualization of the result and output data
formats
3. Stages of image processing
Relying on initial requirements we can
define following stages of solar corona image Fig. 1. Prominence sample image
processing in developed software. The diagram at
II-164
Initial image quality estimation include some benefits against other filters like blur because
estimation of noise level and level of shining from anisotropic filtering blurs image without damaging
camera lenses. Noise level estimation is performed edges between objects in image [1][2][6][8].
with stochastic approach.
Undesirable shining at the edges of the
image can be removed by estimation of these pixel 3.2. Meta-data
values and following normalization against average For the purposes of the processing and in
values available in entire image. particular tracking changes over time are necessary
metadata to images (date and time of capture,
necessary adjustments, etc.). The date and time can
both be obtained from the file or from meta-data (in
cases when the data file does not correspond to
actual shooting time). In implemented application
when as input the data files FITS time are obtained
from the file. In cases when using JPEG and
processed series of images to store metadata using
text file containing JSON (JavaScript Object
Notation) with the necessary meta-data (in the basic
version it is a list of files for processing and
shooting time).
Format used contains data preprocessing:
• image contrast requirements - min and max
values
• noise level
• possibility for preprocessing: pre-filter:
shining removing, blur, median, Mexican
hat, anisotropic (2 types with parameters)
• motion map stage
• output dir
• output file format
Fragment of a JSON file metadata is
presented below:
Fig. 2. Data processing flow
Noise removing is performed via filtering

input images with one or combination of multiple
filters like Mexican hat, median, anisotropic filter,
etc. Common task of this operation is removing of
pixel values which are out-of desired range without
damaging structure of captured objects (lines,
edges). In this situation anisotropic filtering has
II-165
Fig. 4. Generated motion map with

prominence height estimation markers
With these data points are build arc that

defines the boundary of occulting disk and allows to
calculate the correlation in size.
Estimation of connections between changes
in Sun brightness and solar prominence is
performed via processing of number of images in
selected interval and discovering prominence in this
interval.
For the implementation of the software used
language C++, OpenCL environment for parallel
processing, OpenCV library for image processing
Fig. 3. JSON file with metadata and number of additional libraries for FITS and
JSON processing. For the implementation of key
3.3. Image analysis filtering functionality used OpenCL so calculations
Image processing core of the application can be performed on CPU and GPGPU. Using of
include algorithms like generation of motion map, OpenCL provides needed flexibility in selection of
discovering of prominence at the boundary of different computational platforms. Also each filter
occulting disk, estimation of prominence parameters can be defined as one separate kernel function in
like height, direction and their evaluation in time OpenCL which provides standardized approach in
and estimation of connection between changes in filter execution and opportunity for filter
Sun brightness and probability for solar replacement.
prominence.
Generation of motion map is performed via
algorithm described in [3][4][5]. In application is
used its OpenCL implementation with platform
independent optimizations.
Once built the map with the movement of
the particles may be estimated height of the
prominence and its direction.
Discovery the borders of occulting disk
(artificial moon). In assessing the images and
special about calculating the data for the height of
the protuberance having matching the size of the
solar disk image and the actual size of the Sun. In
this case the work is performed with parts of images
that include approximately 1/2 to 2/3 of the entire
solar disk.
In this case, be carried out two operations,
alignment of the images with respect to each other
and detecting the boundary of the artificial moon. Fig. 5. Kernel processing model
Artificial moon itself is depicted as a dark area
defining the border is done by detecting the two Additionally be can perform changes in
endpoints that are located in border areas of the filter algorithm (kernel function implemented in
image and the central point, which is positioned OpenCL language) during development process
between them. without performing changes in entire application
and we don't recompile it.
II-166
3.4. Post-processing Implementation of additional pre-
Post-processing of results include processing and common filtering algorithms.
generation of out data in human readable format or Integration of distributed computations like
format which can be used in another application. MPI (Message Passing Interface)
Developed application creates output in common
images formats like JPEG or FITS (Flexible Image
Transport System) [9] and also generates video REFERENCES
based on frame sequences. For numerical data we
use CSV (Comma-Separated Values) [11] format 1. Al Bovik. (2004). Handbook of image and
which is human readable and can be imported in video processing, Academic Press. Canada.
most of data processing software like Microsoft 2. Rafael C. Gonzalez Richard E. Woods (2002)
Excel, it's open-source alternatives or other Digital Image Processing, 2nd Edition,
applications. Basic example of generated CSV is Prentice Hall
shown at image. 3. Petya Pavlova, Elena Duncheva, Kostadinka
Koleva (2010). Method for tracking and
Table 1. Sample CSV data format generated by the mapping a motion based on images of the solar
application corona. Proceedings of the VII Bulgarian-
Serbian Astronomical Conference (VII BSAC),
1 2017-02-02 20:23:25 2017-02-02 21:24:25 128 46720
Chepelare, Bulgaria.
2 2017-02-02 20:24:25 2017-02-02 21:27:25 134 48910 4. Petya Pavlova, Kristina Staneva (2005).
Dependence of the Colour in Computer Models
CSV data format generated by the on the Colour-reproducing Signals
application is shown in Table 1. Individual columns Discretization. E+E.
are separated by the comma or tab-space depending 5. Petya Pavlova, Dimitar Garnevski, Kostadinka
on preferences. Shown in Table 1 file contains Koleva (2016). Optimization of a motion
information amount measured high of discovered tracking and mapping method based on images
coronal mass ejection (CME). First column of the solar corona. Bulgarian Astronomical
represents sequence number, 2nd and 3rd start and Journal, Volume 24
end time of measured interval (filled against input 6. S. G. Hoggar, Mathematics of digital images.
meta-data or timestamp of the input images), 4rt Creation, Compression, Restoration,
columns represent height of the CME in pixels and Recognition, Cambridge
the last columns shows height in km, which was 7. http://opencv.org/, OpenCV (Open Computer
calculated depending of scale factor (filled in meta- Vision) Reference
data) and image resolution. 8. Frery A. C, Perciano T. (2013). Introduction to
3.5. Result visualization Image Processing Using R, Chapter 2, Springer
9. https://fits.gsfc.nasa.gov/, The FITS Support
As we mentioned in 3.4 application Office, NASA/GSFC
generates images depending on desired result. So
10. http://www.json.org/, ECMA-404 The JSON
they can be viewed by external image viewers. Soft Data Interchange Standard.
has embedded viewer which uses OpenCV [7] 11. https://tools.ietf.org/html/rfc4180, Common
image processing library for additional image
Format and MIME Type for Comma-Separated
processing before visualization. This type of Values (CSV) Files
additional processing includes conversion between 12. Markus J. Aschwanden, A Code for
color spaces, basic filtering or adding of Automated Tracing of Coronal Loops
geometrical objects (lines, arc, markers, etc.) to the Approaching Visual Perception, Springer
image. Science+Business Media B.V. 2010
For in example image at the Fig.4 are
visible object (arcs and markers) which has been
added with usage of OpenCV image processing Dimitar Garnevski
library. Department of Electrical Engineering
Generation of video files (.avi format) is Technical University–Sofia, Branch Plovdiv
also performed via OpenCV library. 25 Tsanko Diustabanov St.
4000 Plovdiv
4. Conclustions
E-mail: garnevsky_dm@abv.bg
Adaptation of the application to work with
web-based storage systems and organization of
images. This will include implementation of API
which can be used by external systems.
ADVANCED ALGORITHM FOR CANNY EDGE

DETECTION GRADIENT MAGNITUDE
COMPUTATION TO BE IMPLEMENTED ON
FPGA
DIMITRE KROMICHEV
Abstract: The paper presents a Pythagorean theorem computing algorithm aimed at

simultaneously satisfying both the requirement for mathematical exactness and the
demand for maximum speed of gradient magnitude computations in the FPGA
orientated Canny edge detection realization. Proposed are several criteria to be
satisfied for a computational algorithm to be labelled as advanced. The focus is on
advantageously utilizing the FPGA integer arithmetic capabilities for providing a
hundred per cent accuracy of square root calculations at optimal speed.
Key words: Canny edge detection, gradient magnitude, FPGA, accuracy, speed, algorithm,
square root, interval
1. Introduction orientated Canny which is described in the literature

[7][8][9][10] relies on the approximation
Canny edge detection is a low level digital
image processing technology. It was intended to be |GM| = |GX| + |GY| (1)
used in software. Field Programmable Gate Array
(FPGA), as a representative of the widely where Gx and Gy are the x- and y-gradients.
applicable programmable logic, has been increasing Although fast, this technique is very inaccurate. So
its market share over the last two decades. The two far, in the literature there has been no algorithm
capital requirements for FPGA-based hardware addressing both the requirement for mathematical
implementation of Canny edge detection are exactness and the demand for speed.
accuracy and speed, the former being a function of The objective of this paper is to present an
the mathematical exactness of computational advanced gradient magnitude algorithm aimed at
algorithms within each of the modules framed by speeding up the FPGA-based Canny on the basis
the maximun utilization of FPGA capabilities. of providing a hundred per cent mathematical
Speed depends upon several factors, among them of accuracy of results. The task is to describe in detail
tremendous importance being the algorithms the sequence of steps, to thoroughly analyze the
calculations are realized through. FPGA computational reliabilty, to expose the
functionalities and characteristics should be taken characteristics, and to point out the applicability of
into account by the computational algorithms in the proposed algorithm. taking into account the
terms of utilizing those favourable to fast FPGA functionalities. Essential criteria to be
executions whereas avoiding operations and satisfied for an algorithm to be labelled as advanced
methods serving as bottlenecks. Speed is worthy of are set forth. The targeted hardware is Altera
being set as an accomplishable goal only on the FPGAs. Relevant to the analyses and conclusions
peremptory platform defined as total accuracy of arrived at in this paper are only gray-scale images.
calculations and veracity of detected contours.
Thus, in FPGA-based Canny every single algorithm
requires a multifaceted approach targeting the 2. Criteria for advanced algorithm
results’ being reliable as well as feasible.
The implementation of Pythagoras aimed at Canny edge detection includes these
computing the gradient magnitude in the FPGA consecutive modules: Gaussian smoothing,
computing the orthogonal gradients, computing
II-168
gradient magnitude and direction, non-maximum should optimally commence computations with
suppression, hysteresis thresholding. Labelling an integer arithmetic which takes exactly one clock
algorithm should reflect its essential computational cycle to execute, whatever the width of input
characteristics, and rely upon definite criteria, the values. Fulfillment of that requirement is well
latter in the FPGA-based Canny being as follows. manifested for the Pythagoras, the FPGA
1) Satisfying the peremptory demand for a hundred implementation of the algorithm peremptorily
per cent mathematical accuracy of results. This is starting with multiplication.
the solid guarantee for veracity and reliability of
detected contours. In this respect, gradient magnitde
is of tremendous importance for the overall 3. The proposed algorithm
plausibility of Canny computations in terms of
procuring the values to be compared in the non- Mathematically, gradient magnitude GM is
maximum suppression module, thus directly and computed using
immediately impacting one of the few most
significant and applicabale qualities of contour – its
precise localization. , (2)
2) Serving as a tangible tool for optimally where
enhancing the speed parameter of Canny Gx and Gy are the x- and y-gradients.
computations on FPGA. This criterion has a
quantitatvely measured characteristic presented by Altera provides a specialized integer square
the smallest possible amount of clock cycles root function – ALTSQRT [5]. Although the 8-bit
requred for executing the surveyed algorithm. Two gray scale image values are definitely advantageous
particular aspects are to be taken into account here. to the speed parameter, the maximum clock rate the
Speed is a function of the appropriate and expedient function can be executed at [5] is not campatible
organization of calculations in the algorithm being with the accomplishment of the speed optimization
analyzed. On the other hand, fast and goal – calculating the correct result requires nine
computationally sofisticated as it can seem, any clock cycles. Thus, an optimized approach is
algorithm should comply with the integer arithmetic needed.
peculiarities and benefit from the advantageous
functionalities of the FPGA hardware 3.1. Input data
implementation. Thus, the more mathematically
complicated the algorithm (such as Pythagoras), the Gx and Gy are computed in the Canny
heavier the speed dependence on the programmable orthogonal gradients module by employing two
logic’s computation accelerating capabilities. approaches:
3) Boosting the efficiency of pipelining, the latter 1) Exact mathematics. The equations used to
being indispensable to the FPGA–based Canny accurately calculate the x gradient and the y
edge detection implementation focused on gradient are:
speed. This criterion proves to be of particular
importance in case of algorithms being
executed in parallel within the main flow of
(3)
computations on FPGA, gradient magnitude and
gradient direction algorithms being
indispensable here. Taking into account that
calculating the local maximum requires a square , (4)
neighbourhood of pixels, the calculated gradient
magnitude values have to be ready for use at a rate where
that allows for a comparison of the central pixel N1..N8 are the neighboring pixels
with the adjacent pixels along the axis of gradient in a 3x3 neighborhood.
direction without any delays.
4) Minimizing the impact of input data width. The values calculated here are within the
Confining the input data to 8 bits is of significant intetval [-255,255].
importance in the Canny modules as a tool for 2) Approximation.
diminishing delays. On Altera FPGAs, the fastest
integer arithmetic execution in terms of clock cycles Gх = [(N3 + 2N4 + N5) – (N1 + 2N8 + N7)] (5)
is guaranteed for 8-bit values [1][2][3][4][5]. In Gу = [(N7 + 2N6 + N5) – (N1+ 2N2 + N3)] , (6)
respect to this criterion, an advanced algorithm
II-169
where capabilities of hardware implementation – parallel

N1..N8 are the neighboring pixels computations. Thus the proposed algorithm satisfies
in a 3x3 neighborhood. the criteria for being labelled as advanced. It
guarantees a hundred per cent mathematical
With division by the power of 2 being accuracy of gradient magnitude computation. It also
dropped out, the values calculated here are within utilizes the fastest integer arithmetic on Altera
the intetval [-1020,1020]. Consequently, proper FPGAs. The presented algorithm is not susceptible
scaling is required for all the results to fit within the to any input data width driven delays. Being that
maximum gray scale image pixel range. fast, it is capable of increasing the efficiency of
pipelining in the FPGA-based Canny.
3. 2. Advanced gradient magnitude
computation 4. Conclusions
On FPGA, computing Pythagoras starts with The paper focuses on presenting and
multiplying |GX| by |GX|, and |GY| by |GY| . The analyzing a new gradient magnitude algorithm
largesr positive pixel value being 255, the results of addressing the goal of speed enhanced and
both multiplications are within the interval [0, mathematically exact Canny edge detection
65025]. computations intended to be implememted on
In the next step, the multiplication results are FPGA. Essential criteria to be satisfied for an
added, and consequently the numbers to be square algorithm to be labelled as advanced are defined. A
rooted are within [0, 130050]. thorough description of the algorithm’s peculiarities
Thus, all the possible values at the input of the is set forth. The algorithm is scrutinized in terms of
square root function are 130051, And all the its applicability, computational reliability and speed
possible values at its output are , taking into characteristics. The technological capabilities of
account that only numbers in the interval [0, 255] efficiently avoiding speed eroding computational
are relevant for pixels in the gray scale matrix. approaches and satisfying the peremptory demand
Consequently, the exact mathematical results of the for mathematical accuracy and plausibility of results
square root function can be determined by 256 pinpoints the feasibility of the proposed algorithm
closed intervals, each of them representing a single as a tangible tool for enhancing Canny’s
integer from the interval [0,255]. These intervals performance on FPGA.
contain all the 130051 values calculated by adding
the squared gradients.
Defining the smallest and the largest in each REFERENCES
interval is based on the fact that the difference
between squares of two consecutive integers is 1. Altera Corporation. Cyclone IV Device
equal to the smaller integer multiplied by 2, and Handbook. 2014, Volume 1
then 1 is added to the result. Thus, the smallest of 2. Altera Corporation. Cyclone V Device
all the consecutive values contained in an interval is Handbook. 2014, Volume 1
determined through: 3. Altera Corporation. Stratix III Device
1) squaring the gray scale image pixel value Handbook. 2014, Volume 1
represented by this particular interval; 4. Altera Corporation. Stratix IV Device
2) the gray scale image pixel value represented by Handbook. 2014, Volume 1
this particular interval is decremented, and the result 5. Altera Corporation. Integer Arithmetic IP
is subtracted from the result computed in 1) . Cores User Guide, 2014
The largest of all the consecutive values 1. Altera Corporation. Cyclone II Device
contained in an interval is determined through: Handbook, Volume 1, 2014
1) squaring the gray scale image pixel value 7. Chandrashekar N.S., K. R. Nataraj. Design
represented by this particular interval; and Implementation of a Modified Canny
2) the gray scale image pixel value represented by Edge Detector based on FPGA,
this particular interval is subtracted from the result International Journal of Advanced
computed in 1) . The only exception is the value in Electrical and Electronics Engineering,
the interval representing 255 – here the largest (IJAEEE). 2013, Vol.2, (1), pp. 17-21
possible number is 130050. 8. Chandrashekar N.S., K. R. Nataraj. NMS and
The comparisons of all values for computing Thresholding Architecture used for FPGA
the gradient magnitude are executed simultaneously based Canny Edge Detector for Area
on FPGA, utilizing one of the most significant Optimization, Proceeding of International
II-170
Conference on Control, Communication and Science and Research (IJSR). 2014, Vol. 3,
Power Engineering. 2013, pp. 80-84 pp. 1198 -1205
9. Divya. D., P. S.. Sushma. FPGA
Implementation of a Distributed Canny
Edge Detector, International Journal of
Advanced Computational Engineering and Plovdiv University Paisii Hilendarski
Networking. 2013, Vol. 1, (5), pp. 46-51 Plovdiv 4000
10. Shamlee V., Jeyamani.. A Split Canny Edge 24 Tzar Asen Street
Detection: Algorithm and its FPGA e-mail: dkromichev@yahoo.com
Implementation. International Journal of
APPROXIMATION IN THE FPGA BASED CANNY

EDGE DETECTION COMPUTATIONS
DIMITRE KROMICHEV
Abstract: The paper deals with the technology of approximation and its impact on the
computational results in the FPGA based Canny edge detection modules. Presented are
four general types of approximation and their essential functions and characteristics are
exposed. The impact of approximation on the mathematical exactness of calculations in the
Canny algorithm is scrutinized with respect to the peculiarities of hardware
implementation and the targeted veracity of the mapped contours. Analyzed is the
relevance of approximation in terms of speed vs. accuracy.
Key words: Canny, FPGA, approximation, type, algorithm, accuracy, speed
1. Introduction Canny attemps to spare the on-chip memory at

the expense of approximation orientated
For decades now, Canny edge detection has calculations and maintaining constant throughput
gained a well deserved reputation of being an through performing the computations at block level.
efficient and reliable low level digital image In [5], area optimized Canny is mainly focued on a
processing technology. Inasmuch as this precise yet new architecture of the non-maximum suppression
complicated operator is orientated towards being and approximation framed hysteresis thresholding
used in software, its hardware implementation on modules. In modified Canny [4] [7], the image is
the Field Programmable Gate Array (FPGA) is partitioned into q sub-images and each sub-image is
particularly difficult and requires that the further divided into p m×m blocks. Approximation
computations be executed by strictly taking into is applied to gradient magnitude and direction
account the capabilities and functionalities of that module, as well as the calculation of high and low
widely used representative of programmable logic. thresholds.
The goal to enhance performance while sparing the The objective of the paper is to thoroughly
area of FPGA generaly involves certain precision analyze the approximation patterns employed in the
decreasing approaches in the realization of the basic FPGA based Canny calculations. The task is to
computational algorithms Canny relies upon. Thus, scrutinize the computational results which are close
addressing the quality and plausibility of detected but do not coincide with the exact numbers obtained
edges, the technique of purposely replacing exact through using the standard mathematics in terms of:
mathematical values with numbers that are close types, distribution by Canny’s modules, attitude
demands an in depth analysis. In the FPGA based towards speed optimization, effect on the utilized
Canny, approximation has several aspects of FPGA area, influence on the precision of detected
utilization, each of them impacting the calculations contours, impact on speed and accuracy relation
in its own way and worthy of being duly assessed. with respect to FPGA’s advanced capabilities. It is
Described in the literature are different uses of the Altera FPGAs that are referred to throughout
approximation related to Canny edge detection on this study. Relevant to the conducted analyses and
FPGA. In [10], the approximation is focused on conclusions arrived at are only gray scale images.
computing the gradient magnitude in terms of
replacing the square root function with the sum of
2. Types of approximation
positive values of x- and y-gradients, and is aimed
at enhancing the speed performance. In [8], split
Canny uses approximation gradient magnitude and Canny edge detection includes five modules:
direction computations for the purpose of increasing Gaussian smoothing, computing the orthogonal
pipelining efficiency. In [6][3][9], distributed gradients, computing gradient magnitude and
direction, non-maximum suppression, hysteresis
II-172
thresholding. With respect to its utilization and 2) approximation to (3) for the purpose of defining
effect on the calculations in FPGA orientated two 1-D masks: -1 0 1 and -2 0 2. They are
realization of Canny, approximation is to be the platform of the two 2-D filters employed in the
presented within four types. calculation of the vertical and horizontal gradients
(Fig. 1.)
2.1. Mandatory approximation
-1 0 1 -1 -2 -1
It is imposed by the requirements of the -2 0 2 0 0 0
discrete domain in digital image processing. In that
respect, actually two out of the five computational -1 0 1 1 2 1
modules in the Canny algorithm are entirely based
on approximation. Gх Gу
Any weighted average filter employed in the Fig. 1. Sobel filters for x and y
Gaussian smoothing module is a discrete
approximation of the 2-dimensional function
2. 2. Targeted approximation
, (1)
where Precise and reliable, Canny is
G is the Gaussian mask weight with computationally complicated, extensively
coordinates x and y, demanding in terms of hardware implementation,
ơ is the standard deviation of Gaussian and relatively slow algorithm. Thus, approximation
distribution; it determines the filter’s is deliberately employed to address different aspects
size, of performance being optimized.
is normalization constant.
2.2.1. Speed enhancing approximation
On FPGA, the filter performs a ZxZ
neighborhood operation (Z is an odd number and Z
Accomplishing the goal of a fast FPGA based
3)
Canny is favoured by a set of approaches related to
approximation.
1) Reducing and exploiting only the fastest integer
arithmetic needed to fulfil a particular
(2)
computational task. With respect to both the highest
frequency and clock cycles taken for execution, the
where
most speed advanced on FPGA are multiplication,
u(m,n) is the input image,
addition and subtraction [1] [2]. In terms of speed,
v(m,n) is the filtered image,
division is definitely a weak point. Using a divider
by employing the Altera LPM_DIVIDE function [2]
, w = {1,2,..},
will immensely erode the speed parameter of
g(k,l) is the Gaussian filter ,
Canny’s algorithm execution. Consequently, the
s is the sum of all
conventional integer division is typically omitted.
coefficients in the
Thus, in computing the orthogonal gradients the
mask; the coefficients
normalization division is dropped out, the gradient
are integers based on
magnitude calculation is modified to rely on a
the binominal series.
single addition, and the gradient direction value is
The Sobel approach for computing the
the result of utilizing only the signs of the vertical
orthogonal gradients is actually the result of a two
and horizontal gradients.
stage approximation:
2) Avoiding those Altera LPM functions [1] [2]
1) the central difference approximation
which, despite being handy, take more than two
∆f(х) = [ f(x+1) - f (x-1) ]*0.5 (3)
clock cycles to output a correct result and lead to
to the derivative of a continuous function
speed eroding delays. The focus here is on omitting
the square root function for the gradient magnitude
(4) and the division in the gradient direction
for ∆х = 1 as the smallest possible value of ∆х in computations.
the discrete domain of digital image pixels;
II-173
3) Keeping the sequential logic input data width
optimally within 8 bits for executing the FPGA |GM| = |GX| + |GY| . (6)
integer arithmetic at the highest frequencies [1] [2].
The impacted values are those of the vertical and The accurate calculation of gradient direction
horizontal gradients which need to be confined to GD is
the interval [-255, 255].
4) Replacing a slow computational procedure with
values determined in advance. To this point, fixed
(7)
values for the high and low thresholds tend to be
where
used instead of assessing the entire image statistics
Gx and Gy are the x- and y-
in the non-maximum suppression module.
gradients; ;
.
2.2.2. FPGA hardware sparing
approximation Calculating the gradient direction by applying
expression (7) has the following inauspicious
aspects:
Typically, the less the utilized FPGA area, the 1) The FPGA implementation is computationally
more efficient the Canny implementation. This expensive, and therefore leads to substantial delays.
aspect of approximation in the FPGA based Canny, 2) The divisor GX can be equal to 0. In that case
despite sometimes being colateral and far from calculating the gradient direction is practically
particularly intended, is important in two ways. It is irrelevant, and, consequently, additional
an implicit indicator of the efficiency of a computational conditions should be introduced to
computational mechanism, thus presenting tangible ensure the correctness of results.
measurement for the practical applicability of any For hardware implementation, (7) is typically
technique or algorithm aimed at optimizing the replaced with the sign relations of the gradients.
organization of Canny calculations to be
implemented on FPGA. The hardware sparing Gy > 0 & Gx > 0
approximation can also serve as guidelines for Gy < 0 & Gx < 0
assessing the extent a proposed computational Gy > 0 & Gx < 0
approach is functionally advanced. Gy < 0 & Gx > 0 . (8)
This approximation type is present mainly in
the calculations of the gradient magnitude and
direction wherein, instead of the very expensive 2. 4. Partial approximation
hardware realization of exact mathematical
algorithms, the results are achieved through relying This approximation type deals with a limited
on addition and comparison. It is also appropriate portion of the algorithm’s computations and has its
for sparing the limited in quantity on-chip memory indispensible application in the calculations for
in terms of using predefined low and high threshold determining the angular values 0, 45, 90, 135 which
values to avoid computations involving the entire define the four axes of gradient direction. Two
image statistics after executing the non-maximum values are utilized to optimally approximate the two
suppression. angles of crucial importance to the mathematical
accuracy of gradient direction calculation.
2.3. Complete approximation Taking into account that four intervals are
needed to contain all results of computing the Gy
and Gx ratio, two referece points are employed. One
This approximation type is focused on of them should represent an angle of 22.5°, and the
replacing a whole computational algorithm and its other - 67.5°. In the integer domain of FPGA
territory is the gradient magnitude and direction based Canny, the most appropriate approximation
module. Mathematically, gradient magnitude GM is to angle 22.5° is the fraction
calculated using
( = 21.8014094°) (9)
(5) and to angle 67.5° - the fraction
where
Gx and Gy are the x- and y-gradients.
In the FPGA based Canny, Pythagoras is ( = 68.1985905) . (10)
generally replaced with
II-174
Gх = [(N3 + 2N4 + N5) – (N1 + 2N8 + N7)]
(13)
Gу = [(N7 + 2N6 + N5) – (N1+ 2N2 + N3)] .
3. Assessment of approximation (14)
This approach presents two flaws. On the one
3.1. Approximation in terms of impact on hand, omitting division leads to lack of additional
computational results averaging, thus impeding a certain extra noise
suppression to be applied to the image. On the other
Inasmuch as the weighted average filter hand, taking into account that computing the
approximation has steadily defined its presence in gradient magnitude refers to the Pythagorean
the Gaussian smoothing, the appropriate selection theorem, and magnitude values larger than 255 are
of the fiter’s size is of significant importance as a irrelevant to the calculations of local maxima in the
tool to compensate for the omission of not to be next Canny module, all the results equal or
underestimated subtleties imposed by the discrete exceeding are scaled down to the appropriate
domain of digital image processing. Thus, taking 255. Consequently, the division’s being omitted
into account the numerical characteristics of the bell requires that all the results larger than 255 should be
shaped curve, the most suitable variants prove to be cut down to fit the maximum value of a gray scale
sizes 5x5 and 7x7. With respect to the fact that the image pixel, and this entails serious
larger the size the worse the localization, and taking disproportionality among this module's output
into account the amount of computations required results.
for the hardware implementation, the optimal option The gradient magnitude and direction module
is filter of size 5x5. is of capital importance for the quality of detected
As gradient filters, the two Sobel matrices contours. In computing the gradient magnitude,
have positive and negative coefficients. If in a 3x3 replacing (5) with (6) leads to calculated numbers
neighborhood C(x,y) is the central pixel, and the which differ from the exact mathematical results in
numbers of neighboring pixels are as shown (Fig. the range of 0.2-30 per cent. In terms of quantity,
2.) , this lack of precision is irregularly distributed
across the calculated gradient magnitudes, and any
x computational mechanism aimed at offsetting the
inaccuracy is practically irrelevant. The largest
N1 N2 N3
differences are calculated for pixel values that are
С among the most frequently to be met in a gray scale
N8 (х.y) N4 image. In view of the fact that it is the gradient
magnitude results that are compared in the non-
N7 N6 N5 maximum suppression module to perform edge
y thinning, the overall effect is poor localization and
lack of plausibility, thus resulting in detected
Fig. 2. 3x3 neighborhood pixels contour’s getting unreliable for production line
applications wherein taking correct measurements is
the exact equations to calculate the x gradient and of crucial importance. Depending only on (8) to
the y gradient for C(x,y) are: compute gradient direction instead of (7) halves
down the accuracy by reducing the required
intervals from 8 to 4. The consequence is decreased
correctness in the results of the algorithm involved
(11) in determining the local maxima, with eroded
precision and gaps in the mapped contours.
The values calculated for the gradient
magnitude and direction are straightforwardly used
. (12) in the non-maximum suppression module. Utilizing
(13), (14), and (6) can result in computing values
On FPGA, the approximation in computing
the orthogonal gradients practically reduces the exceeding As long as only comparison of values
necessary integer arithmetic to addition and limited of up to 255 is relevanly employed, keeping
subtraction with division by the power of 2 being within 8 bits has no alternative.
dropped out: Convenient as it is, utilizing two static values
instead of a costly algorithm employing the entire
image statistics to compute the high and low
II-175
thresholds results in both omitting real edges and imposed by the requirements of the discrete domain
defining noise as contours, the latter being capable in digital image processing. In that respect,
of considerably compromising the output of Canny Gaussian smoothing and orthogonal gradients are
in terms of veracity, particuarly with respect to uses entirely based on approximation. Accomplishing the
orientated towards machine-to-machine interface. goal of a fast FPGA based Canny is favoured by a
Inasmuch as every pixel in the processed set of approaches related to approximation.
image is surrounded by 8 other pixels, the Complete approximation is focused on replacing a
approximation in (9) and (10) is actually whole computational algorithm and its territory is
characterized by accuracy that is enough to provide the gradient magnitude and direction module.
computational results commensurable to those Partial approximation deals with a limited portion
achieved through utilizing the conventional of the algorithm’s computations and has its
mathematical expression in (7). indispensible application in the calculations for
determining the gradient direction. Two values are
utilized to optimally approximate the two angles of
3.2. Approximation in terms of crucial importance to the mathematical accuracy of
speed vs. accuracy gradient direction Each approximation type is
analyzed and assessed in terms of: Canny’s module
Overtly or in the background, the analysis of it is used in, characteristics, expediency, departure
approximation and its impact on the FPGA based from the mathematical exactness of results, impact
Canny calculations eventually comes down to the on the reliability of calculations, relevance to the
speed vs. accuracy model. In this respect, there are quality of detected contours. The attitude of
two important facts. approximation towards speed and accuracy relation
Computationally expensive and slow as Canny with respect to FPGA’s advanced functionalities is
is, addressing the accomplishment of speed scrutinized.
optimization goal should peremptorily focus the
proposed approaches on the framework of detected REFERENCES
contours’ being optimally plausible. Achieving
reliability of the edge detected image especially for 1. Altera Corporation. Cyclone II Device
more demanding production line applications is Handbook, Volume 1, 2014
feasible on the basis of using mathematically 2. Altera Corporation. Integer Arithmetic IP Cores
accurate algorithms to realize the calculations in the User Guide, 2014
Canny modules. This being taken into account, 3. Aravindh G., C. S. Manikandababu, Algorithm
approximation is a tangible technique as long as it and Implementation of Distributed Canny Edge
does not violate the pattern of exactness and Detector on FPGA, ARPN Journal of
veracity at each successive stage in the flow of Engineering and Applied Sciences, Vol. 10 (7),
computations. Approximation for speed 2015, pp. 3208-3216
enhancement purposes which functions as a hurdle 4. Chandrashekar N .S. and Nataraj K. R., Design
to the accuracy of results is unacceptable. And Implementation of a Modified Canny Edge
Even though being hard to implement on Detector based on FPGA, International Journal
FPGA, Canny can afford avoiding approximation of Advanced Electrical and Electronics
that compromises accuracy through utilizing the Engineering, (IJAEEE), Vol.2, (1), pp. 17-21
programmable logic’s typical characteristic - 5. Chandrashekar N.S. and Nataraj K.R., NMS and
executing computations in parallel. Whatever the Thresholding Architecture used for FPGA based
complexity of an algorithm in a Canny’s module, Canny Edge Detector for Area Optimization,
the simultaneity of calculations is quite an efficient Proceeding of International Conference on
tool to accelerate calculations without having to rely Control, Communication and Power
on accuracy decresing approximation. The Engineering, 2013, pp. 80-84
additional requirement to be satisfied here is for the 6. Divya. D. and Sushma P. S., FPGA
parallel computations to utilize the fastest integer Implementation of a Distributed Canny
arithmetic on FPGA. Edge Detector, International Journal of
Advanced Computational Engineering and
Networking, Vol. 1, (5), 2013, pp. 46-51
4. Conclusions 7. Ramgundewar, Pallavi, S. P. Hingway,
Presented in this paper are four general types K. Mankar. Design of Modified Canny Edge
of approximation used in the FPGA based Canny Detector based on FPGA for Portable
edge detection computations: mandatory, targeted, Device, Journal of The International
complete, and partial. Mandatory approximation is Association of Advanced Technology and
II-176
Science. Vol. 16 (2), 2015, pp. 210-214 Research Based on Canny Operator,
8. Shamlee V. and Jeyamani, A Split Canny International Conference on Computer
Edge Detection: Algorithm and its FPGA Science and Service System, 10-13 Aug.
Implementation, International Journal of 2012, pp. 602-610
Science and Research (IJSR), Vol. 3 (12),
2015, pp. 1198-1205
9. Veeranagoudapatil, Chitra Prabhu.
Distributed Canny Edge Detector:
Algorithm & FPGA Implementation, Plovdiv University Paisii Hilendarski
Internationa Journal for Research in Plovdiv 4000
Applied Science & Engineering Technology 24 Tzar Asen Street
(IJRASET), Vol. 3 (5), 2015, pp. 586-588 e-mail: dkromichev@yahoo.com
10. Xianghua Hou, Honghai Liu, Welding
Image Edge Detection and Identification
INTRUSION DETECTION AND PREVENTING

SYSTEM
KOLYO RAYCHINOV, HRISTO VALCHANOV
Abstract: The attacks against computer and network security are a real threat in the
modern Internet world. Detection of attacks is a process of monitoring of computer systems
or networks and analyzing them for signs of possible events that are violations or threats.
Developing of intrusion detecting and preventing systems (IDPS) requires the availability
of resources that are needed to be integrated within the protected network infrastructure.
This paper presents the architecture and functionality of IDPS, which offers affordable,
flexible and efficient solution for building such as systems in different fields.
Key words: Network Attacks, Intrusion Detecting and Preventing System, Raspberry Pi
1. Introduction Internet, or intrusion by unauthorized people who

Besides the advantages offered by the are using incorrect privileges.
Internet technologies, they hide many risks that The system for detecting attacks (Intrusion
professionals in the IT field have to deal. Detection System - IDS) is software that automates
Cybercrime is a growing criminal activity, and the the process of detecting the attacks. The system to
convenience of fast connections to multiple prevent attacks (Intrusion Prevention System - IPS)
resources connected to Internet makes hackers to is software that has all the capabilities of IDS and
invent increasingly clever and high-tech ways to can take to deter possible incidents. IDS and IPS
exploit vulnerabilities in the systems. The attacks technologies offer many similar features and
are a real threat for the information, which is a administrators can usually exclude options for the
valuable resource in today's world, both for prevention of IPS products, forcing them to function
individuals and for businesses. The corporate sector as IDS. Today in cyber security is widely used the
often falls under similar attacks because of financial term “Intrusion Detection and Prevention Systems”
reasons - companies own resources and information (IDPS), which combines two technologies [2].
that can be stolen or to speculate with them. For this
reason the companies pay special attention on used
hardware and software - they are usually of
significantly higher class and price compared to
widely used ones. There is need for effective, smart
and cheap solutions for threat recognition and
preventing.
This paper presents the basic architecture
and functionality of intrusion detection and
preventing system, which offers simple, flexible and
efficient solution for building similar systems in
different fields.
2. Attacks detection and prevention
Detection of attacks is the process of
monitoring of events in computer systems or
networks and analyzing them for signs of possible
violations or threats of computer security policies
[1]. The events may be due to many reasons such as Fig. 1. IPDS infrastructure
malware (spyware and worms) which acquire
unauthorized access to the system through the

II-178
The implementation of IDPS requires is the simplest method because there is only a
certain resources that need to be integrated within comparing the current element of activity as a
the protected network infrastructure (Figure 1). package or a log entry with a list of signatures using
Typical components are as follows: strings compare operations.
 Sensor or agent. They monitor and
analyze activity. The term "sensor" is typical for the 3. Architecture of the system
IDPS who monitor networks and the term "agent" - The architecture of the presented system is
for those tracking only one device. modular (Figure 2). It comprises a multiple
 Management Server. It is a device that modules, each having a specific functionality. This
receives information from sensors or agents and allows easy modification of system functionality in
manages them. future extensions.
 Database server. It is a repository of
information regarding the events detected by
sensors, agents and management servers.
 Console. This is a program that provides
an interface for access to IDPS of users and
administrators.
The building of such infrastructure requires

substantial costs, both from financial and
organizational point of view. A number of
manufacturers offer complete solutions on the
market, including special devices and software that
are however acceptable only for large companies. Fig. 2. Architecture of IPDS
The IPDS technology uses several basic
The system consists of the following
methods for detection of attacks. Anomaly based
components:
detection is the process of comparing definitions of
what activity is considered normal for watched  Operating System under which runs the
events and then to recognize large deviations. Such software implementation.
a system has profiles that represent the normal  Software performing detection of
behavior of elements such as users, devices, attacks.
network connections or applications. The profiles  Events translator.
have been developed by monitoring the  Console for showing the results of the
characteristics of a typical activity for a given analysis of the network traffic.
period of time. A major problem in the generation
of profiles is that it can be quite difficult in some The main idea of the developed system is its
cases to be made precise, because of complexity of easy integration into any network infrastructure. For
the computer activity. this purpose is proposed a simple and efficient
Detection based on state of the protocols is solution using a single board computer Raspberry Pi
the process of comparing the pre-set common [3]. A characteristic feature of this system is its
definitions of normal actions for each protocol’s integrity - the concentration of all software
state with observed events to identify deviations. components on the powerful hardware, at the same
Unlike anomalies based detection that uses network time with small size and low cost of ownership.
or host-based profiles, this analysis relies on There are possible alternative
universal profiles developed by manufacturers that configurations of the presented architecture. One
describe how each protocol should or should not be configuration is a simplified version of that where
used. The main disadvantage of this type of analysis IDPS records the output messages directly in the
is that they require a great intensity of resources, database, skipping the binary file and the translator.
because of the complexity and the additional load In this case the performance of the software is low,
during the simultaneous tracking of multiple as it concentrates more functionality. In another
sessions. architecture a file for direct analysis at a later stage
A more efficient method is the detection is generated (for example, csv or tcpdump) and
based on the signature. This is the process of completely are absent translator, database and front-
comparing the signature and the observed events to end interface. The proposed system takes account of
be identified a possible attack. This approach is the convenience and functionality it provides the
very effective for detection of known attacks. In graphical console for the analysis of the detected
terms of implementation and use of resources, this
II-179
threats. Thus, it is realized an architecture with high has only one Ethernet interface, but for the
efficiency. realization of the system is required at least two (in
As an operating system is used Debian the best case and three network communication
Linux based system - Raspbian [4] which is one of interfaces). For this purpose, additional external
the officially supported according to the Raspberry USB modules are added - an USB-to-Ethernet
Pi Foundation and provides opportunities to meet adapter, an Ethernet card and a single USB dongle
the requirements of current implementation. for wireless WiFi connection. Thus achieves the
Under the operating system is running intended functionality via two interfaces through
another important component for the IDPS – the which to pass network traffic checked for problems
software for control of the traffic that passes and an interface to access the system for its
through the device. For fulfilling this role is used management. Both the Ethernet interfaces have no
Snort, which is a fundamental component in the IP addresses and are connected as a bridge running
network IPDS [5]. In this implementation the IDPS on the link layer of the OSI model, which is
interacts with network traffic through a transparent transparent to the upper layers.
bridge through which the traffic passes. The network system for detecting attacks is
For greater efficiency of Snort, the data situated in a part of the network that carries traffic
from the system are generated in binary file in between internal and external to the organization
unified2 format. This file is processed by translator networks. The Snort sensor must be behind the
- it parses recorded events and particular traffic and devices communicating through encrypted
redirects them to a database. The used translator is connection (VPN routers), so watched traffic is not
Barnayrd2, running parallel with Snort and encrypted. At the same time, it must be located
redirecting new events from the binary file to the behind a firewall that prevents any attempts to
database in real time [6]. As a database attack. Thus, in a security breach, the system will
management system is used MySQL [7]. detect the incidence and will notify by messages the
Key for each signature-based IDPS are lists responsible personnel.
of rules according which the events are defined as There are simulated Denial of Service
having possible degree of threat or not. These (DoS) attacks. Large amount of requests (hundreds
signatures are stored by Snort in several types of of thousands) from spoofed sources are generated
local files. Because of the dynamics of the security for the shortest period of time. This results in
systems, the information about each newly flooding of network. The attack is implemented by
discovered vulnerability or malicious code must sending only SYN packets, as the attack is directed
reach the protected systems as soon as possible. To to the router in the network.
satisfy this requirement, in the system is
implemented a tool which automates the process.
The company, which develops Snort, supports a
repository with signatures for detection of attacks.
The updates are performed by the tool PulledPork,
which connects to the central repository, downloads
the new rules and applies them in IDPS [8].
4. Experimental study and results
Testing of the presented system is done in a
real network infrastructure (Figure 3).
Fig. 4. Events classification
Over 65,000 messages (Figure 4) for

Fig. 3. Integration of IPDS into a network unusual behavior in the network are generated. Its
number continues to grow because the translator
An important issue is the management of continues to read from the binary file and fill the
the system. In order to avoid physical accessibility database with new messages, although the
to Razpbery Pi (keyboard or console terminal) there simulated attack is over. The reason for detected by
is applied a solution by adding third WiFi network the system anomaly are data in SYN packets,
interface. The standard model of Raspberry Pi rev.2 intercepted by Stream preprocessor of Snort.

II-180
The proposed IPDS is designed to be
integrated easily into any existing network
infrastructures. These can be both small business
networks and organizations, as well as large
corporate networks. The compactness of the system
allows its use in different segments of the network,
without the need for a physical change of its
topology. This is especially effective for networks
where is very important their operations to be
uninterruptable, such as bank branches, hospitals,
etc. The combination of cheap hardware and open
source software is a possibility for widespread use
of the system.
The goal of future work is the development
of architecture by using multiple Raspberry Pi
sensors distributed in various key network nodes to
redirect the output data to a centralized database,
thereby optimizing use of the IPDS system.
Fig. 5. Events before and after classification
REFERENCES
After the detection of the events it is
necessary to classify them. In this case, they are 1. Sanders C., Smith J. (2014). Applied Network
defined as DoS attack due to the amount of requests Security Monitoring: Collection, Detection,
from multiple addresses without completed TCP and Analysis, Syngress.
dialogue. This classification is done through the 2. Richard B. (2013). The Practice of Network
management console - Snorby and is illustrated on Security Monitoring: Understanding Incident
Figure 5. After the process of mass classification of Detection and Response. No Starch Press.
new events they are marked as a DoS attack. 3. RaspberryPi Project. http://www.raspberrypi.org.
4. Raspbian OS. http://www.raspbian.org.
5. Conclusions 5. Snort. http://www.snort.org.
The computer security will continue to be a 6. Barnyard2. http://www.github.com/firnsy/
serious problem in the coming years. However, with barnyard2
tools such as systems for attacks detection, 7. MYSQL. http://www.mysql.org.
countering threats would be an achievable goal. In 8. PulledPork. http://www.github.com/shirkdog/
this aspect, network security products developed for pulledpork/.
small organizations have great potential value.
This paper presents the architecture and
basic functionality of an affordable, portable and Contacts
easy to use system for detection and prevention of Hristo Valchanov, Kolyo Raychinov
network attacks. The system is based on open Technical University of Varna
source code, allowing tremendous flexibility in 9010, Varna, 1 Studentska Str.
need of modifications and further developments. phone: +359 52 383 278
E-mail: hristo@tu-varna.bg
SIMULATION FRAMEWORK FOR

REALIZATION OF PRIORITY-BASED LTE
SCHEDULER
VENETA ALEKSIEVA, AYDAN HAKA
Abstract: The LTE Technology provides simultaneously voice, data and video with
different priority on networks. LTE applies the QoS bearers technique that provides high
performance in packet delivery based on prioritization of the traffic. In this paper is
proposed a simulation framework for LTE technology, which realized a priority based
algorithm for LTE Scheduler, which reorders packets, based on classification mechanism.
Key words: LTE, Scheduler, prioritization, QoS
1. Introduction essential to focus on network traffic prioritizes for

In the middle of 2016 the European different types of communication streams.
Commission in [1] presents coordinated designation
2. Priority in LTE networks
and authorization of the 700 MHz band for wireless
broadband by 2020 and coordinated designation of After the first implementations of LTE the
the sub-700 MHz band for flexible use which focus in the allocation of resources is shifting
safeguards the provision of audiovisual media towards to the profit maximization and user
services to mass audience, as well as investments satisfaction [7]. Even with the developing of LTE
into more efficient technologies, which are needed technology the QoS for uplink is discussed by many
in order to vacate the current use of the 700 MHz authors [8,9,10,11,12].
band by digital terrestrial television. The prognosis In 3GPP, the QoS Class Indicator (QCI)
of these estimates require the search of optimal 4G consist of basic classes, which are defined as
solutions in terms of QoS offered by telecom "default", "expedited forwarding", and "assured
providers. According to published from Ericsson forwarding". It means: expedited forwarding is used
paper [2], suggests that by 2020 it is expected a for "strict" priority (video and voice), and "assured
growth of LTE subscriptions up to 3.7 billion. In forwarding" is used for business differentiation
February 2015 Cisco System published, in turn, (e.g., weighted-fair priority).
prognosis for the period 2014-2019 [3], where the In LTE network QoS is between end-user
Global mobile data traffic will grow three times devices and Packet Data Network (PDN) Gateway
faster than global fixed IP traffic from 2014 to applying the "bearers". "Bearers" is a set of network
2019. The Mobile data traffic will grow 10-fold configurations to provide a special handling of
from 2014 to 2019, a compound annual growth rate traffic to its set prioritization. Their hierarchy is
of 57%, and it will reach an annual run rate of 291.8 presented in Table 1. Default bearer is established
Exabytes by 2019, up from 30.3 Exabytes in 2014. when the user equipment (UE) connects to the LTE
Requirements of IMT-Advanced [4] are 1 network, while Dedicated bearer is established
Gbit/s speeds for fixed and 100 Mbit/s for mobile whenever must be set QoS for a specific traffic type
users. Up to 2016 under these requirements, (service) as VoIP, video and etc.
providers of 4G services choose between two GBR (Guaranteed Bit Rate) provides
advanced wireless technologies - LTE [5] or guaranteed bandwidth and monitors two parameters
WiMAX [6], but since 2016 widely used 4G in directions uplink and downlink:
technology is LTE and many vendors implement • GBR- minimum GBR for EPS (Evolved Packet
only this technology in their end devices. switched System) bearer,
Solutions are needed to improve QoS in • MBR- maximum GBR for EPS bearer,
terms of delays in larger loads and any packet loss. Non-GBR bearer does not provide
For communications to be successful, it is also guaranteed bandwidth and also monitors two
parameters in directions uplink and downlink: A-
II-182
AMBR-general maximum speed permitted for the into Resource Blocks (RBs) in the frequency
entire non-GBR throughput for specific APN domain. The Data is transmitted in the Transport
(Access Point Name) and UE -AMBR- overall Blocks (TB) in one transmission time interval (TTI)
maximum speed permitted for the entire non-GBR for 1ms. Each RB consists from 12 subcarriers
throughput for all of APN particularly UE. (each of them is 15kHz). The frame is 10ms and
divides into 10 equal subframes. Each subframe
Table 1. LTE QoS contains 2 slots*0.5ms. Each RB is related to one
Dedicated Bearer Default Bearer
slot in time. One TB is related to 1 subframe and it
Non-GBR GBR Non GBR
is the minimum unit to schedule. The serve rule is
QCI 5-9 QCI 1-4 QCI 5-9
to find first space that can fit the TB. If there are not
APN-AMBR GBR APN-AMBR
enough RBs in the current TTI, the scheduler tries
UE-AMBR MBR UE-AMBR to find resources in the next TTI. This strategy
TFT TFT APN minimizes the response latency, which is the best
ARP ARP IP Address
L-EBI L-EBI ARP practice for delay sensitive traffic.
But in 1 Timeslot number of RB depends
In LTE networks for differentiation of QoS
from frequency. It is presented in the Table 3.
same as in WiMAX are applicable classes which
here are called QoS Class of Identifier (QCI). They Table 3. Number of RB in
define the basic characteristics of the IP packet LTE
level, as presented in Table 2.
frequency RBs for users traffic RB for overhead
Table 2. QCI classes in LTE 25MHz 4 RB 2 RB
2. 5MHz 10 RB 2 RB
QCI Bearer Priori Delay of Packets Example of 5MHz 23 RB 2 RB
Type ty the Packet Loss Traffic Type 10 MHz 48 RB 2 RB
1 GBR 2 100ms 10-2 VoIP 15MHz 73 RB 2 RB
2 4 150ms 10-3 Video call 20 MHz 98 RB 2 RB
3 3 50ms 10-3 Real time But this procedure is not applicable for
games
4 5 300ms 10-6 Video stream
beacon transmissions (it is sent among devices each
5 Non- 1 100ms 10-6 IMS Signaling 100ms), because of emergency information it
6, 8, GBR 6, 300ms 10-6 TCP based conveys, therefore the reserved resource blocks
9 8, services – chat,
9 ftp...
exist to accommodate the temporary overload.
7 7 100ms 10-3 Voice, video, The present paper offers an algorithm for
interactive UEs service in the distribution of resources in the
games
uplink of LTE network as composed of two
Then in the base station (eNodeB) is modules - by a control mechanism for admission
applied a preemption algorithm, which allows high (admission control) and Scheduler.
priority requesting bearers to displace low priority
connected bearers in order to reduce the cell load.
This algorithm coupled with a priority-based
admission control can achieve low dropping and
blocking probabilities.
3. Proposed algorithm for prioritization of
UEs in the LTE
The aim of the proposed algorithm is to
achieve keeping the network throughput as high as
possible at a small price of only a bit more
handovers. The functions for management of QoS
in access networks are responsible for the efficient
allocation of resources in a wireless interface. They
are generally defined as the control algorithms of
radio resources and incorporate power
management, control of the transfer connection,
access control , load control and the management Fig. 1. Traffic Prioritization in the Scheduler
packet , but directly related to QoS level cell are the
last three. They are used to ensure a maximum According to the network load, the
throughput for individual services. admission control for the reception of orders
LTE uses multiple access technology manages the number of UEs, which can enter into
(OFDMA), where the total bandwidth is divided
II-183
the Scheduler, in order to avoid overloading the The module "resource Allocation
system with too many UEs. Generator", based on class "eNodeBdata", realizes
The Scheduler allocates RBs among UEs the proposed algorithm for priority. The class
according to UEs needs. Resource allocation in the contains methods for sorting UEs, adding it's data in
Scheduler is based on the priority, which is array and arranging them.
presented on the Figure 1. Thus, the input of data for each device starts
from initial parameters for the eNodeB. This is
4. Simulation Framework of LTE Scheduler represented in the example of Figure 4. Another
In this approach a simulation environment eNodeB (the next in the order) receives serial
was established for implementation and exploration number. After that the user must type the number
of the proposed algorithm. Used software tool is of UEs that will participate in the network of this
Visual Basic 2010. The architecture of the simulator base station, according to the limits imposed by the
is presented on the Figure 2. standard. The choice of bandwidth sets limits in the
cell radius of the eNodeB and the transmission
speed (bandwidth).
Fig. 2. The LTE simulator's architecture
The modules "Topology Maintenance" and

"Topology Modification" are realized with classes
"Form1" and "Form2". The classes contain
methods for adding parameters of eNodeB and
related UEs. A database for storing data from
individual experiments for each eNodeB and its
connected UEs is created. The tables from database
with parameters for coordinator and related devices Fig. 4. Input form of the parameters of eNodeB
are presented on the Figure 3.
The tabular presentation of base stations
and the information for UE (Figure 5) is suitable for
displaying of multi aspected information that can be
edited (Figure 6).
Fig. 3. Database of LTE Prioritization in the

eNodeB
The module "Traffic Management" is

realized with class "Form3". It loads data into the
'UserEquipment' table and visualizes the chart of the Fig. 5. Database with information of eNodeB and
timing diagram. related UEs

II-184
prioritization. When the number of users grows up,
the numbers of rejected beacons grow up too. This
is the reason to increase the drop ratio when the
number of the users is increased. The observed
parameters degradation when connecting more
users is related to the priority implemented
scheduler, in which case the less priority queues
may not be served in the case of network overload
or congestion.
In assessing the performance of proposed
algorithm of LTE Scheduler it also is assumed that
UEs are evenly distributed within the range of the
cell according to Section 7.1.6.1 of [13]. Each UE
sends a signal periodically to know the channel
condition of the UE for each TTI period of 1ms.
Fig. 6. Data of UE, connected to the eNodeB Figure 9 shows the dependence of the
average time to establish a connection from UE to
One example of resource allocation and
the eNodeB at different average intensity of
transmission matrix is presented on figure 7.
applications when under area of the eNodeB are 100
UEs. An increase of the time for establishing a
connection from the UE to eNodeB when the
intensity of the requests is increased, as for the first
10% of the increase is from 10ms to 17ms, then the
increase is significantly smoother and it amended in
the range of 18ms to 23ms.
Fig. 7. Resource Allocation and Transmission

Matrix
5. Experimental Results
The data from different experiments are
send in .xls format to the next estimation.
Fig. 9. Average time for connection establishment

with 100 UEs
The results give reason to conclude that the

presented algorithm for admission control in the
Scheduler for LTE network can be applied
successfully in the number of UEs under 100,
because regardless of the intensity of the requests of
the active UEs the average connection time is under
25ms-time, fully satisfying the requirements of [13].
Fig. 8. Drop Ratio In the Figure 10 it presents the dependence
of the average time to establish a connection from
On Figure 8 is presented the Drop ratio. It is the UE to eNodeB depending on the number of the
possible to compare the number of drops when it is UE, which are within the scope of the eNodeB for
applied the prioritization, based on distance three different values of intensity of applications -
between the user and Base station and drops without 10%, 50% and 100%.
II-185
http://www.cisco.com/c/dam/assets/sol/sp/vni/f
orecast_highlights_mobile/index.html#~Regio
n, February 2015.
4. (2008). Document IMT-ADV/1-E, Background
on IMT-Advanced, ITU Radiocommunication
Study Groups, ps. 3, 7 March 2008.
5. (2014). LTE: capacity and cell-edge
performance improvements, February 2014,
http://www.3gpp.org/.
6. Ramadas K., R. Jain, (2008). WiMAX System
Evaluation Methodology v2.1, WiMAX
Forum, July 2008.
Fig. 10. Average time for connection establishment 7. Abu-Ali N., Abd-Elhamid M. Taha, Mohamed
with requests intensity of 10%, 50% and 100% Salah, and Hossam Hassanein (2014). Uplink
Scheduling in LTE and LTE-Advanced:
It is easy to see a trend of increase in the Tutorial, Survey and Evaluation Framework,
time for establishing connection with an increase in IEEE Communications Surveys & Tutorials,
the number of UE within the eNodeB, such as at a vol. 16, Issue:3, pp. 1239-1265, Aug. 2014.
higher intensity time is significantly larger. If 8. Al-Rawi M., R. Jantti, J. Torsner, and M.
number of UEs is 200 and 100% intensity is Sagfors (2008). On the Performance of
reached 92ms, while number of UEs is 200 and Heuristic Opportunistic Scheduling in the
10% intensity the average time is 3 times less - 32 Uplink of 3G LTE Net- works, in Personal,
ms. Indoor and Mobile Radio Communications,
6. Conclusion 2008. PIMRC 2008. IEEE 19th International
Symposium, pp. 1 –6, 15-18 2008.
In this paper is proposed a simulation
9. Calabrese F., C. Rosa, M. Anas, P. Michaelsen,
framework for LTE technology, which realized an
K. Pedersen, and P. Mogensen (2008).
efficient method for QoS for LTE service classes. In
Adaptive Transmission Bandwidth Based
this framework is realized an algorithm for
Packet Scheduling for LTE Uplink, in
Scheduler to prioritize users in order to fit the
Vehicular Technology Conference, 2008. VTC
bandwidth requirements, which are satisfying the
2008- Fall. IEEE 68th, pp. 1 –5, 21-24 2008.
application needs, based on the number of users and
10. Calabrese F., C. Rosa, K. Pedersen, and P.
the prioritization of users. Simulation’s results show
Mogensen (2010). Performance of Proportional
that the proposed mechanism improves QoS, but the
Fair Frequency and Time Domain Scheduling
observed parameters degrade when the use of more
in LTE Uplink, pp. 271 –275, May. 2009.
subscribers is related to the priority implemented
11. Liu F., X. She, L. Chen, and H. Otsuka (2010).
Scheduler, in which less priority queues may not be
Improved Recursive Maximum Expansion
served in the case of network overload or
Scheduling Algorithms for Uplink Single
congestion. There are presented drop ratio with
Carrier FDMA System, pp. 1 –5, may. 2010.
prioritization and without prioritization. It was
12. Ruiz de Temino L., G. Berardinelli, S. Frattasi,
always assured a minimum transmission for all the
and P. Mogensen (2008). Channel-Aware
service classes, although with different
Scheduling Algorithms for SC-FDMA in LTE
performances due to prioritization.
Uplink, in Personal, Indoor and Mobile Radio
Communications, PIMRC 2008. IEEE 19th
REFERENCES
International Symposium, pp. 1–6, 15-18 2008.
1. EC (2016). Use of the 470-790 MHz frequency 13. 3GPP (2015). 3GPP, Tech. Specif. Group
band in the Union, http://eur- Radio Access Network; Physical layer
lex.europa.eu/legal- procedures, 3GPP TS36.213, ver.12.4, 2015.
content/EN/TXT/?qid=1454410061980&uri=C
OM%3A2016%3A43%3AFIN, June 2016.
Veneta Aleksieva, Aydan Haka
2. P. Cerwall and others (2015). Ericsson
Mobility Report-On the Pulse of the Organization: Technical University of
Varna
Networked Society,
http://www.ericsson.com/res/docs/2015/ericsso Address:Str. Studentska 1, Varna , 9010
n-mobility-report-june-2015.pdf, June 2015. Phone (optional): +35953383439
E-mail: valeksieva@tu-varna.bg,
3. Cisco System (2016). VNI Mobile Forecast
Highlights 2016-2021, aydin.mehmed@abv.bg
АВТОДИАГНОСТИКА ПОСРЕДСТВОМ
ДЕЛФИ DS150E
ИБРЯМ АХМЕДОВ, ЕБРУ АДЕМ, НИКОЛИНКА ЯНКОВА
Резюме: Delphi DS150E е софтуер, предназначен за лесно диагностициране на

моторни превозни средства. Той улеснява многократно работата на човека
предназначен обучен да работи с този софтуер . Целта на софтуера е да открива
грешки съответно и причината за тази грешка , чрез този софтуерен инструмент
откриването на електрически повреди се намалява многократно. Посредством
софтуерния инструмент можем да постигнем желания положителен резултат за
отрицателно време също така без това устройство с просто око не може да се
рискува при поправка на автомобилна електроника.
Ключови думи: автомобилна диагностика, интерфейс за превозни средства
DIAGNOSTICS USING DELPHI DS150E

IBRYAM AHMEDOV, EBRU ADEM, NIKOLINKA YАNKOVA
Abstract: Delphi DS150E – software designed to easily diagnose vehicles. It makes it

much easier for people who are trained to work with this software. The aim of this software
is to detect errors and the reasons for those errors respectively. With this software tool
finding electrical failures is much easier. Through this software tool we can achieve the
desired positive results in no time. Also, without this device we can’t risk to make repairs of
car electronics with a naked eye.
Key words: On-board diagnostics, Vehicle Communication Interface
1. Увод Инструментът използва оригиналните

DS150E е най-новия модел диагностичен данни от производителите на превозни средства.
интерфейс от продуктовата гама на Delphi. Това е необходимо условие, за да бъде в
Общият му вид е показан на Фиг 1. състояние да предложи качество и
функционалност.
За да направи работата по-лесна,
инструментът съдържа редица функции, за да
помогне на потребителя да направи правилните
решения. Базата данни често се актуализира с
оригиналните данни от производителите на
превозни средства. Софтуера е 100% преведен
на БЪЛГАРСКИ език в софтуера за леките коли
и около 40% при тежкотоварните.
Устройството има функцията Flight
Recorder, за запис на „живи данни“ по време на
движение, чрез microSD карта памет. Като
допълнение е монтиран светодиод на OBD
конектора, за по-лесно намиране на OBD
Фиг. 1. Делфи DS150E буксата. Поддържани интерфейси от
устройството за диагностика са USB и като
допълнителна възможност – Bluetooth.
II-187
2. Особености на устройството за - изтрива текущите кодове на грешки
диагностика: - тестване на изпълнителните механизми
 Конектор с LED – за да намерите - адаптация на отделни параметри и
конектора за диагностика на автомобила с изтриване на адаптационните стойности
лекота, дори ако той е разположен в тъмно и - показва каталожния идентификационен
недостъпно място номер, записан от производителя на
 Проверка на напрежението – когато тестваното ECU, версията на записаната в
DS150 е свързан към превозното средство, него програма, кодировката, датата, на
уредът ще провери напрежението на която е програмиран и други.
автомобила и автоматично се настройва към
3. Диагностика на дебитомер (MAF)
нивото на напрежение на превозното средство
от 12 или 24 волта. Ако напрежението стане 3.1. Какво е дебитомер.
твърде високо или твърде ниско, DS150 ще ви Дебитомерът (MAF) е отговорен за
предупреди със звук, светлина и чрез иконата преносът на данни които казват на ECU-то какво
на батерията в нашия диагностичен софтуер. количество въздух постъпва. Това е относително
 Номер на шасито VIN – софтуерът за прост процес. Основната част в този сензор е
коли поддържа функция, която ви позволява горещия елемент. Когато колата се движи и
да прочете номера на шасито от превозното въздух минава през сензора, той има охлаждащ
средство, което бихте искали да се ефект върху него. ECU-то компенсира това
диагностицира. Това гарантира, че охлаждане като изпраща точна информация за
правилният модел и година е избран горещия елемент и това променя температурата.
автоматично. Това повишение в текущите данни се отчита от
 ISS – Intelligent System Scan – сканира ECU-то и се конвертира в колко въздух е
всички системи в автомобила и показва преминал през сензора (и от там постъпил в
кодовете за грешки, които се съхраняват във двигателя).
всяка система. Това спестява време и можете
да получите бърз преглед на актуалното 3.2. Причини поради което сензорът не
състояние на цялото превозно средство. отчита правилни параметри.
 ISI – интелигентна система за Елемента на дебитомера (MAF) може да
идентификация - (ISI) идентифицира и избира се зацапа с прах и масло (от нередовна смяна на
автоматично вида на контролера на въздушния филтър или от маслените пари от
превозното средство. Това гарантира, че отдушника на картерните газове). За съжаление
диагностичната сесия се извършва правилно с тези малки проблеми могат да доведат до
точните параметри. повреда в дебитомера (MAF), също проблем
 Функция ДОКЛАД – с помощта на тази може да създаде и масивното преминаване на
функция, вие ще бъдете в състояние да видите вода през него. Тези фактори играят ролята на
адаптациите и корекциите, които са възможни изолатор или охлаждащ ефект. При Bosch
за конкретен автомобил, без да се налага да дебитомерите (MAF), тези причини могат да
сте в близост до превозното средство . Заедно доведат до прегряване и последствие да бъде
с помощните текстове като ръководство, физически деформиран елемента в сензора. Това
можете да планирате и да бъде ефективни в ще доведе до некоректни данни които се
работата си дори и при сложни ситуации. изпращат до управляващия блок (ECU). В
 OBD функции – Диагностичният следствие на това се намалява горивото за да се
софтуер DS150 е проектиран с технология, предотврати черния пушек поради намаления
която му позволява да комуникира с обем въздух.
всички видове протоколи. За лекотоварни повреден. Сменете го.
и тежкотоварни средства, които не 3.3 Почистване на дебитомер
използват стандартния 16-пинов конектор, Почистването на дебитомер е важно
са предназначени други видове конектори поради факта замърсяване на сензора
като например : конектор за BMW кръгъл следователно което води до грешни отчитания
20 пинов, конектор за Mercedes-benz на данни.Вади се дебитомера инсталиран на
кръгъл 38 пинов, конектор за Opel кръгъл входящата връзка към смукателните колектори .
10 пинов, и др. Изважда се връзката куплонга и след
 Други функции на продукта са: това с спрециален спрей предназначен за
- връща базовите настройки на автомобила измиване на дебитомер като например този:
- нулиране на сервизни интервали
- чете текущите кодове на грешки
II-188
Така ЕБУ на двигателя, разполага с максимално
прецизна и мигновено опресняваща се
информация за количеството постъпващ в
двигателя въздух.
Измива се сензора и се оставя да

изсъхне. Тази маниполация се препоръчва да се
извърши на изключена клема на батерия
(акумулатора).
4. Принцип на работа на MAF – Mass air
flow sensor (дебитомер)
Дебитомер е сензор, който преобразува
количеството въздух постъпващо в двигателя в
напрежение подавано към ECU на двигателя
(или ЕБУ – електронен блок за управление).
Към ECU на двигателя трябва да
постъпва информация за количеството входящ
въздух в двигателя, за да може да се изчисли
натоварването на двигателя (engine load). Това е
важно, за да се определи колко гориво да се
впръска, кога да се подаде искра за запалване (за
бензинов двигател) и дори кога да превключи
трансимията ако е автоматична. Дебитомера
обикновено се намира между въздушния филтър
и тялото на дросела. Фиг. 2. Mass air flow sensor (дебитомер)
Съществуват различни видове
дебитомери – механични и електронни. Пример Този тип дебитомер има и вграден
за механичен такъв е VAF или vane air flow датчик за температурата на постъпващият в
meter – по-стар модел дебитомер, като типа му двигателя въздух или IAT – intake air
може да бъде определен по формата му. При temperature. Той подава информация към ЕБУ на
него измервателна плоскост се отмества двигателя и така допълнително спомага за
пропорционално на обема въздух постъпващ в поддържането на коректното отношение
двигателя. Демпферна камера намалява рязкото гориво/въздух.
придвижване на измервателната плочка. Възможни причини и решения на
5. Дебитомер – тип MAF с нагреваема проблема са:
жичка 1. Винаги използвайте високо качествени
хартиени филтри. Това ще държи прахта далеч
Основните компоненти на този тип от дебитомера (и далеч от вашият двигател).
дебитомер са термистор, платинена нагреваема Когато монтирате филтърът убедете се, че е
жичка и контролна електрическа верига, хубаво легнал в кутията си.
показани на Фиг. 2. 2. Проверявайте често въздушният филтър
Термистора измерва температурата на и почиствайте кутията му.
входящият въздух. Нагреваемата жичка е с 3. Ако сензора не е прекалено замърсен или
константна температура поддържана от ЕБУ, с повреден, понякога може да бъде почистен
помощта на сигнала от термистора. Повишаване (понякога дава добри резултати). Може да бъде
на потока входящ въздух, причинява понижение почистен с контактен спрей, но не докосвайте
на температурата на нагреваемата жичка и блока горещата част с нищо (пръст, парцал и т.н.).
за управление на двигателя компенсира това Оставете го напълно да изсъхне преди обратно
незабавно, чрез подаване на по-голям ток през да го монтирате.
жичката. Контролна електрическа верига 4. Буксата също може да се замърси или да
същевременно отчита подаваният ток и не е добре вкарана. Това може да доведе до
пропорционално го преобразува в напрежение.
II-189
некоректни стойности. Почистете я с контактен ориентира авторемонтните работници къде по
спрей, това може да нормализира нещата. цялата съвкупност от механизми и електроника
да търси проблема.
6. Измерване на живи данни на изряден Откритите кодове на грешки, записани в
дебитомер (MAF) контролен блок от превозното средство, могат
Delphi DS150E позволява следенето на да посочат единицата, която е дала отказ.
„живи“ данни в реално време, както са показани Не на последно място е цената на
по-долу. При невъзможност за излизане на трасе диагностичния интерфейс, която е достъпна за
с компютър, производителите на устройството малките и средни автосервизи.
са дали възможност то самостоятелно да записва
върху SD карта памет.
Величина на измерваната единица ЛИТЕРАТУРА
количество въздух на софтуера DELPHI DS150E
1. User Guide for DS150E.
е : mg/stroke .
2. http://europe.delphi-dso.com/
Измереното количество въздух
3. http://www.obd-codes.com/trouble_codes/
294mg/stroke искано при 2112 оборота,
съответно, по този начин ЕКУ-то изисква 294
mg/stroke въздух Фиг. 3. Ибрям Сюлейманов Ахмедов
Студент в Шуменски университет
„Епископ Константин Преславски“, гр.
Шумен 9712, „Университетска“ 115
Телефон: +359 988 372 577
E-mail: ibrqm_96@abv.bg
Ебру Гюнай Адем

Телефон: +359 896 554 297
E-mail: ebru.9500@abv.bg
Фиг. 3. Измерване на количеството въздух
Работа на празен ход 800 оборота в Николинка Спасова Янкова

минута съответното желано (поискано)
количество въздух 251 mg/stroke дадено на Фиг.
4.
Телефон: +359 877 884 519
E-mail: niki_qnkova@abv.bg
Фиг. 4. Измерено количество на празен ход

(работа на място)
7. Заключение:
Полезността на софтуера е особено
голяма в използването му при всички
съвременни леки и товарни автомобили.
Проследяването на текущите параметри и
сравняването им с желателните много бързо
BENCHMARKING HASH FUNCTIONS

SVETOSLAV ENKOV, TONY KARAVASILEV
Abstract: This paper presents the most effective use cases of hash functions. The purpose
of the developed practical tests is evaluating hash functions by speed and security. They
show how the different cryptographic algorithms, application level hashing, database layer
hashing and even disk encryption can impact both the overall performance of a system and
are crucial for the security of the data implied in it. The results of this experimental
research are presented in this paper.
Key words: hash functions, cryptographic algorithms, security, performance, encryption
1. Introduction  MD5 – not suitable for securing data [4], but

Hash functions are any functions that can be usable for caching, verify data integrity, etc.
used to map data of various size to data of fixed  SHA-1 – not suitable for protecting data [5], but
size. The value returned by a hash function is called usable for detection of corruption, identifying
a digest, hash value or just hash and are commonly duplicates, etc.
represented as a hexadecimal number. [1]  SHA-2 – a set of six hash functions that are
There is a special class of cryptographic cryptographically secure and can be used for all
hash functions that can map data with arbitrary size purposes. Every one of them has a different
to a fixed bit string. This class is characterized by output size and level of security. They produce
the one-way mathematical design used in the 224, 256, 384 or 512 bit hash values. Those
generation of digests and can also be referred as a algorithms are SHA-224, SHA-256, SHA-384,
digital fingerprint of the input data. Said in others SHA-512, SHA-512/224, SHA-512/256. The
words, every input should have an associated last two are rarely used.
unique digest output without a chance of finding
duplicates. [2] The main purpose of this article is testing
Hash functions are used a lot in computer the performance of different algorithms when used
science and have many information-security in various ways. This way we can define the correct
applications, such as: use cases for each algorithm and find its best use.
The main tests are focuses on the hashing of 100000
 Identifying duplicated files or substrings; strings in three lengths:
 Detection of accidental data corruption;
 Password type size – 20 symbols;
 Providing forms of authentication;
 Identifier type size – 50 symbols;
 Secure storage of sensitive data;
 Comment type – 100 symbols.
 Fingerprinting of data;
 Digital signatures; The hashing is tested both from an
 Caching identifiers. application point with PHP web pages and from
database level point with MySQL stored
Nowadays every modern programming procedures. For both PHP and MySQL, we used the
language, database script language, software internal realizations for the algorithms MD5, SHA-
framework or plugin library has a set of 1, SHA-244, SHA-256, SHA-384 and SHA-512,
cryptographic functions that comes preinstalled being the mostly used hash functions. We have
with it. Also, every hash function can have a performed two tests for each layer:
different kind of complexity, amount of required
resources, security level, performance grade and  Read and hash the values;
realization for the its algorithm.  Read, hash and update the values.
The most frequently used functions in
software development currently are: [1] [3] Besides that, this paper also includes an
overview of the time that would be needed when
combining both application and database hashing
II-191
for higher security, other problems that may change update. The next two sections show how PHP and
the software climate and how can using disk MySQL handle those two with different string
encryption effect the performance and resources of lengths.
the software system, while boosting protection.
All the practical tests in this article have 3.1. Read and hash experiment
arisen from the need of implementing both the This experiment will test the situations
protection of sensitive data, passwords verification when you need to get some string data and hash it
and generating caching identifiers for an improved so that you can do some sort of verification,
second version of a private online website. comparison or uniquely map the data. For PHP, we
used an array with the strings to iterate and hash
2. Testing environment specification them. While for MySQL, a table with the data in it,
For the results to be adequate we have so that we could do a select statement with an alias
chosen to run the test on a virtual machine, created that calls the current hash function and returns the
with Oracle VM VirtualBox version 5.1.14, that has result for all rows.
a typical web development stack installed on it. The
specification of the allocated resources for the 3.1.1. Password length results
virtual machine and the software on it are shown on For the password type size of 20 symbols
Table 1. experiment, the results for application and database
hashing times are shown on Table 2.
Table 1. Virtual machine specification
Table 2. 20 symbols hashing results
Detail
CPU Intel i7-6700HQ, 2 cores, 2.59GHz PHP MySQL
RAM DDR4, 2048 MB, 2.40 MHz MD5 0.026375 0.050560
GPU Intel HD Graphics 530, 32 MB, 2.40 MHz SHA-1 0.033895 0.058299
HDD 20GB, 7200 RPM, 32 MB cache SHA-224 0.059067 0.078799
OS Ubuntu Server 16.04.2 LTS, Kernel 4.4.0 SHA-256 0.059711 0.081613
LAMP Apache 2.4, PHP 7.0.15, MySQL 5.7.17 SHA-384 0.072731 0.096270
SHA-512 0.075571 0.113985
The virtual machine has installed all
available updates, kernel drivers and virtualization As we can see from the results, the
needed packages. All settings for the LAMP stack application level hashing is faster than the database
(Linux Apache MySQL PHP) are by default, with one. The MD5 and SHA-1 can be easily used for
the exception of boosting the values for maximum generation of identifiers from PHP, but not for
memory usage by both PHP and MySQL to allow passwords. The SHA-2 algorithms performance is
the allocation of all available random-access pretty close on both sides, so you can easily use it
memory (RAM). for protecting data and moving some of the pressure
The tests times for PHP are just for the away from the application code to the database.
section of the program that does iteration, hashing Because this test is overviewing sensitive
and updating of values. The time needed for data hashing, stick with the SHA-2 algorithms for
generating 100000 strings with the various length is better security.
explicitly excluded.
The MySQL database has three tables 3.1.2. Identifier length results
having per 100000 strings, for each of the three When using hashing for caching purposes,
length types tested. The time for generating digests you would probably need an identifier for data of
and running the select or update statements is only size about 50 symbols. The performance of all the
taken under consideration. chosen hashing algorithms can be seen on Table 3.
Each single experiment is executed 10 times
and the average time of those runs is taken as final. Table 3. 50 symbols hashing results
All the results will be in seconds with 6-digit PHP MySQL
precision after the decimal point. MD5 0.026706 0.049312
3. Application vs. Database level hashing SHA-1 0.034188 0.055153
SHA-224 0.059884 0.082104
Most of the developers are feed up with the SHA-256 0.060874 0.094240
dilemma where to put the encryption logic and how SHA-384 0.072884 0.107898
much will it cost the system. The two daily SHA-512 0.076474 0.122992
operations that every kind of cryptography logic
needs to imply are read-hash-use and read-hash-
II-192
The final results show that it is better to use paper. The point of this test is mostly for identifying
application layer hashing when using multiple time the generation times differences on each side.
generation of identifiers and that the MD5 and For PHP, we used an array of arrays to
SHA-1 are the fastest to compute. However, if you contain both the input data and the computed hash
are using it for one-time generation for some third in the same subarray. As for MySQL, we used a
party in memory table or MySQL memory table, it table with the data in it and added a second column
probably will not slow the software system much. for the computed digests. This way we could easily
This type of strings is probably used for do an update statement that reads the first column,
caching or mapping and this means security is only computes the hash of each value and saves it to the
needed for the cached data, but not for the second column.
identifier. This means you can use SHA-1 for It should also be noted that when using hash
balancing performance with ease. functions with longer digest output size, you should
prepare for bigger storage sizes.
3.1.3. Comment length results
Sometimes the developer needs to hash 3.2.1. Password length results
longer data, for example finding duplicate For the password type size of 20 symbols
comments that can be about 100 long and above. experiment, the final results are shown on Table 5.
This is why we will use string length of 100
symbols for this experiment. The performance of Table 5. 20 symbols hashing results
the hash functions for this experiment is shown on
PHP MySQL
Table 4. MD5 0.028027 0.317091
SHA-1 0.035334 0.363700
SHA-224 0.057294 0.398448
PHP MySQL SHA-256 0.061983 0.441395
MD5 0.037308 0.063266 SHA-384 0.075343 0.450580
SHA-1 0.048937 0.075928 SHA-512 0.079736 0.592357
SHA-224 0.094610 0.128636
SHA-256 0.100509 0.130748 Since MD5 and SHA-1 algorithms are
SHA-384 0.072727 0.105875 security broken, it is recommended to use the SHA-
SHA-512 0.076085 0.106118 2 ones. As we can see from the test, the stronger the
algorithm, the more computation time is needed.
The length of input data seems to be Having in mind that passwords are always
affecting every algorithm in a way. The first thing sensitive data, you would probably want to use
that can be notice is that SHA-224 and SHA-256 application data encryption to reduce the time of
computation time is slower that the more secure insert or update in SQL. Still, adding a second layer
algorithms SHA-384 and SHA-512. The second of encryption from the database side will boost
thing is that you can use SHA-512 hashing from the security a lot, but will cost you only a bit of
database side and be as fast as using SHA-256 from performance.
the application layer.
Comments sometimes may be sensitive 3.2.2. Identifier length results
information and you should be careful what you are When needing to map unique data and save
comparing, use SHA-2 algorithms. You can drop it somewhere for further use, you would need an
some of the computation pressure by using SHA- identifier generation operation. For this test, the
512 from the database side or stick to the faster needed data to map should be at least 50 symbols.
SHA-512 from application level. The computation and save time can be seen on
Table 6.
3.2. Read, hash and update experiment
The second test will apply to the scenarios Table 6. 50 symbols hashing results
where you are obliged to save the computed hashes.
PHP MySQL
It is way normal of saying that the application side MD5 0.028937 0.343682
results will be faster here. This is only because SHA-1 0.034090 0.375937
resaving data for use there will happen in the SHA-224 0.061613 0.374757
random-access memory which is faster than any SHA-256 0.063669 0.381966
available storage drives today, but is only temporary SHA-384 0.075568 0.434025
media. We will talk about combining application SHA-512 0.078516 0.459434
and database hashing for more security later in this
II-193
For this data length, using a stronger The process of encrypting and decrypting is
algorithm needs more time. When generating directly in the random-access memory and this
identifiers, you can easily choose MD5 or SHA1. creates the so called transparent encryptions effect,
If you do want to use another algorithm, only because of the high speed of RAM. When you
have in mind that you would need more space for need to write something on your disk device, the
the output digests. You would probably need to map central processor will first encrypt it and after that
the data once and only add some new records over send it to the storage device. [7]
time, but you should definitely use the application If you are a professional, using this kind of
side computing to save time. Security is not a factor encryptions can only boost your security status and
for the identifier, only for the underlining data that it will not affect your performance in any fatal way.
it maps to. The only risk is that when rebooting you need to
input your encryption password to boot into the
3.2.3. Comment length results operating system and it would be fatal if you have
When needing to map comments or having lost it or forgot it somehow. If you get any resource
their hash for faster comparison, the developer shortage, switch to partition encryption instead.
needs to compute digests for longer strings. We will In any way, using disk encryption will
use a 100 symbols length for this experiment. The protect your data, but can also be a double-edged
computation results are shown on Table 7. sword. Use it with caution and do not try to
optimize things before they need to.
4. Coping with the combined approach and
PHP MySQL real world problems
MD5 0.038966 0.332903
SHA-1 0.050947 0.359913 When dealing with sensitive and top secret
SHA-224 0.102294 0.381809 data in the real world some other needs,
SHA-256 0.104354 0.401970 specifications and problems may occur. Even when
SHA-384 0.075037 0.424538 money and people power are practically unlimited,
SHA-512 0.078616 0.488914 you can get bad results if you do not give
professionals the time needed to get rid of all your
The results show that SHA-384 and SHA- problems at once.
512 behave better with bigger strings on the The next sections show the most frequent
application side, even faster than SHA-224. The situations and the extra needs that may occur.
database level hashing time seems to be affected
4.1. Combined approach
only by the complexity of the current algorithm.
If you will be using the hashing for In a software system that deals with private
comparison only, it is not a problem using even and security information, an extra need for
MD-5 or SHA-1. In any other situation stick to multilayer security may apply. In this case,
SHA-2. Because comments can be edited often, the combining hashing from both application and
smartest thing is avoiding database computations. database levels is a good choice.
When faced with this challenge, it is better
3.3. Disk Encryption aftermath to lay the complex encryption algorithm in
We cannot skip that having a full software application level and just send it to the database for
disk encryption is a big factor in securing your data. a one-time hashing before saving. As said before,
So we did a research on the performance of disk while both sides do calculations in the random-
encryption and retested the above results on a fully access memory, the database side will need more
encrypted virtual machine, equivalent to the time because it saves data on a disk for further use.
machine we tested before on. An overview of a real world situation would
The interesting thing that we found out be where PHP generates a digest, sends it to the
from both theory and practice is that it will not database for MySQL to generate a hash value of it
affect computation time in any way. This is because and saves the final hash to disk. Using the data from
modern processors have a special set of instructions the experiments and calculating an average of the
called AES-NI [6] that allows encryption of up to 1 time needed for application generations, database
Gbit of data per second fully transparently. computations and saving the values to storage, the
final results per algorithm can be seen on Figure 1.
II-194
4.2. Real world complications
Sometimes reality surprises us and changes
our needs in an unpredicted way. That said: “If you
believe everything you read, better not read.” -
Japanese Proverb.
For an example, let say we have a fast
application that needs 3 seconds for application
computations and 1 second for database querying.
You decide to add four time hashing in the
application level and one-time digest generation at
the database layer.
Your expectations should be that the whole
software will load for maximum of 5 seconds. You
profile it and it turns out to loads for 7 seconds. But
Fig. 1. Combined approach times how can this be possible? This is only because you
have not taken these several factors:
It is easy то conclude that the heaviest part
of this approach is the process of saving data into  Network time – when the connection between
the database. Application and database hashing are the application and the database is over a
not the slowest operations. Having the results for network or a socket, you may get some extra
each algorithm, we can easily calculate in delay for networking.
percentage the average time taken from the three  Domain Name System resolution time – this is
subprocesses. The result is shown on Figure 2. one of the most underestimated threats. After
you ship your application online, some of your
clients may notice a bigger delay than others.
This may be because of the amount of request
you do involving DNS queries.
 Database driver – have in mind that your
database driver may behave differently on
bigger amounts of data or because of your local
network quality;
 Virtualization – although it can give you
network isolation and less physical devices in
your server room, sometimes the virtualization
Fig. 2. Average time for each part of the combined drivers cause extra slowness and may behave
approach work irradically. Placing software on top software
may only slow your machine, but placing
When using both side calculations, we can hardware on top of hardware is the real deal;
conclude that the two most important parts of any  Slow hardware – your oldest enemy. This may
kind of combined solution are: break your clients experience and security. Do
not save money for hardware or system
 Choosing a number of secure algorithms that administrators and you will have less problems.
are proven to be effective and distributing them
as needed. This is done in order to have a real The above and other unknown problems
point of using multilayer encryption; may cause your system to get more than two time
 Profiling your application before and after slower than you have expected.
realization, just to see if your software is Main point is, before you do anything, test
dealing with more pressure on application or your code in a real world situation and try to
database level and double check final results simulate the worst case scenarios before shipping it
before you shipping it. to the real world. Keep the balance between security
and performance or you will slowly start losing
Be cautious when having a security your clients.
paranoia, because if things are not done right, this
may damage your overall software performance. 4.3. Professional data protection side notes
Finding equilibrium is the price for success. When taking care of sensitive data,
sometimes just hashing your passwords may not be
enough. When your information is strictly
II-195
confidential, just combine compression, multilayer 6. Conclusion
hashing, symmetrical encryption, secure This paper has tested the performance of all
connections, network isolation, firewalls, password the most used and standardized hash algorithms
policies and double side application validations. [1] from both application and database realization
Do not assume that you are safe. Include sides. It maps the best uses of each algorithm and
thread assessment and insert security measures in shows the consequences a developer may be faced
every single part of your development cycle. with by using them.
Starting from the design phase, all the way to The most interesting results from the
release and maintenance. This is the professional experiments are:
way of creating secure applications and not just
adding security features on top of by design  Application layer hashing is faster that
vulnerable software. [3] database, but are pretty close;
We are not going to drill more into this  SHA-384 and SHA-512 seem to deal faster with
topic, because the point of this paper is focused on longer input data than others;
hash functions.  Disk encryption does not affect your application
speed, but requires more computation power;
5. Best hashing use case scenarios
 Using just hashing may not be enough for the
Based on the results of this paper, we can security of your data.
say there are three main uses for hash function and
the best ways of choosing algorithms for each one REFERENCES
are summarized in the next sections.
1. Katz, J., and Lindell, Y. (2007). Introduction to
5.1. Generating identifiers Modern Cryptography: Principles and
Every developer comes to the need of Protocols. ISBN: 9781584885511.
mapping data and computing unique identifiers to 2. Menezes, A., P. van Oorschot, and Vanstone,
associate the data with. Using MD5 or SHA-1 S. (1996) Handbook of Applied Cryptography.
seems to be the smartest choice only because they ISBN: 9781439821916.
the fastest and do not have a great chance of 3. Howard, M., and LeBlanc, D. (2004) Writing
duplicated digests results. Of course using either Secure Code: Practical Strategies and Proven
SHA-224 or SHA-256 would not hurt your Techniques for Building Secure Applications in
performance much. a Networked World. ISBN: 9780735617223
4. Wang, X., and Yu, H. (2005) How to Break
5.2. Data verification MD5 and Other Hash Functions. Advances in
For verifying strings or easily comparing Cryptology – EUROCRYPT 2005. ISBN:
them, using any of the tested algorithms will do the 9783540259107
trick for small amounts of data and be fast enough. 5. Wang, X., Yin, Y.L, and Yu, H. (2005)
However, we would suggest SHA-256, because it is Finding Collisions in the Full SHA-1.
currently used for professional application and Advances in Cryptology – EUROCRYPT 2005.
digital signatures. When dealing with longer data, ISBN: 9783540259107
switch to SHA-384 instead. 6. http://www.intel.com/content/dam/doc/white-
paper/enterprise-security-aes-ni-white-
5.3. Securing data paper.pdf
When using hash functions for protecting 7. https://wiki.archlinux.org/index.php/disk_encr
sensitive data, it is recommended to use only the yption
SHA-2 algorithms. Choosing between SHA-224,
SHA-256, SHA-384 or SHA-512 are the best
choices for secure storage and data transfers. Of Contacts:
course in a real software system you would UNIVERSITY OF PLOVDIV PAISII
probably want more than one type of cryptographic HILENDARSKI
algorithm involved. 24 TZAR ASEN
PLOVDIV
E-mail: enkov@uni-plovdiv.bg
E-mail: tony.karavasilev@gmail.com
CLUSTERING TECHNIQUES FOR ANALYSIS OF

LARGE DATASETS
MILENA ANGELOVA
Abstract: Constant advance in science and technology makes collection of data and storage
much easier and very inexpensive than before. This led to the information of enormous
datasets in science, government and industry, which should be processed or sorted to get
useful information. The goal of this survey is to provide a comprehensive review of different
techniques in data mining and how they can deal with large data sets.
Keywords: traditional clustering techniques, data stream clustering, incremental clustering
1. Introduction requirements for large applications. There exist two

Constant advance in science and technology approaches which deal with these issues: data stream
makes collection of data and storage much easier and clustering and incremental clustering.
very inexpensive than ever before. This led to the Data stream clustering has recently attracted
information of enormous datasets in science, attention for emerging applications that involve large
government and industry, which should be processed amounts of streaming data. Data stream model has
and/or sorted to get useful information. some constraints. Data from the streams cannot be
For example, if we consider the results stored due to the large volume, therefore only
generated by a search engine for a particular query, summaries should be processed and stored. The
user has to sift through the long lists and find the speed of arrival is fast thus each item has to be
desired solution. But this job can be very difficult for processed in real time or only once. Data from past
the user if there are millions of web pages displayed might become irrelevant and can even affect
as solutions for a given query. Thus clustering negatively the current summaries.
techniques can be very useful in grouping the closely Incremental clustering, sometimes called
related solutions of a given query and displaying the data stream clustering, is a clustering problem, where
results in the form of clusters so that the unrelated the data to be clustered arrive continuously. In data
documents can be avoided even without taking a stream clustering, the data is expected to be finite,
glimpse at them. but for the incremental clustering we have to expect
The main idea behind clustering any set of the data stream to be infinite. Therefore, incremental
data is to find inherent structure in the data, and clustering supposes two approaches online and
interpret this structure as a set of groups, where the offline clustering of streaming data.
data objects within each cluster should show very In this work, first we present all the major
high degree of similarity known as intra-cluster traditional clustering techniques in brief and then we
similarity, while the similarity between different discuss how they can deal with a large data sets. Then
clusters should be reduced. There exist several we analyze how the data stream and incremental
clustering algorithms which can be used for large techniques can deal with issues from traditional
datasets. They can be separated in three different techniques and large data sets.
groups: partition methods, hierarchical and density
2. Traditional techniques for clustering
based clustering algorithms. They can be called
traditional clustering algorithms. Clustering can be done in many different
The main problems associated with the ways. There exists different kind of clustering
traditional clustering algorithms are handling techniques deal with data sets. Some take input
multidimensionality and scalability with rapid parameters from the user like number of clusters to
growth in size of data. The increase in size of data be formed etc., but some decide on the type and
increases the computational complexities which have amount of data given. Clustering algorithms can be
a devastating effect on the runtime and memory classified in five distinct types: partitioning,
Copyright©2017
Copyright by Technical University
by Technical University-of
Sofia,
Sofia,Plovdiv branch,Bulgaria
Plovdiv branch, Bulgaria ISSN Online: 2535-0048
II-197
hierarchical, model-based, density based methods sensitive to outliers. For example, an object with
and grid-based methods. In this section we have some extremely large value may substantially distort
described partitioning, hierarchical and density the distribution of data by k-means method.
based methods and how they can deal with large data Therefore, instead of taking the mean value of
sets. objects in a cluster as a reference point, an object that
is most centrally located in the cluster can be taken
2.1. Partitioning clustering as a representative object, called as medoid. Thus the
Let a database containing n data objects is partitioning method can be performed by minimizing
given, then a partitioning method constructs k the sum of dissimilarities between each object and
clusters of the data where k <= n and k is the input with its corresponding reference point.
parameter provided by the user. That is, it classifies The algorithm creates k partitions for n given
the data into k groups (clusters) which should satisfy objects. Initially k-medoids are selected which are
the following conditions: (1) each group must located more centrally in each cluster. Then the
contain at least one data object and (2) each data algorithm repeatedly tries to make a better choice of
object should belong to only one group. The most medoids by analyzing all the possible pairs of
well-known and commonly used partitioning objects.
methods are k-means proposed by Mac Queen [1] The k-medoids method is more robust than
and k-medoids proposed by Kaufman and k-means because it is less influenced by outliers or
Rousseeuw [2]. other extreme values than mean. But its processing is
very costly than k-means method and it also has the
2.1.1. The k-means method drawback of user providing the input parameter k.
The k-means algorithm takes input k from The main advantage of k-means and k-
the user and partitions n data objects into k clusters medoids is that their complexity is linear. Therefore,
so that the resulting intra-cluster similarity is very their execution time is proportional to the number n
high and inter-cluster similarity is very low. The of data objects, so that they can be used with large
cluster similarity is calculated based on the mean volumes of data. The second advantage is that it is
value of the objects in the cluster. First, it randomly possible to detect outliers, which appear in the form
picks k data objects as the mean or centroid points. of singleton clusters. There is also a third advantage.
For each of the remaining objects, an object assigned Unlike hierarchical methods, in which the clusters
to the centroid to which it is most similar based on are not altered once they have been constructed, the
distance between the object and the cluster. This reassignment algorithms constantly improve the
process iterates till good clusters are formed. quality of the clusters, which can thus reach a high
Typically, squared root function is used for this level when the form of the data is suitable.
which can be defined as The first disadvantage is that the final
partition depends greatly on the more or less arbitrary
𝒌𝒌
initial choice of the centers ci. Consequently, we do
𝑬𝑬 = ∑ ∑ |𝒙𝒙 − 𝒎𝒎𝒊𝒊 |𝟐𝟐 , (1) not have a global optimum, but simply the best
𝒊𝒊=𝟏𝟏 𝒙𝒙 ∈ 𝑪𝑪𝒊𝒊
possible partition based on the starting partition. The
where x is the point in space representing the given second drawback of partitioning methods is that the
object, and mi is the mean of cluster Ci. This function number of clusters, k, is fixed in these methods, and
tries to make the clusters as separate as possible. it is not less than k unless certain clusters are empty.
If this number does not correspond to the actual
The method is relatively scalable and configuration of the cloud of data objects, the quality
efficient in handling large data sets because the of the clustering may be adversely affected. The third
computational complexity of the method is O(nkt), drawback of these methods is that they are only good
where n is the total number of iterations, k is the at detecting spherical forms. Even convex forms such
number of clusters and t is the number of iterations. as ellipses cannot be detected well if they are not
Normally k<<n and t<<n so, the method often ends sufficiently separated.
up at local optimum. However, this method has some
drawbacks: 1) it can be applied only when mean of a 2.2. Hierarchical clustering algorithms
cluster is defined, but when data with categorical Hierarchical clustering, also known as
attributes is involved it cannot be the case; 2) the user connectivity based clustering, is based on the core
should specify the number of clusters k in advance idea of objects being more related to nearby objects
and 3) it is sensible to noise and outlier data points. that to objects farther away. These algorithms
2.1.2. The k-medoids method connect “objects” to form “clusters” based on their
K-medoids algorithm was developed to distance. A cluster can be described largely by the
overcome the drawbacks of k-means which is very maximum distance needed to connect parts of the
II-198
cluster. Strategies for hierarchical clustering clusters. If n is the number of data objects (elements)
generally fall into two types: to be clustered, the complexity of the basic algorithm
is of the order of n3, and it will soon exceed the
 Agglomerative clustering is a “bottom-up” capacity of even a powerful computer. The
approach: each observation starts in its own difference from the moving centers method is that we
cluster, and pairs of clusters are merged as one are comparing individuals with each other instead of
moves up the hierarchy. comparing them with the centers of the clusters only.
 Divisible clustering is a “top-down” approach: The situation can be improved to a certain extent by
all observations start in one cluster, and splits are the nearest-neighbor algorithm, which reduces a
performed recursively as one moves down the complexity of n3 to n2, by a judicious combination of
hierarchy. more than two observations on each iteration. It can
be used with the Ward, average linkage, single
linkage and complete linkage methods. A second
drawback is that, at each step, the partitioning
criterion is not global, but depends on the clusters
obtained so far: two individuals placed in different
clusters will no longer be compared. In other words,
this type of clustering into n clusters in not
necessarily the best possible outcome, but only the
best of those obtained by combining the clusters of a
clustering procedure into n + 1 clusters. Thus some
natural clusters may be hidden by an earlier
Fig. 1. Examples of agglomerative “Bottom-up” branching.
and divisive “Top-down” clustering
2.3. Density based clustering algorithms
In general, the merges and splits are Density-based clustering [7] methods are
determined in a greedy manner. The results of based on a local cluster criterion. Clusters are
hierarchical clustering are usually presented in a assumed as regions in the data space in which the
dendrogram. The complexity of agglomerative objects are dense and the clusters are separated by
clustering is O(n2log(n)) [3], which makes them too regions of low object density. These regions have an
slow for large data sets. Divisible clustering with an arbitrary shape and the data points inside a cluster
exhaustive search is O(2n), which is even worse. may be arbitrarily distributed.
However, for some special cases, optimal efficient
The most popular [8] density based
agglomerative methods (of complexity O(n2)) are
clustering method is DBSCAN [9]. In contrast to
known: SLINK [4] for single-linkage and CLINK [5]
many newer methods, it features a well-defined
for complete-linkage clustering.
cluster model called “density-reachability”. Similar
The agglomerative clustering does not suffer
to linkage based clustering, it is based on connecting
from the two major drawbacks of the moving centers
points within certain distance thresholds. However,
method, namely its dependence on the choice of
it only connects points that satisfy a density criterion,
initial centers and the fixed number of clusters
in the original variant defined as a minimum number
chosen in advance. The second advantage is that it
of other objects within this radius. A cluster consists
can detect clusters of different shapes, according to
of all density-connected objects (which can form a
the distance chosen. The best possible distance can
cluster of an arbitrary shape, in contrast to many
be chosen according to previous knowledge of the
other methods) plus all objects that are within these
shape of the clusters to be detected, but the choice is
objects’ range. Another interesting property of
mainly made by carrying out a number of trials and
observing the results. It is possible to use a given
distance because the resulting clusters are more
easily described or because the number of clusters
appears more natural with respect to the graphic
indicators. The third advantage is that it enables us to
cluster data elements, variables or centers of clusters Fig. 2. Examples of (a) directly density-reachable,
obtained by using a moving centers algorithm. (b) density-reachable, (c) density-connected in
The main disadvantage is its algorithmic density-based clustering.
complexity, which is non-linear: in order to move
from k + 1 clusters to k clusters, we must calculate (k DBSCAN is that its complexity is fairly low – it
+ 1)k/2 distances and combine the two closest requires a linear number of range queries on database

II-199
– and that it will discover essentially the same results The main disadvantage of data streaming
in each run, therefore there is no need to run it algorithms is to fit a large data into main memory. It
multiple times. OPTICS [10] is a generalization of is not possible for the algorithm to “remember” too
DBSCAN that removes the need to choose an much of the data scanned in the past. This scarcity of
appropriate value for the range parameter ε, and space necessitates the design of a novel kind of
produces a hierarchical result related to that of algorithm that stores only summary of past data,
linkage clustering. DeLi-Clu, [11] Density-Link- leaving enough memory for the processing of future
Clustering combines ideas from single-linkage data [16, 17]. Researchers in [31] are developed a
clustering and OPTICS, eliminating the ε parameter clustering method which is called CluStream. In the
entirely and offering performance improvements literature, CluStream is defined as algorithm which
over OPTICS by using an R-tree index. summarizes streams into micro-clusters to deal with
The key drawback of DBSCAN and memory limitations. The CluStream model has a
OPTICS is that they expect some kind of density wide functionality in characterizing data stream
drop to detect cluster borders. Moreover, they cannot clusters over different time horizons in an evolving
detect intrinsic cluster structures which are prevalent environment. They use a pyramidal time window
in the majority of real life data. A variation of assures that the essential statistics of evolving data
DBSCAN, EnDBSCAN, [12] efficiently detects streams can be captured without sacrificing the
such kinds of structures. On data sets with, for underlying space and time efficiency of the stream
examples, overlapping Gaussian distributions – a clustering process. [31] As advantages, CluStream is
common use case in artificial data – the cluster very flexible for real time transactions and pyramidal
borders produced by these algorithms will often look time window guarantees efficient time and space.
arbitrary, because the cluster density decreases Another approach for these constraints have
continuously. On a data set consisting of mixtures of a direct impact in memory usage and data “age”
Gaussians, these algorithms are nearly always relevance. Some techniques have been developed to
outperformed by methods such as EM clustering that deal with these problems such as sliding window
are able to precisely model this kind of data. combined with low memory functions. Sliding
window is a technique where a window of size W
3. Data Stream and Incremental clustering
keeps the last W items that arrived and run a desired
algorithms
algorithm on these items. The summary is then kept
Traditional clustering techniques are not and the items are then discarded. In this way an up to
suitable for large data sets. Many applications have data statistics is always computed. These statistics
massive amount of data which increases every day. are usually kept by updating the model; in case of
This causes limitation in data storage capacity and classification is could be the decision tree and in
processing time. In this section, we have described clustering could be the final clustering centroids.
two techniques data stream and incremental Advantage of data stream algorithms instead
clustering which can be used to deal with a big of k-partitioning clustering methods is that we do not
amount of data. need to process the data multiple times to receive
final clustering decision. Different techniques were
3.1. Data stream clustering
created to deal with evolving data such as one pass
Data stream clustering algorithms have been processing and summarization. In literature,
developed as an adaption of traditional clustering researches in [16] give a constant-factor
algorithms to fit the streaming model and comply to approximation algorithms for the k-means problem
its constraints. The data stream problem has been in the data stream model of computation in a single
researched in recent years because of the large pass.
number of relevant applications [13, 14, 15, 16, 17]. As a conclusion of this section, clustering
Furthermore, data stream clustering has recently data streams is gaining a lot of importance and a lot
attracted attention for emerging applications that of research is done on improving these algorithms to
involve large amounts of streaming data. In the apply them on real time applications.
literature, one of the first results for data streaming
appeared in 1980 [18] but the model was formalized 3.2. Incremental clustering
in 1998 [19]. Incremental clustering is based on the
A formal definition for data stream S is that assumption that it is possible to consider instances
of a sequence of arriving data objects x1, x2, …, xn, one at a time and assign them to existing clusters.
rendering S = {xi}, where i = 1, 2, …, m and m → ∞. Here, a new instance is assigned to a cluster without
Each data object is an m-dimensional attribute vector significantly affecting the existing clusters. Only the
j cluster representations are stored in the main
xi = [xi ] where j = 1, 2, …, n, belonging to an attribute
space that can be continuous, categorical, or mixed. memory to alleviate the space limitations.
II-200
One of the most famous example is the The algorithm gets its name from the fact
COWEB[20]. It keeps a classification tree as its that cluster coordinates are set equal to the input used
hierarchical clustering model. Then the algorithms to create the cluster. This input is known as the
place new points in a top-down fashion using a leader. Once, the cluster is established new members
category utility function. can be assigned to it, but its coordinates are never
Another incremental hierarchical clustering adjusted from those of the leader. The decision to
algorithm that works in a bottom-up fashion is assign an input to an existing cluster instead of
described in [21]. creating a new cluster is made based on distance. If
SLINK [22] is the most time-wise effective the current input’s distance is within a threshold from
implementation of single linkage hierarchical an existing cluster then it is assigned to that cluster,
clustering. It works incrementally, building several otherwise a new cluster is created. It should be noted
linear indexes. that the algorithm assigns inputs to the first cluster
BIRCH [23] also uses a hierarchical that satisfies the threshold requirement rather than
clustering, but the hierarchy is built on so called searching for the closest cluster.
clustering features. A clustering feature statistically The leader algorithm requires only one pass
summarizes the underlying data. through the sample data so it has speed advantage
DBSCAN [24] searches for its nearest over non-incremental techniques. A second
neighbors when placing a new point. If there are advantage is that it does not require prior knowledge
sufficiently enough points under a minimal distance of the number of expected clusters. A drawback,
of the new point, such point is then added into the however, is that the clustering is highly dependent on
respective cluster of the nearest nodes. A the other of the input samples. This dependence
generalization of DBSCAN called OPTICS [25] exists because samples are assigned to the first
work with a varying density of clusters. cluster that meets the threshold requirement rather
CURE [26] uses yet another approach to than finding the nearest cluster.
clustering. It lies between BIRCH and SLINK, as it It exists many different incremental
uses hierarchical clustering, but instead of clustering algorithms. Each of them attempts to
representing the cluster with once center as BIRCH, improve clustering accuracy by adjusting the
or considering all points as SLINK, it chooses only algorithm to the specific characteristics of the desired
several representatives of the cluster that are then application. Though the overall flow of most
moved closed to the center. incremental clustering algorithms is similar, they
Algorithms such as DBSCAN, OPTICS and differ in their initialization of their conditions for
DENCLUE, STING, CLIQUE, Wave-Cluster and creation of deletion of clusters.
OPTIGRID do not optimize the k-means objective.
An overview article of incremental data stream 4. Conclusion
algorithms [27, 28, 29]. This paper has presented a short review in most
Order-independences is an important of all clustering techniques which can be deal with
property of clustering algorithms. An algorithm is large data sets which requires a large memory, arrive
order-independent if it which the data is presented, continuously, and might change over time. The
otherwise, it is order-dependent. Most of the tradional clustering methods do not scale very large
incremental algorithms presented above are order- data sets since they either need several passes over
dependent. the data or they create data structures that do not
The major advantage for incremental scale lineraly with the number of objects.
algorithm is that, it can operate without having to Incremental and data stream clustering methods are
revisit old data it is more applicable to situations with gaining a lot of research which is done on improving
very large data sets that cannot fit in memory, and these algorithms to apply them on real time
streaming data sets which are being produced for applications.
long periods of time. Additionally, these algorithms In the future work, we are planning to research
are better suited to data whose characteristics might in this area, to seek new clustering algorithms which
be evolving over time. can deal with a large amount of data.
One of the first incremental clustering
algorithms was the leader algorithm describes by
Hartigan [30]. This algorithm attempts to partition a
set of data samples into a number of disjoint clusters. REFERENCES
The desired number of clusters is not specified but a
distance metric and maximum threshold must be 1. MacQueen, J. B. (1967). Some Methods for
defined. classification and Analysis of Multivariate
Observations. Proceedings of 5th Berkeley

II-201
Symposium on Mathematical Statistics and 19. Henzinger, M., Raghvan, P., and Rajagopalan, S.
Probability. volume (1), 281–297. (1998). Computing on Data Streams. Digital
2. Kaufman, L., and Roussew, P. J. (1990). Finding Equipment Corporation.
Groups in Data - An Introduction to Cluster 20. Douglas H Fisher. (1987). Knowledge
Analysis. A Wiley-Science Publication. acquisition via incremental conceptual
3. Rokach, L., and Maimon, O. (2005). Clustering clustering. Machine learning, 139-172.
methods. Data mining and knowledge discovery 21. Dwi H Widyantoro, Thomas R Ioerger, and John
handbook, 321-352, Springer US. Yen. (2002). An incremental approach to
4. Sibson R. (1973). SLINK: an optimally efficient building a cluster hierarchy. In Data Mining
algorithm for the single-link cluster method. The ICDM., 705–708.
Computer Journal. British Computer Society. 16 22. Robin Sibson. (1973). Slink: an optimally
(1), 30–34. efficient algorithm for the single-link cluster
5. Defays D. (1977). An efficient algorithm for a method. The Computer Journal, 30–34.
complete-link method. The Computer Journal. 23. Tian Zhang, Raghu Ramakrishnan, and Miron
British Computer Society. 20 (4), 364–366. Livny. (1996). Birch: an efficient data clustering
6. Steinbach, M., Karypis, G., and Kumar, V. method for very large databases. In ACM
(2000). A comparison of document clustering SIGMOD Record, volume 25, 103–114.
techniques, 525-526. In KDD workshop. 24. Martin Ester, Kriegel Hans-Peter, Jörg Sander,
7. Kriegel, H., Kröger, P, Sander, J. and Zimek, A. and Xiaowei Xu. (1996). A density-based
(2011). Density-based Clustering. WIREs Data algorithm for discovering clusters in large spatial
Mining and Knowledge Discovery. 1 (3), 231– databases with noise. In KDD, volume 96, 226–
240. 231.
8. Microsoft academic search: most cited data 25. Mihael Ankerst, Markus M Breunig, Hans-Peter
mining articles: DBSCAN is on rank 24. Kriegel, and Jörg Sander. (1999). Optics:
9. Ester, M., Kriegel, H., Sander, J., and Xu, X. Ordering points to identify the clustering
(1996). A density-based algorithm for structure. In ACM SIGMOD Record, volume 28,
discovering clusters in large spatial databases 49–60.
with noise, 226-231. 26. Sudipto Guha, Rajeev Rastogi, and Kyuseok
10. Ankerst, M., Breunig, M.M., Kriegel and H., Shim. (1998). Cure: An efficient clustering
Sander, J. (1999). OPTICS: Ordering Points To algorithm for large databases. SIGMOD Rec.,
Identify the Clustering Structure, 49-60. volume, 27(2), 73–84.
11. Achtert, E., Böhm, C., and Kröger, P. (2006). 27. Mohamed Medhat Gaber, Arkady Zaslavsky,
DeLi-Clu: Boosting Robustness, Completeness, and Shonali Krishnaswamy. (2005). Mining data
Usability, and Efficiency of Hierarchical streams: A review. SIGMOD Rec., volume
Clustering by a Closest Pair Ranking, 119-128. 34(2), 18–26.
12. Roy, S. and Bhattacharyya, D. K. (2005). An 28. Sudipto Guha, Nina Mishra, Rajeev Motwani,
Approach to find Embedded Clusters Using and Liadan O’Callaghan. (2000). Clustering data
Density Based Techniques. Springer Verlag. streams. In Foundations of computer science,
volume (3816), 523–535. 359–366.
13. Aggarwal, C. C. (2003) A Framework for 29. Moses Charikar, Chandra Chekuri, Tomás
Diagnosing Changes in Evolving Data Streams. Feder, and Rajeev Motwani. (1997). Incremental
ACM SIGMOD Conference. clustering and dynamic information retrieval.
14. Babcock, B. and Other, (2002). Models and ACM symposium on Theory of computing, 626–
Issues in Data Stream Systems, ACM PODS 635.
Conference. 30. Hartigan J. A. (1975) Clustering Algorithms.
15. Domingos, P. and Hulten, G. (2000). Mining New York: John Wiley and Sons.
High-Speed Data Streams. ACM SIGKDD 31. Aggarwal C. and Other, (2003). A framework for
Conference. Clustering Evolving Data Streams. Proceedings
16. Guha, S., Mishra, N., Motwani, R., O’Callaghan of the 29th VLDB Conference.
L. (2000). Clustering Data Streams. IEEE FOGS
Conference.
17. O’Callaghan, L. and Other, (2002). Streaming- Authors‘ contacts:
Data Algorithms for High-Quality Clustering. E-mail: mangelova@tu-plovdiv.bg
ICDE Conference. University: TU – Sofia, Plovdiv branch
18. Munro, J., and Paterson, M. (1980). Selection Department: Computer Systems and
and Sorting with Limited Storage. Theoretical Technologies
Computer Science, 315-323.
ALGORITHMS IN DEEP LEARNING

T. HRISTEVA, M.MARINOVA
Abstract: Deep Learning is a software tool for learning in neural networks that
utilizes multiple layers of abstraction. Today it has impact on many research areas.
There are two broad types of machine learning – supervised learning and
unsupervised learning. The supervised learning uses labeled datasets to produce
the desired results. The supervised learning is useful for the classification and the
regression. In the unsupervised learning the main purpose is to find out hidden
models of data. The unsupervised learning is used for clustering and associative
tasks.
Key words: deep learning, algorithm, neural network
INTRODUCTION
Credit: https://cuteprogramming.wordpress.com
Supervised Learning
In supervised learning we have input
variables (x) and an output variable (Y) and you use
an algorithm to learn the mapping function from the
Credit: http://blog.algorithmia.com
input to the output(Y = f(X))
The goal is to approximate the mapping
Deep learning is impacting everything from function so well that when you have new input data
healthcare to transportation to manufacturing, and (x) that you can predict the output variables (Y) for
more. Companies are turning to deep learning to that data.
solve hard problems, like speech recognition, object
recognition, and machine translation. It might seem It is called supervised learning because the
like this technology is still years away, but we are process of an algorithm learning from the training
beginning to see its commercial use. Such is the dataset can be thought of as a teacher supervising
case with self-driving cars. Companies like Google, the learning process. We know the correct answers,
Tesla, and Uber are already testing autonomous cars the algorithm iteratively makes predictions on the
on the streets. training data and is corrected by the teacher.
Learning stops when the algorithm achieves an
acceptable level of performance. Supervised
learning problems can be further grouped into
II-203
regression and classification problems. correct decision, most of the time. As a method, it
Classification: A classification problem is when the allows you to approach the problem in a structured
output variable is a category, such as “red” or and systematic way to arrive at a logical conclusion.
“blue” or “animal” and “human”
Decision trees have many advantages.They
Regression: A regression problem is when are simple to understand and interpret. People are
the output variable is a real value, such as “dollars” able to understand decision tree models after a brief
or “weight”. Some common types of problems built explanation, allow the addition of new possible
on top of classification and regression include scenarios easily.Help determine worst, best and
recommendation and time series prediction expected values for different scenarios, can be
respectively. combined with other decision techniques.As a
disadvantages can be given that the calculations can
A popular example of supervised machine get very complex if many values are uncertain or if
learning algorithms is support vector machines for many outcomes are linked.
classification problems
Unsupervised Machine Learning Logistic Regression

Unsupervised learning is where you only Logistic regression is a powerful statistical
have input data (X) and no corresponding output way of modeling a binomial outcome with one or
variables.The goal for unsupervised learning is to more explanatory variables. It measures the
model the underlying structure or distribution in the relationship between the categorical dependent
data in order to learn more about the data. These are variable and one or more independent variables by
called unsupervised learning because unlike estimating probabilities using a logistic function,
supervised learning above there is no correct which is the cumulative logistic distribution.
answers and there is no teacher. Algorithms are left
to their own devises to discover and present the Regressions can be used in applications
interesting structure in the data. such as:
Unsupervised learning problems can be  Credit Scoring

further grouped into clustering and association  Measuring the success rates of marketing
problems. campaigns
 Predicting the revenues of a certain product
Clustering: A clustering problem is where
you want to discover the inherent groupings in the Support Vector Machines
data, such as grouping customers by purchasing SVM is binary classification algorithm.
behavior. Given a set of points of 2 types in N dimensional
place, SVM generates a (N — 1) dimensional
Association: An association rule learning
hyperlane to separate those points into 2 groups.
problem is where you want to discover rules that
Say you have some points of 2 types in a paper
describe large portions of your data, such as people
which are linearly separable. SVM will find a
that buy X also tend to buy Y.
straight line which separates those points into 2
An examples of unsupervised learning types and situated as far as possible from all those
algorithms are: k-means for clustering problems. points.
In terms of scale, some of the biggest
problems that have been solved using SVMs (with
MOST IMPORTANT ALGORITHMS suitably modified implementations) are display
advertising, human splice site recognition, image-
Supervised Learning based gender detection, large-scale image
Decision Trees classification, etc.
A decision tree is a decision support tool
that uses a tree-like graph or model of decisions and
their possible consequences, including chance-event Unsupervised Learning
outcomes, resource costs, and utility.A decision tree Clustering Algorithms
is the minimum number of yes/no questions that one
has to ask, to assess the probability of making a
II-204
usually considered to be noise and border points.
The most popular density based clustering method
is DBSCAN. In contrast to many newer methods, it
features a well-defined cluster model called
"density-reachability". Similar to linkage based
clustering, it is based on connecting points within
certain distance thresholds.
 Neural networks.
Neural networks are inspired by the

structure of the cerebral cortex. At the basic level is
the perceptron, the mathematical representation of a
biological neuron. Like in the cerebral cortex, there
can be several layers of interconnected perceptrons.
The first layer is the input layer. Each node in this
layer takes an input, and then passes its output as
the input to each node in the next layer. There are
generally no connections between nodes in the same
layer and the last layer produces the outputs.
Credits: http://www.machinelearningtutorial.net/ The middle part is called the hidden layer.
These neurons have no connection to the outside
Clustering is the task of grouping data (e.g. input or output) and are only activated by
objects in such a way that objects in the same group nodes in the previous layer.
(cluster) are more similar to each other than to the
object in other groups. Neural networks have the ability to derive
meaning from complicated or imprecise data and
can be used to extract patterns and detect trends that
are too complex to be noticed by other computer
Every clustering algorithm is different, and here are techniques. A trained neural network can be thought
a couple of them: of as an "expert" in the category of information it
 Centroid-based algorithms has been given to analyse.
In centroid-based clustering, clusters are
represented by a central vector, which may not
necessarily be a member of the data set. When the
number of clusters is fixed to k, k-means clustering
gives a formal definition as an optimization
problem: find the {\displaystyle k} k cluster centers
and assign the objects to the nearest cluster center,
such that the squared distances from the cluster are
minimized.
 Connectivity-based algorithms
Connectivity based clustering, also known
as hierarchical clustering, is based on the core idea
of objects being more related to nearby objects than
to objects farther away. These algorithms connect
"objects" to form "clusters" based on their distance.
A cluster can be described largely by the maximum Credit: Michael A. Nielsen, “Neural Networks and
distance needed to connect parts of the cluster. Deep Learning”
 Density-based algorithms
In density-based clustering, clusters are Deep learning can be described
defined as areas of higher density than the as the technique for learning in neural networks that
remainder of the data set. Objects in these sparse utilizes multiple layers of abstraction to solve
areas - that are required to separate clusters - are pattern recognition problems.It is a specific
II-205
approach used for building and training neural address the data analysis and learning problems
networks, which are considered highly promising found in massive volumes of input data. More
decision-making nodes. An algorithm is considered specifically, it aids in automatically extracting
to be deep if the input data is passed through a complex data representations from large volumes of
series of nonlinearities or nonlinear transformations unsupervised data. This makes it a valuable tool for
before it becomes output. In contrast, most modern Big Data Analytics, which involves data analysis
machine learning algorithms are considered from very large collections of raw data.
"shallow" because the input can only go only a few
levels of subroutine calling. BIBLIOGRAPHY
Deep learning removes the manual 1. https://www.udacity.com/course/machine-

identification of features in data and, instead, relies learning--ud262
on whatever training process it has in order to 2. http://machinelearningmastery.com/supervi
discover the useful patterns in the input examples. sed-and-unsupervised-machine-learning-
This makes training the neural network easier and algorithms/
faster, and it can yield a better result that advances 3. http://blog.algorithmia.com/introduction-to-
the field of artificial intelligence. deep-learning-2016/
4. http://www.kdnuggets.com/2016/08/10-
algorithms-machine-learning-
CONCLUSION engineers.html
5. http://www.iro.umontreal.ca/~pift6266/H10
/notes/mlp.html
The digital era has brought un important 6. http://neuralnetworksanddeeplearning.com
amount of data in all forms and from every region 7. http://deeplearning.net
of the world. This data, known simply as Big Data, 8. Ian Goodfellow, Yoshua Bengio, Aaron
is gotten from sources like social media, internet Courville-Deep Learning,MIT Press
search engines, e-commerce platforms, online 9. Deep Learning Tutorial,LISA lab,
cinemas, etc. This enormous amount of data is University of Montreal
readily accessible and can be shared through fintech 10. Deng,Dong Yu, Deep Learning: Methods
applications like cloud computing. However, the and Applications
data, which normally is unstructured, is so vast that
it could take decades for humans to comprehend it Teodora Hristeva
and extract relevant information. Companies realize Technical University - Sofia, Branch
the incredible potential that this information has and Plovdiv
are increasingly adapting to systems for automated Telephone:+359 899 862 656
support. This is where deep learning fits perfectly.In Email: thisteva@gmail.com
contrast to more conventional machine learning and E-mail: thristeva@gmail.com
feature engineering algorithms, deep learning has an
advantage of potentially providing a solution to
БЕЙСОВ МОДЕЛ ЗА ПРЕДВИЖДАНЕ НА

ИЗХОДА ОТ ФУТБОЛНИ МАЧОВЕ
ЙОРДАН КЪРПАРОВ, ДИЛЯНА БУДАКОВА, ЛЮДМИЛ ДАКОВСКИ
Резюме: В този доклад се разглежда Бейсова мрежа за предсказване на

резултатите от футболни мачове. От направените експерименти се вижда, че с
помощта на предложения модел се постигат прогнози с повече от 50% точност.
Програмната система е реализирана в средата на Visual Studio.NET и с помощта на
езикът за програмиране C#, а базата данни е реализирана с помощта на MS SQL.
Ключови думи: Бейсови мрежи, футбол, причинно-следствени връзки директен
ацикличен граф.
A BAYESIAN NETWORK MODEL FOR

FOOTBALL GAMES OUTCOMES PREDICTION
JORDAN KYRPAROV, DILYANA BUDAKOVA, LYUDMIL DAKOVSKI
Abstract: The present paper describes a Bayesian network model for prediction of football
games outcomes. It can be seen from the experiments made, that predictions with more than
50 % accuracy are achieved using the proposed model. The programming system is
realized in Visual Studio.NET environment and the programming language C#. The
database is realized via MS SQL.
Key words: Bayesian networks, football games, causal connections, directed acyclic
graphs
1. Увод математическото очакване за броя на головете,

В тази статия се изследва Бейсов модел които отборът се очаква да реализира и средния
за предсказване на изхода от футболни мачове. брой успешни голове реализирани от отбора в
Използвана е база данни със статистики и миналото, броя допуснати голове от
наблюдения за мачовете от Висшата лига и противниковия отбор, предимството на
испанската ЛаЛига за сезони 2009-2010, 2010- домакинството. Изчисляват се вероятностите
2011, 2011-2012, 2012-2013, 2013-2014, 2014- двата отбора да реализират по 0, 1, 2 или повече
2015, 2015-2016 [4]. В съществуващите голове и се построява разпределението на
вероятностни модели за предсказване на изхода Поасон. Изчисляват се коефициентите за победа,
от футболни срещи се вземат предвид загуба или равен мач [3]. В [5,6] се доразвива
постиженията на всички играчи, психическото методът [2] като се вземат предвид повече данни
им състояние, средната възраст на играчите, кои и се изследват изчислените коефициенти и
са пострадалите играчи, резултатите от техните отклонения. Някои от характеристиките
последните пет игри на разглежданите отбори на вече проведените футболни срещи, които се
[1,2]. Изследва се влиянието на временната използват в предложения вероятностен модел
спортна форма на отборите върху текущото им са: брой отбелязани голове, резултатите на
представяне като се вземат резултатите от срещите, имена на отборите, брой отправени
последните 5-6 мача. Въвежда се рейтинг на точни удари към вратата на противника, текущо
отборите според тази тяхна моментна спортна класиране на отборите. За да се верифицира,
форма. Доказва се силната корелация между работата на реализирания Бейсов модел се
II-207
сравнява с работата на модел базиран на естествен език, компютърното зрение, търсенето
напълно случаен принцип. За програмната на евристики. В един Бейсов модел възлите
реализация е използвана средата Visual Studio показват състоянието на разглеждания проблем,
.NET Framwork, езикът C#, Windows Form а стрелките се разглеждат като причинно-
Application и е разработена релационна SQL следствени връзки и не са абсолютни. Моделът
Data Base (фиг.3). се използва за предсказване какво ще се случи
или за да се намерят причините, които са довели
до наблюдаван ефект.
Възлите в предложеният тук модел са
(фиг.2): 1. Спортна форма на отбора домакин,
определена на базата на последните 6 мача и
наблюденията колко гола е реализирал и колко
гола е допуснал неговият бъдещ противник в
своята врата; 2. Спортна форма на отбора гост –
определена по аналогичен начин; 3. Класиране
на домакина за текущото първенство; 4.
Класиране на госта за текущото първенство; 5.
Фиг. 1. Данни със статистики и Брой точни удари във вратата отправени от
наблюдения за мачовете от Висшата лига и единия отбор. 6. Аналогично - брой точни удари
испанската ЛаЛига за сезони 2009-2010, 2010- от във вратата отправени от противниковия
2011, 2011-2012, 2012-2013, 2013-2014, 2014- отбор. 7. Вероятност за победа на домакина,
2015, 2015-2016 [4]. вероятност за победа за госта и вероятност за
равен резултат.
2. Бейсов модел за предвиждане на Причинно-следствените връзки в модела
изхода от футболни мачове може да се разтълкуват по следния начин:
Бейсовите мрежи [9] са метод за извод, Очаква се, че отборът, който е по-напред в
който ползва вероятности. Известни имена класирането е в по-добра спортна форма и
които също се използват в зависимост от статуса следователно ще отправи повече удари във
на мрежата са Belief networks, Knowledge maps, вратата на противника. А колкото повече удари
Probabilistic causal networks, и т.н. Този метод за отправи към вратата на противника даден отбор,
извод е изключително популярен в общността толкова по вероятно е той да реализира повече
на учените в областта на AI probability and голове и толкова по вероятно е да спечели
uncertainty community. Бейсовите мрежи играта. Това, обаче не е сто процента сигурно,
намират приложение за решаване на проблеми в защото противниковият отбор може да е също в
медицинската диагностика, разбирането на отлична спортна форма.
Фиг. 2. Бейсов модел за предвиждане на изхода от футболни мачове.

II-208
или просто да има повече късмет в даден мач, на противника и ще се увеличи вероятността за
или защото футболът е силно конкурентен неговата победа.
спорт, противниковият отбор може да се Ако отборът гост е в отлична спортна
противопостави активно. Когато играят два форма, класиран е на първите места, а отборът
отбора, може да се предположи, че отборът, домакин е по назад в класирането и не е в
който е по-добре класиран е по вероятно да толкова добра спортна форма, то може да се
спечели, но това също не е сигурно. очаква, че отборът гост ще има по голяма
Предложеният модел може да се вероятност за победа. Ако даден отбор е
използва както за предсказване така и за победил в определена среща, то той вероятно се
намирането на причините, които са довели до е намирал в добра спортна форма и е отправил
очакваната победа, правилния залог или много повече удари във вратата на противника.
допуснатите грешки в предположенията на
потребителите. Например (фиг. 2) какво ще се 3. Програмна реализация на Бейсовата
случи ако отборът домакин е в отлична спортна мрежа и експериментални резултати
форма, класиран е на първо място, а За реализацията на Бейсовия модел
противниковият отбор е по-назад в класирането (фиг.3) е разработена функция Update, която
и не е в толкова добра спортна форма. обновява данните в таблиците на модела,
изпълнява SQL заявките за да се
определи информацията за
прогнозирането; функция Calculate,
която изчислява вероятностите за
победа на даден отбор на базата на
статистическите оценки от данните;
функция Simulate_Prediction, която
изчислява прогнозата според
изчислените вероятностни
стойности и проверява в каква
степен прогнозата съответства на
реалните постигнати резултати. За
сравняване и оценяване на работата
на степента на успешност на модела
е реализирана функцията
Фиг. 3. Програмна реализация на Бейсов Simulate_Random, която изчислява
модел за предвиждане на изхода от футболни прогнозата за резултатите от футболните срещи
мачове. на случаен принцип. Направени са два
експеримента, целта на които е да се сравнят
Очаква се, че в този случай отборът прогнозите направени на базата на модела на
домакин ще отправи повече удари към вратата реализираната Бейсова мрежа
Фиг. 4. Бейсов модел за предвиждане на изхода от футболни мачове.

II-209
и прогнозите направени чрез напълно случаен 2. MAHER M. J. (1982). Modelling association

избор. Резултатите (фиг. 4) от многократното football scores, 109-118, Statistica
провеждане на експериментите показват Neerlandica 36 (1982), nr. 3.,
следното: С помощта на предложения Бейсов http://www.90minut.pl/misc/maher.pdf
модел се постигат 50% точни прогнози; С 3. Joe Buchdahl, (2003) Fixed Odds Sports
помощта на случайния избор се постигат Betting: The Essential Guide, High Stakes
максимум 33% точни прогнози. Изводът е, че Ltd., 2003.
предложеният модел значително подобрява http://www.football-data.co.uk/ratings.pdf
вероятността да се направят точни прогнози. 4. Футболни прогнози (2016)
Ако се вземат предвид още повече факти за http://www.football-data.co.uk/
отборите, които се състезават този процент на 5. Langseth H., (2013). ”Beating the bookie A
точно направените прогнози ще се повиши. look at statistical models for prediction of
football matches”, Norwegian University of
4. Заключение Science and Technology SCAI 2013,
Предложен е Бейсов модел за http://www.idi.ntnu.no/~helgel/slides/Langseth
предвиждане на изхода от футболни мачове. .pdf
Използвани са статистически данни от мачовете 6. Mirza J., Fejes N., Sumpter D., (2016).
от Висшата лига и испанската ЛаЛига за сезони ”Statistical Football Modeling A Study of
2009-2010, 2010-2011, 2011-2012, 2012-2013, Football Betting and Implementation of
2013-2014, 2014-2015, 2015-2016. Statistical Algorithms in Premier League”,
За изчисляване на прогнозите Бейсовият Project in Computational Science: Report,
модел взема предвид спортна форма на отбора UPPSALA University 2016.
домакин и на отбора гост; класирането на http://www.it.uu.se/edu/course/homepage/proje
отбора домакин и отбора гост в текущо ktTDB/ht15/project16/Project16_Report.pdf
разглежданото първенство; брой точни удари 7. http://www.smartodds.co.uk/About/Overview,
отправени към вратата съответно на отбора гост Statistical research and sports modelling
и на отбора домакин. Изчислява се вероятността services.
за победа на отбора гост, отбора домакин и 8. https://bg.wikipedia.org/wiki/Хазартна_завис
вероятността за равен резултат. От направените имост
експерименти може да се направи извода, че 9. Charniak E.(1991), Bayesian Networks without
предложеният Бейсов модел увеличава tears, AI MAGAZINE, 0738-4602/91/$4.00
вероятността да се направят точни прогнози в ©1991 AAAI,
сравнение със случайно направените. Процентът http://www.cs.ubc.ca/~murphyk/Bayes/Charni
на правилните прогнози направени с помощта ak_91.pdf
на разгледания Бейсовия модел достига 50%,
докато при случайно направените прогнози
процентът на успешно залагане достига най- Контакти:
много до 33%. Моделът може да се подобри Име: маг. инж. Йордан Кърпаров
като се вземат предвид повече обстоятелства Организация: ТУ-София, Филиал
при прогнозирането. Моделът може да се Пловдив
използва от инвестиционни компании и за E-mail: crazy_jordan@abv.bg
профилактика на зависимост от хазарта.
Име: доц. д-р инж. Диляна Будакова
ЛИТЕРАТУРА Организация: ТУ-София, Филиал
Пловдив
1. Owramipur F., Eskandarian P., and Mozneb F. Адрес:бул. „Санкт Петербург“ 61, КСТ
S. (2013). Football Result Prediction with E-mail: dilyana_budakova@yahoo.com
Bayesian Network in Spanish League-
Barcelona Team, International Journal of Име: проф. дтн инж. Л. Даковски
Computer Theory and Engineering, 812-815, Организация: ТУ-София
Vol. 5, No. 5. DOI: 10.7763/ IJCTE. 2013. V5. E-mail: l.dakovski@gmail.com
802.
http://www.ijcte.org/papers/802-N30016.pdf
DEVELOPMENT OF CLOUD COMPUTING

BASED SCADA IN ELECTRICAL POWER
SYSTEMS
DIMITAR GROZEV, GRISHA SPASOV, NIKOLAY KAKANAKOV, GALIDIYA PETROVA
Abstract: The paper presents a development of experimental platform for implementation

of Cloud Computing in SCADA systems for Electrical Power Systems. The general idea of
the experimental study is to complete operation tests of traditional SCADA systems on
cloud computing platforms, and integrate the main application modules of SCADA systems
for electrical power metering on the cloud platforms. The logical networking structure of
the Control Center of SCADA systems is presented. The virtualization of the different type
of servers in the Control Center together with physical networking solution of the cloud
platform realization is described.
Keywords: Cloud computing, SCADA, Power systems, Cross vC NSX, SRM
functions. The software application system is

I. INTRODUCTION constructed and developed by using the standard
An electric power system typically involves CIM model, and software architecture has evolved
generation, transmission, transformation, from the Client/Server architecture to the current
distribution, consumption, and dispatching Browser/Server architecture. The new generation
processes. Electric power is generated and consumed Smart Grid [2] dispatching and control system uses
simultaneously and is not available for mass storage multi-core computer cluster technology to improve
or transportation. Thus, as the key to the electric system reliability and processing capacity, and uses
power system, monitoring and dispatching guarantee a service oriented architecture (SOA) to enhance
the reliability and security of electric power system interoperability and achieve horizontal
generation, transmission, distribution, and integration and vertical interconnection of power
consumption, and play a key role in providing high grid dispatching services. Hardware infrastructure of
quality and economic electric power. An electric these application systems involves high performance
power dispatching and monitoring system is often servers, complex high speed computer networks,
called a supervisory control and data acquisition high performance and highly reliable data storage
(SCADA) system, energy management system systems, and workstations. The SCADA software is
(EMS), or distribution management system (DMS) developed based on the platform which consists of
globally. In this article, we call it the electric power Windows, UNIX, and Linux operating systems, and
SCADA system. In the early stage, the electric is based on relational databases.
power SCADA system mainly used multi computer The whole system is connected through computer
architecture, consisting of single server and two networks for data exchange and sharing, and
server cluster systems. Currently, the system uses application programs share information through an
computer systems with a distributed open enterprise service bus (ESB). With the development
architecture [1]. of the traditional IT application systems, IT-based
The electric power SCADA system provides applications have been expanding deeply to another
SCADA, automatic generation control (AGC), industry field, and encountered various problems
automatic voltage control (AVC), EMS, DMS, and bottlenecks. The same is true for the electric
dispatcher training system (DTS), geographic power SCADA system, which is a professional IT
information system (GIS), and other useful application system. The future electric power
II-211
dispatching center should have high computing 1. RTU – remote terminal units are connected
capacity, and powerful information acquisition, to sensors. They have telemetry hardware
integration, and analysis functions especially with capable of sending digital data to the
the advancement of Smart Grid applications. supervisory system, as well as receiving
Existing centralized computing platforms of digital commands from the supervisory
electric power systems can hardly meet the above system.
requirements, which have become one of the major 2. PLCs – programmable logic controllers are
bottlenecks in the Smart Grid. Some of the major also connected to sensors. PLCs have more
disadvantages of these platforms are as follows[3]: sophisticated embedded control capabilities
• Low basic resource utilization and poor than RTUs. PLCs do not have telemetry
scalability. A large amount of basic hardware, although this functionality is
computing resources to meet the demands in typically installed alongside them. PLCs are
peak hours are idle during off peak hours. sometimes used in place of RTUs as field
To ensure reliability, lots of resources are devices because they are more economical,
redundant, and cannot be fully utilized. versatile, flexible, and configurable
Contradictions are growing between energy 3. Telemetry system is used to connect RTUs
demands and conservation policies. Due to and PLCs to control center or data
the upgrade of business, the existing IT warehouse systems.
infrastructure cannot be reused. Currently, 4. A data acquisition server is a software
analysis and computing in the electric power service which uses industrial protocols to
system rely on the centralized computing connect software services, via telemetry,
platform in the dispatching center. Due to with field devices such as RTUs and PLCs.
limited computing capacity, poor scalability, It allows clients to access data from these
and high upgrade costs, large scale power field devices using standard protocols.
systems suffer from insufficient data storage 5. HMI is the apparatus or device which
and analysis capabilities. presents processed data to a human operator,
• Poor system interoperability leading to and through this, the human operator
information islands. Parallel application monitors and interacts with the process.6. A
systems have their own architectural Historian is a software service which
features, and therefore resources cannot be accumulates time-stamped data in a database
exchanged or reused. which can be queried or used to populate
• Increasing management costs and risks, and graphic trends in the HMI. The historian is a
decreasing equipment utilization. client that requests data from a data
• acquisition server.
II. SCADA SYSTEM COMPONENTS 6. A Historian is a software service which
accumulates time-stamped data in a database
The traditional components of SCADA systems which can be queried or used to populate
are shown on figure 1. graphic trends in the HMI. The historian is a
client that requests data from a data
acquisition server.
7. A supervisory (computer) system, gathering
(acquiring) data on the process and sending
commands (control) to the SCADA system.
8. Communication infrastructure.
III. CLOUD COMPUTING TECHNOLOGIES

FOR ELECTRIC POWER INDUSTRY
Cloud computing can be used to solve various

problems occurred during application deployment,
use, and innovation process of the electric power
SCADA system. Nowadays most of companies have
their own cloud platform and virtual infrastructure,
Fig. 1. SCADA network architecture where they run business critical application and store
data.
Private cloud advantages [4]:
II-212
• It is reliable and scalable. All resources are
virtualized and in case of demand more
storage or computing without downtime or
impact can be added.With latest
technology,virtual infrastructure can easily
span across multiple data centers, without
need to change/mapping IP addresses, FW
rules, or deploy sophisticated clusters.
Enhanced disaster recovery is supported.
• Fast provisioning. Using techniques like
templates can deploy thousands of machines
with few clicks.
• Automation. Pretty much everything can be
automated using powercli, REST API,
powernsx, ansible and more. Common user
interface: decoupling the computation
infrastructure and the input system, enables
multiple user interfaces to exist side by side
allowing user-centric  customization.  
• Cost effective.
Fig. 2b. Primary site cloud physical topology

IV. PRIVATE CLOUD
INFRASTRUCTURE Primary site private cloud in use is shared by the
university departments and students. Everyone can
For the purposes of proposed experimental request computing capacity and after an approval
study,we use two virtual environments. Primary site process one could receive VMs with requested
- private cloud infrastructurein Technical University parameters. Secondary site is dedicated only to
of Sofia, branch Plovdiv and secondary site - private SCADA system and it’s used to cover HA
cloud infrastructure in Technical University of Sofia. requirements implementing disaster recovery site.
Physical topology of two sites is similar. Difference The following types of traffic are recognized:
is in used hardware. In Primary site, we use
Enterprise class storage, switches (Ethernet and FC) Local in each site:
and servers, while in secondary site we have mini • Management – this include all traffic
computers Intel NUC and Synology storage. generated by virtual infrastructure: vMotion,
Figure 2 (2a and 2b) shows secondary and primary iSCSI, NFS, fault tolerance, web-based
site physical topology. graphical user interface, etc.
• VM traffic – this is traffic generated from all
VMs connected to VLAN based port groups
– various types, generated by different
applications.
• VXLAN traffic – this include traffic
generated from all VMs connected to logical
switches - various types, generated by
different applications.
• SCADA traffic – this include traffic
generated from SCADA system servers
Between sites using WAN connectivity:

 Traffic generated by VMware Site Recovery
Fig. 2a Secondary site cloud physical topology Manager (SRM)
 VMs and SCADA traffic
Because of the importance and sensitivity of the

SCADA traffic we implement partial application
failover disaster recovery scenario using VMware
II-213
site recovery manager (SRM) and Cross vCenter (uDLR) [6][9] that optimize routing traffic between
NSX deployment [5] [8]. All SCADA components different SCADA components. Universal distributed
are protected by SRM. Cross vCNSX technology FW [7][9] is used to control the communication.
solves issues like IP address mapping and FW
policy. Universal Distributed FW allows deployment V.CONCLUSION
of security policy that span between primary and
secondary site. Universal DLR along with universal
The paper presents an implementation of a cloud-
distributed virtual switch span across sites too and
based SCADA system for use in electrical power
allow same universal SCADA virtual switch on both
control systems. The private cloud infrastructure in
locations. This way SCADA servers will keep their
use is based on VMware platform. Deploying
IP settings. (figure 3).
SCADA in cloud infrastructure automatically
inherits all cloud advantages:
1. Most of SCADA components are virtual
machines: DB server, Application Server,
OPC server, Engineering and HMI VMs.
This means that we can easily create new
ones, clone, backup, take snapshots, HW
upgrade - change parameters like CPUs,
RAM, HDD, software upgrades.
2. All SCADA components can be protected
by technologies like HA and FT[7].
3. All communication nodes are virtual nodes
and thus are also protected by HA - scada-
dsnet-esr1 and scada-dsnet-dlr1.
4. VMware Site Recovery Manager and cross
vC NSX technology is used to ensure
SCADA service is available even if we have
Data Center or site failure.
The presented test bed platform will allow various
types of information flows and configurations to be
investigated and analyzed and the best set-up for the
production services to be selected. Based on the
experiments with different configuration of network
services in the cloud the traffic flows will be
optimized for best performance in terms of delay
and resource usage.
Fig. 3. Private cloud network topology with
universal objects: uDLR, uDVS
GLOSSARY
WAN entry point of SCADA network aresite1-
scada-dsnet-esr1 and esr2 - ECMP configured edges. vSphere - VMware vSphere is the brand name for
They run dynamic routing protocol BGP and send VMware's suite of virtualization products.
updates for SCADA networks. Manipulate different ESXi/ESX – Elastic Sky X. An enterprise-class,
BGP attributes in order to control SCADA networks type-1 hypervisor developed by VMware for
entry point. Same pair of ECMP edges we have in deploying and serving virtual computers.
both sites.Site1-scada-fw1 is NSX router connected
to internet that has multiple features enabled: VXLAN (Virtual Extensible LAN) – A Virtual
firewall, IPSec, SSL VPN support and load- Network that emulates an Ethernet broadcast
balancing. It’s an entry point for all internet domain.
accessible SCADA components like some RTUs and vMotion– VMware vSphere live migration allows
PLCs. To keep traffic private, we use IPSec and SSL you to move an entire running virtual machine from
tunnels. Through this secure channel data for one physical server to another, without downtime.
controlled processes are collected. Universal-scada-
dsnet-dlr1 is a universal distributed logical router
II-214
NSX - VMware NSX is the network virtualization Computer Society, Washington, DC, USA, pp. 868-
platform for the Software-Defined Data Center 871, 2012.
(SDDC).
[4] Alamri A, Ansari WS, Hassan MM, Hossain MS,
DRS (Distributed Resource Scheduler) – technology Alelaiwi A, Hossain MA. A survey on sensor-cloud:
for balancing the computing capacity by cluster to architecture, applications, and approaches,
deliver optimized performance for hosts and virtual International Journal of Distributed Sensor
machines. Networks, Article ID 917923, Feb 2013.
VDS – VMware vSphere Distributed Switch (VDS) [5] VMware NSX Technical Product Management
provides a centralized interface from which you can Team. VMware® NSX for vSphere Network
configure, monitor and administer virtual machine Virtualization Design Guide ver 3.0, Aug 21, 2014,
access switching for the entire data center. online [04.02.2016]:
DLR – Distributed Logical Router. It separates www.vmware.com/files/pdf/products/nsx/vmw-
Control and Data plane. Control plane is a VM, but nsxnetwork-virtualization-design-guide.pdf
data plane is part of hypervisor kernel.
[6] B. Hedlund. Distributed virtual and physical
ECMP – Equal Cost Multi Pathing. Routing data routing in VMware NSX for vSphere, The Network
across multiple links with same metric. Virtualization Blog, November 25, 2013. Online:
[04.02.2016]:
ACKNOWLEDGMENTS https://blogs.vmware.com/networkvirtualization/201
3/11/distribut ed-virtual-and-physical-routing-in-
The presented work is supported by the National vmware-nsx-for-vsphere.html
Science Fund of Bulgaria project “Investigation of
[7] DebashisBasak, RohitToshniwal, Serge
methods and tools for application of cloud
Maskalik, and AllwynSequeira.Virtualizing
technologies in the measurement and control in the
networking and security in the cloud. SIGOPS Oper.
power system” under contract Е02/12
Syst. Rev. 44, 4 (December 2010), pp. 86-94, 2010.
(http://dsnet.tu-plovdiv.bg/energy/).
[8] VMware NSBU team.Recovering NSX backed
REFERENCES Data Ceneters utilizing SRM (Site Recovery
Manager). Online [21.4.2016]:
[1] Arghira N., et al. Modern SCADA philosophy in
https://blogs.vmware.com/networkvirtualization/201
power system operation – a survey, Scientific
6/04/enhanced-disaster-recovery-with-nsx-and-
Bulletin, University POLITEHNICA Bucharest,
srm.html#.WKWv6hD4Eg0
Series C: Electrical Engineering, vol. 73, no. 2, pp.
153-166, 2011. [9] VMware NSX for vSphere 6.2 Documentation
Center, online [02.02.2017]:
[2] Fang Xi, MisraSatyajayant, XueGuoliang, and
https://pubs.vmware.com/NSX-62/index.jsp
Yang Dejun. Smart Grid – The New and Improved
Power Grid: A Survey, IEEE Communications
Surveys & Tutorials 2012, vol. 14, no. 4, pp. 944 –
980, 2012. Authors’ contacts
[3] Gang Tan, Li Zhao, Yaduan Wang, Jingwei Xu, Department of Computer Systems and Technologies,
and Honggang Liu. A Safety Design of Electric Department of Electronics, Faculty of Electronics
Cloud Computing Platform, In Proceedings of the and Automation
Fourth International Conference on Computational Technical University of Sofia, branch Plovdiv
and Information Sciences (ICCIS '12). IEEE 25TsankoDiustabanov Str., 4000 Plovdiv, Bulgaria
dimitargr@gmail.com, {gvs, kakanak, gip}@tu-
plovdiv.bg
II-215

Mechanical, Transport and Aviation Engineering
Mашиностроителна техника и технологии;

Машиностроене и уредостроене;
Транспортна и авиационна техника
NUMERICAL ANALYSIS OF BI-METALLIC PARTS USING A

FINITE ELEMENT SOFTWARE
CENK MISIRLI, ISIK CETINTAV, YILMAZ CAN, CEM MISIRLI
Abstract: Numerical estimation of upsetting processes have very important role in

manufacturing world. In this investigation, bimetallic cylindrical parts have an outside
material and a different solid material inside has been upset in one operation and behavior
of this component during the operation investigated. Two bimetallic component models
were considered. Each model has a ring material as steel. Model 1 has a solid core
material as brass and Model 2 has a solid core material as copper. Parts were upset to
different heights by using parallel flat dies in press machine. Press load was recorded
during the process. Loads and metal flows have been compared and shown reasonable
results for two models. Different finite element software calculated and run very close
results with th experiment.
Key words: bimetallic, upsetting, finite element
In metal forming world, open die forging is the some of the cold-forging features of composite steel
most important and fundamental technique. Over preforms of varying titanium carbide contents
decades, about open die forging of single materials, during cold upsetting under triaxial stress state
such as steel, copper, aluminum, lots of works were conditions.
studied both theoretically and experimentally. Nowadays, compared to single components,
Vilotic et al.[1] have given particular attention to especially in marine industry, chemical industry and
process of upsetting of cylindrical billets. K. electrical equipment, bimetallic components are
Baskaran and R. Narayanasamy investigated study mostly preferred in industry due to their some better
the bulging characteristics of preformed irregularly properties. For example, bimetallic parts especially
shaped aluminium billets (Elliptical billets) during provide corrosion resistance and better conductivity.
cold upset forging using lubricanton both sides of Bimetallic components have an advantage due to
elliptical billets of pure commercial aluminium each two materials consist of different chemical
under different stress state conditions namely compositions. For example, impurities and different
uniaxial, plane and triaxial stress states [2]. S. grain size in materials show different plastic
Malayappan et al. studied experimentally the deformation characteristics. But bimetallic
bulging effect of aluminium solid cylinders by components are not composite materials, because in
introducing an extrusion die at one end [3]. Aksakal composite materials reinforcing elements spread
et al. investigated upsetting of polygonal samples by uniformly in matrix materials.
using dual stream functions theoretically and In recent years, plastic forming of bimetallic parts
experimentally [4]. Thaheer and Narayanasamy[5] have studied by a lot of researchers. Jingcai Wang
studied barrelling effect of truncated cone billets of et al.[11] investigated hot forging of multi-material
various materials. Also Thaheer and Narayanasamy cladded work pieces using upsetting tests, Yang et
explained barrelling effect of truncated cone billets al. [12] studied by using upper bound method.
of various materials [6]. Chang and Bramley[7] Eivani and Tahiri [13] used equal channel angular
investigate the determination of the heat transfer extrusion process to produce bimetallic rods and
coeffcient at the workpiece-die interface for the compared to one produced by general extrusion
forging process. Dyja et al. [8] worked theoretical process. Plancak et al. [14] studied experimental
and experimental analysis of rolling of bimetallic analysis of joining of two bimetallic axisymmetric
Cu-Al and Cu- Steel rods. Kocanda et al.[9] components from various materials by upsetting
examined contact pressure distribution in upsetting bimetallic components in a closed die. Plancak et al.
of compound cylindrical and cubic metals. [15] examined the behaviour of bimetallic
Senthilkumar and Narayanasamy[10] evaluates components which consist of an outer ring of mild
II-217
steel C45 and inner core of softer C15 steel during
the cold upsetting process. Plancak et al. [16] then After prepared all the metals in lathe machine with
extended to this research and compared all the appropriate dimensions, inner solid cylinder was
experimental data to one obtained by using finite pressed by low pressure into outer ring material in
element method. order to produce these bimetallic components.
Produced bimetallic components are shown in Fig.
3.
Fig. 4 shows the 150 tons’ hydraulic press that the
experiments were carried out. It has a constant
speed of 5 mm/s. The hydraulic press was equipped
with a pressure current transducer in order to
measure and record the experimental load. The
pressure or load value was read by an I/O card of
Fig.1. Plancak Bi-metallic Model [16] the personal computer during the experiment.
This value was converted into a resistance value,
In this work, the open die forging of bimetallic then this resistance value, consequently the pressure
cylindrical components producing using different value was provided as an input value to the
materials designed as two models was examined software.
both experimentally and analytically. This process
may provide many parameters for metal forming
operations, such as formability and barreling. The
aim of this study was to examine and compare the
metal flow and load requirement for these two
models when upsetting of bimetallic cylindrical
components.
2. EXPERIMENTAL WORK
For the interaction between different metals, two

bimetallic models have been considered for this
work. Material combinations and properties of these
models have shown in Table 1. Basic design of
Model 1 and Model 2 are shown in Fig. 2.
Table 1. Combinations of Model 1 and Model 2
Model 1: Ring material AISI 1020 STEEL

Solid inner cylinder material BRASS Fig.3. Produced Bimetallic Components
Pb(%) Al(%) Fe(%) Ni(%) Sn(%) Zn(%) Cu(%)
1,6-2,5 0,05 0,3 0,3 0,3 Rest 57-59 There were no special lubricants used but
Model 2: Ring material AISI 1020 STEEL specimens and flat dies were sanded before the
Solid inner cylinder material COPPER experiments. The fixation of components to
O(%) Cu(%) hydraulic press is also shown in Fig. 4. Then
0,04 99,9 bimetallic components were forged with reduction
ratios %15, %30 and %40.
Fig.2. Design of Bimetallic Models

II-218
Fig.5. Meshed component and interactions
4. RESULT AND DISCUSSION
Open die forging of bimetallic cylindrical parts

producing using different metals was studied
experimentally and analytically for two different
models. Three different metals have been used in
order to produce bimetallic part models that have a
ring and a solid cylinder. Model 1 has a ring
material brass, and Model 2 has copper, each of
these has steel solid cylinder. These parts were
Fig.4. Hydraulic Press and fixation of parts forged in an open die with reduction ratios %15,
%30 and %40 experimentally and FE simulation
3. FE SIMULATION model has run for the same conditions.
In order to validate experimental work, a simulation

model was designed using finite element softwares-
DEFORM and ABAQUS. This gives more insight
and helps to compare with experimental data of
open die forging of bimetallic parts, also provides
conducting further analyses about this work.
First, bimetallic parts were designed in Deform
software as 2D model and due to the symmetry only
one half of the parts were performed. All the metals Fig.6. Meridional cut after %15, %30 and %40
reductions
properties were selected from Deform software
library and they were assigned to the designed parts.
For meshing, 8 node axisymmetric elements were After reduction, samples were cut and barrel
used. For all the FE models, at the part-die profiles of the inner solid cylinder and outer ring
interfaces a friction coefficient 0.1 is considered, were observed. Fig.6 and Fig. 7 shows that
but between the inner solid cylinder and the outer deformations and barreling of bimetallic parts after
ring interface zero friction was assumed. Surface- %15, %30 and %40 reductions. Clearly it is shown
to-surface contact was applied to all the friction that Model 2 has much better material flow and zero
locations. Fig.5 shows meshed component and cavity between the inner and outer interface. It is
interactions between all the contact surfaces. observed that the meridional cross section of inner
cylinder material flows very fast and occurred a
barrel profile is different from a normal barrel arc.
Inner cylinder material barrel profile is not the same
as outer ring material barrel profile. It can be also
seen that in Model 1 there is no contact between
inner and outer material and a cavity occurred in the
inner meridian part of inner material. This cavity
can be explained that inhomogeneous deformation
II-219
occurs in the centerline and this cavity is attributed
to a state of hydrostatic stress. At %40 reduction  Obtained experimental results showed
Model 2 has a defect on the right side. It is guessed that metal flow has inhomogeneity for these
that high speed of hydraulic press and the close two models especially for solid cylinder
flow curves of copper and steel may cause this materials. The shift of ring and solid
fracture. cylinder materials also affects metal flow
and load requirement.
 A defect occurred at %40 reduction of
Model 2 as seen in Fig.6. It is considered
that due to the high speed of hydraulic press
this defect has taken place. In the future
works about this topic experimental works
will be carry on lower hydraulic press
speed.
 As considered the load-stroke and load-
reduction graphs, Model 2 or Steel-Copper
bi-metallic sample needs lower load to
produce or use. Also it can be seen that
Model 2 has better material flow than
Model 1. Copper and steel are more useful
to produce bi-metallic materials.
 This work has shown that under certain
conditions, in a range of tolerances,
(a)
bimetallic cylindrical components may be
forged while carrying on a good interface
contact between the two components. This
establishes the possibility of producing bi-
metallic components via this process either
as a finished component or, more likely, as
a pre-forming operation prior to further
processing.
 In future works, flow curves of single
metals copper, brass and steel will be obtain
and investigate how effect the flow curves
to form of bi-metallic samples.
 In order to obtain further information for
producing different bimetallic components
for different materials, more FEM
simulation and more experimental work are
planning.
6. ACKNOWLEDGEMENT
(b)
The authors would like to thank to Trakya
Fig.7. Comparison of FE simulation and
experimental work after %40 reduction, (a) Model 1 (b)
University-Turkey (TUBAP-2011-193) for their
Model 2 help in experimental part of the study.
Fig.7 shows that comparison of FE simulation and 7. REFERENCES

experimental work after %40 reduction. FE
simulation has really similar agreement with the [1] Vilotic, D.; Vujovic, V. & Plancak M. (1992).
experimental results. Analysis of Upsetting of Cylinder by Conial Dies.
Journal for Technology of Plasticity. Vol., 17 (1-2),
pp. 67-72
[2] K. Baskaran, R. Narayanasamy (2008). Some
aspects of barrelling in elliptical shaped billets of
5. CONCLUSIONS
II-220
aluminium during cold upset forging with lubricant. [10] Senthilkumar, V., Narayanasamy, R., (2008).
Materials and Design 29, 638–661 Influence of titanium carbide particles addition on
[3] S. Malayappan, R. Narayanasamy, K. the forging behaviour of powder metallurgy
Kalidasamurugavel (2007). A study on barrelling composite steels. Proc. IMechE Vol. 222 Part B: J.
behaviour of aluminium billets during cold Engineering Manufacture, 1333-1345.
upsetting with an extrusion die constraint at one [11] Jingcai Wang, Laurent Langlois, Muhammad
end. Materials and Design 28, 954–961 Rafiq, Régis Bigot, Hao Lu (2014). Study of the hot
[4] Aksakal, B.; Sezek, S. & Can Y. (2005). forging of weld cladded work pieces using upsetting
Forging of Polygonal discs using the dual stream test. Journal of Materials Processing Technology.
functions. Materials & Design. Vol., 26, pp 643-654 214, 365– 379
[5] Thaheer, A. S. A. & Narayanasamy, R. (2007). [12] Yang, D. Y.; Kim, Y. G. & Lee, C. M. (1991).
Barrelling in truncated lubricated zinc cone billets An upper-bound solution for axisymmetric extrusion
during cold upset forging. Materials & Design. of composite rods through curved. International
Vol., 28, pp. 434-440 Journal of Machine Tools and Manufacture. Vol.
[6] Thaheer, A. S. A. & Narayanasamy, R. (2007). 31, pp. 565-575
Barrelling in truncated lubricated zinc cone billets [13] Eivani, A. R. & Taheri, A. K. (2007). A new
during cold upset forging. Materials & Design. method for producing bimetallic rods. Materials
Vol., 28, pp. 434-440 Letters. Vol. 61, pp. 4110-4113
[7] Zhang, H.G., Dean, T.A.,(1995). Interaction of [14] Miroslav PLANČAK, Zlatan CAR, Dragiša
process and machine in multi-stage forging on a VILOTIĆ, Dejan MOVRIN, Marko KRŠULJA
mechanical press. Part B: Journal of Engineering (2011). Upsetting of bimetallic components in a
Manufacture. Proc Instn Mech Engrs Vol 209, 19- closed die. Engineering Review Vol. 31, Issue 2,
32. [8] Dyja, H.; Mroz, s. & Milenin, A. (2004). 105-109
Theorical and experimental analysis of the rolling [15] Miroslav PLANCAK, Igor KACMARCIK,
process of bimetallic rods Cu-steel and Cu-Al. Dragiša VILOTIC, Marko KRSULJA (2012).
Journal of Materials Processing Technology. Vol. Compression of bi-metallic components- analytical
153-154, pp. 100-107 and experimental investigation. International
[9] Kocanda, A.; Presz, W.; Adamczyk, G.; Journal of Engineering. 1584-2665
Czyzewski, P. & Mazurek, W. (1996). Contact [16] K. Essa, I. Kacmarcik, P. Hartley, M. Plancak,
pressure disrtubition in upsetting of compound D. Vilotic (2011). Upsetting of bi-metallic ring
metals. Journal of Materials Processing billets. Journal of Materials Processing Technology.
Technology. Vol., 60, pp. 343-348 212 817– 824
COMPUTER AIDED RESEARCH OF THE SHAFT

OF A ROTOR CRUSHER OF AN ECCENTRIC
TYPE
ZHIVKO ILIEV, GEORGI DINEV
Abstract: The present article reviews the basic classification of a rotor crusher of an
eccentric type as well as its usage in the field of mining industry. The analytical active
forces of the eccentric shaft have been determined as well as the reactions in the bearing
supports have been calculated. A 3D model of the crusher has been made as it has been
used for the purpose of computer aided simulation research of one of the most stressed
details, namely, the eccentric shaft. The obtained results have been analyzed as well as
some recommendations have been given as they concern the possibilities for prolongation
of the durability of the researched unit.
Key words: mine transport, rotor eccentric crusher, finite element method
1. Basic classification fixed on axes 5 that are fixed within the frame of
According to their pattern of action, the the crusher. Each jaw has hinged connection with
rotor crushers of eccentric origin are classified the hydraulic cylinders 1 in its lower part as the
somewhere between the conic types and the jaw jaws are fixed within the frame of the crusher by the
types of crushers. The process of grinding for this use of special cylindrical pins.
type of crushers is conducted by the use of two
steep working chambers that shrink up and down as
they are placed symmetrically in accordance with
the vertical plane (fig 1). The vertical plane is the
symmetrical plane of the machine itself. The
operation of that type of a crusher is driven by an
eccentric type of mechanism. There are four basic
cinematic groups: grinding jaws, a rotor, an
eccentric type of a shaft and a support protection
system. Fig. 2 and fig.3 show cinematic sketches of
an eccentric type of a rotor crusher along with a
hydraulic type of a support protection system and an
outlet that is hydraulically controlled.
The grinding chambers [1, 4, 16, 17] are
placed between the curved operational surfaces (fig.
1) of the immobile jaws 1 and the cylindrical
mobile rotor 3 that is mounted between them along
with the shaft that is placed horizontally. Rotor 3 is
located into the eccentric part of shaft 2 which is
Fig.1 Common look of a rotor crusher of an
located in accordance with the basic pins. The shaft
eccentric type
is fixed within the frame of the crusher in
accordance with the pins. The crushing jaws 4 are
located next to the both sides of the rotor as they are
symmetrical to the basic pins of the eccentric shaft.
They are mounted on the upper part as they hang

II-222
Fig. 5 A sketch that is used for determination of the

Fig. 2 A cinematic sketch of a rotor crusher
reactions of the bearing supports within the plane of
of an eccentric type and a hydraulically controlled
X-Y
outlet
Fig. 6 A sketch that is used for determination of the

reactions in the bearing supports within the plane of
Fig. 3 Forces of grinding of the rotor crusher of an X-Z
eccentric type
The reactions Az and Bz within the bearing
Figure 2 depicts a cinematic sketch of a supports (Fig. 4) are to be defined by the moment
rotor crusher of an eccentric type with hydraulic equations:
control of its outlet as it has the following Regarding plane X-Z (Fig.6) we have the following:
composite details: 1 a grinding jaw, 2 mobile brake
of a rotor, 3 hydro cylinders, 4 pneumatic-hydro ∑ 𝑀𝑀𝐵𝐵 = 0; (1)
accumulator.
Fig.3 illustrates a sketch of the forces of 0,5. 𝐺𝐺𝑚𝑚 (𝑐𝑐 + 𝑏𝑏 + 2. 𝑎𝑎 + 𝑏𝑏) − 𝐴𝐴𝑧𝑧 (2. 𝑎𝑎 + 2. 𝑏𝑏) +
crushing of the crusher that shows 𝜃𝜃 = +0,5. 𝑃𝑃𝑒𝑒 . 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠. (2𝑎𝑎 + 𝑏𝑏) + 0,5. 𝑃𝑃𝑒𝑒 . 𝑠𝑠𝑠𝑠𝑠𝑠. 𝑏𝑏 −
100 𝑎𝑎𝑎𝑎𝑎𝑎 𝛾𝛾 = 250 – angles which combine forces Q −(0,5. 𝐺𝐺𝑚𝑚 + 𝑄𝑄. 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠). 𝑐𝑐 = 0;
and Pe along with axis OY. The forces of the belt 𝐴𝐴𝑧𝑧 = 53 504, 𝑁𝑁
gearing are depicted on Fig 4.
Force Q = 7452 N is the maximum force where 2.а=460 mm; b=320mm; c=350mm is the
which is applied upon the shaft as a result of the geometrical size of the rotor head along with the
forces of the belt gearing [3, 21]. shaft
2. Determination of the reactions in the bearing ∑ 𝑀𝑀А = 0; (2)

supports of the eccentric shaft.
0,5. 𝐺𝐺𝑚𝑚 . 𝑐𝑐 − 0,5. 𝑃𝑃𝑒𝑒 . 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠. 𝑏𝑏 − 0,5. 𝑃𝑃𝑒𝑒 . 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠.
(2. 𝑎𝑎 + 𝑏𝑏) + 𝐵𝐵𝑧𝑧 . (2. 𝑎𝑎 + 2. 𝑏𝑏) −
−(0,5. 𝐺𝐺𝑚𝑚 + 𝑄𝑄. 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠). (2. 𝑎𝑎 + 2. 𝑏𝑏 + 𝑐𝑐) = 0;
𝐵𝐵𝑧𝑧 = 55 620, 𝑁𝑁
where 𝐺𝐺𝑚𝑚 = 6800𝑁𝑁 –is the weight of the flywheel

[2,3]
1000. 𝜂𝜂𝑚𝑚 . 𝑁𝑁𝑒𝑒𝑒𝑒𝑒𝑒

𝑃𝑃𝑒𝑒 = = 239 062𝑁𝑁; (3)
4. 𝑒𝑒. 𝑧𝑧
Fig. 4 Forces within the belt gearing
II-223
Force 𝑃𝑃𝑒𝑒 is perpendicularly [2] applied The material that is assumed to be used for
according to the operational plane of the grinding the explored detail is steel type S 235 J2G3 БДС
jaw (Fig.3). EN 10025, as it sustains allowable stresses of
where 𝜂𝜂𝑚𝑚 = 0,85 – a mechanical К.П.Д.; approximately [ AL ]  160MPa .
- 𝑁𝑁𝑒𝑒𝑒𝑒𝑒𝑒. = 90𝑘𝑘𝑘𝑘 – power of the drive;
- е =10 mm – eccentricity of the shaft; 2.2. Results from the static analysis that
- z = 8 Hz –number of swaying movements of the was conducted.
shaft;
Regarding plane X-Y (Fig. 5) we have the

following:
∑ 𝑀𝑀А = 0; (4)
𝐵𝐵𝑦𝑦 . (2. 𝑎𝑎 + 2. 𝑏𝑏) − 0,5. 𝑃𝑃𝑒𝑒 . 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐. 𝑏𝑏
−0,5. 𝑃𝑃𝑒𝑒 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐(2𝑎𝑎 + 𝑏𝑏) + 𝑄𝑄. 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐(2. 𝑎𝑎 + 2. 𝑏𝑏 + 𝑐𝑐)
=0
𝐵𝐵𝑦𝑦 = 206 637, 𝑁𝑁
∑ 𝑀𝑀𝐵𝐵 = 0; (5)
Fig.8 Loading applied upon the eccentric shaft
𝐴𝐴𝑦𝑦 . (2. 𝑎𝑎 + 2. 𝑏𝑏) − 0,5. 𝑃𝑃𝑃𝑃. 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐. (2. 𝑎𝑎 + 𝑏𝑏) +
+0,5. 𝑃𝑃𝑒𝑒 . 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐. 𝑏𝑏 + 𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄𝑄. 𝑐𝑐 = 0 The capabilities for MKE investigation of the
𝐴𝐴𝑦𝑦 = 105 997, 𝑁𝑁 software product will be applied throughout the
process of researching by simulation. These are as
follows:
- Equivalent stress according to the Theory of
von Mises [5, 6, 18, 19, 20];
- Reliability (Safety) coefficient according to the
Theory of Mises-Henky
Fig.7 A drawing of the eccentric shaft According to the Theory of von Mises-Henky,
the limit of protraction is to be defined by the
correspondence between the amount of equivalent
2.1. Methodology of researching
stress according to von Mises  VON and the
In order to increase the values of the amount of allowable stress [ AL ] [7]
exploitation indexes throughout the process of
designing of the machine elements, one can use the
 VON  [ AL ]
capabilities of contemporary software products. The stress types, according to von Mises,
The first stage of execution of the research can be defined by the three main types of stress in
is creation of a 3D CAD model of the eccentric correspondence with dependency [6, 11, 12, 13].
shaft of the rotor crusher (fig. 7). The instance
applies to a detail that is shown on fig. 8. which has ( 1   2 ) 2  ( 2   3 ) 2  ( 1   3 ) 2 (6)
 VON 
been drawn by usage of Solid Works. Limitation 2
with zero displacements has been imposed using the
option of 'fixed' as it is required for the purpose of
execution of the computer aided simulation. The
consequent reactions within the bearing supports
will affect the stress upon the shaft significantly.
Concerning that fact, one has to assume the greater
values, namely the results from plane X-Y, as they
represent a more serious case of operation of the
shaft.
A computer aided simulative and structural
analysis has been conducted in order to determine Fig. 9 Distribution of the equivalent types of stress
size and distribution of the equivalent stresses. within the eccentric shaft

II-224
3. Application of Expert system for design Unity. One rule cannot be used more than once
filet curve. for one and the same data.
Actuality. The elements in operational memory
Production rules are a formalism used in in OPS5 contain a field for time, thus it is known on
design of program languages before it was used in the period of wich cycle they are added to
psychological modeling [7] and in Expert systems operational memory.
[8],[9]. Specificity. The concrete elements, drawn from
more specific rules, for example rules with greater
number of condition, are more difficult to be
satisfied and therefore are prefered before more
general rules with smaller number of conditions.
3.2 Production Ruls

In the process of designing the shaft
designer may use the production rules by carrying
out simulation analysis by the finite element method
searching the optimum value of the filet radius from
the standpoint of stress concentration.
This is illustrated by the following
Fig. 10 Distribution of the Factor of algorithm:
Reliability (Safety) within the eccentric shaft
If: 1. The radius of curve is 0,5 mm;
2. The stress   0,260Е9.MPa
Then: To change the radius of 1,5 mm.
If: The stress   0,200Е9.MPa
Then: To chose Steel grade 41Cr4
If: The stress  0,200Е9.MPa.
Then: The concentration of stress are lucky.
In next Fig.12 are shown the results from

investigation of shaft with Finite Element Methods-
ANSYS by 3D geometrical model and determined
Fig. 11 Distribution of the equivalent displacements theoretical distribution of stresses in character
within the eccentric shaft points and surfaces concentrators. The designer can
to used expert system included some algorithms.
For example let to illustrate choice of fillet curve of
Some of most popular model from shaft.
“production rules” include: Rules from kind “IF”
and “THEN”; Different kinds of network; Logistic
equations for coding of facts and principles”[10].
In the literature concerning expert systems
sometime they are call rules from type “condition-
action” or “situation-action”. This is so because in
general they are used to code empiric relations
between models of data presented to the system and
actions, which the system must execute as a
consequence.
3.1. Solving of conflicts
The strategy for solving conflicts is of

particularly importance for the work of production
system and for this purpose must be chosen
precisely. The mechanisms for conflicts solving Fig. 12 Distribution of stresses on fillet curve of
varied but three of them are very popular and often shaft
are used in combination thus forming a global The results shown that maximum value for
direction regime. stresses are in left and right zone of fillet curve.
II-225
This value are from 49 MPa to 53 MPa.in the both “Technologies and Practices in Underground
sides of shaft. The designer can to used the same Mining and Mine Construction”, Devin, Bulgaria.
algorithm.
7. Dinev G. An approach for simulation
design of mechanical assembled unit, Journal
4. Analysis of the obtained results and Advanced Material Research , Vol. 463-464, Trans
conclusions Tech Publication, Switzerland, ( 2012), pp.1085-
1088.
Fig. 9 shows clearly that, according to Von
8. Jackson P. Introduction to expert systems.
Mises, the maximum amounts of stress within the
Allison Wasley Longman, 1998.
shaft are of the value of 156,7 MPa. All types of
stress are concentrated within the bearing supports. 9. Dinev G. Expert systems used throughout
In Fig. 10, it is evident that the minimum factor of the training of master engineering in the field of
reliability (Safety) FOS = 1,1 is quite noticeable at industrial engineering, International virtual journal
the place of fixation. However, the maximum of for science, technics and innovations for the
equivalent displacements appears at the middle industry, MTM, Year IX Issue 1/2015, ISSN 1313-
shoulder of the shaft (Fig. 11). This is due to the 0226, pp. 25-28.
resilience of the detail. The maximum of 10. Dinev G. Course on principles of
displacements is of a value around 0, 12 mm. These mechanical design. In: AVANGARD PRIMA,
results are normal and quite anticipated. It is 2011.
possible that the factor of reliability (Safety) could 11. Iliev Zh., Dinev G.(2016) “Improving the
be raised by constructive changes. An example of maintance of drive sheave of hanging ropeway”,
such changes is enlargement of the diameter of the International Conference on Engineering,
vulnerable shoulders. Technologies and Systems, TECHSYS 2016,
In the process of constructing machine parts Technical University – Sofia, Plovdiv branch,
using the capabilities of expert systems and the Bulgaria.
method of finite elements is improved the 12. Iliev Zh., Dinev G.(2015) „About an
concentration of stress in their operation. education of engineering specialists for a computer
design of the technical products in mining
REFERENCES production“, Technologies and Systems,
TECHSYS 2015, Technical University – Sofia,
1. Iliev Zh., Perenovski N. (2014) Plovdiv branch, Bulgaria.
“Opportunities for using the finite element method 13. Iliev Zh.,(2012) “Opportunities for using
by computer research on rotary crusher of the finite element method by computer research of three
eccentric type 800D x75”, International roll directive devices of mine skip winding
multidisciplinary scientific symposium machines” 12th INTERNATIONAL SCIENTIFIC
“Universitaria Simpro 2014”, Petrosani, Romania.. CONFERENCE VSU'.
2. Цанев Ц. (1973) „Върху някои 14. Iliev Zh.,(2012) ”Kомпютърно
кинематични и конструктивни особености на моделиране по метода на крайните елементи
роторна трошачка ексцентиков тип“, сп. на двуролково направляващо устройство на
Рудодобив, бр. 9, София. подемnи съдове за руднична клеткова уредба”,
3. Цветков Хр., Цанев Ц. (1976) “Technologies and Practices in Underground
„Кинематика на роторна трошачка Mining and Mine Construction”, Devin, Bulgaria.
ексцентриков тип при задействане на
15. Iliev Zh., Perenovski N. (2016) “Computer
предпазната й система.“ Годишник на ВМГИ,
modeling and investigation of the operating parts of
София.
the screw conveyor by finite element method”,
4. Dinev G, Zh. Iliev (2016) “About computer
“Technologies and Practices in Underground
design of machine and equipment the general
Mining and Mine Construction”, Devin, Bulgaria.
machine building”, International Journal
“MACHINE, TECHNOLOGIES AND 16. Iliev Zh., Bogdanov Iv.(2015)” Аnalysis of
MATERIALS”. the vibration state of the eccentric shaft with the
5. Lawrence, K.,(2005) ANSYS Workbench bearings of a complex pendulum jaw crusher”, XVI
Tutorial, SDC Publications, Texas Balkan mineral processing congress, Belgrad,
6. Iliev Zh., (2016) “ Computer modal Serbia.
analysis of a eccentric shaft of a complex pendulum 17. Tsvetkov Kr., Iliev Zh., Perenovski
jaw crusher””5- National scientific and technical N.,(2013)”Power of electric motor and forces of
conference with international participation”, breakage in jaw crushers with direct drive of the

II-226
movable jaw.” International Scientifis Conference, технологични и технически параметри на
UNITECH 2013, Gabrovo, Bulgaria. роторна трошачка ексцентриков тип
18. Iliev Zh., Perenovski N. (2015) “Computer “,Дисертация, ВМГИ-София,
simulation research of hydrostatic thrust bearings
of cone crusher type "kubria" by finite element
method”, Scientific journal 'Mechanics Transport Faculty of Mining Electromechanics
Communications': 3 / 2015. University of Mining and Geology”St.Ivan
19. Iliev Zh., Perenovski N. (2014) “Using Rilski”, Sofia
finite element method by design of impact crusher Prof. Boyan Kamenov St.
for average crushing”, Scientific journal 1700 Sofia
'Mechanics Transport Communications': 3 / 2014 BULGARIA
Faculty scientific conferences. E-mail: halkopirit@mail.bg
20. Radlov K., Iliev Zh. (2011) “Opportunities
for using finite element method by design of disk Faculty of Mechanical Engineering
feeder for mining and building materials”, Technical University of Sofia
International Scientific Conference, UNITECH Bvd. Kliment Ochridski-8
2011, Gabrovo, Bulgaria. 1000 Sofia
21. Цанев Ц. (1976), “Изследване на BULGARIA
функционалната връзка между основните E-mail: gdinev@tu-sofia.bg
COMPARATIVE STUDY OF THE MECHANICAL PROPERTIES

OF TI AND CR BASED HARD COATINGS DEPOSITED BY
CLOSE FIELD UNBALANCED MAGNETRON SPUTTERING
VASILIY CHITANOV, LILYANA KOLAKLIEVA, ROUMEN KAKANAKOV,
TETIANA CHOLAKOVA
Abstract: Two quaternary hard coating structures Ti/TiN/TiCrAlN and Cr/CrN/CrTiAlN,

were deposited on HSS substrates by close field unbalanced magnetron sputtering
(CFUBMS). A comparative study of their mechanical characteristics, nanohardness, a
coefficient of friction and adhesion was carried out. The effect of the nitrogen flow
controlled by an Optical Emission Monochromator (OEM) on the adhesion and transition
layers was investigated. Among both investigated structures, the Cr-based coating showed
better mechanical properties pronounced by higher nanohardness of 27 GPa, a lower
coefficient of friction of 0.089 and excellent adhesion up to 30N loading.
Key words: coatings, nanohardness, adhesion, CFUBMS
800-4 equipment (Teer Coatings, UK). Figure 1

1. Introduction presents a scheme of the CFUBMS vacuum
chamber used for deposition of hard coatings.
Chromium nitride (CrN) and titanium nitride
Two Al targets, one Cr and one Ti targets were
(TiN) coatings have well established background
used for the coatings deposition.
in advance machining operations [1]. The
subsequent ternary nitride coatings like TiAlN [2]
and CrAlN [3] exhibit improved tribological,
mechanical, thermal and oxidation properties
compared to the binary nitride based coatings.
Besides, CrTiN coatings show enhanced
mechanical behaviour [4]. The development of
TiAlCr alloys allowed applications at
o
temperatures as high as 1000 C [5]. The CrTiAlN
coatings obtained by Close Field Unbalanced
Magnetron Sputtering (CFUMS) exhibit a high
hardness and strong adhesion between the coating
and substrate [6]. Furthermore, this technology
allows deposition of compositionally graded
structures, which improves the tribological Fig. 1. A vacuum chamber scheme of the
characteristics of the coatings [7]. In this study, UDP 800-4 equipment.
Ti/TiN/TiCrAlN and Cr/CrN/CrTiAlN graded
Fig. 1.
coatings deposited by CFUBMS were
comparatively studied in respect to their Before the coating deposition the specimens were
mechanical properties, a nanohardness and ultrasonically cleaned in an alkaline solution,
adhesion. rinsed in deionised water and dried at temperature
of 130 °C. After that the substrates were loaded
and the vacuum chamber was evacuated to
2. Experimental 1.5x10-5 torr. The ion cleaning in an Ar plasma
The quaternary coating structures, and at a bias voltage of − 500 V was performed
Ti/TiN/TiCrAlN and Cr/CrN/CrTiAlN, named Ti- for 30 min immediately before the coating
based and Cr-based, respectively, were deposited deposition. To obtain better adhesion with the
on high speed steel (HSS) specimens by close substrate an adhesion layer of Ti or Cr was
field unbalanced magnetron sputtering in UDP
II-228
deposited in an Ar plasma. After that a transition µm and cone angle of 120о. The equipment allows
layer of TiN or CrN was grown by a gradiently loading in the interval of 0.1-30 N. The precision
increased N2 flow. The internal stress between the of the measurement of the loading is 0.3 mN. The
transition layer and the active coating layer was maximum loading depth is – 200 µm. The
reduced with deposition of a gradient TiCrAlN or software allows display of the penetration depth,
CrTiAlN layer. In this step, both parameters, the friction and load forces, acoustic emission and
target power and nitrogen flow were increased to coefficient of friction. The scratch tests were
values optimal for the active top layer of the made with a progressive load from 1 to 30 N and
coating. The active coating CrTiAlN and TiCrAlN a length of 1 mm for each investigated sample.
layers were deposited at constant values of the The scratch velocity was 0.5 N/min.
regime parameters. The total thickness of the Ti-
based coating is 2.0 µm, while of the Cr-based it 3. Mechanical properties
is 2.2 µm. During deposition, the nitrogen flow
The load–displacement curves of the Ti-
was controlled by an OEM monochromator,
based coating at indentation loads of 15 mN and
which was tuned to the Ti peak at 501 nm or to
200 mN are shown in Fig. 2.
the Cr peak at 421nm. The target’s current was
kept the same and the OEM set points were
chosen to fix the same N2 flow during the graded
layer deposition. The sputtering process was
adjusted by the power and was controlled by the
current. It was found out that best mechanical
parameter were achieved at a current of 5 A, 8 A
and 3 A of the Cr, Ti and both Al targets,
respectively. The currents of the different material
targets differ due to the different sputtering
material rates and their reaction activity to N2 [8].
The argon (Ar) flow was controlled by a Mass
Flow Controller (MFC) and was kept at 25 sccm
during the whole process. The pulsed bias voltage
was maintained at −70 V during deposition. The
working preassure during the process was
between 1.6x10-3 and 1.9x10-3 torr. The carousel Fig. 2. Load-displacement curves of
was rotating with 5 rpm. the Ti/TiN/TiCrAlN coating measured
The mechanical parameters were determined at: 1 - 15 mN, 2 - 200 mN.
with a Compact platform CPX- MHT/NHT –
CSM Instruments, Anton Paar, Austria. The The nanohardness H, elastic module E and
nanohardness was measured with a maximum indentation depth hmax are summarized
nanoindentation module. A microindentation in Table 1. A highest nanohardness of 23 GPa
module with a Rockwell indenter was used was measured at hmax=202 nm, which value
performing a scratch test in order to study the evidences that the influence of the substrate could
adhesion and to determine the coefficient of be accepted negligible according to the 10 % rule.
friction against the diamond indenter. The The corresponded module of elasticity was 271
nanoindentation module has a diamond Berkovich GPa.
indenter that allows indentation in the interval
0.01-500 mN. The equipment software Table 1. Nanohardness
automatically calculates the nanohardness and and elastic module
elastic module depending on the penetration
depth. The Oliver and Pharr methodology is Fn H E hmax
implemented for the calculations. An indentation [mN] [GPa] [GPa] [nm]
with a load of 200 mN was made for Ti/TiN/TiCrAlN
characterization the complex nanohardness 15 23.3 271 202
including the coating and substrate influence. 200 16.5 292 875
Besides, a loading of 15 mN was performed to Cr/CrN/CrTiAlN
eliminate the substrate effect keeping the well- 15 27 385 185
established 10% rule of penetration in the coating. 200 15.4 305 889
The scratch module has a certified diamond
spherical Rockwell indenter with a radius of 200
II-229
The are few published results about the
nanohardness of the TiCrAlN coatings deposited For the Cr based coating, the measured
with CFUBMS. A value of 32 GPa is reported for nanohardness at the maximum indentation depth
the same coating structure obtained by cathodic of 185 nm was 27 GPa and the elastic module was
arc evaporation [9]. This technique allows 385 GPa. The obtain result for the nanohardness
achievement of high ionisation rates and is comparable with that reported in other
deposition of dense coatings with an increased publications for CFUBMS [10, 11] but still could
nanohardness, but in a combination with be improved by an optimisation of the N2 partial
formation of macro particles. pressure [12].
The scratch test trends of the Ti based
coating are presented in Fig. 4. Four signals are
presented in the graph: the load force Fn increased
from 1 to 30 N, coefficient of friction μ, friction
force Ft and acoustic emission AE. As it is seen,
there is a very good adhesion between the coating
and the substrate. There is no delamination of the
coating. The AE signal has several peaks, which
correspond to rare small and short semi-circular
and angular cracks, which are not considerable for
the very good resistance of the coating during the
scratch test. For the whole test loading up to 30N,
coating chipping and spalling do not appear. The
Fig. 3. Load-displacement curves of measured coefficient of friction at the end of the
the Cr/CrN/CrTiAlN coating measured test was 0.167.
at: 1-15 mN, 2- 200 mN.
The results from measurements of the Cr-
based structure are presented in Fig. 3 and Table
1. The inserted photo presents the residual indent
on the coating surface after indentation with
200mN.
Fig. 5. Scratch test results for the

Cr/CrN/CrTiAlN coating:
a) Graphic trends of the test process, 1-Load
force, 2- Coefficient of friction, 3-Friction force,
4-Acustic emission;
b) A panoram photo of the scratch test track;
Fig. 4. Scratch test results for the
Ti/TiN/TiCrAlN coating: The scratch test results of the
a)Graphic trends of the test process, 1-Load Cr/CrN/CrTiAlN coating are given in Fig. 5. This
force, 2- Coefficient of friction, 3-Friction force, coating also have not delamination. The AE signal
4-Acustic emission; is clear and no cracks were found during the test.
b) A photo of the scratch test track; The only one pick at the end of AE signal could

II-230
be attributed to the unloading process of the friction and excellent resistance to the progressive
indetor or defect in the substrate and it is not load from 1 to 30 N, while the coefficient of
corresponding to a crack in the scratch track. The friction is higher and the scratch resistance is very
resistance of the coating during the scratch was good with small non-essential cracks of the Ti-
excellent, i.e. the adhesion is excellent. The based coating. This difference in the scratch
coefficient of friction at the end was 0.089. resistance is because of the better adhesion
The deposition of Ti–based and Cr-based strength and coefficient of friction of the CrN
quaternary structure at the same technological layer than the TiN one. No delamination was
conditions allows the comparative study of the observed in both scratch tracks, which implies the
mechanical behavior resulting from the good adhesion to the substrates.
interactions between the Ti and Cr layers and the
HSS substrate, as well as from the influence of Table 2. Comparison of
TiN/ TiCrAlN and CrN/CrTiAlN transition layers. the Scratch results
The N2 flows during the main coating layers
Layer Coefficient Scratch Adhesion
deposition were 21.2 sccm and 21 sccm for Ti- of friction resistance
based and Cr-based coating, respectively. Thus, it to
can be consider that these layers were deposited at progressive
the same nitrogen flow. load
The dependence of the hardness and TiCrAlN 0.167 Very good yes
elastic modulus on the loading is presented in Fig. CrTiAlN 0.089 Excellent yes
6. The mechanical parameters were measured in
two characteristic points 15 mN and 200 mN,
corresponding to the depth penetration giving 4. Conclusion
information about the coating hardness itself and The investigated quaternary Ti- and Cr-
the hardness influenced by the substrate. It is seen based coatings consist of an adhesition, transition
that no significant difference of the coating and main layers. The technological parameters
hardness were observed at 15 mN loading. It is during the main layer deposition were set the
expectable because there are no differences in the same for both coating structures.
technological parameters at which these layers The study of the mechanical properties of
were deposited. With the increase the load the both structure types showed that the same
composite hardness decreases due to the increased deposition conditions affect the same coating
influence of the substrate. hardness. However, the different composition of
the adhesion and transition layers lead to different
adhesion to the substrate.
Because of the better scratch properties of
the Cr/CrN interlayers the Cr/CrN/CrTiAlN
structure has lower friction coefficient of 0.089
and excellent scratch resistance to loads up to 30
N, which defines it as more suitable for industrial
applications.
REFERENCES
1. Cakan, A., Ozkaner V. and Yildirim M. M.
(2008). Quantifying Cutting and Wearing
Behaviors of TiN- and CrN Coated AISI
Fig. 6. Dependence of the nanohardness and 1070 Steel. Sensors, volume(8), 6984-
elastic modulus of the Ti- and Cr-based coatings: 6998.
1- Nanohardness of Ti/TiN/TiCrAlN coating, 2. Chauhan, K. V. and Rawal, S. K. (2014). A
2- Nanohardness of Cr/CrN/CrTiAlN coating, review paper on tribological and mechanical
3- Module of elasticity of Ti/TiN/TiCrAlN coating, properties of ternary nitride based coatings,
4- Module of elasticity of Cr/CrN/CrTiAlN coating. Procedia Technology, volume (14), 430-
437.
The scratch parameters for both coatings 3. Chim Y.C., Ding X. Z., Zeng X. T. and Zhang
S. (2009). Oxidation resistance of TiN, CrN,
are presented in Table 2. It can be seen that the
Cr-based coating has a better coefficient of TiAlN and CrAlN coatings deposited by
II-231
lateral rotating cathode arc, Thin Solid Films, 10. Wu W., (2010) Development and
volume (517), 4845–4849. characterization of novel low-friction wear-
4. Zhanga, G.A., Yana, P.X., Wangb P., resitant multilayer nanocomposite CrAlTiCN
Chenb, Y.M. and Zhang J.Y. (2007). The coatings. Ph.D. thesis, University of
structure and tribological behaviors of CrN Birmingham, 2010.
and Cr–Ti–N coatings. Applied Surface 11. Zhou Z., Tam P., Shum P., Li K., (2009)
Science, volume (517), 7353–7359. High temperature oxidation of CrTiAlN hard
5. Lee, J.K., Oh, M.H. and Wee D.M. (2001). coatings prepared by unbalanced magnetron
Long-term oxidation properties of Al–Ti–Cr sputtering, This Solid Films, volume (517),
two-phase alloys as coating materials for 5243-5247.
TiAl alloys. Intermetallics, volume (10), 12. Kim Y., Lee H., Kim S.and Han J., (2007)
347-352. Microstructural and Mechanical Properties of
6. Bai L., Zhu X., Xiao J. and He J. (2007). CrTiAlN Nanocomposite Thin Films
Study on thermal stability of CrTiAlN Synthesized by Closed Field Unbalanced
coating for dry drilling. Surface & Coatings Magnetron Sputtering, Journal of the Korean
Technology, volume (201), 5257–5260. Physical Society, volume (51), 1183-1186.
7. Kabir, M.S., Munroe, P., Zhou, Z. and Xie
Z.(2016), Structure and mechanical
properties of graded Cr/CrN/CrTiN coatings
synthesized by close field unbalanced
magnetron sputtering, Surface & Organization:
Coatings Technology, volume (309), 779- Central Laboratory of Applied Physics,
789. Bulgarian Academy of Sciences
8. Danek M., Fernandes F., Cavaleiro, A., Address: Bulgaria, Plovdiv, 4000,
Polcar T., (2017) Influence of Cr additions 61, Sankt Petersburg Bld.
on the structure and oxidation resistance of Phone: 00359(0) 893611035
multilayered TiAlCrN films, Surface & 00359 (0) 32 265515
Coatings Technology, volume (313), 158- E-mails: vchitanov@gmail.com,
167. ohmic@mbox.digsys.bg,
9. Saleem H., (2014) Thermal stability of cubic ipfban@mbox.digsys.bg,
and nanocrystalline arc evaporated TiCrAlN ipfban-dve@mbox.digsys.bg
coatings, Master’s Thesis, Department of
Physics, Chemistry, and Biology, Linköpings
universitet.

TOPOLOGICAL FEATURES OF PHOSPHATE TREATED

SPECIMENS OF PRELIMINARY CERIUM OXIDE COATED
AIRCRAFT ALLOY
STEPHAN KOZHUKHAROV, CHRISTIAN GIRGINOV
Abstract: The highly doped constructive aluminum alloys are from crucial importance for
the aircraft, automobile and marine transport. These alloys, however, are very susceptible
to localized corrosion, due to occurrence of various intermetallic inclusions, which form
galvanic elements with the basic Al matrix. That is the reason for the necessity for reliable
corrosion protective systems development. In this sense, the present brief research
represents the results of the finishing phosphate sealing of cerium oxide coated aircraft
alloy.
Key words: aircraft alloy, anodizing, cerium oxide coating, phosphatation, SEM, EDX
1. Introduction In addition, the nowadays environmental

Aluminum (Al) is a lightweight, relatively restrictions impose new challenges regarding the
easily treatable metal that possesses an aptitude for elaboration of environmentally benign efficient
passivation by formation of a natural oxide layer. corrosion protective coatings [9].
Nevertheless, the pure Al is inapplicable in the In this sense, the aim of the present research
industrial practice, since it does not present work is to determine the impact of the subsequent
satisfying mechanical properties. phosphating on cerium oxide coated anodized
Recently, besides in the aircraft industry [1- specimens of aircraft alloy.
3], the aluminum alloys have found continuously
2. Experimental
increasing applications for car body panels [4], and
even in the modern shipbuilding [5, 6]. Besides, it is The experimental procedures were
a common practice to apply multilayered coating performed with nine square coupons with
systems for corrosion protection of metallic details, dimensions 30 x 30 mm (3 mm thick) A2024-T3
assemblies, and entire vehicle constrictions [7]. aircraft alloy. All samples underwent preliminary
The remarkably heterogeneous chemical treatment for removal of the temporal corrosion
composition of these alloys causes galvanic protective films. It was performed by subsequent
coupling between the intermetallics and the basic etching in 50 g dm-3 of NaOH solution at 60 °С and
Al-matrix, which gradually converts to localized activation in diluted with water 1:1 v/v HNO3 at
corrosion phenomena, as is established by D. room temperature. Both operations were done for 2
Balgude, A. Sabnis, [8]. minutes each. The specimens were then cleaned
with tap and distilled water after each treatment.
The anodizing was performed in galvanostatic
regime at 15mA cm-2 for 1 hour with intensive
stirring in 15% H2SO4. The already anodized
samples were subsequently dip coated in 150 ml of
mixed cerium ions containing solution.
Following the main objective of the present
research, the already anodized and coated samples
were divided in two groups: (i) reference samples,
Fig. 1. Schematic representation of (a) conventional without additional treatments and (ii) additionally
chromate-based coating system, (b) newly phosphated samples;
developed sol–gel based films and (c) The further phosphatation procedures were
environmentally friendly chrome-free carried out as follows: immersion in 0.3 mol dm-3
super primer technology [8] (NH4)H2PO4 for 12 minutes at 60 °С.

II-233
In order to perform systematical

investigations, all specimens were submitted to
methodical research activities, which were divided Fig. 2. Low resolution topographical SEM
into two groups: images of the investigated samples (a) reference
Structural and topographical observations, CeCC coating; (b) phosphated CeCC coating; (c)
including Scanning Electron Microscopy (SEM), CeCC coating after phoisphatation and subsequent
Compositional analysis, by Energy thermal treatment
Dispersion X-ray (EDX) spectroscopy.
The investigations were performed by The films reveal identical morphology,
TESCAN, SEM/FIB LYRA I XMU working at 30 kV. being strongly cracked possessing a multitude of
The SEM observations were combined by elemental concavities. Nevertheless, the films maintain
analysis, with energy dispersion spectroscopy excellent coverage and uniformity. The film
(EDX), using Energy Dispersion Spectrometer uniformity was confirmed by the element
Quantax 200 (BRUEKER). distribution, established by the EDX map analysis,
shown in Fig. 3. The main film elements selected
3. Results and discussion for observation were: oxygen from the anodization
The low resolution SEM images in Fig. 2 process, cerium from the subsequent cerating and
show comparable film surfaces, revealing lack of phosphorus from the finishing treatments.
distinguishable impact of the subsequent
phosphatation and thermal finishing on the film
morphology.
II-234
Acknowledgements: The authors are grateful for

the funding of this research to the Bulgarian
National Scientific Research Fund, under contract
DFNI-T02-27.
LITERATURE
1. Wildey, J. F., Aging Aircraft, Materials

Performance, 1990, 29, 80–85
2. Komorowski, J. P., et al., Quantification of
Corrosion in Aircraft Structures with Double
Pass Retroreflection, Canadian Aeronautics
and Space Journal, 1996, 42, 76–82
3. Starke E. A., Jr., and J. T. Staley, Application
of Modern Aluminum Alloys to Aircraft, Prog.
Aerospace Sci., 1996, 32, 131-172
4. Davies G., Materials for Automobile Bodies,
Linacre House, ed. (2003), ISBN 0 7506 5692
1, pp. 146 – 156.
5. Eyres D., “Ship construction”, 6th ed. 2007,
Linacre House, p. 50–52, ISBN 10:0- 75-
068070-913.
6. “Aluminium in shipbuilding”, access via:
http://www.aluminiumleader.com/en/around/tr
ansport/ship 7.
7. G. Tsaneva, V. Kozhukharov, S. Kozhukharov,
M. Ivanova, J. Gerwann, M. Schem, T.
Schmidt, Functional nanocomposite coatings
for corrosion protection of aluminum alloy and
steel, J. Univ. Chem. Technol. Met. 43–42
(2008) 231–238.
8. D. Balgude, A. Sabnis, Sol–gel derived hybrid
coatings as an environment friendly surface
treatment for corrosion protection of metals
and their alloys, J. Sol-Gel Sci. Technol. DOI
10.1007/s10971-012-2838-z
9. Kozhukharov, S. V. (2016) Deposition of
Fig. 3. EDX map analyses of the element Environmentally Compliant Cerium-
distribution of the investigated samples (a) Containing Coatings and Primers on Copper-
reference CeCC coating and (b) phosphated CeCC Containing Aluminium Aircraft Alloys,
coating PUBLISHED IN: Biobased and Environmental
Benign Coatings (eds A. Tiwari, A. Galanis
4. Conclusions and M. D. Soucek), John Wiley & Sons, Inc.,
Hoboken, NJ, USA.
The proposed approach in the present brief doi: 10.1002/9781119185055.ch2
research of subsequent combination of anodizing,
cerium oxide coating and finishing phosphating
enables formation of uniform coating primers with Eng. Stephan Kozhukharov, Ph.D.
remarkable coverage and equal coating ingredient’s Department of Analytical Chemistry
distribution of: University of Chemical Technology and
- Oxygen from the anodizing procedure Metallurgy, Sofia, Bulgaria
- Cerium from the subsequent cerium oxide 8 Kliment Ohridski blvd., 1756 Sofia,
coating Bulgaria
- Phosphorous from the finishing coating +359 899 837282
sealing stephko1980@abv.bg
II-235
8 Kliment Ohridski blvd., 1756 Sofia,

Assist. Prof. Eng. Christian Girginov, Ph.D. Bulgaria
Department of Physical Chemistry +359 899 293908
University of Chemical Technology and girginov@uctm.edu
Metallurgy, Sofia, Bulgaria
SENSORS MASS INFLUENCE ON THE

NATURAL FREQUENCY OF A CANTILEVER
BEAM
SVETLIN STOYANOV
Abstract: The influence of relation between the sensor mass and the beam mass to the
natural frequencies of a cantilever beam is investigated theoretically. This influence is not
simple, because it depends from the mode number, and a three dimensional plot is used to
present the results.
Key words: natural frequency, cantilever beam, modal analysis, Fourier transform,
Matlab.
1. Introduction where L is the beam length. The values of k for

Over recent decades, modal analysis has which this equation is satisfied are his characteristic
been widely used in aerospace, mechanical and civil numbers. Each of the characteristic numbers
engineering. One of its advanced applications is for corresponds to one natural frequency :
damage detection in beams [1].
The experimental modal analysis makes EI
use of input (excitation) and output (response) n  kL2 , (2)
mL4
measurements to estimate modal parameters:
natural frequencies, mode shapes, and damping where m is the mass of the one meter of the beam.
ratios [2]. Usually, the input is applied through an The acceleration sensor acts as lumped
impulse hammer. mass in the end of the cantilever beam. As a result
The operational modal analysis carries out of this, the cantilever beam is loaded with a constant
in operational environment. In this case, structures force – the gravitational force of the sensor.
are exited naturally due to external dynamic forces Therefore, the differential equation of the forced
or displacements. Therefore, the information about vibration of the beam is used for the frequency
these input excitations is absent or incomplete [2, 3, equation derivation. The frequency of the distortion
4]. force is set to be zero and the frequency equation
The theoretical modal analysis tries to becomes [7, 5]:
predict the values of the modal parameters by
solving the eigenvalue problem of structures [3].
m1
In [5], theoretical and experimental modal AkL  kL kL BkL
analysis of a cantilever beam is conducted and the mL
0, (3)
results are compared. It is raised the problem about m1
the influence of sensor mass on the beam natural DkL  kL C kL AkL
mL
frequencies. This justifies the aim of this work – to
investigate theoretically the influence of relation where:
between the sensor mass and the beam mass to the
natural frequencies of a cantilever beam. 1
AkL  coshkL  coskL , (4)
2. Theoretically obtaining of the natural 2
frequencies
1
The frequency equation of a cantilever BkL  sinhkL  sinkL , (5)
beam without a lumped mass added (i.e. without 2
sensor mass accounting) is [6]
1
coskLcoshkL  1  0 , (1)
C kL  coshkL  coskL , (6)
2
Copyright©2017
University - Sofia,
of Sofia, Plovdiv
Plovdiv branch,
branch, Bulgaria
Bulgaria ISSN 1310
ISSN Online: - 8271
2535-0048
II-237
1 the spectrograms are calculated using Fourier

DkL  sinhkL  sinkL (7) transform (fig. 3).
2
are the Krilov’s functions and m1 is the mass of the
sensor.
3. Experimental evidence of sensor mass
influence
In [5], an experimental setup is created (fig.
1) and the acceleration of the free vibration of the
end point of a cantilever beam is measured as a
function of the time (fig. 2). From this time signal,
Bending oscillationsFig. 1. A
around photography
z axis of the experimental setup
40
30
20
10
a, m/s2
-10
-20
-30
-40
0.4 0.5 0.6 0.7 0.8 0.9 1
t, s
Fig. 2. Time-diagram of the experimentally obtained acceleration of the end point of the beam
Bending oscillations around z axis - 26.2451Hz, 167.8467Hz, 463.2568Hz
3.5
2.5
2
a, m/s2
1.5
0.5
0
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
f, Hz
Fig. 3. The experimentally obtained acceleration (red line) and the theoretically predicted natural
frequencies without sensor mass accounting (blue marker lines)
The peak points of the red line on Fig. 3 frequencies are shown in the second row of Table 1.
indicate the frequencies fexp of the harmonics The third table row presents the relative difference.
contained in the measured acceleration. These Similarly, Fig. 4 and Table 2 show the
frequencies are shown in the first row of Table 1. corresponding frequencies for the case with sensor
The blue straight lines marks frequencies ft, this is mass accounting, according to Eq. 3. One can
theoretically calculated without sensor mass observe significant better results in this case. This
accounting, according to Eq. 1 and Eq. 2. These justifies the need for a detailed investigation.
II-238
Table 1. Numerical comparison of the case without Table 2. Numerical comparison of the case with
sensor mass accounting sensor mass accounting
fexp, Hz 26.25 167.85 463.26 Fexp, Hz 26.25 167.85 463.26
ft 32.173 201.85 564.562 ft 27.55 178.64 510.89
(fexp-ft)/fexp, % -22.6 -20.1 -21.9 (fexp-ft)/fexp, % -5.0 -6.4 -10.3
Bending oscillations around z axis - 26.2451Hz, 167.8467Hz, 463.2568Hz
3.5
2.5
2
a, m/s2
1.5
0.5
0
0 100 200 300 400 500 600 700 800 900 1000 1100 1200
f, Hz
Fig. 4. The experimentally obtained acceleration (red line) and the theoretically predicted natural
frequencies with sensor mass accounting (blue marker lines)
4. An investigation of sensor mass influence Fig. 6 presents the influence of the mode
To evaluate the influence of sensor mass on shape number to the frequency relation factor for
natural frequencies, two quantities are defined as six values of the mass relation factor. One can
follows: observe that the influence grows with sensor mass
increasing.
- mass relation factor
The relations from Fig. 5 and Fig. 6 are
combined together in a three dimensional plot – Fig.
m1
 (8), 7.
m
5. Conclusion
- frequency relation factor An experimental evidence of significant
sensor mass influence on the cantilever beam
f1 natural frequencies is presented. This justifies the
 (9),
f necessary to investigate theoretically this influence.
The investigation is done through the frequency
where f1 is the natural frequency for the case of equation describing the eigenvalue problem.
sensor mass accounting (according to Eq. 3) and f The frequency equation and the equation of
is the natural frequency for the case without sensor the forced vibrations of the beam are solved
mass accounting (according to Eq. 1 and Eq. 2) numerically by the help of the software system
Fig. 5 presents the influence of the mass Matlab. The results obtained are presented and
relation factor to the frequency relation factor for discussed. The influence of the sensor mass is not
the first three natural frequencies. As can be seen simple, because it depends from the mode number,
from this figure, the influence is the greatest for the and a three dimensional plot is used to present the
first mode (red line). For the second mode and for results. It was found that the influence is the
the third mode, when the mass relation factor is greatest for the first mode. Furthermore, it is of
greater than about one, the relation is close to a interest to obtain theoretically and experimentally
straight line, i.e. the change of the frequency the natural frequencies of a truss structure.
relation factor can be neglected.
Copyright  2016 by Technical University - Sofia, Plovdiv branch, Bulgaria ISSN 1310 - 8271
II-239
1 1
m1 = 0.05m
0.8 0.8
m1 = 0.1m
m1 = 0.5m
0.6 0.6


Mode 1 m1 = 1m
Mode 2 m1 = 2m
0.4 Mode 3 0.4 m1 = 3m
0.2 0.2
0 0.5 1 1.5 2 2.5 3 1 2 3 4 5 6 7
 Mode number
Fig. 5. Dependences of the frequency relation Fig. 6. Dependences of the frequency relation
factor on the mass relation factor factor on the mode number
0.8
0.6

0.4
0.2
8
6 3
2.5
4 2
1.5
2 1
0.5
0 0
Mode mumber 
Fig. 7. Three dimensional representation of the frequency relation factor dependence
REFERENCES 5. Stoyanov, S. (2016) ‘Experimental setup for

the determination of the natural frequencies of
1. Yang Z., Radzienski M. and Kudela P., a cantilever beam’, Scientific proceedings of
Ostachowicz. (2017) ‘Fourier spectral-based University of Ruse, Ruse, ISSN 1311-3321. (In
modal curvature analysis and its application to Bulgarian. In press.)
damage detection’, Mechanical Systems and 6. Zannon M. (2014) ‘Free Vibration of Thin
Signal Processing, vol. 84, pp. 763-781. Film Cantilever Beam’, International Journal
2. De Vivo A., Brutti C. and Leofanti J.L. (2013) of Engineering and Technical Research
‘Modal shape identification of large structure (IJETR), vol. 2, pp. 304-314.
exposed to wind excitation by operational 7. Repetto S. E., Roatta A. and Welti R.J. (2012)
modal analysis technique’, Mechanical ‘Forced vibrations of a cantilever beam’,
Systems and Signal Processing, vol. 39, pp. European Journal of Physics, vol. 33, pp.
195–206. 1187-1195.
3. Aenlle M.L. and Brincker R. (2013) ‘Modal
scaling in operational modal analysis using a
finite element model’, International Journal of Authors’ contacts
Mechanical Sciences, v76, pp. 86-101. Address: 36 Zahari Stoianov str.,
4. Dziedziech K., Staszewski W.J. and Uhl T. POB 7005, Ruse, Bulgaria
(2015) ‘Wavelet-based modal analysis for Phone: 887-549-965
time-variant systems’, Mechanical Systems and E-mail: sstoyanov@uni-ruse.bg
Signal Processing, vol. 50-51, pp. 323–337.
© International
International Scientific Conference
Conference on Engineering,
on Engineering, TechnologiesTechnologies
and Systemsand Systems
TECHSYS
TECHSYS2017, Technical
2017, TechnicalUniversity ––
University Sofia,
Sofia,Plovdiv
Plovdivbranch
branch
18 –18
20–May 2017,
20 May Plovdiv,
2017, Bulgaria
Plovdiv, Bulgaria
THE APPLICATION OF THE MODAL ANALYSIS IN

DETERMINING THE DYNAMIC
MODULE OF ELASTICITY OF
POLYMER-CONCRETE COMPOSITES
ILIA POPOV, SABI SABEV
Abstract: This article presents the quantitative values of the dynamic modulus of elasticity
(Young's modulus) of gamma polymer-concrete composites. Experimental studies are based
on the experimental modal analysis method. Using the similarity and comparison methods,
the results of the experimental determination of the dynamic modulus of elasticity of
samples of gray cast iron GF20 and C45 steel with the same dimensions as PC samples are
presented.
Keywords: Polymer-concrete Composites, Dynamic Modulus of Linear Deformations,

Experimental Modal Analysis.
1. Introduction research laboratory of Metal cutting-machines at

Knowing the material constants of structural TU-Sofia, branch Plovdiv.
engineering materials is a must when defining the
strength deformation behaviour of the parts and
structures produced by these materials. 2. Theoretical prerequisites
In this aspect, the study and the determination Theoretically, natural beam frequencies with a
of the quantitative values of the elastic rectangular cross section are determined by the
characteristics (such as the modulus of the linear dependence [2]:
deformations E, the modulus of the angular
deformations G, and the Poisson μ coefficient, Ed I Ed I
which are in simple algebraic relations) of the  
n
2
(  n l )2 (1)
A  Al 4
gamma polymer-concrete composites is an up-to-
date engineering task. These features, as well as the
From equation (1) Ed is expressed and the
density of the material, are input data for any
software product for static or dynamic engineering following is obtained:
analysis (CAD, CAM CAE), which greatly assists
in designing and enables reliable results to be 2 b h l 4
Ed  (2)
obtained quickly. (  n l )4 I
The application of the polymer-concrete (PC)
composites as an alternative non-metallic structural Where
material for the production of bodies and body parts
(B and BP) of the support system of the production  - is the frequency of vibration
equipment is an innovation activity which creates b - beam thickness
prerequisites for improving the technical and h - height of the material
operational qualities of the manufactured machines  - density of the material
such as:
 static stability; n - density of the material
 dynamic sustainability; I - beam moment of inertia around y
 thermal behaviour. (  nl ) - depends on the boundary conditions of
the beam.
The article has an experimental-research
character and its main objective is: Quantification ( 1l 4.730;
 2l 7.853;
 3l 10.996)
of the dynamic elastic modulus Ed of gamma PC
composites, created and realized in the testing and
II-241
The assembly of measuring equipment includes:
 Microphone of the company "Audio

Technica" - AT2031;
 Sound card M-Audio Audiophile 192
 Specialized software "Spectra PLUS".
Fig. 1.1. Forms for bending a rectangular beam with
designated detail lines Experimental results for the dynamic modulus
of linear deformations Ed of PC samples are
obtained by pulse excitation, Fig. 1.2 and processed
3. Methodology of the experiment with the specialized software "Spectra PLUS".
By impulse (shock) excitation of the system SpectraPLUS is a powerful 2-channel spectral
with a subsequent analysis of free damping analyser. Software interfaces provide real-time
oscillations. The impulse method is realized with an spectral analysis. Spectrum analyser is a tool used
unattached exciter - an impact (power) hammer. to convert a time domain signal (amplitude-time)
Thus, the generated energy is distributed over a into a frequency domain (amplitude-frequency).
wide frequency spectrum and excites all the mods The Audio Spectrum Analyser, by definition, is
in the considered frequency range of the sample limited in signal bandwidth processing. The limit of
[4,5]. the analysed frequency is determined by the
For the quantification of the dynamic elastic capabilities of the sound card used.
The software uses Fast Fourier Transform
modulus, the following experimental approach was
(FFT), which converts the signal from the time
used: - A test piece of dimensions 30x30x350 of PC domain into the frequency domain.
is suspended freely in strings, fig.1.2. At one end of
the beam the impulse dynamic effect is applied and
at the opposite side at the other end the repercussion
is measured by the microphone.
The method is easy to implement, and only
requires some basic components: a computer with a
suitable sound card and a microphone.
Fig.1.3. Test bench
Fig. 1.2. Transverse vibration study

II-242
Table 1 Experimental results 4. Experimental results
Sample f [Hz] r [m3 ] ω[rad] E [Pa]
d
The obtained experimental results for the
953 5985 3,61E+10 dynamic modules of the linear deformations Ed of
1 2537,5 2381 15936 3,37E+10 the PC composites are presented in Table 1. In
4751 29836 3,07E+10 Figure 1.4 these are represented graphically.
866 5438 2,93E+10
2 2311 2339 14513 2,74E+10
4294 26966 2,46E+10 С 45
977 6136 3,77E+10 FG20
3 2576 2365 16177 3,44E+10 15
4817 30251 3,13E+10 14
899 5646 3,17E+10
13
4 2385 2355 14978 2,94E+10
12
4456 27984 2,67E+10
11
815,6 5122 2,52E+10
10
Sample
5 2165 2268 13596 2,33E+10
4044 25396 2,12E+10 9
761,7 4783 2,11E+10 8
6 2027 2181 12730 1,97E+10 7
3792 23814 1,79E+10 6
767 4817 2,22E+10 5
7 2129 2258 13370 2,25E+10 4
3398 21339 1,49E+10 3
799 5018 2,36E+10 2
8 2137 2216 13420 2,22E+10
1
3958 24856 1,98E+10 1,00E+00 5,00E+10 1,00E+11 1,50E+11 2,00E+11
839 5269 2,70E+10 dynamic modulus
9 2184 2300 13716 2,41E+10
4090 25685 2,20E+10 Fig.1.4.. Experimental results for Ed
864 5426 2,92E+10
10 2319 2348 14563 2,77E+10 We obtain the values, in which we are
4344 27280 2,53E+10 interested, of the sampling frequencies of the
815 5118 2,49E+10 experimental PC composites from the auto-spectral
11 2184 2252 13716 2,36E+10 characterization, fig.1.5 and 1.6.
4068 25547 2,13E+10
785 4930 2,34E+10
12 2114 2280 13276 2,24E+10
3980 24994 2,06E+10
858 5388 2,88E+10
13 2281 2342 14325 2,67E+10
4250 26690 2,42E+10
939 5897 3,51E+10
14 2504 2387 15725 3,29E+10
4688 29441 3,00E+10 Fig.1.5. Autospectrum of sample 1
788 4949 2,33E+10
15 2105 2245 13219 2,18E+10
3937 24724 1,99E+10
1001 6286 1,20E+11
FG20 2643 7203 16598 1,10E+11
4891 30715 9,84E+10
1262 7925 2,07E+11
С 45 3342 7803 20988 1,91E+11
6225 39093 1,73E+11
Fig.1.6. Autospectrum of sample С45
3
II-243
The theoretical verification of the first three 5. Conclusion
forms (own frequencies) of the lateral oscillations
of the 15 composites was performed using the The results of this work are limited to:
dynamic simulation analysis module of SolidWorks • Quantitative values for the dynamic elastic
2014, fig.1.7. The results are shown in Table 2. modulus of 15 different PC composites were
obtained experimentally.
• By increasing the sequence of the sample's
own resonance frequency, the module decreases.
• The method of experimental modal analysis
is used with priority in the quantitative
determination of the dynamic modulus of elasticity
of the range of PC composites.
• The dynamic elasticity modulus of PC
composites is determined on the basis of the
dynamic response (response) of the free damping
oscillations of the impulse excited dynamic system
of the experimental samples.
• Opportunity to obtain reliable and realistic
information about simulation shapes and own
oscillation frequencies of each PC composites
through SolidWork Simulation Module has been
demonstrated.
• By a comparative analysis of the conducted
tests and the results obtained for the elastic
characteristics of the samples of gray cast iron
BS20, the steel C45 and the PC composites it was
found that the modulus of linear deformations is
one level lower.
REFERENCES
Fig. 1.7 Graphics of the first three own forms of 1. S. Timoshenko, D. H. Young, and W. Weaver,
vibration for the transverse vibrations of sample 1. Jr., Vibration Problems in Engineering (4th
ed.), Wiley, New York, 1974.
Table 2 Modal frequencies
2. Singiresu S. Rao, Mechanical Vibrations 5th
Measured values Simulated values Ratio to Percentage
number
Sample
f f1 f f2 f f3 f S1 f S2 f S3 f f1 /f S1 f f2 /f S2 f f3 /f S3 Edition, University of Miam, 2011.

[Hz] [Hz] [Hz] [Hz] [Hz] [Hz] [%] [%] [%] 3. L. Klein, Transverse vibrations of non-
1 953 2538 4751 956 2526 4692 100% 100% 101% uniform beams, Journal of Sound and
2 977 2311 4294 868 2296 4268 113% 101% 101% Vibration,Vol. 37, 1974.
3 899 2576 4817 980 2591 4814 92% 99% 100% 4. Ulbrich H. Vibration, Control of Nonlinear
4 870 2385 4456 900 2381 4428 97% 100% 101%
Mechanism and Structure. Technical
5 816 2165 4044 819 2166 4030 100% 100% 100%
University, Munich, 2005.
6 762 2027 3792 764 2023 3766 100% 100% 101%
7 797 2129 3398 770 2039 3795 104% 104% 90%
5. Попов Ил., „Изследване и оптимизиране на
8 799 2137 3958 801 2122 3946 100% 101% 100% полимербетонни състави по отношение на
9 839 2184 4090 842 2227 4144 100% 98% 99% демпферните им характеристики”,
10 864 2319 4344 866 2292 4265 100% 101% 102% Докторска дисертация, (2013).
11 815 2184 4068 816 2159 4013 100% 101% 101%
12 785 2114 3980 786 2080 3869 100% 102% 103%
13 858 2281 4250 859 2275 4232 100% 100% 100%
14 939 2504 4688 938 2478 4600 100% 101% 102% Адрес : ТУ-София, филиал Пловдив
15 788 2105 3937 786 2081 3871 100% 101% 102% E-mail: ilgepo@abv.bg
FG20 1001 2643 4891 998 2636 4889 100% 100% 100% E-mail: sabi_sabev@abv.bg
ЧИСЛЕНО МОДЕЛИРАНЕ И ПАРАМЕТРИЧЕН

АНАЛИЗ НА КАЛИБРОВЪЧНА ДВУСЛОЙНА
РОЛКА
ИВО ДРАГАНОВ, ЮЛИЯН АНГЕЛОВ
Резюме: Създаден е крайноелементен модел на калибровъчна двуслойна ролка от

стенд за проверка на ходовата и свързаните с нея системи на автомобил. Чрез
решаване на контактна задача е определено изменението на периферната
повърхнина на ролката в зоната на контакта с автомобилната гума, като е
отчетено нелинейното поведение на покриващия слой от ебонит. Извършен е
параметричен анализ за влиянието на дебелините на ебонитовия и стоманения слой.
Ключови думи: калибровъчна двуслойна ролка, метод на крайните елементи
NUMERICAL MODELING AND

PARAMETRICAL ABALYSIS OF CALIBRATED
DOUBLE LAYERED ROLL
IVO DRAGANOV, YULIAN ANGELOV
Abstract: Finite element model of calibrated double layered roll for testing of automobile’s
transmission is created. The change of roll’s external diameter in the tire’s area is
calculating using contact algorithm. The nonlinear behavior of covering ebonite layer is
assumed. The parametrical analysis for influence of ebonite’s and steel’s thickness is made.
Key words: calibrated double layered roll, finite element analysis
1. Основни положения материалните характеристики на ебонитовия

Калибровъчните ролки намират редица слой, върху напреженията и деформациите, е
приложения, като обект на изследване в целесъобразно да се използва методът на
настоящата работа е такава, използвана в крайните елементи (МКЕ) [5].
стендове за изпитване на спирачна ефективност
на автомобили [1], за проверка на таксиметрови 4
апарати [2] и за други устройства, за които е 1
необходимо да се отчитат оборотите на колелата
2
на автомобила. Компановъчна схема на подобен
стенд е дадена на Фиг. 1 [3], [4].
Двуслойните ролки се състоят от 3 5
стоманена основа, даваща нужната коравина на
конструкцията и вулканизиран ебонитов слой,
чиято функция е да се намали преплъзването,
между ролката и гумата на автомобила. Фиг. 1. Местоположение на ролката в стенд
Калибровъчните ролки влизат в за проверка на таксиметрови апарати (1 –
устройството на различните по вид стендове, електродвигател; 2 – верижна предавка; 3 –
като изчислителната им схема е греда на две редуктор; 4 – задвижваща ролка; 5 –
упори с приложена сила в средата [4]. За да се калибрираща ролка)
отчете по-прецизно влиянието на геометрията и
II-245
Ебонитовият слой, с който е покрита еднокамерни гуми. Прието е, че съществено

стоманената основа, има нелинейни материални влияние върху резултатите ще окаже ширината
характеристики, описани от Давие [6]. на гумата, като е прието тя да е 155 mm, с радиус
Деформациите на покритието са най- на закръгление 20 mm. Поради симетрията се
големи в зоната на контакт между колелото на разглежда една четвърт от гумата.
автомобила и ролката, което налага решаването
на контактна задача [7]. 2.2. Материални характеристики
Материалът, от който се изработват
2. Крайноелементен модел на заварените детайли на ролката, е стомана S235 с
калибровъчната ролка якост σB = 360 MPa [9].
За изследване на напрегнатото и Прието е, че стоманената част на ролката
деформирано състояние на ролката е използвана и автомобилната гума ще действат само в
програмна система Абакус (ABAQUS) [8]. границата на еластичност и връзката между
напреженията и деформациите ще е линейна. В
2.1. Геометричен модел таблица 1 са дадени използваните материални
Необходимостта от решаване на характеристики.
контактна задача налага построяването на
Таблица 1. Линейни материални
геометричен модел на ролката и на част от гумата
характеристики
на автомобила.
Поради наличието на симетрия в Материал Модул на Юнг, Коефициент на
надлъжно и напречно направление е разгледана GРа Поасон, [-]
една четвърт от ролката – Фиг. 2. Построяването
Стомана 200 0,27
на геометричния модел е извършено в средата на
програмна система Абакус. Външният диаметър Гума
на ролката, диаметърът между ебонитовия и 0,01 0,45
стоманения слой и вътрешният диаметър на Ебонитът представлява материал с
стоманения слой приемат стойности, които могат нелинейна връзка между напрежения и
да се видят на Фиг. 6. Останалите размери са деформации и тъй като се очаква деформациите
избрани съобразно компановъчната схема и в него да окажат най-съществено влияние върху
някои други съображения [10], [11], [12]. изследваните параметри, за него се решава
физически нелинейна задача с материални
650 характеристики дадени в таблица 2. Прието е, че
10 коефициентът му на Поасон е 0,39.
Таблица 2. Материална нелинейност на

4 ебонит
Точка Напрежение, Дeформация, [-]

МРа
1 3 0
48 692
Ø48
Ø40
2 4 0,33
Dm
Dg
Dc
3 10 5
Фиг. 2. Геометрия на ролката (Dg – външен
диаметър на ролката, Dc – диаметър между 2.3. Гранични условия
ебонита и стоманата, Dm – вътрешен
Ролката е закрепена като са ограничени
диаметър на ролката)
всички степени на свобода по челото на шийката
Тъй като оебктът на изследване е ролката, на вала и са зададени равнините на симетрия –
геометричният модел на гумата е построен силно Фиг. 3. Гумата е закрепена като са ограничени
идеализирано, с цел да се минимизира степените на свобода в равнините на симетрия и
изчилсителният обем на задачата. По степените на свобода в нормално напрвление за
съществените допускания са следните: прието е, външната равнина на гумата.
че контактната повърхнина на гумата с ролката е В гумата е зададен разпределен товар в
равнина, не се отчитани наичието на грайфери по напречно направление с интензитет 0,3871 МРа,
гумата, дебелината на гумата е значително по- еквивалентен на приетата съсредоточена сила с
голяма от съществуващите в практиката
II-246
големина 3000 N – Фиг. 3, определена от

максималната товароносимост на стенда.
Фиг. 3. Гранични условия Фиг. 5. Премествания в радиално направление

2.4. Дискретизация На база получените стойности за
Дискретизацията на ролката и гумата е преместването е определен периметъра на
извършена с хексаедърни крайни елементи с контура - P’, за деформирантата ролка, по
един междинен възел (C3D20R) – Фиг. 4 [8]. формулата:
Направена е проверка за сходимостта на
решенията като са сравнени резултатите за n
изменението на външния диаметър на ролката P'   u ri  / n , (1)

i 1
при две гъстоти на мрежата – 13996 и 83596
възела. Разликата в резултатите е под 2%.
където ur е преместването в радиално
направление, а n е броя на точките в окръжно
направление, за които е определено
преместването.
В таблица 3 са дадени измененията за
отделните сечения при Dc = 171 mm и Dm = 159
mm, определени като разлика между началния
периметър, който е Р = 600,04 mm и периметъра
на деформиранта ролка.
Фиг. 4. Дискретизация с 13996 (в ляво) и 83586 Таблица 3. Промени на диаметрите на

възела (в дясно) ролката при Dc = 171 mm и Dm = 159 mm
2.5. Контакт
Отместено на
Деформираното състояние на ролката е

Средно
10 mm
30 mm
60 mm
определено чрез решаване на контактна задача за Сечение

взаимодействието между калибровъчната ролка
и автомобилната гума, без да се отчита триенето.
За по-добра работа на изчислителния
алгоритъм, е избрано гумата да се дискретизира с Периметър (Dg),
599,46
599,46
599,46
599,47
елементи по-големи, от тези на ролката [8]. mm

2.6. Резултати
Отчетено е преместването в радиално
Изменение, mm
0,58
0,58
0,58
0,57
направление на външните контури на средното и

на още 6 сечения, отместени в осово направление
през 10 mm - Фиг. 5.
II-247
3. Параметричен анализ от 16.12.2011 г. за периодичните прегледи

Създаденият крайноелементен модел е за проверка на техническата изправност на
използван за изследване влиянието на пътните превозни средства. София, 2011.
дебелините на ебонитовия и стоманения слой, 2. Министерство на транспорта. Наредба № 34
върху изменението на външния периметър на от 06.12.1999 г. за таксиметров превоз на
средното сечение на ролката. Резултатите от пътници. София, 1999.
параметричния анализ са дадени на Фиг. 6. 3. Trzeciak K. Diagnostika samochodow
osobowych. WKL, Warszawa, 2005.
1,2 4. Сестримски, Д. Диагностика на
автомобила. Техника, София, 1989.
Разлика в периметъра на
1
0,8
5. Иванов, И. Компютърно моделиране на
непрекъснати среди. Русенски университет,
ролката, mm
0,6
2004.
0,4 6. Davies, B. The Stress-Strain Relationship in
0,2 Ebonite. Transactions of the Institution of the
0 Rubber Industry. Vol. 9 No. 2, pages 130-140,
1 2 3 4 5 6 1933.
Дебелина на ебонитовия слой, mm 7. Wriggers, P. Computational Contact
dc=2 mm dc=4 mm dc=6 mm Mechanics. Springer, 2006.
dc=8 mm dc=10 mm 8. ABAQUS. Analysis User’s Manual. ver. 6.12,
Dassault Systemes Simulia Corp., Providence,
Фиг. 6. Параметричен анализ RI, USA.
9. БДС EN 10025-2:2005. Горещовалцувани
4. Анализ на резултатите и изводи продукти от конструкционни стомани.
Извършената проверка за сходимост на Част 2: Технически условия на доставка за
резултатите от задачата за определяне на нелегирани конструкционни стомани.
напрегнатото и деформирано състояние на 10. Ангелов, Ю., И. Борисов. Моделиране
калиброваща двуслойна ролка, показва изследване и оптимизация на ролков стенд
удовлетворяваща, за практическите нужди, за измерване на спирачна ефективност. Част
резултати. I. Моделиране и изследване. Научни трудове
Извършеният параметричен анализ на Русенски университет, том 52, серия 2,
показва, че изменението на периметъра на стр. 118-121, 2013.
ролката е по-чувствително към дебелината на 11. Ангелов, Ю. Моделиране изследване и
стоманения слой, отколкото към ебонитовия. оптимизация на ролков стенд за измерване
Получените резултати могат да се на спирачна ефективност. Част II.
използват както за качествена оценка, така и Параметрична оптимизация. Научни
непосредствено за избора на дебелина на трудове на Русенски университет, том 52,
ебонитовия слой в калибровъчна ролка за серия 2, стр. 122-125 , 2013.
автомобилни стендове с различно 12. Драганов, И., Ю. Ангелов. Моделиране на
предназначение. ролката на стенд за измерване на спирачна
В настоящата работа не е разгледано ефективност по метода на крайните
увличането, породено от триенето между елементи. Научни трудове на Русенски
автомобилната гума и ебонитовия слой. университет, том 53, серия 4, стр. 183-187,
2014.
5. Благодарности
Авторите на доклада изказват
благодарност на Фонд „Научни изследвания“ за гл. ас. д-р инж. Иво Драганов,
предоставените средства по договор № 2017- доц. д-р инж. Юлиян Ангелов
МТФ-02. Русенски университет „Ангел Кънчев“
Адрес: гр. Русе, ул. „Студентска“, №8
ЛИТЕРАТУРА Телефони: 082/888 572, 082/888 224
1. Министър на транспорта, информационните E-mail: iivanov@uni-ruse.bg,
технологии и съобщенията. Наредба № H-32 julian@uni-ruse.bg
ДИНАМИЧЕН СИНТЕЗ НА АГРЕГАТА НА

ЕЛЕКТРОМЕХАНИЧНА ПРЕСА ЗА
ПРОИЗВОДСТВО НА ПРОСЕЧЕНА МРЕЖА ПО
ЗАДАДЕНА СТЕПЕН НА НЕРАВНОМЕРНОСТ
ЗДРАВКО ВИТЛАРОВ
Резюме: Извършен е динамичен синтез на агрегата на електромеханична преса за
производство на просечена мрежа при зададена степен на неравномерност.
Определени са параметрите на едномасовия динамичен модел на агрегата. Решено е
диференциалното уравнение на движение и е определен необходимия маховик
реализиран като втора ремъчна шайба.
Ключови думи: динамичен синтез, маховик
DYNAMIC SYNTHESIS OF THE AGGREGATE OF

ELECTROMECHANICAL PRESS
FOR EXPANDED METAL AT A SET LEVEL OF INEQUALITY
ZDRAVKO VITLAROV
Abstract: Dynamic synthesis is made of the aggregate of electromechanical press for
expanded metal at a set level of inequality. Parameters of the mass dynamic model are
determined of the aggregate. The differential equation of motion is solved and it was
determined necessary flywheel realized like a second pulley.
.
Key words: dynamic synthesis, flywheel
1. Основни положения възбуждащи трептения и допълнителни

Съществуват три режима на движение на динамични натоварвания.
машинните агрегати – пусков, стационарен и От направения кинематичен синтез [1]
спирачен. За настоящата разработка интерес на параметрите на механизмите от
представлява стационарния режим на работа на електромеханична преса (фиг.1) и последващия
електромеханичната преса. За оценка на кинетостатичен анализ [2] даващ възможност да
отклонението на ъгловата скорост от средната се направи и якостно оразмеряване елементите
такава е въведена безразмерната величина δ на пресата (звена и кинематични двоици) е
степен на неравномерност: установена необходимостта от извършването и
на динамичен синтез за да се постигне
(1) зададената степен на неравномерност.
Решението на тази задача, определя масовия
инерционен момент на маховика, чрез
При работа на агрегата с постоянна поставянето на които се постига зададената
ъглова скорост (δ = 0) в него възникват степен на неравномерност.
инерционни сили, които възбуждат трептения и
създават динамични натоварвания на звената. 2. Изложение
Колебанията на ъгловата скорост ( ) Към степента на неравномерност δ се
създават допълнителни инерционни сили, предявяват изисквания, които се дефинират и от
II-249
изискванията за нормална работа на машината. моделиране на механизмите на машината за
В литературата съществуват известни производство на просечена мрежа. По принцип
емпирични данни за допустимите степени на при кинематичния синтез [1] и кинетостатичния
неравномерност (δ) – за металообработващи анализ [2] машината беше разделена условно на
машини от 1/30 до 1/40. три механизма. За да се получи масовия
инерционен момент на агрегата трябва да се
добави и задвижването (двигател и ремъчна и
верижна предавки). Ще приведем всички масови
и силови параметри към главния вал на агрегата
(фиг.1), където ще се разположи и маховика.
2.1.Механизъм за просичане (фиг.2)

Приведеният масов инерционен момент
към началното звено 1 ( главния вал на агрегата)
се определя от:
( )
( ) ( ) ( ) , (4)
фиг.1 Обща кинематична схема
Степента на неравномерност зависи от

масовите и силовите параметри на агрегата.
Тази зависимост може да се получи чрез
интегриране на диференциалното уравнение за
движение в границите, съответствуващи на
абсолютните екстремуми на скоростта. След Фиг.2 Кинематична схема на механизма за
известни преобразовани се получава: просичане
където:mi – маса на i-то звено (i=2, 3);
( ) ( )
JSi – масов инерционен момент на i-то
∫
(2) звено (i=1, 2);
( ) ( ) wi – ъглова скорост на i-то звено (i=1, 2);
vS2 – скорост на масовия център на звено
От получената зависимост (2) следва, че 2;
с увеличаване на постоянната част на масовия vB - скорост на т.В от звено 3.
инерционен момент на агрегата степента на а приведеният момент:
неравномерност се намалява. Освен това е
видно, че най-лесно може да се влияе на δ чрез ( )
(5)
изменение на общия масов инерционен момент
на агрегата J0, който включва и масовия
инерционен момент на маховика , където: Fp- сила на просичане.
където: JC – масов инерционен момент на С оглед частично уравновесяване на
агрегата без маховик. механизма е необходимо масовият център S1 на
Задачата за синтез на машинния агрегат звено 1 да се доведе до неподвижната т.О.
по зададена степен на неравномерност се свежда
до определяне на необходимия масов 2.2.Механизъм за структурно оформяне
инерционен момент на маховика. За целта се (фиг.3)
решава уравнение 2 относно J0: Приведеният масов инерционен момент
към началното звено 1 се определя от:
( )
∫ [ ( )( ) ( ) ( ) ( )
( )( )] (3) ( ) ( ) ( )
За да се определи масовия инерционен ( ) ( ) ( ) , (6)
момент на агрегата е необходимо да се извърши
II-250
където:mi – маса на i-то звено (i=2, 4, 6, 7); JSi – масов инерционен момент на i-то
JSi – масов инерционен момент на i-то звено (i=1, 2, 3);
звено (i=1, 2, 3, 4, 5, 6); wi – ъглова скорост на i-то звено (i=1, 2,
wi – ъглова скорост на i-то звено (i=1, 2, 3, 3);
4, 5, 6); vS2 – скорост на масовия център на звено
vS2 – скорост на масовия център на i-то 2;
звено (i=2, 4, 6); а приведеният момент:
vH - скорост на т.H от звено 7.
а приведеният момент: ( )
(9)
( )
(7) където: Mpl - необходим момент за подаване на
ламарината в зоната на рязане.
където: Fso- сила за структурно оформяне. С оглед частично уравновесяване на
С оглед частично уравновесяване на механизма е необходимо масовите центри Si
механизма е необходимо масовите центри Si (i=1, 3) на звена 1 и 3 да се доведат до
(i=1,3, 5) на звена 1, 3 и 5 да се доведат до съответната неподвижна т.О и т.C.
съответната неподвижна т.О, т.D и т.F.
2.4. Задвижване (Фиг.5)
към началното звено 1 ( главния вал на
машината) се определя от:
( )
( ) ( ) , (10)
Фиг.3 Кинематична схема на механизма за

просичане
2.3.Механизъм за подаване на ламарината

(фиг.4)
към началното звено 1 (главния вал на Фиг.5 Кинематична схема на задвижването
машината) се определя от: където:Jдв – масов инерционен момент на ротора
на ел.двигателя;
( ) Jрш1 – масов инерционен момент на
( ) ( ) ( ) , (8) ремъчна шайба 1;
Jрш2 – масов инерционен момент на
ремъчна шайба 2;
wдв – ъглова скорост на двигателя;
w1 – ъглова скорост на главния вал на
машината;
а приведеният момент:
( )
(11)
където: Mдв – въртящ момент на двигателя.

Фиг.4 Кинематична схема на механизма за Ремъчна шайба 2 е необходимо да бъде
подаване конструирана така, че да представлява и
където:m2 – маса на звено 2; маховик.

II-251
2.5.Параметри на модела за изследване и
[( ) ( ) ] (18)
определяне на масовия инерционен
момент на маховика. Окончателно за масовият инерционен момент на
Полученият едномасов динамичния маховика се получава:
модел ще има следните параметри:
-приведен масов инерционен моментна където:γ-специфично тегло (за чугун –
агрегата: 73600N/m3);
( ) ( ) ( ) ( )
, (12) g=9,81m/s2;
k-брой отвори;
-приведен момент на агрегата: z-брой ремъчни канали.
( ) ( ) ( ) ( ) 3.Заключение.
, (13)
На база на разработения динамичен
където: – предавателно отношение на модел на механизмите на електромеханична
верижната предавка. преса за производството на метална
След извършване на голям обем от просечена мрежа е определен необходимия
изчисления е определен маховик с J=……. и е масов инерционен момент на маховик с
разработен маховик оформен като втора оглед постигане на нужната степен на
ремъчна шайба (фиг.1). неравномерност.
2.6.Проектиране на маховика Авторът изказва своята благодарност
Избираме принципна конструкция на на НИС при Технически университет -
маховика (фиг.6). Разделяме маховика на
София (проект №162ПД0018-24), с чиято
следните части: главина средна част и венец.
Определяме инерционните моменти [4] спрямо
финансова помощ са проведени настоящите
оста на въртене за така определените отделни изследванията.
части.
1. Витларов З., Методики за синтез на

механизмите на електромеханична преса за
производство на просечена мрежа,
International Conference on Engineering,
Technologies and Systems TECHSYS 2016,
Technical University – Sofia, Plovdiv branch
26– 28 May 2016, Plovdiv, Bulgaria.
2. Петришки Н., Витларов З.,
Кинетостатичен анализ на механизмите на
електромеханична преса за производство на
просечена мрежа, Mechanical Engineering &
Фиг.6 Принципна конструкция на маховика
Science, сп.Механика на машините , кн.117,
-главина
В., 2017.
[( ) ( ) ] (14) 3. Минчев Н. и др., “Теория на механизмите
-средна част и машините“, „СОФТТРЕЙД“, 2011.
4. Кисьов И., Наръчник на инженера, част II,
[( ) ( ) ] (15) Механика, „Техника:, С., 1979.
От средната част трябва да се извадят отворите, Контакти
като за един отвор, спрямо оста на отвора: Адрес: 4000 Пловдив, БЪЛГАРИЯ
( )
(16) ул. Цанко Дюстабанов № 23
а спрямо оста на маховика използвайки ТУ-София, филиал Пловдив
теоремата на Щайнер: катедра МУ
( )
E-mail: z_vitlarov@abv.bg
( ) ( ) (17)
-венец
От венеца трябва да се извадят каналите за
ремъците:
 International
TECHSYS Conference
2017, Technical on Engineering,
University Technologies
– Sofia, Plovdiv branch and Systems
TECHSYS 2017, Technical University
18 – 20 May 2017, Plovdiv, Bulgaria – Sofia, Plovdiv branch
ЕДИН ПОДХОД ЗА ИЗБОР НА КРИТЕРИИ ЗА

ОЦЕНЯВАНЕ НА СРЕДСТВА ЗА ЗАЩИТА НА
ОБЕКТИ
НИКОЛАЙ ГЕОРГИЕВ, АЛЕКСАНДЪР КОЛАРОВ, ВЕНЦИСЛАВ ПЕХЛИВАНСКИ
Резюме: Използването на разнообразни системи за защита прави актуален

проблемът за избор на подходящи критерии за тяхното качество. В разработката
се предлага подход за определянето им базиран на обобщени показатели за тяхното
качество. Подходът е свързан с адаптиране на известни критерии за качество,
отчитайки особеностите на системите за защита на обекти.
Ключови думи: критерии, качеството, система, защита
AN APPROACH FOR SELECTION CRITERIA

FOR EVALUATION OF REMEDIES OF
PROJECTS
NIKOLAY GUEORGUIEV, ALEKSANDAR KOLAROV, VENCISLAV PEHIVANSKI
Abstract: Using a variety of protection systems makes current problem of selecting

appropriate criteria for their quality. The paper proposes an approach for their
determination and based on generalizes indicators of their quality. The approach is related
to the adjustment of certain quality criteria, taking into account the characteristics of the
systems to protect objects.
Key words: criteria, quality, system, protection
1. Въведениe публикациите на Американското общество за

Известно е, че основната част на качеството и с много други организации[1,2,3],
стандартите от Системите за управление на качеството се определя като съвкупност от
качеството ISO 9000, които установяват свойства и характеристики на дадено изделие
принципи и основни понятия на управлението или действие, определящи способността му да
на качеството, се състои от пет международни задоволи определени потребности на
стандарта (Фигура 1), които по същество са потребителя при конкретно дефинирани
ръководство за разработка и внедряване на условия. От това определение следва, че не
ефективна Система за управление на всички свойства и характеристики на изделието
качеството [1,2]. Европейскиятстандарт EN ISO или действието влизат в състава на тяхното
9001:2015 е въведен в България като БДС EN качество, а само тези, които обуславят
ISO 9001:2015. В него е въведено изискването свойствата им да удовлетворяват определена
за идентифициране на рисковете и потребност.
възможностите, които могат потенциално да Използването на разнообразни
повлияят на функционирането и ефективността системи за защита прави актуален проблемът за
на системите за управление на качеството, избор на подходящи критерии за тяхното
както и съответните „пропорционални“ качество както и подход за определянето им,
действия за справяне с тях. В съответствие с базиран на обобщени показатели за тяхното
тези международнин стандарти, с качество.
II-253
излизането му от употреба), към разходите за
създаването, поддържането, експлоатацията,
ремонта и утилизацията на този модул.
Основната класификация на сложните
системи, към които можем да отнесем и
системите за защита на критичната
инфраструктура, по отношение на целите и
ресурсите е [6]:
a. По отношението на системата
към обкръжаващата среда:
Фиг. 1 Основни стандарти от Системата - отворена – системата има обмен на
за управление на качеството ресурси с обкръжаващата среда;
- затворена - системата няма обмен на
2. Изложение ресурси с обкръжаващата среда.
b. По произход на системата
Показателят за качество е количествен
(елементи, връзки, подсистеми):
израз на едно или на няколко свойства, а
- изкуствени – оръжия, механизми,
критериите за качеството са изисквания за това роботи, транспортни срдства и т.н.;
даден показател, или съвкупност от показатели - естествени – живи, неживи,
да имат определени стойности. В зависимост от екологични, социални и т.н.;
природата на свойствата, които количествено - виртуални – въображаеми и
характеризират, критериите и показателите за независимо че в действителност те
качеството биват физични, химични, реално не съществуват,но
функционират така, все едно че
механични, и т.н., а в зависимост от техния
съществуват реално;
характер - показатели за надеждност, за - смесени – икономически,
ергономичност, за функционалност и др. Освен биотехнически, организационни и др.
това всички критерии и показатели за качество c. По описанието на неизвестните
се разделят на единични, комплексни и (променливите) в системата:
интегрални[4,5]. - с качествени променливи – имащи
Единичният показател за качество само съдържателно описание;
характеризира само едно от свойствата на - с количествени променливи – имащи
продукта. Комплексният показател за качество дискретно или непрекъснато
количествено описание;
характеризира определена съвкупност от - смесено описание – количествено-
свойства, влизащи в структурата на качеството. качествено.
Той може да бъде групов и обобщаващ. d. По типа на законите, който
Груповият показател за качество характеризира описват функционирането на системата:
няколко прости свойства или сложно - тип „черна кутия” – напълно
(комплексно) свойство. Обобщаващият неизвестни закони по който
показател за качество характеризира в функционита системата, известни са
количествено отношение цялата съвкупност от само входните и изходните данни;
- непараметизиран – законите не са
свойства, по които е прието да се оценява
описани, а са известни само някои
дадено изделие или действие. Най-често априорни негови свойства;
интегралният показател за качество отразява - параметризиран – законуте са
отношението между общия полезен ефект на известни с точност до параметрите му
дадено изделие или действие и сумата от и техните зависимости;
разходите по неговото създаване и - тип „прозрачна кутия” – напълно
експлоатация или провеждане. Например известни закони за функциониране на
системата.
интегрален показател за качеството на дадена
e. По способа на управление на
система за защита може да бъде обемът на
системата (в системата):
„защитаваната” зона с определена ефективност
за определен период от време (напр. до

II-254
- управляема отвън – без обратна на вероятностния подход е системния анализ на
връзка, регулируема, със структурно възможните сценарии с отчитане на
управление, функционално или количествения анализ на надеждността на
информационно;
системата. Когато управлението на
- с вътрешно управление –
самоуправляваща (саморегулираща) безопасността на критичната инфраструктура
се, програмно или автоматическо протича в условията на неопределеност и
управляема, самоорганизираща се, неяснота на информацията, свързана с нейното
адаптируема, и др.; функциониране, се налага използване на метода
- с комбинирано управление – на така наречените „меки изчисления”. Този
автоматическо, автоматизирано, метод се основава на теорията на нейронните
организационно и др.
мрежи (Neural Network), логиката на неясните
От своя страна системите на
събития (Fuzzy Logic) и вероятностното
критичната инфраструктура по характера на
описание (Probabilistic Reasoning).
своето поведение могат да бъдат
Основните характеристики на
класифицирани като:
методите за анализ на безопасността в
- детерминирани – поведението им е
известно по определени правила и критичната инфраструктура са показани в
закони. Множеството от състояния на Таблица 1.
всеко елемент и на системата като Таблица 1 Основни характеристики на
цяло са известни; методите за анализ на безопасността
- вероятностни – поведението им може в критичната инфраструктура
да бъде описано с методите на
теорията на вероятностите; МЕТОДИ: Детерминиран анализ Вероятностен подход „Меки изчисления”
- хаотични – характеризират се с това, Входни данни: Детерминирани Вероятностно Неясни случайни
че не големи изменения в тяхното разпределение значения на
параметрите
текущо състояние могат да доведат
Множество на Само с най-лошите Всички прогнозни Всички прогнозни
към непредсказуеми и значителни разглежданите последствия събития събития
изменения в последващите състояния. събития
Класифицирането на дадена система Честота Достоверни събития Вероятността се Лингвинистични
оценява в съответствие оценки, числа с
на критичната инфраструктура по нейното с приетия закон за неопределена стойност
разпределение
поведение определя в голяма степен и методите
Тежест на Предполага се известна Предполага се известна Лингвинистични
за нейната оценка. последствията променливи
Характерното за детерминирания Оценка на риска Качествен анализ Количествен анализ Анализ на неточни
числа
анализ е, че при него не се използват
Оценка на Не се разлежда Стохастическа Нестохастическа
количествени вероятностни данни за описание неопределеността неопределеност, неопределеност,
неопределеност от неопределеност от
на събития или тяхното съчетание. Определят първи род втори ред
се възможни сценарии включващи базовото
множество от събития и последствията от тях
Поради това, че в повечето случаи
върху критичната инфраструктура. Сценариите
разходите по създаване и експлоатация на
обиновенно се описват с стествен език с
системите за защита са относително добре
песимистичен уклон на изразяване –
дефинирани в теоретичен аспект те не
ориентирани са към най-лошия сценарий на
представляват интерес. От друга страна
развитие на събитието. Обикновенно се
критериите за качество, базирани на единични
планират по-голямо количество средства от
и групови показатели са непълни и биха могли
необходимото и не се отчитат
да се използват или за решаване на частни
неопределенностите в ситуацията, което прави
задачи, или чрез комбинирането им в система
този подход неефективен от гледна точка на
от критерии. Ето защо в доклада се предлага
управлението. Когато поведението на
адаптиране на известни критерии за качество,
системата не може да бъде описано с набор
базирани на обобщаващи показатели за
определени правила, за оценка на риска се
качество, отчитащи особеностите на системите
използва подход характеризира се с
за защита на обекти с използването на
използването на веоятностни оценки. Основата

II-255
вероятностния подход за анализ на - - загуба нанесена от системата за
безопасността на критичната инфраструктура. защита върху -ти тип средства за
Основното предназначение на атака.
системите за защита на обекти е да се При известни стойности на
предотвратят атаките срещу тях или да се посочените параметри е възможно
намали ефектът от тях. Това можа да се формулирането на следните основни групи от
постигне чрез предотвратявана на атаките критерии за качеството на системите за защита:
посредством унищожаване или повреждане на a. Критерий за минималните
атакуващите средства, както и чрез средни общи загуби (на отбраняваните обекти и
възпрепятстване на достъпа им до зони, от модулите за защита).
които те могат да извършат атака. В общия Критерият отчита средните загуби
случай описването на тези характеристики нанасяни от атакуващите средства върху
включва вероятностите и степента на всички собствени обекти - и защитаваните
поразявания нанасяни както на атакуващите обекти и компонентите на системата за защита.
Общият му вид е:
средства върху защитаваните обекти и върху
=ƩƩƩ /j /j = min , (1)
модулните системи за тяхната защита, така и
където сумирането е за
тези, нанасяни от средствата и системите за
защита върху атакуващите средства. Известно е [3,4], че Qi/j може да се
Нека приемем, че съществуват I разгледа като зависеща от вероятността
възможни типа средства за атака върху J броя подсистемата за разузнаване да открие
прикривани обекти. Нека освен това приемем, своевременно i-тото средство за защита ( ),
че съществуват М типа модулни системи за вероятността тази информация да се предаде с
необходимото качество и на нейна база да се
защита, като всяка от тях е предназначена да
вземе решение за въздействие върху i-тия тип
прикрива един или няколко обекта. средства за атака ( ) и от вероятността за
Нека въведем следните обозначения: това въздействието на системата за защита да е
- - вероятност за това -ти тип ефективно, т.е. да реализира загубата
средства за атака да извърши реално . Освен товаможе да се приеме, че в процеса
въздействие върху j-ти прикриван на операцията стойността на Pijи Rim/jса с
относително постоянни стойности - и
обект ( );
.Тези стойности могат да се променят
- - вероятност за товасистемата за съществено с коефициент , зависещ основно
защита да извърши реално от действията на системата за защита ( =1-
въздействие върхуi-ти тип средства за =1-
атака, прикривайки j-ти обект; Въвеждайки гореописаните параметри,
- - вероятност за това ти тип критерият за минимални средни общи загуби
добива вида:
средства за атака да извърши реално
/ =
въздействие върху -ти тип модул за
ƩƩƩ /
защита при опита си за въздействие
върху -ти прикриван обект; )=ƩƩƩ (2)
- - загуба нанесена от -ти тип където:
средства за атака при реално - = / /j - коефициент
въздействие върху j-ти прикриван на важност на -то средство за атака
обект без върху него да въздейства - при действието му срещу -ти
ти тип модул за защита; защитаван обект и -ти модул от
- - загуба нанесена от -ти тип системата за защита;
средства за атака при реално - / / / -
въздействие върху m-ти тип модул за коефициент на противодействие на
защита, при опита си за въздействие
системата за защита срещу -то
върху -ти прикриван обект; средство за атака при действието му

II-256
срещу -ти защитаван обект и -ти Този критерии се базира на хипотезата,
модул от системата за защита. че зашитата на обектите следва да се постигне
Формула (2) характеризира средния чрез намаляване в максимална степен на
риск не само чрез действията на атакуващите възможностите на атакуващите средства, т.е.
средства, а и чрез противодействието на критерият е:
системите за защита ( и ). Следва да се (6)
отчита, че зависимостта между и от където сумирането е за
една страна и и от друга страна е по- и .
сложна и определянето и е целесъобразно да се e. Критерий за осигуряване на
осъществява за фиксирани ситуации, или чрез
минимално необходими средни загуби на
използване на теорията на игрите. Въпреки това
използваното приближение не променя общата противника
качествена зависимост между качеството на Критерият е модификация на критерия за
системата и параметрите участващи във максимални средни загуби на противника и е
Формула (2). свързан с въвеждането на изискване за
Очевидно минимума на средния риск се гарантиране на определена минимална
постига с отчитане както на важността на стойност на тези загуби, т.е.:
отделните средства за атака , така и на , (7)
възможностите на системата за защита за където е гранична, допустима
противодействие срещу него стойности на средни загуби на противника.
b. Критерий за минималните f. Критерий на максималното
средни загуби отношение между загубите
Критерият отчита само загубите на или
отбраняваните обекти и е от вида:
= min, (3) = / (8)
Вероятните сценарии на приложение
където сумирането е за
на предлаганите критерии за качеството на
Отчитайки използваните за Формула (2) системите за защита и изключенията са
параметри можем да запишем критерияза показани в Таблица 2.
минималните средни загуби като: Таблица 2 Вероятни сценарии на
приложение на предлаганите критерии
Критерий Вероятен сценарий на Ограничения
(4) приложение
- защитаваните обекти и - неподходяш за
където = е коефициент на модулните системи са защита на особено
със сравними по важни обекти от
важност на то средство за атака при стойности важност; критичната
действието му срещу -ти защитаван обект. - запазване капацитета инфраструктура;
c. Критерии за максимално на системите за - слабо приложим за
защитата. епизодични
допустими средни загуби агресивни действия.
За преодоляване на слабостите на - защитаваните обекти - при защита на
критериите за минималния среден риск, както и са с относително еднаква единични особено
важност; важни обекти.
за осигуряване на определена степен на защита - въздействия с - при невъзможност
се използва метод за фиксиране на допустимите единични или с малки от бързо
групи ; възстановяване на
загуби. Методът предвижда средните загуби - системите за защита са средствата за
(общи или само на защитаваните обекти) при с висока степен на защита.
произволен сценарии на агресия да не устойчивост - при цена на
- епизодични агресивни средствата за защита
превишават предварително зададени стойности, дейстия, след които съизмерима с тази на
т.е.: системата за защита се защитаваните
= < или < (5) възстановява обекти.
където и са граничните, Да не се допусне Когато приоритет е

намаляване на защитата на много
допустими стойности на съответните средни остатъчната стойност под важни единияни
загуби. определена (допустима): обекти.
d. Критерии на максималните - на защитаваните Когато има
обекти; възможност за бързо
средни загуби на противника - на защитаваните възстановяване на
обекти и системите за обектите или на
защита. средствата за защита.

II-257
Намаляване в Нанасянето на хидро- и аеродинамика, национална
максимална степен на значителни загуби на
възможностите на противника е сигурност“ ISSN:1313-8308, стр. 374-378.
атакуващите средства. самостоятелна цел, 3. Kiril Stoichev, Valeri Panevski, Dmitar
Приоритет на за изпълнението на Dimitrov (2016). Contemporary approach
психологическо, медийно която не е
и морално въздействие. задължително да се for complex analysis and evaluation of
Защита на територия защитават конкретни hazardous environments (CAEHE),
обекти.
Малка интензивност на Нанасянето на
International Journal of Economics,
конфликта. предварително Commerce and Management, IV, 4,
Създаване на условия за планираните загуби ISSN:2348 0386, pp. 500-516.
прекратяване на на противника не е
конфликта. пряко свързано със 4. Кирил Стойчев, Димитър Димитров,
Демонстрация на защитата на Валери Пъневски (2016) Интегрирана
възможности за защита. конкретни обекти. сигурност и защита на критична
При относителен паритет Постигането на
на силите в началото на максимално инфраструктура, ИМСТЦХА-БАН,
конфликта и стремеж за съотношение между ISBN:978-619-90310-6-3, стр. 340.
повишаване на нанесените и 5. Kiril Stoichev, Valeri Panevski, Dmitar
съотношението между претърпените загуби
възможностите на не означава нито Dimitrov (2016) Contemporary approach
защитаваните обекти и сриване на for complex analysis and evaluation of
средства спрямо възможностите на
атакуващите. противника, нито
hazardous environments (CAEHE),
При стремеж за надеждна защита на International Journal of Economics,
демонстрация на конкретни обекти. Commerce and Management, IV, 4,
превъзходство на
защитата и доказване на International Journal of Economics,
несъстоятелността на Commerce and Management, ISSN:2348
агресията.. 0386, pp. 500-516.
6. Георгиев Стилиян (2016) Сензорни
3. Заключение
системи за защита на обекти от
Посочените критерии се базират на критичната инфраструктура. Научни
обобщени показатели за качеството на известия на НТСМ, 1, 187, ISSN:1310-
системите за защита. Използването им зависи 3946.
от конкретния сценарии за развитие на
агресията - напр. при очаквана агресия от Контакти:
терористични групи подходящ е критерия за
минимални средни загуби или за максимално
допустими такива[5]. При възможна кратка 1. Николай Георгиев
агресия от друга държава е подходящ критерия Адрес: София 1574, ул. "Шипченски
за максимални средни загуби нанесени на проход" 67. Институт по металознание,
противника или критерия за максимално съоръжения и технологии „Акад. Ангел
отношение между загубите. При очаквани Балевски“ с Център по хидро- и
продължителни агресивни действия би могъл аеродинамика – БАН (ИМСТЦХ–БАН),
да се използва критерия за минималните средни тел: (+359 2) 46-26-250
общи загуби. Възможно е използването и на факс: (+359 2) 46-26-202
други критерии, получени чрез различни E-mail: niki0611@abv.bg
комбинации от посочените, както и чрез
групиране на определени съвкупности от 2. Александър Коларов
обекти за защита, модулни системи и Адрес: София 1574, ул. "Шипченски
атакуващи средства и задаване на различни проход" 67. ИМСТЦХ–БАН.
частни критерии за всяка отделна група. тел.: + 359 52 370 512
факс: + 359 52 370 514
ЛИТЕРАТУРА E-mail: aleksandar_kolarov@abv.bg
1. http:/asq.org/glossary/q.html. 3. Венцислав Пехливански

2. Dimitar Dimitrov, Valeri Panevski (2015) Адрес: София 1574, ул. "Шипченски
The need for unformity of languange in проход" 67. ИМСТЦХ–БАН,
security analysis, risk management and тел: (+359 2) 46-26-250
critical infrastructure protection-an факс: (+359 2) 46-26-202
overview Пета национална конференция с E-mail: venpeh@abv.bg
международно участие „Металознание,

 International
TECHSYS Conference
2017, Technical on Engineering,
University Technologies
– Sofia, Plovdiv branch and Systems
TECHSYS 2017, Technical University
18 – 20 May 2017, Plovdiv, Bulgaria – Sofia, Plovdiv branch
НОВИ ПОДХОДИ ЗА ПОВИШАВАНЕ

ПРОТИВОМИННАТА УСТОЙЧИВОСТ НА
МОРСКИ ДЪННИ МИНИ
СТЕФАН ВОДЕНИЧАРОВ, АЛЕКСАНДЪР КОЛАРОВ, НИКОЛАЙ ГЕОРГИЕВ, ВЕНЦИСЛАВ
ПЕХЛИВАНСКИ
Резюме: При конструирането на съвременните морски дънни мини обикновенно се

търси компромис между тактическите изисквания към този вид оръжие и
изискванията за висока устойчивост срещу евентуални противоминни действия,
например, правят се с цилиндрична форма когато трябва да се поставят на позиция
от торпедни апарати на подводници, корпусът им се конструира в така
наречената „стелт“ технология и се изработва от неметални материали,
увеличава се количеството на взривното вещество когато дълбочината на района е
по-голяма и т.н. В статията се предлага подход, който дава общо решение на
всички тези изисквания и позволява конструирането на една моногофункционална,
универсална и с висока противоминна устойчивост морска дънна мина, която няма
аналог във военноморската практика.
Ключови думи: морска дънна мина, хидроакустична сянка, фрагментиране,
дефрагментиране, подводен реактивен двигател, аербег, противоминен сонар.
NEW APPROACHES TO INCREASE ANTI-

MINING RESISTANCE TO SEA BOTTOM MINES
STEFAN VODENICHAROV, ALEKSANDAR KOLAROV, NIKOLAY GUEORGUIEV, VENCISLAV
PEHLIVANSKI
Abstract: In constructing the modern sea bottom mines usually seek a compromise between
tactical requirements for this type of weapon and requirements for high resistance against
possible mine-action e.g. make it a cylindrical shape when you need to put an item on the
torpedo apparatus submarines, their shell is constructed in the so-called "stealth"
technology and is made of non-metallic materials, increasing the quantity of the explosive
when the depth of the region is greater, etc. The article proposes an approach that provides
a common solution to all these requirements and allows the construction of a highly
functional, versatile, high anti-mining resistance sea bottom mine that has no analogue in
the naval practice.
Key words: sea bottom mine, hydro acoustic shadow, fragmentation, defragmentation,
underwater jet engine, airbag, anti mining sonar.
1. Въведениe Съвременните противоминни системи

Разработването на нови типове дънни за откриване и унищожаване на дънни морски
морски мини е актуален проблем поради факта, мини представляват информационен комплекс
че този тип средства за защита на пристанища и организиран на няколко нива. Простото
подстъпите към тях стават част от откриване на мината като локална
компонентите за защита на морска критична хидролокационна цел на фона на
инфраструктура[1,2,3]. обкръжаващата среда все повече се
преобразува в мониторинг на подводната
Copyright©2017
University - Sofia,
of Sofia, Plovdiv
Plovdiv branch,
branch, Bulgaria
II-259
обстановка и създаване на детайлна хидроакустични вълни в диапазоните
хидролокационна картина на участъци от на работа на съвременни минотърсещи
морското дъно с подробно обследване за всяка сонари – 30 kHz ÷ 600 kHz;
подозрителна цел и фиксиране на всички - профилът на тялото на мината да
локални изменения. отговаря на изискванията на “стелт”
Най-новите тенденции при технологията за разсейване на
конструирането на корабните подкилни звуковата енергия в направление,
противоминни сонари и тези от състава на различно от ъгъла на падане;
буксируемите или автономни подводни апарати - да има голяма отразяваща повърхост,
и роботи, свързани с обработването на която да води до маскиране на
отразените от целта хидроакустични сигнали, сянката, при което филтрите за
използват следните основни показателите за големина на целта (относно габарити
класификация, заложени в алгоритмите за на известни мини) няма да я
откриване на дънна мина [4,5]: класифицират като миноподобна цел;
1. Хидроакустична сянка на целта - да има симетричност на формата
върху морското дъно - основно при използване поради това, че обикновенно се
на буксируеми сонари или на монтирани на извършва двукратен поиск на
автономни самоходни подводни апарати перпендикулярни галсове и се
(Autonomous Underwater Vehicle – AUV); оценяват (изваждат) сигналите от
2. Сила на целта – съотношение целите;
между падналата върху целта и отразената в - да има ниска отразаваща енергетична
направление към антената хидроакустична способност, защото се извършва
енергия (основно за подкилни сонари); оценка на центъра на тежестта, т.е. ако
3. Оптическа забелижимост една маса е с по-голяма площ тя е по-
(контраст) върху морското дъно – при слабо контрастна.
използване на оптико-електронни (оптически) Влиянието на вертикалния профил на
средства за поиск и на сонари с много висока тялото на мината върху големината на
честота (над 600 kHz), които правят хидроакустичната сянка е показано на Фиг.1,
хидроакустична снимка на целта; където:
4. Пространствени размери на
- H [m] – отстояние на буксируемия
целта, които са заложени във филтрите за
сонар със страничен обзор (AUV) от
автоматична класификация на миноподобни
дъното;
тела.
- h [m] – височина на дъната от дъното;
Отчитайки изложеното, подходите за
- D [m] – разстояние между сонара и
повишаване на противоминната устойчивост на
мината;
морските дънни мини, наричани още „защитно
- S [m] – дължина на хидроакустичната
оръжие на бедните държави” са свързани с
сянка.
намирането на технически решения, които да
намаляват значително признаците за тяхната H D h S H D h S
класификация като цел на фона на морското 2 10 0.2 1.1 4 20 1.0 6.67
дъно. 2 10 0.3 1.8 4 40 0.2 2.13
2 10 0.5 3.3 4 40 1.0 13.34
2. Изложение 2 10 1.0 10.0 4 60 0.2 3.16

2 20 0.2 2.2 4 80 0.2 4.25
С цел намаляване ефективността на 2 20 0.5 6.7 6 10 0.2 0.34
противоминните сонари при откриване на 2 20 1.0 20.0 6 10 1.0 2.0

2 40 0.2 4.4 6 20 0.2 0.69
дънни мини по големината и формата на 2 40 1.0 40.0 6 20 1.0 4.0
хидроакустичната им сянка и по нивото на 2 60 0.2 6.7 6 30 0.2 1.03
отразения сигнал, съвременна дънна мина 2 80 0.2 8.8 6 30 1.0 6.0

4 10 0.2 0.53 6 50 0.2 1.72
трябва да отговаря на следните изисквания [6]: 4 10 0.3 0.81 6 50 1.0 10.0
- да има нисък вертикален профил, за да 4 10 0.5 1.43 8 50 0.2 1.28
формира по-малка хидроакустична 4 10 1.0 3.33 8 50 1.0 7.14
сянка върху морското дъно; 4 20 0.2 1.06 8 100 0.2 2.56
- да е с конструкция от материал, който 4 20 0.5 2.87 8 100 1.0 14.28
поглъща и/или има малка отразяваща

способност при облъчване с
II-260
Фиг. 1 Хидроакустична сянка на цел, в пространството. От резултатите може да се
позиционирана на дъното направи извод, че пораженията, които могат да
се очакват от дънна мина с тегло на заряда 500
Вижда се, че големината на сянката от
кг на разстояние 30 метра от кила на кораба, са
мина с височина 0.1 м е 10-15% от тази на най-
приблизително равни на пораженията от
разпространените дънни мини с височина
подводен заряд с тегло 10 кг на разстояние 5 м
около 1 м. Следователно формата на мината
от корпуса на кораба.
трябва да е с минимална височина.
За удовлетворяване изискването за
симетричност е целесъобразно мината да има
звездовидна форма във фрагментирано
състояние (когато е на позиция на дъното), а за
да отразява на звуковите вълни в странични
направления е необходимо повърхността на
фрагментите да сключва походящ ъгъл с
равнината на дъното. Корпусите на отделните
фрагменти трябва да са от материал с намалена
отразяваща способност. При групиране на
фрагментите (дефрагментиране) мината трябва
да придобива цилиндрична форма с диаметър,
позволяващ тя да бъде изстрелвана от торпеден
апарат.
От гледна точка на бойната
ефективност на дънна мина може да се приеме,
че поразяващата и способност е обратно
пропорционална на дистанцията до подводната
част на кораба и право пропорционална на
големината на заряда. В съответствие с
методиката на разчет на налягането на ударната Таблица 1 Поразяващо действие на
тротилов заряд
вълна, свързано с формата и размерите на
газовия балон, се доказва, че зарядът трябва да Следователно съвременната дънна
е компактен с форма близка до сферичната мина трябва да е компактна и да се детонира на
(цилиндрична) и с минимален радиус. Известно оптимална (минимална) дистанция до дъното
е, че поразяващото действие на тротилов заряд на кораба-цел, което ще позволи тя да бъде
може да се изчисли по формулатa [7]: позиционирана на дълбочини съответстващи на
зоната на действие на неконтактните сензори –
3 1.16
Pmax = 23800
W
(1) 100 ÷ 200 м, а не на зоната на действие на
R
поразяващото действие на концентриран заряд
където: – 30 – 40 м за заряд 1000 кг.
- Pmax [dB] – акустично налягане От гледна точка на тактиката на
на фронта на ударната вълна; постановка на минни заграждения е
- 𝑊𝑊 [lb] – тегло на тротиловия целесъобразно дънните мини освен по
заряд; традиционния способ на миноспускане – по
- 𝑅𝑅 [ft] – разстояние до корпуса на релсови пътища от палубата или с подемно-
кораба. пускови устройства, да могат да се поставят и
Някои зависимости между Pmax , 𝑊𝑊 и 𝑅𝑅 през торпедните апарати на бойни кораби и
са представение в Таблица 1, в която чрез к е подводници, т.е. да имат цилиндрична форма
обозначен пространствен коефициент на със съответния радиус. Най-разпространения
разпространение на енергията на газовия балон калибър торпеда, а съответно и торпедни

II-261
апарати, е 533 мм, но са известни и образци с изделието на вода от палубата на
диаметър от 254 до 660 мм. кораб или от торпедните апарати на
За удовлетворяване на всички тези бойни кораби и подводници, като
подбраните геометрична форма на
понякога противоречиви изисквания се
фрагментираното тяло и
предлага нова инженерна конструкция на масогабаритни характеристики,
фрагментираща се дънна мина /ФДМ/, която се съобразени с размерите на торпедните
състои от N на брой бойни тела, които са апарати, осигуряват плавно
съединени посредством подвижни връзки с удълбочаване и заставане стабилно на
централно цилиндрично тяло - групиращ дъното;
неконтактен взривател /ГНВ/, в което се - процесът на дефрагментиране
(групиране на елементите) на мината
намират сензорите, захранващия блок на
след сигнал от сензорната система и
мината и реактивен двигател и/или аербег подизплаването й на оптимална
(airbag), който служи за отлепване на корпуса дълбочина осигуряват ефективното
от дъното и задвижване във вертикално използване на поразяващата сила на
направление и групиране на бойните тела. подводния взрив при минен вариант
Предназначението на ФДМ е да или подходящо положение на
поразява подводната част на корпуса и за елементите при торпеден вариант.
Могат да бъдат реализирани следните
нанасяне на бойни повреди на кораби със
варианти според изискванията за бойно
средно и голямо водоизместване, на
използване на ФДМ в конкретна операция:
подводници и на десантни кораби, в това число
- ГНВ без реактивен взривател – за
и на въздушна възглавница.
противодесантни морски минни
заграждения на дълбочини до 10 м,
когато не се налага групиране;
- ГНВ с реактивен двигател с малка
мощност и/или с използване на аербег,
който осигурява групирането в
компактно състояние на бойните тела
след повдигане на ГНВ на височина от
дъното, равна на дължината им –
стандартен вариант на дънна мина,
позиционирана на дълбочини от 10 ÷
40 м;
- ГНВ с реактивен двигател с голяма
Фиг. 2 Конструкция на ФДМ мощност, осигуряващ движение на
мината към повърхността на морето и
Конструктивните особености при доближаване на целта;
реализиране на идеята за ФДМ (Фиг.2), освен - ГНВ с реактивен двигател с малка
мощност и/или с използване на аербег,
за удовлетворяване на изискванията за
на която в обемите на бойните тела
намаляване ефективността на противоминните вместо взривно вещество са поместени
сонари при откриване на дънни мини по ракетни подводни снаряди, които
големината и формата на хидроакустическата реализират зона на поражение с
им сянка и ниво на отразен сигнал, я правят програмируеми характеристики;
многофункционална защото: - ГНВ с реактивен двигател с малка
- съставните елементи могат да бъдат мощност и/или с използване на аербег,
профилирани с максимален полезен на която в обемите на бойните тела
обем при вариант “мина” и с вместо взривно вещество са поместени
подходящи пускови контейнери за малогабаритни високоскоростни
варианти “подводни реактивни торпеда.
снаряди” и “малогабаритни Проведените експерименти показват,
високоскоростни торпеда”; че за дълбочини в районите за минна
- минната постановка може да се постановка до 70 - 80 м, е целесъобразно
извършва със свободно спускане на използването на вариант на ФДМ с ГНВ с
II-262
реактивен двигател с малка мощност, който да показват, че те спомагат в значителна степен за
осигурява преодоляване на засмукващата сила повишаване ефективността и противоминната
на тинесто дъно, и аербег за осигуряване на устойчивост на морските дънни мини в
групирането и изплаване към повърхността. В следните направления:
този случай е важно вертикалната скорост на - универсалност – могат да се поставят
движение да е достатъчна за да се реализира на позиция на целия диапазон
детонация под кила на кораба цел на оптимална оперативни дълбочини на района на
постановка от надводни кораби и
дистанция или най-общо казано за времето, за
подводници за въздействие
което целта изминава дистанцията на откриване практически срещу всички видове
от сензорите, мината трябва да изплава под надводни кораби и подводници;
него на оптимално разстояние за поражение на - многофункционалност – възможно е
подводната част. Значенията на средната да бъдат използвани във оптимален
скорост на изплаване на балон пълен с въздух вариант според изискванията за бойно
при различни значения на неговия обем и използване в конкретна операция или
комбинирано;
напречно сечение са показани в Таблица 2 [8]:
- в групиран вид ФДМ има формата на
цилиндър с диаметър в съответствие с
приетите на въоръжение торпедни
апарати (прототипът е изработен за
торпеден апарат за изстрелване на 533
мм торпеда), което позволява
извършване на минна постановка
както от кораби, така и от подводници;
- постига се значително намаляване
ефективността на противоминните
сонари при откриване на дънни мини
по големината и формата на
хидроакустическата им сянка и ниво
на отразен сигнал. На Фиг. 3 се вижда,
че хидроакустичната сянка от ФДМ е
по-малка от тази на 50 литров метален
варел и значително-помалка от тази на
стандартна дънна мина със заряд с
еднакъв тротилов еквивалент;
Таблица 2 Средна скорост на изплаване на

аербег в морска вода
Разчетите показват, че е достатъчна

скорост на изплаване от порядъка на 3 – 4 m/s,
което е напълно достижимо при използване на
аербег с обем 50 ÷ 100 литра и налягане на
въздуха в него съответен на дълбочината на
активирането му.
Фиг. 3 Хидроакустичен контакт с ФДМ и с
Проведените експерименти с прототип, 50 литров метален варел
изготвен под ръководството на доктор Стоян
Делиев от ИМСТЦХ-БАН, за оценяване на - след пускане (изстрелване) на мината
на вода бойните тела се разтварят като
новите технически решения при
образуват звездообразно тяло, което
конструирането и функционирането на ФДМ осигурява плавно потъване на ФДМ и
II-263
заемане на стабилно положение на 4. J. A. Fawcett (2001) Image-based
дъното. Процесът на групиране на classification of sidescan sonar detections,
заряда и придвижването на мината Proceedings of CAD/CAC 2001, Halifax,
към повърхността, съответно към Nov. 12-14,.
целта, което е показано на Фиг.4, 5. S.Reed, Y.Petillot, J.Bell (2004) A model
значително повишава нейната бойна based approach to the detection and
ефективност. classification of mines in sidescan sonar,
Applied Optics, Vol. 43, Issue 2, p237-246.
6. G.J.Dobeck (2001) Algorithm fusion for
automated sea mine detection and
classification, Proc. MTS/IEEE Oceans
Conf. and Exhibition, pp.130-134.
7. R. J. Urick (1983) Principles of underwater
sound, McGraw-Hill, 3rd ed. Pp.401-412
ФДМ на позиция Стартиране
8. Tyce, R., Coelho, E., Bovio, E. MREP
(2003) Maritime reconnaissance for NATO
recognized environmental picture,
SACLANTCEN CD-7.
Движение нагоре На повърхността
Фиг. 4 Активиране на ФДМ, групиране и

движение към целта
Реализираните в ФДМ нови технически

решения я правят без аналог във военното
производство на този вид оръжия.
1. Kiril Stoichev, Valeri Panevski, Dimitar

Dimitrov. Contemporary approach for
complex analysis and evaluation of
hazardous environments (CAEHE).
International Journal of Economics,
Commerce and Management, IV, 4, pp. 500-
516ISSN:2348 0386,
2. Кирил Стойчев, Димитър Димитров,
Валери Пъневски (2016) Интегрирана
сигурност и защита на критична
инфраструктура ИМСТЦХА-БАН стр.
340, ISBN:978-619-90310-6-3.
3. Николова В., Г. Дамянов (2013)
Оперативен риск - оценка и
предотвратяване в обекти от
критичната инфраструктура. Трета
национална конференция с международно
участие „Металознание, хидро- и
аеродинамика, национална сигурност
’2013“, София стр. 294-298, ISSN 1313-
8308.

DETERMINATION OF PROPELLER
CHARACTERISTICS FOR MULTIROTOR
DRONE DESIGN
HRISTIAN PANAYOTOV, STANIMIR PENCHEV
Abstract: A methodology for determination of multirotor drone propeller characteristics is

described. A special formula with correction coefficients is derived to calculate propeller
thrust and power at different flight modes. The calculated theoretical results are compared
with experimental data and thorough analysis is carried out.
Key words: multirotor drone, propeller, characteristics
1. Basic theory velocity 𝑉𝑉𝑗𝑗 [1]. The described control volume is

Like helicopter and airplane propellers, shown in Fig. 2.
multirotor drone propellers are well described by
momentum and blade element theory. However in
multirotor drone design it is often necessary to
assess the main propeller characteristics such as
propeller thrust, power and efficiency in hover and
translational non-axial flight. Hence the aim of the
present study is to derive some approximate
dependencies that could be useful in early
multirotor design stage for determination of
propeller characteristics.
For propellers in hover mode the main
assumptions for the flow field are shown in Fig. 1.
Fig. 2. Propeller control volume
Applying the governing conservation laws

for the fluid mass and momentum to this control
volume yields the following equations [2]:
�⃗ . 𝑑𝑑𝑆𝑆⃗ = 0
∬𝑆𝑆 𝜌𝜌𝑉𝑉 (1)
�⃗ . 𝑑𝑑𝑆𝑆⃗)𝑉𝑉
𝐹𝐹⃗ = ∬𝑆𝑆 𝑝𝑝𝑑𝑑𝑆𝑆⃗ + ∬𝑆𝑆 (𝜌𝜌 𝑉𝑉 �⃗ , (2)
�⃗ is the local velocity, 𝜌𝜌 is the density of the

where 𝑉𝑉
fluid across the surface 𝑆𝑆, and 𝐹𝐹⃗ is the net force.
According to eq. (1) for the control volume
in Fig. 2 the air mass flow at the station of the
Fig. 1. Propeller flow field propeller disc is:
Far upstream the air flow is undisturbed and
𝑚𝑚̇ = 𝜌𝜌𝜌𝜌𝑉𝑉𝑖𝑖 , (3)
passing through the propeller disc the flow induces
speed – 𝑉𝑉𝑖𝑖 . The airflow is further accelerated until it
where 𝜌𝜌 = 𝜋𝜋𝑅𝑅 2 is the propeller disc area provided
reaches the far wake at slipstream velocity or the jet
the propeller has a radius of 𝑅𝑅.
II-265
Applying eq. (2) the thrust of the propeller at a certain flow velocity, rotational speed and angle
is: of attack of the propeller the measured properties
are: propeller thrust, power and efficiency. Also at
𝑇𝑇 = 𝑚𝑚̇𝑉𝑉𝑗𝑗 = 𝜌𝜌𝜌𝜌𝑉𝑉𝑖𝑖 𝑉𝑉𝑗𝑗 . (4) steady conditions with no wind tunnel flow the
slipstream velocity of the jet (𝑉𝑉𝑗𝑗 ) could be measured
For hover mode (Fig. 2) the theory [1] yields using calibrated pitot-static tube. According to eq.
that 𝑉𝑉𝑗𝑗 = 2𝑉𝑉𝑖𝑖 , so finally: (4) and (5) the induced speed (𝑉𝑉𝑖𝑖 ) could be
calculated as well.
𝑇𝑇 = 2𝜌𝜌𝜌𝜌𝑉𝑉𝑖𝑖2. (5)
Additionally the power exerted to the flow

from the propeller is [1]:
𝑇𝑇 𝑇𝑇 3/2
𝑃𝑃 = 𝑇𝑇𝑉𝑉𝑖𝑖 = 𝑇𝑇� = . (6)
2𝜌𝜌𝜌𝜌 �2𝜌𝜌𝜌𝜌
Fig. 4. Propeller test bed
Firstly the properties of a propeller are

measured to validate the theoretical model from eq.
(1) to (8). The chosen drone propeller is well known
APC 10’’x4,5’’ (Diameter=10’’, Pitch=4,5’’). The
propeller is shown in Fig. 5.
Fig. 3. Propeller in translational flight
In the case of translational non-axial flight the

flow field assumption is given in Fig. 3. The disc is
inclined at some angle of attack 𝜌𝜌 and the resultant
velocity across the propeller disc is 𝑉𝑉 �⃗ = 𝑉𝑉
��⃗ ��⃗
∞ + 𝑉𝑉𝚤𝚤 .
Hence the mass airflow in that case is 𝑚𝑚̇ = 𝜌𝜌𝜌𝜌𝑉𝑉
and propeller thrust in translational flight will be:
𝑇𝑇 = 2𝜌𝜌𝜌𝜌𝑉𝑉𝑉𝑉𝑖𝑖 . (7)
Alternatively from eq. (5) provided the thrust

of the propeller is known the induced speed can be
calculated:
𝑇𝑇
𝑉𝑉𝑖𝑖 = �2𝜌𝜌𝜌𝜌. (8) Fig. 5. Propeller APC 10x4.5
2. Wind Tunnel Propeller Test Bed

The Department of Transport and Aviation The results from the experiment are given
Engineering in the Technical University of Sofia, in Table 1. Here the angle of attack of the propeller
Plovdiv Branch is equipped with state-of-the-art disc is zero (𝜌𝜌 = 0). The variables are propeller
laboratory wind tunnel used for aerodynamic rotational speed in RPM and flow velocity. This
measurements both for fix-wing and rotary-wing represents the hover mode and translational non-
aircraft. There is a specially developed propeller axial flight. In hover mode (𝑉𝑉 = 0) thrust T,
test bed, mounted in the open test area of the tunnel consumed electric power 𝑃𝑃𝑒𝑒𝑒𝑒 , and jet slipstream
(Fig. 4) that allows orientation of the axis of the velocity 𝑉𝑉𝑗𝑗𝑒𝑒𝑗𝑗𝑗𝑗 are measured. Additionally using eq.
propeller at arbitrary inclination towards the vector (8) the induced speed (𝑉𝑉𝑖𝑖 ) is calculated, based on
of undisturbed velocity flow (the angle of attack of the measured thrust. This is to compare the induced
the propeller – 𝜌𝜌). The test bed is designed so that
II-266
speed based on momentum theory and actual same then eq. (7) should yield the propeller thrust
measured speed. for a given airspeed. However two problems arise
that influence the thrust and power of the propeller
Table 1. Propeller experimental data in translational flight. First at a given airspeed at
n, rpm 𝟒𝟒𝟒𝟒𝟒𝟒𝟒𝟒 𝟔𝟔𝟒𝟒𝟒𝟒𝟒𝟒 𝟕𝟕𝟕𝟕𝟒𝟒𝟒𝟒 zero angle of attack the flow field around the rotor
A=0 is not axisymmetric. There is a region with reversed
𝑇𝑇, 𝑁𝑁 3,15 7,20 9,45 flow defined by the regime coefficient:
𝑃𝑃𝑒𝑒𝑒𝑒 , 𝑊𝑊 34,885 100,41 162,41 𝜇𝜇 =

𝑉𝑉∞ 𝑐𝑐𝑐𝑐𝑐𝑐𝜌𝜌
, (9)
𝜔𝜔𝜔𝜔
𝑇𝑇 𝑑𝑑𝑑𝑑𝑁𝑁 9,03 7,17 5,82
, where 𝜔𝜔 is the rotational speed in 𝑟𝑟𝑑𝑑𝑑𝑑/𝑠𝑠. The
𝑃𝑃𝑒𝑒𝑒𝑒 𝑊𝑊
V=0
diameter of the reversed flow region is 𝑑𝑑 = 𝜇𝜇𝑅𝑅 [2].

𝑉𝑉𝑖𝑖 , 𝑚𝑚/𝑠𝑠 5,07 7,66 8,78 So it is as if at higher airspeed the actual radius of
the disc is reduced. This phenomenon has to be
𝑉𝑉𝑗𝑗𝑒𝑒𝑗𝑗𝑗𝑗 , 𝑚𝑚/𝑠𝑠 9,17 14,17 17,22 taken in account.
The second problem is that if the propeller
𝑉𝑉𝑖𝑖𝑒𝑒𝑒𝑒𝑒𝑒 , 𝑚𝑚/𝑠𝑠 4,58 7,08 8,61 disc were at a given angle of attack towards the
undisturbed flow then the axial component of the
𝑇𝑇, 𝑁𝑁 4,05 9,00 10,35 resultant flow velocity decreases the angle of attack
of the propeller blade elements, which reduces the
V=10 m/s
𝑃𝑃𝑒𝑒𝑒𝑒 , 𝑊𝑊 34,10 99,86 163,35

actual induced speed. So if formula (7) is used to
𝑇𝑇 𝑑𝑑𝑑𝑑𝑁𝑁 11,88 9,01 6,34 calculate propeller thrust in translational flight the
, abovementioned phenomena should be considered.
𝑃𝑃𝑒𝑒𝑒𝑒 𝑊𝑊
As we can see from Table 1 with the airspeed
𝑇𝑇, 𝑁𝑁 5,85 12,60 13,95 increased the resultant speed V is increased and so
the thrust is increased, however the reverse flow
area is also increased and if the propeller is at some
V=20 m/s
𝑃𝑃𝑒𝑒𝑒𝑒 , 𝑊𝑊 34,78 102,41 164,80

angle of attack (negative for forward flight) then the
𝑇𝑇 𝑑𝑑𝑑𝑑𝑁𝑁 16,82 12,30 8,46 induced velocity should be decreased.
,
𝑃𝑃𝑒𝑒𝑒𝑒 𝑊𝑊 We can take advantage of the blade element
theory in order to take in account the
abovementioned problems. The propeller thrust and
For the other non-hover mode cases at flow power are usually defined in terms of thrust and
velocity 10 m/s and 20 m/s (Table 1) the induced power coefficients – respectively 𝐶𝐶𝑇𝑇 𝑑𝑑𝑎𝑎𝑑𝑑 𝐶𝐶𝑃𝑃 :
and jet velocities are not measured due to obvious
difficulties. 𝑇𝑇 = 𝐶𝐶𝑇𝑇 𝜌𝜌𝑎𝑎2 𝐷𝐷4 (10)
It is apparent from Table 1 for the hover 𝑃𝑃 = 𝐶𝐶𝑃𝑃 𝜌𝜌𝑎𝑎3 𝐷𝐷5.
mode that the errors between the measured and
calculated induced speeds from the propeller are If we apply the first assumption that the
acceptably small, and the assumptions from the effective diameter of the propeller is reduced due to
momentum theory are reliable and applicable to reverse flow region with diameter 𝜇𝜇𝑅𝑅 and bear in
early design phases. mind that the momentum theory assumes even
distribution of the induced velocity across the
3. Methodology for determination of propeller disc then the following is true:
propeller characteristics.
In Table 1 induced velocities in hover mode 𝑇𝑇′ 𝐷𝐷−𝜇𝜇𝜔𝜔 4 𝜇𝜇 4
are easily measured and calculated provided the =� � = �1 − � ≡ 𝑘𝑘𝜇𝜇 , (11)
𝑇𝑇 𝐷𝐷 2
propeller thrust is a priori known. However the
calculation of the induced velocities in translational where T’ is the thrust of the propeller due to reverse
flight at some angle of attack is a difficult task. In flow region. In other words the thrust in eq. (7) has
the case of translational flight eq. (7) is used to to be corrected with coefficient 𝑘𝑘𝜇𝜇 equal to the ratio
calculate propeller thrust. If we have correct in eq. (11).
measurements of the propeller characteristics in The second problem – the reduction of the
hover mode then the induced speed can be induced velocity with the increase of axial
calculated using eq. (8) or measured directly. If we component of the resultant speed is connected with
assume that in translational flight the induced the blade element angle of attack or with the thrust
speeds for a certain rotational speeds remain the coefficient. If we assume that due to increased axial

II-267
velocity the initial blade angle of attack at hover the mechanical work exerted to the flow. In fact the
mode (𝜑𝜑𝐻𝐻 ) is decreased by ∆𝜑𝜑 then: ratio of the mechanical power to electric power
gives the efficiency of the propulsion (propeller,
𝑇𝑇′
=
𝐶𝐶𝑇𝑇 ′ 𝐶𝐶𝑇𝑇 𝜑𝜑 𝜑𝜑′
= 𝐶𝐶 =
𝜑𝜑𝐻𝐻 −∆𝜑𝜑
=1−
∆𝜑𝜑
, (12) electric motor and ESC) that is shown with the
𝑇𝑇 𝐶𝐶𝑇𝑇 𝑇𝑇 𝜑𝜑 𝜑𝜑𝐻𝐻 𝜑𝜑𝐻𝐻 𝜑𝜑𝐻𝐻 dotted line in Fig.7. In that manner the described
methodology is capable to predict the overall
where 𝐶𝐶𝑇𝑇 𝜑𝜑 is the derivative
𝜕𝜕𝐶𝐶𝑇𝑇
, which for a certain efficiency of the propulsion.
𝜕𝜕𝜑𝜑
airfoil is accepted to be a constant.
The decrease of the blade angle of attack
can be found if the velocity triangle for a blade
element is viewed:
∆𝑉𝑉𝑖𝑖 𝑉𝑉∞ sin 𝜌𝜌

tan ∆𝜑𝜑 ≈ ∆𝜑𝜑 =
𝜔𝜔𝜔𝜔
= = 𝜇𝜇 tan 𝜌𝜌. (13)
𝜔𝜔𝜔𝜔
Finally the ratio in eq. (12) will take the

form:
𝑇𝑇′ ∆𝜑𝜑 𝜇𝜇 tan 𝜌𝜌
=1− =1− ≡ 𝑘𝑘𝜌𝜌 . (14)
𝑇𝑇 𝜑𝜑𝐻𝐻 𝜑𝜑𝐻𝐻
In other words provided we use the induced

speed in hover mode to calculate the thrust in Fig.6. Thrust vs RPM & Airspeed
translation flight the formula in eq. (7) has to be
corrected with 𝑘𝑘𝜇𝜇 𝑑𝑑𝑎𝑎𝑑𝑑 𝑘𝑘𝜌𝜌 and the corrected thrust
will be:
𝑇𝑇 ′ = 2𝑘𝑘𝜇𝜇 𝑘𝑘𝜌𝜌 𝜌𝜌𝜌𝜌𝑉𝑉𝑉𝑉𝑖𝑖 (15)
Regarding to the propeller power from

Table 1 it is obvious that the required electric power
for all flight regimes including hover is almost
identical so with some precision the power of the
propeller in translational flight could be considered
approximate to the power at hover mode for a given
rotational speed (Fig. 7).
Fig.7. Power vs RPM & Airspeed
4. Results and discussion
To asses the accuracy of the hereby Once again Fig. 7 shows that the power
methodology, propeller characteristics can be consumed by the propeller at different flight modes
evaluated using formula (15) with the corrections is approximately the same.
(11) and (14). Figure 6 shows experimental data for So far the angle of attack of the propeller
the thrust from Table 1 (solid lines) compared to disk was considered to be zero. In order to check
momentum theory results, corrected with 𝑘𝑘𝜇𝜇 the assumption made in derivation of (14)
(dashed lines). The angle of attack of the propeller additional experimental and theoretical research is
is zero and the 𝑘𝑘𝜌𝜌 coefficient is zero respectively. It done. Propeller characteristics are measured at
can be seen (Fig. 6) that with the increase of the angle of attack 𝜌𝜌 = −25 𝑑𝑑𝑑𝑑𝑑𝑑 (forward flight). The
flow velocity the thrust generated by the propeller is experimental data are compared to theoretical
increased due to the increased mass airflow through (Table 2). The corrected thrust, calculated using eq.
the propeller disc. If uncorrected the predictions for (15) is shown in the last row, compared to the
the thrust are too high. In this case this is a experimental in the first row. Here the hover
satisfactory result. induced speed are used (𝑉𝑉𝑖𝑖ℎ ) to calculate the thrust.
Figure 7 shows the power of the propeller – This is also plotted in Fig. 8. The results show
both experimental (electric) and theoretical – acceptable convergence and accuracy.
according to momentum theory, calculated from the
induced speeds, corrected with 𝑘𝑘𝜇𝜇 . The measured
electric power (solid lines) is far bigger than the
theoretical, which gives the time rate of change of
II-268
Table 2. Propeller experimental data at The present paper deals with a methodology
A=-25 deg for determination of a multirotor drone propeller
n, rpm 𝟒𝟒𝟒𝟒𝟒𝟒𝟒𝟒 𝟓𝟓𝟒𝟒𝟒𝟒𝟒𝟒 𝟔𝟔𝟒𝟒𝟒𝟒𝟒𝟒 𝟕𝟕𝟕𝟕𝟒𝟒𝟒𝟒 characteristics at different flight modes. Specially
A=-25 deg derived coefficients are proposed to correct the
𝑇𝑇𝑒𝑒𝑗𝑗𝑗𝑗 , 𝑁𝑁 2,70 4,05 4,95 8,55 results from momentum theory. Theoretical results
are compared with experiment.
𝜇𝜇 0,18 0,14 0,12 0,10
𝑘𝑘𝜇𝜇 0,69 0,74 0,78 0,81 LITERATURE
1. Leishman J. Gordon (2006), Principles of

V=10
𝑘𝑘𝜌𝜌 0,57 0,65 0,71 0,76

Helicopter Aerodynamics, 826 p., Cambridge
𝑉𝑉𝑖𝑖ℎ , 𝑚𝑚/𝑠𝑠 4,58 5,97 7,08 8,61 Aerospace Series, USA.
2. W.Z. Stepniezki, C.N. Keys (1979). Rotary-
𝑇𝑇, 𝑁𝑁 6,22 8,59 10,71 14,03 Wing Aerodynamics, Vol.1, 380 p., NASA,
Philadelphia, USA.
𝑇𝑇 ′ , 𝑁𝑁 2,43 4,17 5,96 8,75
3. Homiao Huang, Gabriel Hofmann et al. (2008),
Aerodynamics and Control of Autonomous
Quadrotor Helicopters in Aggressive
Maneuvering, Stanford, California
Contacts:
Hristian Panayotov
TU - Sofia, Plovdiv Branch
Address: 25 Tsanko Dyustabanov Str,
Plovdiv, Bulgaria
Tel.:+359 32 659 518
E-mail: hristian@tu-plovdiv.bg
Stanimir Penchev
TU - Sofia, Plovdiv Branch
Address: 25 Tsanko Dyustabanov Str,
Fig. 8. Propeller thrust at V=10 m/s & A=-25 deg Plovdiv, Bulgaria
5. Conclusion Tel.:+359 32 659 632
E-mail: spenchev@tu-plovdiv.bg

СИСТЕМА ЗА ИЗМЕРВАНЕ НА НАЛЯГАНЕТО

ПРИ РАБОТА И ПОСТРОЯВАНЕ НА
ИНДИКАТОРНАТА ДИАГРАМА НА ДИЗЕЛОВ
ДВИГАТЕЛ
КРАСИМИР АМБАРЕВ, ВЪЛЬО НИКОЛОВ
Резюме: В настоящата статия е описана системата за индициране на дизелов

двигател чрез измерване на налягането при неговата работа и компютърната
обработка на получените резултати със създадения софтуер в средата MATLAB.
Представени са основните технически данни на всеки един от компонентите на
системата и са дадени връзките между тях. Извършени са измервания на
налягането при работа на дизеловия двигател и са получени индикаторни диаграми.
Ключови думи: дизелови двигатели, измерване на налягане, индикаторна диагарама
SYSTEM FOR MEASURING THE PRESSURE AT

WORK AND CONSTRUCTION OF INDICATOR
DIAGRAM OF DIESEL ENGINE
KRASIMIR AMBAREV, VALYO NIKOLOV
Abstract: This article describes the system for indicating of diesel engine by measuring the
pressure in its operation and computing the results obtained with the created software in
the MATLAB environment. The basic technical data of each of the components of the
system are presented and the connections between them are given. Diesel engine pressure
measurements have been made and indicator diagrams are obtained.
Key words: diesel engines, pressure measurement, indicator diagrams
1. Основни положения налягане AVL QC43D, едноканален усилвател и

Въз основа на измерването на налягането преобразовател AVL FI PIEZO, многоканален
на работното вещество по време на работа на аналогово-цифров преобразовател (АЦП) NI
ДВГ се построява така наречената индикаторна 6343, софтуер за запис и визуализация на
диаграма. От получените данни за налягането в получените резултати, разработен в средата
цилиндъра могат да се изчислят аналитично MATLAB.
мощностните показатели на изследвания
двигател при различни режими на работа, както
и да се анализира горивния процес в двигателя,
да се изследват различни явления, свързани с
етапите от протичане на горивния процес в ДВГ
[2], да се провери адекватността на математични
модели [3] и др. Фиг. 1. Общ вид на сензора за налягане QC43D
2. Система за индициране на ДВГ Сензорът за налягане QC43D (фиг. 1) е
Системата за измерване налягането на производство на фирмата AVL и е от категорията
работното вещество се състои от сензор за сензори, предназначени за развойна дейност в

II-270
областта на ДВГ [4]. Присъединяването на функция от ъгъла на завъртане на коляновия вал
QC43D към цилиндровата глава на изследвания за един работен цикъл или по-малък интервал
ДВГ се извършва посредством резбово (фиг.5), дефиниран предварително, както и
съединение M14х1,25. други параметри.
За защита на кварцовия елемент на
QC43D от прегряване е необходимо да се
охлажда чрез принудителна циркулация на
охлаждащата течност. Това гарантира дълъг
експлоатационен срок и висока достоверност на
получените резултати.
Някои основни данни от техническата
спецификация на сензора за налягане QC43D са
представени в таблица 1.
Таблица 1. Техническа спецификация на

AVL QC43D
Диапазон на измерване 0...200 bar
Максимално допустимо 250 bar Фиг. 2. Размери на сензора за налягане

работно налягане AVL QC43D
Чувствителност 71,32 Възможността за динамична drift

pC/bar компенсация позволява безпроблемна работа с
Линейност ≤ ±0.2 % дължина на кабела от сензора до устройството
до 20 m. От друга страна, здравата конструкция
Калибрирани диапазони 0 ... 80 bar на устройството позволява монтирането му в
на измерване 0 ... 140 bar непосредствена близост до изследвания
0 ... 200 bar двигател. Устройството може да се използва,
Собствена честота 50 kHz както за стендови измервания, така и за
измервания в пътни условия.
Размер на
присъединителната M14 x 1.25
резба
Присъединителен 10-32 UNF

куплунг Micro-Dot
Дебит на охлаждащата ≥ 20 l/h

течност
Маса 34 гр. (без

кабел)
Необходим въртящ 20 Nm
момент за затягане на
сензора
Фиг. 3. Общ вид на AVL FI PIEZO
Основните геометрични размери на Чрез ъпгрейд на фърмуера на AVL FI

сензора за налягане AVL QC43D, които трябва да PIEZO, предлагащ се като опция е възможно
се вземат в предвид при монтажа му са показани отчитане на максималното налягане в
на фиг. 2. цилиндъра при процеса горене. Тъка отчетената
Другият елемент от системата - AVL FI стойност се подава към CAN изхода на
PIEZO (фиг. 3) в конкретния вариант 2P2E устройството.
включва едноканален усилвател и Определянето на коефициента на
преобразовател на електрическия сигнал [4]. усилване A става по формула (1).
AVL FI PIEZO е снабден с 8000
A (1)
течнокристален дисплей, на който се S .pn
визуализира налягането в цилиндъра във където S е чувствителността на сензора, pC/bar
II-271
pn - максимална стойност на измервания инсталирано приложение IndiSignal или
сигнал, bar. посредством бутоните и дисплея от лицевия
Формула (1) е в сила при 0V офсет. При панел на устройството.
избор на 8V офсет, константата 8000 във
формула (1) се заменя с 16000.
Коефициентът на преобразуване от
електрически заряд в напрежение SCF се
изчислява по формула (2)
800 bar
SCF  , . (2)
S.A V
Основни данни от техническата
спецификация на AVL FI PIEZO са представени
в таблица 2.
Таблица 2. Техническа спецификация на Фиг. 4. Блокова схема на AVL FI PIEZO

AVL FI PIEZO
Аналоговият сигнал от измерването на
Маса 1,6 kg изхода на AVL FI PIEZO се подава към един от
диференциалните входове на АЦП на мулти-
Размери 86х109х271 mm функционалното устройство NI 6343,
9.5 V ... 36 VDC производство на фирмата National Instruments
Захранване или 100 ... 240 VAC [7].
чрез AC адаптер Връзката между NI 6343 (фиг. 5) и
10 W (при работа) компютъра се осъществява посредство USB
Консумирана мощност
20 W (при пускане) интерфейс.
Температурен
диапазон на работа -10°C...+60°C
SID (Идентификация
на сензора) не
Аналогови входове
(ел. заряд/напрежение) 1 брой
Фиг. 5. Общ вид на NI 6343
Диапазон на
измерване на 0 ... 14,400 pC Основни данни от техническата
електрически заряд спецификация на NI 6343 са представени в
таблица 3.
„Low-pass“ филтър 2 kHz, 5 kHz,
10 kHz, 20 kHz, Таблица 3. Спецификация на NI 6343
50 kHz или 100 kHz
Диапазон на Аналогови входа 32 еденични
измерване на -10 ... +10 V (16 броя диференциални)
напрежение Резолюция на
АЦП 16 bits
Резолюция на АЦП 12 bits
Скорост на
Drift компенсация постоянна или измерване 500 kS/s
циклична
Цифрово-аналогови Резолюция на
изхода 1 брой измерване 10 ns
Аналоговите изходи -10 ... +10 V, Диапазон на

на АЦП измерване ±0.2 V, ±1 V, ±5 V, ±10 V
BNC куплунг
Максимално вх.
Настройките на AVL FI PIEZO се правят напрежение ±11 V
от компютър, свързан чрез USB интерфейс и
II-272
Елемент от системата е и закрепения към файлове с разширение “mat”. Чрез създадения
края на коляновия вал на изследвания ДВГ диск скрипт в MATLAB е възможна обработка и
с един зъб, който при работа на двигателя визуализация на резултатите от измерването на
генерира импулси в намиращия се в близост до по-късен етап.
диска индуктивен сензор (фиг. 6). Тези
синхронизиращи импулси се подават към друг
от аналоговите входове на NI 6343 и служат за
синхронизиране не измерванията по горна
мъртва точка (ГМТ).
Фиг. 7. Визуализиране в реално време на

налягането на работното вещество на
течнокристалния дисплей на AVL FI PIEZO
Фиг. 6. Разположение на индуктивния сензор за За допълнителна възможност за

ГМТ 1 – диск с един зъб; 2 – индуктивен сензор. визуализация в реално време на сигнала на
изхода на AVL FI PIEZO при измерванията
Индуктивният сензор е свързани към служи свързан към аналоговия му изход цифров
NI 6343 посредством високочестотен екраниран осцилоскоп (фиг. 8).
проводник - тип RG58U (с вълново
съпротивление 50 Ω), което гарантира
минимални загуби при преноса на
високочестотния електрическия сигнал, както и
минимални смущения в измервания сигнал.
Техническите данни за индуктивния сензор са
показани в таблица 4.
Таблица 4. Техническа
спецификация на индуктивния сензор
Производител Автоприбор
Модел 191.3847
Минимална амплитуда на
променливото напрежение при
честота на въртене на вала ≤ 0,28 V
30±5 min-1 и разстояние от
диска 1,4±0,05 mm Фиг. 8. Визуализиране на налягането на
Максимална амплитуда на работното вещество за работен цикъл на
променливото напрежение при цифров осцилоскоп
честота на въртене на вала ≤ 250 V
7000±30 min-1 и разстояние от 3. Работа на системата
диска 1,4±0,05 mm За проверка на работоспособността,
системата за индициране е монтирана на
Създадената програма в приложението експериментална уредба [1], състояща се от
Simulink в средата MATLAB [6] позволява дизелов двигател и натоварващо устройство -
визуализиране в реално време и възможност за генератор за променлив ток (фиг. 9). Стендът е
съхраняване на данните от измерванията във разработен на базата на дизелов агрегат за
II-273
променлив ток, производство на фирмата KIPOR Сензорът за налягане AVL QC43D е
(Китай) - модел „KDE 6500T“ [5]. монтиран на цилиндровата глава на двигателя в
специално изработен за целта отвор (фиг. 11).
Фиг. 11. Разположение на сензора за налягане

AVL QC43D
Местоположението на отвора към

сензора за налягане е показано на фиг. 12.
Фиг. 9. Общ вид на експерименталната уредба

1 - дизелов двигател; 2 - генератор;
3 - контролен панел; 4 – горивен резервоар;
5 – динамометър
За натоварване на електрическия
генератор и свързания с него дизелов двигател
се използва външен електрически товар, показан
на фиг. 10. Електрическият товар представлява
седем на брой нагреватели, свързани успоредно
в две електрически вериги.
Фиг. 12. Разположение на отвора в
цилиндровата глава към сензора за налягане
Снемането на индикаторните диаграми

при този стенд се осъществява при
стабилизирана честота на въртене n=3000 min-1,
тъй като двигателят задвижва електрически
генератор за променлив ток с честота на
генерираното напрежение f = 50 Hz.
За осигуряване на сравнимост на
получените резултати при различна стойност на
електрическия товар е необходимо
индикаторните диаграми да се снемат при:
- еднаква регулировка (според
предписанията на фирмата производител) на:
газоразпределителния механизъм (хлабини в
кинематичната верига) и горивната уредба на
двигателя (началния ъгъл на изпреварване на
Фиг. 10. Външен електрически товар
впръскването и налягане на впръскване pвп);
Регулирането на големината на тока през - еднакво топлинно състояние
електрическия товар се осъществява чрез два (температура на маслото в картера Тм) на
регулатора на мощност. Потенциометърът на двигателя;
единия е разположен на контролния панел на - еднакви атмосферни условия (снемане на
стенда, а другият е външно разположен. индикаторните диаграми в един ден).

II-274
Визуализация на резултатите от 4. Изводи
извършените измервания при големина на Представената система за индициране на
електрическия ток през товара 20А за шест ДВГ, изградена на основата на система за
поредни измервания са представени на фиг. 13. измерване на налягането в цилиндъра на
двигателя на фирмата AVL, чийто възможности
Cylinder Pressure са разширени с опции за визуализация на
резултатите от измерванията на екрана на
70
60
компютър, възможност за запис на резултатите
от измерванията и последваща тяхна обработка.
50 Системата позволява изследване на бутални
двигатели с вътрешно горене чрез снемане на
40
индикаторна диаграма и автоматизирана
обработка на резултатите от измерването.
p, bar
30
Представената система може да се използва,
20 както за научни изследвания, така и при
обучението на докторанти и студенти.
10
0
-10
0 720 1440 2160 2880 3600
 , deg
1. Амбарев К., Николов В., Димитров Е.,

Фиг. 13. Индикаторна диаграма на двигателя Панчев С., Ташев А. (2014). Уредба за
при пълно натоварване за шест последователни експериментално изследване на дизелов
работни цикъла двигател. Научно списание "Механика на
машините", год. XХII, кн. 2, бр. 106, стр. 8-
Усреднена стойност за налягането в
12.
цилиндъра на двигателя за 20 последователни
2. Zaccardi, J., Duval, L., and Pagot, A. (2009).
работни цикъла при работа на двигателя са
Development of Specific Tools for Analysis and
представени на фиг. 14.
Quantification of Pre-ignition in a Boosted SI
Cylinder Pressure Engine. SAE Int. J. Engines 2(1):1587-1600,
70
2009, doi:10.4271/2009-01-1795.
60
3. Yousefi A., Birouk M. (2016). Numerical study
of the performance and emissions
50 characteristics of natural gas/diesel dual-fuel
engine using direct and indirect injection
40
systems. Proceedings of Combustion Institute -
Canadian Section Spring Technical Meeting
p, bar
30
University of Waterloo, May 10-12.
20 4. https://www.avl.com
5. https://www.kipor.com
10
6. https://www.mathworks.com
0 7. http://www.ni.com/en-rs.html
-10
0 180 360 540 720
, deg
Контакти:
Фиг. 14. Индикаторна диаграма на двигателя Красимир Амбарев, Въльо Николов
при пълно натоварване за един работен цикъл Технически университет-София,
филиал Пловдив
Създадените програми позволяват избор Адрес: гр. Пловдив
на броя работни цикли при работа на двигателя ул. Цанко Дюстабанов, 25
на установен режим, за които се извършва Телефони: 032/659 518, 032/659 555
измерването, записването и усредняване на E-mail:
стойностите на налягането в работното kambarev@tu-plovdiv.bg,
пространство и построяване на индикаторната vnikolov@tu-plovdiv.bg
диаграма.
ПОКАЗАТЕЛИ ОЦЕНЯВАЩИ
ГАЗООБМЕННИТЕ ПРОЦЕСИ ПРИ
ДВИГАТЕЛИТЕ С ВЪТРЕШНО ГОРЕНЕ
РАДОСЛАВ КОСТОВ
Резюме: В статията са анализирани, показателите оценяващи газообмения процес,

по които се осъществява усъвършенстването на газообмена при двигатели с
вътрешно горене с газотурбинно пълнене.
Ключови думи: двигател с вътрешно горене, газообмен, газотурбинно пълнене
INDICATORS EVALUATING GAS EXCHANGE

PROCESS IN COMBUSTION ENGINES
RADOSLAV KOSTOV
Abstract: The article analyzes the main physical principles and indicator of gaseous
exchange processes and leads to the improvement of the gas exchanges in the combustion
engines turbocharged.
Key words: combustion engines, gas exchange, turbocharged
1. Основни положения На фиг. 1 е представена индикаторна

Основните критерии за подобряването на диаграма на двигател с вътрешно горене и са
газообменните процеси са качеството на отбелязани фазите на газоразпределение.
напълване и очистване на цилиндрите.
Мощността на двигателя при равни други
условия е толкова по-голяма, колкото
количеството на прясно работно вещество,
намиращо се в цилиндъра след завършване на
напълването e по-голямо. Количеството на
прясно работно вещество зависи от
подобряването на очистването на цилиндъра от
продуктите на горене. Част от обема на
цилиндъра се заема от отработилите газове,
които не са успели да напусната цилиндъра. Фиг. 1. Индикаторна диаграма на
четиритактов двигател
2. Цел на анализа
Целта на настоящата статия е да се От представената индикаторна диаграма
разгледат показателите, по които да се оценят за начало на процеса изпускане се приема
показателите на газообменните процеси при точката , а краят на процеса е точката .
двигателите с вътрешно горене с газотурбинно Началото на процесът пълнене е в точка , а
пълнене. края – в . Процеса на припокриване на
пълнителния и изпускателния клапан обхваща
3. Същност на проблема - .

II-276
На фиг. 2 е изобразена опростена схема Ако при четиритактовите двигатели се
на газообменните процеси на четиритактов пренебрегне продухването на горивната камера
двигател. поради незначително припокриване на
При четиритактовите двигатели с клапаните около горно мъртво положение
вътрешно горене без свръхпълнене поради (ГМП), може да се приеме, че в края на процеса
съпротивленията в пълнителния тракт изпускане обемът на остатъчните газове при
(разяснения за символите) ≥ . налягане и температура , е равен по обем на
горивната камера . Тогава от
характеристичното уравнение за състояние на
газовете се получава:
; (3)
= . (4)
Като се вземе под внимание

определянето на коефициента на пълнене
, както и това, че ( ) се
получава формулата:
Фиг. 2. Схема на газообменните процеси в ( )

. (5)
четиритактов двигател
Разсъждавайки върху получената
Очистването на цилиндъра се оценява формула може да се изкаже следното:
чрез коефициента на остатъчните газове ( ) – - с увеличаването на степента на
отношението на количеството киломолове сгъстяване, обемът на горивната камера се
остатъчни газове към количеството намалява и при други условия намалява и
киломолове прясно работно вещество : количеството на остатъчните газове - ;
- с повишаване на налягането на
. (1) остатъчните газове се увеличава и тяхната
плътност, а следователно и количеството им,
Ако количеството на остатъчните газове поради което се увеличава;
и прясното работно вещество се изразят в - налягането на остатъчните газове
килограми то: нараства при увеличаване на съпротивлението в
изпускателната система и на честотата на
, (2) въртене на кояновия вал;
- при увеличаване на температурата на
където: са съответните молекулни маси; остатъчните газове намалява тяхната
, – прясното работно вещество, плътност, а следователно и количеството им,
действително постъпило в цилиндъра с поради което намалява;
параметри т.а – налягане и температура , - температурата на остатъчните газове
т.е с плътност . зависи от степента на сгъстяване и от работния
Действителното количество прясно режим на двигателя. Като се увеличава ,
работно вещество, което постъпва в цилиндъра увеличава се и степента на разширяване на
се отличава значително от теоретичното, защото продуктите на горене, поради което намалява.
налягането в края на напълването при всички С увеличаване на натоварването и честотата на
четиритактови двигатели е по-малко от въртене на двигателя се повишава
налягането в пълнителния тръбопровод , в температурата на нагретите повърхнини, до
следствие на наличието на хидравлични загуби в които газовете в цилиндъра се допират, и
пълнителния тракт. Температурата на прясното температурата се повишава.
работно вещество в цилиндъра в края на При бензиновите двигатели, където
пълненето е по-висока , тъй като по време на регулирането е количествено, при намаляване на
напълването се загрява от стените на цилиндъра, натоварването и има наличие на дроселна клапа
чиято температура е по-висока от . - се притваря, количеството прясно работно
вещество намалява, поради което се
увеличава. При дизеловите двигатели
II-277
регулирането е качествено и при намаляване на Степента на запълване на цилиндъра с
натоварването се увеличава, в резултат на прясно работно вещество се характеризира с
което намалява. При двигателите със коефициента на пълнене . При двигатели без
свръхпълнене с увеличаване на налягането на свръхпълнене представлява отношението на
прясното работно вещество се намалява. количеството прясно работно вещество
При двутактовите двигатели ( ) (в kmol или kg), действително постъпило
коефициентът на остатъчните газове зависи в цилиндъра (с параметри на т. a – налягане и
главно от системата на продухване и честотата температура , т.е с плътност към
на въртене на коляновия вал. теоретичното количество ( ), което може да
Коефициентът на остатъчните газове е в запълни работния обем на цилиндъра, при
следните граници: условие е има параметрите на околната среда
- за четиритактови двигатели: (налягане и температура , и плътност ),
бензинови без принудително пълнене т.е.
дизелови без принудително пълнене . (6)
бензинови и дизелови с свръхпълнене Количеството от прясното работно

вещество, постъпило в цилиндъра през
- за двутактови двигатели: пълнителния клапан по време на продухването
с картерно продухване не остава цялото в цилиндъра. Част от него
изтича през изпускателния клапан в
с правотоково клапанно продухване изпускателния тръбопровод и не участва в
осъществяването на работния цикъл на буталния
с правотоково продухване през прорези двигател.
Коефициента на пълнен може да се
с възвратно продухване през прорези определи по два начина:
1. Определяне на от уравнението за
В Таблица 1 са представени баланса на работното вещество в края на
ориентировъчни зависимости на основните пълненето – т. а.
параметри на газообменния процес, определени
експериментално при работа на двигателя на
( ) – без свръхпълнене (7)
номинален режим [3].
Таблица 1 и
Четиритактов двигател с Двутактов ДВГ 2. Определянето на от уравнението за
вътрешно горене (ДВГ) с правотоково
Показател
продухване
баланса на топлината в края на пълненето
Бензинов Дизелов през прорези
Коефициент 0,04…0,08 0,03…0,06 0,03…0,10 ( ) – без
на остатъчни
газове свръхпълнене (8)
Налягане на 0,105…0,120 0,105…0,120 0,105…0,120
остатъчните
газове Отношението на количеството въздух
Температура 900…1000 600…900 600…900 ( ) постъпило в цилиндъра по време на
на
остатъчните
продухването и напълването, към количеството
газове прясно работно вещество ( ), се нарича
Коефициент 0,75…0,85 0,8…0,9 0,75…0,85* коефициент на продухване - . Той отчита
на пълнене
Налягане в 0,085…0,09 0,085…0,095 (0,85…0,105)рк изменението на количеството остатъчни газове
края на от забавянето на затваряне на изпускателния
пълненето клапан:
Подгряване на 0…25 20…40 5…10
прясното
работно . (9)
вещество
Температура в 320…380 310…350 320…400
края на Реципрочната стойност на коефициента
пълненето
на продухване се нарича коефициент на
* - отнесено към пълния обем на цилиндъра
използване на продухвания въздух:
. (10)

II-278
Този коефициент определя частта въздух – количеството въздух, теоретично
останал в цилиндъра от цялото количество необходимо за пълното изгаряне на един
постъпило в цилиндъра по време на газообмена. килограм гориво, kg/kg.
Коефициента на продухване се изменя в
граници (при идеално продухване 4. Заключение
), т.е. в цилиндъра няма остатъчни газове. Оценката на газообмена, направена само
Максималните стойности на по качеството на очистването и напълването е
коефициента на пълнене се определят при пълно недостатъчна. Протичането на газообмена е
натоварване на четиритактовите двигатели без неразривно свързано и с други процеси,
свръхпълнене (при честота на въртене, при съставящи действителния цикъл на двигателите
която се постига максимален въртящ момент) се с вътрешно горене.
намират в следните граници: Някой от коефициентите,
- за бензинови двигатели характеризиращи газообмена са директно
- за дизелови двигатели . свързани с температурните напрежения в такива
детайли на двигателя като изпускателни
Факторите, които оказват влияние върху клапани, бутало и др. При избора на параметри
коефициента на пълнене са: и показатели за протичане на газообмена при
1. Влияние на степента на подгряване на съвременните високооборотни двигатели е
прясното работно вещество и средната му необходимо да се отчитат и факторите на
скорост в проходното сечение на пълнителния надеждност и дълговечност на двигателите.
клапан;
2. Влияние на фазите на ЛИТЕРАТУРА
газоразпределение;
3. Влияние на режимите на работа на 1. Бояджиев К.Г. и др. (1990),
двигателя; Автотракторни двигатели, Техника,
4. Степен на сгъстяване; София;
5. Налягане в края на пълненето; 2. Орлин А.С. и др. (1971), Теория робочих
6. Продухване на горивната камера; процессов поршневых и комбинировоных
7. Температура на остатъчните газове; двигателей, Москва;
8. Коефициент на остатъчните газове; 3. Орлин А.С. и др. (1973). Системы
9. Загряване на прясното работно поршневых и комбинировонных
вещество. двигателей, Москва.
4. Луканин В.Н. и др. (2007), Двигатели
Коефициент на въздушно отношение е внутреннего сгорания – кн.1, Москва.
отношението на разхода на въздух на двигателя
към разхода на въздух, теоретично необходим за
пълното изгаряне на горивото Контакти:
Радослав Костов
Технически университет - София
, (11)
Адрес: гр. София, 1756
бул. ,,Климент Охридски“ 8
където: е часов разход на въздух, kg/h; Учебен блок 9, Факултет по транспорта
– часов разход на гориво, kg/h; E-mail: radi@tu-sofia.bg
ВЛИЯНИЕ НА КОНСТРУКТИВНИТЕ И
ЕКСПЛОАТАЦИОННИ ФАКТОРИ НА
ГАЗООБМЕННА ПРИ ДВИГАТЕЛИТЕ С
ГАЗОТУРБИННО ПЪЛНЕНЕ
РАДОСЛАВ КОСТОВ
Резюме: В настоящата статията са анализирани, конструктивните и

експлоатационни фактори на газообмена при двигателите със свръхпълнене, които
се явяват решаващи за повишаването на мощностите и икономичните показатели
на двигателите с вътрешно горене с газотурбинно пълнене.
Ключови думи: двигател с вътрешно горене, газообмен, газотурбинно пълнене
INDICATORS EVALUATING GAS EXCHANGE

PROCESS IN COMBUSTION ENGINES
RADOSLAV KOSTOV
Abstract: In this article have been analyzed, the design and operational factors of the gas
exchange in the engine with turbocharging, which appear crucial for increasing the
capacity and economical performance in the combustion engines turbocharged.
Key words: combustion engines, gas exchange, turbocharged
1. Основни положения газорезпределение оказващи влияние не само

Конструктивните и експлоатационни към показателите на очистване на напълване на
фактори на газообмена, характеризиращи газа цилиндрите. Охлаждането на въздуха след
при постъпването им в цилиндъра и при компресора и влиянието на някои
напускането му от него, влияят на експлоатационни фактори върху газообмена при
газообменните процеси. Решаващ фактор за двигателите газотурбинно пълнене.
повишаване на мощностните и икономическите
3. Същност на проблема
показатели на двигателя е доброто конструиране
и експлоатация на двигателя. Това определя Конструктивни фактори:
газообмена, като обект на непрекъснати 1. Изисквания към пълнителните и
изследвания. Чрез принудителното пълнене изпускателните тръбопроводи.
съществено се увеличава ефективната мощност Сложността на газодинамичните
на двигателя, повишава се горивната процеси, протичащи в пълнителната и
икономичност и се намалява отделянето на изпускателната система на двигателите с
токсични вещества в отработилите газове. В газотурбинно пълнене, затруднява определянето
същото време няма съществено увеличаване на на необходимите размери на тръбопроводите
масата и габаритите на двигателя. само по аналитичен път. Проектирането на тези
системи се извършва основно на базата на
2. Цел на анализа експериментални изследвания.
Да се разгледат изискванията, които се Пълнителните тръбопроводи на
предявяват към пълнителните и изпускателните четиритактовите двигатели с газотурбинно
тръбопроводи. Влиянието на фазите на пълнене по конструкция са аналогични на
II-280
тръбопроводите на двигателите без Дефазирането е на 360°. В момента на
принудително пълнене. отваряне на изпускателния клапан, налягането в
При проектирането на изпускателните секцията на колектора рязко се повишава,
тръбопроводи за двигателите със свръхпълнене, достига максимума и започва да пада до
трябва да се отчитат следните изисквания: момента, в който започва изпускането от
1. Загубите на топлина и кинетична следващия цилиндър в същата секция. За да се
енергия от отработилите газове в тръбопровода намалят загубите от енергия при изтичане на
от изпускателните органи до направляващия газовете, обемът на секцията трябва да бъде
апарат на газовата турбина, трябва да бъдат възможно най-малък. Газовете от секцията на
минимални; колектора се отвеждат към отделни канали в
2. Процесът на изпускане от един корпуса на турбината и така се осъществява
цилиндър на двигателя не трябва да затруднява парциално отвеждане на работното тяло в
газообмена в другите цилиндри; направляващия апарат на турбината.
3. Обемът на изпускателния тръбопровод За осъществяване на най-ефективното
трябва да бъде минимално допустим с цел използване на разполагаемата енергия на
намаляване влиянието на инертността на отработилите газове, целесъобразно е
системата при работа на двигателя на преходни турбокомпресора да бъде поставен в
режими и за увеличаване на импулса на непосредствена близост с изпускателните
налягането при изпускане. клапани.
За изпълнението на тези условия, Размерите на изпускателния
изпускателните тръбопроводи се проектират с тръбопровод се избират по емпирични
минимална дължина без резки извивки и зависимости. Площта на напречното сечение на
промени в сечението. изпускателния тръбопровод при обединяване в
При използването на импулсна система една секция на колектора на две
на пълнене, газовете се изпускат в общ неприпокриващи се по фази изпускания, при
тръбопровод от не голямо число цилиндри. изпълнените конструкции за четиритактов
Изпускателния колектор се дели на части двигател с принудително пълнене е 1-1,3 от
(сектори), които не са свързани помежду си. максималната площ на проходното сечение на
Всяка част обединява цилиндри, при които изпускателния клапан. Площта на
процесът на изпускане се редува без направляващия апарат на турбината на един
припокриване, т.е. отместени са по фаза на не тръбопровод е 0,4-0,5 от максималната площ на
по-малко от продължителността на изпускане проходното сечение на изпускателния клапан.
(при четиритактовите двигатели 220° - 240° от Изменението на налягането в
завъртането на коляновия вал). При изпускателната система може да се установи
четирицилиндровия двигател една секция на ориентировъчно ако са определени
колектора обединява два цилиндъра (фиг. 1). конструкцията, разположението и размерите на
тръбопровода и скоростта на изпускане на
газовете. Средната скорост на газовете в
изпускателния колектор за комбинирани
двигатели е 70-100 m/s.
Изменението на налягането в
изпускателния тръбопровод (MPa) e
, (1)
Редуването на изпускателния: 1-3-4-2 или 1-2-4-3
Фиг. 1. Схема на разпределение на изпускателен където: е изменение на налягането,
колектор при импулсно пълнене съответстващо на местните съпротивления,
МРа:
, (2)
– коефициент на местни
съпротивления за отделните участъци от
тръбопровода;
– скорост в (m/s) на i-я участък;
– плътност на газа, kg/m3;
Фиг. 2. Изпускателни колектори при импулсно – изменение на налягането на i-я
пълнене [5, 6] участък на тръбопровода, МРа:
II-281
(3)
– коефициент на триене на i-я участък;

– дължина на i-я участък на
тръбопровода, m;
– диаметър на тръбопровода на i-я
участък, m.
Обемът на изпускателния тръбопровод с
газотурбинно пълнене обикновено е 0,8 – 2,0 от
Vh. За да се избегнат резонансните колебания на
стълба газове в тръбопровода, неговата дължина
трябва да удовлетворява съотношението: Фиг. 3. Диаграма на изпускането на
отработилите газове
, (4)
При закъсняването на затварянето на
изпускателния клапан след ГМП от една страна
където: l e дължината на изпускателния
се обезпечава достатъчно време-сечение за
тръбопровод, m;
изтичането на газовете от цилиндъра в процеса
– честота на изпускане в min;
на изпускане и от друга вследствие на това, че
a – скорост на звука при температура
количеството отработили
на газа в тръбопровода, m/s.
2. Фази на газоразпределение. газове може допълнително да изтече от
Фазите на газоразпределение оказват цилиндъра, а това води до намаляване на
влияние не само на показателите на очистване и остатъчните газове. При закъснението на
напълване на цилиндрите, но и на големината на затварянето на изпускателния клапан може да се
работата, необходима за работа на газообмена, използва и разреждането, образуващо се при
на температурата на детайлите, на условията на наличието на ускорена маса газ, което създава
работа на турбината и компресора и т.н. благоприятни условия за по-пълно очистване на
Предварителното отваряне на цилиндъра от продуктите на горенето.
изпускателния клапан преди долно мъртво От размерите на клапанната система,
положение (ДМП), обезпечава ефективното зависи характера на колебателните движения на
очистване на цилиндъра по време на свободното отработилите газове и това трябва да се отчете
изпускане, когато изтичането на газовете е в при избора на момента на затваряне на
следствие разликата на наляганията в цилиндъра изпускателния клапан.
и изпускателната система. То се определя от Ъгъла на отваряне на пълнителния
изискването време-сечение на свободното клапан се подбира така, че в момента, който
изпускане да е достатъчно за необходимото налягането на остатъчните газове в цилиндъра
понижаване на налягането в момента, когато стане по-ниско от това на пълнителната система,
буталото започва движение от ДМП към горно проходното сечение да бъде достатъчно за
мъртво положение (ГМП). При правилно избран постъпване на прясното работно вещество. По
ъгъл на отваряне на изпускателния клапан тази причина пълнителния клапан се отваря с
(точка 3 фиг. 3) се намалява загубата на енергия, изпреварване – т.е. преди достигането на
необходима за изтичането на газовете. От буталото в ГМП.
индикаторната диаграма на изпускането фиг. 3 В комбинираните двигатели при ⁄
се вижда, че при увеличаване на изпреварването , припокриването на клапаните е по-голямо в
(точка 2) загубената работа от разширението се сравнение с това на двигателите, при които
увеличава (площ 1241). При отваряне на клапана ⁄ . Голямото припокриване в този
в ДМП (точка 1) се увеличава работата за
изтласкване на газовете (площ 1451). случай се създава главно чрез увеличаване ъгъла
Показателите на двигателя ще бъдат оптимални, на предварително отваряне на пълнителния
ако клапана се отваря в точка 3. клапан, тъй като по-високото налягане в
пълнителния тръбопровод ( ) в сравнение с
налягането в изпускателния тръбопровод ( ),
непозволява при по-голяма част режимите
връщане на отработили газове в пълнителния
тръбопровод.

II-282
Затварянето на пълнителния клапан след значително по-висока степен на пълнене и се
ГМП т.е. със закъснение, подобрява явява най-ефективно и евтин способ за
напълването на цилиндъра с прясно работно увеличаване на мощността на двигателя с
вещество защото, първо проходното сечение на принудително пълнене. Трябва да се добави, че
пълнителния клапан остава голямо до края на охлаждането на въздуха способства и за
движението на буталото към ДМП и второ намаляване на съдържанието на вредните
налягането в началото на хода сгъстяване в компоненти в отработилите газове.
цилиндъра е по-малко от и през отворения
пълнителен клапан продължава постъпването на Експлоатационни фактори:
допълнително количество прясно работно При четиритактовите двигатели с голяма
вещество. степен на сгъстяване ε, коефициентът на
3. Охлаждане на въздуха след пълнене на цилиндъра е по-голям. Обяснява се с
компресора. това, че с повишаването на сгъстяването,
Нарастването на температурата на намалява обмена от камерата на горене Vc, който
въздуха в компресора зависи от степента на заемат остатъчните газове в края на
повишаване на налагането, КПД на компресора изпускането. Коефициентът на остатъчните
и топлообмена със стените, т.е. от газове се намалява, а коефициентът на
конструкцията на компресора. пълнене се повишава. При голяма степен на
Прилагането на газотурбинно пълнене с сгъстяване температурата на стените е по-ниска
променливо налягане в колектора може да се вследствие на по-пълното разширение на
разгъне в областта на по-високите от средните продуктите на горенето, което увеличава също
ефективни налягане, посредством въвеждане на коефициента на пълнене.
охлаждане на въздуха след компресора. Увеличаването на оборотите на
Понижаването на температурата на въздуха при коляновия вал при постоянно положение на
пълнене, довежда до увеличаване заряда на регулиращия орган, довежда до понижаване на
цилиндъра с въздух и понижаване на коефициента на пълнене, тъй като се намалява
температурата в началото на сгъстяването. За отношението ⁄ и нараства ⁄ при едно и
напълването на цилиндъра определящ се явява също (вследствие увеличаване на загубите на
плътността на заряда при пълненето енергия при протичането на газа през
пълнителния и изпускателния тракт).
( ) . (5) Произтичащото намаляване на заряда, което
( )
повишава коефициента на пълнене, не
компенсира намаляването му вследствие
хидравличните загуби. Текущото изменение на
От тук се вижда, че повишаването на
плътността при определени условия може да налагането е пропорционално на
бъде значително по-малко от повишаването на квадрата на оборотите n на коляновия вал и
налягането. Ориентировъчно може да се смята, обратнопропорционално на квадрата на площта
че при еднакво налягане на пълнене и на проходното сечение f1:
понижаване на температурата на въздуха след
компресора с 10°С, плътността на въздуха се , (6)
увеличава около 3%. Благодарение на това при
постоянен коефициент на пълнене и при където: е постоянна величина;
постоянен специфичен разход на гориво да се f1 – площ на най-тясното проходно
повиши мощността с 3%, а отчитайки сечение на тръбопровода.
останалите промени на различни параметри, Влиянието на натоварването на
произтичащи от понижението на температурата двигателя и на коефициента на пълненето е
на въздуха, мощността може да се повиши и до различно в зависимост от начина на регулиране
4,2%. на двигателя. Съпротивлението на пълнителната
Използвайки вода в качеството си на система независи от натоварването (при n =
охладител на въздуха след компресора в const). При всички натоварвания на двигателя
повечето случаи позволява без особено високи ( ⁄ ) , коефициентът на пълнене
загуби да се осъществи охлаждане на въздуха до зависи от изменението на интензивността на
температура, превишаваща само няколко подгряването на въздуха при напълване. В даден
градуса средната температура на водата. случай с увеличаването на натоварването на
Може да направим извода, че двигателя, коефициентът на пълнене малко се
охлаждането на въздуха на пълнене при равни понижава, тъй като отношението на
други критерии, позволява да се осъществи
II-283
температурата се намалява в резултата на ЛИТЕРАТУРА

по-интензивно подгряване от стените на
цилиндъра. 1. Цинер К., (1978), Наддув двигателей
внутреннего сгорания , Ленинград;
4. Заключение 2. Орлин А.С. и др. (1971), Теория робочих
От организацията на газообменния процессов поршневых и комбинировоных
процес, зависят мощностните и икономически двигателей, Москва;
показатели на двигателя. Влиянието на 3. Орлин А.С. и др. (1973). Системы
поршневых и комбинировонных двигателей,
конструктивните и експлоатационни фактори
Москва;
върху газообмена, го правят обект на 4. Куросов В.С. и др. (2013), Теория
непрекъснати изследвания, независимо от двигателей внутреннего сгорания ,
наличието на трудност свързани с КубГАУ;
нестационарността на подходите. При 5. www.full-race.com
двигателите с газотурбинно пълнене това е от 6. www.turbozenturu.de
значение за получаването на надеждни данни за
процесите и за системата на принудително
пълнене. Контакти:
Радослав Костов
Технически университет - София
Адрес: гр. София, 1756
бул. ,,Климент Охридски“ 8
Учебен блок 9, Факултет по транспорта
E-mail: radi@tu-sofia.bg

АКТУАЛНИ ГРАЖДАНСКО-ВОЕННИ ПРОЦЕДУРИ В ОВД ПРИ

ДЕЙСТВИЯ НА ВОЕННАТА АВИАЦИЯ ЗА ОСИГУРЯВАНЕ НА
ВЪЗДУШНИЯ СУВЕРЕНИТЕТ. ЧАСТ 1
ДАНЧО КОЛИБАРОВ
Резюме: Представени са гражданско-военните процедури и координация, като

част от обслужване на въздушното движение, при действия на военната авиация,
за осигуряване на въздушен суверенитет при различни видове вероятни
терористични заплахи.. Разгледан е процеса на оценка на опасността и механизма
на протичане на информационния поток.
Ключови думи: Гражданско-военни процедури и координация, обслужване на
въздушното движение, управление на въздушното движение, вероятни
терористични заплахи, въздушен суверенитет.
CURRENT CIVIL-MILITARY PROCEDURES IN AIR AT THE ACTION OF

THE MILITARY AIRCRAFT TO ASSURING AIR SURVEY. PART 1
DANCHO KOLIBAROV
Abstract: Civic-military procedures and coordination are presented as part of air traffic
services in military aviation activities to provide air sovereignty for various types of likely
terrorist threats. The process of assessing the hazard and the mechanism of flow of the
information flow is considered.
Key words: Civil-military procedures and coordination, air traffic services, air traffic management.
possible terrorist threats, air sovereignty.
1. Характеристика на гражданско- 1.1. Законова основа и регламентация

военната координация при управление на на функционирането на двете подсистеми.
въздушното пространство и обслужване на Функционирането на гражданската и
въздушното движение. военната подсистеми за УВД и ОВД се основава
Включва координация между на член 2(а), ал. 2 на Закона за гражданското
упълномощени органи по въпросите на въздухоплаване и Наредба № 19 на министъра
разделянето на ВП, за осигуряване на безопасно, на отбраната и министъра на транспорта и
ефективно и хармонизирано ползване на ВП. съобщенията от 27 януари 1999 г., определяща
Стратегическата координация между реда за въвеждане и правилата за работа на
граждански и военни органи представлява единната система за гражданско и военно
дефинирането и прегледа на държавната управление на въздушното пространство. С тази
политика по ВП, като се вземат под внимание наредба се определят основната цел, задачите,
потребностите на държавните и чуждестранните съставът и организацията на работа на единната
потребители. Предтактическата координация система за гражданско и военно управление на
между гражданските и военните органи е въздушното пространство в мирно време. В
работата по оперативното управление на дейността си единната система за гражданско и
съществуващите структури и процедури – военно управление на въздушното пространство
ежедневно разпределение на ВП. ползва принципите, залегнали в програмата на
Координацията в реално време е активирането, Евроконтрол за хармонизация и интеграция на
деактивирането и преразпределението на ВП, системите за управление на въздушното
както и по разрешаването на конфликти и движение, Концепцията на Евроконтрол за
проблемни ситуации. гъвкаво използване на въздушното пространство
и на препоръките на Регионалната инициатива
на САЩ за модернизация на организацията и
II-285
управлението на въздушното пространство в други органи за координация между
Централна и Източна Европа. сьответните РП.
Инструкция №24 за реда за даване на
1.2. Система за разпространение на заявки от ползвателите до Службата за
данни за ВП и гражданско-военна планиране на военното използване на
координация. въздушното пространство (СП ВИВП) определя
Разпространението на данни за ВП и редът за заявяване намерение за използване на
гражданско-военната координация са двата ВП от ползвателите.
главни компонента на системата. Гражданско- Тези ползватели на въздушното
военната координация се подразделя на два пространство, които дават заявки за използване
компонента, които са обмен на гражданско- на временно отделени зони в СП ВИВП, са:
военни полетни данни и координация при авиационните бази на военновъздушните сили
пресичане на ВП. и/или военноморските сили; щабът на отбраната
Оперативните изисквания за системата на Българската армия и командванията на
произтичащи от концепцията за ГИВП могат да сухопътните войски, ВВС и ВМС; началниците
се разделят на две категории: на гарнизоните, в които се провеждат
Първата категория изисква система за илюминации; ръководителите на местната
предтактическо УВП (денят преди деня на изпълнителна власт, физически и юридически
действието). Оперативните изисквания в тази лица, отговорни за организиране и провеждане
област са за система за разпространението на на взривни работи и/или илюминации.
данни за ВП към различните заинтересовани СП ВИВП е структурна единица от
органи. военното „Ръководство на въздушното
Националният план за използване на движение", която събира заявките от
ВП, определящ как ВП ще ползвателите за използване на ВОЗ.
бъде използвано през деня на полетите и Полетите и другите дейности във ВОЗ,
неговата актуализация - актуализиран план за заявени в СП ВИВП се планират и изпълняват
използване на въздушното пространство се само след осигуряване на безконфликтност с
изпраща до различните заинтересувани органи полетите на гражданската авиация, като
за УВД. Съобщение за действащите условни безопасността на полетите и другите дейности
маршрути/трасета, включено в международно във ВОЗ се осигуряват от органите на ВРВД.
обединено съобщение, изброява тези от тях, Заявките за ВП се събират от оторизираните
които ще бъдат приложени в Националният органи и се изпращат до Центровете за УВП.
план за използване на ВП. То също трябва да Плановете за използване на ВП се
бъде изпратено до всички заинтересувани изпращат всеки ден в определено време от
районни центрове за УВД, работните позиции центъра за УВП до Централния орган за
за управление на потоците ВД и до всички управление на потоците и до оторизираните
центрове за УВП. органи. Актуализирания план за използване на
Целта на системата е да минимизира ВП се изпраща всеки ден, във време
натоварването свързано с рутинни задачи. определено от Централния Орган за
Тази цел може да се постигне чрез Управление на Потоците ВД на адресите, на
използване на система за съобщения, която които е изпратен и плана за използване на
позволява автоматична проверка и въздушното пространство.
разпространение на данни и визуализация, както Актуализираният план за използване на
и обработващите подсистеми, където те могат да ВП съдържа само изменения за гъвкавите
се използват (работни позиции, АТС структури на ВП, описани в съответния план за
съоръжения и т.н.). използване на ВП. На фиг. 1 е показано
Втората категория изисква система за схематично как се осъществява обмена на данни
тактическа (в деня на полетите) военно- и информация – плана за използване на ВП
гражданска координация. В тази област се (АUP), актуализирания план за използване на
определят два типа изисквания за системата: ВП (UUP), съобщение за планиране
за обмен на полетни данни между гражданските използването на УМТ (CRAM) и др. между
и военни органи за ОВД и за координация при националните планиращи използването на ВП
пресичане на ВП. органи, органите за УВД, операторите на ВС
Практическата полза в тази област е (потребителите на ВП) и централизирания пункт
намаляване на натоварването на ръководител за планиране на потоците ВД (CADF).
полетите (РП), свързано с осигуряване на
информация за ОВД, осъществявано от
II-286
случаи се извършва чрез въвеждането на данни
от определен РП.
Кординацията при пресичане на ВП се
състои от обмен на съобщения между РП. Той
има избора или да уведоми другия орган, под
чийто контрол полетът ще премине през
контролираното от него ВП или да изиска
разрешение за пресичане на това ВП. Във
втория случай РП към втория орган ще разчита
да се въведе оперативен отговор за искането за
пресичане на ВП.
Тези оперативни отговори могат да са
или одобрения към искането, контра-
предложения или отказ. Контра-
предложението изисква одобрение или отказ от
РП, който е предложил това искане.
Съответните области на пресичане
може да са корегирани от РП, отговорен за
полета или от РП, отговорен за ВП, което ще се
пресича.
Целия обмен на полетни данни и на
координацията при пресичане на ВП се
потвърждават автоматично, ако могат да бъдат
Фиг. 1. Схема на разпространение на информацията правилно получени и обработени в адресирания
за ВП. орган.
Съобщение с условни маршрути или 1.3. Нормативна база за преминаване

трасета се изпраща всеки ден от централния под военен контрол.
орган за управление на потоците ВД /Централна Преминаването на националната система
функция за въздушна информация на ЕКГА до за УВД под военен контрол се основава на
изброени оператори на ВС, до всички участващи “Наредба № 23 за реда за използване на
районни центрове за УВД или работни позиции въздушното пространство при прехода към
за управление на потоците ВД и до всички военен контрол и ръководство на въздушното
центрове за УВП. То съдържа всички условни пространство при повишаващи се нива на
трасета предоставени в областта на страните от извънредни условия”.
ЕКГА по време на валидния период на В съответствие с посочената наредба,
плановете за използване на ВП. редът за използване на въздушното
Предтактическите съобщения за ВП пространство при прехода към военен контрол и
се очаква да се получават до определено време ръководство при повишаващи се нива на
всеки ден. Те не са време критични, извънредни условия, се определя от СУВП по
следователно не се изисква потвърждение. предложение на министъра на отбраната
Информационната система за гражданско- съгласувано с министъра на транспорта и
военна координация изпълнява следните включва последователността от дейности и
функции: правила за планиране, разпределение,
 Обмен на полетни данни; предоставяне и контрол на използването на
 Координация при пресичане на ВП. въздушното пространство и изпълнението на
Обменът на полетни данни полетите.
представлява предаването и приемането на Нивата на извънредни условия, при които
базова група информация, когато полетът е се извършва преходът към военен контрол и
навлязал в областта на приемащия орган, ръководство на въздушното пространство, са:
допълнена с подновена информация, докато - военно положение при заплаха от
полетът е през предаваното ВП. Допълненията въоръжено нападение или война на територията
могат да съдържат намерението на РП, когато на страната или на част от нея;
това е регламентирано. Наличието на тази - положение на война при въоръжено
информация ще намали нуждата от словестна нападение срещу страната;
координация РП-РП. Обменът на данни
преимуществено е автоматичен и в известни

II-287
- положение на война при необходимост потоците ВД на ЕВРОКОНТРОЛ в Брюксел за
от неотложно изпълнение на международни активирането и деактивирането на обемите ВП,
задължения; отделени за специфични нужди – ВОЗ, ГЗП,
- внезапно въоръжено нападение или ЗНК. Това позволява при възникване на
нахлуване на чужди войски на територията на определени ситуации или кризи, имащи
страната или на част от нея. продължителност повече от 24 часа да се
Повишаващите се нива на извънредни осъществи надеждна сепарация на гражданския
условия се обявяват с решение на Народното въздушен трафик, наричан още основен трафик
събрание или указ на президента на Р България от военния, или този, използван за специални и
по установения от Конституцията ред и се специфични задачи означаван като оперативен
оповестяват чрез средствата за масово трафик.
осведомяване. Трябва също да се отбележи, че
За начало на прехода към военен контрол гражданско-военната координация и обмен на
и ръководство на въздушното пространство при полетни данни и информация, осъществявани
повишаващи се нива на извънредни условия се между гражданските и военните служби за ОВД
приемат часът и денят на обявяването на осигуряват постигане на по-високи нива на
военното положение, на войната, ако не е сигурност и икономичност за гражданския
предхождано от обявяване на военно положение въздушен трафик от една страна и изпълнението
или часът и денят на началото на внезапното на учебните и бойни задачи от оперативния
въоръжено нападение. въздушен трафик
За край на военния контрол се приемат Друг важен извод е, че гражданско-
часът и денят на отменянето на военното военната координация се осъществява основно в
положение, часът и денят на обявяване края на две направления: първо – на принципа на
военните действия или времето на тяхното резервиране и отделяне на ВП за специфични
фактическо прекратяване. нужди и дейности; второ – по пътя на
Продължителността на прехода към подобряване на т.н. “тиха координация”. Тя
военен контрол се определя в директивата за може да се осъществява чрез въвеждане на
бойната готовност на въоръжените сили на Р ефективни комуникационни системи и системи
България. за обмен на данни в реално време, както и с
При преход към военен контрол на използването на еднозначна радиолокационна
въздушното пространство се извършва картина за всички потребители на
преструктуриране на системата за УВД. радиолокационна информация за ОВД.
Целта на прехода към военен контрол и Използването на такава автоматизирана система
ръководство на въздушното пространство при ще намали необходимостта от вербална
повишаващи се нива на извънредни условия е да координация и съответно ще редуцира времето
осигури изпълнението на задачите на за координация и вероятността от допускане на
държавните въздухоплавателни средства, както грешки.
и безопасността на полетите на гражданските
въздухоплавателни средства. 2. Предотвратяване на конфликтни
От направената характеристика на ситуации при УВД.
функционирането на военната и гражданската Ефективното функциониране на НС УВД
подсистеми за УВП и ОВД и на системата за се характеризира с намаляване на броя на
разпространение на данни за използването на конфликтните ситуации при опериране на
ВП и осъществяване на гражданско-военна военни ВС в гражданско ВП. Решаването на
координация може да се направят следните задачите по предотвратяване на конфликтни
изводи. ситуации за осигуряване на безопасността на
На първо място трябва да се отбележи, че полетите на оперативното и основно ВД е една
системата дава възможност за гъвкаво реагиране от основните функции на военните РП.
както при управлението на ВП – отделяне на Към конфликтните ситуации при ОВД се
определени обеми ВП за специфични цели или отнасят:
използване на УМТ в зависимост от  опасно сближение на ВС в полет;
възможността за планиране, така и при  попадане на ВС в опасни за полета
обслужване на ВД – координация за полетите на МТО условия;
гражданските и военните ВС.  попадане на ВС в зони забранени
От изключителна важност е за полети.
своевременното оповестяване на Решение за предотвратяване на създала се
централизирания орган за управление на конфликтна ситуация се взема в случаите,
II-288
когато величината и характера на текущите или От фигурата се вижда, че скоростта на
екстраполирани параметри на движението на ВС маневриращото ВС - Vман може да се опердели
застрашават безопасността на полетите. по следната формула:
В практиката по регулиране на ВД най- Vман 
Д 
(1),
t
често се решава задачата по предотвратяване на
опасно сближение между ВС. Ето защо задачата където: Д – разстояние от текущото място на ВС
по предотвратяване на конфликтна ситуация „А” до разчетната точка на среща;
може да се разглежда като опасно сближение  - радиус на зоната на вероятните
между ВС. стълкновения.
За решаване на задачата по отстраняване Следователно ВС, изпълняващо полет със
на конфликтна ситуация е необходимо да се скорост Vман и в момента на прелитане на
знаят положенията на ВС, направлението на конфликтната точка F, другото ВС ще се намира
движението им, времето на полета и на границата на зоната на вероятните
разстоянието до точката на пресичане на стълкновения, т.е. или в точка 1, или в точка 2.
траекториите им с отчитане на вероятните За отстраняване на конфликтна ситуация
отклонения за сметка на неточното поддържане чрез изменение на направлението на полета при
на режима на полета и влиянието на вятъра. постоянни скорост и височина, схемата на
Конфликтните ситуации се маньовъра се разчита и изпълнява така, че
предотвратяват чрез: траекторията на маневриращото ВС да минава
 изменение на височината на извън границите на зоната.
полета на ВС; Ако областта на вероятните стълкновения
е на голямо разстояние от изходната точка А,
 маневриране по скорост на едното
маньовъра по направление е целесъобразно да се
или двете ВС;
изпълни чрез изменение на курса под ъгъл
 изменение на направлението на
,определен от следната зависимост:
полета на едното или двете ВС при
постоянни скорости. 
  arcsin (2),
Скоростта за маневриране на военното ВС Д
се подбира така, че в момента на преминаване където: Д – разстояние от текущото място на ВС
на едното ВС през изчислената точка на „А” до разчетната точка на среща,  - радиус на
конфликтна ситуация (фиг. 2), разстоянието зоната на вероятните стълкновения.
между двете ВС да не е по-малко от На ВС, намиращо се на по-голямо
определения радиус на зоната на вероятни разстояние от точката F, се осигурява прелитане
стълкновения . извън зоната.
На фиг. 2 е показана схема на Когато разстоянието до зоната е малко,
маневриране по скорост и направление на две необходимо е да се изпълни вираж от зададената
ВС. траектория във възможно най-кратко време,
С (точка В).
Виражът трябва да се изпълни при
Rm
in
известни ограничения. Минималният радиус на
1 
виража се определя по следното съотношение:
А
В V2
R min  (3)
g  tg mah
F
При това е необходимо да бъде спазено
Д- следното условие:
3 Rmin  Д   (4),
r
2 където: Д – разстояние от текущото място на ВС
„А” до разчетната точка на среща,  - радиус на
зоната на вероятните стълкновения.
Предотвратяването на възможността за
Д възникване на повторна конфликтна ситуация и
осигуряване на безопасността на полета при
разделяне на ВС е необходимо виража на ВС да
Фиг.2. Маневриране по скорост и се изпълни на страната на намаляване на
направление от ВС курсовия ъгъл на визиране.

II-289
Един от пътищата за намаляване на броя доп.ДВ, бр.34 от 2001.Инструкция
конфликтните ситуации е представяне на по- №24/01.11.1999г. за работа на Центъра
високи изисквания към техническите средства за за планиране и разпределяне на
контрол и навигация, както на земята, така и на въздушното пространство на Република
борда на ВС. Изпълнението на тези изисквания България, Обн., ДВ, бр.101 от 1999.
доведе до усъвършенстване на организацията на 3. Инструкция №25/01.11.1999г. за
структурите на ВП и повишаване на капацитета структурата и правилата за работа на
на зоните и районите за ОВД. Центъра за координиране използването
В изпълнение на програмата на на въздушното пространство на
ЕВРОКОНТРОЛ за намален минимум за Република България, Обн., ДВ, бр.101 от
вертикална сепарация (RVSM) от 300 м (1000 ft) 1999.
между полетни нива FL290 и FL 410, 4. Наредба №19/27.01.1999г. за реда за
гражданските ВС бяха оборудвани със въвеждане и правилата за работа на
съвременни бордни системи за предотвратяване единната система за гражданско и
на сблъсквания. военно управление на въздушното
Бордната система за предотвратяване на пространство, Обн., ДВ, бр.13 от 1999.
сблъсквания ACAS 7 е система, която 5. Наредба №11/05.05.1999г. за обслужване
анализира информацията за ВД през 300 м (1000 на въздушното движение във
ft) до полетно ниво FL 410, а над него през 600 м въздушното пространство на Република
(2000 ft). България, Обн., ДВ, бр.50 от 1999.
Информацията се обменя чрез 6. Наредба №2/10.03.1999 г. Правила за
транспондерите на ВС в радиус от 50 мили. При провеждане на полети на Министъра на
наличие на потенциално конфликтна ситуация и транспорта и съобщенията – Обн., ДВ,
след изтичане на време 45 s екипажа на ВС не бр.50 от 1999.
предприеме мерки, то апаратурата подава 7. Пенев П. Б., Ячев Р. Я., Основи на
команди на автопилота за отстраняване на теорията и използването на
конфликтната ситуация. авиационните навигационни системи и
Следователно около ВС се описва средства,учебник, ВА “Г. С. Раковски”,
цилиндър с радиус 6 км при обслужване по 2004.
контрол на подхода и 18,5 км при обслужване по 8. Унгорян С.Г., Маркович Е.Д., Волевач
контрол на района и 300 м (1000 ft) под и над А.И, Анализ и моделирование систем
него до полетно ниво FL 410. управления воздушным движением,
Възможно е получаване на препокриване Транспорт, М., 1980.
на части от различни цилиндри, при което 9. Ячев Р. Я., Управление на въздушното
възниква риск от конфликтна пространство по време на кризи и военен
ситуация.Съществуват следните четири нива на конфликт, Лекция, ВА „Г. С. Раковски”,
риска: С., 2002.
 риск от сблъскване; 10. Procedures for air navigation services; Doc
 безопасността не е осигурена; 8168-OPS/611, Vol II, IKAO.
 отсъства риск от сблъскване; 11. Airspace Management Modernization for
 рискът не е определен. Central and Eastern Europe, Volume II –
Държавните ВС не са оборудвани с Annex K Air Sovereignty Modernization
бордната система за избягване от сблъскване Recommendation for Bulgaria, sept. 1997.
ACAS 7 . По тази причина е необходимо да се 12. Наредба № 23 от 1.11.1999 г. за реда за
увеличи нормата за радарно сепариране между използване на въздушното пространство
държавните и гражданските ВС. при прехода към военен контрол и
ръководство на въздушното
ЛИТЕРАТУРА пространство при повишаващи се нива
на извънредни условия, издадена от
1. Заеков В.С., Чилингиров Д. И., Теория за министъра на транспорта и министъра
управление на въздушното движение, на отбраната, обн., ДВ, бр. 99 от
учебно пособие, ВИ, София, 1992. 16.11.1999 г.
2. Закон за гражданското въздухоплаване,
Обн., ДВ, бр. 94 от 01.12.1972г., изм. и
Данчо Колибаров
доп. бр. 30 от 1999г., бр. 16 от 1997г., бр.
ТУ- София, филиал Пловдив
85 от 1998г., в сила от 01.01.1999г., бр.
E-mail: danchokol@abv.bg
12 от 2000г.,в сила от 15.02.2000г., изм.
АКТУАЛНИ ГРАЖДАНСКО-ВОЕННИ ПРОЦЕДУРИ В ОВД ПРИ

ДЕЙСТВИЯ НА ВОЕННАТА АВИАЦИЯ ЗА ОСИГУРЯВАНЕ НА
ВЪЗДУШНИЯ СУВЕРЕНИТЕТ. ЧАСТ 2
ДАНЧО КОЛИБАРОВ
Резюме: Представени са гражданско-военните процедури и координация, като

част от обслужване на въздушното движение, при действия на военната авиация,
за осигуряване на въздушен суверенитет при различни видове вероятни
терористични заплахи.. Разгледан е процеса на оценка на опасността и механизма
на протичане на информационния поток.
Ключови думи: Гражданско-военни процедури и координация, обслужване на
въздушното движение, управление на въздушното движение, вероятни
терористични заплахи, въздушен суверенитет.
CURRENT CIVIL-MILITARY PROCEDURES IN AIR AT THE ACTION OF

THE MILITARY AIRCRAFT TO ASSURING AIR SURVEY. PART 2
DANCHO KOLIBAROV
Abstract: Civic-military procedures and coordination are presented as part of air traffic
services in military aviation activities to provide air sovereignty for various types of likely
terrorist threats. The process of assessing the hazard and the mechanism of flow of the
information flow is considered.
Key words: Civil-military procedures and coordination, air traffic services, air traffic management.
possible terrorist threats, air sovereignty.
1. Установяване на гражданско-военни жизненоважно значение за последващите

процедури по управление на въздушното действия се приема, че националната система за
движение. УВД ще допълва интегрираната развърната
система на НАТО за ПВО (NATO Integrated
От изложеното в част първа възниква
Extendent Air Defence System - NATINEA, DS),
въпросът, дали така представената НС УВД ще
когато това е възможно.
бъде способна да поддържа и осигурява
Временното ограничено ВП за
военните изисквания за организиране и
провеждане на хуманитарни, спасителни и др.
провеждане на операции, и нарастването на
операции различни от война, трябва да бъде
гражданския трафик в едно и също време. За да
осигурено с непрекъснат радиолокационен
удовлетвори тези изисквания НС УВД, трябва да
контрол и обмен на данни между гражданските
осигури максимални нива на ефективност при
и военните центрове. Границите му трябва да
ръководството и подобряване на стандартите за
бъдат известни на гражданските органи за ОВД
сигурност. Въпреки това в определени части от
и координирани с ЕВРОКОНТРОЛ и съседните
ВП и в определени периоди това може да не е
РЦ за ОВД.
възможно. Това налага стриктно спазване на
При заявяване на височините на
сепарацията между основното и оперативното
полетите, на които ще се изпълняват
ВД, а при определени условия забрана за
поставените задачи, сектора за ОВД може да
полетите на граждански въздухоплавателни
бъде ограничен по височина (напр. в долното
средства в дадена част от ВП.
ВП). На фиг. 1 са показани минимумите за
От ключово значение е да се установят
радарно сепариране осигуряващи безопасност
националните правила по УВД при операции
между оперативния и основния трафик до
различни от война. Тъй като времето за
полетно ниво FL410.
реагиране и разпространяване на информацията
до органите, които вземат решенията е от
II-291
развитието на инцидента и методите на
прилагане на активните средства за ПВО, това
ВП може да бъде ограничено по височина.
FL
4
Отменят се всички УВД разрешения в
ограниченото ВП и се вземат мерки за
∆Н= управление на потоците въздушно движение, а в
FL R=18,5 300 м контролираните летищни райони се прекратяват
3 км
полетите в зоните за изчакване (ЗИ),
отклоняване от маршрути и др.
За ефективното изпълнение на задачите
FL ∆Н= си е необходимо да се осигурят по 1000 м по три
2 600 м полетни нива под и над подозрителното
въздухоплавателно средство. Също така трябва
FL да се отчита факта, че ВП може все още да не е
1
освободено от другите въздухоплавателни
средства. В този случай РП е длъжен да осигури
тяхната безопасност и минималните норми за
сепарация, както е показано на фиг. 2.
Фиг. 1. Минимуми за радарно
сепариране.
Разгледаният пример на фиг. 1 ограничена

FL 4 зона
показва,че граждански ВС летят на полетни
нива FL3 и FL4.То следва военното ВС да лети
на FL1. Това ще осигури безопасността на 600м
гражданското ВС и няма да се получи сигнал за FL 3
забране
опасно сближение от бордната система за на
избягване от сблъскване. 900м зона
R=25NM
1.1. Установяване на гражданско FL 2
военни процедури по УВД при терористична
заплаха 900м
Анализът на събитията от 11.09.2001 г. в
САЩ, наложи подобрения в гражданско- FL 1
военните процедури за УВД при
идентифицирането, воденето и внедряването на Фиг. 2. Зона за изпълнение на задачата.
мерки на ПВО по отношение на
въздухоплавателно средство, което се използва Ако подозрителното въздухоплавателно
като оръжие за извършване на терористична средство лети на полетно ниво FL 2, то за
атака. Това налага предварително да бъдат изтребителната авиация се заделя ВП между
установени подходящи комуникационни връзки полетни нива FL 1 и FL 3, в което то ще
между гражданските и военните органи, както и изпълнява задачата си. Това ВП е обявено като
между органите от предтактическото и забранена зона за другите граждански
тактическото ниво за УВД с органите от въздухоплавателни средства. ВП между FL 3 и
системата за ПВО. Необходимо е каналите да са FL 4 е обявено за ограничено ВП за
със съответното ниво за защита. гражданските въздухоплавателни средства,
Въздушното пространство около когато на FL 4 лети такова.
подозрителното ВС, трябва да бъде освободено ВП, което се осигурява за действие на
от граждански въздухоплавателни средства, активните средства на ЗРВ е с R=10 NM и без
което би улеснило внедряването на мерките на ограничения по височина. То се обявява като
ПВО, като прихващане, разузнаване, забранена за полети зона.
идентификация принудително кацане на С оглед предотвратяване получаването
определено летище. на предварителна сигнализация на системата за
Въвежда се ограничена зона около ВС, алармиране и предотвратяване на конфликтна
предмет на заплаха с размери R= 25 NM, а по ситуация TCAS или АCAS, военният РП
височина до всички полетни нива от информира гражданския РП, че прехващащото
регулираното ВП над и под
въздухоплавателното средство. В зависимост от
II-292
въздухоплавателно средство може да ограничи Ръководните насоки, посочени в
информацията в режим С в ограниченото ВП. програма “Ренегат” се концентрират върху две
Като част от процедурите по фази:
прехващането, подозрителното - Оптимизиране визуализацията на
въздухоплавателно средство ще трябва да въздушната обстановка;
осъществи контакт с прехващащото, в - Внедряване на дейности по УВД,
съответствие с процедурите на Международната свързани с осигуряване на ПВО срещу
организация за гражданска авиация - ICAO. граждански ВС, използвани като оръжие за
Единият от начините за осъществяване на извършване на въздушна терористична атака
контакти е на Международната аварийна (ВС-РЕНЕГАТ).
честота 121,5 MHz или 243 MHz. Ако е Цялата информация относно
възможно, за предпочитане е използването на подозрително ВС трябва да се разпространи до
отделна честота. съседните сектори за УВД и да се уведоми
2. Участие на ВВС в програмите на системата за ПВО (включително извън
НАТО “Air Policing” и “Ренегат”. националните граници), а по командна линия и
На 18.07.2002 г. Северноатлантическият на съответните военни командири на НАТО и
съвет утвърди ръководни насоки за националните органи, за да се провокира
установяване на гражданско-военни процедури своевременна реакция на ПВО.
по управление на въздушното движение в В този процес от основно значение е
светлината на новата обстановка по отношение установяването на подходящи комуникационни
на сигурността, което осигурява политиката на връзки между гражданските и военните органи,
Алианса за подобряване на краткосрочните както и между по-висшите нива на органите за
антитерористични мерки по УВД. УВД и ПВО. В национален мащаб това налага
Събитията от 11.09.2001 г. в САЩ установяване на съответни процедури и
показват, че се разширява диапазона от определяне на фразеология, за да се
опасности за националната сигурност на всяка предотвратят евентуални недоразумения.
държава, свързани с различните прояви на Разпространяването на информация трябва да се
тероризма. осъществява при необходимост и да се използват
Една от основните задачи за канали със съответното ниво на защита.
въоръжените сили на НАТО, посочена в От основно значение е да не се губи
Коалиционната военна стратегия на съюза, е радиолокационната връзка (радарен трак) с
регулирането на кризи, включително и на тези, подозрително ВС. Това означава резервиране на
предизвикани от терористични действия. вторична радиолокационна информация чрез
От изключително значение за активна и пасивна радарна информация.
ефективността на работата на летателния състав Задължително е поддържането на първично
и органите, ръководещи полетите, както и за радиолокационно покритие и обмен на данни
адекватността на действията им в отговор на между гражданските и военните центрове.
нестандартни ситуации е доброто познаване на Препоръчва се установяването на
структурата, принципите и подчинеността на телеконферентни връзки, за да се улесни
системата за УВД, тяхното място в нея и постоянното, изчерпателно и еднакво познаване
евентуалните задачи, които биха им били на въздушната обстановка от всички
поставени в съответните ситуации. заинтересовани участници.
Новост в подготовката на летателния 2.2. Прехват на гражданско ВС.
състав и на ръководителите полети е Внедряване на дейности по УВД в резултат
подготовката им за действие при терористичен на усложнената обстановка и изпълнението
акт на борда на ВС. Програма “Ренегат” съдържа на мерки по ПВО.
ръководните насоки за политиката на НАТО за В интерес на безопасността на полетите
краткосрочни подобрения на гражданско-военните въздушното пространство около
процедури за УВД в новата обстановка по подозрителното ВС трябва да бъде освободено,
отношение на сигурността. което би улеснило и внедряването на мерки на
ПВО като прихващане за разузнаване,
2.1. Идентифициране на идентификация или принудително кацане на
въздухоплавателно средство (ВС), определено летище.
представляващо вероятна терористична Въздухоплавателно средство, навлязло
заплаха. без разрешение във ВП на Р България или е
нарушило правилата за полети, без държавен
регистрационен отличителен знак или не се

II-293
подчини на указанията на органите за от прехващача се изпълнява, както следва:
управление на ВД, се счита за нарушител и се а) Фаза I
принуждава да кацне. Прехващачът приближава прехващаното
При маневриране за изпълнение на ВС в задна полусфера. Прехващачът или
прехват трябва да се отчитат следните водачът на прехващащата група заема позиция
принципи: от лявата страна, малко по-високо и отпред
а) прехват на гражданско ВС се спрямо прехващаното ВС, в границите на
предприема само като последна мярка; видимостта на пилота на прехващаното ВС и
б) ако се предприеме прехват, той се първоначално не по-близо от 300 m.
ограничава до идентифициране на ВС, с Другите ВС от прехващащата група
изключение на случаите, когато е необходимо трябва да поддържат безопасно отдалечение от
ВС да бъде върнато на зададената пътна линия, прехващаното ВС, като заемат позиция в задна
да бъде изведено извън границите на полусфера над прехващаното ВС. След
въздушното пространство на Република достигане на необходимата позиция и скорост и
България или извън забранена, ограничена или при необходимост се преминава към Фаза II.
опасна зона, или да бъде принудено да извърши б) Фаза II
кацане на посочено летище; Прехващачът или водачът на
в) забранени са прехватите на прехващащата група започва бавно сближение с
граждански ВС с тренировъчна и учебна цел; прехващаното ВС до отдалечение не по-малко
г) насочването на прехванатото ВС и от абсолютно необходимото за получаване на
даването на свързана с това информация се нужната информация, като поддържа нивото си.
извършва с помощта на радиотелефония, когато Прехващачът или водачът на прехващащата
може да се установи радиовръзка; група трябва внимателно да избягва
д) прехванато гражданско ВС се причиняването на уплаха на екипажа и
принуждава да кацне, когато определеното за пътниците на прехващаното ВС, като има
целта летище отговаря на изискванията за предвид, че маньоврите, приемани за нормални
безопасно кацане на съответния тип ВС; от прехващача, могат да бъдат възприети като
е) забранена е употребата на оръжие опасни от пътниците и екипажа на гражданското
срещу граждански ВС, изпълняващи полет по ВС.
смисъла на чл. 3, ал. 1 от Закона за Другите ВС от прехващащата група
гражданското въздухоплаване. трябва да поддържат безопасно отдалечение от
С цел осигуряване на необходимата прехващаното ВС. След завършване на
безопасност на въздушната навигация и процедурата по опознаване прехващачът се
ограничаване до минимум случаите на отдалечава от прехващаното ВС съгласно Фаза
изпълнение на прехват за опознаване на III.
гражданските ВС, в района на държавната в) Фаза III
граница трябва да се използва вторичен радар, Прехващачът или водачът на
доколкото това е практически възможно. прехващащата група започва бавно
2.2.1.Маневриране за изпълнение на отдалечаване с плавно пикиране. Другите ВС от
прехват прехващащата група трябва да поддържат
При маневриране за изпълнение на безопасно отдалечение от прехващаното ВС и да
прехват трябва да се отчитат следните фактори: се присъединят към водача.
а) ограниченията в летателните 2.2.3.Маневриране за указване
характеристики на гражданските ВС; направлението на полета
б) необходимостта от поддържане на След изпълнение на маньоврите за
определена дистанция от прехващаното ВС с опознаване, описани във Фаза I и Фаза II, ако се
цел предотвратяване на опасността от счете, че е необходима намеса в навигацията на
сблъскване; прехващаното ВС, прехващачът или водачът на
в) необходимостта от избягване прехващащата група заема позиция от лявата
пресичането на траекторията на полета на страна малко по-високо и отпред спрямо
гражданското ВС или изпълнение на маньовър прехващаното ВС, позволяваща на пилота на
по такъв начин, че турболентната следа да прехващаното ВС да наблюдава подаваните
представлява опасност, особено ако визуални сигнали.
прехващаното ВС се класифицира като леко. Командирът на прехващача е длъжен да
2.2.2.Маневриране за визуално се увери, че командирът на прехващаното ВС е
опознаване разбрал за прехвата и потвърждава подаваните
Визуално опознаване на гражданско ВС сигнали. Ако опитите да се привлече
II-294
вниманието на командира на прехващаното ВС с отношение на дължината на ПИК и масата на
помощта на сигналите, са неуспешни, могат да ВС към този момент.
се използват други методи за подаване на Особено важно е цялата необходима
сигнали за тази цел, включително, като крайна информация за изпълнението на безопасен
мярка, визуалният ефект, получаван при подход и кацане да бъде предадена на
включване на форсажа, като не се създава прехванатото ВС посредством радиотелефония.
опасност за прехващаното ВС. 2.2.5.Действия на екипажа на
Допустимо е поради метеорологични прехванатото ВС
условия или релефа на местността прехващачът Екипажът на ВС, прехванато от друго
или водачът на прехващащата група да заеме ВС, е длъжен незабавно да:
позиция от дясната страна малко по-високо и - изпълнява инструкциите на
отпред спрямо прехващаното ВС. В този случай прехващача, да интерпретира и отговаря на
командирът на прехващача е длъжен да визуалните сигнали;
предприеме всички необходими мерки, така че - уведоми при възможност съответния
прехващачът да бъде наблюдаван постоянно от орган за ОВД;
командира на прехващаното ВС. - направи опит да установи радиовръзка
2.2.4.Насочване на прехванато ВС с прехващащото ВС или със съответния орган за
Насочването на прехванатото ВС и управление на прехвата, като се обади на
даването на свързана с това информация се аварийната честота 121.5 MHz и съобщи
извършват посредством радиотелефония, ако позивната си и вида на полета; в случай че не
може да се установи радиовръзка. може да установи контакт и ако е възможно, да
Насочване на прехванато ВС се повтори това съобщение на аварийната честота
изпълнява така, че насочваното ВС да не 243 MHz;
попадне в условия, при които видимостта може - постави режим А, код 7700, ако ВС е
да бъде под изискваната за полет при ВМУ, и оборудвано с транспондер, освен когато е
така, че изискваните маньоври от прехванатото получило други инструкции от съответния орган
ВС да не увеличават вече съществуващата за ОВД.
заплаха за безопасността на полета, в случаите Когато инструкциите, получени по
когато летателно-техническите характеристики радиото от друг източник, противоречат на тези,
на прехванатото ВС са влошени. подавани от прехващача посредством визуални
В изключителни случаи, когато сигнали, командирът на прехванатото ВС е
прехванатото ВС се принуждава да кацне, се длъжен незабавно да поиска разяснение, като
спазват следните изисквания: същевременно спазва визуалните инструкции,
- определеното за кацане летище да подавани от прехващача.
отговаря на изискванията за безопасно кацане на Когато инструкциите, получени по
съответния тип ВС, особено ако това летище не радиото от друг източник, противоречат на тези,
се използва от граждански ВС; подавани от прехващача посредством
- местността в района на летището да е радиотелефония, командирът на прехванатото
подходяща за полет по кръга, подход и ВС е длъжен незабавно да поиска разяснение,
минаване на втори кръг; като същевременно спазва инструкциите,
- прехванатото ВС да има достатъчен подавани от прехващача по радиото.
запас от гориво за достигане до летището; На процедурно ниво, между военното
- за прехванато гражданско транспортно ведомство и министерството на транспорта и
ВС ПИК на определеното за кацане летище да съобщенията съществува необходимост от
има дължина, еквивалентна на не по-малко от договори за взаимодействие, протоколи за
2500 м дължина на ПИК, разположена на комуникация и планиране за внедряване на
средното морско равнище; временно ограничено въздушно пространство
- за кацане, когато това е възможно, се Анализирайки настоящите технически
определя летище, за което е публикувана възможности за осигуряване на
подробна информация в Сборника за радиолокационна картина за нуждите на ПВО е
аеронавигационна информация и публикация. необходимо да се извърши група от технически
Когато гражданско ВС се принуждава да подобрения с цел улесняване координирането
кацне на непознато летище, е необходимо да му на мерките по УВД и ПВО. Това може да
се предостави достатъчно време за подготовка за включи преразглеждане, в някои случаи, на
кацане, като се има предвид, че само настоящото планиране за спиране на граждански
командирът на гражданското ВС има право да и военни първични радари и необходимостта от
вземе решение за безопасността на кацането по по-добър обмен на радиолокационна

II-295
информация (филтрирана въздушна картина) за планиране и разпределяне на
между системата за командване и контрол на въздушното пространство на Република
НАТО и системата за УВД. България, Обн., ДВ, бр.101 от 1999.
Поставянето на задачи за бойно 3. Инструкция №25/01.11.1999г. за
използване на активните средства за ПВО – структурата и правилата за работа на
дивизионите ЗРВ (GBAD) и ескадрилите Центъра за координиране използването
изтребителна авиация (QRA), от на въздушното пространство на
многонационалния въздушен оперативен център Република България, Обн., ДВ, бр.101 от
(САОС) и оперативния център за въздушен 1999.
суверенитет (ASOC), чрез бригадните 4. Наредба №19/27.01.1999г. за реда за
оперативни центрове (SAMOC) и базовите въвеждане и правилата за работа на
оперативни центрове (WOC) схематично е единната система за гражданско и
показан на фиг. 3. военно управление на въздушното
пространство, Обн., ДВ, бр.13 от 1999.
5. Наредба №11/05.05.1999г. за обслужване
на въздушното движение във
въздушното пространство на Република
България, Обн., ДВ, бр.50 от 1999.
6. Наредба №2/10.03.1999 г. Правила за
провеждане на полети на Министъра на
транспорта и съобщенията – Обн., ДВ,
бр.50 от 1999.
7. Пенев П. Б., Ячев Р. Я., Основи на
теорията и използването на
авиационните навигационни системи и
средства,учебник, ВА “Г. С. Раковски”,
Фиг. 3. Схема на процеса на протичане на 2004.
информационния поток 8. Унгорян С.Г., Маркович Е.Д., Волевач
А.И, Анализ и моделирование систем
Специално внимание трябва да се управления воздушным движением,
отдели на експортирането на УВД картината на Транспорт, М., 1980.
въздушната обстановка до центровете за ПВО. 9. Ячев Р. Я., Управление на въздушното
Трябва да се установят защитени директни пространство по време на кризи и военен
комуникационни връзки . конфликт, Лекция, ВА „Г. С. Раковски”,
Не е изключено да възникне С., 2002.
необходимост от централизирано събиране на 10. Procedures for air navigation services; Doc
УВД информация на ниво НАТО и/или 8168-OPS/611, Vol II, IKAO.
ЕВРОКОНТРОЛ, както и от развръщане на 11. Airspace Management Modernization for
военен персонал при гражданските органи за Central and Eastern Europe, Volume II –
УВД, както и на граждански РП в центрове на Annex K Air Sovereignty Modernization
ПВО. Recommendation for Bulgaria, sept. 1997.
12. Наредба № 23 от 1.11.1999 г. за реда за
използване на въздушното пространство
ЛИТЕРАТУРА при прехода към военен контрол и
ръководство на въздушното
1. Заеков В.С., Чилингиров Д. И., Теория за пространство при повишаващи се нива
управление на въздушното движение, на извънредни условия, издадена от
учебно пособие, ВИ, София, 1992. министъра на транспорта и министъра
2. Закон за гражданското въздухоплаване, на отбраната, обн., ДВ, бр. 99 от
Обн., ДВ, бр. 94 от 01.12.1972г., изм. и 16.11.1999 г.
доп. бр. 30 от 1999г., бр. 16 от 1997г., бр.
85 от 1998г., в сила от 01.01.1999г., бр.
12 от 2000г.,в сила от 15.02.2000г., изм. Данчо Колибаров
доп.ДВ, бр.34 от 2001.Инструкция ТУ- София, филиал Пловдив
№24/01.11.1999г. за работа на Центъра E-mail: danchokol@abv.bg
АКЦЕНТИ ПРИ ОБУЧЕНИЕТО ЗА ОПТИМАЛНО

УПРАВЛЕНИЕ НА ЕЛЕКТРИЧЕСКИ РЕЛСОВ
ТРАНСПОРТ С БЕЗСТЕПЕННО РЕГУЛИРАНЕ НА
ТЕГЛИТЕЛНАТА СИЛА
ИЛКО ТЪРПОВ
Резюме: Различните технически решения за управление на теглителната сила на

електрическите транспортни средства до голяма степен определя поведението и
стила на шофиране от страна на водача. В доклада ще бъдат разгледани някои
акценти при обучението за оптимално управление с безстепенно регулиране на
теглителната сила.
Ключови думи: оптимално управление, тренажор, обучение
HIGHLIGHTS IN THE TRAINING FOR OPTIMAL

CONTROL OF ELECTRICAL RAILWAY SMOOTH
VARIABLE TRACTION
ILKO TARPOV
Abstract: The various technical solutions to control the thrust of electric vehicles, largely
determine the behavior and driving style of the driver. The report will be discussing some
highlights in the training for optimal, smooth controlling of the thrust for constant
acceleration
Key words: optimal control of smooth acceleration, simulator training
1. Увод превозите, чрез тренажорното обучение може да

Важен фактор при реализиране на се постигне овладяване и натренирване на
оптимално управление на електрически умения за оптимално управление на ЕТС.
транспортни средства (ЕТС) се оказва техният Според [1] фактора – психомоторика влияе
водач. Техническите знания, свързани с около 24% от всички психологически фактори,
управлението на тяговото оборудване, съчетани разглеждани като възможности за намаляване на
с опита и отлично познаване характеристиките енергийния разход. Този подход със сигурност
на пътния участък, гарантират реализирането на ще подобри енергийната ефективност на
оптимално управление на транспортните транспортните процеси.
средства. Постигане на конкретни умения у
2. Оптимално управление при безстепенно
водача в тази посока, става чрез добро обучение
регулиране на теглителната сила
и продължителна практика. По-бързото
преодоляването на този дълъг процес може да Фактори оказващи влияние върху
бъде осъществено, чрез използването на разхода на електроенергия са теглителната сила
технологиите за глобално позициониране, на локомотива, скоростта на движение и
добавени към оборудването на транспортното коефициента на полезно действие на
средство и чрез обучение с тренажор. транспортното средство. При диодните
Наред с придобиването на устойчиви транспортни средства органа за въздействие
познания и умения свързани с безопасността на върху теглителната сила е контролера и
II-297
неговото положение, което се определя от средства с безстепенно регулиране на тяговите
водача. Докато при напълно управляемите двигатели.
вентили максималният въртящ момент се
поддържа от управляващата електроника. 2.1. Режими на движение
Водача определя само процента тяга. При съставите с безстепенно регулиране
След като разхода на електроенергия ни на напрежението на тяговите двигатели водача
е основен критерий, определен от времето за избира процента тяга, който може да зададе
придвижване, съгласно книжка разписание, то посредством лоста за „движение –спирана”.
той придобива вида: Поддържането на желаната скорост се
осъщиствява автоматично, чрез устройство
tk наречено темпомат. Тези технически особености
FkV
E 
tн

dt предопределят до голяма степен малкото
влияние на водача при ускоряване на
, (1)
транспортното средство. Движението на такъв
където tн и tк са начално и крайно време състав е показано на Фиг. 1. На нея се
за движение; наблюдават характерните четири основни фази
Fк - теглителна сила на локомотива; на траекторията на движение на релсово
 - среден коефициент на полезно транспортно средство във всяко едно
междугарие. Специфичното тук, е, че
действие на локомотива; ускоряването е максимално бързо и след
V - средна скорост на движение. достигане на максимално допустимата скорост
Според [2] „Прилагането на режим за ясно се очертава втора зона на така нареченото
водене на транспортното средство с подходящи „трионовидно” движение, подсигурено от
скорости и ускорения, и ограничено използване темпомата. Третата зона се характеризира с
на спирачните средства, намаляват разхода за движение по инерция и се ограничава от
енергия.”
Оптимизирането на енергетичният
разход за тягови нужди е целесъобразно да се
направи, чрез аналитично изследване на
различните стратегии за управление на
транспортното средство. За да се постигне този
вид управление на ЕТС е необходимо да се
определят критериите, по които ще се прави
оценка на енергетичния разход. От своя страна
той зависи от две основни групи фактори -
обективни и субективни.
Фиг. 1. Графично изобразяване движението на
Към обективните фактори се отнасят
влак с диаграма V(s)
профилът на пътя, допустимата скорост на
движение, натовареността на ЕТС, пътната стартиране на спирането, което представлява
обстановка, условията на сцепление, нивото на четвъртата зона. Именно в тези последни две
напрежение в контактната мрежа и др. зони водача може да влияе на енергийната
Към субективните фактори се отнася ефективност, чрез определяне на тяхното начало
стилът на управление, неговото съобразяване с и продължителност. Във втарата зона също
техническите възможности на транспортното може да се повлияе,чрез понижаване на
средство и усетът към използването на скоростта, ака съществеват времеви буфери в
натрупаната потенциална енергия на състава по разписанието и не е нарушен графика за
време на движение от страна на водача. движение. Съществуват и други стилове на
Влияние върху постигането на движване, които са производни на описания по
оптимално управление, както бе споменато, горе.
оказват коефициента на полезно действие на
състава, намаляване на основното 2.2. Фактори за ефективно
съпротивление на движение, намаляване на управление
загубите при спиране и пускане както и Основни фактори, оказващи влияние
подобряване на стратегията за управление на върху разхода на енергия сa скоростта на
състава. В настоящия доклад ще се разглежда движение, загубите при потегляне и спиране,
единствено въпроса свързан с правилно както и увеличението на спирачното закъснение.
подбраните режими на движение на превозните
II-298
Увеличението на спирачното закъснение wi-wo-wR – съпротивление на движение
също води до намаляване на разхода за енергия. при наклон, основно съпротивление и
Както се вижда от Фиг. 2, нa която крива 1 ce съпротивление от криви.
отнася за по-голямо спирачно закъснение, a След намиране на размера на
крива 2 - за по-малко, увеличението на рекуперативната енергия можем да определим
закъснението позволява да се нейното влияние върху общия разход чрез
формула 3.
Eоб  Едв  Есн  Е рк , (3)
където Еоб е общ разход на енергия;

Едв- разход за движение;
Есн- разход за собствени нужди;
Ерк- енергия от рекуперация.
Определянето на ефективността на
рекуператевното спиране става чрез
коефициента крк.
Фиг. 2. Увеличаване на спирачното закъснение
намали скоростта на началото на спиране от Vн2

E рк
K pк  , (4)
до Vн1. Поради това се намаляват и загубите за Eоб
спиране и съответно намалява времето, през
което двигателите работят под ток ti2 на ti1. По- В планински участъци Крк може да
ранното изключване на ТД е отбелязано с достигне до 10-20 %, ако използваемостта на
индекса ti1, а по- дългата работа на ТД с индекса върнатата енергия е висока и ако товаропотока
ti2.Диаграмата на фигура 2 касае времетраенето на спускане е по-голям от този на изкачване.
на третата условна зона, движение по инерция. При спиране с рекуперация
Скоростта на започване на спирането оказва ефективността се характеризира с отношението
голямо влияние върху общия баланс на на върнатата енергия към пълната, необходима
консумираната енергия, особено при участъци с за спиране на състава. Тук коефициента може да
малка дължина, поради възможността за достигне до Крк~ 0,49 [3], но благодарение на
рекуперативно спиране при ЕТС с напълно загубите от порядъка на 40÷60 % от цялата
управляемите вентили. изразходвана енергия за рекуперация
икономията на енергия достига до 15÷30 % от
2.3. Рекуперативно спиране общата изразходвана.
Служебното спиране се задейства от
водача на ЕТС посредством лоста "движение- Vн2  Vk2
спиране”. Спирането се осъществява K pк   рк л К и , (5)
преимуществено, чрез електрическата спирачка Vн2
с цел да се върне в мрежата рекуперативен ток и
да се сведе до минимум нивото на шума и където Vн, Vк-начална и крайна скорост
износването на накладките в процеса на на рекуперативно спиране;
спиране. ηрк, ηл – коефициент на полезно действие
Размера на върнатата енергия при на рекуперацията и на локомотива.
спускате по наклон- Ерк се изразява чрез Водача на ЕТС трябва да се стреми
формула 2. частта от спирачната сила от рекуперация през
целия процес на спиране да е по-голяма от
Е рк К И .У рк . рк .  wi  wO  wR  .Tспу общата при използване и на друг вид спирачки.
, (2) Това показва, че рекуперацията е по-
ефективна при спиране отколкото при спускане
където Ки - коефициент на използваемост в надолнище или поддържане на максимално
на рекуперативната енергия; допустима скорост на движение поради по-
Урк - частта от спирачната сила от голямата разлика на скоростите в началото и
рекуперация през целия процес на спиране; края на спирането.
ηрк - среден коефициент на полезно Уменията за правилно определени и
действие на рекуперацията; подбрани режими за управление на ЕТС от
Tспу - товаропоток при спускане неговия водач, придобити след обучение с

II-299
тренажор, ще гарантират повишена енергийна експериментално изследване, чрез съпоставяне
ефективност на транспортните процеси. разхода на енергия на различни водачи, при
Отрицателно влияние върху икономията управление на едни и същи влакове с мотриса №
на енергия оказват още непредвидените 31- 019, показани в Табл. 1.
спирания на състава, нарушаването на графика Получената разлика от 20 % в
на движение, временните намаления на реализирания разход на електрическа енергия
скоростта и др. Тези фактори няма да бъдат доказват необходимостта от прилагане на
разглеждани подробно поради по-малкото им предложената методика и внедряване на
влияние върху електропотреблението и тренажорно обучение. Резултатите от
случайният им характер на възникване. проведеното изследване са потвърдени и в други
Намаляването на разхода на енергия за направления. Те подчертават влиянието на
собствени нужди също е обвързан със стила на субективния фактор върху разхода на енергия и
управление на железопътните състави и зависи, необходимостта от обучение по енергийно
както от техническите решения за управление на ефективно управление.
електрическите устройства за собствени нужди,
така и от начина на тяхното използване от 3. Заключение
локомотивния машинист. Правилната употреба Получените теоретични резултати дават
и целесъобразното използване на системите за основание за въвеждане на тренировъчен стенд
климатизация в мотрисните състави могат да за обучение на водачи на релсови транспортни
доведат до оптимизиране на общият енергиен средства;
Получената икономия на енергия с
Таблица 1. Разход на въвеждането на тренажор ще е в порядък от
електроенергия по направление 5 % ÷ 10 %;
Пловдив-Асеновград Посочените предимства и доказания
икономически ефект, дават основание да
маш. час час енергия
дата № начало край в kW килом. влак № продължат изследванията в тази посока и от
19201- гледна точка на необходимите инвестиции за
17.01 378 04:15 07:30 510 74
19202- внедряване и използване на обучителен
19205- тренажор в железопътния транспорт.
19206
19201-
19202- ЛИТЕРАТУРА
03.01 762 04:20 07:35 520 76
19205-
19206 1. Петров И., И. Лалов, С. Сребрекова, (2015),
19201- Изследване на психофизическите
11.01 737 04:15 08:15 582 76
19202- показатели на локомотивните машинисти
19205- за подобряване на техниката на водене на
19206
19201-
влак и енергийната им ефективност,
19202- Механика транспорт комуникации, том 13,
06.01 746 04:15 08:25 500 77
19205- Брой 3/3.
19206 2. Ненов Н., (2008), Движение на влаковете и
19201- оптимални режими на управление, София
19202-
18.01 319 04:15 08:20 623 77
19205- 2008.
19206 3. Българанов Л. и кол. (2009), Електрически
транспорт, София.
разход. Той може да се окаже значителен пpи
пътническите влакове, които са снабдени с
климатични системи, ако остават включени при Контакти:
служебните придвижватия и продължителни ас. д-р инж. Илко Търпов
престои без пътници. При тези състави Технически университет – София,
правилното регулиране на температурата, филиал Пловдив
използването на автоматично управление на Адрес: Цанко Дюстабанов 25
вратите, добрата топлоизолация и др., могат Телефон: 0889 075605
чувствително да намалят разхода на E-mail: stsb_plovdiv@abv.bg
електроенергия.
За да се провери верността на
получените теоретични резултати и
дългогодишни наблюдения е направено
ПОВИШАВАНЕ ЕФЕКТИВНОСТТА ОТ
РЕКУПЕРАТИВНОТО СПИРАНЕ ЧРЕЗ
ПРОМЕНИ В ГРАФИКА ЗА ДВИЖЕНИЕ НА
ВЛАКОВЕТЕ
ИЛКО ТЪРПОВ, СИЛВИЯ САЛАПАТЕВА
Резюме: В доклада е разгледано рекуперативното спиране и възможностите за
подобряване на коефициента за ефективност на този вид електродинамично
спиране чрез промени в графика за движение на влаковете.
Ключови думи: рекуперация, график за движение на влаковете
INCREASING THE EFFICIENCY OF THE

RECUPERATIVE BRAKING BY ALTERING THE
TRAIN SCHEDULES
ILKO TARPOV, SILVIYA SALAPATEVA
Abstarcts: In the report the recuperative stopping and the possibilities of increasing the
efficiency of this kind of electrodynamic braking, by changing the train schedules are
analyzed.
Key words: recuperation, train schedules
1. Въведение достигне до 30% от консумираната такава, в

От края на 2007 година в България се зависимост от стила на управление на
експлоатира тягов подвижен състав, които транспортното средство. Пълното консумиране
притежават електродинамична спирачка с на енергията от рекуперация може да се осигури
рекуперативен ефект. ако в същата фидерна зона има състав, който в
Рекуперативното спиране, като вид този момент се намира в процес на ускоряване.
електрическо спиране се осъществява В доклада ще бъде обърнато внимание
благодарение на обратимостта на на възможността за повишаване на коефициента
електрическите машини. Натрупаната кинетична на рекуперация, чрез промени в графика за
енергия в състава през периода на ускоряване и движение на влаковете (ГДВ).
потенциалната такава при спускане се
2. Изследване на ефективността от
преобразува в електрическа по време на
рекуперация
спиране. Тази енергия върната в контактната
мрежа или съхранена в акумулаторни батерии, За изследване на ефективността от
наричаме рекуперативна. рекуперативното спиране са проследени
Значителен ефект от този вид спиране електрическите мотрисни влакове (ЕМВ)
може да се наблюдава при продължителни “Дезиро” на фирмата “SIMENS”, собственост на
стръмни спускания или в участъци с голям брой БДЖ. На фиг. 1 са показани графичните
спирки. Размерът на върнатата енергия в изменения на трите вида мощности (активна – P,
мрежата, според някои автори [2], може да W, реактивна – Q, VAr и привидна – S, VA) на
II-301
ЕМВ получени от експериментално измерване. Ерк - енергия от рекуперация, кWh.
Вижда се, че в няколко участъка активната
мощност P сменя знака си, което показва, че За да се повиши ефективността от
състава реализира успешно рекуперация и рекуперативното спиране според формула (2),
връща енергия в мрежата. трябва да се повиши енергията от рекуперация
Ерк, тъй като общата консумирана енергия
зависи от свършената работа.
В планински участъци коефициента Крк
Pemv, W Qemv, Var Semv, VA може да достигне до 10-20 %, ако
1200000
използваемостта на върнатата енергия е висока
1000000
800000
и товаропотока на спускане е по-голям от този
600000
на изкачване.
400000 Това се установява от израза (3):
200000
0
Vн2  Vк2
К рк   рк л Ки , (3)
17:22
17:24
17:26
17:28
17:30
17:32
17:34
17:36
17:38
17:40
17:45
17:47
17:49
17:51
17:53
17:55
17:57
18:00
-200000
времепътуване, min Vн2

Фиг. 1. Изменение на активната P, където Крк е коефициент на ефективност на
реактивната Q и привидната S мощности на рекуперативното спиране;
ЕМВ “Дезиро” Vн, Vк, - съответно начална и крайна
скорост на рекуперативно спиране, km/h;
Големината на върнатата енергия при
ηрк, ηл - коефициенти на полезно
спускане на състава по наклон Ерк се определя
действие съответно на рекуперацията и на
чрез формула (1).
локомотива;
Ки - коефициент на използваемост на
Eрк  КиУ рк рк( wi  wo  wR )Tспу , (1)
рекуперативната енергия.
където Ки е коефициент на използваемост на Локомотивният машинист трябва да се
рекуперативната енергия; стреми частта от спирачната сила от
Урк - частта на рекуперацията за целия рекуперация Урк през целия процес на спиране
процес на спиране; да е по-голяма от общата при използване и на
ηрк - среден коефициент на полезно друг вид спирачки:
действие на рекуперацията;
Tспу - товаропоток при спускане; T2 wi  ( wo  wR )
К рк   рк л К иУ рк , (4)
wi, wo, wR - съответно съпротивление на T1 wi  wo  wR
движение при наклон, основно съпротивление и
съпротивление от криви, kg/t. където Т1 и Т2 са годишните товаропотоци в
От формула 1 се вижда, че може да се двете посоки;
влияе на размера на върнатата енергия по два wi - относително съпротивление на
начина. Първият е чрез увеличаване движение от наклона, kg/t;
коефициента на използваемост Ки на wR - съпротивление от крива, kg/t;
рекуперативната енергия, като се осигури wo - основно съпротивление на
консуматор, а втория чрез използване на движение, kg/t;
рекуперацията Урк през целия процес на спиране Урк - частта на рекуперацията от общия
без да се прилага друг вид спирачка. Методите спирачен процес.
за повишаване стойностите на останалите При Крк = 0,49 може да се достигне
параметри от формула (1) са обект на икономия на енергия от порядъка на
разглеждане в други разработки. 15÷30 % [2], необходима за движение по
Определянето на ефективността на участъка. Създаването на подходящи условия,
рекуператевното спиране се извършва чрез при които произведената енергия от
коефициента Крк: рекуперация може да се консумира от друг
състав в същия момент, води до увеличаване на
Е рк коефициента на използваемост на
К рк  , (2)
Еоб рекуперативната енергия Ки, а от там и до
реализиране на енергийно ефективен
където Еоб е общия разход на енергия, кWh; транспортен процес.

II-302
Това може да се постигне, чрез прецизно коефициента на използваемост Ки на
съставяне и изпълнение на ГДВ за всеки отделен рекуперативната енергия.
участък. Изисква се още при съставяне на Рекуперацията е по-ефективна при
графика да се съобрази големината на спиране отколкото при спускане в надолнище
железопътното трасе, захранвано от една или поддържане на максимално допустима
подстанция и намиращите се в същата зона скорост на движение. На фиг. 3 е представено
електрически подвижен състав (ЕПС), съгласно изследване, което показва нарастването на
предварително подготвения график. Движението върнатата енергия в контактната мрежа. При
по наклон на един състав трябва да бъде рекуперативно спиране от 100 до 0 km/h
съчетано с изкачването на друг състав по нарастването достига до 4,8 kWh за 42 s. [3].
същото време, с цел да се използва
произведената от рекуперация електрическа
енергия. Също така е необходимо, спирането на
един ЕПС да съвпада с процеса на ускорение на
друг състав. Оптимизирането на ГДВ с цел
повишаване коефициента на ефективност на
рекуперативното спиране Крк е представен на
фиг.2.
Фиг. 3. Нарастване на върнатата в

контактната мрежа енергия при спиране
За нуждите на анализа в държавния

железопътен превозвач се проведе измерване на
доставената енергия от ЕМВ “Дезиро” за период
от една година и върнатата такава от състава.
Данните са отчетени от електромери LEM
монтирани във всяко едно ЕМВ, съгласно
справка [1]. Общо мотрисите са 25 броя, от
които 15 са серия 30-00 (три вагонни) и 10 -
серия 31-00 (четири вагонни). Резултатите от
проведеното аналитично изследване са
представени в таблица 1.
Таблица 1. Анализ на консумираната и
Фиг. 2. Оптимизиране на ГДВ с цел повишаване върната енергия от ЕМВ “Дезиро”
коефициента на ефективност на
рекуперативното спиране Крк ЕМВ Еоб, kWh Ерк, kWh L, km Крк М, t
30 9827510 1922921 1688741 0,1956 133
От лявата страна на фиг. 2 е представена 31 9540453 2044247 1407681 0,2142 155
част от ГДВ между три съседни гари А, В и С
свързани с двойна железопътна линия. От След сравняване на двете серии ЕМВ се
дясната страна на фигурата е изобразен профила констатира, че при серия 31-00 консумацията на
на участъка, като от гара А към гара В е в електрическа енергия за километър е около 16%
нагорнище, а от гара В към гара С в надолнище. по-голяма от тази на серия 30-00. Също така
Чрез отместване на влакове с номера 10114 и рекуперираната електроенергия Ерк за изминат
10116, изобразени с червени линии, стремежът е километър е с 21,58% по-голяма при серия 31-00
влак №10116 да започне изкачване от гара А [2]. Тези разлики се дължат на по-голямата маса
към гара В в момент, в който влакове №10111 и М и завишените разходи за собствени нужди от
№10114 се спускат и произвеждат регенерирана допълнителния вагон.
електрическа енергия. С тази корекция в ГДВ
влак №10116 ще оползотвори максимално 3. Заключение
енергията от рекуперация и така ще се повиши От направените експериментални
измервания и анализ на получените резултати се
II-303
вижда, че ефективността на реализираното 2. Ненов Н., (2008) Движение на влаковете и
рекуперативно спиране от ЕМВ “Дезиро” е оптимални режими на управление, София.
значителна. Рекуперираната електическа 3. Търпов И., (2015) Оптимизация на
енергия Ерк е приблизително 20 % от общата експлоатационните и енергийни показатели
консумирана енергия от състава. на електрифицирания транспорт,
При създаване на условия за максимално Дисертационен труд за продобиване на
използване на енергията от рекуперация образователна и научна степен „доктор“,
енергийната ефективност може да се подобри с ВТУ „Т. Каблешков”, София.
30 %. Такива стойности на ефективност могат да
се наблюдават в „Метрополитен”- София поради
свързването в паралел на захранващите Контакти:
подстанции. ас. д-р инж. Илко Търпов
Заинтересована страна от този вид Технически университет – София, филиал
оптимизация на ГДВ са превозвачите, които Пловдив
притежават ЕПС с рекуперативна спирачка и Адрес: Цанко Дюстабанов 25
значително влаково движение в профилни Телефон: 0889 075605
участъци. E-mail: stsb_plovdiv@abv.bg
ЛИТЕРАТУРА доц. д-р инж. Силвия Салапатева

Технически университет – София, филиал
1. Генова С., М. Красимирова, (2011) Месечни Пловдив
справки за отчетената енергия на Адрес: Цанко Дюстабанов 25
“Дезиро”, ППП София. Телефон: 0893 691913
E-mail: sisisal@tu-plovdiv.bg

УДЪЛЖАВАНЕ НА РЕСУРСА НА ВОЕННА

АВИАЦИОННА ТЕХНИКА В УСЛОВИЯТА НА
ИКОНОМИЧЕСКА КРИЗА
ДОБРИН СЕЙЗИНСКИ, СТОЯН АВРАМОВ
Резюме: Разглежда се въпросът за увеличаване на ресурса на военната авиационна

техника в Българската армия в края на миналия и в началото на този век като
основен начин за обосновано използване на заложения при проектирането и
производството и запас от дълготрайност. Посочват се факторите, обуславящи
удължаването на този ресурс. Анализира се удължаването на ресурса по време на
авиационни изделия и преди всичко техническите аспекти на този въпрос.
Ключови думи: авиационна техника, авиационни изделия, ресурс, дълготрайност.
MILITARY AVIATION EQUIPMENT LIFESPAN

PROLONGATION IN ECONOMICAL CRISIS
SITUATION
DOBRIN SEYZINSKI, STOYAN AVRAMOV
Abstract: Analysis of the ways of increasing the Bulgarian militaryaviation equipment

lifespan in the end of the last and at thebeginning of this century is shown as the main way
of justifiable useof the full design and production limits. Indicated are the
factorsunderlying the prolongation of this lifespan. It is analyzed theprolongation of the
time lifespan mainly in the technical aspects ofthis issue.
Key words: Aviation equipment, aviation products, time resource, durability
1. Основни понятия този по време – в календарна продължителност

Ресурсът е важна технико-икономическа от време.
характеристика за всички обекти, които имат Съществуват следните видове ресурс:
висока производствено-експлоатационна стой-  ресурс до първи ремонт, който представлява
ност и от надеждната работа на които зависи наработката (календарното време) от началото
живота и здравето на хора. Към този вид обекти на експлоатацията на обекта до постъпването
спада и авиационната техника (АТ). му за първи ремонт;
Тъй като съществуват различни  междуремонтен ресурс, който е наработката
тълкования на понятието ресурс, в настоящата (календарното време) между два последова-
работа под ресурс на авиационната техника ще телни ремонта на обекта;
разбираме сумарната наработка на АТ от нача-  назначен ресурс, който представлява сумар-
лото на експлоатацията до прехода и в пределно ната наработка или календарното време, при
състояние. Той се използва за количествена достигането на които експлоатацията на обекта
оценка на дълготрайността на АТ. трябва да се прекрати.
Най-общо ресурсът се дели на две основ- В настоящия доклад се прави опит да се
ни групи: ресурс по наработка; ресурс по време, анализира изминатия път при решаване на
който се нарича още срок на служба. проблема с удължаване със собсвени сили на
Ресурсът по наработка се измерва се в ресурса на военната АТ в Българската армия
пролетени часове, брой кацания, брой цикли, а (БА) през последните 20 години, за които са
II-305
характерни условията на икиономическа криза в правата на производителя и свързаните с това
страната. Поради ограничения си размер той негови задължения да предоставя на потреби-
разглежда само някои основни аспекти от този теля всякаква информация за експлотираните от
въпрос и няма претенциите да бъде обстоен и
него ЛА, включително и тази за удължаване на
изчерпателен.
ресурсите; за спазване на изискванията на
2. Проблеми на експлоатацията на стандартите; за органите, осигуряващи този вид
авиационна техника в условия на дейност и др.
икономическа криза в страната Организационните фактори включват:
В исторически план разглежданият  провеждане на техническа, общоинженерна и
период се характеризира с непрекъснато намаля- специална подготовка на инженерно-
ване на финансовото осигуряване на екслоата- техническия състав;
цията на АТ във всичките и съставни елементи.  спазване или промяна на стратегията и
Най-значително това намаление е в областта на методите на техническо обслужване и ремонт
нейната летателна и техническа експлоатация. и ритмичността на тяхното провеждане;
Затруднена е доставка на резервни части  своевременност в осигуряване на запасни
и окомплектовки. Възникват проблеми с лиценз- части при поява на неизправности и при
ното производство на рективно гориво РТ в изпълнение на текущи ремонти;
тогавашния „Нефтохим” – Бургас.  създаване на идеология на цялостния процес
Пролетените часове непрекъснато на удължаване на ресурса на АТ – от
намаляват, достигайки минимума за запазване методиката до завършващия етап;
на летателната годност на пилотите.  използване на методи и средства за
От друга страна, структурната реформа в механизация и автоматизация на процесите на
армията се изрази в тотално ликвидиране на подготовка на ЛА за полет, за търсене на
авиационните части и няколкократно намаля- неизправностите и за тяхното отстраняване и,
ване броя на летателните апарати (ЛА). особено, използване на автоматизирани
В резултат на всичко това ЛА престояват средства за контрол на техническото състояние
на земята, ресурсът им по време се изразходва, на всички функционални системи и др.;
без да се използва този по наработка.
 информационно осигуряване; наличие на орган
При тези условия основният път за
за контрол и узаконяване на разглежданата
ресурсно осигуряване на АТ, експлоатирана по
дейност.
планово-предупредителната система, е техни-
Информационното осигуряване на
чески обосновано и ефективно използване на
удължаването на ресурса на АТ е един от основ-
заложената при проектирането и производството
ните проблеми за успешното решаване на тази
и дълготрайност, както и увеличаване на тази
задача. При изчислителния (разчетния) ресурс
дълготрайност чрез удължаване на ресурса на
на конструкцията или на ЛА като цяло, напри-
АТ.
мер, се въвеждат коефициенти на запас от дъл-
3. Фактори, от които зависи готрайност, които отчитат влиянието на различ-
удължаването на ресурса на ните неотчетени фактори и степен на достовер-
авиационна техника ност на изходните данни. Тъй като тези коефи-
циенти са приведени в конструкторската доку-
Удължаването на ресурса на АТ зависи
ментация, то за удължаване на ресурса е необхо-
от множество фактори, които могат да се
дима информация, свързана с проектирането и
подразделят на следните: нормативно-правни;
производството на ЛА. Такава своевременна и
организационни; експлоатационни; научно-
актуална информация може да бъде получена от
технически.
заводите-производители, създаващи експлоати-
Нормативно-правните фактори са най-
раната у нас АТ. Подобна информация може да
слабо проучени и изяснени.
бъде получена и чрез обмен със съответните
През разглеждания период законовата
органи във ВВС на страните, експлоатиращи
основа за решаване на този въпрос липсва или еднотипни на нашите ЛА. Трябва да се
не е достатъчно ясна. До средата на периода отбележи, че в нача-лото на разглеждания
дейността на Главния инженер на авиацията е период органите на АТВ във ВВС имат известни
юридически защитена с промяната на НИАС-80 успехи, но в послед-ствие тази дейност няма
и на други документи, позволяващи удължава- никаква реализация.
нето на ресурса на АТ. През повечето време Експлоатационните фактори включват:
 режимите на полета (скорост, височина,
остават нерешени следните въпроси: по защита
използвани маневри, полетна маса и др.);
II-306
 метеорологичните и климатични условия на игнорираше тотално, като ресурсната политика
полета, в това число турболентността на се свеждаше до безусловно изпълнението на
атмосферата, градиентите на температурата по изискванията на експлоатационно-техническата
височина, влажност и др.); документация, на НИАС-80 и на изпращаните от
 индивидуалните особености на членовете на заводите-производители бюлетини.
екипажите и тяхната професионална Липсата на целенасочена научноизследо-
подготовка; вателска работа е причина и за ограничения
 квалификацията на инженерно-техническия брой специалисти, занимаващи се с разглеж-
състав, определяна, в частност, от знанието на даната материя. Тези специалисти са преди
конструкцията на ЛА, от пълнотата, с която се всичко офицери, успешно защитили докторски
откриват неизправностите, своевременното и дисертации в тогавашната ВВИА „Н. Е. Жуков-
ефективно откриване, локализиране и ски” – Москва. Това е изключително актуален
отстраняване на пукнатини, корозии и др.; проблем, тъй като скромният ни кадрови потен-
 качеството и пълнотата на изпълняваните циал по това време не стига дори за следене и
профилактични мероприятия, а също така и адаптация на чуждите научни постижения в тази
качеството на използване на техническите област у нас. От друга страна, в тогавашния
средства за контрол състоянието на ЛА. момент голяма част от разглежданите въпроси
Научно-техническите фактори включват: се решават на експертно ниво, за което също са
 конструктивни, производствени и необходими компетентни лица.
технологични особености; Основен изпълнител и координатор на
 наличие на научноизследователска база за работите и дейностите по удължаване на
провеждане на уникални за целта изследвания; ресурса на АТ е отделението за анализ и контрол
 достижгения на научноизследователска на АТ при Авиационноизследователска база –
дейност в разглежданата област – у нас и в ВВС. За решаването на тази задача са
световен мащаб. привлечени специалисти и е използвана
Например, конструктивните, производ- материално-техническата база на повечето от
ствените и технологичните фактори при изброените по-горе научни звена.
удължаване ресурса на планера се обуславят от Към нерешените чисто наши, български
концетрацията на напрежения в елементите на проблеми следва да добавим и тези, свързани с
конструкцията, от остатъчните напрежения, нерешените или недостатъчно изследвани
задачи на теорията на надеждността на АТ в
възникващи поради несъвършенства в
технологиите и за сметка на пластичните международен план. Сред тях можем да
деформации при монтаж на сглобените единици посочим следните: разработване на методи за
или при ремонта. Те произлизат и от прогнозиране на откази; оценка на надеждността
зависимостта на тези напрежения от свойствата на изделията, изработвани в неголеми серии;
на материалите и на тяхното изменение в разработване на методи за ускорени изпитвания.
процеса на експлоатация. Специфични са тези Независимо от множеството публикации
фактори за електрониката и елементната база. у нас и в чужбина за разработване на методи за
В началото на разглеждания период у прогнозиране на отказите, не можа да се
нас има сравнително добре развита научноизсле- представи апробиран теоретичен метод, който
дователска база, която е предназначена за да се приложи при удължаване на ресурса на
която и да е комплексна група или подгрупа
изследване на надеждността на различни
технически системи и на техните елементи. Това авиационни изделия. В световен мащаб
е материално-техническата база на различни разработването на математични модели за
научни звена в институтите на БАН (Институт прогнозиране на отказите се намира на
по металознание, Институт за космически първоначално ниво и обхваща далеч не всички
изследвания), във висшите учебни заведения реално срещащи се ситуации. Например, има
(тогавашното ВНВВУ „Г. Бенковски”, модел за развитие на корозията, но няма такъв за
Технически университет – София, включително появата и в закрити и трудни за осъществяване
и филиала му в Пловдив), в тогавашния Военен на контрол отсеци от конструкцията на ЛА.
Разработването на такива методи е
научно-технически институт – МО, в бившите
ведомствени институти и бази („Химатех” ООД изключително трудно и изисква следното:
– София).  дълбоко изучаване на физичните явления,
Като проблем при удължаване на водещи до отказите;
ресурса на АТ се оформи слабата научноизсле-  изключително прецизно изследване на причи-
дователска работа у нас в дадената област. В ните за всяка отделна аварийна ситуация;
миналото, преди 10.11.1989 г, тази дейност се

II-307
 монтиране в конструкцията на ЛА на датчици неговата конструкция; служи за материално-
за контрол на напрегнато-деформационното техническа база на провеждането на уникални за
състояние на най-отговорните и елементи; целта изследвания; явява се основен потребител
 натрупване на необходимия материал за на създаваните със значителни финансови
разработване на правилата за поведение на АТ средства авиационни специалисти, пряко
в процеса на нейната експлоатация. свързани с разглежданата задача; захранва се с
За продукцията, произвеждана в големи множество научноизследователски разработки,
серии и още повече за тази от масовото част от които неминуемо засягат тези въпроси.
производство, са разработени добре действащи Практиката показва, че липсата на всяка
статистически методи на надеждността. една от изброените предпоставки е проблем за
Статистическите методи са непригодни, обаче, решаване на задачата за удължаване на ресурса
когата няма необходимия брой наблюдения. на АТ.
Например, отказите по самолета и двигателя са
малко на брой, но за сметка на това решаващи 4. Алгоритъм на удължаването на
при определяне нивото на безопасност на полета ресурса
на ЛА. В такива случаи се спазват следните Удължаването на ресурса през
препоръки: да се използват знанията за физиката разглеждания период се извърши по алгоритъм,
и механиката на процесите, в които работи който се състои от следните елементи:
авиационно изделие; да се въведе статистическо  разработване на методически указания за
изпитване на елементите от изделията, цялостното организиране на процеса;
изработвани в масови серии; да се използват  провеждане на организационни мероприятия;
данни за стари аналогични изделия.  провеждане на инженерно-технически меро-
Методите за ускорени изпитвания, които приятия (инженерен анализ на експлоатаци-
се използват доста отдавна, са особено важни за онната надеждност на конкретния тип ЛА,
решаване на задачата за удължаване на ресурса. моделиране на някои процеси, изслед-вания
В тях е заложен принципът, че чрез прекомерно на отделни елементи и системи на ЛА и др.);
влошаване на условията на работа на изделията  оформяне на регламентиращи документи
се ускорява изразходване на ресурса им. При (технически ръководства или програми за
такава по-напрегната работа тези изделия удължаване на ресурса и указания на Главния
излизат по-бързо от строя и се получава инженер на авиацията или на други
необходимата информация за по-кратко време. ръководни лица, изпълняващи неговите
Това, обаче, налага от своя страна функции).
предварително изучаване на закономерностите Удължаването на назначения ресурс по
на ускоряване на процеса, в зависимост от наработка на ЛА като цяло е уникална и трудно
влошаването на режима. Добре известно е, че до решима за страната ни задача, поради което през
определени граници на изделието могат да последните 20 години не е решавана само с
действат едни закономерности, а при по-високи наши сили.
– други. Срещат се публикации, в които се През разглеждания период удължаването
привеждат изящни резултати и прости формули, на ресурса по време е реализирано за авиацион-
по които трябва да се направят изчисленията на ни изделия (системи, агрегати, елементи), за
основата на ускорените изпитвания. Но които има създадена материално-техническа
организираните по специален начин проверки база и съответно научно осигуряване [1].
показват, че резултатите не се потвърждават. С над 120 указания и други ръководни
Разгледаните досега проблеми биха се документи е удължен ресурса на летателния
решили успешно у нас, ако има подкрепата на апарат като цяло, на негови агрегати, блокове и
авиационния производител. Авиационното функционални системи на почти всички типове
производство (самолетостроенето) осигурява ЛА (Миг-23, Миг-21, Миг-29, Су-22, Су-25К,
удължаването на ресурса на АТ като създава Ми-8, Ми-14, Ми-17, Л-29, Л-39, Л-410, Ан-24 и
следните предпоставки за това: притежава др.), намиращи се на въоръжение в БА.
техническа документация (конструкторска,
технологична и експлоатационна), чрез която в 5. Удължаване на ресурса по време на
производството на АТ се залага и съответния авиационни изделия
назначен технически ресурс. Освен това Удължаване на ресурса по време (срока
разполага с якостните изчисления на на служба) може да се приложи при авиационни
произвеждания ЛА, което е изключително изделия (АИ), чийто ресурс по наработка не
важно при каквото и да е изследване на може да се изразходва в назначения от
напрегнато-деформационното състояние на производителя календарен срок.
II-308
Удължаването на срока на служба на АИ корозиране. При извършване на този анализ
се извършва по методики, използващи различни особено внимание се отделя на създаването на
методи, но включващи в най-общ план следните адекватна представа за различните въздействия
работи и дейности: изследователски работи; и следващите от тях последствия върху
вземане на решение за удължаване на ресурса; безотказността и дълготрайността на АИ.
работи и мероприятия, извършвани преди и след В процеса на своята експлоатация АИ са
удължаването на ресурса; разработване и подложени на въздействието на комплекси от
приемане на принципи на експлоатация на фактори. Най-характерен е следният:
изделията с увеличен ресурс. температура на въздуха; относителна влажност;
Изследователските работи зависят от скорост на вятъра; слънчева радиация; химическо
вида на изследваното изделие т.е. от неговото замърсяване на атмосферата.
предназначение, конструкция, използваните Резултатите от пресмятането на
материали, технологични решения и т.н. При коефициентите на корелация за този комплекс
удължаване на срока на служба основно показват, че няма пряка зависимост между
значение има изучаването на процесите на въздействията на различните фактори в
стареене в пластмасите, гумените изделия, различните климатични зони. Обаче, съвмест-
херметиците, електронните елементи и др. От ното въздействие на няколко от тези фактори
съществено значение е изучаването на може да усили тяхното влияние върху изделие-
корозионните процеси в конструкционните то. Това се проявява изключително силно при
елементи на металните конструкции и тяхното действие на висока влажност заедно с
влияние върху срока на служба на изделието. химически фактори (съдържание на морски
През разглеждания период изследова- соли, на SO2 , H2S, NH3, NO2). Влагата, попаднала
телската работа най-общо се свежда до след- на повърхността или проникнала във вътреш-
ното: статистически и технически анализ на ността на изделията и материалите може да
неизправностите на АИ в процеса на неговата предизвика: снижение на повърхностното или
експлоатация; анализ на условията на експло- обемно съпротивление на изолация, снижение
атация и съхранение; анализ на техническото на електрическата якост на материалите и др.;
състояние; икономически анализ. изменение на физическите свойства на материа-
Статистическият и техническият анализ лите, изменение на размерите и изкорубване на
на неизправностите на изследваното АИ в отделни неметални детайли и др.; ускорено
процеса на експлоатацията му се извършват с стареене, окисляване и корозиране на материа-
цел отделяне на отказите, свързани с лите, повишаване износването на детайлите,
понижаване нивото на безопосност на полетите покритията, изолациите, смазочните материали
и определяне на елементите от изделието, и т.н.; увеличение на нивото на относителните
оказващи съществено влияние върху това електрически загуби в материалите и изделията.
понижаване. Трябва да се отбележи, обаче, че Трябва да се отчете, че анализът на
чрез тези анализи се отделят само тези откази и условията на експлоатация и съхранение на
свързаните с тях елементи, които се обуславят военната АТ у нас показва, че тази техника е
от процесите на стареене, корозия и др. т. е. от експлоатирана и съхранявана при строго спазва-
процесите, определящи назначаването на срока не на изискванията на съответната за всеки тип
на служба. Анализът се извършва на базата на експлоатационна документация. В резултат на
конкретни данни за всяко отделно изделие. При това тя е възможно най-добре защитена от
липса на такова се използва експлоатационната вредното въздействие на отделните външни
информацията за откази за дадения тип изделие, фактори, както и на тяхната комбинация, което
събирана от авиобазите. от своя страна я предпазва от възникването на
Резултатите от анализа са толкова по- изброените по-горе състояния.
правдоподобни, колкото по-пълна е информа- Анализът на техническото състояние на
цията и колкото повече изделия от дадения тип АИ цели получаване на данни за определени
подлежат на наблюдение. За повишаване техни параметри, на базата на които да се
достоверността на тези резултати към удължи ресурсът им. Този анализ се извършва
статистическите данни от експлоатацията се индвидуално за всеки тип изделие и включва
добавят и тези от ремонта на изделията. различни по вид и по обем лаборатории,
Анализът на условията на експлоатация имитационни и други изпитвания. Видът и
и съхранение се провежда от гледна точка на обемът на изпитванията в голяма степен зависят
неблагоприятните въздействия на атмосферните от резултатите на статистическия и техническия
условия върху неговата конструкция и елементи анализи и на анализа на условията на експлоата-
и свързаните с това процеси на стареене и ция и съхранение. В повечето случаи това са

II-309
сложни и специфични изследвания, които специалисти от нашите научни, научнопроиз-
изискват използването на специална апаратура и водствени и учебни звена в сферата на каучука.
методики. Коректността на тези изследвания в Работите и мероприятията, извършва-
най-голяма степен определя правилността на ни преди и след удължаването на ресурса са
решението за удължаване на ресурса. осъществявани от съответните експлоатационни
Икономическият анализ позволява да се звена в системата на ВВС без или с помощта на
определи икономически критерий, въз основа на авиационите военно-ремонти заводи. Обемът и
който да се вземе решение за реализация на периодичността на тези дейности са определяни
работите, свързани с удължаването на ресурса както от решението, забележките и препоръките
на определен тип АИ. на експертната комисия, така и на базата на
Най-общо, икономичческият критерий за натрупания опит в експлоатацията на типовете
ефективността от удължаването на срока на АИ.
служба на АИ е По методика [1,2], отчитаща горното, са
Е=С2 – С1 , удължени ресурсите на шлангове от горивната
където С1 – стойност на разходите за система на самолет МиГ-23 БН, на мекия
удължаване на ресурса на изделието, както и за горивен резервоар на самолет L-39 ZA, на мекия
неговата експлоатация след това; С2 – стойност горивен резервоар на самолет МиГ-21, на
на разходите за ремонт на изделието с изтекъл противопретоварващите и на височинните
ресурс или на доставката на ново изделие, което костюми на пилотите и др. Правилността на
да го замени, както и за неговата експлоатация. взетите решения се потвърждава от практиката.
Стойността на разходите за удължаване
на ресурса на изделието се състои от ЛИТЕРАТУРА
стойностите на: елементите, подлежащи на
подмяна; работите, които трябва да се изпълнят 1. Сейзински, (2012) Д. Анализ на удължаването на
на изделието; изследователските работи, ресурса по време на авиационни изделия през
свързани с удължаването на ресурса; последните две десетилетия. Jubilee International
експлоатацията на изделието с удължен ресурс. Congress Sciens, Education, Tehnologies „40 Years
Понятно е, че при Е ≥ Еmin ≥ 0, Bulgaria – Space Country”, стр. 342.
удължаването на ресурса е икономически 2. Сейзински, Д.Т. (1998). Удължаване на ресурса
целесъобразно. на гумени изделия от авиационната техника. Сб.
Вземането на решение за удължаване на от докл. „Ресурсна политика, методи и
ресурса на АИ е най-отговорния момент в целия програми за увеличаване на техническия ресурс
комплекс от работи и дейности. За всеки тип на авиационната техника и въоръжение”,
изделие то става индивидуално. За някои АИ стр.53.
има апробирани методики, които позволяват да
се оцени тяхната дълготрайност и да се
Добрин Сейзински
прогнозира ресурса им. За повечето, обаче,
ТУ – София, филиал Пловдив
решаването на този въпрос може да се
Адрес: гр. Пловдив,
осъществи само по експертен път. В този случай
ул. Цанко Дястабанов – 25
решението се взема от експертна комисия на
Телефон: 0895 587 410
базата на резултатите от изследователските
E-mail: dobsey@abv.bg
работи и на определените от нея критерии за
удължаване на ресурса. Правилността на взетото
Стоян Аврамов
решение се определя в голяма степен от
ТУ – София, филиал Пловдив
компетентността на експертите от комисията.
Адрес: гр. Пловдив,
Поради това, в качеството на такива се
ул. Цанко Дястабанов – 25
назначават най-добрите специалисти в дадените
Телефон: +359 32 632404
области. Например, в комисията за удължаване
E-mail: stav@tu-plovdiv.bg
ресурса на гумените изделия в качеството на
експерти са привлечени най-добрите учени и
МЕТОДОЛОГИЧЕСКИ АСПЕКТИ НА
БЕЗОПАСНОТО И ЕФЕКТИВНО ГАСЕНЕ НА
ПОЖАРИ С ВЕРТОЛЕТ С ОКАЧВАЕМА
СИСТЕМА ЗА ПОЖАРОГАСЕНЕ
ИВАЙЛО ГЕОРГИЕВ, ДОБРИН СЕЙЗИНСКИ
Резюме: Гасенето на пожари в горски и планински райони с вертолет с окачваема

система за пожарогасене е сложна задача. Настоящият доклад е посветен на този
проблем и представя актулните в момента предимно практически аспекти на
методологията на неговото изследване.Той цели разгледаната алгоритмизация на
това изследване да бъде използвана за разработване на методика за провеждане на
числени експерименти за определяне на най-съществените фактори, влияещи върху
безопасността на полетите и ефективността на разглежданото гасене на
пожари.
Ключови думи: вертолет, пожари, система за пожарогасене
METHODOLOGICAL ASPECTS OF SAFE AND

EFFECTIVE FIRE FIGHTING BY MEANS OF A FIRE
FIGHTING SYSTEM ATTACHED TO A HELICOPTER
IVAJLO GEORGIEV, DOBRIN SEJZINSKI
Abstract: Extinguishing fires in forest and mountainous areas by means of a fire fighting
system attached to a helicopter is a complex task. This report focuses on this problem and
presents the current, mainly practical aspects of the methodology applied in the research.
The aim of this report is to use the described below algorithmics so as to develop
methodology for conducting numerical experiments which will determine the most
significant factors that affect flight safety and the effectiveness of the firefighting method in
question.
Key words: helicopter, fire, firefighting systems
1. Основни положения стоварването им в района на пожара, както и по-

Пожарите, възникнали и разпрострели се ниските относителни разходи за доставка на 1
на голяма надморска височина, в гори и на тон вода.
трудно достъпни места, могат да бъдaт овладяни В България за първи път гасене на пожар
и потушени предимно с използването на с вертолет с окачваема система е извършено
въздухоплавателни средства. Гасенето става или през 2010 г, като за период от шест години
чрез изхвърляне на вода (пяна) в огнището на вертолетите са участвали в гасенето на десетки
пожара, или чрез направата на заградителни пожари на територията на страната.
полоси, препятстващи разпространението на Използването на вертолети Ми-8 и Ми-
огъня. Предимствата на вертолетите в тези 17 с окачваема система за пожарогасене (ОСП) е
случаи пред самолетите са тяхната маневреност, показало висока ефективност при гасене на
възможността за зареждане на вода от местни пожари на неголеми площи и в труднодостъпни
източници, извозване на пожарникари и залесени местности, където използването на

II-311
наземни сили и средства е много трудно или определяне на най-съществените фактори,
невъзможно. То има забележим ефект само в влияещи върху безопасността на полетите и
началните стадии на разгарянето на огнището на ефективността на разглежданото гасене на
пожара. Ефективност на гасенето на разрастващ пожари.
се интензивен пожар може да се достигне при
изхвърляне на вода в режим на непрекъснато 2. Постановка и задачи на изследването
подлитане. Това изисква използването на голям Обект на изследването е вертолетът Ми-
брой въздухоплавателни средства, осигуряващи 17 с външна окачваема система за пожарогасене
необходимата разчетна интензивност на Bambi Bucket при гасене на горски и планински
подаване на вода и организация на защитни пожари (фиг. 1).
бариери на още незасегнатите от огъня площи
пред фронта на разпространение на пожара. При
голяма височина изхвърлянето на течността
рязко намалява ефективността, тъй като
значителна част от водата се изпарява и не
достига до земната повърхност. При ниска
височина на полета горещият въздух е беден на
кислород, поради което съществува опасност от
помпаж на двигателите и възникване на
катастрофална ситуация. Това може да се
отстрани като се увеличи дължината на въжето
между вертолета и сливния резервоар, но и това
има своята отрицателна страна – опасност от
закачаване в електропроводи, клони на
дърветата, при което точността и
производителността са по-ниски, а вероятността
от попадане на ОСП в опашния витло е доста
голяма. Прокарването на заградителни полоси е
много по-безопано, но е ефективно само при
използване на пяна, което изисква допълнително
специално оборудване. Като цяло гасенето на
пожари с вертолет е крайно опасно. Фиг. 1. Гасене на горски планински
Основният проблем при гасене на пожар пожар с вертолет с ОСП Bambi Bucket
в горски и планински райони с вертолет с ОСП е
осигуряване на високи нива на безопасност и Предмет на изследването е процесът на
ефективност на летателната експлоатация. летателна експлотация на вертолета при гасене
Актуалността на този проблем е обоснована от на планински и горски пожари под
спецификата на летателната експлоатация на въздействието на външни фактори в условията
вертолета при извършване на тази екстренна на извънредна ситуация.
авиационна работа, свързана с използването на Целта на изследването е методическото
опасни режими на малки скорости, с осигуряване на безопасността на полетите и
транспортиране на специални устройства на ефективноста на гасене на планински и горски
външно окачване, с полети в усложнени условия пожари от вертолет с окачваема система за
на експлоатация (на пределно малки височини, пожарогасене.
при намалена видимост, в сложни метеоусловия) За постигане на целта трябва да се решат
и характеризираща се с извънредна срочност, следните основни задачи:
непредвидимост и екстремални организационно-  определяне на характерните особености
технологични условия. на изследвания обект;
Настоящият доклад е посветен на този  разработване на математичен модел на
проблем и представя актулните в момента системата „вертолет – окачваема система
предимно практически аспекти на за пожарогасене” (В – ОСП) с отчитане
методологията на изследване на гасенето на на особеностите на процеса на гасене на
планински и горски пожари с помощта на планински и горски пожари;
вертолет с ОСП. Той цели разгледаната  разработване на математични модели за
алгоритмизация на това изследване да бъде моделиране на отделните съставящи
използвана за разработване на методика за задачи (на потока от носещото витло на
провеждане на числени експерименти за вeртолета, на изтичането на течността от
II-312
системата за пожарогасене, на неразривно свързани една с друга и
колебанията на външно окачената непосредствено зависят от качествата на самата
система за пожарогасене, на възходящия система и от качеството на пилотирането.
поток от огнището на пожара и др.); Качествата на системата се характеризират с
 провеждане на числени експерименти с нейната динамика на полета при въздействие на
математичните модели на системата в външните фактори и съществено зависят от
различни ситуации при приетите надеждната и безопасна работа на нейната
условия; конструкция, силова установка и функционални
 доказване адекватността на модела на системи. От своя страна, качеството на
обекта на изледването чрез съпоставяне пилотиране се определя от теоретичната и
на резултатите от числените и натурни практическа подготовка на членовете на
експерименти; екипажа, управляващ системата, както и от
 решаване на практически задачи за тяхното разбиране за динамиката на полета в
повишаване на ефективността на гасене различни ситуации и знанието на съответните
на пожари с използване на предлаганите инструкции по експлоатация. Това налага чрез
теоретични подходи при осигуряване на аналитични методи да бъдат обработени голям
необходимото ниво на безопасност на брой параметри и експлоатационни
полетите; ограничения, влияещи на режимите на полета и
 разработване на нови изисквания, изискващи повишено внимание от страна на
препоръки и предложения по пилотите. На базата на тази обработка трябва да
летателната експлоатация на вертолета с се определят тези от тях, които определят
окачваема система за пожарогасене при ефективността и безопасността на системата при
провеждане на дейности, свързани с гасене на горски и планински пожари.
гасенето на горски и планински пожари. За изследване динамиката на полета на
За решаване на тези задачи е необходимо всеки летателен апарат в областта на пределните
добро познаване на физичните процеси, режими на полета, включително и на системата
протичащи при разглежданото изследване, както В – ОСП, най-рационално е използването на
и използваното на надеждно математично теоретичните методи и най-вече
моделиране. математическото моделиране. Достойнството на
теоретичните методи е в следното: ниска
3. Средства и методи на изследването себестойност; възможност да се моделират
За решаване на задачите на изследването особени ситуации, включително катастрофални;
като цяло се използва системният подход с възможност да се задават необходимите външни
използване на следните методи: въздействия. Тези методи успешно се използват
 аналитични; от всички автори [1...8], занимаващи се с
 математично моделиране; изследване динамиката на полета на вертолет с
 за решаване на системи обикновени товар на външно окачване, включително с ОСП
диференциални уравнения; при гасене на пожари.
 за решаване на трансцедентни При разработване на математичните
уравнения; модели на индуктивния поток от носещото
витло на вертолета, на възходящия поток от
 от теорията на вероятностите и
огнището на пожара, на изтичането на течността
математическата статистика;
от системата за пожарогасене, на колебанията на
 на летателния експеримент;
външно окачената система за пожарогасене
 на лабораторния експеримент; може да се използват подходите и насоките,
 на математичната обработка на използвани в [1...6], като се изберат най-
резултатите; подходящите за постигане на поставената цел.
 на системния анализ. Например, при моделиране на облака от водни
Анализът на резултатите, получени чрез капки, образуван след пускането на водата от
приведените по-горе теоретичние и емпирични ОСП Bambi Bucket и покриването на
научни методи на изследване, позволява определената площ, може да се приеме, че газът
взаимно да се съпоставят качествените и е непрекъсната среда, а течността (водата) е
количествените показатели. дискретна среда, при изследването на която се
При определяне на характерните прилага подходът на изучаване поведението на
особености на изследвания обект трябва да се отделните капки (фрагменти) [1]. В този случай,
отчита, че високата ефективност и безопасност за икономия на изчислителни ресурси се приема,
на полетите на системата В – ОСП са че траекториите на големи групи от капки са

II-313
еднакви т.е. не се отчита сблъсъкът между Пловдив. Резултатите от лабораторния
самите капки вътре в групата. За решаване на експеримент се съпоставят с тези, тези получени
системата от уравнения се използва методът на при численото моделиране с предварително
крайните обеми. изработения математичен модел.
Провеждането на числените При решаване на практически задачи за
експерименти с математичните модели на повишаване на ефективността на гасене на
системата в различни ситуации при приетите пожари при осигуряване на необходимото ниво
условия може да се осъществи с използване на на безопасност на полетите може да се използва
утвърдили своята работоспособност и следният подход. С използване на метода на
ефективност програмни продукти. Например, математичното моделиране се получават
численият експеримент с модела на облака от параметрите на движение на ОСП за различни
водни капки, образуван след пускането на значения на интензивността и направлението на
водата от ОСП, може да се осъществи с ветровия порив при различни скорости на
помощта на програмния продукт Flow vision [9]. полета вертолета. Определят се условията на
Този продукт позволява да се отчита цялото възникване на явлението на динамична
поле на течението без да се отделят неустойчивост на ОСП при нейното
особеностите и има подходящи възможности транспортиране на вертолета-носител. На базата
при задаване на движението на телата и на това се формулират основните препоръки за
граничните условия. За числения експеримент с използване на системата В – ОСП.
моделите на движение на вертолета и взаимното По аналогичен начин се разработват
положение на вертолета и ОСП може да се нови изисквания, препоръки и предложения по
използва програмният продукт HeliCargo [6]. летателната експлоатация на вертолета с
Доказването на адекватността на окачваема система за пожарогасене при
разработените математични модели става чрез провеждане на дейности, свързани с гасенето на
съпоставяне на резултатите от числените и горски и планински пожари.
натурни експерименти.
Оценката на адекватността на 4. Очаквани резултати
математичния модел на движение на системата Като резултат от числените и натурни
В – ОСП може да се направи с използване на експерименти се очаква:
следния подход. Първо се оценява адекватността  отделяне на основните фактори, влияещи
на модела при полет на вертолета без ОСП чрез върху ефективността на използване на
сравняване на резултатите от числения вертолета с ОСП за гасене на пожари
експеримент с тези от записите на параметрите (скорост и височина на полета, маньовър
на полета от летателния експеримент. След това за пускане на водата, секунден разход на
същият модел преминава проверка на течност);
адекватност на моделирането на поведението на  пълно изследване на особеностите на
ОСП чрез сравняване на резултатите от вертолет с ОСП, както в нормални, така
числения експеримент и от летателния и в екстремни ситуации;
експеримент с празна окачваема система.  осигуряване икономия на ресурси чрез
Летателните експерименти се провеждат използване на математически модели за
с вертолет Ми-17 с ОСП – Bambi Bucket на сметка провеждане на ЛИ;
летище Крумово (надморска височина 182 m), в  провеждане на анализ на особените
района на Асеновград (надморска височина 1000 условия на експлоатация на В – ОСП на
m) и в района на Баташки снежник (надморска различни етапи на полета в рамките на
височина 2000 m). Експериментите се ограниченията, с цел оценка пределните
извършват при хоризонтален полет със скорост експлоатационни възможности на
Vпол= 60-180 кm/h и при различно запълване с вертолета;
вода на резервоара на ОСП. Дължината на  използване и внедряване на получените
въжето от мястото на окачване на вертолета до резултати от изследванията в научни
ОСП се променя с цел постигане на най-добри организации и експлоатационни
резултати за пожарогасене с отчитане предприятия във вид на лекции, учебни
непопадането на празната ОСП в областта на пособия, инструкции и методики по
опашното витло. осигуряване безопасността на полетите;
Лабораторните експерименти се  разработване на нови методи за
провеждат в аеродинамичната тръба УЛАГ-1 на съкращаване на работния цикъл с цел
катедра Транспортна и авиационна техника при повишаване ефективността и
Технически университет – София, филиал надеждността на екипажа;
II-314
 използването на нови способи за пускане неравновесным процессам в соплах и струях
на вода, различни от хоризонтален полет (NPNJ-2008), 24 – 31мая 2008 г, стр. 98.
(от горка или разворот на горка), да 2. Борисов, И.В. и Ципенко, А.В. (2012).
осигури необходимото безопасно Вычислительный эксперимент для анализа
отдалечение на вертолета от опасната работы вертолета с водосливным
конвективна зона на пожара; устройством. Электронный журнал „Труды
 провеждане на разследване на АП и МАИ“, № 53.
инциденти с помощта на решаване на 3. Володко, А.М. (1992). Вертолет в особой
обратната задача от динамиката на ситуации, 262с. Транспорт, Москва.
полета на вертолета с ОСП и на тази
основа да се направят препоръки за 4. Володко, А.М. (2008). Влияние
недопускане в бъдеще подобни събития; транспортируемого груза на эффективность
управления вертолетом. Научный вестник
 разработване на препоръки и указания за
МГТУ ГА, № 125, стр. 191.
обучение и тренировки на екипажите на
вертолети при използване на ОСП в 5. Ефимов, В.В. (2012). Исследование условий
нормални и особени случай на полета. возникновения особых ситуаций при
транспортировке груза на внешней
5. Заключение тросовой подвеске вертолета. Научный
Използването на вертолети за гасене на вестник МГТУ ГА, № 175, стр. 36.
планински и горски пожари е абсолютно 6. Ефимов, В.В. (2012). Исследование
необходимо. Местата, където може да се автоколебаний вертолета с грузом на
използва вертолетът, в повечето случаи са внешней тросовой подвеске. Научный
недостъпни за пожарникарите с техните вестник МГТУ ГА, No 177, стр. 65.
мобилни средства. От друга страна бързо 7. Козловский, В.Б., Паршенцев, С.А.,
променящите се метеорологични условия в Ефимов, В.В. (2008). Вертолет с грузом на
планината, липсата на достатъчно кислород внешней подвеске, 304с. Машиностроение-
вследствие на пожара или надморската Полет, Москва.
височина, може да окажат влияние върху
8. Паршенцев, С.А. (2005). Надежность
експлоатационните и аеродинамични
функционирования системы „экипаж -
характеристики на вертолета. Практическата
вертолет - груз на внешней подвеске“ в
значимост на изследванията се състои в това, че
условиях развития неблагоприятного
получените резултати могат да бъдат
фактора. Общероссийский научно-
използвани за усъвършенстване на външната
технический журнал „Полет“, № 4, стр. 34.
система за пожарогасене, технологията и
тактиката за използването и, както и да се даде 9. Система моделирования движения
отговор за температурния диапазон в района на жидкости и газа FlowVision.
пожара, имащ пряко отношение върху работата http://www.tesis.com.ru/software/flowvision/.
на двигателите. Могат да се определят и
факторите, оказващи най-голямо влияние върху
ефективността и безопасността на полетите на
системата В – ОСП. На тяхна основа могат да се Ивайло Георгиев
разработят нови изисквания, препоръки и ТУ – София, филиал Пловдив
предложения по летателната експлоатация на Адрес: гр. Пловдив
вертолетите с ОСП. ул. Николай Коперник - 12
Телефон: 0898 741 400
ЛИТЕРАТУРА E-mail: iv4o_markov@abv.bg
Добрин Сейзински
1. Борисов, И.В., Паршенцев, С.А., и Ципенко, ТУ – София, филиал Пловдив
А.В. (2008). Моделирование слива Адрес: гр. Пловдив,
жидкости из контейнера ВОП-3 на внешней ул. Цанко Дюстабанов – 25
подвеске вертолета с учетом потока от Телефон: 0895 587 410
несущего винта. Материалы VII E-mail: dobsey@abv.bg
международной конференции по

II-315

INDUSTRIAL MANAGEMENT
Индустриален мениджмънт
PROBLEMS IN RECRUITING HUMAN

RESOURCES FROM BULGARIAN INDUSTRIAL
ENTERPRISES
TONI MIHOVA, VALENTINA NIKOLOVA - ALEXIEVA
Abstract: The report is an attempt to analyze the problems associated with finding suitable
human resources by industrial enterprises in Bulgaria. For this purpose, the authors were
based on two national surveys conducted by ManpowerGroup and Bulgarian Industrial
Association, as well as its own investigation on the causes of this problem. Have been
formulated conclusions and guidelines for improving the activity of selection of human
resources in Bulgarian industrial enterprises.
Keywords: human resources, recruitment, motivation, remuneration, training and
development
1. Introduction selection of human resources. Extremely a thorough

studies on this issue have been issued by Dimitar
Modern industrial enterprises operate in an
Shopov, Margarita Atanasova, Tatiana Hristova
extremely dynamic development and application of
Dimitar Kamenov and Jordan Bliznakov.
new equipment, technology, communication and
According to the authors, definition that most
information systems. Nevertheless, the human
accurately reflects the nature of the selection of
factor occupies a leading position as the main
human resources is: "The selection of human
productive and creative force in the enterprise.
resources in the enterprise be an activity to attract
The management of human resources is
suitable candidates for a post evaluation of their
directly related to the competitiveness and
merits in terms of job requirements and then
efficiency of each company. Activities on
deciding whom to make a job offer. [1]
recruitment and selection of staff play a leading role
The main objective of selection is to ensure
in attracting candidates who meet the requirements
"the best man" for the job. It depends on the
for education, experience, training, contributing to
objective and accurate forecast of future job
the achievement of organizational goals. This
performance of candidates.
determines the topicality of the theme of the
The selection process of candidates includes
report.
a number of activities that can be defined as its
The subject of research are the difficulties and
major stages:
problems faced by industrial enterprises in selecting
the most appropriate human resources. The authors
base their findings on two national studies as well •Determination of requirements for applicants
1
as own research conducted in industrial enterprises
in the Plovdiv region. •Attracting good candidates
2
2. Definition of the recruitment and •Assessing the qualities of the candidates

3
selection of human resources
•Pre-screening of applicants on the basis of the submitted
In connection with the objective to 4 documents
understand what the problems and difficulties in
recruitment and selection of human resources will 5
•Deciding who should be offered the vacant post.
be clarify in a systematic way the essence of this
process. In the specialized literature has several
publications about the nature of recruitment and
II-317
Basic methods and tools for contacting potential Rising sales of powerful smartphones and
candidates for vacancies in the enterprise are: tablets will probably lead to a time when their use
- Advertise the vacancy; will overtake those of desktops. This in turn brings
- Oral advertising by relatives and friends; to the fore the question of mobile selection, and
- Attracting good candidates with the help of expectations are to become a major recruitment
employees in the enterprise; channel in the future.
- Contacts with employment exchanges;  Gamification
- Contacts through social networks; According to one of the leading analytical
- Contacts with students in their respective companies in the world - Gartner, the so-called
schools. gamification is the trend of using different gaming
Business practices shows that this is an mechanisms in the non-gaming situations, such as
extremely complex and responsible process, which the selection of personnel. The idea is thus to
depends on the overall performance of the increase the motivation of the people or to provoke
enterprise. Therefore it is necessary human change in their behavior. Gartner forecast that in the
resources specialists have appropriate training and next five years the use of such methods will become
competence to be able to effectively carry out "a pretty significant trend." Factor for this is the
activities in the selection and recruitment. growing market of developers, offering
opportunities for the use of game elements in
3. Contemporary tools in recruitment and processes like recruitment and management of staff.
selection of human resources The use of such methods for example can allow
Besides the necessary training and more rapid identification of candidates who possess
competence recruiters also need to monitor the necessary skills for a position. Game elements
emerging technological novelties on the process of can be used also for better communication and build
recruitment and selection of staff. One such new relationships between talent within an organization.
challenge is recruitment through the use of so-called  Collaboration tools
„social and mobile channels“. According to British Despite its great potential for sharing and
electronic edition HRmagazine, "the market already exchanging information currently professional
has many social tools that allow you to quickly find social networks are not as used by recruiters.
people - this is not a problem" [2]. However, if used effectively, this type of networks
In this sense, the question is not only which can contribute to a better integration of the
channels to select are the most suitable for the processes of selection and management of human
needs of industrial enterprise, but also how to be resources with other departments and activities
used in the most effective manner. For example, within an organization, which ultimately contribute
according to marketing firm 4MAT, currently to increasing the efficiency of the selection process .
between 20% and 30% of traffic to websites for  Technology to determine the location
careers and recruitment is generated by users who The idea of using this type of tool is that it
visit mobile versions of the relevant pages [2]. allows both employers and HR professionals in an
We investigated various technologies and tools easy and inexpensive way to target jobseekers who
for selection, which we believe have the potential to are often on the move. The use of technology to
change the face of the HR industry and provide a determine the location (most often through the IP
rating as follows: address of the mobile phone or the computer of a
 Systems for tracking applicants (applicant user) enables companies to provide potential
tracking systems, ATS) employees information that is relevant to the area in
The first such instruments appear early 90s which they are located. "When you ask people
of the last century and are among the earliest looking for work, what is important to them three
opportunities for online selection. At the very things that always stated are salary, position and
beginning their functionality is limited to tracking location",explains John Salt from the
incoming company CV-s, but lately it offers TotalJobs.com. According to him the possibility of
opportunities to expand in various areas, including receiving targeted ads depending on their location is
for example compliance checks skills, Analysis of already something jobapplicants seeking and HR
CV-th and posting job advertisements through professionals should be aware.
various communication channels, including social  Video interviews
media. In practice, modern ATS is a platform that In 2016, according to the head hunting
allows HR professionals to connect various tools company OfficeTeam UK, 45% of recruiters have
and services used in the selection process, and thus increased their use of video to conduct job
streamline operations. interviews compared to 2013. While it should not
 Instruments for mobile selection be taken entirely as an alternative to actual face to
II-318
face meetings, interest in such tools grows and - the majority - 50% - provide additional
factors are speed and convenience they offer. training and development of current staff;
Furthermore, the video channels are a natural - second, employers indicated hiring people
environment for the majority of the representatives outside the traditional range of talent - older or
of the younger generations, which means that HR younger than usual employees;
professionals will have to accept their increasing - one-third of employers (33%) use
importance in the selection process. outsourcing activities;
 Internet sources - 30% of surveyed employers offer a larger
For a number of recruiters using social salary to employees, and 29% - offer additional
networks like LinkedIn, Twitter, Facebook etc. is motivational packages.
now living part of their work. At a later stage in this - 22 percent say changing existing business
category includes the use of a wider range of models - such as flexible or teleworking;
sources and head-hunting techniques for more - оnly 9% of respondents said they did not
difficult to detect talents. The challenge for HR follow any strategies, and 2% said they did not
professionals in this case is rather in how to know.
optimize the use of such sources, so to maximize In 2016 most sought after by employers in
efficiency. Possible solutions to this effect are as Bulgaria are skilled workers - they change their
applications that enable semantic search, where place in the rankings with engineers who were most
results are based on the meaning of search scarce last year. In 2016 IT talents are in 10th place
keywords rather than specific matches word for to fourth position in 2015. Sales representatives and
word. For now, however the use of such tools is not project managers were replaced by laborers and
yet extensive, indicated by HRmagazine [2]. operators proceedings. In 2016 remained large
4. Results from research imbalance between demand and supply of skills.
To realize the goal of the report and to The second study was conducted by the
reveal the main difficulties and problems in the Bulgarian Industrial Association (BIA) in 2015,
recruitment and selection of human resources from whose results were announced at a round table with
industrial enterprises, we can use three surveys. the participation of the World Bank and the "Open
The first study is a national survey of Society" Institute [4]. The conclusions from it are:
ManpowerGroup [3]. According to him, nearly two - Businesses looking for staff with
thirds (62%) of employers in Bulgaria have interdisciplinary knowledge and skills hybrids -
difficulties to fill their vacancies in 2016. In combinations of technical capabilities and
comparison, the percentage of Bulgarian employers marketing flair and entrepreneurial attitudes;
shared these difficulties was 50% in 2015. About - Labor market is valued also skills in
one-fifth (24%) of respondents consider that they development, creation and implementation of new
are more difficult to find suitable candidates for the technologies, willingness to learn and improve;
job in 2016. The situation has facilitated only in 6% - Required skills for teamwork, leadership,
of Bulgarian employers. decision making, problem solving, communication,
The main reasons for difficulties in filling emotional intelligence, commitment to change;
the vacancies, cited by employers, are: - The advantages are quickness, flexibility,
- first, with 32% - lack of so-called required. adaptability, organization, personal effectiveness,
hard skills (eg qualifications relating to certain courage to assume responsibility, reliability,
tasks, IT skills, or language and mathematical attention to detail and focus on the customer.
skills); Important to many areas are also ability to work
- secondly with 30% - lack of candidates; under pressure and stress, self-control and time
- third with 13% - lack of so-called required management.
soft skills (professionalism, enthusiasm, According to this study do not reach
interpersonal skills, flexibility / adaptability); professionals in the software industry,
- 12% of employers indicate lack of experience nanotechnology, mechatronics, mobile communi-
and talent in candidates; cations, applied sciences, energy, chemistry,
- 11% of employers indicate higher expectation construction, mechanical engineering.
than the remuneration offered. The third study was conducted by the
The results of this study also indicate that: authors of the report during the period February
Employers in Bulgaria suffer most from a lack 2016 - December 2016 into industrial enterprises in
of skilled workers, engineers and drivers. the Plovdiv region. In particular industrial
To deal with difficulties in finding people enterprises are from the "Trakia economic zone"
Bulgarian employers use several strategies: and the collection of empirical material is
accomplished with the help of students from

II-319
Technical University - Sofia, Plovdiv branch and the location of businesses; they do not think that
the University of Food Technologies and pupils they will receive the expected salary);
from two vocational schools in Plovdiv. It covers 65 - 19% do not have information about these
managers (from middle and operational level), 587 companies and cannot decide;
employees (121 - 466 engineers and skilled - 11% did not answer the survey questions.
workers) and 332 students (175 students and 157 On the same question (fig.2), students from
pupils). vocational schools have responded as follows:
The methods used are interview and questionnaire. - 13% would apply for a job at the declared
We'll point out some highlights from the survey. vacancy;
The main problem that indicate 92% of the - 61% share the opinion that it would apply (they
managers is the difficulty in finding quality and have other plans for their development; they are not
skilled human resources. Only 3% do not think convinced that will handle the job; have
there is a problem to find what specialists, and 5% expectations for higher remuneration from the
did not answer the question of selection of human alleged);
resources (fig. 1). - 21% are unaware of these enterprises;
[CATEGORY [CATEGORY
NAME] NAME] - 70 [VALUE] %
[PERCEN 5% 60
TAGE] 50
40
30 [VALUE] %
20 [VALUE] %
10 [VALUE] %
[CATEGORY
NAME] 0
[PERCE apply for a share the unaware of did not
job opinion that these answer
NTAGE] it would enterprises
apply
Fig. 1. "What kind of problems are you experiencing
in Fig. 2. "Are you interested to work in the industrial
the staff selection?" enterprises from "Trakia industrials zone"?"
On question of job satisfaction, the results are as - 5% did not answer the poll question.
follows:
- 78% of engineering professionals are largely What are the specific findings of the present
motivated for the most effective job performance. studies?
They are satisfied with the remuneration, training - major problem for industrial companies is
opportunities and career development and working finding quality and skilled human resources;
conditions; - for the most part, employees are motivated the
- 16% are not motivated due to stress at work, most effective job performance and are satisfied
poorly constructed teams and ineffective commu- with the remuneration, opportunities for training
nication; and career development and working conditions in
- 6% of respondents did not answer the question. industrial enterprises;
Highly skilled workers meet the same issues as - the majority of potential job applicants
follows: (students from vocational schools and students of
- 67% are highly motivated to work effectively. universities) are not sufficiently confident in their
Satisfied are from the salary, benefits, training professional employability in the industrial
opportunities and working conditions. enterprises, and they lack the necessary information
- 24% have expectations for higher about.
remuneration; The findings are reason enough to formulate
- 9% did not answer the questions in the survey. some guidelines to solve the problem - the difficulty
Study analysis of the results conducted with of finding the most appropriate human resources.
students enables to point out the following: Above all, it is necessary to build
- 17% would apply for a job at the declared mechanisms for extremely strong connection
vacancy into industrial enterprises from "Trakia between business on the one hand and vocational
economic zone"; schools and universities - on the other. The training
- 53% do not believe they will apply for jobs in of engineers and technical specialists creates a
these companies (do not feel sufficiently prepared;
have other plans for their development; do not like
II-320
serious potential for recruitment for industrial economic cluster and educational board between
enterprises in the Plovdiv region. Plovdiv Municipality, Ministry of Education,
The problem with insecurity of the students in "Trakia Economic Zone" and "Industry Watch".
their preparation is solvable also with the direct
This will create conditions for young people
involvement of business representatives who can
to see alternative for its development as a highly
actively participate in updating curricula, visits
qualified human resources in Bulgarian industrial
enterprises to conduct practices, offering training
enterprises.
programs, organizing various joint events to
increase the practical orientation of the training.
Bibliography
Moreover, this relationship between the two
parties will produce information for industrial
1. Shopov, D. and co., How to manage human
enterprises will construct attitude and willingness to
work in them. resources in the enterprise, Sofia, 2013
The other direction is the activities of HR 2. Seven tool selection that you need,
specialists who can use more advanced and modern https://www.capital.bg/
techniques to attract staff as discussed in the 3. Firms increasingly difficult to find qualified
previous section of this report. staff, https: //technews.bg/article-
The use of social networks is also an 94426.html
important factor in relation to students and students
4. Businesses seeking employees with hybrid
who had been thoroughly studied in contemporary
publications [5]. skills,
Last but not least is the use of various https://technews.bg/article-80429.html
motivational techniques by the management of 5. Tepavicharova, M., Boykova, L., Social
industrial enterprises detained successful staff – networking in the context of increasing
more attractive payment, flexible system of bonuses student motivation in Modern Education,
for stimulation, opportunities for training and career Scientific Works of the Union of Scientists
development. The analysis of the relationship
in Bulgaria - Plovdiv, Serie A: Social
between motivation of human resources and
commitment to the business organization is Sciences, Arts and Culture, Volume II,
thoroughly investigated problems in modern 2015
scientific works. [6] 6. Tepavicharova, M., Analysis of correlation
This is a summary of some of the basic between the motivation of human resources
guidelines for dealing with problems with finding and their commitment to business
the most appropriate human resources that the
organizations: Scientific papers of the
effective functioning of industrial enterprises will
depend on. Union Scientists, Volume II, Serie A,
Plovdiv, 2016
5. Conclusion Contacts
The report is an attempt to explain the Name of the authors:
essence of one of the major problems of industrial Toni Bogdanova Mihova
Valentina Nikolova-Alexieva
enterprises, the difficulty in finding qualified and
Organization:
quality human resources. Technical University – Sofia, branch
The dynamic development of these Plovdiv
enterprises will make this problem even more acute University of Food Technologies -
and limiting the activity in the event it does not Plovdiv
prepare adequate strategies, policies and tactics of Address:
human resources management. 4000 Plovdiv, ul. "Tcanko Dyustabanov"25
4000 Plovdiv, bul. "Maritsa" № 29
The analysis of publications shows that the
Phones:0893 69 06 55 и 0885 696 696
management of these companies is making effective E-mail:
efforts in the right direction. Proof of this are the expert2009@abv.bg
meetings at universities and vocational schools, as valentina_nikolova@abv.bg
well as the formation of a joint partnership with the

CHALLENGES AND OPPORTUNITIES FOR

FLEXIBLE CREDITING OF SMALL AND
MEDIUM-SIZED ENTERPRISES IN BULGARIA
MINA ANGELOVA, DANIELA PASTARMADZHIEVA
Abstract: The small and medium-sized enterprises (SMEs) play an important role in the
economic processes and in this sense the optimization of their functioning is extremely
important for raising the competiveness of the Bulgarian economy as a whole. The purpose
of this study is to analyze, generalize and systematize challenges and opportunities for
flexible financing of SMEs by European funds. A particular focuses of the study are
challenges related to political instability and opportunities coming from European Union
(EU) funding. Тhe major goal of this study is subordinate to a major project, namely to
create a useful model of an information platform that enables SMEs to get information
about the best options for European funding in a quick and accurate manner.
Key words: business financing, credit instruments, small and medium-sized enterprises,
smart information platform, human resources, European financial programs, political
instability
1. Introduction “Juncker plan” (such as COSME) and the

The maturing of markets and market reaffirmation of SMEs as priority in the Rome
relations lead to intense competition and increasing declaration from 26 March 2017 [2] give ground for
consumer requirements, turning the quality into a further examination of capabilities for optimizing
factor for the success and survival of organizations. their role and functioning and the effect of political
The significance of the studied topic is central to the instability on their growth.
discussions, research and organizational activities in In these terms, the objects of the analysis
the field of manufacturing and services, because the are SMEs and the particular focus is on main
opportunities for flexible financing of organizations theoretical concepts regarding SMEs and the
are preconditions for growth of quality and conceptualization of correlation political instability-
competitiveness. economic growth.
The topicality of the study is grounded in The purpose of this study is to analyze,
the current situation in the global market. generalize and systematize challenges and
Dynamically changing economic and political opportunities for flexible financing of SMEs by
environment requires more flexibility in the European funds. But the existence of opportunities
companies and willingness to change in accordance is not enough for the growth of SMEs. The latter
with the new conditions. Organizations face the need accurate, accessible and easy to use
challenge to "fight for survival" in a rapidly information in order to put the possibilities into
changing and uncertain environment. This process effect. Thus, the major goal of this study is
is enhanced in terms of the continuing economic subordinate to a project, namely to create a useful
crisis. model of an information platform that enables
Optimizing the performance and SMEs to get information about the best options for
opportunities for flexible financing of SMEs is European funding in a quick and accurate manner.
extremely important to increase the competitiveness In this sense, the research tasks of this
of the organizations and hence to the Bulgarian analysis are:
economy as a whole. 1) To systematize the theoretical concepts
Despite the circumstances mentioned regarding SMEs;
above, the development of small and middle sized 2) To conceptualize the effect of political
enterprises (SMEs) has been an important area of instability on the growth of SMEs;
EU’s policies [1]. The recent measures under
II-322
3) To study empirical data related to access Considering the unstable political environment in
to finance of SMEs; many European countries and in European Union at
4) To list some general options for large it is an up to date issue for Bulgaria as well.
financing from EU. According to a recent analysis in the
opinion to the companies worldwide there are five
2. Role of SMEs in the national economy main obstacle of their growth: access to finance,
According to Commission electricity, political instability, competition and tax
Recommendation 2003/361/EC, as published in the rate [7]. The relationship between political
Official Journal of the European Union L 124, p. 36 instability and economic growth has been examined
of 20 May 2003 “The category of micro, small and in numerous scientific studies over a long period of
medium-sized enterprises (SMEs) is made up of time. This correlation is bilateral. If an economy
enterprises which employ fewer than 250 persons works poorly this may cause “government collapse
and which have an annual turnover not exceeding and political unrest”. And when a political
EUR 50 million, and/or an annual balance sheet environment is unstable usually this leads to
total not exceeding EUR 43 million.” [3] decrease of “investments and the speed of economic
This definition was transposed in Bulgarian development” [8]. In fact national political
legislation and in particular in article 3 of Bulgarian instability may cause some kind of positive effect. It
Law on Small and Medium-Sized Enterprises. can be a factor for internationalization of SMEs,
At European level SMEs are seen as drivers because when national political environment is
of growth, employment and innovation. According unreliable business is motivated to go abroad. [9]
to European Investment Bank they represent over Some major factors of political instability
90% of businesses in the EU and also two thirds of are “political motivated violence”, “mass civil
the active working population is employed in them. protest”, “instability within the political regime”,
[4] and “instability of the political regime” [10].
SMEs play an essential role in economic Nevertheless, it may be caused by various factors
processes and therefore optimization of their but Alesina et al measure political instability as
operation is extremely important to enhance their “propensity of government changes” [8]. The latter
competitiveness and sustainable development. They is a major factor of political instability in Bulgaria
are not only the backbone, but they are nearly the in recent years. In the period January 2013 – March
entire economy of Bulgaria, as according to NSI 2017 there have been six governments, three of
data for 2015 they represent 90.3% of all business which were caretaker governments. The frequent
organizations [5]. SMEs are a major source of change of governments has affected the economic
added value and the largest employer in the country, environment and caused difficulties in financing of
but at the same time are experiencing serious the SME’s.
difficulties in many areas. Undoubtedly,
management must focus on the opportunities for 4. EU instruments supporting SMEs
flexible business financing, and access to various According to analysis commissioned by
financial instruments that could help to overcome Bulgarian Small and Medium Enterprises
the main difficulties and discover horizon for Promotion Agency (BSMEPA) the access to finance
innovation and investment. is a major problem for SMEs in Bulgaria.
According to Bulgarian National Strategy
for Small and Medium-sized Enterprises 2014-2020 30%
25%
a main weakness regarding the state of SMEs in 20%
Bulgaria is underdeveloped information 15%
environment in terms of opportunities to provide 10%
affordable services and project financing. This also 5%
0%
leads to another weakness - support programs for
Costs of production or
Other
Competition
Finding customers
Access to finance
Regulation
Availability of skilled staff
or experienced managers
SMEs funded by the EU, remain largely

underutilized [6]. In this sense, the increase of
labour
awareness among SMEs about EU funding should

be a priority and specific measures should be taken.
3. Political instability as a challenge faced
by SMEs
There are number of studies dedicated to EU 28 Bulgaria
the effects of political instability on economic
growth and particularly on the growth of SMEs.
II-323
Fig. 1. Most important problems for the firms, intermediaries in Bulgaria for loan/guarantee under
2016 COSME programme are CIBANK and
Raiffeisenbank.
Most commonly utilized sources of funding Another option is InnovFin SME Guarantee
are banking instruments, government funding and Facility. This program is specially designed for
international programs to support SMEs and the use research-based and innovative SMEs and Small
of resources of the owner’s family and his/her Mid-caps. It also is carried out through
relatives. [11] Access to finance is a major issue at intermediaries who can be banks, leasing
European level as well. Empirical data from Survey companies, guarantee institutions, etc. The type of
on the Access to Finance of Enterprises shows that support that SMEs can receive is loan/guarantee.
in 2016 “access to finance” is challenge for 8,7% of Bulgarian partners of the programme are DSK
SMEs at European level and for 9,8% companies in Bank, CIBANK, ProCredit Bank, Piraeus Bank,
Bulgaria (Fig. 1.). [12] Eurobank Bulgaria, UniCredit, Raiffeisenbank,
The results on Fig. 2 indicate that there is a United Bulgarian Bank, Deutsche Leasing and
decrease in the significance of this problem in the European Investment Bank [15]
period 2013-2015, but in 2016 there is a slight The options shortly reviewed above can
increase. [12] only be effective if the SMEs in Bulgaria are
informed about them in an accurate and accessible
16% manner.
15%
14%
13% 5. Expected results and effects
12%
11%
The study of opportunities for flexible
10% crediting of small and medium-sized enterprises in
9% Bulgaria can generate new knowledge of the studied
8%
2013 2014 2015 2016 economic and socio-political phenomena and the
relationship between socio-political environment
EU 28 Bulgaria
and funding opportunities for SMEs. Challenges
that might be faced are associated with the ability to
Fig. 2. Dynamics in "access to finance" as a propose alternative, improved versions of existing
problem to the firms information tools concerning opportunities for
European funding organizations. The research
The European Union provides many interest was provoked by the possibility of the
opportunities for grants, loans and guarantees application of new methods and technologies to
available for small and medium enterprises. The detect dependencies between unexplored
financing options often are not direct funding. phenomena of economic, social and political
Usually various national and sub-national spheres.
institutions are intermediaries in this process. These The expected results of the continuing
can be authorities or financial institution such as research work this area are as follows: creating
banks. [13] lasting and sustainable relationships with businesses
Two major programs under “Juncker plan” and construction of model of an innovative
can be mentioned. One of them is Program for the information platform for flexible financing of
Competitiveness of Enterprises and small and SMEs. The results can be used in practice:
medium-sized enterprises (COSME) (2014 - 2020). management and control of business and public
It aims at “strengthening the competitiveness and organizations; financial investments; development
sustainability” of the companies in EU with of marketing campaigns and social programs;
emphasis on SMEs. The program has potential to business decisions with incomplete and inaccurate
support the access to finance for the firms but also information and more.
to widen their access to foreign markets (in EU and
outside). Alongside COSME is expected to 6. Conclusion
encourage the formation of new SMEs and they can In order to continue to carry out its role as
assure economic growth in the European Union. the backbone of the economy SMEs need a special
[14] support. The changing international and national
COSME is a mechanism for guaranteeing environment constitutes an obstacle to their access
loans financed by the European Union and managed to finance. Although European Union provides
by the EIF. It can contribute to reducing structural various measures for promoting creation of SMEs
weaknesses in the financial market for SMEs and to and for support of their growth, the existence of
become more diverse and accessible. Financial

II-324
such is not enough. First of all companies need to 4. http://www.eib.europa.eu/projects/priorities/sm
be inform about them. e/index.htm
A study on the topic will increase 5. http://www.nsi.bg/
knowledge of mechanisms for financing SMEs. In 6. National Strategy for Small and Medium-sized
this case the general purpose should be to develop Enterprises 2014-2020 - Small Business Act,
an innovative information platform. It will work in adopted by Decision № 37 of the Council of
favor of the business to quickly and easily orient Ministers of 23 January 2014;
them to the most appropriate financing option. 7. Wang, Y. (2016) What are the biggest
The research has the potential to enhance obstacles to growth of SMEs in developing
and develop the scientific knowledge in the countries? e An empirical evidence from an
following three areas: enterprise survey, Borsa Istanbul Review 16-3,
1) Displaying the challenges facing SMEs pp. 167-176
in the Bulgarian context, providing opportunities to 8. Alesina, Alberto, Sule Ozler, Nouriel Roubini,
enhance their competitiveness and sustainable and Phillip Swagel. (1996) Political instability
development and positioning in the global market. and economic growth. Journal of Economic
2) Systematization of conceptual framework Growth, 1(2), pp. 189-211
concerning the relationship and mutual influence of 9. Shirokova G. and T. Tsukanova (2014)
political instability and certain economic indicators Internationalization of SMEs from transition
concerning SMEs. In addition to systematization of economies: institutional perspectives, In:
scientific knowledge on the subject an empirical Etemad, H, T. K. Madsen, E. S. Rasmusen, P.
evidence based on Bulgarian political process and Servais (eds.) Current Issues in International
economic development in the period from 2013 Entrepreneurship, Cheltenham: Edward Elgar
onwards should be provided. Pub;
3) Enrichment of knowledge of Information 10. Jong-A-Pin, R. (2008). Essays on political
and Communication Technology can be applied in instability: Measurement, causes and
favor of SMEs. consequences Enschede: Print Partners
Future research effort will be focused on Ipskamp B.V., Enschede, The Netherlands, pp.
building innovative information platform for 4-5
flexible financing SMEs to orient organizations in 11. Vladimirov, Zh, K. Ganev, R. Simeonova-
funding opportunities, grants, guarantees, etc. Ganeva (eds) (2013) Study of entrepreneurship
Furthermore, a "bank" with information on potential and development prospects of innovation in
future employees in organizations (including CVs, SMEs (2012-2013), Sofia: Bulgarian Small and
experience, key competencies) can be integrated in Medium Enterprises Promotion Agency
such platform. (BSMEPA);
The tasks that should be fulfilled in order to 12. http://ec.europa.eu/growth/access-to-
realize such objective are: finance/data-surveys
1) to further analyze, summarize and 13. https://ec.europa.eu/growth/access-to-
systematize opportunities for flexible SME finance_en
financing from European funds; 14. http://eur-lex.europa.eu/legal-
2) to conduct a survey among content/EN/TXT/?uri=CELEX:32013R1287
representatives of small and medium-sized 15. http://www.eif.org/what_we_do/guarantees/sin
organizations and to identify problem areas in gle_eu_debt_instrument/innovfin-guarantee-
connection with flexible financing; facility/
3) to study the experience of EU Member
States in building such information platforms.
Mina Angelova, Daniela Pastarmadzhieva
REFERENCES University of Plovdiv Paisii Hilendarski
24 Tzar Asen
1. https://ec.europa.eu/growth/smes_en 4000, Plovdiv
2. http://www.consilium.europa.eu/en/press/press 00359887461272, 00359884727096
-releases/2017/03/25-rome-declaration/ mina.marinova@abv.bg,
3. European Commission (2015) User Guide for daniela.pastarmadjieva@gmail.com
the SME Definition, Luxembourg: Publications
Office of the European Union;
ПРИЛОЖЕНИЯ НА ИМИТАЦИОННОТО
МОДЕЛИРАНЕ ПРИ ИЗСЛЕДВАНЕ НА
КРЕДИТОРСКАТА ЗАДЛЪЖНЯЛОСТ
В БЪЛГАРИЯ
АНТОНИЯ ЛАЗАРОВА
Резюме: Целта на доклада е да покаже използването на имитационното

моделиране при изследване на кредиторската задлъжнялост.
Ключови думи: кредиторска задлъжнялост, имитационно моделиране, счетоводни
отчети, Закон на Бенфорд
APPLICATIONS IMITATION MODELING IN

INVESTIGATION OF A LENDER LEVERAGE IN
IN BULGARIA
ANTONIA LAZAROVA
Abstract: The purpose of the report is to demonstrate the use of imitation modeling in the
study a lender's debt.
Key words: lender's debt, imitative modeling, accounting records, Benford's Law
1. Имитационно моделиране да отговаря на следните най-общи условия: да

Имитационното моделиране е отговаря на определението за действащо
изследователски метод, при който системата на предприятие, съгласно Закона за счетоводството
изследване се заменя с модел с достатъчна и да е вписан в Търговския регистър, да е
точност. Целта е да се експериментира с модела, кредитоспособен, да се ползва с добра
за да се получи информация за реалната репутация. И не на последно място да има
система. Имитационното моделиране е необходимата за кредитиране гаранция - за
формализирано описание на изучаваното възможността да си погаси кредита. Тя изразява
явление. способността му да посреща задълженията си
към контрагентите и кредиторите, в това число
2. Кредиторска задлъжнялост – към банките своевременно при настъпване на
определение падежа на кредитите.
Общото разбиране за кредиторска
3. Възможности за имитационно
задлъжнялост е сумите, които принадлежат на
моделиране при изследване на
доставчиците за закупени от тях на кредит стоки
кредиторската задлъжнялост в
или услуги.[Error! Reference source not found.]
България
От гледна точка на търговските банки -
те предоставят кредити при спазване на От дефинираните условия става ясно, че
условията на кредитиране, регламентирани в се касае за действащо предприятие по смисъла
посочената нормативна база, даващи на Закона за счетоводството, но при прилагане
възможност за правилното обслужване на на различните показатели за ликвидност от
кредита и нормалното му погасяване. При страна на банките, не се създават възможности в
отпускане на кредит, кредитоискателят трябва цялост да се илюстрира наличието и

II-326
нарастването на кредиторската задлъжнялост в
следващ времеви период. Поради тази причина
предлагаме един теоретичен подход, които
предстои да се апробира, чрез прилагането на
който:
- Ще е възможна синхронизация при работа
със СС 13 - Отчитане при несъстоятелност и
ликвидация [1] и при общите разпоредби на
МСФО е предвидено дружествата да
прилагат стандартите само при сигурни
индикации за “действащо предприятие”,
ето защо такъв стандарт липсва в МСФО.
Съгласно хипотезата на чл. 35, ал. 1 от
Закона за счетоводството СС 13 се прилага и
от предприятия, които като действащи
предприятия са прилагали МСФО.
- Ще се създаде възможност за изглаждане на
някой разлики при отчитане на и
представянето във финансовите отчети, като
например - МСС 1 Представяне на
финансови отчети Регламент [12]
предприятието представя в баланса в състава
на текущите (краткосрочни) пасиви статия
“текуща част от дългосрочен дълг”,
съдържаща подлежащата на погасяване през
следващата година част от дългосрочни
заеми и други финансови пасиви СС 1 - Фиг. 1. Модел на база на имитационното
Представяне на финансови отчети не налага моделиране за изследване на
такова изискване. кредиторската задлъжнялост
- Ще се сведе до минимум възможността за
намаляване на допуснати в хода на От фигура 1става видно, че нормалното
счетоводното отчитане грешки, които по – и законово установеното отчитане на
късно ще намерят отражение в баланса и кредиторската задлъжнялост се извършва в
Отчета за приходи и разходи на рамките на възприетата от предприятието
дружествата. счетоводна политика и изключително много
- Ще се създаде възможност за устойчивост на зависи от факта дали предприятието използва
счетоводната политика на дружествата и ще Националните или Международните счетоводни
се създадат необходимите условия на стандарти. Подходът е изцяло стандартизиран и
съпоставимост на получените данни. не предлага алтернативи. На практика, обаче
- Ще се създадат обективни възможности за при промяна и задълбочаване на кредиторската
метрирано представяне на данните и задлъжнялост на дружествата или при промяна
съпоставимост с оповестяванията в рамките на счетоводното обслужване в рамките на
на отчетите. текущата година, доста често започва да работи
паралелният алгоритъм, който доста често е
4. Имитационно моделиране при съпътстван с първоначална парализа на анализа.
изследване на кредиторска a. “Парализа на анализа” още е наричана
задлъжнялост „информационно претоварване”. Често
При приложение на имитационно „парализа на анализа” възниква и тогава,
моделиране за изследване на кредиторската когато ние прекалено усложняваме
задлъжнялост, счетоводството се разглежда като всичко и се опитваме да анализираме
единна система от взаимосвързани елементи на онези действия, които трябва да се
отчитане. Ще предложим модел на база на правят автоматически.[Error! Reference
имитационното моделиране за изследване на source not found.] Оказва се,
кредиторската задлъжнялост. че независимо колко време и по кои
стандарти се отчита кредиторската
задлъжнялост на дружеството, грешките,
които се допускат в неговите рамки
II-327
нараства, което най-често се дължи на въздействията на отделните елементи.[4]
прекомерно вглеждане в първичната Емерджентните свойства могат да бъдат
счетоводна информация. На практика се овладени и използвани при
случва така, че само се стига до алтернативните подходи, така че
обсъждане на проблема, а не до реални системата да придобие и нови,
действия по отношение на допълнителни свойства. Това на
задлъжнялостта към кредиторите. практика може да се случи на база на
б. Погубване на инстинкти – Обикновено едновременното използване на
това се случва в края на счетоводната стандартизирания и добре познат подход
година, когато се установи, че при при използването на финансово –
анализа на динамиката на показателите, счетоводния анализ на рентабилността,
третиращи кредитната задлъжнялост, задлъжнялостта и т.н. и алтернативните
показват отклонения от нормите. подходи – анализ на база на верижни
Страхът от обявяване на статистически индекси и динамично
несъстоятелност и изпадане в програмиране.
ликвидация доста често блокира д. Алтернативни подходи при изследване
адекватните действия не само от страна на кредиторската задлъжнялост
на счетоводният отдел, но и от страна на  Използване на верижни
ръководството. За преодоляване на статистически индекси за изследване
парализа на анализа и потенциалната на кредиторската задлъжнялост
възможност за погубване на инстинкти След отпадане на НСС 13 –
следва да се използва AGIL. Показатели за финансово -
„Парализа на анализа“ и „погубването на счетоводен анализ на предприятието
инстинкти“ много често се свързват с човешкия [5] все още е възможно в рамките на
фактор в счетоводния процес и им се предава оповестяванията в годишният
оценъчна психологическа характеристика. финансов отчет и като част от
Не бива да се забравя, че именно вътрешнофирменият счетоводен
счетоводителите са онази връзка между данните анализ да се изградят верижни
от счетоводните отчети и мениджмънта, на индекси, като стойностите на
който тези данни се предоставят. Поради тази интересуващите ни коефициенти се
причина наличието само на едно от тези изградят като текущата година към
психични „блокатори на анализа“ е възможно да базисна година. Използването на
деформира не само преценката по отношение на познатите коефициенти, приложими
данните, които да се предоставят на трети лица, при финансово – счетоводен анализ,
но и да компрометира анализа в неговата счетоводителите прилагат повече по
цялостност. навик, отколкото на база на реално
в. AGIL – Това е една цялостна система на взети управленски решения. Така
действие, чрез която се стремим да предложеният модел способства за
отговорим на въпроса – Как да се интеграция на реално приложимите
избегне информационното натоварване и до момента показатели за
да се сведе до минимум страха от финансово – счетоводен анализ и
обявяване в несъстоятелност на статистическите методи за оценка.
дружеството при наличие на висока Този интегриран подход подпомага
кредиторска задлъжнялост? вземането на управленски решения
AGIL е неразделна част от системата за във връзка с кредиторската
действие при наличие на високи нива на задлъжнялост на дружествата като
кредиторската задлъжнялост и по този разкрива следните три важни
начин се проявява само като подсистема елемента – падеж на задължението,
в рамките на по-голямото цяло на стойността му и вероятната
системи в рамките на счетоводната безнадеждност по отношение на това
отчетност. Тя създава възможност за задължение.
нагаждане, постигане на целите,  Използване на динамично
интеграция, следване на модела на оптимиране при изследване на
интеграция .[3] кредиторската задлъжнялост
г. Емерджентно поведение на системата –
Възникване на нови свойства на В много случай процесът на вземане на
системата, различаващи се от сбора на управленско решение се разделя на няколко

II-328
етапа в зависимост от характера на изследване на кредиторската задлъжнялост да се
разглежданата система. Тъй като процеса на интегрират и други модели, чрез които да се
вземане на решение е разложен на отделни изследват всички емерджентни промени на
етапи, разположени във времето и взетите поведението на системата.
решения трябва да доведат до оптимален Най-често фракталът се генерира от
резултат разделът от математиката, който дава повтаряща се схема, обикновено рекурсивен или
методите за решаване на такива задачи се итерационен процес. Това позволява
нарича динамично оптимиране. В него се представяне на неговите характерни
разглеждат широк кръг от екстремални задачи, особености - самоподобността и
по-голямата част от които имат динамичен възпроизвеждането на цялостната структура
характер в смисъл, че при тяхното решаване независимо от увеличението.[7]
трябва да се отчита фактора време или На практика тук става дума за
последователност на операциите. Възприето статистическа самоподобност на фракталите при
като самостоятелно съществуваш елемент в кредиторската задлъжнялост.
системата, той изпълнява двете функции на Фракталът има числени или
AGIL – интеграция и следване на модела на статистически характеристики, които се
интеграция. запазват в различни мащаби, което може да
Системата може да действа в дискретни доведе до прецизна преценка от страна не само
моменти от времето к=1,2,3...n, (n-етапен на вътрешните анализатори на дружеството, но
процес). На всеки етап или момент с-мата се и от външните ползватели на счетоводна
характеризира с някакъв елемент P (състояние информация за промените на кредиторската
на системата), принадлежащ на задлъжнялост.
множеството P(pP). Множеството Р може да има Решенията само на база на
различна природа: фраксипариален анализ за нивото на
- множество от реални числа; кредиторска задлъжнялост, обаче са
- Р може да бъде векторно пространство. изключително трудни са математическо
Тогава състоянието на системата се дефиниране, защото:
определя от параметрите на вектора. - специфичните характеристики на
фракталите са интуитивно разбираеми;
Дискретни процеси, при които - няма точно значение на „прекалено
променливите, характеризиращи състоянието на неравномерен”;
системата приемат стойности само в определени - няма единствено определение на
интервали от време. Непрекъснати процеси „размерност”;
когато интервалите между отделните етапи - има много начини, по които един обект
клонят към 0. Процес на вземане на решения, може да бъде самоподобен;
при който изборът на решение определя - не всеки фрактал е дефиниран
еднозначно резултата от решението, се нарича рекурсивно.[8]
детерминиран процес. На практика често се Чрез рекурсия е възможно да се достигне
срещат задачи на многоетапно планиране, до решение, когато назад в рамките на
в които важна роля играят случайни фактори. счетоводното записване на първични данни се
При тях резултатът от решението не е напълно открият грешки, които следва да бъдат
определен. Изходът от приетото решение е коригирани. Прилагането на пряка рекурсия
определен само във вероятностен смисъл. създава обективни възможност за сторниране,
Такива процеси се наричат стохастични коригиране на стойности и обективизиране на
(вероятностни) (6) процеси, без това да е в разрез с действащата
На практика чрез динамичният анализ счетоводна политика на предприятието и
при изследване на кредиторска задлъжнялост се актуалната нормативна уредба в България.
създава и възможност за точно, ясно и Важно условия е, че на практика прилагането на
конкретно дефиниране на емерджентното пряка рекурсия може да се случи само със
поведение на счетоводната система на счетоводни записвания, които са в рамките на
дружеството. текущата година и поради тази причина пряката
рекурсия следва да се прилага изключително
5. Решения на база на фраксипариален внимателно.
анализ На практика се работи с индиректна,
Фраксипариалният анализ дава косвена рекурсия, която създава възможности за
възможност върху вече съществуващият връщане и прецизиране на счетоводната
класически финансово – счетоводен модел за информация до три години назад. Това, обаче
II-329
означава, че е възможно да възникне финансово отчитане 8. Оперативни
най-слабата форма на самоподобност, а сегменти (12). От друга страна чисто
именно – статистическата. От гледна точка на математическият подход към проблема за
реалността днес, в България, това е кредиторската задлъжнялост на дружествата
изключително масово прилагане на способства за многостранни и различни по
фраксипариален анализ в рамките на задълбоченост анализи, които рано да откроят
използвания счетоводен софтуер, който обаче е тревожни тенденции, ако съществуват такива.
с индиректна рекурсия.
6. Приложение на Закона на Бенфорд В заключение може да се каже,
при имитационното моделиране при че отпадането на НСС 13 създава обективни
изследване на кредиторската възможности за по-пълно използване на
задлъжнялост възможностите, които дава имитационното
Строго математически погледнато в моделиране при изследване на кредиторската
дефиниционно отношение Закона за Бенфорд задлъжнялост в България.
може да се представи като: съществуват такива
случайни величини, при които вероятностното ЛИТЕРАТУРА
разпределение на дробната част на логаритъма
клони към единица в границите на 1. СС 13 - Отчитане при несъстоятелност и
разпределение от нула до едно. ликвидация Обн. ДВ. бр. 30 от 7.04.2005 г.,
Съществуват математически обекти, изм. ДВ. бр. 86 от 26.10.2007 г.
които отговарят на изискванията на Закона на 2. Регламент (ЕС) № 475/2012 на Комисията
Бенфорт: от 5 юни 2012 година за изменение на
- Числата на Фибоначи, известни и като Регламент (ЕО) № 1126/2008 за приемане на
Прогресия на Фибоначи;[9] някои международни счетоводни стандарти
- Факториелите - Факториел е функция на в съответствие с Регламент (ЕО)
цялото число n, равна на № 1606/2002 на Европейския парламент и
произведението на всички естествени на Съвета по отношение на Международен
числа, по-малки или равни на n.[10] счетоводен стандарт (МСС) 1 и
Факториел може да бъде определена и Международен счетоводен стандарт (МСС)
чрез пряка и чрез индиректна рекурсия. 19 (текст от значение за ЕИП)
Приложението на Закона на 3. Talcott Parsons, On the Concept of Value-
Бенфорд [11] при имитационното моделиране Commitments. Sociological Inquiry. Vol.38.
при изследване на кредиторската задлъжнялост Issue 2. pp.135-160. April 1968.
следва да се използва и при класическият 4. Дянков, А. Методи, модели и метафори за
финансово – счетоводен подход по отношение мениджъри, 2010, ПОВВИК АД, pp.26
на публичните отчети на дружествата. 5. НСС 13 – Показатели за финансово -
Истината е, обаче че това се случва счетоводен анализ на предприятието ,приет
изключително рядко. с ПМС No 65 от 1998 г. - обн., ДВ, бр. 36 от
Чрез Закона на Бенфорд се създава 31 март 1998 г.; издание на Народното
обективна възможност в рамките на събрание - притурка към ДВ, бр. 36 от
имитационното моделиране да се установят 1998 г.
отклонения при изготвяне на междинните 6. Дюлгерова, Г. Икономическо моделиране,
счетоводни отчети, при показаните данни за научна статия, 2010, София, с. 5-6.
кредиторската задлъжнялост и при анализа на 7. Mandelbrot B., (1967). How Long Is the Coast
показателите за рентабилност, of Britain? Statistical Self-Similarity and
платежоспособност и задлъжнялост на Fractional Dimension, Science, New Series,
дружеството. Vol. 156, No. 3775. (May 5, 1967),
pp. 636-638.
7. Изводи 8. Манделброт Б., Фракталните обекти:
Основните изводи могат да бъдат Форма, случайност и размерност (прев. от
направени от две различни гледни точки. фр. ез. Петър Лазаров и др.) София: Унив.
От гледна точка на действащата в момента изд. К. Охридски, 1996, ISBN 954 07 0469-3)
нормативна уредба в България имитационното 9. Черноокий, А. Закон Бенфорда и
моделиране при изследване на кредиторската распределения под него попадающие, 2014.
задлъжнялост е възможност за по-пълно с. 4
прилагане на Международните стандарти за

II-330
10. Николов,Т. Естествени, цели и рационални връзка с Международен стандарт за
числа. Математическа логика и анализ, финансово отчитане (МСФО) 8
София, 2011, с. 2 13. Young, C. (2001) English Heritage position
11. Theodore P. Hill. The First Digit Phenomenon. statement on the Valletta Convention, [Online],
A century-old observation about an unexpected Available:
pattern in many numerical tables applies to the http://www.archaeol.freeuk.com/EHPostionSta
stock market, census statistics and accounting tement.htm [24 Aug 2001].
data, pp. 358-363
12. Регламент (ЕО) № 1347/2007 на Комисията
от 16 ноември 2007 година за изменение на Гл.ас. д-р Антония Лазарова
Регламент (ЕО) № 1725/2003 относно Адрес: гр. Пловдив ул. „Задруга“ 18
приемането на някои международни Европейско висше училище по
счетоводни стандарти в съответствие с икономика и мениджмънт
Регламент (ЕО) № 1606/2002 на Телефон: +359 893 336 424
Европейския парламент и на Съвета във E-mail: lazarova.a@abv.bg
.
МОДЕЛ ЗА ИЗБОР НА ДОСТАВЧИЦИ

ОТМАН АХМЕД, СВЕТОСЛАВ ДИМКОВ
Резюме: Настоящата работа се занимава с проблема за избор на доставчици. Цел

на работата е разработване на пределно прост и практически приложим подход за
избор на доставчици като се отчита променливата резултатност на дейността
им. На база на контакти с мениджъри са избрани показатели за оценяване на
резултатността на дейността на доставчиците. Разработени са модели за
оценяване на максималната и минималната производителност на доставчиците.
Извършено е апробиране на моделите на база на предварително набавени реални
данни за мебелна компания в Либия и шест нейни доставчика в Италия, Турция,
Испания и Португалия.
Ключови думи: Избор на доставчици, Показатели за резултатност, Макс-Мин
подход.
A MODEL TO SOLVE THE SUPPLIER

SELECTION PROBLEM
OTMAN AHMED, SVETOSLAV DIMKOW
Abstract: This work deals with the problem of choice of suppliers. Aim of the work is to
develop a very simple and practicable approach to selecting suppliers taking into account
the variable performance of their activity. Based on contacts with managers indicators to
assess the performance of providers are selected. Models for assessing the maximum and
minimum performance of suppliers are developed. There has been committed approbation
of the models based on real data previously supplied from furniture company in Libya and
six of its supplier in Italy, Turkey, Spain and Portugal.
Key words: Supplier selection; Performance measures; Max-min approach.
1. Въведение В литературата са познати множество

Понастоящем разходите за закупуване на модели за избор на доставчици, но тези модели
материали и компоненти от доставчици имат не включват показатели за отчитане на
ключово значение, защото в много различната резултатност на дейността на
индустриални сектори формират значима част доставчиците. За една компания купувач е от
от себестойността на крайните продукти. критично значение да дефинира и използва
Изборът на доставчици има два основни такива показатели в процеса за избор на
подхода. Първият е определяне на оптимален доставчици. Това ще спомогне за получаване на
брой доставчици и вид на взаимоотношенията с точна представа за възможностите на всеки
тях съобразно мащабите на организационната доставчик.
дейност, характеристиките на продуктите и на Настоящата работа запълва тази
пазарите се избира оптимален брой доставчици. празнина като предлага подход за оценяване на
Вторият подход се основава на избор на най- доставчици, отчитащ максималната и
добрите доставчици между познатите. В минималната резултатност на дейността на
настоящата работа се взема под внимание този всеки от тях. Основното предимство на такъв
втори подход като се счита, че броят на подход е, че осигурява на купувача избор на
доставчиците предварително е набелязан. алтернативни доставчици в идентифицирани
групи от доставчици.
II-332
2. Показатели за резултатност на оценяване на дейността на доставчиците варира
доставчици според нивото на оценяване (отделна дейност,
Въпреки изобилието от публикации функционална област, множество функционални
върху избора на доставчици, малцина автори области, организации). Извършеното от Канън и
разглеждат въпроса за показатели за оценяване Перо [5] изследване доказва, че в различни
на резултатността на дейността на доставчиците. ситуации следва да се използват различни
Но и в ограничения брой публикации се показатели. Харланд [6] счита, че ролята и
забелязва значително разнообразие от приносът на определена връзка между
показатели. Доминиращи показатели са цените и доставчик и купувач за достигане на крайните
качеството заедно с навременността на бизнес цели е основен фактор при избора на
доставките и гъвкавостта на доставките. В показатели за резултатност. Това още повече
публикациите през 70-те и 80-те години на 20- увеличава сложността за стандартизиране на
век основно се акцентира върху цените, докато в моделите за оценяване на резултатност.
началото на 90-те години на 20-век акцентът Добре познатият SCOR модел спомага за
пада върху цикъла за доставка и реакцията на разрешаване на проблемите между бизнес
потребителските изисквания. В края на 90-те организации като спомага за обективно
години на 20-век акцентът се премества върху оценяване на резултатността и за
гъвкавостта на доставчиците. Понастоящем идентифициране на възможностите за
освен всички тези показатели се добавя и подобряването й [7]. Това е йерархичен модел
екологичния аспект. В крайна сметка на четири нива. Ниво 1 дефинира пет типови
резултатността на дейността на доставчиците мениджърски процеси. Ниво 2 декомпозира
следва да се оценява като се използват различни типовите процеси в 26 категории базови
показатели в зависимост от характеристиките на процеси, които могат да се използват за
създаваните продукти, влаганите материали, конфигуриране на логистични вериги. Ниво 3
възможностите за удължаване на живота на дава детайлно описание на елементите на
продуктите (рециклиране, многократно базовите процеси от 26-те категории заедно с
използване, повторно производство и др.) и др. показатели за диагностика, целеви стойности
Тенденцията е към разработване на всестранни и (бенчмарки), добри практики и др. Ниво 4 е
по-детайлни показатели за резултатността на реализационно ниво, на което всяка бизнес
дейността на доставчиците. организация къстомизира своите операционни
Така например Холмберг [1] изгражда практики. Понастоящем ниво 1 включва 13
всестранен поглед върху управлението на показателя в пет категории – надеждност на
взаимоотношенията с доставчици и доставките, реактивност, гъвкавост, разходи,
идентифицира проблемите свързани с липсата ефикасност на управлението на управлението на
на системно третиране на логистиката. Той активите.
счита, че липсата на синхрон между системите
3. Max-min подход за оценяване на
за оценяване на доставчиците и общо
доставчици
организационните бизнес цели на купувача води
до дефиниране на изолирани и несъвместими В настоящата работа авторите предлагат
показатели. Освен това акцентирането върху оригинален подход за избор на доставчици чрез
финансови показатели води до предоставяне на използване на непостоянството на
непълни данни на мениджмънта относно резултатността на доставчиците в процеса на
слабостите на взаимоотношенията с оценяване. Max-min подходът, представен в
доставчиците. Бимън [2] констатира, че настоящата работа, позволява всестранно
показателите трябва да притежават четири оценяване на резултатността на доставчиците
характеристики – всеобхватност (отчитане на чрез отчитане на техните максимална и
всички аспекти); универсалност (възможност за минимална ефикасност спрямо поставени от
сравнение при различни условия за дейност); купувача целеви стойности. Друг значим
измеримост (предоставяне на количествени елемент при този подход е, че могат да се
данни); адекватност (съответствие с идентифицират групите от сходни доставчици,
организационните цели). което предоставя на купувачите възможност да
Ремко [3] предлага модел във формата на правят алтернативни подбори при вземане на
матрица, който свързва конкурентността на окончателно решение.
логистични вериги с мястото на доставка в Концепцията върху която се базира Max-
рамките на цялостната логистична верига. Ван min подходът се заключава в максимизиране и
Амстел и Дхерт [4] установяват, че типът на минимизиране на резултатността на
показателите за резултатност, използвани за доставчиците спрямо възможни екстремни
II-333
стойности, дефинирани от купувача. параметър, br изразява тегловния коефициент на
Съчетанието от модели, което се използва при r-тия входен параметър.
този подход спомага да се получат по два В модел (1) целевата функция изразява
показателя за резултатност на всеки доставчик, производителността на доставчик p , която е
като по-високите стойности показват по-добра максимизирана целева стойност, която може да
степен на резултатност. Първият модел е се постигне при отчитане на група от
структуриран по начин да идентифицира ограничения. Първата група ограничения
областите, в които доставчиците показват изразява стойността на производителността на
превъзхождаща резултатност, а вторият служи база на дефинирани от купувача най-добри
за идентифициране на областите, в които стойности (стойност 1). Това показва, че
доставчиците показват слаба резултатност. дейността на доставчика по отношение на
Двата показателя се използват съвместно за дефинирани от купувача най-добри стойности е
постигане на цялостна представа за ефикасна, тъй като нито отделен доставчик или
резултатност на всеки доставчик. група доставчици не могат да доминират [8].
Ако са дадени n доставчици с m Втората група ограничения предотвратява
характеристики купувачите са в състояние да стойностите за ефикасност на всички
идентифицират най-добрите стойности за всяка доставчици да не надхвърлят стойност 1, което е
характеристика сред всички доставчици. повече от нормализиране на ограничение. На
Купувачите използват тази група от оптимални последно място за модела са дефинирани
стойности като целеви стойности при оценяване ограничения за избягване на отрицателни
на резултатност на алтернативните доставчици. стойности.
Също така купувачите могат да използват свои Модел (1) лесно може де се приведе в
стойности за добра практика (бенчмарки) при нормализирана форма, което е показано с
дефиниране на целевите стойности, които са представения по-долу Модел (2):
различни от намерените оптимални стойности.
 ay
v
За всеки случай резултатността на max r 1 r rp
доставчиците, която се дефинира като
 bx
u
s.t. 1
отношение между претеглени изходни s 1 s sp
параметри спрямо претеглени входни  a y  b y 0

v u
r 1 r r* s 1 s s*
параметри, се максимизира и минимизира за
r 1 ar yri  s1bs ysi  0
v u
получаване на двоични групи от стойности за i,
резултатност. Случаят за максимизиране е
показан по-долу като Модел (1).
a ,b  0
r s
r , s
Модел (2) се решава за всеки доставчик
за да се получат стойности за
 ay
v
max
s 1 r rp
производителността. Моделът отчита
 bx
u
s 1 s sp тегловните коефициенти на отношенията
„входни параметри / изходни параметри”, което
 ar y r*  1
v
s.t. su1 не само спомага за поддържане на максимална

s1bs xs* ефикасност по отношение на дефинираните от
 ay купувача група от целеви стойности, но и
v
s 1 r
s.t. ri
 1 i, максимизира ефикасността на доставчика, който
 bx
u
s 1 s si се оценява. Накратко, този модел спомага за
a r , bs  0  r , s , идентифициране на силните страни на
доставчиците, които също са силни страни
спрямо групата от целеви стойности,
Където p показва доставчика, който се
дефинирани от купувача. Така доставчик, който
оценява, v изобразява броя на изходните
постигне високи стойности на ефикасност се
параметри на доставчика, u изразява броя на
счита за добър контрагент при сравняване с
входните параметри на доставчика, yri изразява
изискванията на купувача.
стойността на r-тия изходен параметър на r-тия
Втората част от Max-min подхода е
доставчик, xri изразява стойността на r-тия
свързана с минимизиране на стойности за
входен параметър на r-тия доставчик, yr*
ефикасност на доставчиците спрямо същата
изразява най-добрата стойност за r-тия изходен
система от ограничения, която се използва при
параметър между всички доставчици, xr*
Модел (2). Това е изразено в представения по-
изразява най-добрата стойност за r-тия входен
долу модел (3):
параметър между всички доставчици, ar
изразява тегловния коефициент на r-тия изходен

II-334
min a y
v
1 

( N 1) 
2
,
( )
2
S  X
2
r rp  R ij  N
r 1 N  1  allranks 4 

 
Модел (3) определя тегловните
R( X ij ) е рангът, присвоен на
коефициенти на отношенията „входни
параметри / изходни параметри”, което не само наблюдението j в i-тата група; стойността t е
спомага за поддържане на максимална (1   / 2) квантилът на t разпределение с N — k
ефикасност по отношение на дефинираните от степени на свобода. Стойността  се използва
купувача група от целеви стойности, но и както при теста на Кросъл-Уолис.
минимизира ефикасността на доставчика, който
се оценява. Накратко, този модел спомага за
идентифициране на слабите страни на 4. Апробиране на модела
доставчиците. По този начин доставчици, които
постигат високи стойности за ефикасност се Извършва се оценяване на доставчиците на
считат за добри контрагенти. либийско мебелно предприятие. Това са шест
За пълно решаване на проблема е доставчици на метален обков за мебели от
необходимо прилагането на статистически Италия, Испания, Португалия и Турция.
метод за идентифициране на групи от Данните за доставчиците са взети чрез
показатели на база на двете групи резултати за непосредствени контакти с мениджъри от
ефикасност. мебелното предприятие. Мениджърите на
В настоящата работа се използва мебелното предприятие са счели за ключови
непараметрична статистическа процедура, фактори при избора на доставчици цените на
позната като „Тест на Кросъл-Уолис” за компонентите; качество на компонентите;
тестване на хипотезата, че поне един доставчик коректността на доставките. Цените на
е в състояние да постигне по-високи резултати компонентите са изразени като единична цена за
за ефикасност от поне един друг доставчик. всеки компонент. Качеството на компонентите
Тестът на Кросъл-Уолис, който е базиран на се изразява чрез процент на върнатите
рангове, се използва за анализиране на компоненти при доставката. Коректността на
различията в две или повече независими извадки доставките се оценява чрез процент на
[9]. Базовите единична и нулева хипотези са: закъснелите доставки на компоненти.
H0: Функциите на разпределение на Първичните данни са представени на Таблица 1
всички к съвкупности са идентични. В извършените анализи цената се използва
Ha: Поне една от съвкупностите проявява за входен параметър, а качеството на
склонност за по-високи стойности на компонентите и коректността на доставките –
наблюденията от поне една друга съвкупност. като изходни параметри. Цената се взема за
Статистическа проверка: входящ параметър, защото изразява разходите,
12 k
Ri  3( N  1)
2
които прави купувача. Качеството на
T 
N ( N  1) i 1 ni компонентите и коректността на доставките се
считат за изходни параметри защото изразяват
Където: N е пълният брой на наблюденията; ползите за купувача. Изхождайки от тези
k е броят на групите; ni е броят на наблюденията постановки на модела, по-високите стойности на
в i-тата група, където i = 1 до k; а Ri е сумата от изходящите параметри и по-ниските стойности
ранговете на наблюденията във всяка група на входящите параметри са индикатори за
Правило за вземане на решение: желаните характеристики.
Ако T   2 (k 1, 1   ) тогава се отхвърля
нулевата хипотеза H0, в противен случай се Таблица 1. . Данни за доставчиците
елиминира отхвърлянето на H0, където  е
вероятността за допускане на грешка от първи
тип.
За идентифициране на за различията между
групите се извършват множество сравнения.
Групите i и j са различни ако е удовлетворено За да се спази този принцип е извършено
неравенството: линейно трансформиране на двата изходни
показателя в изследването. Вместо да се
1/ 2
1 1
1/ 2
, използват процент на върнатите доставки и
R  R t  2 N 1T     процент на закъснелите доставки се използват
S 
i j
n n 1( / 2 )
 N k   ni n j 
i j
  разликите от 100%. Така трансформираните
Където: показатели изразяват процент на приетите
II-335
доставки и процент на навременните доставки. на ред „Min. еф.”. на база на тези резултати
Тъй като по-високите стойности на тези може да се направи извода, че доставчик 2
производни показатели изразяват по-високи отново се представя най-добре с резултат 0.930
нива на резултатност, така се постига следван от доставчици 1, 4, 6, 5 и 3 с резултати
съгласуваност с дефинирания модел. Таблица 2 0.913, 0.885, 0.862, 0.861, и 0.853, съответно. В
представя данните за доставчиците след контраст на Модел (2), Модел (3) може да се
извършената трансформация на показателите. определи като модел, който идентифицира
тегловните коефициенти, които правят всеки от
Таблица 2. Трансформирани данни за шестте доставчика да изглежда най-слабо при
доставчиците поддържане на целевата ефикасност 1. Така този
модел акцентира върху слабите страни на всеки
от доставчиците. Фигура 1 изобразява графично
резултатите от двата модела. На фигурата е взет
за пример доставчик 1.
Двойките резултати за ефикасност за всеки
В таблица 2, като е използван модел (2), са доставчик се използват за извършване на теста
изчислени стойности за максимална за на Кросъл-Уолис. Тъй като тестът се базира на
ефикасност. Целевите стойности използвани в рангове резултатите в таблица 3 се преобразуват
модела се получават чрез избиране на най- в рангове, както е показано на таблица 4.
добрите стойности за всеки от трите показателя
за резултатност между всички доставчици. В Таблица 4. Рангове за ефикасност
таблица 2 целевите стойности са 0.1881 $/брой
за цена; и 100% за показателите за качество и
навременност. След шесткратно проиграване на
Модел (2) се получават резултатите за
Присвояването на рангове се извършва като
максималната ефикасност, които са представени
се започне с ранг 1 за най-ниския резултат, ранг
в Таблица 3.
2 за следващия по големина и т.н. Осреднен ранг
Таблица 3. Резултати от max-min се присвоява при наличието на връзки. При
оценяването стойност на теста 0.1 нулевата хипотеза се
отхвърля, което показва, че поне един от
доставчиците е склонен да постига ефикасност
по-висока на поне един от останалите
доставчици.
От таблица 3 се вижда, че доставчик 2
постига най-високи резултати за ефикасност
0.992 следван от доставчици 1, 4, 6, 5 и 3 с
резултати 0.949, 0.904, 0.887, 0.868, и 0.853,
съответно. Тези резултати изразяват
максималните ефикасности, които се постигат
от всеки доставчик, когато се сравняват с
целевите стойности, дефинирани от купувача.
При оценяване на тези стойности модел (2)
избира тегловни коефициенти, които правят
всеки от шестте доставчици да изглежда
възможно най-добре при поддържане на
целевата стойност за ефикасност 1. Така в
Фиг. 1. Резултати от моделите
определен смисъл модел (2) представлява
средство за акцентиране върху силните страни Извършени са множество сравнения и
на всеки от шестте доставчици. Следващият резултатите са представени графично на фиг. 2.
етап от процеса за вземане на решение включва От фигура 2 е видно, че доставчиците от
оценяване на стойностите за минимална група 1 (доставчици 1 и 2) се представят най-
ефикасност. добре, следвани от групи 2, 3 и 4. Въз основа на
С цел да се получат стойностите за извършения тест на Кросъл-Уолис се
минимална ефикасност за всеки доставчик, констатира, че не съществуват значими
модел (3) се прилага за данните от таблица 2. различия на резултатността на доставчиците в
резултатите от анализа са показани в таблица 3 групите. Купувачите могат да използват тази

II-336
информация за ефективно вземане на решение Използването на предпочитанията на
при избор на доставчици. мениджъри ще направи разработения модел по-
реалистичен и по-универсален.
1. Holmberg, S., 2000. A systems perspective on

supply chain measurements. International
Journal of Physical Distribution & Logistics
Management 30 (10), 847-868.
2. Beamon, B., 1999. Measuring supply chain
performance. International Journal of
Operations and Production Management 19
(3), 275-292.
3. Remko, I.V.H., 1998. Measuring the
unmeasurable—measuring and improving
Фиг. 2. Групиране на доставчиците performance in the supply chain. Supply Chain
Management: An International Journal 3 (4),
5. Заключение 187-192.
Въпреки, че бе разработен и успешно 4. van Amstel, R.P., D'hert, G., 1996.
апробиран в практиката Max-Min подход за Performance indicators in distribution. The
оценяване на алтернативни доставчици според International Journal of Logistics Management
показаната от тях резултатност на дейността, е 7 (1), 73-82.
необходимо в бъдеще да се извърши 5. Cannon, J.P., Perreault Jr., W.D., 1999. Buyer-
експериментално изследване за намиране на seller relationships in business markets. Journal
ефективни начини за отчитане на of Marketing Research 36 (4), 439-460.
непостоянството на параметрите на 6. Harland, C, 1997. Supply chain operational
резултатността на дейността им. Разумно е да се performance roles. Integrated Manufacturing
приложи итеративен подход при използване на Systems 8 (2), 70-78.
разработения модел за да се идентифицират 7. Stephens, S., 2001. Supply Chain Council and
онези доставчици, които показват най-добра Supply Chain Operations Reference (SCOR)
резултатност. Model Overview. Supply Chain Council, Inc.
В бъдеще трябва да се работи и върху 8. Talluri, S., 2002. A buyer-seller game model
практическото приложение на разработения for selection and negotiation of purchasing
метод. За целта следва да се проведат множество bids. European Journal of Operational
срещи с мениджъри от различни индустриални Research 143 (1), 171-180.
сектори за идентифициране на специфични 9. Conover, W.J., 1980. Practical Nonparametric
проблеми (входящи и изходящи параметри за Statistics. Wiley, New York.
модела). В крайна сметка трябва да се определи
оптимален брой на параметрите за модели и да
се идентифицират параметрите с най-силно Контакти – Светослав Димков,
влияние. Разумно е да се използват Технически университет – София,
предпочитанията на мениджърите при Стопански факултет
оценяването на значимостта на параметрите. Адрес:1756 София, Климент Охридски 8
Самите мениджъри следва внимателно да Телефон: 02 965 3537
оценят и подберат факторите, които ще E-mail: sdim@tu-sofia.bg
използват с модела съобразявайки се със своите
приоритети, цели и конкретни задачи.
МАРКЕТИНГОВИЯТ МИКС И
СЪВМЕСТИМОСТТА МУ С МАРКЕТИНГА НА
ВЗАИМООТНОШЕНИЯТА ПРИ БАНКОВИТЕ
УСЛУГИ
МАЯ ДИЛКОВА-ПАВЛОВА
Резюме: Настоящият доклад има за цел да разгледа доколко класическата

концепция за маркетингов микс, използванa при транзакционния маркетинг, е
приложима при маркетингa на взаимоотношенията в сектора на услугите и
конкретно при банкирането. Освен основната концепция 4Р са представени още два
варианта на маркетинг микса – 7Р и 4С, които по някои свои аспекти се
доближават до маркетинга на взаимоотношенията.
Ключови думи: маркетингов микс, концепция 4Р, концепция 7Р, концепция 4С,
маркетинг на взаимоотношенията, сектор услуги, банков сектор
MARKETING MIX AND ITS COMPATIBILITY

WITH RELATIONSHIP MARKETING IN
BANKING SERVICES
MAYA DILKOVA-PAVLOVA
Резюме: This paper aims to examine whether the classic concept of marketing mix used in
transactional marketing is relevant to relationship marketing in the services sector and
especially in banking. Besides the main concept (4P) two more variants of the marketing
mix are presented – 7P and 4C. They in some aspects are close to relationship marketing.
Ключови думи: marketing mix, concept 4P, concept 7P, concept 4C, relationship
marketing, services sector, banking sector
1. Предпоставки за избор на сектора на

банковите услуги
По данни на НСИ [1] секторът на
услугите продължава да е водещ с най-голям дял
на БДС. От Фиг. 1 се вижда, че за 2015г. БДС в
сектор услуги е 2/3 от целия за страната.
Голямата по обем добавена стойност, бързото
развитие на сектора и специфичните
характеристики на банковите услуги -
неосезаемост, неделимост, хетерогенност, Фиг. 1. БДС по икономически сектори в млн.лв.
несъхраняемост, доверителна отговорност и за 2015г.
двупосочност на информацията са едни от
главните причини да се фокусираме именно
върху него.

II-338
2. Концепции за маркетингов микс разглежда инструментите от гледна точка на
2.1. Класическа концепция 4P на проф. организацията и е вътрешно ориентирана.
Джером Маккарти Миксът е разработен за масови продукти и не
предлага помощ за персонификация на
Концепцията за маркетинговия микс е
маркетинговите дейности. Потребителите се
една от основополагащите теории в маркетинга.
възприемат като пасивна категория и не се
За нейните поддръжници оспорването ú е
предлагат подходи за взаимодействие и
равносилно на критиките, които е понесъл
изграждане на трайни взаимовръзки с тях. Тези
Коперник за твърдението си, че Земята се движи
аспекти правят класическия маркетинг микс
[2].
неприложим при разработване на маркетинга на
Въпреки това към нея са отправяни
взаимоотношенията.
много нападки от различен характер, основните
от които ще бъдат представени по-долу. 2.3. Други концепции за маркетинговия
Mаркетинговият микс представлява микс
набор от взаимосвързани контролирани
Критиките към класическия 4P модел на
променливи, чрез които организацията се
маркетинговия микс и измененията в бизнес
стреми да задоволи изискванита на
средата налагат предлагане на нови варианти на
потребителите от целевите пазари и по този
микса. В литературата се стрещат много
начин да постигне своите цели.
различни вариации на базовия модел -
Концепцията за маркетинговия микс в
комбинации от думи с първи букви С, Е, Р, R,
производствена фирма за пръв път е
V, акронимът SAVЕ и др.
представена от Нейл Бордън [3] и оригинално
В настоящата разработка сa обхванати
включва 12 елемента: планиране на продукта
две от тях, които имат връзка с маркетинга на
(product planning); ценообразуване (pricing);
взаимоотношенията в сектора на услугите, за да
брендинг (branding); канали за дистрибуция
се изследва дали и доколкото тези две
(channels of distribution); лични продажби
концепции са съвместими:
(personal selling); реклама (advertising);
Маркетингов микс при услугите 7P [6] –
промоции (promotions); опаковане (packaging);
традиционните елементи са допълват от още 3 –
дисплей (display); обслужване (servicing);
хора (people), процеси (processes) и материални
физическа работа (physical handling); откриване
свидетелства (physical evidence)
на факти и анализ (fact finding and analysis).
От гледна точка на потребителите –
Самият Бордън е на мнение, че предложените от
концепция 4C [7] – пречупване на променливите
него елементи на маркетинговия микс могат да
през погледа на купувачите. Елементите в нея
се допълват или някои от тях да отпадат в
са потребителска стойност (customer value), цена
зависимост от конкретната ситуация.
за потребителя/разходи (cost), удобство
Приетият за класически модел на
(convenience), комуникации (communication).
маркетинговия микс е разработен от проф. Е.
Джером Маккарти [4]. Той редуцира 3. Маркетингов микс при услугите (на
елементите, на които организацията пряко може примера на банковия сектор) –
и трябва да влияе чрез разработване на концепция 7Р
съответните стратегии и политики, до четири на Услугите по своята същност се
брой: отличават от физическите продукти с няколко
- Продукт – включващ елементи като свои присъщи качества – неосезаемост,
качество, опаковка, асортимент, сервиз и др. неделимост от дейността, която ги създава,
- Цена – включваща отстъпки, намаления, хетерогенност, несъхраняемост плюс две
условия за кредитиране, състав на цената, характерни за банковите услуги черти –
ценови фактори и др. доверителна отговорност и двупосочност на
- Дистрибуция – обхваща каналите за информацията. Това създава специфични
доставка, складови наличности, транспорт и др. изисквания и открива нови възможности пред
- Промоция / комуникация – разработват маркетинга на услугите. Именно тези
се комуникационните политики: реклама, особености на услугите водят до създаването на
насърчаване на продажбите, лични продажби, нови варианти на класическия маркетинг микс,
връзки с обществеността. така че да се отчитат присъщите им свойства.
2.2. Критики към концепцията 4P На Фиг. 2 са представени елементите на
7Р концепцията за маркетинговия микс и как те
Moller [5] отправя няколко основателни
си взаимодействат с основните характеристики
критики към разработения от проф. Маккарти
на услугите и в частност банковите услуги.
маркетинг микс. Концепцията на Дж. Маккарти
II-339
с едно и също качество. При услугите в
Особености на банковия сектор качеството им силно зависи от
Елементи на 7Р компетенциите на персонала и процесите при
услугите в концепцията
банковия сектор осъществяване на услугата.
1.Продукт Несъхраняемостта на услугите или
1. Неосезаемост 2.Цена невъзможността им да се поддържат като запаси
2. Неделимост 3.Комуникация е предизвикателство пред банките. Тъй като
3. Нееднородност 4.Дистрибуция част от предлаганите услуги са със сезонен
4. Несъхраняемост 5. Хора характер – изплащане на работна заплата,
5. Доверителна 6. Материални данъчни плащания и др. един от начините за
отговорност свидетелства
6. Двупосочност на преодоляване на сезонноста е да се влияе върху
7. Процеси предлагането чрез наемане персонал на непълно
информацията
работно време. Друга мярка е „увеличаване
участието на клиентите за ускоряване на
Фиг. 2. Връзка между характеристиките на транзакциите” [8], което отново ни насочва към
банковите услуги и елементите на концепцията маркетинга на взаимоотношенията.
7Р Специално внимание заслужават двете
допълнителни характеристики на услугите при
Необходимостта от добавяне на хората, банкирането – доверителната отговорност и
материалните свидетелства и процесите към двупосочността на информацията.
елементите на маркетинговия микс произтича от Двупосочният обмен на информация и
характеристиките на услугите. изграждането на доверие са два от основните
Услугите са неосезаеми и съответно елементи на МВ. Именно те са препратки към
основната задача на маркетинга е да намали тази прилагане на маркетинг на взаимоотношенията.
неосезаемост. Това може да се постигне чрез
ефективна комуникация на персонала с 4. Концепция 4 С – маркетингов микс
клиентите, „предоставяне на информация за през погледа на потребителите
лицата, ангажирани с осъществяването на Класическата концепция на четирите P
услугата, в т.ч. за броя, квалификацията” [8], на проф. Маккарти и представената концепця 7Р
както и чрез подходящи материални са подложени на основателна критика поради
свидетелства за извършените/извършващите се факта, че те представят гледната точка на
услуги – кредитни и дебитни карти, договори, производителя/продавача, а не на клиента.
рекламни материали и други. Пример от Погледнато от страната на клиента, тези
банкирането на дребно е превод на работна елементи на маркетинговия микс придобиват
заплата по сметка. Докато преди години други измерения.
работната заплата се изплаща на каса и човек Докато компаниите предлагат
може да пипне и усети парите, при преводите се продукти/услуги, клиентите търсят решение на
работи с абстракции. С цел намаляване на свои проблеми и купуват стойност – напр. при
неосезаемостта банките изпращат напр. една от най-разпространените банкови услуги,
уведомителни смс-и, е-mail-и и др, в които се кредитирането, потребителят получава не
визуализира изплатената сума. толкова пари, колкото напр. средства и
Услугите са неделими от дейността, възможност за инвестиране.
която ги създава. Финансовите услуги първо се Потребителят се интересува от
купуват (сключва се договор) и след това се удобството при намиране и получаване на
произвеждат. В процеса по тяхното услугата, а не от дължината на
осъществяване пряко се включва персоналът. дистрибуционните канали. Именно затова някои
Хората имат ключова рола за приближаването „активни“ банкови компании използват по-
на услугата до изискванията на отделния клиент. гъвкави канали за дистрибуция като посещения
За да се създаде в максимална степен полезния на място при корпоративните клиенти [8].
ефект продавачът (банката в лицето на своите Важен за клиента е разходът, който той
служители) и клиените трябва максимално да си прави, за да бъде извършена услугата, а не
взаимодействат. Тук отново имаме препратка компонентите на цената. Тук се включват
към прилагане на маркетинга на всички разходи – по примера с кредитирането:
взаимоотношения, изградени на система от от привличане на вниманието към ГПР, а не
доверителни отношения. само на лихвите по кредита, към разходите по
Хетерогенността на услугите води до придвижване до банковия клон и не на последно
невъзможността всеки път да се предлага услуга

II-340
място отделеното време за получаване на персонализация, комуникация и интеракции с
услугата. точно определени клиенти.
Промоцията, разбирана като
еднопосочно предаване на информация от 6. Заключение
фирмата към клиентите, отстъпва място на От тази гледна точно концепциите 7Р,
процес на двустранна комуникация. 4С могат да се разглеждат като преход от
Поставянето на клиента, а не на класическия транзакционен маркетинг, основан
продукта в центъра, предлагането на стойност, на 4P концепцията, към маркетинга на
отчитане усилията и разходите на клиента и взаимооношенията, който има нови
разбирането за двупосочна комуникация, а не за концептуални рамки като 30R [9] (разглеждаща
едностранно предаване на информация прави 30 различни типа взаимодействия, илюстриращи
концепцията 4С една стъпка по-близо до ролята на маркетинга като смес от отношения,
ориентирането на фирмата към маркетинг на мрежи и взаимодействия) или като изцяло нова
взаимоотношенията. маркетингова парадигма [2].
5. Маркетинг на взаимоотношенията ЛИТЕРАТУРА

Маркетингът, използван до сега, се
базира главно на отделната сделка и представя 1. http://www.nsi.bg/bg/content/2215/бвп-
една транзакционна перспектива, в чиято основа регионално-ниво
е концепцията за маркетинговия микс. Това ясно 2. Gronroos, C. (1994). From Marketing Mix to
се вижда от определението за маркетинг на Relationship marketing: Towards a Paradigm
Американската организация по маркетинг от Shift in Marketing. Management Decision, vol.
1985г.: „Маркетингът представлява процес по 32, no. 2, pp. 4-20
планиране и реализиране на замисъла, 3. Borden, N. H. (1984). The Concept of the
ценообразуването, придвижването и Marketing Mix, Journal of Advertising Research,
реализацията на идеи, стои и услуги vol 2, Septemberpp. 7-12.
посредством размяна, удовлетворяващи целите 4. Goi, C. (2009). A review of Marketing Mix:
на отделни лица и организации.“ 4Ps or More?International Journal of
В последните години глобалната Marketing Studies, vol. 1, no. 1, pp. 2-15
конкуренция, бързото технологично развитие, 5. Möller, K. (2006). Comment on: The
хетерогенните пазари, все по-изискващите Marketing Mix Revisited: Towards the 21st
потребители и други фактори водят до „промяна Century Marketing by E. Constantinides.
на ударението от транзакционно-ориентрана Journal of Marketing Management, vol. 22, pp.
размяна към изграждане на отношения“[7]. 439 – 450.
Основата на маркетинга на 6. Booms, B. H., Bitner M. J. (1981). Marketing
взаимоотношенията е в ефективната Strategies and organization for Service firms.
комуникация между двете страни, поемането на Marketing of Services, James H. Donnelly and
отговорност от страна на фирмата и William R. George, eds. Chicago: American
предоставянето на стойност за клиента, Marketing Association,pp. 47-51
изграждане на доверие във фирмата, което води 7. Constantinides, E. (2006). The Marketing Mix
до лоялност на клиентите, и като резултат Revisited: Towards the 21st Century
осъществяване на партньорство между двете Marketing. Journal of Marketing Management,
страни. vol. 22, pp. 407-438
Такава глобална промяна в цялостната 8. Младенова, Г. (2007). Маркетинг на
дейност на фирмата налага използването на финансовите услуги, Университетско
нови методи и подходи и прави „старите“ такива издателство „Стопанство“, София
неподходящи, особено в сферата на услугите. 9. Gummesson, E. (2008). Total relationship
Маркетинговият микс се разработва за масов marketing. 3rd edition, Butterworth-
продукт, насочен към целева група потребители. Heinemann, Oxford
Доколкото маркетингът на взаимоотношенията
използва съответнте политики от микса, то те се
разработват спрямо изискванията на конкретния Контакти
клиент и предложенията са силно Име на автора: Maя Дилкова-Павлова
персонализирани. Маркетинговият микс Организация: ПУ „П.Хилендарски“,
отстъпва своето място на нови концепции. ФИСН, катедра УКМИ
Фокусът се измества от продукта към клиента и Адрес:гр. Пловдив, ул. „Цар Асен“ 24
по-конкретно не към масовия клиент, а към Телефон: 0889 45 79 69
E-mail: maya_dilkova@abv.bg
СЪСТОЯНИЕ НА ПАЗАРА НА ТРУДА В

БЪЛГАРИЯ
ТАНЯ ГИГОВА
Резюме: В статията се изследва състоянието на пазара на труда, като се

анализират основните индикатори на пазара на труда.
Ключови думи: пазар на труда, заетост, безработни, работна сила
STATE OF THE LABOR MARKET IN BULGARIA

TANYA GIGOVA
Abstract: The article examines the state of the labor market by analyzing the main
indicators of the labor market.
Key words: labor market, employment, unemployed, labor force
1. Увод населението във възрастовата група 15-64

Пазарът на труда е икономическото навършени години.
пространство, в което се намират и преговарят Икономически активните лица
лицата, които предлагат и търсят работна сила. (работната сила) включва всички наети (заети) и
Пазарът на труда е фактор, които оказва влияние безработни лица. Заети са лицата на възраст 15 и
върху производството и върху икономическата повече години, които през наблюдавания период
активност на населението като цяло, затова са положили труд дори за едни час, за което са
данните за пазара на труда са особено важни за получили работна заплата или друг доход. Като
оценка на икономическото състояние. заети се считат и лицата, които не работят, но
Световната финансово-икономическа криза се имат работа, от която временно отсъстват
отрази особено тежко на заетостта. Целта на поради болест, годишен отпуск, отпуск при
настоящата публикация е да се изследва раждане на дете, родителски отпуск, стачка или
състоянието и да се посочат основните друг трудов спор и други.
проблеми и предизвикателства на пазара на Безработни са лицата на възраст от 15 до
труда, като използваните основни методи са 74 навършени години, които през наблюдавания
сравнителен анализ и обработка на период нямат работа, както и тези които активно
статистическа информация. търсят работа през период от четири седмици
или вече са намерили работа, която очакват да
2. Изложение започнат до три месеца след края на
Основните показатели за пазара на труда наблюдавания период.
по статистически изследвания на Националния Коефициентът на заетост се изчислява
статистически институт (НСИ) [1], като съотношение на броя на заетите лица към
Статистическа служба на Европейския съюз населението във възрастовата група 15-64г.
(Евростат) [2] и Агенцията по заетостта [3], за При коефициента на икономическа
периода 2010-2016 година са обобщени в активност (Таблица 1) [1] за населението на
Таблица 1. възраст 15-64 навършени години за
Коефициента на икономическа разглеждания период 2010-2016 г. сe запазва
активност се изчислява като отношение на броя тенденцията на нарастването му от 66,5 през
на икономически активните лица към 2010 до 69,3% за 2015 г., а за 2016 спрямо 2015
намалява с 0,6 пункта.
II-342
Таблица 1. Основни показатели за пазара на труда в България

Показател 2010 2011 2012 2013 2014 2015 2016
Коеф. на икономическа активност
15-64, % 66,5 65,9 67,1 68,4 69,0 69,3 68,7
Заети лица на 15 и повече

навършени години, бр. (хил.) 3052,8 2965,2 2934,0 2934,9 2981,4 3031,9 3016,8
Заети лица на 15-64г., бр. ( хил.) 3010,4 2927,5 2894,9 2889,4 2927,4 2973,5 2954,3
Коеф. на заетост 15-64г., % 59,7 58,4 58,8 59,5 61,0 62,9 63,4
Безработни лица – общо, бр.
(хил.) 348 376,2 410,3 436,3 384,5 305,5 247,2
Брой продължително безработни

(над 1г.), бр. (хил.) 161,7 210 226,6 250,3 232,5 187,0 146,2
Коеф. на безработица, % 10,2 11,3 12,3 12,9 11,4 9,1 7,6

Коеф. На продължителна
безработица (над 1г.), % 4,8 6,3 6,8 7,4 6,9 5,6 4,5
Обезкуражени лица на възраст 15-

64г., бр. (хил.) 222,3 233,5 225,6 209,7 188,7 170 163,4
Безработни регистрирани в
бюрата по труда, бр. 350944 332601 364537 371380 366470 330816 284707
Равнище на регистрирана
безработица, % 9,5 101,1 11,1 11,3 11,2 10,1 8,7
Безработни лица с регистрация в

бюрата по труда над 1г., бр. 110848 118135 118832 122738 138473 146010 118278
Равнище на безработица средно за

ЕС-28, % 9,6 9,6 10,5 10,9 10,2 9,4 8,5
Недостиг на работна сила в

промишлеността, % 6,5 7,5 8,7 8,6 11,2 15,8 21,7
Заетите лица (фиг.1) [4] и изследвания с 8,2 хил. и култура, спорт и

коефициентът на заетост (фиг.2) са идентични развлечения с 5,1хил. За съжаление се
с коефициента на икономическа активност. наблюдава по-съществен спад на заетите лица
Общият броя на заетите лица на възраст 15 и в: образованието с 12,8 хил., преработваща
повече навършени години през 2016 г. е промишленост с 8,8 хил. и с по 4,2 хил. в
3016,87 хил., като спрямо 2015 г. се наблюдава селско, горско и рибно стопанство, търговия,
намаляване на общия броя на заетите лица с ремонт на автомобили и мотоциклети.
15,1 хил.. Относителният дял на заетите лица
от населението на 15 и повече навършени
години за четвъртото тримесечие на 2016 г. е
49,2%, което е намаляване с 0,5 процентни
пункта в сравнение с четвъртото тримесечие
на 2015 г.. Като при мъжете този дял е 54,9%, а
при жените – 43,9%. Увеличение на заетостта
за 2016 спрямо 2015 г. се наблюдава в
секторите: транспорт, складиране и пощи с
11,6 хил., професионални дейности и научни
II-343
Равнището на заетост във възрастовата
група от 15 до 64 години в ЕС-28 [5] е най-
високо през 2008 г., когато достига равнище
от 65,7%, като намалява през 2010г. с 1,6
процентни пункта и достига до 64,1%. Този
спад по време на световната финансова и
икономическа криза е последван от период на
стабилност между 2010 и 2013 г., когато
равнището на заетостта в ЕС-28 през 2010г. е
64,1 % и 64,2 % през 2013г. Равнището на
заетост измерено в Наблюдението на
работната сила (НРС) на ЕС [6] показва, че
възходящата тенденция отпреди кризата се
възобновява през 2014г., като равнището на
заетостта достигайки 64,9 %, т.е. се увеличава
Фиг. 1. Заети лица с 0,8 процентни пункта в сравнение с 2013 г..
Равнището на заетост през 2014г. е високо и
При коефициентът на заетост за сред държавите членки на ЕС. Най-висока
населението на възраст 15-64 навършени стойност е отчетена в Швеция, 74,9%,
години единствено за 2011г. спрямо 2010г. се следвана от Германия, Нидерландия, Дания и
наблюдава спад с 9,9 процентни пункта. През Обединеното кралство, които се движат в
останалите години се запазва тенденцията на интервала от 74% до 71%. На другия край на
нарастване. За 2015 г. е 62,9%, като в скалата, равнища на заетост под 60 %, са
сравнение с 2016 г. се увеличава с 0,5 отчетени в четири от държавите членки на
процентни пункта и достига 63,4%. За жените ЕС-28: Италия, Испания, Хърватия и Гърция,
е 60%, а за мъжете 66,7%. При останалите като най-ниската стойност е била в Гърция
възрастови групи, също се наблюдава (49,4 % през 2014 г. и 54,9% през 2015г.)
увеличение на коефициентите на заетост за (фигура 3). Равнището на заетост за България
населението (фиг.2). Коефициентът на заетост през 2014 е 61%, а през 2015 нараства с 6,2
за населението на възраст 20 - 64 навършени процентни пункта и достига до 67,2%, което
години за 2016 г. е 67,7%, съответно 71,3% за отново е под средните стойности за държавите
мъжете и 64,0% за жените. Този коефициент се членки на ЕС (70%) и страната се нарежда
е увеличил с 0,6 процентни пункта спрямо отново в долните нива на скалата - пред
2015 г., като за мъжете е с 0,9 процентни Румъния.
пункта, а за жените с 0,2 процентни пункта. Като продължително безработни се
Коефициентът на заетост за населението на приемат лицата, които са безработни от една
възраст 55 - 64 години за 2016 г. е 54,5%, т.е. или повече години.
се увеличава с 1,5 процентни пункта, като Коефициентът на безработица се
увеличението при мъжете и жените спрямо изчислява като съотношение на броя
2015 г. е 1,5 процентни пункта. През 2016 г. безработни лица към населението във
коефициента на младежка заетост (15-29 г.) възрастовата група 15-64г.
намалява съществено с 18 процентни пункта и Коефициент на продължителна
достига до 21,2%. безработица представлява относителният дял
на продължително безработните лица от
икономически активното население.
Обезкуражени са лицата извън
работната сила, които не търсят активно
работа, защото не вярват, че ще намерят, но
желаят да работят.
Фиг. 2. Коефициенти на заетост по

възрастови групи

II-344
2014 намалява с 1,5 процентни пункта и
достига 11,4%. През 2015 г. коефициентът на
безработица е 9,1%, като в сравнение с 2014 г.
е по-нисък с 2,3. През 2016 спрямо 2015 е по-
нисък с 1,5 процентни пункта. През
четвъртото тримесечие на 2016 г.
коефициентът на безработица намалява както
при мъжете, така и при жените, но стойността
му се запазва по-висока при мъжете (6,8%), в
сравнение с тази при жените (6,5%).
Фиг. 3. Равнище на заетостта във

възрастовата група 20 — 64 години в Фиг. 4. Коефициент на безработица
държавите членки на ЕС, 2015 г.
Коефициентът на продължителна
безработица е идентичен с коефициента на
Броят на безработните лица показва безработица (фиг.5). За 2016 г. коефициентът
устойчиво нарастване до края на 2013г.. През на продължителна безработица е 4,5%, като в
2014г. се наблюдава спад с 51,8 хил. бр. сравнение със 2015 г. намалява с 1,1
спрямо 2013 г.. През 2016 г. безработните лица процентни пункта. Коефициентът на
в страната са 247,2 хил., като в сравнение с продължителна безработица намалява в по-
2015 г. броят на безработните е намалял с голяма степен при мъжете, отколкото при
57,9хил.. От общия брой на безработните лица жените.
през третото тримесечие на 2015 г. 57,0%
(159,1 хил.) са мъже и 43.0% (120,1 хил.) са
жени. 12,7% от всички безработни са с висше
образование, 50,9% са със средно, и 36,4% са с
основно и по-ниско образование. Броят на
продължително безработни лица (от една или
повече години) от 2010 до 2013 г. значително
нараства, като от 2014г. се наблюдава
намаляване на броя на продължително
безработните, като 2016г. те са 146,2 хил. или
59% от всички безработни лица. Като цяло
2016г. приключва с намаляваща безработица
спрямо предходната година.
За разглеждания период коефициентът Фиг. 5. Коефициенти на безработица и
на безработица следва тенденцията на коефициенти на продължителна безработица
безработните лица. За периода от 2010 до
За разглеждания период 2010-2016г.
2013г. коефициентът на безработица (фиг.4)
най-висок е дела на обезкуражените лица на
[4] нараства от 10,2% до 12,9%, след което се
възраст 15-64 години през 2011г., когато
наблюдава постепенното му намаляване, като
достигат до 233,5 хил. бр. (фиг.6). От 2012 г.
II-345
до момента се наблюдава тенденция на бюрата по труда над 1 година за периода от
намаляване на броя на обезкуражените лица. 2010 до 2014 запазват устойчива тенденция на
От населението на възраст 15-64 навършени нарастване, за 2014 г. са нараснали с 15735
години за 2016 г. 31,3% са икономически лица спрямо 2013г., 2015 г. нарастват с 7537
неактивните лица, като от тях 163,4 хил. са лица. За 2016 са 118278 лица, като спрямо
обезкуражени лица. В сравнение 2015 г. броят 2015 г. са намалели значително с 27732 лица.
на обезкуражените лица намалява с 6,6 хиляди. Топ десет на най-търсените професии
от работодателите в бюрата по труда през 2014
г. са: продавачи и посредници в търговията и
продажбите; работници в шивашкото и
текстилното производство; преподаватели в
начално, основно и средно образование;
неквалифицирани работници; работници в
преработващата промишленост; градинари и
растениевъди, животновъди и др.; готвачи и
помощници при приготвянето на храни;
работници в добивната промишленост и
строителството; сервитьори и бармани; друг
обслужващ персонал.
По данни на бизнес анкетите на НСИ
[4] се наблюдава тенденция на нарастване на
недостигът на работна сила в промишлеността
Фиг. 6. Безработни лица и обезкуражени лица за периода 2010-2016г., като нарастването за
2014г. спрямо 2013г. е с 2,6 процентни пункта.
По данни на Евростат [5] в периода
За 2016 г. спрямо 2015 г. нарастването е с 5,9
между 2010 г. и 2013 г. сезонно коригираният
процентни пункта и достига до 21,7%. През
коефициент на безработица в ЕС-28 нарасна
март 2017 г. 23,5% от промишлените
от 9,6 % на 10,9 %. През 2014 г.
предприятия посочват недостига на работната
коефициентът на безработица спадна спрямо
сила като фактор, затрудняващ дейността им,
2013 с 0,7 процентни пункта и достига до
т.е. трудовият пазар не отговаря на нуждите на
10,2%. Безработицата в ЕС-28 се е понижила
работодателите.
през 2016 г. до 8,5%, което е с 0.9%, като това
Основните предизвикателства и
е най-ниското равнище за разглеждания период
проблеми на пазара на труда за 2016 г. са:
от 2010г. Безработицата бавно намалява, но
висок дял на безработните, въпреки че
продължава да бъде на високи равнища в ЕС.
безработицата продължава да намалява.
Коефициентът на безработица в България
Въпреки намалението, броят на неактивните
остава по-висок за средното ниво на ЕС. Най-
лица, в. т.ч. обезкуражените, остава висок.
голяма безработица е регистрирана в Гърция,
Трудовият пазар не отговаря на нуждите на
където повече от една четвърт от хората нямат
работодателите. Бавно е нарастването на броя
работа, а най-ниска безработица в Германия и
на новоразкриваните свободни работни места.
Австрия.
Влошена е ситуацията на пазара на труда в
По данни на Агенцията по заетостта
някои европейски страни, в които работят
(АЗ) безработните регистрирани в бюрата по
много българи – в т.ч. Гърция и Испания.
труда през 2011 спрямо 2010 намаляват с
Безработните в България остават повече от
18343 лица. В следващите 2012 и 2013г. се
средното ниво за ЕС.
наблюдава нарастване броя на регистрираните
безработни в бюрата по труда. През 2014 брой 3. Заключение
на безработните е намалял с 4910 души. През
Пазарът на труда постепенно се
2015 г. продължават да намаляват на годишна
стабилизира. Безработицата бавно намалява,
база регистрираните в бюрата по труда
но продължава да бъде на високи равнища в
безработни – тенденция, която се наблюдава
ЕС. Безработицата в България намалява, а
от 2014 г.. Безработните, регистрирани в
заетостта расте, но въпреки това безработните
бюрата по труда, през 2016 са 284 707 лица,
в страната остават повече от средното ниво за
като в сравнение с 2015 г. те намаляват
ЕС. Въпреки положените усилия държавната
съществено - с 46 109 лица. Въпреки, че
институция все още не успява да повлияе
започва да намалява равнището на
значително върху нивата на продължително
регистрираната безработица е притеснителен
безработните и недостига на работна сила и
факта, че безработните лица с регистрация в

II-346
трябва да продължава да работи в тази насока, www.nsi.bg/sites/default/files/files/pages/Ke
трябва да се търсят възможности за yInd/KeyInd2017-04.pdf
повишаване на конкурентоспособността на 5. http://appsso.eurostat.ec.europa.eu/nui/show.
труда. do
6. http://ec.europa.eu/eurostat/statistics-
explained/index.php/Glossary:Labour_force_
ЛИТЕРАТУРА survey_(LFS)
1. НСИ, www.nsi.bg/bg/,
2. Евростат, Контакти:
http://ec.europa.eu/eurostat/statistics- Таня Бечева Гигова
explained/index.php/Employment_statistics/b Технически университет София –
g филиал Пловдив
3. АЗ, www.az.government.bg/stats/1/ Адрес: ул. „Цанко Дюстабанов” №25
4. НСИ, Ключови показатели за България, E-mail: t_gigova@abv.bg
6стр.,
ИЗСЛЕДВАНЕ ХРАНИТЕЛНОТО ПОВЕДЕНИЕ

НА ДЕЦА И ТИЙНЕЙДЖЪРИ В ЗАВИСИМОСТ
ОТ ТИПА НАСЕЛЕНО МЯСТО, В КОЕТО УЧАТ
ЕЛЕНА ЗЛАТАНОВА-ПЪЖЕВА
Резюме: В статията са описани основните логически конструкта, върху които е

изградена методиката на изследване – разработване на представителна извадка,
конструиране на въпросници и създаване на инструмент за събиране на
информация. Представени са част от получените резултати от проведено
емпирично онлайн изследване на потребителското поведение на деца и
тийнейджъри по отношение на храни. Обсъдени са статистически доказаните
връзки между изследвани аспекти от хранителното поведение на децата и
тийнейджърите и типа населено място, в което учат.
Ключови думи: потребителско поведение, дете-потребител, хранителни
предпочитания
SURVEY OF FOOD BEHAVIOUR OF CHILDREN

AND TEENAGERS IN DEPENDANCE OF THE
TYPE OF POPULATED PLACE THEY STUDY
ELENA ZLATANOVA-PAZHEVA
Abstract: The article explores the main logical constructs, on which basis is developed the
methodology of research – development of representative sample, design of questionnaires
and creating a tool for gathering information. Some of the obtained results from the
empirical online study of the consumer behavior of children and adolescents in relation to
food are presented. The statistically proven relations between the observed aspects of
eating behavior of children and adolescents and the type of populated place they study are
discussed.
Key words: consumer behavior, child-consumer, food preferences
1. Основни положения станал значителен, тъй като младите хора имат

Научаването на децата на моделите на по-голям разполагаем доход от предишните
потребителското поведение представлява поколения.
интерес за изследователите от 60-те години на Поради факта, че покупателната сила на
ХХ век, когато за първи път през 1964 г. е младите хора продължава да расте, маркетолози,
публикувано изследване за развитието на производители и рекламодатели, са значително
моделите на потребителско поведение [1]. В заинтересовани в изработването на ефективни
последствие децата започват да бъдат методи, които да достигнат до пазара на младите
признавани за пазар [2]. хора [3]. Тъй като децата са потребители е
Маркетолозите и изследователите на уместно да се изследва тяхното покупателно
потребителите дълго са пренебрегвали децата поведение като се открият мотиваторите, които
като пазарен сегмент, поради малкия им определят от къде пазаруват и какво купуват [4].
разполагаем доход. През последните Днес интересът към децата и тяхното
десетилетия детският и тийнейджърски пазар е потребителско поведение е толкова голям не
само защото формират значим пазарен сегмент и
CopyrightbyTechnical University - Sofia, Plovdiv branch, Bulgaria
II-348
влияят на решенията на своите родители, но и Министерство на образованието и науката,
защото сегашното им поведение в голяма степен подкрепило провеждането на изследването.
ще формира и повлияе на бъдещото им Като признат метод, използван при
поведение. Необходимо е да се разбере международни оценъчно-диагностични
потребителското поведение на децата и как то се изследвания, измерващи постиженията на
формира, за да могат да се създадат адекватни учениците (Trends in International Mathematics
подходи към задоволяване потребностите на and Science Study (TIMSS), Progress in
децата. International Reading Literacy Study (PIRLS),
Въпросът с хранителните предпочитания Progress in International Student Achievement
и навици на българските деца и тийнейджъри е (PISA) [7]) е избрано да се работи с много
особено актуален, тъй като по данни на стъпкова стратифицирана клъстерна извадка.
Европейски конгрес по затлъстялост в Прага от Същността на възприетия метод за
2015 г. българските деца са класирани на пето формиране на извадка може да се опише по
място по степен на затлъстялост в Европа. Този следния начин:
факт изисква предприемане на сериозни мерки • Извадката е много-стъпкова, тъй като
за изучаване на хранителното поведение на последователно се избират най-напред училище
децата и тийнейджърите и на факторите, които и след това паралелки в тези училища;
им оказват влияние. • Стратифицирана е, защото предлага
Детските хранителни предпочитания са възможност за разделяне на училищата или
силно свързани с техните модели на учениците на групи в зависимост от съществен
потребление [5]. Множество изследвания са признак за измерването. За целите на
открили, че хранителното поведение, изследването училищата са стратифицирани
формирано в ранно детство може да устои през според признака – възрастова група. На тази
периода на късното детство и дори в началото на основа са разработени две страти – Страта 1,
зрялата възраст [6]. Именно това прави включваща всички училища в Южна България, в
хранителните предпочитания на децата и които се обучават ученици от 1 до 3 клас, и
тийнейджърите още по–важни за изследване. Страта 2, съответно за училищата, включващи
Обект на проведеното изследване са две ученици от 6 до 8 клас;
възрастови групи – деца на 7-9 години (1-3 клас) • Клъстерна е, защото се избират цели
и тийнейджъри на 12-14 години (6-8 клас), клъстери, в случая паралелки.
идентифицирани чрез статистическа извадка от Разработването на извадката е
училища, териториално разположени в Южна осъществено в две стъпки:
България. Стъпка 1: Избиране на училища.
Предмет на изследване в настоящата Избирането на училище е извършено с
статия са хранителните предпочитания на метода PPS (Probabilities proportional to their
българските деца и тийнейджъри в Южна size), при който вероятността дадено училище да
България. попадне в извадката, е пропорционална на броя
Целта на настоящата статия е да се на учениците от разглежданата популация,
представят статистически доказаните връзки които учат в това училище.
между хранителните предпочитания на децата и Стъпка 2: Избиране на клъстери
тийнейджърите и факторът - населено място, в (паралелки) от избраните вече училища.
което учат, като резултат от проведено Това е извършено посредством проста
емпирично изследване. случайна извадка от списъка на всички
паралелки в даденото училище. Във втората
2. Методика на изследване
стъпка паралелката, а не отделният ученик, е
Три са основните логически конструкта, самостоятелна единица за извадката. От
върху които се изгражда изследването: избраната паралелка в изследването участват
• разработване на представителна извадка; всички ученици.
• конструиране на въпросници; В настоящето изследване използваната
• създаване на инструмент за събиране на допустима грешка на извадката е 5%, а
информация. доверителния интервал 0,95, използвани при
международни оценъчно-диагностични
2.1. Разработване на представителна изследвания.
извадка Полученият резултат от разработената
Обезпечаването на необходимите извадка за всяка страта е, както следва:
изходни данни за разработване на • Страта 1 - попадат 2340 деца от 115
представителна извадка са предоставени от училища;
II-349
• Страта 2 – попадат 2310 тийнейджъра от което са използвани компютърно създадени
115 училища. добронамерени и усмихнати емотикони (Фиг. 2).
2.2. Конструиране на въпросници

Методът за събиране на данни, който е
използван, е посредством въпросници в онлайн
среда. Разработени са три възрастово
съобразени въпросника – за деца, тийнейджъри
и родители.
При конструиране на въпросниците,
насочени към деца, се е изхождало от тезата, че
„децата могат и предоставят надеждни отговори,
ако са попитани по начин, който разбират и за
събития, които значат нещо за тях” [8]. Фиг. 2. Изглед на въпрос от въпросник за
Някои основни характеристики на „Тийнейджър”
съставените два детски въпросника са:
• брой въпроси: въпросникът за деца е със 2.3. Създаване на инструмент за
застъпени по-малко въпроси в сравнение с този събиране на информация
за тийнейджърите, поради различното ниво на Целта, която е поставена, е
когнитивно и езиково развитие; осъществяване на представително изследване в
• обръщение: и двата въпросника започват Южна България, което предполага събиране на
и завършват с обръщение към респондентите, голям обем информация.
което е различно за двете групи и съобразено с Проектът създаден за провеждане на
когнитивното им и умствено развитие; изследването представлява уеб-базирано
• структура: еднаква е и за двата приложение, изградено с цел по-добро
въпросника. Състои се от 7 части – любими взаимодействие с потребителите. Приложението
храни, избор, знание, на училище, в къщи, се характеризира със сравнително ниски разходи
любимо, ти и твоите приятели; по изграждане и поддръжка. Този продукт
• скала за отговор: при децата са позволява неограничен брой потребители да
използвани биполярни скали от типа “да”–“не”, работят с него едновременно без това да
изобразени с емотикони и номинални скали. утежнява нито сървъра, в който се намира
При тийнейджърите скалирането е отразено в продуктът, нито клиентските устройства, които
ординални, биполярни, номинални и Ликърт се свързват с него.
скали. Съвременното високо ниво на
• визуални стимули: При децата компютърна грамотност, както сред родителите,
задаването на въпрос се извършва от „анкетьор”, така и сред децата създаде възможност
който е компютърно създаден плод, зеленчук организацията на изследването да се извърши
или друга храна, което има за цел предизвикване посредством кодове. По този начин се гарантира
на интерес у децата и създаване на чувство за за обезпечаване на изследването с достоверна
спокойствие (Фиг. 1). информация, за опазване на
конфиденциалността на анкетираните деца и
родители и за проследяване на получените
отговори. Същността на организацията се
състои в това, че администраторът генерира
набор от кодове за всяко училище, което попада
в представителната извадка. Всяко от тези
училища, териториално разположени в Южна
България, получава писмо по пощата,
съдържащо обръщение към директорите,
указания до децата и техните родители с
наименованието на сайта и генерираните от
Фиг. 1. Изглед на въпрос от въпросник за администратора кодове. На следващия етап, за
„Дете” да посети онлайн базираното приложение,
потребителят трябва да въведе наименованието
При тийнейджърите „анкетьорите” са на сайта http://www.food-interests.eu/ (активен в
адаптирани за малко по-големи деца, поради периода 09.2015-12.2015 г.)в адресната лента на
уеб браузъра на компютър, таблет или мобилен

II-350
телефон, което го прави най-често срещания признак и изследван аспект на хранително
начин при изграждане на уеб-базирани поведение. Алтернативната хипотеза ( H 1 )
приложения. [9] съдържа предположение за съществуване на
3. Резултати статистически значима връзка между двете
основания на класификациите и се
Инструментът, посредством който е
противопоставя на H 0 . За всички връзки, които
извършвана статистическата обработка на
получените данни, е статистическият пакет за ще бъдат представени в анализа, се отхвърля
социални науки на IBM SPSS Statistics Trial нулевата хипотеза и се приема алтернативната,
Software. Използваният метод за статистическа тъй като за тях p < 0,05 .
обработка е χ 2 тест, посредством който е Необходимо е да се обърне внимание на
обстоятелството, че при отхвърляне на нулевата
извършвана проверка на поставена нулева
хипотеза се приема алтернативната хипотеза за
хипотеза. Изборът на този статистически метод
се обуславя от факта, че въпросниците, които се връзка между двете основания на
класификациите, която се проявява поне за една
използват са независими и променливите, които
клетка от кростаблицата. Но остава неизвестно
се изследват са неметрични или категорийни.
това по отношение на кои именно клетки от
При статистическа обработка непълни
кростаблицата връзката се проявява [10]. С
анкети (непопълнени до края на анкетата) са
други думи, възниква въпросът за
изключени. Възвръщаемостта на получените
множествените сравнения, поради което е
отговори е представена на Фиг. 3.
направена допълнителна проверка - анализ по
100% двойки или на кростаблици от вида 2х2.
80% участвали
3.1. Териториален признак – възрастова
60% 56,8% група „Дете”
40% 13,6%
64,6%
непълни
Териториалният признак е изграден в
8,6% анкети зависимост от вида населено място, където учат
20% 29,6%
26,8% децата, а не в зависимост от мястото на живеене.
0% Това решение произтича от факта, че при
не
Дете участвали
разработване на представителната извадка
Тийнейджър децата са избирани в зависимост от училището,
в което учат. Териториалният признак е
Фиг. 3. Разпределение на участието в определен на база четири типа населено място:
изследването на попадналите в извадката столица, областен град, град и село.
ученици Най-високо ниво на значимост р=0,000
Получените стойности за сила на
Един от анализите, приложен при зависимостите са степенувани съгласно
обработка на резултатите, е Анализ на връзките Cramers’s V Value.
и зависимостите, имащ за цел да се открие какви Сила на зависимост= 0,2 ÷ 0,3
зависимости съществуват между изследваните Децата, учащи в града предпочитат като
аспекти на хранително поведение за всяка любима храна „Домашно приготвена храна”,
възрастова група и избрани признаци. За целите повече от тези, учащи в столицата. Децата в
на настоящия доклад от интерес представляват столицата предпочитат като любима храна
откритите връзки с териториалния признак. „Плодове” повече от тези в града.
Именно този фактор е избран за обект на
настоящия доклад, тъй като в проведеното
Сила на зависимост= 0,1 ÷ 0, 2
цялостно изследване това е факторът, влияещ Установено е, че децата, учащи в града
най-силно върху хранителното поведение на обичат да ядат хамбургер, повече от тези в
възрастова група „Дете”. столицата.
Представянето на получените резултати Децата в столицата предпочитат да ядат
е на база класифициране по ниво на значимост в голямото междучасие „Плод” повече от тези в
( p < 0,05 ) и сила на зависимостите (Cramers’s V града. Децата в града предпочитат да ядат в
голямото междучасие повече „Солени тестени
Value). Нулевата хипотеза ( H 0 ), която се задава изделия” от тези в столицата.
и подлежи на статистическа проверка е Високо ниво на значимост р<0,05
изградена на предположението за липса на Сила на зависимост= 0,1 ÷ 0, 2
статистически значима връзка между двете
основания на класификациите – териториален
II-351
Децата, учащи в областен град обичат да Таблица 2. Изследвани основания
ядат риба повече от тези, учащи в града. на класификациите –група
Сила на зависимост < 0,1 ”Тийнейджър”
Децата, учащи на село обичат да ядат Населено
зеленчуци повече от тези, учащи в столицата. От място
своя страна децата, учащи на село обичат да
ядат чипс повече от връстниците им в Отговори
столицата.
Предпочитание Да, много Град
3.2. Териториален признак – възрастова към хляб Областен
група „Тийнейджър” Не, изобщо град
Най-високо ниво на значимост р=0,000
= 0,1 ÷ 0,2
Сила на зависимост Сладки тестени и
шоколадови Град
Малки предпочитания за любима храна Любима
реклама на изделия
към „Домашно приготвена храна” имат както Областен
тийнейджърите, учащи в града, така и тези, храна
Нямам град
учащи на село. Последните имат по-големи
предпочитания за любима храна към Областен
„Нездравословна калорична храна” от тези, Предпочитание Да, много град
учащи в града. към чипс
Село
Високо ниво на значимост р<0,05 Не, изобщо
Сила на зависимост < 0,1 Храна, Солени тестени Столица
Тийнейджърите, учащи на село обичат консумирана в изделия
да ядат хамбургер повече от връстниците им в голямо Нездравословна Град
столицата. междучасие калорична храна
Да, много Столица
4. Обсъждане Предпочитание
При прилагане на Анализа на връзките и към зеленчуци Не, изобщо Град
зависимостите и на допълнителната проверка
Избор между Пържени Столица
посредством анализ по двойки, са установени
пържени картофи
съществуващи зависимости, но не и конкретно картофи и
тяхно проявление между представляващите картофена Картофена Село
интерес за изследването основания на салата салата
класификациите. При последващ анализ има
възможност да се изследва за кои клетки се
•Възрастова група „Дете”
проявява съществуващата връзка. На Таблица 1
Налице е статистически значима връзка
и Таблица 2 са представени отговорите, за които
между храната, която детето яде с приятелите си
е извършвана допълнителна проверка чрез
и населеното място, в което учи. Съществува
анализ по двойки.
статистически значима връзка между избора,
Таблица 1. Изследвани основания който детето би направило между вафла и банан
на класификациите- група „Дете” и населеното място, в което учи. Но анализа по
двойки на избраните отговори показва, че
Населено място конкретно за тях не се проявява връзка и би
било некоректно да се правят заключения.
Отговори • Възрастова група „Тийнейджър”
Солени Налице е статистически значима връзка
тестени Град между предпочитанието към хляб, любимата
Храна,
консумирана изделия реклама на храна по телевизията, и населеното
с приятели Сладки място, в което учи тийнейджърът. Съществува
тестени и Областен град статистически значима връзка между
шоколадови предпочитанието към чипс и населеното място,
изделия в което учи детето. Открита е връзка между
храната, консумирана в голямото междучасие и
Вафла Град
Избор между населеното място, в което учи детето. Открита е
вафла и банан връзка между предпочитанието към зеленчуци,
Банан Областен град
изборът, който детето би направило между

II-352
пържени картофи и картофена салата, и patterns in childhood, р. 429. Butterworth-
населеното място, в което учи. Но анализа по Heinemann, an imprint of Elsevier Inc, USA.
двойки на избраните отговори показва, че 3. Михова, Т., Бочева, М. (2013). Техники за
конкретно за тях не се проявява връзка и би привличане на българските потребители. V
било некоректно да се правят заключения. Национална Научна Конференция за
Анализът, извършен при обработка на студенти, докторанти и млади учени,
получените резултати от цялостното изследване, Сборник доклади, стр. 356-360
показва, че при възрастова група „Дете“ 4. Sidin, S.Md., Abdul Rahman, M.K. (2008).
факторът със статистически обосновани връзки Effects of social variables on urban children’s
с най-голяма сила на влияние (0,2-0,3), е consumption attitude and behavior intentions.
териториалният признак. Това е и един от Journal of consumer Marketing, Emerald
факторите, които влияят върху най-много Group Publishing Limited, 25/1, p. 7-15.
аспекти от хранителното поведение на децата. 5. Jansen, P. et al. (2012). Children's eating
За възрастова група „Тийнейджър” е behavior, feeding practices of parents and
установено, че териториалният признак не weight problems in early childhood: results
оказва силно влияние върху хранителното from the population-based Generation R Study,
поведение на тийнейджърите. International Journal of Behavioral Nutrition
and Physical Activity, 9:130.
5. Заключение 6. Owen, S. et al. (1997). Food choice: how to
Представените статистически доказани assess attitudes of pre-adolescent children,
връзки за двете изследвани възрастови групи Nutrition & Food Science, Number 1,
дават информация за хранителните January/February, p. 5–11.
предпочитания на децата и тийнейджърите в 7. Банков, К. (2012). Широко мащабни
зависимост от един важен фактор, влияещ върху оценъчно-диагностични педагогически
потребителското поведение – географският изследвания, стр. 287. Хабилитационен труд
критерии. Получените резултати за за академичната длъжност професор,
хранителните предпочитания на деца и Софийски университет „Св. Климент
тийнейджъри поставят редица въпроси, на които Охридски”, Факултет по математика и
да се търсят отговори не само по отношение на информатика, София.
хранителните предпочитания на обекта на 8. Kellet, M. (2011). Engaging with children and
изследване, но и по отношение на причините за young people, р.23. Centre for children and
изменение силата на влияние на изследвания young people: Background Briefing Series,
фактор върху поведението на децата. Southern Cross University, Lismore, Australia.
Може да се твърди, че поведението на 9. Zlatanova–Pazheva, E et al. (2016). The web-
децата-потребители в съвременния based application as an instrument for
глобализиран се свят е сложен, многопластов и collecting information when conducting online
динамично развиващ се във времето проблем. surveys with children, European Journal of
Динамичността на факторите, влияещи върху Scientific Research, Paris University Press. №
децата потребители при избора им на храни, 1 (13), (January-June). Volume II. p. 728-734.
предопределя нуждата от задълбочаване на 10. Наследов, А. Д. (2004). Математические
изследванията по този проблем, които в методы психологического исследования:
практико-приложен план ще дадат възможност анализ и интерпретация данных, 392 стр.
на фирмите да постигнат по-добри Изд. Речь, Санкт Петербург, Русия.
икономически резултати.
ЛИТЕРАТУРА Контакти
Име на автора: Елена Златанова-Пъжева
1. Шадрина, С.В. (2009). Потребительское Организация: ТУ-София, Филиал
поведение подростоков: основные факторы Пловдив
влияния. Вестник Санкт-Петербургского Адрес:Пловдив, ул. „Цанко Дюстабанов”
Университета, Сер. 8, Вып. 2, стр.53-91. № 25
2. McNeal, J.U. (2007). On becoming a Телефон: +359898 943 956
consumer: Development of consumer behavior E-mail: elyzlatanova@abv.bg
КАЧЕСТВО И УДОВЛЕТВОРЕНОСТ ПРИ

ПРЕНОСИМИ КОМПЮТРИ
КРИСТИАН ЦВЕТКОВ1, ВЛАДИСЛАВ ГЕРАСИМОВ1, ЕЛЕНА КОЛЕВА1,2, ЛИЛЯНА КОЛЕВА1
Резюме: Проведено е изследване за удовлетвореността на потребтелите от

преносимите компютри. За характеризиране на гореупоменатата са използвани
индикатори, като ефективност, използваемост, качество и ефикасност.
Анкетираните са разделени в две групи, до и над 36 годишна възраст.
Ключови думи: удовлетвореност, ефикасност, ефективност, качество,
използваемост, потребители, преносими компютри.
QUALITY AND SATISFACTION FOR

PORTABLE COMPUTERS
KRISTIAN CVETKOV1, VLADISLAV GERASIMOV1, ELENA KOLEVA1,2, LILYANA KOLEVA1
Abstract: A study about complacency of portable computers. For characterization on

complacency were used indicators witch efficiency, usability, effectiveness and quality.
Respondents were split into two groups, up to and over 36 years of age.
Key words: satisfaction, efficiency, usability, effectiveness, quality, user, portable
computers.
са с плоски LCD-TFT монитори, вградени

1. Въведение оптични устройства, мрежова карта,
Преносим компютър (портативен безжична комуникация, сензорни екранни,
компютър, ноутбук, нетбук) или често възможност за включване на външни
наричан лаптоп, е компютър с малки размери устройства, които разширяват
и същата функционалност, като персоналния функционалността на компютъра [1].
компютър. Думата лаптоп идва от lap- скут. В днешно време поради своята
Идеята за мобилните устройства от този тип функционалност те се използват широко по
възниква около 70те години на 20 век. През целия свят от милиарди потребители, което
1973 година компанията IBM демонстрира поражда интерес към спецификата, коя от
проекта SCAMP (Special Computer APL техните характеристики е най- важна за
Machine Portable). След две години задоволяване нуждите на клиента.
компанията предлага IBM 5100, който е Удовлетвореността от мобилното устройство
първия преносим компютър. Първоначално е функция, която зависи от ефективността,
те били с 8 битови процесори, работели с ефикасността, използваемостта и не на
5.25 инчови (13.335 см.) флопидискови последно място от качеството на продукта.
устройства и с малки CRT дисплей. С Ефективността характеризира само
развитието на технологиите търсенето на отношението на постигнатия резултат спрямо
тези устройства става все по- голямо, заради поставената цел, тя не отчита разхода бил то
своите известни предимства, едно от които пари, време и други, а само постигнатата цел.
безспорно е тяхната мобилност. Използваемостта изразява честотата
Съвременните модели преносими устройства на ползване на даден продукт.
II-354
Ефективност
1.При покупка на лаптоп от какво

значение
са параметрите и марката на
процесора му?
2.При покупка на 3.При покупка на лаптоп от
лаптоп от какво какво значение е Ефикасност
значение е с колко RAM големината на HARD- 41.До каква степен за вас
- памет разполага? диска?
качеството на продукта се
4.При покупка на лаптоп от 5.При покупка на лаптоп от определя от това платената
какво значение е с каква какво значение е марката 39.За да сте удовлетворен цена да отговаря на
видеокарта разполага? на дъното? от направената покупка полученото качество?
6.При покупка на лаптоп от 7.При покупка на лаптоп от на устройство? 42.До каква степен за вас
какво значение е какво значение е да има 40.До каква степен за вас качеството на продукта се
издръжливостта на инсталирана операционна качеството на продукта се определя от съотношението
батерията му? система? определя от цената? цена-качество?
24.При покупка на лаптоп 31.До каква степен за вас Удовлетвореност

8.Използвате ли Интернет 16.Използването на
от какво значение за Вас е качеството на продукта се за обучение? Интернет разширява ли
цената му? определя от големината на 9. Използвате ли Интернет кръга от хора, с които
25.При покупка на лаптоп за комуникация? редовно общувате?
от какво значение е марката гаранционното време?
32.До каква степен за вас 10.Използвате ли Интернет 17.Използвам Интернет
му?
26.При покупка на лаптоп качеството на продукта се за търсене на информация? основно по работа.
от какво значение е размера определя от това гаранцията 11.Използвате ли Интернет 18.Използвам Интернет за
на монитора? да е валидна и за чужбина? за работа? получаване на информация
27.При покупка на лаптоп 33.До каква степен за вас 12.Използвате ли Интернет за лични въпроси.
от какво значение е теглото качеството на продукта се за общуване чрез
социалните мрежи? 19.Използвам Интернет за
му? определя от материалите, с
13.Използването на забавление ( филми, книги,
28.При покупка на лаптоп които е изработен?
Интернет пречи на онлайн игри и др.)
от какво значение е да се 34.До каква степен за вас
обичайния ми 20.Използвам често
предлага с аксесоари качеството на продукта се
ритъм на живот. електронната си поща.
(чанта, мишка, слушалки и определя от външния му
14.Използването на 21.Когато съм в Интернет,
др.)? вид?
Интернет редуцира е основно, за да съм в
29.При покупка на лаптоп от 35.До каква степен за вас
времето, прекарано на живо социалните мрежи.
какво значение е да има качеството на продукта се
с приятели. 22.Използвам интернет, за
водоустойчива клавиатура? определя от наличието на 15.Изпитвате ли нужда
30.До каква степен за вас допълнителни функции? непрекъснато да имате да пазарувам онлайн дрехи,
качеството на продукта се 36.До каква степен за вас връзка с Интернет? продукти и услуги.
23Поне веднъж в
определя от дългосрочното качеството на продукта се
седмицата поръчвам храна
използване, без повреди? определя от високото ниво
38.До каква степен за вас през Интернет.
на изпълнение на
качеството на продукта се стандартни функции?
определя от спазването на 37.До каква степен за вас Използваемост
международни стандарти качеството на продукта се
(ИСО-стандарт и др.)? определя от страната, в
която е произведен?
Качество
Фиг. 1. Диаграма на Ишикава
Качеството в нашия случай е показател разпространен инструмент за решаване на

определящ изследваната удовлетвореност, управленски задачи, свързани с
който е свързан с характеристиките на управлението. Причинно-следствената
разглежданото устройство, като например диаграма помага не само да елиминира даден
размер, марка, материал на изработка и проблем, но и да се предприемат действия за
други. неговото отстраняване. Причинно-следствена
Ефикасност- показва отношението на диаграма на Ишикава тип „рибена кост“ е
вложените разходи спрямо постигнатия наречена така, заради структурата си, която
резултат. наподобява рибена кост. За целите на
изследването е използвана диаграма на
2. Методика Ишикава, която е разклонена и разделя
Диаграмата на Ишикава е един от причините, които характеризират проблема
класическите примери и най-широко
II-355
на 4 категории: ефективност, ефикасност, 3. Определяне на удовлетвореност-

използваемост и качество [2, 3]. та на потребителите
На Фиг. 1 е представена диаграма на Всеки от отговорите е разпределен в
Ишикава тип рибена кост, на която са петстепенна скала. На всяка оценка се
представени разглежданите показатели по присвоява съответния коефициент, който е в
които ще се съди за удовлетвореността от граница от 0 до 1. В Таблица 1 са
преносимите компютри. Основните представени възможните отговори и
разклонения са: съответните им коефициенти.
- ефективност – свързана е с
хардуерните компоненти изграждащи Таблица 1. Коефициенти на
разглеждания продукт; удовлетвореност
- използваемост- показва, колко често Оценки Да По- Нито Да, По- Не
и по каква причина потребителите използват скоро нито Не скоро
интернет; Да Не
- качество – според ISO то е Коефициент 1 0.75 0.50 0.25 0
съвкупност от свойства и характеристики на
обектите, които определят пригодността им В Таблица 2 са показани отговорите
да удовлетворяват установени и
групирани с помощта на допълнителен
предполагаеми потребности;
критерий, възраст. Потребителите са
- ефикасност – показва колко е
определяща цената, като вложен ресурс за разделени под и над тридесет и шест
удовлетворението на потребителите. годишна възраст с цел по-голяма
На базата на определените чрез информативност. В тях са дадени
диаграмата на Ишикава основни причини за произведенията на съответните коефициенти
удовлетвореността е съставена анкетна карта по броя на анкетираните отговорили със
с 42 въпроса. На диаграмата е показан, като съответния отговор. В последната колона е
подразклонение всеки един от въпросите със дадено нивото на удовлетвореност според
съответния му номер. Анкетираните са хора съответния критерии в проценти. За да се
от всяка възраст. За попълването на анкетата получи тази удовлетвореност са сумирани
анкетираните са използвали петстепенна положителните отговори за даден критерий,
скала на оценяване: Да; По-скоро Да; Нито след което този сбор е разделен на броя на
Да, нито Не; По-скоро Не; Не. При
анкетираните и е умножен по 100 за бъде в
проведеното запитване са събрани 138
валидни анкети, които са обработени за проценти.
целите на статията. Анкетата е проведен
онлайн.
Таблица 2. Отговори от анкетата, групирани по възраст: до 36 г. / над 36 г.

Да По- Нито По- Не Удовлет-
Въпроси скоро Да, скоро вореност
Да нито Не Не (%)
1.При покупка на лаптоп от какво значение са параметрите и
73/33 13/14 3/0 1/0 0/1 95.55/97.91
марката на процесора му?
2.При покупка на лаптоп от какво значение е с колко RAM - памет
68/32 17/14 3/1 2/0 0/1 94.44/95.83
разполага?
3.При покупка на лаптоп от какво значение е големината на
43/23 28/20 14/4 5/0 0/1 78.88/89.58
HARD-диска?
4.При покупка на лаптоп от какво значение е с каква видеокарта
45/23 29/21 13/4 3/0 0/0 82.22/91.66
разполага?
5.При покупка на лаптоп от какво значение е марката на дъното? 28/18 24/20 19/8 15/2 4/0 57.77/79.16
6.При покупка на лаптоп от какво значение е издръжливостта на
39/29 30/10 11/3 7/2 3/4 76.66/81.25
батерията му?
7.При покупка на лаптоп от какво значение е да има инсталирана
11/12 23/17 6/7 13/2 37/10 37.77/60.41
операционна система?
II-356
8.Използвате ли Интернет за обучение? 60/19 21/16 5/10 4/2 0/1 90/72.92

9.Използвате ли Интернет за комуникация? 85/47 4/0 0/0 1/0 0/1 98.88/97.92
10.Използвате ли Интернет за търсене на информация? 90/46 0/1 0/0 0/0 0/1 100/97.92
11.Използвате ли Интернет за работа? 68/31 8/7 5/2 4/3 5/5 84.44/79.16
12.Използвате ли Интернет за общуване чрез социалните мрежи? 76/38 6/4 3/2 3/1 2/3 91.11/87.5
13.Използването на Интернет пречи на обичайния ми ритъм на
7/4 10/2 20/13 25/6 28/23 18.88/12.5
живот.
14.Използването на Интернет редуцира времето, прекарано на
20/5 22/7 11/7 16/12 21/17 46.66/25
живо с приятели.
15.Изпитвате ли нужда непрекъснато да имате връзка с Интернет? 16/15 20/9 14/8 16/7 24/9 40/50
16.Използването на Интернет разширява ли кръга от хора, с които
28/20 20/10 12/7 19/5 11/6 53.33/62.5
редовно общувате?
17.Използвам Интернет основно по работа. 12/7 18/8 24/19 23/5 13/9 33.33/31.25
18.Използвам Интернет за получаване на информация за лични
43/22 25/11 13/9 5/3 4/3 75.55/68.75
въпроси.
19.Използвам Интернет за забавление ( филми, книги, онлайн игри
70/25 13/11 3/4 3/4 1/4 92.2/75
и др.)
20.Използвам често електронната си поща. 54/32 14/7 7/4 4/4 1/1 75.55/81.25
21.Когато съм в Интернет, е основно, за да съм в социалните
12/13 18/1 18/11 23/6 19/7 33.33/50
мрежи ( Facebook, Twitter и др.).
22.Използвам интернет, за да пазарувам онлайн дрехи, продукти и
24/11 14/9 10/7 20/11 22/10 42.22/41.66
услуги.
23.Поне веднъж в седмицата поръчвам храна през Интернет. 5/3 2/1 4/3 12/7 67/34 7.77/8.33
24.При покупка на лаптоп от какво значение за Вас е цената му? 33/11 25/20 25/15 4/0 3/2 64.44/64.58
25.При покупка на лаптоп от какво значение е марката му? 27/15 23/19 29/9 4/2 7/3 55.55/70.83
26.При покупка на лаптоп от какво значение е размера на
33/16 29/18 19/12 6/2 3/0 68.88/70.83
монитора?
27.При покупка на лаптоп от какво значение е теглото му? 10/8 20/13 22/10 22/10 16/7 33.33/43.75
28.При покупка на лаптоп от какво значение е да се предлага с
7/5 20/7 10/10 19/14 34/12 30/25
аксесоари (чанта, мишка, слушалки и др.)?
29.При покупка на лаптоп от какво значение е да има
4/3 18/17 16/3 27/13 25/12 24.44/41.66
водоустойчива клавиатура?
63/32 21/15 5/0 0/0 1/1 93.33/97.92
дългосрочното използване, без повреди?
20/20 34/17 21/8 7/2 8/1 60/77.08
големината на гаранционното време?
13/10 15/13 17/11 20/5 25/9 31.11/47.92
това гаранцията да е валидна и за чужбина?
38/17 21/18 17/10 10/1 4/2 65.55/72.92
материалите, с които е изработен?
8/8 25/10 22/19 23/4 12/7 36.66/37.5
външния му вид?
19/10 34/19 22/12 9/4 6/3 58.88/60.42
наличието на допълнителни функции?
48/22 27/18 11/6 4/0 0/2 83.33/83.33
високото ниво на изпълнение на стандартни функции?
15/14 24/14 18/12 14/3 19/5 43.33/58.33
страната, в която е произведен?
27/22 33/15 14/6 8/2 8/3 66.66/78.08
спазването на международни стандарти (ИСО-стандарт и др.)?
39.За да сте удовлетворен от направената покупка на устройство? 12/10 26/12 37/17 15/7 0/2 42.22/45.83
16/16 58/16 21/14 10/0 5/2 60/66.66
цената?
48/25 26/16 12/5 2/0 2/2 82.22/85.41
това платената цена да отговаря на полученото качество?
39/20 32/18 11/8 5/1 3/1 78.88/79.16
съотношението цена-качество?
Следващата стъпка е определянето на За всеки от посочените критерии уi се

важността на параметрите, което става по изчислява сума от техните оценки:
метода на Стоянов за определяне на тегловни
коефициенти при многокритериално вземане
на решения, чрез ранжиране на целевите
параметри [4]. и сумата на теглата за всеки целеви
параметър .
II-357
Таблица 3. Тегловни коефициенти, групирани по възраст: до 36 г. / над 36 г.
Vj i Vj i
1 84.5/43.5 1.010/1.015 0.1386/0.1406 22 44.5/24 1.033/1.033 0.0631/0.0630
2 82.75/43 1.013/1.017 0.1391/0.1408 23 11.5/7 1.058/1.056 0.0646/0.0645
3 72.25/40 1.032/1.027 0.1417/0.1423 24 65.25/33.5 1.019/1.021 0.0661/0.0664
4 74/40.75 1.029/1.025 0.1413/0.1419 25 59.75/34.25 1.024/1.020 0.0664/0.0664
5 59.25/37.5 1.056/1.036 0.1450/0.1435 26 65.75/36 1.019/1.017 0.0660/0.0662
6 68.75/38.5 1.039/1.032 0.1426/0.1430 27 41.5/25.25 1.038/1.033 0.0673/0.0672
7 34.5/28.75 1.102/1.066 0.1513/0.1477 28 31.75/18.75 1.046/1.043 0.0678/0.0679
8 79.25/36.5 1.007/1.015 0.0616/0.0620 29 32.25/20.5 1.045/1.040 0.0678/0.0677
9 88.25/47 1.001/1.001 0.0611/0.0611 30 81.25/43.25 1.006/1.007 0.0652/0.0655
10 90/46.75 1/1.001 0.0611/0.0611 31 57.75/37.25 1.025/1.015 0.0665/0.0661
11 77.5/38 1.009/1.013 0.0616/0.0619 32 37.75/26.5 1.041/1.031 0.0673/0.0671
12 82.75/42.25 1.005/1.007 0.0614/0.0615 33 64.75/35.75 1.020/1.018 0.0661/0.0662
13 30.75/13.5 1.043/1.047 0.0637/0.0639 34 43.5/26 1.036/1.032 0.0672/0.0672
14 46/16.75 1.032/1.043 0.0631/0.0637 35 57.75/31.25 1.025/1.024 0.0665/0.0667
15 42/27.5 1.035/1.028 0.0632/0.0627 36 74.75/38.5 1.012/1.014 0.0656/0.0660
16 53.75/32.25 1.026/1.021 0.0627/0.0623 37 45.5/31.25 1.035/1.024 0.0671/0.0667
17 43.25/23.75 1.034/1.033 0.0632/0.0631 38 60.75/36.75 1.023/1.016 0.0663/0.0661
18 69.5/35.5 1.015/1.017 0.0620/0.0621 39 53.75/29.25 1.134/1.130 0.2584/0.2597
19 82/36.25 1.005/1.016 0.0614/0.0620 40 57.5/35 1.120/1.090 0.2552/0.2505
20 69/40.25 1.015/1.010 0.0620/0.0617 41 74/39.5 1.059/1.059 0.2413/0.2434
21 40.25/28.25 1.036/1.027 0.0633/0.0627 42 69.75/37.75 1.075/1.071 0.2449/0.2462
Определянето на тегловните но се вижда също и, че той носи най-малка

коефициенти j става по следната формула: удовлетвореност и на двете групи.
където Vj се определя като:
,
където R е броят на анкетираните хора, а m е
броят на оценяваните критерии.
Получените резултати за сумата на
оценките, тегловните коефициенти и
коефициента Vj , които също са разделени на две
групи от критерия възраст, са дадени в Таблица
3.
4. Анализ на резултатите
Фиг. 2. Ефективност на преносимите
Фиг. 2-5 показват зависимостта между
компютри: „‟ – възраст до 36 г.; „o‟ – възраст
удовлетвореноста във проценти и тегловните
над 36 г.
коефициенти на всеки от основните четири
критерии. Всяка от тях е разделена на четири
квадранта, в които са групирани въпросите.
Квадраните са означение с латинските букви: A,
B, C, D:
- А – висока удовлетвореност и тегловни
коефициенти;
- B – висока удовлетвореност и ниски
- C – ниска удовлетвореност и ниски
- D – ниска удовлетвореност и високи
тегловни коефициенти.
Според ефективноста, за потребителите
с най-голямо значение е продуктът да има Фиг. 3. Използваемост на преносимите
инсталирана операционна система, въпрос № 7, компютри: „‟ – възраст до 36 г.; „o‟ – възраст
над 36 г.
II-358
По въпроса за използваемостта с най- 5. Заключение

голямо „тегло“, но с най-малка удовлетвореност Няма съществена разлика между
е въпрос № 23, поне веднъж поръчвам храна параметрите носещи удовлетворениет на
през интернет, а с най-голяма удовлетвореност потребителите под и над 36 годишна възраст. С
са въпроси № 9 и 10 използвате ли интернет за изключение на използваемостта, може да се
търсене на информация и за комуникация. обобщи, че потребителите над разделящата
възраст са по-удовлетворени от изследваните
параметри.
Единствено според основния критерий
качество тези под 36 години са по-
удовлетворени отколкото тези над тази възраст.
1. Тончев Милен, (2015)

http://www.digital.bg/koga-e-sazdaden-
parviqt-laptop-article435090.html
Фиг. 4. Качество на преносимите компютри: 2. Todorova Maria, Koleva Lilyana, Koleva
„‟ – възраст до 36 г.; „o‟ – възраст над 36 г. Elena, Iliev Veselin, (2015). Organization and
optimization of e-learning at UCTM,
При качеството се наблюдава разлика Proceedings international scientific conference
между тегловните коефициенти на различните “Post - crisis management in business”
възрастови групи, като за тези под 36 години с November 27, 2015 Sofia, Bulgaria; Volume
най-голям тегловен коефициент е въпрос № 29, ІІ; 438-446.
важно ли е при покупка на лаптоп, той да има 3. Koleva Lilyana, Koleva Elena, Dzharov
водоустойчива клаватура, а за потребителите Volodya, Todorova Maria, (2015).
над 36 въпрос № 28, устройството да се предлага Investigation of the satisfaction from the
с допълнителни аксесоари. И двата въпроса education in UCTM - SOFIA, Proceedings
носят най-малка удовлетвореност на international scientific conference “Post -
съответните потребители. Най-важно според crisis management in business” November 27,
този критерий е дългосрочното използване без 2015 Sofia, Bulgaria; Volume ІІ; 399-406.
повреди, въпрос № 30. 4. Стоянов С. (2005), Интелигентни методи
за оптимизация. Въведение в
оптимизацията, Химико-технологичен и
металургичен унверситет – София.
Контакти:
Кристиан Цветков1, Владислав
Герасимов1, Елена Колева1,2, Лиляна
Колева1
1
Химико-Технологичен и Металургичен
Университет – София 1756, бул.
“Климент Охридски” № 8
2
Институт по Електроника - БАН -
София 1784, бул. “Цариградско шоссе”
Фиг. 5. Ефикасност на преносимите №72
компютри: триъгълник – възраст до 36 г.; кръг E-mails: kpcvd@mail.bg,
– възраст над 36 г. vladivd123@abv.bg, eligeorg@abv.bg,
sura@abv.bg
Според критерия ефикасност въпрос №
39 е с най-голямо тегло и най-малка
удовлетвореност, а въпрос № 41 е негова
противоположност.
АСПЕКТИ НА ОПТИМАЛНО ПРОЕКТИРАНЕ

НА АВТОМАТИЧНИ ЛИНИИ
СТОЯНКА КАСАБАДЖАКОВА, ЕКАТЕРИНА ПЕРПЕРИЕВА
Резюме: В предприятията с дискретно производство като правило се произвеждат

изделия със сложна йерархична структура. Тези изделия обикновено са
многовариантни на всички нива на производството. В настоящата статия са
разгледани възможностите и критериите за оптимизация на автоматичните
линии. За оптимален се счита този проект, който води да създаване на обект, най-
добър от реално осъществените за дадени конкретни условия.
Ключови думи: автоматични линии, функционални схеми, целева функция,

управляващи променливи
PRINCIPLES OF AUTOMATIC LINES OPTIMAL

DESIGN
KASABADZAKOVA S., PERPERIEVA EK.
Abstract: The products with a complex hierarchical structure are generally produced in
discrete manufacturing enterprises. These products are usually multivariate at all levels of
the production. The opportunities and criteria for optimization of automated lines are
examined in this paper. This project, which led to the creation of an object, the best of the
actual set of specific conditions, is considered to be the optimal one.
Key words: Automatic lines, functional schemes, target function, control variables
1. Въведение Бизнес процесът е организирана серия от

Дискретното производство има редица действия, изпълнявани от системи и хора за
специфични особености: огромно разнообразие постигане на даден резултат или цел. За
на произвежданите детайли и изделия; голямо ефективността на система (структурата) се
многообразие на прилаганите технологични изисква пълно равновесие между връзките,
процеси, производствени машини и формиращи нивата на автоматизираното
съоръжения. В посоченото производство управление на индустриалната фирма. Създават
съществуват големи изискванията към точността се условия да се интегрират различните видове
на обработка на детайлите. За всеки конкретен бизнес информационни потоци в йерархичните
случай са възможни огромен брой технически структури в дадена система, която вече е
решения и това затруднява избора на най- предмет на управление.
подходящия вариант за даден обект в Обхватът на автоматизация на бизнес
съответствие с поставените изисквания. процесите се ограничава в зоната на описанието
на процесите, което идентифицира различните
II-360
функции на процесите, процедурни правила и при внедряването на всякакви бизнес ИТ
съответстваща информация, необходима за системи. В същото време правилното му
управление на процеса. изпълнение е ключови както за успеха на
внедряването на системата - анализът на
2. Бизнес процеси в индустриалната процесите е особено важен за това, както и за
фирма по-добрата работа след внедряването на
Бизнес-процесът обединява на практика системата - промяната на начина на работа е
потока от дейности или функции, служителите и факторът, даващ възможност организацията да
оборудването (ресурсите), нужната за вземане се възползва от новата система.
на решения информация (знания), както и В днешния бизнес нуждата от оптизация е постоян
правилата за изпълнение на тези дейности и налага както поради постоянната промяна на бизнес среда
функции. Една или няколко свързани помежду технически средства, водещи до по-висока производителнос
си процедури или операции (функции), които на индустриалната фирма.
съвместно реализират определена бизнес задача
Изборът на стратегия и средства за оптимизаци
и реализирането им води до конкретни
отношение на ИТ решенията. Факт е обаче, че ИТ системите
резултати.
дейностите, правейки процесите по
Рационалната им организация е решаващ
фактор за превръщането им в конкурентно
предимство за индустриалната фирма. За да Вначалото на XXI век основният
бъде постигната такава обаче, е необходимо стремеж на всяка организация е да направи
повече с по-малко ресурси. Постигането на тази
бизнес процесите да бъдат детайлно
цел изисква процесите и работните потоци да
анализирани и постоянно оптимизирани. бъдат проектирани, така че да се осигурява
Бизнес процесът е организирана серия от максимална оперативна ефективност, а това е
действия, изпълнявани от автоматизирани същността на концепцията управление на
системи и хора за постигане на даден резултат бизнес процесите Отделните бизнес дейности се
или цел. Бизнес процесите по същество са привеждат в състояние да работят по-бързо, по-
типични за човека, но в днешния автоматизиран качествено и по-евтино, което води до общо
и свързан свят, използването на технологията за подобряване на технико-икономическото състояние
автоматизиране на процесите е неизбежно в на индустриалното предприятие.
индустриалната фирма.
Бизнес процесите са основния двигател 3. Управление на бизнес процесите
на работа в една организация. Добрите бизнес Добре дефинираните и безупречно
процеси са гаранция, че нещата се случват изпълнени процеси в дадена компания са
бързо, качествено и надеждно. Процесите в една предпоставка за по-конкурентоспособни
организация не е задължително да са продукти и услуги, намалени разходи, по-добро
формализирани. Колкото по-голяма е дейността обслужване на клиентите и бързи реакции при
обаче и колкото по-сложна е тя, толкова по- промяна на пазарните условия. Затова е
висока е нуждата от ясни и ефективни процеси необходимо вътрешните бизнес процеси да
на работа. бъдат системно проверявани за проблеми и
Координацията на сложни групи от грешки, възникнали в резултат на ръчна
дейности, извършвани от независими един от обработка.
друг хора или машини, в никакъв случай не е Съвременните средства за управление на
просто. Бизнес процесите по същество са бизнес процеси дават възможност на фирмата
типични за човека, но в днешния автоматизиран изцяло да оркестрира своите процеси от гледна
и свързан свят, използването на технологията за точка на бизнес анализа и техническия дизайн.
автоматизиране на процесите е неизбежно. Обърнете внимание, че тук не става въпрос само
Технологията помага за автоматизирането на за рутиране на документи във вътрешната ИТ
рутинни операции и динамично човешко инфраструктура. Решенията трябва да
взаимодействие. рефлектират не само върху техническата логика,
Търсенето на оптимален процес е но и върху целия работен поток.
основна цел за всяка една организация. За Правилното решение трябва да
постигането му могат да се използват различни позволява: извличане на максимална
средства- нови бизнес ИТ системи насочени към ефективност от процесите, като се дава
подобряване на дейността. възможност на крайните потребители директно
Първоначалният анализ и впоследствие да проектират, управляват, наблюдават и
промяната и подобряването на вътрешните анализират бизнес процесите.
процеси е почти винаги пренебрегван фактор
II-361
4. Ролята информационните технологии Обикновено се прибягва до построяване на
Техническото и обществено развитие е симулационни модели, даващи възможности за
изградено на базата на комуникационните и визуализация на бизнес процеса. Архитектурата
информационните технологии. От друга страна, на производствения процес се проектира на база
стимулираните от глобалните приоритети разработен моделиращ алгоритъм и програма.
технологични решения напълно разкриват своя Използват се универсални езици за
потенциал, само в рамките на световното програмиране.
икономическо пространство. Процесът на оптимизация се включва на
Времето, в което живеем е период на всеки етап на проектирането Резултатът от
големи икономически и социални промени. симулационното моделиране е симулационният
Създадоха се общи световни мрежи и системи модел, служещ за изследване поведението на
от отделни части и елементи, при които индустриалното предприятие при различни
глобалните интереси все повече преобладават ситуации на работа и различни външни
над националните. смущения.
В резултат на това се формират точно
определени международни приоритети, при 6. Съвременни технологии за бизнес
които света се вижда като единно икономическо моделиране
пространство, или единна сфера за Компаниите все повече се нуждаят от
осъществяване на технически и икономически анализ на архитектурата на организацията,
растеж. Това се обяснява с „прехода от преди да внедрят каквито и било промени. За
индустриално общество към общество на множество средни и големи компании
знанието, което още повече усложнява използването на подобни средства е стандарт, а
преплитащите се информационни бизнес самите инструменти са се превърнали в част от
процеси в сложната йерархична индустриална информационните им ресурси. Реализацията на
структура”[1]. проекти, свързани с описание структурата на
В условията на съвременната глобална организацията, реинженеринг на бизнес
икономика много компании са изправени пред процесите и др. Би било изключително трудно
предизвикателството непрекъснато да прилагат без използването на средства за бизнес за бизнес
нови стратегии, стремейки се да бъдат първи в моделиране. По данни на Gartner Group през
пазарната надпревара за предлагане на нови 2015 г. Обемът на продажбите на средства за
продукти и услуги. Добре известен факт е, че бизнес моделиране е бил около 500 млн. долара
първите на пазара получават по-висок процент с очакван годишен ръст около 15%.
от общия пазарен дял. В бъдеще най-голямо
значение ще има не толкова увеличаването на 7.Автоматизация на управлението на
броя и сложността на бизнес процесите, а бизнес процесите
създаването и използването на знания за Автоматизираното управление на бизнес
подреждането им в система, дефинираща процесите може да се приеме като комплексно
възможността за ефективното им системно решение само от гледна точка на
автоматизирано управление. системния подход, защото отделни бизнес
процеси са част или елементи от други
5. Принципи на оптималното подсистеми, или от общата автоматизирана
проектиране на бизнес процеси система на индустриалното предприятие.
За оптимален се счита този проект, който Обхватът на автоматизация на бизнес
води до създаването на обект, най-добър от процесите се ограничава в зоната на описанието
реално осъществимите при реални на процесите, което идентифицира различните
производствени условия. Процесът на функции на процесите, процедурни правила и
оптимизация на всеки етап от проектирането съответстваща информация, необходима за
включва следните основни етапа: управление на процеса. Автоматизирания
 Формиране на съвкупност от процес пък е съвкупност от една или повече
технически възможности и свързани помежду си процедури, които
целесъобразни проектни варианти, съвместно и автоматизирано реализират
които се различават по съчетаването определена задача на бизнеса [2].
на управляващите променливи – В автоматизираната система бизнес
структурни, технологични, процесите се състоят от множество
конструктивни и др. функционално зависими елементи, които са
 Избор на оптимален вариант по логически свързани така, че да осигурят
зададена целева функция. автоматизираното управление на целия бизнес

II-362
процес. Тогава функционалната зависимост на 4. Изводи
автоматизираното управление на бизнес Изяснени са теоретичните въпроси,
процесите приема вида: свързани с моделирането на работни потоци за
автоматизирани производствени системи.
FS
  Br  Ba  (1) Формулирани са етапите за разработване
на симулационни модели и са съставени
където: симулационни модели за изследване работата на
FS - автоматизирано управление на бизнес конкретни автоматични комплекси за
процесите; дискретното производство.
Br - ръчно управлявани бизнес процеси;
Ba - автоматизирани бизнес процеси. ЛИТЕРАТУРА
1.Лесидренска С., С.Демирова,
Автоматизираните бизнес процеси
М.Маринова, Процесният подход – условие за
включват цялата автоматизирана обработка на
автоматизирано управление на бизнес процесите
информацията в системата за управление на
в индустриалното предприятие, ТУ-Варна, 2011.
индустриалното предприятие. Нивото на
2.Чакърски, Д, Т. Вакарелска. Инженерни
автоматизирано управление, или степента на
изследвания, ИК на ТУ - София, 2008.
автоматизация на бизнес процесите ще
определим по следния начин:
Ba  Br Контакти – стил „Contacts”

K (2) Име на автора/авторите
Ba Стоянка Касабаджакова, доцент
където: Организация - Висше училише по
K - степен на автоматизация на бизнес сигурност и икономика, Пловдив
процесите. Адрес: бул. Кукленско шосе 13
Устойчивите тенденции към глобализация Телефон:0889219886
на икономиката създават добри предпоставки за E-mail: tania_sm@mail.bg
ускорено въвеждане на решения за
автоматизирана обработка на информацията и Екатерина Перпериева, асистент
разширяване на възможностите за обхващане на Варненски свободен университет,
по-голям брой бизнес процеси за филиал Смолян,
автоматизирано управление. E-mail: ksmolian@abv.bg
ПРОБЛЕМИ И РЕШЕНИЯ ПРЕД

УПРАВЛЕНИЕТО НА ОТПАДЪЦИТЕ В
ГЛОБАЛЕН И НАЦИОНАЛЕН АСПЕКТ
МАРИЯ ПЕТКОВА-КОЗОВСКА
Резюме: Настоящият доклад има за цел да даде представа за проблемите и

решенията пред управлението на отпадъците в глобален и национален аспект. На
първо място, са разгледани проблемите, свързани с управлението на отпадъците в
световен мащаб. Посочено е, че обемът на отпадъците в голяма степен се определя
от два фактора – брой на населението и структура на потреблението, но
увеличеното количество отпадъци не би имало толкова катастрофални
последствия, ако не се наблюдаваше и промяна в състава им. На следващо място, се
очертават политиката и инициативите в областта на управлението на
отпадъците в България. Системата за разделно събиране и рециклиране на
отпадъците в страната все още не функционира достатъчно добре, като
превръщането им в полезна суровина е слабо разпространена практика и все повече
изисква иновативни управленчески и технологични решения. В края на доклада е
изтъкнато, че България разполага с много добри условия за въвеждане на
иновативни методи за преработка на отпадъци, изразяващи се преди всичко в добра
инфраструктура и подходяща законова рамка за управление на отпадъците.
Ключови думи: отпадъци, управление на отпадъци, рециклиране, околна среда
PROBLEMS AND SOLUTIONS OF WASTE

MANAGEMENT IN GLOBAL AND NATIONAL
ASPECT
MARIYA PETKOVA-KOZOVSKA
Abstract: The current report aims to provide insight into the problems and solutions of
waste management in global and national aspect. First, the paper discusses the problems
related to waste management worldwide. It is stated that the volume of waste is largely
determined by two factors – the size of the population and the structure of consumption, but
the increased volume of waste will not have such disastrous effects if it was not for the
change in waste composition. Next, the policies and initiatives in the sphere of waste
management in Bulgaria are outlined. The separate collection and recycling of MSW,
however, is still not very popular in the country. Their transformation into useful raw
materials is not a widespread practice and requires an increasing number of innovative
management and technological solutions. At the end of the report, it is highlighted that
Bulgaria offers very good conditions for introduction of innovative methods for waste
recycling like good infrastructure and favorable legal framework for waste management.
Key words: waste, waste management, recycling, environment
1. Въведение тя често не е призната като такава. Разходите за

Управлението на отпадъците е водеща общественото здраве и свързани с щетите върху
комунална услуга и критичен елемент на околната среда от безконтролното изхвърляне и
инфраструктурата в подкрепа на обществото, но открито изгаряне са далеч по-големи от
разходите на разумното управление на
II-364
отпадъците. Постигнат е значителен напредък съсредоточено в девет страни: Индия, Нигерия,
през последните десетилетия, но от 2 до 3 млрд. Пакистан, Демократична република Конго,
души, често в най-слабо развитите страни, все Етиопия, Обединена република Танзания, САЩ,
още нямат достъп до редовно събиране на Индонезия и Уганда, изброени в зависимост от
отпадъци и/или услуги за контролирано размера на техния принос към цялостния растеж
обезвреждане на битови твърди отпадъци. Това [1].
е глобален приоритет за общественото здраве, Около 80 % от това нарастващо
който изисква координиран подход, а не е само население ще живее в градове, повечето от
проблем на национално или местно ниво. които все още предстои да бъдат построени. От
Развитите страни също са постигнали напредък тези прогнозирани почти 9 млрд. души, 3 млрд.
в повишаване на нивата на рециклиране и ще принадлежат към средната класа, с
стабилизиране на растежа на отпадъци, но все достатъчно разполагаем доход за закупуване на
още има много да се направи в целия свят при потребителски стоки, които другите вече
осъществяването на прехода от управление на употребяват по света, допълнително източвайки
отпадъците в линейна икономика към вече обтегнатите природните ресурси на
интегрирано и устойчиво използване на планетата. Преминаването в кръгов модел на
ресурсите и управление на отпадъците в кръгова развитие, който работи за намаляване на
икономика. отпадъците, преди те да се образуват и който ги
третира като ресурс, когато те вече са станали
2. Проблеми, свързани с управлението на отпадъци, е от съществена важност, като
отпадъците цялостното и интегрирано устойчиво
Глобалният проблем със събирането и управление на отпадъците ще бъде от решаващо
извозването на отпадъците в световен мащаб значение [1].
спешно изисква иновативни решения. С В допълнение към пренаселеността се
въпросите, свързани с управлението на очаква БВП да се увеличи значително, особено в
отпадъците, са ангажирани най-големите развиващите се страни, като до 2050 г.
световни изследователски центрове и известни световното производство вероятно ще е четири
учени. През последните години става все по- пъти по-голямо от нивата си в началото на века.
очевидно, че съществуващата система за Значителният прираст на населението и
управление на отпадъци не може да се справи глобалният растеж на БВП неминуемо ще
със задачата. Разразилата се през 2008 г. – 2009 доведат до увеличаване на обема на отпадъците.
г. финансова и икономическа криза само Според наблюдения, при увеличаването на
влошава ситуацията – преработката на отпадъци националния доход с 1 % количеството на
винаги е била изключително скъпа, и твърдите отпадъци нараства с 0,69 % [2].
влошаващата се икономическа обстановка води Въпреки това са налице и положителни
до затваряне на много преработващи перспективи – известно е, че колкото по-голям е
предприятия, най-вече в развиващите се БВП на една държава, толкова по-сложна и
държави. В същото време обемът на отпадъците ефективна е нейната система за управление на
стремглаво се увеличава, в резултат на което отпадъците. Въз основа на тази закономерност
огромните депа за отпадъци в много страни може да се твърди, че нарастването на БВП със
неминуемо водят до екологична катастрофа. сигурност ще принуди правителствата на новите
Обемът на отпадъците в голяма степен членки на групата на развитите държави да
се определя от два фактора – брой на предприемат мерки за подобряване на системата
населението и структура на потреблението. По за събиране и управление на отпадъците.
данни на ООН населението в света, възлизащо Настоящите икономически реалности обаче са
на 7,3 млрд. души, се очаква да достигне 8,5 такива, че за развитите страни е по-изгодно да
млрд. до 2030 г., 9,7 млрд. през 2050 г. и 11,2 изнасят отпадъците си към държавите от третия
млрд. през 2100 г. Повече от прогнозираното свят, отколкото да хвърлят огромни средства за
увеличение в глобален мащаб може да се отдаде тяхната преработка. Следователно са налице
на кратък списък от държави с висока сериозни опасения, че усъвършенстването на
раждаемост, най-вече в Африка, или на страни с технологиите без подкрепящи мерки няма да
вече многобройно население. С най-висок доведе до разрешаването на проблема. Възниква
процент на нарастване на населението, Африка и въпросът доколко новите технологии ще са
се очаква да осигури повече от половината от достъпни за развиващите се страни.
световния ръст между 2015 г. и 2050 г. През Увеличеното количество отпадъци не би
този период половината от увеличението на имало толкова катастрофални последствия, ако
населението в света се очаква да бъде не се наблюдаваше и промяна в състава им.
II-365
Преди началото на ХХ в. по-голямата част от рециклирането на всички произведени на тяхна
изхвърляните от човечеството отпадъци територия отпадъци и започват да извозват
естествено и лесно се разграждат от бактерии в голяма част от тях за преработка в развиващите
почвата – храните в рамките на една година, се страни. Най-голям е делът на изнасяните
хартията и дървото за няколко години, а металът електронни отпадъци, тъй като те съдържат
за няколко десетилетия. С появата на нефта и тежки метали и вредни вещества и
синтетичните материали обаче ситуацията се преработването им в съответствие с
променя коренно и към настоящия момент ограничителните норми за безопасност на
повече от 80 % от отпадъците не подлежат на националното законодателство се превръща в
естествено разлагане поради факта, че не много скъп процес. За развитите страни е по-
съществуват бактерии, способни да разградят изгодно да извозват тези вид отпадъци към
пластмасите. Проблемът се усложнява от страни от третия свят, където те се сортират и
огромните емисии на токсични вещества, обработват ръчно с помощта на най-
изхвърляни при производството на този вид примитивни методи. Въпреки че Конвенцията от
продукти, както и от съдържанието им в редица Базел забранява износ на токсични отпадъци
предмети от бита – живак в луминесцентните към развиващите се страни, нейните разпоредби
лампи и термометрите, тежки метали в често са пренебрегвани, като отпадъците се
батериите и акумулаторите и др. извозват под формата на хуманитарна помощ
Според различни оценки образуването или друг „благороден“ претекст. Като се има
на твърди битови отпадъци се оценява на около предвид, че опасните отпадъци представляват
2 млрд. тона годишно в световен мащаб. потенциална заплаха за здравето на хората и за
Въпреки вариацията на нивата на образуване в околната среда, един от водещите принципи на
рамките на и между отделните страни, Базелската конвенция е, че за да се сведе до
генерирането на твърди битови отпадъци на минимум тази заплаха, опасните отпадъци
човек от населението се намира в силна следва да бъдат третирани възможно най-близо
корелационна зависимост с националния доход. до мястото, където се произвеждат. Въз основа
В страните с високи доходи нивата на на концепцията за предварително информирано
образуване на битови отпадъци сега започват да съгласие трансграничното движение на опасни
се стабилизират, или дори показват лек спад, или други отпадъци може да се осъществява
което може да показва началото на „отделянето“ само след предварително писмено уведомление
на ръста на отпадъците от икономическия от страна на компетентните органи на държавата
растеж. Въпреки това, тъй като икономиките износител към същите на държавата вносител и
продължават да растат бързо в страните с ниски на транзитните държави (ако има такива), като
и средни доходи, може да се очаква им се предоставя подробна информация за
генерирането на отпадъци на човек от предвиденото движение. То може само да
населението да се увеличава постоянно. продължи, ако и когато всички засегнати страни
Органичните фракции представляват по-голям са дали своето писмено съгласие. В някои
процент от образуваните твърди битови страни съществуват и окончателни забрани за
отпадъци в страните с ниски доходи (където износ на тези отпадъци [4].
органичните отпадъци са обикновено от 50 до 70 С настъпването на световната финансова
% от всички битови отпадъци), отколкото в и икономическа криза от 2008 г. – 2009 г.
страните с високи доходи (където органичните проблемът с извозваните за развиващите се
са обикновено 20 до 40 %). От своя страна, държави отпадъци се влошава. Развитите
рециклирането може да осигури източник на държави вече не могат да си позволят да заделят
доходи, да спомогне за опазването на оскъдните огромните средства, необходими за
ресурси и да намали количествата на рециклиране на отпадъци на тяхна територия,
отпадъците, изискващи депониране. Нивата на държавите от третия свят стават още по-силно
рециклиране в държавите с високи доходи зависими от значителните приходи в
непрекъснато се увеличават през последните 30 националния си бюджет от сортиране и
години, дължащо се главно на законодателни и преработка на отпадъци, а за местните жители
икономически инструменти. В страните с ниски работата на сметища често е единствена
доходи, неформалният сектор често постига възможност за прехрана. По този начин, докато
нива на рециклиране от 20 % до 30 % за и двете страни имат интерес от „миграцията на
твърдите битови отпадъци [3]. боклуци“, справянето с тази порочна практика е
В края на миналия и началото на невъзможно.
настоящия век развитите държави започват да Посочените по-горе тенденции, свързани
изпитват трудности при обезвреждането и с проблема с отпадъците, ще се проявяват и

II-366
занапред. Нарастването на количеството на югоизточна Европа и в частност в България се
отпадъците не може да бъде спряно поради наблюдава сравнително слаба иновативна
ръста в световното население, като в дейност в тази насока. Депонирането на
допълнение навиците на консуматорското отпадъци все още е „мръсна“ и неизгодна
общество ще продължат да се разпространяват с дейност, както в екологичен, така и в
увеличаване на просперитета в някои от икономически план. То е източник на
развиващите се държави с многобройно значителни емисии от парникови газове (главно
население. Съвременната икономическа система въглероден диоксид, сероводород и метан),
изисква постоянен растеж, за който е нужно да е отделящи се в атмосферата при анаеробното
налице и постоянно нарастващо потребление. разграждане на депонираните отпадъци, като в
Стагнация в потреблението, съответно в световен мащаб, приблизително 20 % от
растежа, би довела до рухването на много годишното количество емитиран метан в
национални икономики, до безработица и глад. атмосферата се дължи именно на процеса
Стига се до затворен порочен кръг, който не би депониране. В същото време в държавата
могъл да се прекъсне без иновации, водещи до продължават да се изразходват значителни
кардинална промяна в потреблението на средства за изграждане на регионални депа за
ресурсите и управлението и преработването на битови отпадъци, поради недостиг на такива,
отпадъците. изчерпване на лимита им или масовото
Що се отнася до промяната в закриване на тези, които не отговарят на
структурата на отпадъците, и тук перспективите екологичните изисквания на ЕС. Това
не са окуражаващи. Опитите да се замени противоречи на стратегията за устойчиво
пластмасата с лесно разграждащи се в околната управление на отпадъците в страната, насочена
среда материали към настоящия момент не дават към нулево депониране, максимално
съществени резултати. Много изкуствени рециклиране, повторна употреба, ограничаване
материали не могат да бъдат заменени, а при на енергийното оползотворяване до
тези, при които това е възможно, разходите за неподлежащите на рециклиране отпадъци и др.
производство на екологично чисти заместители Основните трудности при изпълнението на тези
са 4-5 пъти по-високи. цели са свързани главно с някои недостатъци на
Извозването на отпадъците в системата за разделно събиране, чието правилно
развиващите се страни също ще продължи, поне функциониране е важен фактор за постигането
в рамките на това десетилетие. За да бъде им.
ограничено е необходимо значително да се В България обезвреждането и
повиши стандарта на живот във всички слоеве преработката на твърдите битови отпадъци
на населението на държавите от третия свят, а (ТБО) става чрез депонирането им на
това е изключително трудна и дори определени за целта площадки близо до
непостижима за момента задача. Много от населените места, както и на множество малки
бедните страни не разполагат с природни неконтролируеми сметища. Разделното
ресурси, развита промишленост или качествено събиране на ТБО в страната обаче все още не
образование и няма какво да предложат на функционира достатъчно добре, като
световния пазар, освен евтина работна ръка. превръщането им в полезна суровина е слабо
разпространена практика и все повече изисква
3. Политика и инициативи в областта на иновативни управленчески и технологични
управлението на отпадъците в решения. Във връзка с амбициозните цели на ЕС
България за устойчиво развитие, управлението на
Проблемът за устойчивото управление отпадъците и опазването на околната среда се
на отпадъците е един от основните в превръщат в първостепенни задачи за общините.
съвременния свят. Развитието на технологиите в През 2013 г. около 60 % от отпадъците,
това направление дава широки възможности все постъпващи в сметищата, се състоят от
повече отпадъци да се трансформират в нови употребени опаковки, като голяма част от тях са
ресурси и eнергия. Заедно с това е налице от бързооборотни стоки – хранителни продукти
необходимост от промяна в начина на мислене и и стоки за бита. При това положение пред
културата на потребление, която да подтикне общините възниква проблемът как да се
фирмите да внедряват и използват материали и мотивират производителите на бързооборотни
продукти, незаплашващи екологичната стоки да намалят превантивно отпадъците от
сигурност. За разлика от останалата част на опаковките на техните стоки. Така „битката с
Стария континент, където за тези проблеми вече отпадъците“ и усвояването на съвременни
се намират множество адекватни решения, в опаковъчни технологии се явяват общ,
II-367
комплексен проблем на общините и бизнеса на [5] При добива на енергия от алтернативни
техните територии (без да се смятат търговските източници особено място заема преработката на
вериги и дистрибуторските системи, които са биомаса (селскостопански органични отпадъци)
слабо заинтересовани). За да се справят с при която, използвайки процеса на ферментация
проблема, свързан с ТБО, българските общини се получава биогаз, съдържащ 70 % метан.
се насочват към стимулиране и подпомагане на Теоретичният потенциал за добив на биогаз в
бизнеса на своите територии да обновява и България се оценява въз основа на общото
внедрява опаковъчни системи с олекотени, производство на биомаса. Селскостопанското
рециклируеми, деградируеми и многооборотни производство и свързаните с него индустрии и
опаковки. потребление генерират значително количество
През последните години вниманието на изхвърляна органична материя. Тяхното
държавата и бизнеса в областта на ТБО се използване за производство на биогаз е
пренасочва в унисон със стратегическите цели целесъобразно и препоръчително. В България
на ЕС за управление на отпадъците – стремежът най-голям потенциал за производство на биогаз
е те не само да се обработват ефективно, но и в от първични и вторични селскостопански
самия производствен цикъл на преработка на отпадъци има в Североизточен и Южен
суровините да се получават по-малки централен регион на планиране, поради силно
количества от тях. Всяко производство крие развития селскостопански сектор в тях. По
потенциална опасност от генериране на нови отношение на твърдите битови отпадъци и
отпадъци, затова днес големият въпрос пред отпадъчните води, най-значителни предимства
продуктовите дизайнери е какво ще се прави с при производството на биогаз има в
новото изделие, когато то завърши жизнения си югозападната част на България (най-вече
цикъл и се превърне в отпадък. В сферата на столицата София), където гъстотата на
ТБО тази политика води до повишени населението е висока. Топлоенергийната
изисквания към опаковките. Стремежът е те да промишленост е един от най-крупните
имат минимален обем и размери, да са генератори на твърди отпадъци в България.
направени от рециклируеми и/или Ежегодно се генерират големи количества
биодеградируеми материали, да се използват твърди отпадъци, от които само 25 % (основно
многократно и др. гипс) биват оползотворявани като вторични
България разполага с много добри суровини, а останалите се депонират.
условия за въвеждане на иновативни методи за Отпадъците от ТЕЦ могат да бъдат
преработка на отпадъци, изразяващи се преди използвани като ценен ресурс в много
всичко в добра инфраструктура и подходяща производства с цел извличането на многобройни
законова рамка за управление на отпадъците. екологични и икономически ползи. Пепелите и
Както бе отбелязано по-горе обаче, системата за сгуриите са подходяща суровина за
разделно събиране и рециклиране на отпадъците производството на различни видове изкуствени
все още не функционира достатъчно добре. добавки, с които се изготвят леки бетони.
Около 74 % от генерираните отпадъци през 2014 Използването им предоставя предимства като:
г. се транспортират към депата, при положение по-добри кохезионни свойства; по-добра
че голяма част от общото количество отпадъци текстура; по-лесен контрол на влагата в
са биоразградими. Площадките за депониране изделията; по-голяма здравина на изделията и
на отпадъци, както и броят на земеделските други. Пепелта и сгурията от ТЕЦ могат да се
стопанства в България са достатъчни използват успешно и в пътното и уличното
предпоставки за изграждане на централи за строителство самостоятелно като свързващо
производство на биогаз от земеделски продукти, вещество или в смеси като инертен пълнител,
както и на вторични биопродукти (компост и както за долните слоеве на основите на всички
течен органичен тор). Препоръчително е категории пътни настилки, така и за горни
съоръженията за производство на биогаз да са в слоеве на основите на пътна настилка на леко и
непосредствена близост до сметищата или в средно натоварени пътища.
самите промишлени или земеделски цехове, Сред иновационните технологии,
като по този начин се ограничават прилагани в България, особено място заема
транспортните разходи. активният демонтаж посредством т.нар. „умни
С интензивните услилия за гарантиране материали”. Използването им прави възможно
на енергийната сигурност и ефективност, които върху свързващите елементи да бъде приложена
през последното десетилетие се полагат в сила, така че продуктът да бъде разделен на
глобален мащаб, все по-голямо значение се съставните си части. На този етап с активен
отдава на възобновяемите енергийни източници. демонтаж могат да бъдат демонтирани

II-368
подвъзли, основни възли (включително среда, ефективното и високотехнологично
корпуси), PCB, LCD, антени, екрани, управление на отпадъците от човешката дейност
трансформатори, механични компоненти, би допринесло за формиране на нов
дискови устройства, бутони, фитинги, различни възобновяем ресурс и алтернативен източник на
видове винтови или други връзки. Тази енергия за националното и световно стопанство.
технология ще бъде особено ефективна в
следните случаи: когато замества ръчния
демонтаж, което води до намаляване на ЛИТЕРАТУРА
демонтажното време, а оттам и до намаляване на
разходите; когато се прилага вместо 1. United Nations, Department of Economic and
раздробяването и разделянето, като по този Social Affairs (2015). World Population
начин се повишава ефективността на Prospects: The 2015 Revision, Key Findings
рециклирането на материалите. and Advance Tables. New York.
2. International Monetary Fund (IMF) (October
2013). World Economic Outlook: Transitions
4. Заключение and Tensions. Washington. Available:
Ресурсите на земята са ограничени, а http://www.imf.org/external/pubs/ft/weo/2013/
това намалява и възможностите за развитие на 02/pdf/text.pdf.
постоянно нарастващото световно население. 3. United Nations Environment Programme
Отпадъците от една страна се образуват от (2015). Global Waste Management Outlook.
използването на ресурси или продукти, а от Available:
друга – те имат негативни ефекти върху http://unep.org/ietc/ourwork/wastemanagement
околната среда в резултат от тяхното /GWMO.
преработване или обезвреждане. Прекъсването 4. http://www.basel.int/.
на връзката между икономическия растеж и 5. Петков, В. (2014 г.). „Енергийна сигурност
нарастващото използване на ресурсите, което и пазарите на енергоносители в страните
води до негативно влияние върху човека и от Черноморския регион“. ГорексПрес,
природата, е основна цел в областта на София, стр. 228.
политиката по околната среда. Този 6. Чкорев, Н. (2015 г.). „Интегрираното
ръководен принцип за устойчиво развитие управление на отпадъците в българските
представлява не само ориентир за държавни региони – предизвикателства и последици“.
действия, но е и мерило за решенията на БАН, Институт за икономически
икономиката и обществото. Отпадъците изследвания, София.
представляват икономическо, екологично и 7. Закон за управление на отпадъците, Обн.
социално предизвикателство. Растящото ДВ, бр.53 от 13.07.2012 г.
потребление и развиващата се икономика 8. Министерство на околната среда и водите,
продължават да произвеждат големи количества Изпълнителната агенция по околна среда.
отпадъци, като все повече усилия са необходими „Национален доклад за състоянието и
за редуцирането и предотвратяването на опазването на околната среда 2015“.
образуването им. Макар в миналото отпадъците София. Available
да са приемани като непотребни, днес те все http://eea.government.bg/bg/soer/2013.
повече се възприемат като ресурси. Това се 9. Министерство на околната среда и водите.
вижда в смяната на фокуса при управлението на Available: http://www.moew.government.bg/;
отпадъци от обезвреждане към рециклиране и 10. Евростат. Available:
оползотворяване. http://ec.europa.eu/eurostat/data/database.
Управлението на отпадъците е една от
областите, в които въвеждането на иновативни
технологии не е лукс, а необходимост, Mariya Stefanova Petkova-Kozovska, PhD
продиктувана от условията на съвременния Department of International Economic
живот. Нуждата от иновации не е породена Relations and Business
единствено от нарастващото количеството University of National and World
отпадъци, в това число и такива, застрашаващи Economy–Sofia
екологичното равновесие, нито само от 8-mi dekemvri Str.
замърсяването на въздуха с вредни вещества от 1700 Sofia
изгаряне на отпадъците в инсталациите. Освен BULGARIA
че би спомогнало за разрешаването на E-mail: marypetkova@gmail.com
гореспоменатите проблеми, свързани с околната
ЕДИН ПОДХОД ЗА МОДЕЛИРАНЕ НА

СИСТЕМИ ЗА УПРАВЛЕНИЕ ПРИХОДА НА
ХОТЕЛ
СВЕТЛАНА ВАСИЛЕВА
Резюме: Социалните науки в България са слабо развити от гледна точка на

експерименталната наука. С проекта, представен тук се цели внедряване на
имитационното моделиране в мениджмънта и образованието за целта на
развитието на устойчива туристическа индустрия както като отделен
икономически отрасъл, така и от гледна точка на държавно управление. В
представения GPSS World симулационен модел на Система за управление прихода на
хотел входящите потоци GPSS транзакции могат да моделират различните видове
туристически потоци, които ще се изследват симулационно, за да бъдат взети
определени решения, касаещи мениджмънта на изследваните събития и
организации от значение за целите на съответните изследователски дейности.
Представени са резултати от изпълнението на GPSS модела в средата на
Разширения редактор на GPSS World.
Ключови думи: Управление на прихода на хотел, моделиращ алгоритъм, GPSS
World
ONE APPROACH FOR MODELLING HOTEL

REVENUE MANAGEMENT SYSTEMS
SVETLANA VASILEVA
Abstract: Social sciences in Bulgaria are underdeveloped in terms of experimental

science. The project presented here aims to introduce simulation modeling in management
and education for the purpose of development of a sustainable tourism industry as well as a
separate economic sector and from the perspective of governance. In the presented GPSS
World simulation model of Hotel Revenue Management System GPSS transactions‘ inflows
can model various types of tourist flow in simulation way will be explored to certain
decisions concerning the management of the examined events and organizations relevant
for the purposes of the relevant research activities. The results from the implementation of
the GPSS model in the mid Extended editor of GPSS World are presented.
Key words: Hotel revenue management, modeling algorithm, GPSS World
1. Основни положения външната среда [1]. Това е изключително важно

Една от актуалните задачи в в началния етап на развитието на всяка
Информатиката е проблемът за създаване на иновация, особено ако тя касае развитието на
Методология на изследването на реални обекти икономически отрасъл в дадена географска
със средствата на компютърната техника и област (и дори държава, а защо не и
наука. Т.е. това е проблем на създаването на международни проекти и колаборации).
информационни модели на основата на които Според Девятков [1]: “Пазарът на
може да се определи теоретичната значимост и приложенията на Имитационните изследвания
практическата ценност на всяка (произволна) на сложните икономически системи е
система в условията на въздействията на потенциално огромен. Имитационните
изследвания са изключително важни, а понякога
II-370
са и единствен метод за оценка и анализ на управление на прихода на хотел (Hotel Revenue
иновации, оптимизация на разходите на Management (HRM) Systems).
предприятия и проекти, подобряване на
производителността и увеличаване обема на 2. Концепция за моделиране на HRM
продукцията…“. системи
Методите на имитационното моделиране Всяка система (техническа,
позволяват на изследователите да създават информационна, икономическа, социална и др.)
модели на сложни системи. Но това все още и нейния алгоритъм на работа могат да бъдат
остава научна област [1]. Целта е внедряване на описани в термините на Теорията за масовото
разработваните модели в бизнеса – реално обслужване като Система за масово обслужване
използване на самостоятелни имитационни [7]. Имитационното моделиране позволява да се
приложения в качеството на системи за изследват системи за масово обслужване при
поддържане вземането на решения. Затова в различни типове входни потоци и
доклада се разглежда един пример за интензивности на постъпване на заявките на
приложение на системата за имитационно входовете на системите и определяне на
моделиране (ИМ) GPSS World [2] в основните характеристики на същите [7].
туристическата индустрия. Една от най-използваните системи за
За да се развива една индустрия (в ИМ – GPSS World [2] предоставя на
случая - хотелиерството) и да бъде ефективна е потребителите средства за взаимодействие със
необходимо да се подготвят съответните системата в процеса на симулационните
специалисти – мениджъри. Едно от основните изследвания и програмни средства, реализиращи
умения, които трябва да развиват всички непосредствено процеса на изпълнение на
бъдещи мениджъри на първо място в модела. Като обекти на езика GPSS се използват
икономиката и всички нейни отрасли това е аналози на такива стандартни компоненти на
системното мислене. По характеристики то е: системите за масово обслужване, като заявки,
контекстно-обусловено, неемоционално, обслужващи прибори, опашки и др.
операционно. [3] Изключително силно средство Достатъчният набор от подобни компоненти
за развиване на такъв тип умения предоставя позволява да се конструират сложни
имитационното моделиране. Появиха се много симулационни модели, при запазване на
нови системи за (не просто) имитационно привичната терминология на системите за
моделиране, а дори и „мултиподходно масово обслужване.
имитационно моделиране“, обединяващо Концепцията за (параметрични)
Системната динамика, Агентното и Дискретно- имитационни модели на хотелски системи за
събитийното моделиране – в лицето на AnyLogic управление на приходите е те да се използват
[4]. Обикновено тези системи са изключително първо за обучение на мениджъри в хотелския
скъпи, дори и пакетите с учебна цел. бизнес и на следващ етап да бъдат внедрени в
Но с появата на т.нар. системи за хотелиерския бизнес и управление.
автоматизация на имитационните изследвания Имитационното моделиране на HRM системи [8]
има възможност имитационното моделиране да включва решаването на следните задачи по [9]:
стане инструмент за обучение и - Дефиниране и систематизиране на
експерти/специалисти в конкретна предметна числените показатели, участващи в измерването
област (като разглежданата тук Хотелиерство и на състоянието на един хотел. При което се
Управление на туризма) да са активна част от отчита спецификата на: ареала, за който ще се
екипа, разработващ симулационни приложения. разработва хотелския бизнес; на туристическите
Такива са Разширеният редактор [5] и потоци, потребители на услугите, предлагани в
Универсалният редактор на формите [6], ареала; външната по отношение на хотела среда,
разработени над системата GPSS World [2]. обект на интерес от различните туристически
Системата за ИМ и нейните разширения потоци
създават възможност симулационните - Съставяне концептуалния модел на
приложения да са много по-ефективни, защото система за управление на приходите на хотел за
логиката на експертите/обучителите се залага целите на симулациите със средствата на
директно в разработваните продукти. средата GPSS World и определяне на входните
Съществуващите симулационни приложения, параметри и изразите, в които участват;
разработвани от фирми не притежават - Моделиране на модулите, формиращи
гъвкавостта и „индивидуалната насоченост“ приход на хотел;
каквато е целта на разглежданите имитационни - Съставяне на алгоритми, моделиращи
приложения – моделиране на системи за работата на система за управление на приходите
II-371
на хотел. Настройка на моделите, валидация и транзактите и задаване на стойности на
верификация; параметрите на всеки постъпил транзакт по [9]:
- Имплементация и апробация на P1, P2 – Месец и ден от месеца
имитационните модели ще се извършва със „заявявани“ от транзакта за първа нощувка;
средствата на Разширения редактор и P3 – Продължителност на заявявания от
Универсалния редактор на формите на GPSS транзакта престой;
World. Универсалният редактор на формите P4, P5 - Месец и ден от месеца
предоставя изключително полезните за целите „заявявани“ от транзакта за последна нощувка;
на имитационните проектни изследвания P6 – Номер на колоната в GPSS
средства: Форма за въвеждане на данни; Форма матрицата MX$Booking, моделираща базата от
за планиране на експеримента (единичен и данни, където в реда с номера на резервираната
множество експерименти); Форма за динамика стая ще се постави запис за първата нощувка;
на експеримента; Анимационна форма за P7 - Номер на колоната в GPSS
представяне работата на модел; матрицата MX$Booking2, където в реда с номера
- Формулиране на изводи за бъдещи на резервираната стая ще се постави запис за
разширения и приложения на методиката на последната нощувка. Разглежданият модел е на
параметричните имитационни GPSS World сезонен (морски) хотел;
модели в областта на креативните и P7 - Вероятност резервацията
рекреативните индустрии. (моделирана от GPSS транзакта) да бъде
канцелирана по-късно (в модела - след
3. Елементи на GPSS World модела на приблизително 1÷2 месеца). Стойност 0 –
HRM системи означава, че GPSS транзактът ще канцелира
Разработеният в средата на GPSS World резервацията след изтичането на 62800±21600
имитационен модел включва следните стъпки по моделни единици (В модела за моделна единица
[9]: е приета 1 минута), т.е. след 43±15 дни. След
- Генериране на поток транзакти Тi, успешно поставен запис за резервиране на стая в
моделиращи заявките за резервиране на стаи в матриците MX$Booking и MX$Booking2,
моделирания хотел. След постъпването на моделиращи базата от данни с резервации,
транзакт в модела (програмирано чрез GPSS транзактът излиза от опашката “Rezervacii”, но
блок GENERATE) на параметрите на транзакта остава в общата опашка „TotalB“ в модела. И
се задават стойности – в тази част от алгоритъма след посочения по-горе нормално разпределен
се моделират желаните от туриста дата на първа интервал параметърът P8 ще получи стойност 2
нощувка, продължителност на престоя (дата на и се изпраща към съответния сектор в модела.
последна нощувка); Транзактът ще нулира записите за резервацията
- Търсене в матриците, моделиращи на стаята в матриците MX$Booking и
базата от данни на хотелската информационна MX$Booking2 и ще напусне модела; Стойност 1
система дали има свободно количество стаи за – Транзактът, ако успее да резервира стаята за
желаните дати; датите, зададени в параметрите P1, P2, P4 и P5,
- При намиране на свободна стая/стаи, те няма да канцелира резервацията и прихода,
се резервират чрез поставяне на съответните формиран от тази резервация ще остане добавен
записи в матриците, моделиращи базата от към общия приход на хотела.
данни, след което броячът на успешните заявки В моделиращия алгоритъм се задават и
се увеличава с 1. Изчисляват се изразите, които други стойности за параметри на транзактите,
показват за моделирания хотел прихода от които са помощни. Предвижда се за
резервации; разширяването и параметризирането на модела
- Ако транзактът не е успял „да да бъдат задавани и други стойности в
резервира“ стая/стаи, съответно не поставя параметрите на генерираните транзакти.
никакъв запис. Увеличава се броячът на Например, предвижда се моделирането на
отказаните заявки с 1; генериране на приход и от други услуги в хотела
- Транзактът напуска модела (с GPSS (ресторант, спа и други). Освен това се
блок TERMINATE). предвижда отчитането и на други параметри на
туристическия поток като представените в [10]:
3.1. Параметри на генерираните Тип на престоя (бизнес, почивка и др.); Тип
транзакти националност; Ниво на интерес към хотелските
На фиг. 1 е показан началният сегмент от услуги; и други характеристики на
моделиращия алгоритъм – постъпване на туристическите резервации и туристическите

II-372
потоци, които влияят на формирания приход на броя на стаите при реализацията на цикъла за
хотела. проверка на свободни стаи в достъпните крила
на моделирания хотел;
OccupRoomN – изчислява заетостта на
моделирания хотел Occupancy rate по формула
(1) [8, с.43]:
NumberOfRoomnights (1)
Occupancy roomnights  * 100%
NumberOfRoomsAvailable
ADRroomnights – изчислява средната

цена за една нощувка (Average daily rate - ADR)
по формула (2) по [8, с.45]:
Room _ revenues (2)

ADR roomnights 
NumberOfRoomnights
Фиг. 1. Задаване стойности на параметрите

на GPSS транзактите в модела на HRM
3.2. Променливи и изрази

Някои от най-важните променливи, чрез
които се изчисляват формирания приход и
помощни стойности, участващи в различни
изрази, а също и при определяне пътя на
транзактите в GPSS модела [9] са (фиг. 2):
DayMonths – използва се за изчисляване
на месеца на първата заявявана нощувка;
DayBook1 –за изчисляване на деня в
месеца на първата заявявана нощувка;
DayBook2 – използва се за изчисляване
на деня в месеца на последната заявявана Фиг. 2. Дефиниране на променливи в
нощувка; средата на Разширения редактор на GPSS с
Day1TabM1 – изчислява датата на показана контекстуална подсказка
първата заявявана нощувка от началото на
кaлендарната година в брой дни (като серийно 3.3. Функции
число); Езикът GPSS предоставя много мощен
DayTabM2 – изчислява датата на първата инструмент за описание на (статистически)
заявявана нощувка от началото на сезона в брой разпределения на случайни величини:
седмици; равномерно, експоненциално, нормално и други
DayTabCol1 – определя номера на [7]. За представения модел са определени
колоната в матрицата за резервации следните функции: за описание на броя на дните
MX$Booking1, където ще се постави записа за във всеки календарен месец; вероятностно
първата заявявана нощувка; разпределение на резервациите по месеците от
DayTabCol2 – определя номера на сезона; задаване броя на дните от началото на
колоната в матрицата за резервации годината до всеки от месеците на летния сезон
MX$Booking2, където ще се постави записа за (фиг. 2, горе):
последната заявявана нощувка; DayMonths – задава броя на дните във
DayMonths0 – за определяне на деня от всеки месец от годината;
месеца за последната нощувка, когато датата е MonthSummer – разпределение на
от следващ месец; вероятностите, че генерирания транзакт
Krilo2, Krilo3 – изчисляват номера на (моделиращ заявка към хотела) ще бъде за
първата стая от второто и третото достъпно месеца, присвояван на параметъра P1 (за месеци
„крило“ на моделирания хотел; 5 и 9 – вероятността е 0.10, за месец юни - 0.20,
Krila1and2, Krila, RoomKr2, RoomKr3 – и за месеците 7 и 8 - 30% вероятност).
допълнителни променливи за изчисляване на
II-373
Вероятностното разпределение, задавано от На фиг. 4 е показана част от
функцията може да се промени при наличие на моделиращия алгоритъм – цикъла на търсене на
други статистически данни за реалните свободна стая в един сектор (в примера – първи
туристически потоци; сектор). Може да се види как се използват
DaysFromBeg – определя броя на дните описаните променливи и съхранявани стойности
от началото на календарната година до първия при търсенето на свободна стая и поставянето на
ден на месеца, чийто номер е записан в резервация в матриците, моделиращи базата от
параметъра P1. данни на информационната система на хотела.
3.4. Клетки и съхранявани стойности
Освен променливи и изрази за
определяне на различните метрики на HRM на
моделирания хотел се използват и други
средства, предоставяни от GPSS World, като
клетки и съхранявани стойности [9]. На фиг. 3 e
показан прозорецът на Съхраняваните величини
в Стандартния отчет от симулация в средата на
Разширения редактор на GPSS World със
стойностите на началните „параметри“ и
„броячите“, използвани в модела:
WeekOpen – Начална стойност 120 –
брой дни от началото на годината до отварянето Фиг. 4 Моделиране търсенето на свободна
на хотела за туристи; стая
WorkWeeks – брой седмици (на сезона),
в които хотелът е достъпен за туристи; 4. Изпълнение на моделиращия
BookingsAll – брой на генерираните алгоритъм
транзакти по време на моделирането; Резултати от изпълнението на
BookingsEject – брой транзакти, не синтезирания алгоритъм, моделиращ HRM
успели да резервират стая за датата, посочена в система могат да бъдат наблюдавани в
параметрите им P1 и P2; прозорците на стандартния отчет от симулация в
BookRooms – брой транзакти, напуснали Разширения редактор на GPSS World, като
модела, успешно резервирали стая в GPSS показаните на фиг. 3 и фиг. 5. Средата на
матриците Booking и Booking2; Разширения редактор, както и на GPSS World
Roomnights – общ брой на резервираните предоставят отчет от симулациите и прозорци, в
нощувки; които може да се наблюдават резултатите от
RoomReven – Сума на прихода, симулацията (в отчета) и динамиката на
генериран от резервации на стаи. На фиг. 3 може изпълнението на модела в прозорците (Прозорец
да се види формирания приход на хотела – на съхраняваните стойности SaveValues Window
стойността на RoomReven. (фиг. 3), Прозорец на блоковете Blocks Window,
Прозорeц на матрицa Matrix Window и други).
Фиг. 3 Прозорецът на Съхраняваните

величини в Стандартния отчет от
симулациите в средата на Разширения Фиг. 5 Прозорец на матрицата RoomList в
редактор на GPSS средата на Разширения редактор

II-374
На фиг. 5 е показан прозореца на 5. Федотов, М. и Девятков, В. (2013) САИИ
Матрицата RoomList, която е част от „базата от для профессионалов – развитие
данни“ за резервации и моделира достъпността возможностей расширенного редактора
на стаите на хотела и цените им. GPSS World. Сборник докладов шестой
всероссийской научно-практической
5. Заключение конференции «Имитационное
Най-важната задача при разработването моделирование. Теория и практика»
на всяка една индустрия е задачата за оценка на (ИММОД-2013). Т. 1, Изд. «ФЭН»
показателите на функционирането на системата, Академии наук РТ, Казань, c.356-360.
които накрая се използват за решаване на http://simulation.su/files/immod2013/material/i
приложната задача. Имитационното моделиране mmod-2013-1-356-360.pdf.
дава възможност да се проведат предварителни 6. Исаев, Ф. и Девятков, В. (2013). Основные
изследвания и да се оценят: характеристики на возможности универсального редактора
работата на системата: пропускателна форм для GPSS World. Сборник докладов
способност, вероятност за обслужване, време за шестой всероссийской научно-
обслужване; икономически показатели, като практической конференции
приходи и разходи и др. Дейностите по «Имитационное моделирование. Теория и
разработваните имитационни модели имат за практика» (ИММОД-2013). Т. 2. Изд.
цел не само Развитие на теорията, «ФЭН» Академии наук РТ, Казань, 2013,
методологията и практиката на компютърния c.149-154.
експеримент в рекреативните индустрии, но и в http://simulation.su/files/immod2013/material/i
социално-икономическите изследвания и в mmod-2013-2-149-154.pdf.
задачите на управлението на България. 7. Томашевский, В.Н., Жданова, Е.Г. (2003).
Работата по симулирането на HRM Имитационное моделирование в среде
системи поставя като генерална цел създаването GPSS. Бестселлер, Москва.
на експериментални комплекси и 8. Ivanov, S.H. (2014). Hotel revenue
разработването на модели на експериментална management – from theory to practice.
икономика и определянето на ефективни Zangador, Varna, Bulgaria.
направления за развитие. 9. Vasileva, S.Z. (2016). GPSS World Extended
Editor’s Opportunities to Simulate Hotel
ЛИТЕРАТУРА Revenue Management Systems. International
Conference Automatics and Informatics’16,
1. Девятков, В.В. (2014). Развитие Conference Proceedings, October 4-5, Sofia,
методологии имитационных исследований Bulgaria, p. 267-270.
сложных экономических систем. 10. Walek, B., Hosek, O., Farana, R. (2016).
Диссертация на соискание ученой степени Proposal of Expert System for Hotel Booking
доктора экономических наук, Москва, РФ. System. Proceedings of the 17th International
http://elib.fa.ru/avtoreferat/devyatkov.pdf/dow Carpathian Control Conference (ICCC), IEEE,
nload/devyatkov.pdf p.804-807
2. Minuteman Software. (2010). GPSS World
Tutorial Manual.
http://www.minutemansoftware.com
3. Каталевский Д. (2015). Основы Контакти –
имитационного моделирования и Светлана Василева-Бояджиева
системного анализа в управлении, Изд.дом Висше училище по мениджмънт
Дело, Москва. Адрес: Оборище 13А, 9000 Варна
4. AnyLogic. (2017). Многоподходное Телефон: +359 878491157
имитационное моделирование. E-mail: svetlanaeli@abv.bg
http://www.anylogic.ru/overview.
ИЗПОЛЗВАНЕ НА ПЕКТИНОВИ ПОКРИТИЯ

ЗА ПРЕСНИ ПЛОДОВЕ И ЗЕЛЕНЧУЦИ
РАДОСЛАВ РАДЕВ
Резюме: Увеличеното потребителско търсене на висококачествени храни с

продължителен срок на съхранение е основна предпоставка за разработване на
иновативни технологии в хранително-вкусовата промишленост. Ядивните
покрития представляват тънък слой съставен от биополимерни компоненти, който
може да се консумира със стоката, върху която е нанесен. Той намалява загубата
на влага, селективно контролира обмена на газове (кислород, въглероден диоксид и
етилен), които участват в процеса на дишане. Ядивните покрития удължават
съхраняемостта, поддържат свежестта, инхибират микробната развала на
продукта, намаляват загубата на ароматични вещества и на други важни
компоненти. Целта на настоящото изследване е да се проучи научната литература
относно използването на пектинови покрития за пресни плодове и зеленчуци.
Ключови думи: пресни плодове и зеленчуци, пектинови покрития, ядивни покрития
USE OF PECTIN COATINGS FOR FRESH FRUIT

AND VEGETABLES
RADOSLAV RADEV
Abstract: Increased consumer demand for high quality food products with a long shelf life
is a fundamental prerequisite for the development of innovative technologies in the food
industry. Edible coatings are thin layer composed of biopolymer components that can
consume the goods to which they apply. This reduces the loss of moisture, selectively
controls the exchange of gases (oxygen, carbon dioxide and ethylene). Edible coating to
prolong the storage, maintenance of freshness, inhibit microbial spoilage of the product,
reducing the loss of aromatic substances and other essential components. The aim of this
study was to investigate the scientific literature on the use of pectin coatings for fresh fruits
and vegetables.
Key words: fresh fruits and vegetables, pectin coatings, edible coatings
1. Въведение което оказва положително влияние върху

Увеличеното потребителско търсене на съхраняемостта им. Увеличеното използване на
висококачествени храни с продължителен срок синтетични опаковки, води до сериозни
на съхранение е основна предпоставка за екологични проблеми, дължащи се на тяхната
разработване на иновативни технологии в небиоразградимост. Потреблението на
хранително-вкусовата промишленост. Те целят хранителни продукти и изхвърляните след
да запазят естествен и свеж външния вид на консумация полимерни опаковки замърсяват
храните възможно най-дълго време, като околната среда. Това поражда необходимост от
основно изискване към тях е да са безопасни. използването на биоразградими материали
Опаковката е важен елемент в концепцията за (възобновяеми източници), суровини от
осигуряване на подходяща (механична и селското стопанство, морски отпадъци от
функционална) защита на различни храни и хранително-вкусовата промишленост и др.,
суровините, от които те се произвеждат. Тя които могат да се употребяват за създаването на
предпазва от окисление и микробна развала, биоразградими опаковки. Въпреки, че пълна
II-376
подмяна на синтетичните опаковки е задържат летливите съединения, с което се
невъзможна, използването им може да бъде запазват вкусово-ароматичните свойства на
ограничено с разработване на подходящи по пресните плодове и зеленчуци за по-дълъг
състав и свойства ядивни покрития за период от време и се намалява микробното им
определени стокови групи [1,2]. замърсяване. С цел да се реши проблема свързан
Синтетичните филми се използват за с намаляване загубата на влага при пресни
опаковане на пресни плодове и зеленчуци, но те плодове и зеленчуци, редица изследователи се
не са биоразградими и се натрупват в околната стремят да разработят ядивни покрития с
среда, което води до екологичен дисбаланс и желаните бариерни свойства. В същото време се
замърсяване на нашата екосистема. Тези извършват проучвания за подобряване адхезията
обстоятелства създават необходимост от и трайността на покритието, нанесено върху
разработване на алтернативни екологични повърхността на пресните плодове и зеленчуци,
опаковки, които са най-добрата екологично както и изследване на сензорните му свойства
чиста алтернатива на пластмасите [3]. [6].
Целта на настоящото изследване е да се През последните години използването на
проучи научната литература относно ядивните покрития, които удължават
използването на пектинови покрития за пресни съхраняемостта и подобрят качеството на
плодове и зеленчуци. пресните плодове и зеленчуци нарастна.
Причините за това са свързани предимно с
2. Значение на ядивните покрития за въпросите, отнасящи се до замърсяване на
пресните плодове и зеленчуци околната среда, дължащо се на използването на
Прилагането на ядивни покрития е един конвенционални синтетични опъковки за храни,
от най-иновативните подходи за повишаване които са трудно разградими.
съхраняемостта на пресните плодове и Ефективността на ядивните покрития за
зеленчуци [4]. Най-често използвани основни защита на пресните плодове и зеленчуци зависи
суровини за създаването им са полизахарди, от мокрещата способност на покритието, което
липиди и антимикробни агенти. Широк спектър се отразява върху дебелината му. По този начин,
от научни изследвания доказват, че ядивните те се разпространяват равномерно върху
покрития се използват за повишаване срока на повърхността им и се изсушават, с което се
съхраняемост и качеството на пресни плодове и постига необходимото сцепление [4].
зеленчуци чрез създаване на модифицирана Пресните плодове и зеленчуци
атмосферна среда във вътрешността им поради притежават основни съставки, които са
техните бариерни свойства за газове, влага и необходими за дневния хранителен прием на
микробно развитие. Те подобряват тяхната човешкия организъм. Те са важен източник на
хранителна стойност и външен вид. Ядивните витамини, минерали, антиоксиданти,
покрития се използват основно за пресни флавоноиди, диетични фибри и вкусови
плодове и зеленчуци, за забавяне на зреенето по вещества. След прибиране на реколтата, в
време на съхранение. Трайността при резултат на въздействието на
зеленчуците се удължава значително чрез микроорганизмите, дишането и загубата на
добавяне на вещества, които потискат растежа влага пресните плодове и зеленчуци бързо
на микроорганизмите, отделят или абсорбират намаляват своята трайност и хранителна
подбрани компоненти. Способността на стойност.
ядивните покрития да удължават Най-важните фактори за качество на
съхраняемостта на пресни зеленчуци зависи от: пресните плодове и зеленчуци, които
техния химичен състав, структурата на допринасят за продаваемостта им са: текстура,
филмообразуващия полимер, дебелината на цвят, външен вид, вкус, аромат, хранителна
покритието, начина на формиране и нанасяне, стойност и микробиологична безопасност.
вида и свойствата на емулгаторите и Качеството на пресните плодове и зеленчуци
пластификаторите, които се използват и от зависи от растителния сорт, етапа на зреене,
условията на съхранение. Добрите резултати на степента на зрялост, условия на съхранение след
ядивното покритие за забавяне на промените прибиране на реколтата. Загубите след
при съхранение зависят от физиологията на прибиране на реколтата на пресните плодове и
пресните плодове и зеленчуци и присъствието зеленчуци са сериозен проблем, защото той
на естествени покрития като восъци, кутикула бързо ги влошава по време на съхранение и
на повърхността им [3,5]. транспортиране. Ядивните покрития нанесени
Покритията забавят дехидратацията, върху повърхността на пресните плодове и
потискат дишането, подобряват текстурата,
II-377
зеленчуци се използват за подобряване на повишава съхраняемостта от 4 седмици при
тяхното качество и срок на годност. [7]. температура 30°C [14].
В друго изследване е доказано, че
3. Същност на ядивните покрития пектиновото покритие е с по-добри резултати от
Ядивните покрития се използват за нишестено покритие с добавка на растителна
удължаване на срока на годност на пресните смола. Нанесеното покритие върху два вида
плодове и зеленчуци. Те могат да бъдат стафиди съхранявани при подходящи условия ги
консумирани като част от продукта върху който запазва непроменени в продължение на 6
са нанесени без да му влияят неблагоприятно месеца. Проведените микробиологични
[7]. изследвания доказват, че по време на
Ядивните покрития представляват тънък съхранение броя на микроорганизмите се
слой съставен от биополимерни компоненти, понижава. Получените резултати по сензорните
който може да се консумира със стоката, върху показатели цвят, текстура и вкус на стафидите
която е нанесен. Той намалява загубата на влага, покрити с пектин са с най-висока оценка [15].
селективно контролира обмена на газове Сложно покритие съставено от пектин
(кислород, въглероден диоксид и етилен), които (хидрофилен полимер), пчелен восък
участват в процеса на дишане. Ядивните (хидрофобна фаза), сорбитол (пластификатор) и
покрития удължават съхраняемостта, поддържат емулгиращ агент, нанесено върху повърхността
свежестта, инхибират микробната развала на на краставици ефективно намалява загубата на
продукта, намаляват загубата на ароматични влага и удължава съхраняемостта. С покритието
вещества и на други важни компоненти [8, 9, се запазват различни параметри на качеството:
10]. консистенция, цвят на кожата (хлорофил) и
други. Всички промени в обекта на изследването
4. Използване на пектинови намаляват при по-ниски температури на
покрития за пресни плодове и съхранение. Срокът на съхранение на
зеленчуци краставиците при естествена атмосфера е
Пектинът е полизахарид, изолиран от удължен от 2 до 5 дни, а в хладилни условия до
стените на растителните клетки. Получава се от две седмици [16].
кори на цитрусови плодове, ябълки, цвеклови
резени и др. растителни суровини. Пектиновите 5. Заключение
покрития са с висока пропускливост за водна Въз основа на проучената научна
пара, а хидрофилния им характер може да бъде литература се установи, че изследванията за
подобрен чрез добавянето на парафин или пектиновите покрития нанесени на
пчелен восък [11]. повърхността на пресни плодове и зеленчуци не
Пектинови покрития в комбинации с са много. Получени са определени положителни
пчелен восък, сорбитол и моноглицериди резултати при нанасянето на пектинови
намаляват промяната на цвета, омекването на покрития върху повърхността на манго, авокадо,
тъканите и естествените загуби (фири) при домати, стафиди и краставици. Това дава
манго. Те удължават съхраняемостта на основание да се планират бъдещи научни
изследвания продукт с над една седмица, след изследвания на неизследвание до момента
като в съчетание с другите компоненти пресни плодове и зеленчуци с нанесени на
осигуряват контролирано дишане и повърхността пектинови покрития.
пропускливост за водна пара [12].
Авторски колектив доказва, че авокадо с ЛИТЕРАТУРА
нанесени на повърхността пектинови покрития
забавя разпространението на болестта 1. Maftoonazad, N., Badii, F. (2009). Use of
Lasiodiplodia theobromae, както и свързаните с edible films and coatings to extend the shelf
нея промени в качеството (текстура и цвят). life of food products. Recent, patents on food,
Получените резултати са много по-добри в nutrition & agriculture, Volume 1, Number 2,
сравнение с тези на контролната проба [13]. pp. 162 - 170.
В проведено изследване са установени 2. Tharanathan, R. N. (2003). Biodegradable
добри резултати при покритие с включен в films and composite coatings: past, present and
състава си 3% пектин нанесено върху future. Trends in food science & technology,
повърхността на домати. То запазва физико- Volume 14, Issue 3, pp.71 - 78.
химичните параметри, поддържа качеството по 3. Mishra, B., Khatkar B. S., Siddiqui S., (2006).
време на съхранение и зреене, с което се Edible coatings for postharvest preservation of
fresh fruits and vegetables. Advances in

II-378
postharvest technologies for horticultural 12. Moalemiyan, M., Ramaswamy H., Maftoonazad
crops, pp. 131 - 172. N., (2012). Pectin - based edible coatings for
4. Scetar, M., Kurek M., Galic K., (2010). Trends shelf - life extension of Ataulfo Mango.
in fruit and vegetable packaging - a review. Journal of food process engineering, Vol. 35,
Croatian journal of food technology, Issue 4, pp. 572 - 600.
biotechnology and nutrition, 5, (3 - 4), pp. 69 - 13. Maftoonazad, N., Ramaswamy H., Moalemiyan
86. M., Kushalappa, A. (2007), Effect of pectin -
5. Kokoszka, S., Lenart, A., (2007), Edible based edible emulsion coating on changes in
coatings - formation, characteristics and use - a quality of avocado exposed to Lasiodiplodia
review. Polish journal of food and nutrition theobromae infection, Carbohydrate polymers,
sciences, Vol. 57, No. 4, pp. 399 - 404. Vol. 68, Issue 2, 21, pp. 341 - 349.
6. Lin, D., Zhao, Y. (2007). Innovations in the 14. Delina, E. and Mahendran, T. (2009), Physico -
development and application of edible coatings chemical properties of mature green tomatoes
for fresh and minimally processed fruits and (Lycopersicon esculentum) coated with pectin
vegetables. Comprehensive reviews in food during storage and ripening, Tropical
science and food safety, Vol. 6, pp. 60 - 75. agricultural research & extension, 12, (2), pp.
7. Raghav, P. K., Agarwal, N., Saini, M. (2016). 111 - 113.
Edible coating of fruits and vegetables: a 15. Ghasemzadeh, R., Karbassi, A., Ghoddousi H.,
review. International Journal of Scientific (2008), Application of edible coating for
Research and Modern Education (IJSRME), improvement of quality and shelf - life of
Volume I, Issue I, pp. 188-204. raisins. World applied sciences journal, 3, (1),
8. Bourtoom, T., (2008), Edible films and pp. 82 - 87.
coatings: characteristics and properties. 16. Moalemiyan, M., and Ramaswamy, H. 2012,
International food research journal, 15, (3), pp. Quality retention and shelf - life extension in
237 - 248. mediterranean cucumbers coated with a pectin -
9. Kester, J. and Fennema О., (1986), Edible films based Film. Journal of food research, Vol. 1,
and coatings: a review, Food technology, 40, No. 3, pp. 159 - 168.
(12), pp. 47 - 59.
10. Pavlath, А., and Orts W.., (2009), Edible films Контакти:
and coatings for food applications, pp. 1 – 24. Име на автора:
Publisher: Springer, New York, USA. Радослав Радев
11. Baldwin, E., (2007). Handbook of food Организация: ИУ-Варна
preservation, second edition, pp. 477 - 507 Адрес: Варна 9002 бул. „Княз Борис I-
Publisher:,CRS Press Taylor and Francis ви“ 77
Group, New York, USA. Телефон: 0889/126 623
E-mail:r.radev_ue@mail.bg
ПРИЛОЖЕНИЕ НА ЛИПИДНИ ПОКРИТИЯ В

ХРАНИТЕЛНИТЕ ТЕХНОЛОГИИ
РАДОСЛАВ РАДЕВ
Резюме: Липидните покрития са гъвкави и устойчиви на счупване. Те притежават

добри бариерни свойства срещу водна пара и кислород. Това са основните фактори
влияещи върху промените на органолептичните свойства, на физико-химичните
свойства и на микробиологичната развала на храните по време на съхранението.
Целта на настоящото изследване е да се проучи научната литература относно
приложението на липидните покрития в хранителните технологии. Въз основа на
проучената литература се установи, че значително по-голямата част от научните
изследвания за липидни покрития са насочени към приложението им за пресни
плодове и зеленчуци.В заключение са посочени бъдещите направления за изследвания
на липидните покрития.
Ключови думи: ядивни покрития, липидни покрития, храни
APPLICATION OF LIPID COATINGS IN FOOD

TECHNOLOGY
RADOSLAV RADEV
Abstract: Lipid coatings are flexible and resistant to breakage. They possess good barrier
properties against water vapor permeability and oxygen. These are the main factors
influencing changes in the organoleptic properties, of the physicochemical properties and
microbiological spoilage of food during storage. The main purpose of this study was to
investigate the scientific literature about the use of the lipid coatings in food technology.
Based on the research literature found that a significant majority of research on lipid
coatings are focused on their application for fresh fruit and vegetables. In conclusion are
set future directions for research on lipid coatings.
Key words: edible coatings, lipid coatings, food
1. Въведение известен като "larding" - съхранение на различни

Ядивните покрития са нанасяни върху плодове чрез покриване с восъци или мазнини
повърхността на различни плодове в [1]. Восъчните покрития намаляват загубата на
продължение на векове за предотвратяване влага и забавят дишането на пресни плодове и
загубата на влага, за създаване на лъскава зеленчуци, в резултат, на което се удължава
повърхност и за подобряване на външния им съхраняемостта им. Различни восъци са
вид. Тази практика е използвана дълго преди да нанасяни върху повърхността на пресни плодове
се изяснят свързаните с нея биохимични и зеленчуци. Използвани са горещо топени
процеси [1]. восъци или емулсии. През Средновековието
През XII и XIII в. в Китай пресни липидните покрития са прилагани, за да намалят
портокали и лимони са покривани с восък, който естествените загуби и изсъхването на месо и
забавя изсушаването [2,3] и промените в сирена [5].
плодовата тъкан [4]. Методът е най-добрия за През XV в. са създадени ядивни
този период от време и се използва дълго поради покрития от варено соево мляко, които са
липсата на по-ефективен. В Европа процесът е използвани в Япония за поддържане качеството

II-380
на храните и подобряване на външния им вид съхранение. Необходимо е да притежават
[1]. приемливи сензорни характеристики,
През XVI в. в Англия са предлагани подходящи бариерни свойства (CO2, O2, вода),
храни с нанесени на повърхността мазнини за микробна, биохимична и физикохимична
намаляване загубата на влага от продуктите [2], стабилност. Ядивните покрития трябва да бъдат
а през XIX в. в САЩ е издаден първия патент за безопасни и да се получават чрез бърза и лесна
запазване качеството на различни месни технология при ниски цени. Те могат да
продукти [1]. действат, като ефективен носител на
През 1930 г. в търговската мрежа се антиоксиданти, вкус, цвят, хранителни или
предлагат пресни ябълки и круши с нанесени на антимикробни компоненти [10].
повърхността восъци [3], а в САЩ парафина се
използва за първи път за защита на цитрусови 3. Приложение на липидни покриия в
плодове [6]. Значителни изследвания за ядивни хранителните технологии
покрития започват през 1950 г. [2], но до 1967 г. Липидните покрития са гъвкави и
употребата им се ограничава най-вече с восъчни устойчиви на счупване. Те притежават добри
слоеве предназначени за плодове. През бариерни свойства срещу водна пара и
изминалите години приложението на ядивните кислород. Това са основните фактори влияещи
покрития нанасяни върху различни хранителни върху промените на органолептичните свойства,
продукти значително се разраства. През 1986 г. в на физико-химичните свойства и на
САЩ фирмите, които предлагат такива микробиологичната развала на храните по време
продукти са малко над 10, а до 1996 г. броя им е на съхранението [11].
600. Днес използването на ядивните покрития
расте бързо, като основно се цели запазване 3.1. Восъчни и парафинови покрития
качеството на разнообразни хранителни Покритията нанасяни върху
продукти с общи годишни приходи над 100 млн. повърхността на пресни плодове и зеленчуци
$ [1]. често се наричат "восъци", независимо дали
В момента, ядивните покрития се някоя от съставките наистина е восък. С восък
използват най-вече за нанасяне върху пресни най-често се обработват ябълки, авокадо,
плодове и зеленчуци, месо и месни продукти, цитрусови плодове, краставици, патладжан,
бонбони и някои ядки [5]. праскови, сладък пипер и домати, което
Целта на настоящото изследване е да се подобрява външният им вид. При проучване на
проучи научната литература относно връзката на восъците с процесите разваляне,
приложението на липидните покрития в нараняване и покафеняване е установено, че в
хранителните технологии. повечето случаи плодовете покрити с восъци са
с по-ниска честота на дишане, отколкото
2. Същност на ядивните покрития непокритите проби, което подобрява
Ядивните покрития представляват тънък съхраняемостта им [12].
слой съставен от ядивни биополимерни Восъците (минерални масла, парафин,
компоненти, който може да се нанесе върху канделилов и пчелен восък) се използват, като
повърхността на определени храни и да се покрития нанесени върху плодове като
консумира с тях. Той намалява загубата на портокал, лимон, грейпфрут, ябълка, круша,
влага, селективно контролира обмена на газове череша, банан, гуава, манго, кокосов орех,
(кислород, въглероден диоксид и етилен), които праскови, грозде, цяло и прясно нарязано
участват в процеса на дишане. Ядивните авокадо. Нанасят се и върху някои зеленчуци
покрития удължават съхраняемостта, поддържат като морков, краставица, тиква, пъпеш, сладка
свежестта, инхибират микробната развала на царевица, патладжан, пипер, домати, аспержи,
продукта, намаляват загубата на ароматични целина, ряпа и картофи. Восъците и маслата,
вещества и на други важни компоненти [1, 7, 8]. самостоятелно или в емулсия с хидроколоиди
Ядивните покрития се нанасят върху ефективно удължават срока на съхраняемост. Те
повърхността на храната, обикновено чрез подобряват защитните (бариерни) свойства на
метода на потапяне в определен разтвор или покритията за вода, което предотвратява
чрез метода на напръскване с пулверизатор [9]. естествените загуби (фири). Въпреки това,
Биоразградимите ядивни покрития се нанасянето на по-дебел слой восъци върху
създават предимно от естествени полимери и пресни плодове и зеленчуци силно променя
функционални съставки, които защитават бързо обмена на кислород и въглероден диоксид,
развалящи се хранителни продукти, запазват което създава анаеробни условия на съхранение
качество им и забавят промените при [13,14].
II-381
Восъчните покрития са значително по- наблюдава промяна във водната активност.
устойчиви на влага от повечето липидни. Ядивното покритие от пчелен восък
Въпреки това, покритията на восъчна и маслена предотвратява втвърдяването и улеснява
основа имат определени проблеми във връзка с обелването на италианския салам.
дебелина, хомогенност, мазна повърхност и Резултатите от това проучване доказват,
органолептичните свойства. Малко са откритите че ядивно покритие от пчелен восък може да
научнати изследвания по отношение на бъде полезна алтернатива на пластмасови
ефективността на восъци, мазнини, и масла като опаковки за италиански салам. Този естествен
защитни покрития за месо. Покритие от материал за покритие може да се използва за
минерално масло и восък намаляват загубата на увеличаване периода на съхранение на
влага от замразени птици повече от покритие от изследваната стока без да се нарушава
царевично масло или свинска мас. В сравнение структурата. Пчелният восък ефективно
със синтетичните опаковки загубите са по- намалява развитието на липидната
високи [15]. пероксидация в тези колбаси [18].
Пчелният восък е най-известния В друго изследване е извършен
компонент, който засяга непрозрачността, която сравнителен анализ на ефектите на няколко
нараства с увеличаването му. Влага се в концентрация на покрития от пчелен восък,
покрития за храни и е от решаващо значение за които оказват влияние върху условията на
поддържане на свежестта, контролирането на съхранение, качеството и срокът на годност на
микробния растеж и осигуряване на по-добра мандарини. Получените резултати доказват, че
текстура на продукта. Покритията от пчелен комбинацията на 12% восъчно покритие (пчелен
восък контролират дейността на водата в восък, олеинова киселина, триетаноламин) и
продукта, като предотвратява загубата на влага съхранението на пробите при 5°C е с най-добри
или поемането й. Те защитават храната от резултати относно поддържането на качеството
физически наранявания и ограничават и удължаване на срока на годност на мандарини.
миграцията на мазнини, осигуряват бариера за Разработеното покритие оказва благоприятно
газове, спомагат за запазване на вкуса и аромата, влияние върху честотата на гниене, титруемата
могат да съдържат антиоксиданти или киселинност, разтворими твърди вещества,
антимикробни вещества [16]. Восъчните аскорбинова киселина и общата сензорна
покрития намаляват загубата на влага и забавят приемливост пред контролната проба. Въз
дишането на покритите пресни плодове и основа на получените резултати е доказано, че
зеленчуци, в резултат на което се удължава 12%-но восъчно покритие и съхранение при
съхраняемостта им [5]. температура 5 °С най-ефективно поддържа
В проведено изследване е доказано, че качеството и увеличава срока на годност на
запазването на качеството на някои (Generos, мандарини [19].
Starkrimson, Ionagold и Idared) сортове ябълки е Авторски колектив изследва влиянието
възможно чрез нанасяне върху кожицата им на на парафиново покритие (25, 50, 75, 100% -ен
покритие от пчелен восък с концентрация от 1% разтвор и контролни проби) върху качеството на
[17]. краставици при различни условия на съхранение
Ядивно покритие от пчелен восък за (5, 10 и 15 °С). Изследвани са показателите
храни е ефективна бариера за кислород, промяна на диаметъра и дължината, загуба на
светлина и пара, която може да помогне за тегло, напълно разтворими твърди вещества,
предотвратяване на окислението на мазнини, твърдост. Установено е че срокът на годност се
пигменти и загубата на вода. Изследвани са увеличава с повишаване концентрацията на
количествата на 2-тиобарбитурова киселина и парафина и намалява с нарастване на
водната активност (Aw) на италиански салам температурата на съхранение.
покрит с пчелен восък. Получените резултатите Използваният парафин е восъкоподобно
са корелирани с определяне на качеството чрез твърдо вещество с бял цвят, без вкус и мирис, с
сензорна оценка. Нивата на на 2-тиобарбитурова точка на топене между около 46 и 68 °С (115 и
киселина в италианския салам с нанесено 154 °F), и с плътност от около 0.9 g/cm3. Той е
ядивно покритие с пчелен восък са под 0.8 неразтворим във вода, но разтворими в етер,
mg/kg-1 до 6 месеца и значително по-високи в бензен, и някои естери.
непокритата контролна проба. Наблюдава се Получените резултати доказват, че
малка корелация между количеството на 2- промените по показателите промяната на
тиобарбитурова киселина и водната активност диаметъра и дължината, загубата на тегло,
(Aw) при колбасите с покритие от пчелен восък. твърдостта, напълно разтворими твърди
След 7 месеца от съхранението на колбаса се вещества намаляват с увеличаване на

II-382
концентрацията на парафиновия разтвор, но се подобна на восък твърда маса. Повечето липиди
увеличават с повишаване на температурата на в твърдо състояние се удължават с около 102%
съхранение. При срокът на годност тази от първоначалната им дължина преди да се
тенденция се запазва [20]. счупят. Ацетилираният глицерол моностеарат
A. Purvis нанася различни восъци може да се разтяга до 800% от първоначалната
(повърхностно базирани восъци, восъци от си дължина. Водопропускливостта на това
натурални продукти, карнаубов восък, восъчна покритите е много по-ниска от тази на
емулсия) по повърхността на краставици, които полизахаридните покрития с изключение на
са съхранявани при два температурни режима 5 целулозата и етилцелулозата. Покритията от
и 15°С. Получените резултати доказват, че ацетилирани моноглицериди се използват за
восъчните покрития забавят загубата на влага и забавяне загубата на влагата по време на
промените при охлаждане на краставиците. Това съхранението [7].
се запазва с повишаване концентрацията на Ядивните покрития на основата на
восъка, а при намаляването й се наблюдава ацетилирани моноглицериди намират
увеличаване изпарението на влага и вредите от приложение, нанесени върху различни видове
охлаждане. Съхранението на краставици при месо [24], като изследвания на тези покрития
ниски температури води до появата на върху пресни плодове и зеленчуци не са
пукнатини във восъците, което повишава достатъчно изучени.
загубата на влага [21].
Авторски колектив установява, че 3.3. Покрития от смоли
покритие от канделилов восък с елагова Шеллакът е пречистен продукт на лак от
киселина подобрява качеството и удължава естествени смолисти олигомери, който се
срока на съхранение на ябълки в продължение секретира от паразитното насекомо „Kerria
на 8 седмици, без да се променят lacca” на различни дървесни видове в Индия и
органолептичните свойства на плодовете. Тайланд. Съставът му варира в зависимост от
Покритието доказва, че притежава вида на насекомите и дървото, от което се
противогъбични свойства, които инхибират получава суровината [25].
растежа на определени гъбични щамове [22]. Ябълки, покрити с шеллак са с по-добър
В друго изследване плодовете и гланц и твърдост. Шеллакови покрития или в
зеленчуците са представени като бързо съчетание с канделилов восък намаляват
развалящи се селскостопански стоки с висока пропускливостта на O2 и CO2, с което се забавя
хранителна стойност. Акцентирано е на дишането и естествените загуби (фири) [26].
ядивните покрития и възможностите им да Ядивно покритие от шеллак с добавен
удължат срока на годност и поддържат гел от алое вера се използва за удължаване срока
качеството на пресни плодове и зеленчуци чрез на съхранение на домати в условия на
създаване на модифицирана атмосфера във естествена атмосфера (28±2°C). Включването на
вътрешността на плода поради техните гел от алое вера в покритието подобрява
бариерни свойства за газове и влага. Разгледани пропускливостта за кислород, въглероден
са възможностите на покривните материали да диоксид и водни пари. Ядивното покритие,
бъдат носители на добавки за храни, което е нанесено върху повърхността на домати
антиоксиданти и/или антимикробни агенти. забавя промените при съхранение, твърдостта,
В същото проучване е установено, че ограничава дишането и синтеза на етилен [27].
карнаубовото покритие с низин може да В друго изследване е доказано, че
попречи на много микроорганизми, като покритите с арабска гума (20%-ен разтвор)
удължава срока на годност на продуктите. краставици значително забавят естествените
Доказано, е че покритието подобрява загуби (фири) и промените в консистенцията до
външния вид на продуктите, цвят, крехкост, 16 дни при температури на съхранение 10 и
вкус, хранителна стойност, сочност, текстура. 25°С. Сензорната оценка по показателите вкус,
То може да се използва за пресни плодове и цвят, консистенция, външен вид и приемливата
зеленчуци, месни продукти, шоколади, различни оценка от потребителите доказва, че покритието
ядки, вафли, пържени картофи, пиле, риба, от арабска гума запазва качеството на
царевичен чипс, ябълка, и лимон [23]. краставиците по време на съхранението [28].
3.2. Ацетилглицеридни покрития 3.4. Покрития от животински мазнини

Ацетилираните моноглицериди Ядивно покритие от разтопени мазнини
притежават уникални свойства. Те могат да (говежда лой, свинска мас) нанесено на
преминават от разтопено състояние в гъвкава повърхността на прясно нарязани меса, постига
II-383
по-добри резултати от контролните проби без Polish journal of food and nutrition sciences,
покритие по отношение на цвета и загубата на Vol. 57, No. 4, pp. 399 - 404.
влага по време на съхранение при температура 7. Bourtoom, T. (2008). Edible films and coatings:
от 2 до 4 °С. Замразени меса, домашни птици и characteristics and properties. International
риба не претърпят значителна дехидратация food research journal, 15, (3), pp. 237 - 248.
когато са покрити в маслено водна емулсия, 8. Kester, J. and Fennema О. (1986). Edible films
приготвена при температура от 60 до 80 ° С чрез and coatings: a review. Food technology, 40,
смесване на животински мазнини или (12), pp. 47 - 59.
растително масло с емулгатори, вода, подправки 9. Falguera, V., Quintero J., Jimenez A., Munoz
и консерванти. J., Ibarz. A. (2011). Edible films and coatings:
Значително намаляване на загубата на влага се structures, active functions and trends in their
наблюдава при лиофилизирано месо с нанесено use. Trends in food science & tehnology, 22,
покритие съставено от говежда лой, свинска pp. 292 - 303.
мас, с млечни триглицериди и растително масло 10. Maftoonazad, N., Badii F. (2009). Use of edible
[15]. films and coatings to extend the shelf life of
food products. Recent, patents on food,
4. Заключение nutrition & agriculture, Volume 1, Number 2,
Въз основа на проучената литература се pp. 162 - 170.
установи, че значително по-голямата част от 11. Milovanovic, М., Picuric - Jovanovic K. (2001).
научните изследвания за липидни покрития са Lipids and biopackaging. Usage of lipids in
насочени към приложението им за пресни edible films. Joumal of agricultural sciences,
плодове и зеленчуци. Някои от липидните Vol. 46, No 1, pp. 79 - 87.
покрития са нанесени на повърхността на месни 12. Hagenmaier, R., Shaw P. (1992). Gas
продукти. В проучването се установи, че най- permeability of fruit coating waxes. Journal of
използваният липиден компонент е пчелният the american society for horticultural science,
восък. 117, (1), pp. 105 - 109.
Бъдещите изследвания на липидните 13. Debeaufort, F., Quezada-Gallo J., Voilley A.
покрития могат да се извършват в две основни (1998). Edible Films and Coatings: Tomorrow’s
направления: Packagings: A Review. Critical reviews in food
 научни изследвания насочени към science, 38, (4), pp. 299 - 313.
нанасяне на липидни покрития върху 14. Rojas - Molina, R., León-Zapata M., Saucedo -
неизследвани до момента пресни Pompa S., Aguilar-Gonzalez M., Aguilar C.
плодове и зеленчуци; (2013). Chemical and structural characterization
 научни изследвания насочени към of Candelilla (Euphorbia antisyphilitica Zucc.).
нанасяне на неизследвани до момента Journal of medicinal plants research, Vol. 7,
месни продукти, хлебни и тестени (12), pp. 702 - 705.
изделия, захарни изделия. 15. Khan M. I., Adrees M. N., Tariq M. R. and
Sohaib M. (2013). Application of edible coating
ЛИТЕРАТУРА for improving meat quality: A review. Pakistan
journal of food sciences, Volume 23, Issue 2,
1. Pavlath, А., and Orts W. (2009). Edible films page(s): 71-79.
and coatings for food applications, pp. 1 – 24. 16. Maftoonazad, N., Ramaswamy H., Marcotte M.
Publisher: Springer, New York, USA. (2007). Evalution of factors affecting barrier,
2. Ben, A., Kurth L. (1995). Edible film coatings mechanical and optical properties of pectin -
for meat cuts and primals, Meat ’95. CSIRO based films using response surface
Meat industry research conference, pp. 13 - 15. methodology. Journal of food process
3. Park, H. (1999). Development of advanced engineering, Vol. 30, Issue 5, pp. 539 - 563.
edible coatings for fruits. Trends in food science 17. Anghel, R. (2011). The influence of wax
& technology, 10, pp. 254 - 260. protection film on apple fruits, in order to
4. Muranyi, P. (2013). Functional Edible Coatings maintain their quality during cold storage.
for Fresh Food Products. Food processing & Lucrări ştiinţifice - vol. 54, Nr. 1, seria
technology, Volume 4, Issue 1, pp. 1 - 2. Agronomie, 87 - 90.
5. Hаn, J., Gennadios A. (2005). Innovations in 18. Trevisani, M., Cecchini, M., Siconolfi, D.
food packaging, pp. 239 – 262. Publisher: Mancusi, R. and Rosmini R. (2017). Effects of
Elsevier science & technology books. Beeswax Coating on the Oxidative Stability of
6. Kokoszka, S., Lenart A. (2007). Edible coatings Long-Ripened Italian Salami. Journal of food
- formation, characteristics and use - a review. quality, pp. 5.

II-384
19. Hassan, Z. H., Lesmayati, S., Qomariah, R. and технологии, Издателска къща „ Сафо ” -
Hasbianto, A. (2014). Effects of wax coating Ловеч, с. 65 - 90.
applications and storage temperatures on the 25. Farag, Y., Leopold C. (2009). Physicochemical
quality of tangerine citrus (Citrus reticulata) Properties of Various Shellac Types,
var. Siam Banjar. International food research Dissolution technologies, pp. 33 - 39.
journal, 21(2):641-648. 26. Alleyne, V., Hagenmaier R. (2000). Candelilla -
20. Bahnasawy A. H. and Khater E. G. (2014). shellac: an alternative formulation for coating
Effect of wax coating on the quality of apples. Horticultural science, 35, (4), pp. 691 -
cucumber fruits during storage. Food 693.
processing & technology, 5 (6):339-346. 27. Chauhan, O., Nanjappa C., Ashok N., Ravi N.,
21. Purvis, A. (1994). Interaction of waxes and Roopa N., Raju P. S. (2013). Shellac and Aloe
temperature in retarding moisture loss from and vera gel based surface coating for shelf life
chilling injury of cucumber fruit during storage. extension of tomatoes. Journal of food science
Proceedings of the Florida state horticultural and technology, 52(2): 1200–1205.
Society, 107, pp. 257 - 260. 28. Al - Juhaimi, F., Ghafoor K., Babiker E. (2012).
22. Ochoa, E., Saucedo - Pompa S., Rojas - Molina Effect of gum arabic edible coating on weight
R., Garza H., Charles - Rodríguez A., Noé loss, firmness and sensory characteristics of
Aguilar C. (2011). Evaluation of a candelilla cucumber (Cucumis sativus L.) fruit during
wax -based edible coating to prolong the shelf - storage. Pakistan journal of botany, 44, (4), pp.
life quality and safety of apples. American 1439 - 1444.
journal of agricultural and biological sciences,
6, (1), pp. 92 - 98.
23. Puttalingamma.V. (2014). Edible coatings of Контакти:
carnauba wax –– a novel method for Име на автора:
preservation and extending longevity of fruits Радослав Радев
and vegetables - a review. Internet journal of Организация: ИУ-Варна
food safety, Vol.16, pp. 1-5. Адрес: Варна 9002 бул. „Княз Борис I-
24. Пашова, С., (2011). Приложение на ви“ 77
растителните восъци в хранителните Телефон: 0889/126 623
E-mail: r.radev_ue@mail.bg
II-385

NATURAL SCIENCES
Природни науки
QUADRATURE ITERATIVE METHOD FOR A

NUMERICAL SOLUTION OF A NONLINEAR
HAMMERSTEIN FUZZY FUNCTIONAL
INTEGRAL EQUATION
ALBENA PAVLOVA
Abstract: In this paper, we obtain error estimation of the iterative method using a fuzzy
trapezoidal quadrature formula to solve nonlinear Hammerstein fuzzy functional integral
equations. Error estimation of the proposed method is given in terms of uniform and partial
modulus of continuity. Finally, an illustrative numerical experiment confirms the
theoretical results and demonstrates the accuracy of the method.
Key words: Hammerstein fuzzy functional integral equations, approximation solution,
fuzzy trapezoidal quadrature formula, modulus of continuity.
1. Introduction
The study of fuzzy integral equations begins with the investigations of Kaleva [1] and Seikkala [2].
The numerical methods for solving fuzzy integral equations involve various techniques. The method of
successive approximations and iterative methods are applied in [3,4].
In this paper, we investigate the nonlinear Hammerstein fuzzy functional integral equation
b
x(t )  g (t )  f t , x(t )  ( FR)  k (t , s)  H s, x(s ) ds , a  s  t  b , (1)

a
where k (t, s ) is a positive kernel function for s, t   a, b  . The fuzzy-number-valued functions g , f , H such
that g :  a, b   E1 and f , H :  a, b E1  E1 are supposed to be continuous.
The convergence of the iterative numerical method proposed in [5] is based on error estimation in
approximation of the solution of (1) that was derived using supplementary Lipschitz continuous for g , f , k
and H .
The error estimate obtained in this paper is expressed in terms of the modulus of continuity for g
and k .
2. Preliminaries
Definition 1. [6] A fuzzy number is a function u :   [0,1] satisfying the following properties
1. u is upper semicontinuouson  ,
2. u ( x )  0 outside of some interval c, d  ,
3. there are the real numbers a and b with c  a  b  d , such that u is increasing on c, a  ,
decreasing on and for each b, d  , and u(x )  1 for each x  [a, b ] ,
4. u is fuzzy convex set ( that is ( x  (1  ) y )  min{u ( x), u ( y)} , for all x, y  ,   [0,1] ).

II-387
The set of all fuzzy numbers is denoted by E1 . The neutral element with respect to in E1 is denoted
by 0  {0} . For any 0  r  1 an arbitrary fuzzy number is represented, in parametric form, by an ordered
pair of functions (u (r ), u ( r )) , which satisfies the following properties
1. u ( r ) is bounded left continuous non-decreasing function over [0,1] ,
2. u ( r ) is bounded left continuous non-increasing function over [0,1] ,
3. u ( r )  u ( r ) .
Definition 2. [7] For arbitrary fuzzy numbers u  (u (r ), u ( r )) , v  ( v (r ), v ( r )) , the quantity

D(u , v)  sup max{| u ( r )  v (r ) |,| u ( r )  v ( r ) |} is the distance between u and v and also the following
r [0,1]
properties hold
1. ( E1 , D) is a complete metric space,
2. D(u  w, v  w)  D(u , v) , for all u , v , w  E1 ,
3. D(u  v, w  e)  D (u, w)  D (v, e) , for all u , v, w, e  E 1 ,
  D(u ,0)
4. D(u  v,0)   D(v,0)  , for all u , v  E1 ,
5. D( k  u , k  v ) | k | D (u, v) , for all u , v  E1 , for all k   ,
6. D( k1  u, k2  u ) | k1  k2 | D(u ,0) , for all k , k   with k k  0 and u  E1 .
1 2 1 2
Definition 3. [8] A fuzzy-number-valued function f :  a, b  E1 is said to be continuous at

t0   a, b  if for each   0 there is   0 such that D ( f (t ), f ( s ))   , whenever t   a, b and t  s   .
We say that f is fuzzy continuous on  a, b if f be continuous for each t0   a, b  , and denote the space of
all such functions by C ( a, b, E1 ) .
On the set C ( a, b , E1 )   f : a, b  E1 ; f is continuous  , we define
 is denoted by  .  . It is obvious that
D ( f , g )  sup D  f (t ), g (t ) , for all f , g  C (a, b , E1 ) and D (.,0) 
t  a , b 
(C ( a, b, E ), D ) is a complete metric space.

1
Definition 4. [4] Let f :  a, b  E1 , be a bounded mapping, then the

function  a ,b  ( f ,  )  sup  D( f (t ), f (t0 )) : t , t0   a, b, t  t0   is called the modulus of oscillation of f on
 a, b . In addition if f  C (a, b , E 1 ) , then  a ,b  ( f ,  ) is called uniform modulus of continuity of f .
According to [4] the following properties hold
1. D( f (t ), f ( s ))   a ,b  ( f , t  s ) for any t , s   a, b  ,
2.  a ,b  ( f ,  ) is a non-decreasing mapping in  ,
3.  a ,b  ( f , 0)  0 ,
4.  a ,b  ( f , 1  2 )   a ,b  ( f , 1 )   a , b ( f , 2 ) for any 1 , 2  0 ,
5.  a ,b  ( f , n )  n a ,b  ( f ,  ) for any   0 and n   ,
6.  a ,b  ( f ,  )  (  1) a ,b  ( f ,  ) for any  ,  0 ,
7. If [c, d ]  [ a, b] , then [ c ,d ] ( f ,  )  [ a ,b ] ( f ,  ) for all   0 .
We consider the integral equation (1) and assume that k is continuous and therefore it is uniformly
continuous and there exists N K  0 such that N K  max  k (t , s ) : t , s  a, b  .
We introduce the following conditions:
(i) g  C ( a, b, E1 ) , f  C ( a, b E1 , E1 ) , H  C ( a, b  E1 , E1 ) ;
(ii) there exist  f ,  f  0 such that D  f (t1 , u ), f (t 2 , v)   f t1  t 2   f D (u , v ) for all
t1 , t2  a, b  , u , v  E1 ;
II-388
(iii) there exist H ,  H  0 such that D  H (t1 , u ), H (t2 , v)   H t1  t 2  H D(u , v ) for all
t1 , t2  a, b  , u , v  E1 ;
(iv) B   f  N K  H  1 , where   b  a .
Theorem 1. [5] Let the conditions (i)-(iv) are fulfilled. Then the integral equation (1) has unique
solution x  C (a, b , E1 ) and the sequence of successive approximations  xm m  C ( a, b , E 1 )
b
xm (t )  g (t )  f t , xm1 (t )  ( FR)  k (t , s )  H  s, xm1 ( s) ds , t   a, b , m   (2)

a
convergences to x in C ( a, b, E1 ) for any choice of x0  C (a, b, E1 ) . In addition, the following error
estimates hold
Bm
D( x (t ), xm (t ))  D( x1 (t ), x0 (t )) for all t   a, b , m   (3)
1 B
B
D( x (t ), xm (t ))  D( xm (t ), xm1 (t )) for all t   a, b , m   . (4)
1 B
Choosing x0  C (a, b, E1 ) , x0  g the inequality (3) becomes
Bm
M 0  N0 N K (b  a ) for all t   a, b , m   ,
D( x (t ), xm (t ))  (5)
1 B
  M and D( H (t , g (t )),0)
where D( f (t , g (t )),0)  N .
0 0
  and  H   sup D( H (t ,0),

Remark 1. Let  f   sup D ( f (t , 0),0)  0) then the inequality (5) has the form

t  a ,b  t  a ,b 
m
B
D( x (t ), xm (t ))   B  g    f   N K   H   for all t   a, b , m   . (6)
1 B
Theorem 2. [4] Let f :  a, b  E1 be a continuous fuzzy –number-valued function. Then

 b
ba  b  a  b  a 
D ( FR )  f (t ) dt ,   f ( a)  f (b)   a ,b   f , .
 a
2  2  2 
3. Successive approximations and the iterative algorithm

Now, we introduce a numerical method for solving the fuzzy integral equation (1). We consider the
uniform partition of the interval a, b  , a  t0  t1  ...  t n1  tn  b with ti  a  ih , i  0, n , where
ba
h . Then the following iterative procedure gives the approximate solution of equation (1) on the
n
point t
x0 (t )  g (t ),
n1
h , (7)
xm (t )  g (t )  f t , xm1 (t )     k (t , t j )  H t j , xm1 (t j )  k (t , t j 1 )  H t j 1 , xm1 (t j 1 )
j 0 2
 
for m  1 .
Lemma 1. Under the conditions (i)-(iv) we have

a)  a ,b  ( H ( s , xm ( s ), h)   H h   H  a ,b ( xm1 ( s ), h ) ,
1 f  H   B m  g  H M 
b)  a ,b  ( xm ( s), h)   a ,b  ( g , h)  h    
 H  1 ,
1  f 1  f 1   f B f (1 B )(1   f ) 
where M  g    f   N K   H  and 1   a ,b ( k ,  )  sup {| k (t1 , s)  k (t2 , s ) |:| t1  t 2 | }
t1 , t2  a ,b 
for all   0 .
II-389
Proof. a) For s1 , s2   a, b  with s1  s2  h and (iii)

D ( H ( s1 , xm ( s1 )), H ( s2 , xm ( s2 )))   H | s1  s2 | H D ( xm ( s1 ), xm ( s2 ))   H h  H  a , b ( xm , h) .
b) For t1 , t2  a, b  with t1  t2  h , by using Theorem 2 and (2) we obtain
D( xm (t1 ), xm (t2 ))  D ( g (t1 ), g (t 2 ))  D( f (t1 , xm1 (t1 )), f (t 2 , xm1 (t 2 ))) 
 b b 

D ( FR)  k (t1 , s )  H ( s, xm1 ( s) ds,( FR)  k (t2 , s)  H ( s, xm1 ( s ) ds  
 a a 
b
 a ,b ( g , h)   f h   f D( xm1 (t1 ), xm1 (t2 ))   | k (t1 , s)  k (t2 , s ) | D  H ( s, xm1 ( s )),0 ds 
a
b

  a ,b ( g , h)   f h   f D( xm1 (t1 ), xm1 (t2 ))  1  D  H ( s, xm1 ( s)), H ( s,0)
   D  H ( s,0),0
  ds  
a
 a ,b ( g , h)   f h   f D( xm1 (t1 ), xm1 (t2 ))  1H  xm1    H   

  f D( xm1 (t1 ), xm1 (t2 ))  1H  xm1   P , where P   a , b ( g , h )   f h  1  H  .
So, we have
D( xm (t1 ), xm (t2 ))   f D ( xm1 (t1 ), xm1 (t2 ))  1 H  xm1   P
D( xm1 (t1 ), xm1 (t2 ))   f D( xm2 (t1 ), xm2 (t2 ))  1H  xm2   P
...
D( x1 (t1 ), x1 (t2 ))   f D ( g (t1 ), g (t2 ))  1 H  g   P .
Multiplying these inequalities by 1 ,  f ,…,  mf1 , respectively, and summing them, we have
1 f  H  (b  a)
D( xm (t1 ), xm (t2 ))   mf  a ,b  ( g , h)  h 1 
1  f 1  f 1  f . (8)
 1H  xm1   f  xm2  ...   m 2
f  x1   m1
f  g  
Now we calculate  xm1   f  xm2  ...   mf 2  x1   mf 1  g  .

 b 
   
 xm1   D( xm1 (t ),0)  D( g (t ),0)  D( f (t , xm2 (t )),0)  D ( FR )  k (t , s )  H ( s, xm2 ( s) ds,0 
 a 
 g   f  xm2    f   N K (b  a )  H  xm2    H    B  xm2  M .
So, we have
 xm1   B  xm2   M
 xm2   B  xm3  M
...
 x1   B  g   M
Multiplying these inequalities by 1 , B ,…, B m2 , respectively, and summing them, we have
M
 xm1   B m1  g   .
1 B
Hence
 xm1   f  xm2  ...   mf 2  x1   mf 1  g  
M f M
 B m1  g    B m2  f  g    ...  B mf 2  g   mf 2 M   mf 1  g  
1 B 1 B
B m1 M Bm M
  g     g   .
f (1 B )(1   f ) B   f (1 B )(1   f )
1
B
From (8) we obtain
II-390
1 f  H    Bm M 
D( xm (t1 ), xm (t2 ))   mf  a ,b ( g , h)  h 1  1H   g    .
1  f 1  f 1  f  B   f (1 B )(1   f ) 
Hence,
1 f  H   B m  g  H M 
 a ,b  ( xm ( s), h)   a ,b  ( g , h)  h    
 H  1 . □
1  f 1  f 1   f B f (1 B )(1   f ) 
Lemma 2. Under conditions (i)-(iv) we have  xm   B  xm1  M .

Proof.
  D ( g (t ),0)
 xm   D( xm (t ),0)   D( f (t , x (t )),0)  
m1
 h n1   h n1 
 D   k (t , t j )  H t , xm1 (t j ),0  D   k (t , t j 1 )  H t j 1 , xm1 (t j 1 ),0 
 2 j 0   2 j 0 
 g    f   f  xm1  N K H  xm1  N K  H   B  xm1  M . □
Theorem 3. Under conditions (i)-(iv) the iterative method (7) converges to a unique solution x of
(1), and its error estimate is as follows
Bm
D ( x  , xm )   B  g    f   N K   H   
1 B
3N K H 3N K    
  a ,b  ( g , h)   H  H f  h 
4(1  B)(1  f ) 4(1 B )  1  f 
 
 H B  M  (1  B)  H      mB m1  g   H M  (1 B )  H   ,
m1
32 N K H
  1  H
 (1  B)(1  f )  
 2
4(1  B)  B   f  (1  B) 2
where M  g    f   N K   H  , 2   a ,b  (k ,  )  sup {| k (t , s1 )  k (t , s2 ) |:| s1  s2 | } .
s1 , s2 a ,b 
Proof. Considering iterative procedure (7), for all t   a, b we have

D( xm (t ), xm (t ))  D( g (t ), g (t ))  D( f (t , xm1 (t )), f (t , xm1 (t ))) 
 b
h n1 

 D ( FR)  k (t , s)  H ( s, xm1 ( s) ds,   k (t , t j )  H t , xm1 (t j )  k (t , t j 1 )  H t j 1 , xm1 (t j 1 )  
 a
2 j 0   
  f D( xm1 (t ), xm1 (t )) 
 t j1 
n 1
 h
 D ( FR)  k (t , s)  H ( s, xm1 ( s) ds,  k (t , t j )  H t , xm1 (t j )  k (t , t j 1 )  H t j 1 , xm1 (t j 1 )   
 2  
j0  tj 
  f D( xm1 (t ), xm1 (t )) 
 t j1 
n 1
 h
 D ( FR)  k (t , s)  H ( s, xm1 ( s) ds,  k (t , s )  H t , xm1 (t j )  k (t , s)  H t j 1 , xm1 (t j 1 )   
 2  
j0  tj 
n1
h
 D   k (t , s )  H t , xm1 (t j )  k (t , s )  H t j 1 , xm1 (t j1 ) ,
j0  2  
h 
 k (t , s )  H t , xm1 (t j )  k (t , s)  H t j 1 , xm1 (t j 1 )
  
2   
n1
h
 D   k (t , s )  H t , xm1 (t j )  k (t , s )  H t j 1 , xm1 (t j1 ) ,
j0
 2  
h 
 k (t , t j )  H t , xm1 (t j )  k (t , t j 1 )  H t j 1 , xm1 (t j 1 )
  
2   

II-391
 h  
  f D( xm1 (t ), xm1 (t ))  N K  a ,b  ( H ( s, xm1 ( s)), )  N K H D ( xm1 , xm1 )  2 D( H t , xm1 (t ),0)
2 2
3N K h
 BD ( xm1 , xm1 )   a ,b ( H ( s, xm1 ( s)), )  2H  xm1  2  H  .
4 2
3N K h
We denote P   a ,b  ( H ( s, xm1 ( s)), )  2  H  .
4 2
D ( xm , xm )  BD ( xm1 , xm1 )  2 H  xm1   P
 
D ( xm1 , xm1 )  BD ( xm2 , xm2 )  2 H  xm2  P

...
D ( x1 , x1 )  2 H  g  P
Multiplying these inequalities by 1 , B ,…, B m1 , respectively, and summing them, we have
P
D ( xm , xm )  H 2  xm1  B  xm2  ...  B m1  g    .
1 B
We use Lemma 2 and get
M
 xm1   B  xm2  ...  B m1  g   mB m1  g   .
(1  B )2
Hence,
 M  1  3N K h 
D ( xm , xm )  H 2 mB m1  g       a , b ( H ( s, xm1 ( s ) ), )   2  H   .
 (1  B)  1  B  4
2
2 
From Lemma 1 we obtain
3N K  H 3N K    
D ( xm , xm )   a ,b ( g , h)   H  H f  h 
4(1  B )(1   f ) 4(1 B )  1  f 
 
 H B  M  (1  B)  H      mB m1  g   H M  (1 B )  H   .
m1
32 N K H
  1  H
 (1  B)(1  f )  
 2
4(1  B)  B   f  (1  B) 2
Considering the inequality D ( x , xm )  D ( x  , xm )  D ( xm , xm ) and inequality (6) we prove the theorem.□
REFERENCES
1. Kaleva, O. (1987). Fuzzy differential equations. Fuzzy Sets and Systems, vol. 24, pp.301-317.
2. Seikkala, S.(1987). On the fuzzy initial value problem. Fuzzy Sets and Systems, vol. 24, pp. 319-330.
3. Bica, A. M. (2008). Error estimation in the approximation of the solution of nonlinear fuzzy Fredholm
integral equations. Inf. Sci.178, pp. 1279-1292.
4. Bede, B. and Gal, S. G. (2004).Quadrature rules for integrals of fuzzy-number-valued functions. Fuzzy
Sets and Systems, vol. 145, pp.359-380.
5. Enkov, S.,Georgieva, A. and Nikolla, R. (2016).Numerical solution of nonlinear Hammerstein fuzzy
functional integral equations. AIP Conference Proceedings 1789, 030006,
http://doi.org/10.1063/1.4968452, pp. 1551-7616.
6. Goetschel, R. and Voxman W. (1986).Elementary fuzzy calculus. Fuzzy Sets and Systems, vol. 18,
pp.31-43.
7. Wu, C.and Gong, Z.(2001). On Henstock integral of fuzzy-number-valued functions. Fuzzy Sets and
Systems, vol. 120 , pp. 523-532 .
8. Congxin Wu and Gong Wu (1997). The supremum and infimum of these to fuzzy-numbers and its
applications. J. Math. Anal. Appl., 210, pp. 499-511.
Albena Pavlova
Department of MPC
Technical University-Sofia, Plovdiv Branch
25 Tzanko Djustabanov Str., 4000 Plovdiv, Bulgaria
E-mail: akosseva@gmail.com
RIEMANNIAN ALMOST PRODUCT MANIFOLDS

GENERATED BY A CIRCULANT STRUCTURE
DOBRINKA GRIBACHEVA, DIMITAR RAZPOPOV
Abstract: A 4-dimensional Riemannian manifold equipped with a circulant structure,

which is an isometry with respect to the metric and its fourth power is the identity, is
considered. The almost product manifold associated with the considered manifold is
studied. The relation between the covariant derivatives of the almost product structure and
the circulant structure is obtained. The conditions for the covariant derivative of the
circulant structure, which imply that an almost product manifold belongs to each of the
basic classes of the Staikova-Gribachev classification, are given.
Key words: Riemannian metric, circulant matrix, almost product structure.
Mathematics Subject Classification (2010). Primary 53B20, 53C15, Secondary 15B05.
1. Introduction recalled. In Sect. 2, the relation between the

The circulant matrices and the circulant covariant derivative of P and the covariant
structures have application to Vibration analysis, derivative of Q is obtained. In Sect. 3, the
Graph theory, Linear codes, Geometry (for example conditions for the covariant derivative of Q , which
[1], [2] and [3]). Riemannian manifolds equipped
with a circulant structure, whose fourth power is the imply that ( M , P, g ) belongs to each of the basic
identity were considered in [4] and [5]. In particular classes of the Staikova-Gribachev classification, are
case, such manifolds could be Riemannian almost given.
product manifolds. The systematic development of
2. Preliminaries
the theory of Riemannian manifolds M with a
metric g and an almost product structure P was Let M be a 4-dimensional Riemannian
manifold equipped with a metric g and an
started by K. Yano in [6]. In [7] A. M. Naveira
classified the almost product manifolds ( M , P, g ) endomorphism Q in the tangent space T p M at an
with respect to the covariant derivative of P . The arbitrary point p on M . Let the coordinates of Q
Riemannian almost product manifolds ( M , P, g ) with respect to some basis {ei } of T p M form the
with zero trace of the structure P were classified circulant matrix
with respect to the covariant derivative of P by
M . Staikova and K. Gribachev in [8]. The basic 0 1 0 0
classes in this classification are W1 ,W2 and W3 .  
0 0 1 0
The class W0  W1  W2  W3 was called the class Q
0 0 0 1
of Riemannian P -manifolds. Our purpose is to
 . (1.1)
1
 0 0 0 
obtain characteristic conditions for each of these
classes according to the circulant structure. In the
present paper we consider a 4-dimensional Then Q satisfies the equalities
differentiable manifold M with a Riemannian
metric g and a circulant structure Q , whose fourth Q 4  id , Q 2  id .
power is the identity and Q acts as an isometry on
Let the structure Q be compatible with the
g . This manifold we will denote by ( M , Q, g ). We
metric g , i.e.
study the Riemannian almost product manifold
( M , P, g ) where P  Q 2 . The paper is organized g (Qx , Qy )  g ( x, y ). (1.2)
as follows. In Sect. 1, some necessary facts about
considered manifolds ( M , Q, g ) and ( M , P, g ) are
II-393
Here and anywhere in this work x, y, z , u will ( x Q) y   x Qy  Q x y (1.8)

stand for arbitrary elements of the algebra of the
smooth vector fields on M or vectors in T p M . The ( x P) y   x Py  P x y (1.9)
Einstein summation convention is used, the range of
the summation indices being always 1, 2, 3, 4 . Let the structure Q of a manifold ( M , Q, g ) be the
Further, we consider a manifold ( M , Q, g ) covariant constant, i.e. ( x Q) y  0 . Then, from
equipped with a metric g and a structure Q , which (1.8) we obtain successively  x Qy  Q x y,
satisfy (1.1) and (1.2). This manifold is studied in 2
 x Q y  Q x Qy  Q  x y, 2
thus we get
[4] and [5].
We denote P  Q 2 . In [4] it is noted that  x Py  P x y.
the manifold ( M , P, g ) is a Riemannian manifold Therefore, from (1.9) it follows
( x P) y  0 .
with an almost product structure P , because
By using the latter equality and (1.3) we find (1.4).
P 2  id , P  id and g ( Px, Py)  g ( x, y ).
Hence the next theorem is valid.
Moreover trP  0 . For such manifolds is valid the Theorem 1.1. If the structure Q of the
Staikova-Gribachev classication given in [8]. This
classification was made with respect to the tensor manifold ( M , Q, g ) satisfies Q  0 , then
F of type 0,3 and the Lee form  , which are ( M , P, g ) belongs to the class W0 .
defined by As it is known the curvature tensor R of 
is determined by
F ( x, y, z )  g (( x P) y, z),
R( x, y ) z   x  y z   y  x z  [ x , y ] z.
 ( x)  g ij F (ei , e j , x). (1.3)
The corresponding tensor of type (0,4) is defined
Here  is the Levi-Civita connection of g , and
as follows R( x, y, z, u )  g ( R( x, y ) z, u ).
g ij are the components of the inverse matrix of g
Proposition 1.2. [5] If the structure Q of
with respect to {ei } .
the manifold ( M , Q, g ) satisfies Q  0 , then for
The basic classes of the Staikova-Gribachev
the curvature tensor R it is valid
classification are W1 ,W2 and W3 . Their intersection
is the class of Riemannian P -manifolds W0 . A R( x, y, Qz , Qu )  R( x, y, z, u ).
manifold ( M , P, g ) belongs to each of these
We substitute Qz for z and Qu for u in
classes if it satisfies the following conditions:
the latter equality, and using Theorem 1.1, we
W0 : F ( x, y, z )  0, (1.4) obtain
Corollary 1.3. If the manifold ( M , P, g )
1 belongs to W0 , then the curvature tensor R
W1 : F ( x, y, z )  (( g ( x, y ) ( z ) satisfies,
4
 g ( x, z ) ( y )  g ( x, Py) ( Pz ) (1.5) R( x, y, Pz, Pu)  R( x, y, z, u ).
 g ( x, Pz) ( Py)),
i.e. R is a Riemannian P -tensor.
W2 : F ( x, y, Pz)  F ( y, z, Px)  F ( z, x, Py)  0, 3. Relation between F and F

 ( z )  0, (1.6) We consider manifolds ( M , Q, g ) and
( M , P, g ) , where P  Q 2 . We define a tensor F
W3 : F ( x, y , z )  F ( y , z , x ) of type (0,3) , as follows
(1.7)
 F ( z, x, y)  0.
F ( x, y, z)  g (( xQ) y, z),
It is well known that  satisfies the
equalities:  ( x)  g ij F (ei , e j , x), (2.1)
II-394
where  is the Lee form associated to F . Proof. Let ( M , P, g ) W0 , i.e. F  0 .

Theorem 2.1. For the tensors F on the Then, due to (2.2), it follows
manifold ( M , P, g ) and F on the manifold
( M , Q, g ) the following equalities are valid: F ( x, Qy , Qz )   F ( x, y, z ).
The latter equality and (2.1) imply (2.7).

F ( x, y, z )  F ( x, Qy , Qz )  F ( x, y, Qz ), (2.2) Vice versa. According to (2.1) and (2.7) we
find F ( x, y, z )  F ( x, Qy , Qz )  0.
F ( x, y, Q 3 z )  F ( x, Qy , z )  F ( x, y, z ). (2.3)
Then, due to (2.2), it follows F  0 , i.e. the
Proof. From (1.3) and (1.9), due to manifold ( M , P, g )  W0 .
2
P  Q , we get
4. Properties of F
F ( x, y, z )  g ( x Py  P x y, z ) Theorem 3.1. For the tensor F on
( M , Q, g ) the following equalities are valid:
 g ( x Q 2 y  Q 2  x y, z ),
2 2 F ( x, y, Q 3 z )  F ( x, Qy , z )
i.e. F ( x, y, z )  g ( x Q y  Q  x y, z ). (3.1)
Then  F ( x, z, Q 3 y )  F ( x, Qz , y ),
F ( x, y, Qz )  g ( x Q 2 y, Qz )  g (Q 2  x y, Qz ), F ( x, y, z )  F ( x, Qy , Qz )
(3.2)
from which, because of (1.2) we have  F ( x, Q 2 y, Q 2 z )  F ( x, Q 3 y, Q 3 z )  0,
F ( x, y, Qz )  g ( x Q 2 y, Qz ) F ( x, y, Qz )   F ( x, z, Qy ), (3.3)
(2.4)
 g (Q x y, z ),
F ( x, y, Q 3 z )   F ( x, Q 2 z , Qy ). (3.4)
From (1.8) and (2.1) we obtain
Proof. It is known that the tensor F
F ( x, y, z )  g ( x Qy , z ) determined by (1.3) has the properties:
(2.5)
 g (Q x y, z ), F ( x, y, z )  F ( x, z, y), (3.5)
and consequently F ( x, Py, Pz)   F ( x, y, z ). (3.6)
F ( x, Qy , Qz )  g ( x Q 2 y, Qz ) Equalities (2.3) and (3.5) imply (3.1).

(2.6) From (2.3) and (3.6) we get
 g ( x Qy , z ).
Taking the sum of (2.5) and (2.6) we get F ( x, Qy , z )  F ( x, y, Q 3 z )

 F ( x, Q 3 y, Q 2 z )  F ( x, Q 2 y, Qz )  0.
F ( x, y, z )  F ( x, Qy , Qz )
 g ( x Q 2 y, Qz )  g (Q x y, z ). In the latter equality we substitute Qz for z , and
we obtain (3.2). Further, we substitute Qz for z
Then, having in mind (2.4), we find (2.2). Now we into (2.5) and we have
3
substitute Q z for z into (2.2) and using (1.2), we
find (2.3). Hence the next theorem is valid.
Theorem 2.2. The manifold ( M , P, g )
belongs to W0 if and only if Q satisfies
( x Q)Qy  Q( x Q) y. (2.7)

II-395
F ( x, y, Qz )  g ( x Qy , Qz )  g ( x y, z ) This work was partially supported by project FP17-

FMI-008 of the Scientific Research Fund, Paisii
 xg(Qy , Qz )  g (Qy ,  x Qz ) Hilendarski University of Plovdiv, Bulgaria.
 xg( y, z )  g ( y,  x z )
References
  g ( x Qz , Qy )  g ( x z, y ) 1. S. I. R. Costa, J. E. Strapasson, M. M. S.
  g ( x Qz , Qy )  g (Q x z , Qy ) Alves, T. B. Carlos: Circulant graphs and
tessellations on at tori, Linear Algebra Appl.,
  g ( x Qz  Q x z , Qy )   F ( x, z, Qy ). 432(1) (2010) 369-382.
2. A. Kaveh, H. Rahami: Block circulant matrices
Therefore we get (3.3). From (3.3) directly follows and application in free vibration analysis of
(3.4). Using (1.5), (2.1) and (2.2) we obtain the cyclically repetitive structures, Acta Mech.,
following 217 (2011) 51-62.
Theorem 3.2. The manifold ( M , P, g ) 3. R. M. Tanner, D. Sridhara, A. Sridharan, T.
Fuja, D. J. Costello: LDPC Block and
belongs to W1 if and only if the tensor F on convolutional codes based on circulant
( M , Q, g ) satisfies the following conditions: matrices, IEEE Trans. Inf. Theory, 50(12)
(2004) 2966-2984
F ( x, y, Q 3 z )  F ( x, Qy , z ) 4. I. Dokuzova: Curvature properties of 4-
dimensional Riemannian manifolds with a
1 circulant structure. J. Geom. (2016).
 ( g ( x, y ) ( z )  g ( x, z ) ( y )
4 doi:10.1007/s00022-016-0356-9
 g ( x, Q 2 y ) (Q 2 z )  g ( x, Q 2 z ) (Q 2 y ), 5. D. Razpopov: Four-dimensional Riemannian
manifolds with two circulant structures, In:
Proc. of 44-th Spring Conf. of UBM, SOK
 (Q 3 z )  g ij F (ei , Qe j , x)   ( z ). "Kamchia", Bulgaria (2015) 179-185.
6. K. Yano: Differential geometry on complex
We apply (2.2) and (2.3) into (1.6) and we and almost complex spaces. Pergamont Press,
find Oxford, (1965)
7. A. M. Naveira: A classification of Riemannian
F ( x, y, Pz)  F ( x, y, Qz )  F ( x, Qy , Q z ). 2 almost product manifolds, Rend. Math. 3
(1983) 577-592.
Therefore we arrive at the following 8. M. Staikova, K. Gribachev: Cannonical
Theorem 3.3. The manifold ( M , P, g ) connections and conformal invariantson
Riemannian almost product manifolds. Serdica
belongs to W2 if and only if the tensor F on Math. J. 18 (1992) 150-161.
( M , Q, g ) satisfies the following condition
Dobrinka Gribacheva
F ( x, y, Qz )  F ( x, Qy , Q 2 z ) Department of Algebra and Geometry
Faculty of Mathematics and Informatics,
 F ( y, z , Qx )  F ( y, Qz , Q 2 x)
University of Plovdiv Paisii Hilendarski
 F ( z, x, Qy )  F ( z, Qx , Q 2 y )  0. 24 Tzar Asen, 4000 Plovdiv, Bulgaria
E-mail address: dobrinka@uni-plovdiv.bg
We apply (2.3) into (1.7) and we have
Theorem 3.4. The manifold ( M , P, g ) Dimitar Razpopov
Department of Mathematics, Informatics
belongs to W3 if and only if the tensor F on and Physics Faculty of Economics,
( M , Q, g ) satisfies the following condition Agriculture University Plovdiv
12 Mendeleev blvd,
4000 Plovdiv, Bulgaria
F ( x, y, Q 3 z )  F ( x, Qy , z )
E-mail address: razpopov@au-plovdiv.bg
 F ( y, z, Q 3 x)  F ( y, Qz , x)
 F ( z, x, Q 3 y )  F ( z, Qx , y )  0.
VARIATIONAL METHOD FOR A CLASS OF

HIGHER ORDER HYPERBOLIC EQUATIONS
G.P.PASKALEV
Abstract. For considered mixed problem for a higher order hyperbolic equation a
Shalov-type symmetrizing operator[4] is build. Equivalence of the problem to the problem
of minimization of quadratic functional is proved. Existence and uniqueness of the
generalized solution are obtained.
Key words: higher order equation, mixed problem, variatonal method, Seeley’s extension
formula, anisotropic space.
1. Introduction ~
(2). Define the space H 2 n  1, n (GT ) as a closure of
~
Let D  R m , m  1 be a bounded domain C  (GT ) with respect to the norm
with a boundaryD and x  ( x1 ,..., xm ),
 (  D
2
u  j
Dx u ) 2 dtdx. (3)
0  D  {0},   D  (0, T ), GT  D  (0, T ), 2 n 1, n ,T
GT j
t
 1
T  0, n  2 ,  '  (1 ,...,  n ). 2 n 1 n
Let G2T  D  (0,2T ) and let by definition

 a ( ' )  ( ' )   C1  ( ' ) 2  ',
~

  n 1 '
the space C  (G2T ) be composed of extensions of
C1  const.  0 , a are real constants and
~
the functions, belonging to C  (GT ) , obtained by
a  a   ,  :     n .
Let A  0, C  0. the linear Seeley’s extension formula [6]. Then
Consider in GT the following problem. To there exists a positive constant C1 , non depending
find a solution to the equation
on u , such that
Lu  f (t, x), (1) ~
where U 2 n 1, n , 2T
 C1 u 2 n 1, n ,T
u  C  (GT ) ,
where U is the extension, corresponding to u.

Lu   a D  D  u  D
x x t
2n
u  ADt u  Cu ,
 
 n ~ ~
Define H 2 n 1,n (G2T ) as a closure of C  (G2T ) with
satisfying the boundary conditions
respect to the norm (3). Denote by Lref
2 (G2T ) the
~
D x u   0,   n  1; Dtl u closure of C  (G2T ) by the L2  norm and by F
0  0, l  0,2n  1. (2)
2.Function spaces and definition the corresponding extension to f .

~
Let C  (GT ) be the space of infinitely smooth
If   (1 ,...,  n1 ),  '  (1 ,...,  n ) , let
in GT functions, satisfying the boundary conditions
we enter the operator
Copyright©2017
University - Sofia,
of Sofia, Plovdiv
Plovdiv branch,
branch, Bulgaria
II-397
n 1
n
 x j j t n 1 )
i( which means that L also is K  positive.

KV  (2 ) 2
 B ( ).
e j 1
Vˆ ( )d ,
~
R n 1 Definition: The function u  H 2 n1,n (GT ) is called
a generalized solution to the problem (1),(2) if there
(i) j 
( ' )
2
( n 1 ) 2 j
2 n 1, n ~
j
 1 exists an element U  H (G2T ), U  u,
B( )  2 n 1 n
, (4) GT

 
a ( )  ( )   C  iA
 н
' '
n 1 such that
~
~ ( F , KV ) 0, 2T  [U ,V ]2n1,n, 2T V  H 2n1,n (G2T ).
where V  C0 ( R n1 ), V G2T
 C  (G2T ),
and Vˆ ( ) is the Fourier transform of V .

3.Main results
We note that for obtained extensions and
~
Theorem 1: The function u  H 2 n 1,n (GT ) is a
some positive constant we have
generalized solution to the problem (1),(2) if and
~
V  C2 . V V  C  (G2T ).
2 n 1, n , R n 1
2 n 1, n , 2T only if the corresponding element U (from above
~ Definition ) realizes the minimum of the quadratic
Now for U ,V  C  (G2T ) ,by integration
functional
by parts from (4) we have
2
D[U ]  U 2 n 1.n , 2T
 2( F , KU ) 0, 2T (5)
 (   a D
 
( LU , KV ) 0, 2T  x Dx U  ~
G2 T  n on the space H 2 n 1,n (G2T ) .
Dt2 nU  AU  CU ) KVdtdx 
Theorem 2: For each function F  Lref

2 (G2t )
 U [   a D

x Dx KV  Dt2 n KV  ADt ( KV ) 
G2 T  n
there exists a unique solution to the variational
C ( KV )]dtdx    ( D t
j
D x U .Dt j D x V )dtdx. problem to minimize the quadratic functional (5) on
G2 T j ~

2 n 1 n
1
the space H 2 n 1,n (G2T ) . The considered problem
Hence is correctly posed - to the small variance of the
~ function F in Lref
2 (G2T ) corresponds a small
( LU , KV ) 0, 2T  [U , V ]2n1,n, 2T U , V  C  (G2T ). ~
variance of the solution in H 2 n1,n (G2T ).
Changing the places of U and V , obtain 4. Proofs
~ In order to prove the above theorems using

( LV , KU ) 0, 2T  [U , V ]2n 1,n, 2T U , V  C  (G2T ).
the Theorem 1.3 from [7, p.367],we obtain the next
estimate.
The last two equalities mean that the operator L is
Lemma: There exists a positive constant C  0 ,
K  symmetric [4].
such that
For U  V we have ~
KV 0 , 2T
 C. V 2 n 1, n , 2T
V  C  (G2T ).
2 ~
( LU , KU ) 0, 2T  U 2 n 1, n , 2T
U  C  (G2T ), The proofs follow the scheme, used in [1-3].
II-398
For elliptic equations the variational variational method. (Russ.), Diff. Uravnenia,
approach is well known but for hyperbolic equation Minsk, V.28, no.9, P.1640-1641.
(string equation) for the first time variational
method for boundary value problem is constructed 4. Shalov, V.M.(1963) A solving of non
in [5]. For multidimensional parabolic and selfadjoint equations by the use the variational
ultraparabolic equations the considered variational method. (Russ.), Dokl. A. N. SSSR, V.151,
method is presented in [2,3].The present paper no.3, P.511-512.
generalizes the results, obtained in [1]. 5. Shalov, V.M. (1965) A principle of the
minimum of quadratic functional. (Russ.),Diff.
REFERENCES Uravnenia, Minsk, V.1,no.10,P.1338-1365.
1. Paskalev, G.P. (2013) Shalov’s variational
method for the multidimensional wave 6. Seeley, R.T. (1964) Extension of C  
equation. Journal of the Technical University functions, defined in a half space. , Proc.
at Plovdiv, “Fundamental Sciences and Amer. Math. Soc. 15, P.625-626.
Applications”, V.19, P.141-145. 7. Taylor, M. (1985) Pseudodifferential
2. Paskalev, G. P. (1999) Variational method for operators, (Russ.), Mir, Moskva.
the multidimensional heat equation. Journal
of the Technical University at Plovdiv,
“Fundamental Sciences and Applications”, Department of Mathematics,Physics and Chemistry
V.8, P.69-78 Technical University-Sofia, Plovdiv Branch
25,Tsanko Dyustabanov Str.
3. Paskalev, G.P. (1992)On the investigation of 4000 Plovdiv
boundary value problem for ultraparabolic BULGARIA
type equation with constant coeficients using a e-mail: g.p.paskalev@abv.bg

МАТЕМАТИЧЕСКИ МЕТОДИ ЗА ОЦЕНКА НА

СРЕДНИЯ ДОБИВ НА СПАНАК В
ДЪРЖАВИТЕ-ЧЛЕНКИ НА ЕВРОПЕЙСКИЯ
СЪЮЗ ЗА ПЕРИОДА 1961-2014 Г.
НЕЛИ КЕРАНОВА
Резюме: В настоящото изследване е направена математическа оценка на добива на

спанак за периода 1961-2014 г. за всички страни-членки на Европейския съюз.
Установи се, че Холандия (25,73 t/ha) и Белгия (20,8 t/ha) са страните с най-висок
добив от тази култура. България (8,64 t/ha) и Румъния (7,53 t/ha) са страни, в които
добивът от спанак е най-нисък. Средство за постигане на поставената цел е
съставяне на математически модел, чрез който да се анализират
статистическите данни и да се направи оценка на добива на спанак в отделните
държави. Направена е класификация и групиране на държавите от ЕС чрез
йерархичен клъстерен анализ според средния добив на разглежданата земеделска
култура. Получена е оценка на добива за съответния период чрез еднофакторен
дисперсионен анализ.
Ключови думи: среден добив, спанак, математически модел, йерархичен клъстер
анализ, еднофакторен дисперсионен анализ ANOVA
MATHEMATICAL METHODS FOR

ESTIMATING THE AVERAGE YIELD OF
SPINACH IN MEMBER STATES OF THE
EUROPEAN UNION FOR THE PERIOD 1961-2014
NELI KERANOVA
Abstract: This study made mathematical evaluation of the yield of spinach for the period
1961-2014 in all member states of the European Union. It was found that the Netherlands
(25,73 t/ha) and Belgium (20,8 t/ha) are the countries with the highest yield of this crop.
Bulgaria (8,64 t/ha) and Romania (7,53 t/ha) are the countries where the yield of spinach is
the lowest. The means to achieve the set goal is generating a mathematical model for
analyzing the statistical data and evaluating the yield of spinach in each country. There is a
classification and grouping of the EU countries through hierarchical cluster analysis based
on the average yield of the examined crop. There is an assessment of the yield for the
relevant period by a single-factor analysis of variance.
Key words: average yield, spinach, mathematical model, hierarchical cluster analysis,
single-factor analysis ANOVA
1. Въведение Първоначално се е отглеждал в Китай, а много

За родина на спанака се смята по-късно и в Европа – Франция, Испания,
Югозападна Азия, по-точно Персия. Той води Англия. Той е ценна зеленчукова култура,
началото си още от средните векове. разпространена и у нас. Причина за това са
качествата, които притежава: високо
II-400
съдържание на много витамини от групите А, В които ясно се очертават на дендрограмата на
и С, минерални соли, белтъчини, желязо, Фигура 1.
натрий, калий и протеини, което го превръща и Тъй като при клъстерния анализ не се
в лечебно средство при редица заболявания. правят никакви тестове за статистическа
В тази работа се анализира състоянието значимост, ще приложим еднофакторен
на производството на спанак, както в България, дисперсионен анализ.
така и в страните от Европейския съюз. Според резултатите от теста за
Периодът, който се разглежда, обхваща времето хомогенност на Ливин можем да твърдим, че
от 1961 г. до 2014 г. Основната цел на данните за добив на спанак в разглежданите
изследването е да се направи математическа държави са с равни дисперсии и същите могат
оценка, основана на статистическите данни за да бъдат сравнявани по избрания критерий.
среден добив на спанак за 28-те държави от ЕС. Изчислената стойност на критерия на Фишер е
62,903.
Общата статистическа оценка показва
2. Материали и методи ниво на значимост, по-малко от грешката α =
Настоящото изследване е разработено с 0,05, което е достатъчно да считаме, че средните
данни за произведено количество спанак, както добиви от спанак на 28-те страни-членки на ЕС
и за размера на засятите площи, извлечени от притежават статистически различия и че общият
база данни FAOSTAT. Оценката на модел е статистически значим.
статистическите данни е направена на базата на На Таблица 1 е представена оценката на
размера на средния добив за съответната средния добив на спанак по страни в зависимост
европейска държава за периода 1961-2014 г. от принадлежността им към съответния клъстер.
Използван е методът за междугруповото
свързване, а критерият, по който се осъществява
той, е евклидовото разстояние. Известно е, че
при метода на междугруповото свързване
разстоянието между два клъстера А и В се
дефинира като средната стойност на nA.nB на
брой разстояния между nA точки от А и nB
точки от В:
nA nB
1
D( A, B) 
n A nB
 d ( x , x )
i 1 j 1
i j (1)
където
p
d ( xi , x j )   (x im  x jm ) 2
m 1
(2)
i, j  1, n е обикновеното евклидово разстояние

между два вектора xi ( xi1 , xi 2 ,..., xip ) и
x j ( x j1 , x j 2 ,..., x jp ) . Построена е дендрограма,
чрез която се представят графично образуваните
клъстери. Направен е и еднофакторен
дисперсионен анализ [1].
Обработката на данните е извършена Фиг. 1. Дендрограма, представяща
чрез статистическата програма IBM Statistics графично групирането на държавите-
SPSS 23 [2]. членки на ЕС в клъстери според средния
добив на спанак
3. Резултати и обсъждане
Клъстери 5 и 6 включват страните с
След направения клъстерен анализ става най-висок среден добив от спанак – Белгия и
ясно, че според показателя „среден добив на Холандия. Тези два клъстера могат да бъдат
спанак“ разглежданите държави от Европейския обобщени в по-голям клъстер на евклидово
съюз могат да бъдат групирани в шест клъстера, разстояние 4 единици.
II-401
Клъстер 2 включва следващите по обем от най-големите в сравнение с тези на
на производство със среден добив, останалите държави (5,67). Това означава, че
статистически различен от средния добив на получените добиви не са устойчиви във времето.
страните от клъстери 5 и 6. Това са Испания, Същевременно Литва е държавата с най-нисък
Франция и Португалия. среден добив, но има сравнително устойчиви
С най-нисък добив на спанак са страните добиви във времето.
от клъстери 1 и 4. На Фигура 2 са отразени действителните
линии и трендовете за добива от спанак на
България, Гърция и Кипър. За България се
Номер Станд. очертават четири периода, отличаващи се един
Среден откло- от друг по посока на изменение на
на Държава
добив нение производството на спанак. Първият период
клъстер
обхваща 1961-1979 г. Характеризира се с
Дания 10,12 hij 2,44 неособено големи пикове и падове, но с лека
тенденция на нарастване. Вторият период е
Кипър 8,73 jk 3,78 1980-1990 г., в който се наблюдават резки
пикове и падове, но отново тенденцията е към
България 8,64 jk 2,91 задържане на производството.
1
Словакия 9,02 jk 2,9 В началото на третия период (1991-2000
г.) се наблюдава рязко спадане на
Чешка реп. 9,61 ij 2,69 производството и стабилизирането му до края на
Румъния 7,53 k 1,08 периода. Последният период обхваща 2001-2014
г.
Испания 16,7 c 2,01 Целият период се характеризира с резки
спадове и пикове. Трендът за производството на
2 Франция 16,67 c 2,68 спанак в България се описва с полином от втора
Португалия 16,41 cde
1,49 степен:
ghi
(3)
Гърция 11,52 3,17 и показва намаляваща тенденция. Коефициентът
на детерминация (23%) показва слабо изменение
Унгария 11,62 ghi 5,68
на добива във времето.
Словения 11,3 ghi 2,01 Производството на спанак в Гърция
може да бъде разделено на три периода.
Италия 12,54 fg 1,1 Първият обхваща времето от 1961 до 1991 г.
3 Малта 12,01 gh 0,68 Началото на периода се характеризира с
нарастваща тенденция на производството на
Финландия 13,17 efg 3,41 спанак, със слабо изразени пикове и спадове.
Втората половина на този период е със
Швеция 13,21 efg 3,42
силно изразени пикове и спадове, но отново с
Австрия 14,77 cde 6,59 нарастваща тенденция на производството.
def
Вторият период е 1992-2009 г. Той е с изразена
Германия 14,46 2,67 намаляваща тенденция, със слаби пикове и
4 Литва 3,68 l 1,84 спадове. Последният период е 2010-2014 г. Тук
имаме силно изразена намаляваща тенденция.
5 Белгия 20,8 b 2,96 Трендът за производството на спанак в Гърция е
описан с полином от втора степен:
6 Холандия 25,73 a 5,67 . (4)
Изменението в добива на спанак е в
Таблица 1. Оценка на средните добиви силна зависимост от периода. Основание за това
на страни от ЕС по метода на Duncan, дава коефициентът на детерминация, който е
a,b,c,… – степен на доказаност при ниво 69%.
на значимост α = 0,05 На представения тренд за
производството на спанак в Кипър ясно се
открояват три периода. Първият е от 1961 г. до
Страната с най-висок добив на спанак е 1972 г., когато количествата произведена
Холандия, но от Таблица 1 е видно, че
статистическият показател за варирането е един

II-402
20 България След направения йерархичен клъстерен
18 анализ се установи, че страните-членки на
Гърция Европейския съюз се групират в шест клъстера.
16
Разпределението на държавите в клъстерите се
14 Кипър
потвърждава и от проведения еднофакторен
12 дисперсионен анализ. България е в първи
10 клъстер, в който се намират държави с по-нисък
8 среден добив на спанак. Оказва се, че в
6 държавите от Източна Европа производството
на спанак е ниско. Това би могло да се обясни
4
както с водената държавна политика към
2 производителите на тази култура, така и с
0 географското разположение на съответната
1960 1970 1980 1990 2000 2010 2020 страна. Това се доказва и от графичното
представяне на добива в България, Кипър и
Фигура 2. Графично представяне на Гърция. При трите държави нямаме устойчива
изменението на средния добив от спанак за тенденция на нарастване на производството за
периода 1961-2014 г. за България, Гърция и целия период от 1961 г. до 2014 г., а са
Кипър характерни спадове и пикове в кратки интервали
от време.
продукция са с постоянни стойности. Вторият
период обхваща 1973-2004 г. Като цяло ЛИТЕРАТУРА
периодът се характеризира с постепенно 1. Barov, V. (1982). Analysis and schemes
нарастване добивът на спанак, както и със слабо of the Polish experience, NAPS, Sofia.
изразени спадове и пикове на добиваните 2. Haralampiev, K. (2009) Working with
количества продукция. Последният период data in SPSS, Publishing House „St. Kl. Ohridski“,
започва от 2005 г. и продължава до края на 2014 Sofia
г. В продължение на тези десет години има
резки спадове и пикове на производство на
спанак. Това се потвърждава и от коефициента
на детерминация, който за Кипър е 72%. Както Контакти
в България и Гърция, така и тук трендът за Нели Керанова
производство на спанак се описва с полином от Аграрен университет-Пловдив
втора степен: 4000 Пловдив
Бул. „Менделеев“, № 12
. (5)
Факултет по икономика
Катедра „Математика, информатика и
4. Заключение физика“
E-mail: nelikeranova@abv.bg
МОДЕЛИРАНЕ НА ДАННИ
ОТ ЖИВОТНОВЪДНИЯ СЕКТОР
В БЪЛГАРИЯ, СВЪРЗАНИ С ОТГЛЕЖДАНЕТО
НА СЕЛСКОСТОПАНСКИ ЖИВОТНИ
ЗА ПЕРИОДА 2000-2014 Г.
НЕЛИ КЕРАНОВА
Резюме: Основната цел на настоящата работа е да се направи оценка и анализ на

статистически данни от сектор „Животновъдство“. Те се отнасят до брой
отглеждани селскостопански животни /говеда, биволи, свине, овце и кози/ в
България за периода от 2000 г. до 2014 г. При обработката на данните е извършен
йерархичен клъстерен анализ и са построени математически модели, отразяващи
връзките между изследваните обекти. При всички животни има тенденция на
чувствителен спад в техния брой. Изключение прави секторът, свързан с
отглеждане на биволи, където се наблюдава тенденция на слабо повишаване на
интереса на фермерите.
Ключови думи: йерархичен клъстер анализ, математически модели,
селскостопански животни
DATA MODELING IN THE AGRICULTURAL

SECTOR IN BULGARIA, RELATED TO THE
BREEDING OF SOME FARM ANIMALS FOR
THE PERIOD 2000-2014
NELI KERANOVA
Abstract: The main objective of this work is to assess and analyze statistical data in the
stock-breeding sector. They refer to the number of farmed livestock /cattle, buffaloes, pigs,
sheep and goats/ in Bulgaria for the period 2000 to 2014. During the data processing a
hierarchical cluster analysis is made and mathematical models are built, reflecting the link
between the researched objects. There is a tendency to a substantial fall in the number of
all animals. The sector related to the breeding of buffalos is an exception, with a tendency
towards slightly increased interest of farmers.
Key words: hierarchical cluster analysis, mathematical models, livestoc
1. Въведение регресионни модели, отразяващи връзките

България е страна с дългогодишни между разглежданите обекти. В резултат на
традиции и опит в развитието на извършените анализи се установи, че
животновъдството. В настоящата работа са съществуват силни връзки между годините на
установени връзките между брой отглеждани проведеното изследване и броя на
селскостопански животни (говеда, биволи, разглежданите селскостопански животни. При
свине, овце и кози) в България и годините на говедата, свинете, овцете и козите се очертава
изследване (2000-2014 г.). Направен е тенденция на намаляване ръста на отглеждане
йерархичен клъстерен анализ по метода на на тези животни. Само при биволите интересът
междугруповото свързване. Построени са на фермерите се засилва във времето.
II-404
2. Материали методи
Настоящото изследване е разработено с
данни за брой отглеждани селскостопански
животни (говеда, биволи, свине, овце и кози) в
България за периода 2000-2014 г.
Статистическите данни в изследването са взети
от базата данни FAOSTAT.
Построени са математически модели
върху получените данни. За целта са използвани
подходящи регресионни уравнения. Определени
са коефициентите на корелация и детерминация.
Чрез тях са направени изводи за степента на
влияние на годината върху брой отглеждани
селскостопански животни. Дадена е графика, Фиг. 1. Йерархичен клъстерен анализ по брой
представяща изменението на количеството на отглеждани говеда в България
всеки вид животни във времето.
Обработката е извършена чрез а вторият – 2012 и 2014 г. Те се обединяват на
статистическата програма IBM Statistics SPSS разстояние 3 евклидови единици. Вторият
23 [1]. основен клъстер също се състои от два
подклъстера: 2001, 2004, 2005, 2006 и 2007 г. от
3. Резултати и обсъждане една страна и като самостоятелен подклъстер –
2003 г. Третият клъстер съдържа само 2002 г. и
В настоящата работа се моделират и
той е най-отдалечен от останалите два клъстера
анализират данни, свързани с брой отглеждани
– на разстояние 25 евклидови единици.
селскостопански животни в България от 2000 г.
до 2014 г. За оценка на статистическите данни е
използван йерархичен клъстерен анализ.
Направено е групиране на годините в
зависимост от броя на съответните животни,
като се използва методът на междугруповото
свързване и за мярка е използвано евклидовото
разстояние. На Фигури 1-5 са поместени
дендрограмите, представящи графично
организирането на данните в клъстери.
Според резултата за брой говеда в
България се открояват три основни клъстера
(Фигура 1). Първият се разделя на два
подклъстера: единият включва 2009, 2010, 2012
и 2014 г., а вторият – 2011 и 2013 г. Те се
обединяват на евклидово разстояние 3 единици.
Вторият основен клъстер включва 2001, 2002,
2006 и 2007 г., като към тях на разстояние 2 Фиг. 2. Йерархичен клъстерен анализ по брой
евклидови единици се присъединява 2008 г. отглеждани биволи в България
Третият клъстер се състои от два подклъстера:
първият се образува от 2000, 2003 и 2005 г, а в Резултатът от йерархичния клъстер анализ
самостоятелен подклъстер е 2004 г. Тези два за брой отглеждани овце е даден на Фигура 3.
подклъстера формират един обобщен клъстер на Очертават се четири клъстера: първият включва
разстояние 5 евклидови единици. 2002, 2009 и 2010 г., вторият – 2011, 2012, 2013
На Фигура 2 се открояват три основни и 2014 г., които се присъединяват към първия
клъстера. Първият включва два подклъстера: клъстер на разстояние 2,5 евклидови единици.
единият се състои от 2000, 2009, 2011 и 2013 г.,
II-405
Фиг. 3. Йерархичен клъстерен анализ по брой Фиг. 4. Йерархичен клъстерен анализ по брой
отглеждани свине в България отглеждани овце в България
Третият клъстер включва два подклъстера:

първият се състои от периода 2003-2007 г., а
вторият – 2001 г. Присъединяват се на
разстояние 2,5 евклидови единици. Последният
и най-отдалечен клъстер обхваща периода през
2000 г.
В резултат на йерархичен клъстерен
анализ става ясно, че при отглеждането на свине
имаме един клъстер, включващ 2002 и периода
от 2009 до 2014 г.. Втори клъстер обхваща 2001,
2003-2008 г. Двата клъстера се обединяват на
разстояние от 6 евклидови единици. Последният
клъстер се състои само от 2000 г. и той е най-
отдалечен от всички останали.
На дендрограмата, представяща
групирането по брой отглеждани овце, се
оформят два обобщени клъстера. Първият се
състои от два подкръстера: от 2008 до 2014 г. и Фиг. 5. Йерархичен клъстерен анализ по брой
от 2002 до 2007 г.Вторият клъстер включва 2000 отглеждани кози в България
и 2001 г., които се присъединяват към горния
клъстер на разстояние 25 евклидови единици. селскостопански животни в периода 2000-2014
г. При всички от 2000 г. до 2003 г. се наблюдава
От Фигура 5 става ясно, че
спад, като най-силен той е в овцете, а най-слаб –
групирането на годините според брой при говедата и биволите. Следващият период,
отглеждани кози е подобно на това според който може да се разграничи, е от 2003 г. до
брой овце. 2004 г., когато има увеличение в броя на
Получените резултати за групирането на животните от всеки сектор. От 2004 г. до 2006 г.
данните по години се обясняват с резултатите, е период с плавни пикове и спадове, а от 2007 г.
представени на Фигура 6. При йерархичния до края на 2014 г. има спад. Изключение прави
клъстерен анализ за всички животни бе секторът, ориентиран към биволите, където
установено, че 2000 г. - 2001 г. формират според статистическите данни се наблюдава
самостоятелни клъстери и причина за това е по- известен ръст. Като цяло за разглеждания
големият брой отглеждани животни през този период от време се наблюдава намаляване на
период. Голямата отдалеченост от останалите интереса на фермерите към отглеждане на
клъстери се обяснява с факта, че в следващите селскостопански животни.
години рязко спада броят на селскостопанските Статистическите данни в тази работа са
животни в България. На графиката са посочени моделирани чрез различни регресионни модели,
измененията в количеството на отглежданите описани в таблица 1. При всички модели
регресионните коефициенти са статистически

II-406
значими, тъй като степента им на значимост полиномен регресионен модел от втора степен,
(Sig.) е по-малка от грешката (0.05). свине и кози – показателен регресионен модел,
овце – полиномен регресионен модел от втора
3000000 степен.
Говеда
Коефициентите на корелация за всички
2500000 България
животни показват висока степен на зависимост
Биволи между брой отглеждани селскостопански
2000000
България животни от времето. Коефициентите на
1500000 Свине детерминация също показват висок процент на
България влияние на годината върху броя на животните.
1000000
Овце
България
500000
На базата на проведените регресионни
Кози анализи и построените математически модели
0 България
1995 2000 2005 2010 2015 можем да направим заключение, че за
различните сектори на животновъдството се
установяват различни тенденции – на
Фиг. 6. Графично представяне на
нарастване или намаляване количеството
брой (в хил.) отглеждани селскостопански отглеждани селскостопански животни в
животни в България за периода 2000 г. – България през периода 2001-2014 г. Буди
2014 г. оптимизъм нарастващият брой биволи,
отглеждани у нас, имайки предвид както по-
Селскост. високите хранителни качества на биволските
животни Регресионен модел продукти, така и по-високата им себестойност.
Говеда Факторите, оказващи влияние върху
спада в отглежданите количества
селскостопански животни, най-общо могат да се
Биволи обобщят в следните няколко направления:
Свине недостатъчни държавни субсидии, подпомагащи
фермерите, морално остаряло техническо
Овце оборудване на фермите, недостатъчно
количество инвестиции от страна на
Кози производителите, незадоволителна степен на
образование и квалификация на хората, заети в
Таблица 1. Регресионни модели на изменение на този сектор и др.
брой отглеждани животни
Коеф. на Коеф. на
Селскостоп. корелация детерм. 1. Ganeva, Z., (2016). To rediscover the statistics
животни (R) (R2) (%) with SPSS, Elestra, Sofia
Говеда -0,84 71
Биволи 0,77 59 Контакти
Свине -0,88 78 Нели Керанова
Аграрен университет-Пловдив
Овце -0,88 78 4000 Пловдив
Кози -0,98 96 Бул. „Менделеев“, № 12
Факултет по икономика
Таблица 2. Коефициент на корелация и Катедра „Математика, информатика и
коефициент на детерминация физика“
E-mail: nelikeranova@abv.bg
Данните за брой говеда се моделират със
степенен регресионен модел, биволи – с
RECTANGULAR PLATES – SINGLE TRIGONOMETRIC SERIES,

EXPONENTIAL FUNCTION
LILIYA PETROVA
Abstract: A method of single trigonometric series is applied for investigation of a plate,

simply supported at two of their opposite ends and arbitrary at two others. In the investigation
instead a hyperbolic functions are used an exponential function. In a general case the load of
the plate is with arbitrary distribution. The investigation rends an account asymmetrical
location of the load acting of the plate. A rectangular plate, simply supported at two of their
opposite ends and clamped at two others, is investigated automated with asymmetrical
distributed load. It is determines automated the normal displacement and of the forces in a
plate. It is made a verification of the results obtained.
Key words: rectangular plates, single trigonometric series, exponential function
1. THE REATIONSHIPS OF THE PLATE IN M x M xy M y M xy

CARTESIAN COORDINATES
Qx   Qy   ,
x y y x
A differential equation for equilibrium of the 3w 3w  * 3w 3w 
plate Qx*   D  3  2   Q   D   2   ,(2)
xy 2   y 3 x 2y 
y
 x 
For a plate with constant thickness it has a kind
 M xy   M xy 
4w  4 w  4 w q  x, y  Et 3 rx   Qx   ry   Q y   ,
 2   D  , (1)
121 2   y  x0  x  y 0
x 4 x 2 y 2 y 4 D xa y b
where q x, y  is a function of the external load; D  w3

w  3
 w 3
3w 
rx   3  2    r   
 y 3   2    .
is the cylindrical rigidity of the plate; w x, y  is a xy 2  x0 x 2y  y0
y
 y xa
 yb
function of the normal displacements of the points
A boundaries conditions at ends with external
of the plate’s middle surface; t is a thickness of the
normal y .
plate;  is a Poisson’s coefficient.
The forces in the plate with a positive directions Depending on a way of supporting they have a
kind
Qy
M xy M xy
x M xy fixed end simply supported end
My M xy
Qx
w x, y   0 Ym  y   0 w x, y   0 Ym  y   0
Mx Mx
dy w x, y  M y  x, y  0 YmII  y   0
My Qx
dx 0
M xy
y Y  y  0
m
I
M xy
y Qy free end
Fig.1 M y  x, y  0 YmII  y   m2 Ym  y   0 (3)
Q*y  x, y   0
2 2 2 2
 w  w  w  w YmIII  y  2   m2 YmI  y   0
M x   D  2  2  M y   D  2  2 
 x y   y x 
2w 2. A SINGLE TRIGONOMETRIC SERIES
M xy   D1  ,
xy A single trigonometric series is applies at
rectangular plates simply supported in two opposite
ends. At the two other ends the plate, in a case y  0
Copyright©2017
University - Sofia,
of Sofia, Plovdiv
Plovdiv branch,
branch, Bulgaria
II-408
y  b , the supporting in a general case is arbitrary A function Ym  y  from elastic surfасе of the
(fig. 2). plate
An introduced coordinate system xOy is From condition Ym  y  to satisfy the differential
according with the supporting and the loading of the equation from equilibrium of the plate, after
plate. substituing (4) and (5) in (1), is receives the
y following unhomogeneous arbitrary differential
equation about function Ym  y 
qm  y 
YmIV  y  2 m2 YmII  y  m4 Ym  y  Bm  y  , where Bm  y   .(6)
b D
The characteristic equation, corresponding to (6)
and its roots has a kind
x
a r 4  2 m2 r 2  m4  0 (7)
Fig.2 r1, 2   m , r3,4   m (8)
Analytical expression for a function of external load
q  x, y   q f  x  f ( y ) ,
General integral of arbitrary unhomogeneous
differential equation about function Ym  y  accepts a
where q is an intesity of external load;
kind
f  x , f  y  are a functions of load in
Ym  y  C1m e m y  C2m e  m y  C3m ye m y  C4m ye  m y  Y m  y  (9)
directions of axes x и y at q 1 ;
A function of the normal displacement of the where Y m  y  is a partial integral of unhomogeneous
points of the plate’s middle surface differential equation about function Ym  y  ;
m C1m , C2m , C3m и C4m are an integration

w x, y   Ym  y sin( m x) ,  m  , m 1,2,...  (4)
a
m constants in an expressions for Ym  y  .
The choice of the function of elastic surface of the
A determination of an integration constants C1m ,
plate must be to satisfy the boundaries conditions
into his simply supported ends in this case parallels C2m , C3m и C4m are make from the boundaries
of the axi’s y . conditions at arbitrary supported ends of the plate –
A development of the external load in the single in this case parallel of axi’s x .
trigonometric series at sin into axi’s x , m -
Furrier’s coefficient
a
2
q x, y   
m
qm  y sin( m x) qm  y  
a0 
q x, y sin  m x dx (5)
A derivations of functions Ym  y 
I
YmI  y   C1m me m y C2mme m y C3m 1m y e m y C4m 1m y e m y Y m
II
YmII  y C1m  m2 e m y  C2m  m2 e m y  C3m  m 2  m y e m y C4m  m 2  m y e m y Y m (10)
III
YmIII  y C1m  m3 e m y  C2m  m3 e m y  C3m  m2 3 m y e m y  C4m  m2 e m y 3 m y Y m
IV
YmIV  y  C1m  m4 e m y  C2m  m4 e  m y  C3m  m3 4  m y e m y  C4m  m3 e  m y 4  m y Y m
Displacements and forces in the plate, A final normal displacements, forces and
presented in single trigonometric series reactions in the plate accept a kind
The coefficients of the trigonometric series 
w x, y   Ym  y sin( m x) ,
M xm  y   m2 Ym  y YmII  y  , M ym  y   YmIi  y  m2 Ym  y  m
sin( m x) sin( m x)
M xy  y   m2 YmI  y 
m
(11) S  x, y  
m
Sm  y 
cos m x 
, r  x, y  r  y cos x .(12)
m
m
m
Qxm  y    m2 YmII  y  m2 Ym  y  , Qy  y  YmIII  y  m2 Ym2  y 

m In the expressions (12) S  x, y  , r  x, y  is a
Q  y  m 2  Y  y  Y  y  , Q  y   m Y  y  2   Y  y 
*
xm m
II 2
m m
*
y m
III 2 I
m m
general significations for a forces, reactions in a
plate, S m ( y) , rm  y  are the coefficients of
m
rx  y  x0   m  Y  y  2  YmII x0

2
m xa
m m
xa trigonometric series for the same forces and
ry  y 
m
y0   m  Y  y  2  Y
2
m m m
II

y0 . reactions. The multiplication with sin  m x  ,
ya ya
II-409
cos m x  is accords with the differential The matrix of coefficients and the vector of free
relationships of the forces in the plate. terms of equations boundarie’s conditions are
present in a kind
3. AN INVESTIGATION OF THE BILATERAL  1 1 0 0  Y m 0
FIXED PLATE  m  m 1 1   I 
An integration constant in a general integral M   e m e  m m e m m e m  Vm  Y m 0 (14)
Ym  y   m m  Y m b 
  
 me m  m e m e m 1 m  e m 1 m  Y mI b 
A boundary conditions at the ends of the plate  
y  0 y b - clamped has a kind After a solution of the received system equations
w x,0   0 Ym 0   0 w x,b   0 Ym b   0 for a plate, clamped at their two opposite ends, for
w w an integration constants are receive the following
 x,0  0 YmI 0  0  x,b  0 YmI b   0 (13) expressions
y y
It is introduses the signification  m  mb .
 1 m2 1  m  1  m  
  2 2
m  1 2  m  2 1m e m   m  e  

2
e m m em  m  e m  
2
 2  1    m 1  
  2m2  e m 1 2m 2 m  1m e m   m     e m  
1 m em m  e m   (15)
M 1  
 2 1 m  
2
1  1   1  2 m  m 1  2  m

 4m   m2  2  e      2  2m 1 m 2m   2 1   e m   m  m  e  m  
 e   e m  em  e   e  

 2
m
 2
m
 m
 1 2m  e  m  2m  e 1  2m 1e 1  m  (2m 1)e   m 




m 1 
e 
Cm M  Vm  .
1
(16) It follows a determination the coefficients of
trigonometric series and the functions of
displacement, forces and reactions in the plate.
4. NUMERICAL EXAMPLES  q  20 kN 
 .
The load is nonsimmetyrical distributed into  m2 
axi’s y . m - Furrier’s coefficient in development of
The geometry, the supporting and the loading of external load in sin series on the axi’s x
a
the plate are shown in fig. 3. 2
a 
qm  y   q x, y sin  m x dx (18)
0
cos
y kN
q  20 2 x y
m q cos при m 1
2b
x or qm  y   2b (19)
sin
a b 0при m 1
A right part of unhomogenous differential
equation accepts a kind
q y
Bm  y   cos (20)
a D 2b
y
A partial integral Y m  y  of unhomogeneous
Fig. 3
differential equation
A characteristics of a plate
A lengths of the side of the plate Y m  y   K m cos y  (21)
a 6m ; b  6m ; After substitution in (8)
A thicknes of the plate t  0,12 m .
  q
K m  4  2 m2  2  m4 cos y  cos ,
D
y
2b
(22)
kN
Module of elasticity E  2.107 2 ; from where on the method of undetermined
m
A Poisson’s coefficient   0,17 . coefficients, at
A function of external load has a kind, q 1 
Km  4 2 2 4
 , (23)
y x D   2 m   m 2b
q x, y  q cos sin for 0  x a and 0  y  b , (17)
2b a It follows

II-410
K m cosy  при m 1 III IV
Y m  y  . (24) Y m  y  K m  3 sin  y  , Y m  y  K m  4 cos y  . (25)
0 при m 1
A results from the solutions with m 1 terms are
A derivations of Y m  y  shown in a tables 1.1-1.5.
I II
Ym  y    K m  sin  y  , Y m  y  K m  2 cos y  ,
Table 1.1.
m m m mb e m
C1m C2m C3m C4m
m 
a
1 0.5236 3.1416 23.1407 -3.1925 -167.1078 0.7458 -86.5716
I II III
m DY m , DY m , DY m , DY m
y0 y  0,5b y b
1 170.3003 0 -11.6722 0 120.4205 -31.5260 -8.2535 2.1608 1.0428e-14 -44.5845 0 3.0558
Table 1.2.
DYm  y  , DYmI  y  , DYmII  y  , DYm  y 
III
m
y0 y  0,5b y b
1 0 0 33.0776 -47.0593 27.0978 -1.8832 -11.2358 2.3869 -5.0116e-137.1054e-14 21.9896 22.6900
Table 1.3.
Dw x, y 
m x0 x  0.1675a x  0.333a x  0.5a
y0 y  0.5b yb y0 y  0.5b y  b y0 y  0.5b y  b y0 y  0.5b yb
1 0 0 0 0 13.5489 -2.5058E-13 0 23.4674 -4.3402e-13 0 27.0978 3.3184E-15
x  0,667 a x  0.833a xa
y  0 y  0.5b y  b y  0 y  0.5b y  b y0 y  0.5b yb
1 0 23.4674 3.3184E-15 0 13.5489 -2.5058E-13 0 3.3184E-15 0
Table 1.4.
M x  x, y 
m x0 x  0.1675a x  0.333a x  0.5a
1 0 0 0 -2.8116 4.6695 -1.8691 -4.8698 8.0879 -3.2374 -5.6232 9.3391 -3.7382
x  0,667 a x  0.833a xa
1 -4.8698 8.0879 -3.2374 -2.8116 4.6695 -1.8691 0 1.1437E-15 0
M y  x, y 
m x0 x  0.1675a x  0.333a x  0.5a
y0 y  0.5b yb y  0 y  0.5b y  b y0 y  0.5b y  b y0 y  0.5b yb
1 0 0 0 -16.5388 6.2494 -10.9948 -28.6460 10.8242 -19.0435 -33.0776 12.4987 -21.9896
x  0,667 a x  0.833a xa
y0 y  0.5b y  b y0 y  0.5b y  b y0 y  0.5b yb
1 -28.6460 10.8242 -19.0435 -16.5388 6.2494 -10.9948 -4.0507e-15 1.5306e-15-2.6929e-15
M xy  x, y 
m x0 x  0.1675a x  0.333a x  0.5a
1 0 0.8184 -3.0879e-14 0 0.7088 -2.6742e-14 0 0.4092 -1.5440e-13 0 0 0
x  0,667 a x  0.833a xa
y  0 y  0.5b y b y  0 y  0.5b y  b y0 y  0.5b yb
1 0 -0.4092 1.5440e-14 0 -0.7088 2.6742e-14 0 -0.8184 3.0879e-14
Qx  x, y 
m x0 x  0.1675a x  0.333a x  0.5a
II-411
1 -17.3194 9.7729 -11.5137 -14.9990 8.4636 -9.9712 -8.6597 4.8864 -5.7569 -1.0605e-15 0 0
x  0,667 a x  0.833a xa
1 8.6597 -4.8864 5.7569 14.9990 -8.4636 9.9712 17.3194 -9.7729 11.5137
Q y  x, y 
m
x0 x  0.1675a x  0.333a x  0.5a
1 0 0 0 23.5297 -1.4516 -11.3450 40.7546 -2.5142 -19.6501 47.0593 -2.9032 -22.6900
x  0,667 a x  0.833a xa
y0 y  0.5b y  b y0 y  0.5b y  b y0 y  0.5b yb
1 40.7546 -2.51420 -19.6501 23.5297 -1.4516 -11.3450 5.7629e-15 0 -2.7786e-15
Qx*  x, y 
m x0 x  0.1675a x  0.333a x  0.5a
1 -31.6945 14.6558 -21.0701 -27.4482 12.6923 -18.2472 -15.8472 7.3279 -10.5350-1.9407e-15 0 -1.2901e-15
x  0,667 a x  0.833a xa
y  0 y  0.5b y  b y  0 y  0.5b y  b y  0 y  0.5b y  b
1 15.8472 -7.3279 10.5350 27.4482 -12.6923 18.2472 31.6945 -14.6558 21.0701
Q*y  x, y 
m
x0 x  0.1675a x  0.333a x  0.5a
y0y  0.5b y  b y  0 y  0.5b y  b y  0 y  0.5b y  b y  0 y  0.5b y  b
1 0 0 0 23.5297 -1.6658 -11.3450 40.7546 -2.8853 -19.6501 47.0593 -3.3317 -22.6900
x  0,667 a x  0.833a xa
y  0 y  0.5b y  b y  0 y  0.5b y  b y  0 y  0.5b y  b
1 40.7546 -2.8853 -19.6501 23.5297 -1.6658 -11.3450 5.7629e-15 0 -2.7786e-15
Table 1.5.
rx  x, y 
m
x0 xa
y0 y  0.5b yb y0 y  0.5b yb
-31.6945 14.6558 -21.0701 -31.6945 14.6558 -21.0701
1 ry  x, y 
x0 x  0.167a x  0.333a x  0.5a x  0.667a x  0.833a xa
y 0 0 -23.5297 -40.7546 -47.0593 -40.7546 -23.5297 -5.7629E-15
yb 0 11.345 19.6501 22.69 19.6501 11.345 2.7786E-15
A cause a symmetry of the supporting and loading the results of displacements and forces are writed only
for a part of the plate.
For verification are write an equations for equilibrium of the plate or for the part of the plate.
4. CONCLUSIONS REFERENCES
The presented investigation uses only one
function – the exponential. The replacement of 1. Varbanov Chr., A. Tepavicharov, T. Ganev,
hyperbolic trigonometric functions by exponential Theory of elasticity and plasticity, Thechnika,
simplifies the expressions used in the solution and Sofia, 1992.
in the some cases improve the conjuences at 2. Tc. Krustev, T. Karamanski and oth., Guidance
(m 1,..., ) terms of series. for a solution of a problem of theory of
elasticity, stability and dynamics of elastic’s
The solutions are conduct with a compoused PC
systems, Thechnika, Sofia, 1974.
program. The results obtained are compared.
3. Timoshenko S. P., S. Voinovski-Kriger, Plates
The method is applies analogical at arbitrary
and Shells, Моscow, 1966.
distributed loads on the plate.

II-412
4. Szilard R, Theory and Analysis of Plates,
Classical and Numerical Method, Prentice-
Hall, New Jersry, 1976
5. Vainberg D.B., E.D. Vainberg, A solution of a
plates, Budivelnik, Kiev, 1970.
Liliya Petrova, Assoc. Prof. PhD Eng.-math.,

dep. “Mechanics”, Univ. of Transport “T. Kableshkov”,
158 Geo Milev Str., Sofia 1574, Bulgaria
E-mail: lbphr@abv.bg
PARTNERS AND SPONSORS
БЛАГОДАРНОСТИ ??????????

TECHSYS 2017 Proceedings

Uploaded by

Copyright:

Available Formats

TECHSYS 2017 Proceedings

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

TECHSYS 2017 Proceedings

Uploaded by

Copyright:

Available Formats

What is the purpose of the conference?

What is the purpose of the conference?

What organizations are sponsoring/supporting the conference?

What organizations are sponsoring/supporting the conference?

TECHNICAL UNIVERSITY – SOFIA

SIXTH INTERNATIONAL SCIENTIFIC CONFERENCE

PROCEEDINGS СБОРНИК ДОКЛАДИ

ТЕХНИЧЕСКИ УНИВЕРСИТЕТ – СОФИЯ

ШЕСТА МЕЖДУНАРОДНА НАУЧНА КОНФЕРЕНЦИЯ

TECHNICAL UNIVERSITY – SOFIA, PLOVDIV BRANCH

SIXTH INTERNATIONAL SCIENTIFIC CONFERENCE

ТЕХНИЧЕСКИ УНИВЕРСИТЕТ – СОФИЯ, ФИЛИАЛ ПЛОВДИВ

ШЕСТА МЕЖДУНАРОДНА НАУЧНА КОНФЕРЕНЦИЯ

This conference is technically co-sponsored by: Tази конференция е технически

International Programme Committee:

Chairmen: Prof. Dr. Galidia Petrova

Scientific Secretary: Dr. Sevil Ahmed

ISSN Online: 2535-0048

Научен секретар: д-р инж. Севил Ахмед

1. ABDELHAFID BENYOUNES, AHMED HAFAIFA, ABDELLAH KOUZOU,

8. DIMITAR SPIROV, NIKOLAY KOMITOV.......................................................................................II-42

1. EMEL ONAL ..................................................................................................................................II-55

1. TARIK YERLİKAYA, EMRECAN ADLI ........................................................................................II-144

6. DIMITAR GARNEVSKI ..............................................................................................................II-163

1. CENK MISIRLI, ISIK CETINTAV, YILMAZ CAN, CEM MISIRLI....................................................II-216

6. ILIA POPOV, SABI SABEV .........................................................................................................II-240

1. TONI MIHOVA, VALENTINA NIKOLOVA - ALEXIEVA .............................................................II-316

1. ALBENA PAVLOVA .....................................................................................................................II-386

2. DOBRINKA GRIBACHEVA, DIMITAR RAZPOPOV ...................................................................II-392

© International Scientific Conference on Engineering, Technologies and Systems

PLENARY PAPERS • ПЛЕНАРНИ ДОКЛАДИ

Smart Grid and Power Electronics Integration

CONTROL THEORY AND IT APPLICATIONS

18-19.May 2017 TechSys, Plovdiv

Automatic control in information

18-19.May 2017 TechSys, Plovdiv

SMC Industrial Automation Bulgaria

SMC has a clear strategy when it comes to Industry 4.0.

INDUSTRY 4.0 Examples

The content-related drivers for the advance of

INDUSTRY 4.0 is a term generated by German

Nevertheless- Industry 4.0, Machine-to-Machine

Training solutions, adapted to the modern industry

eLEARNING-200 - The perfect theory companion

To be able to develop different technology skills, theoretical knowledge

• SMC-100 – Introduction to industrial automation

A. Training systems in different technologies:

PNEUMATICS AND ELECTROP

PNEUMATE-200, PNEUTRAINER 200, VAC 200

ELECTRICAL DRIVES, SENSORS, LASER, OPTICS, ARTIFICAL VISION

B. Training systems with INTERACTIONS OF TECHNOLOGYES:

AUTOMATION, MANIPULATION, SCADA

ROBOTICS, HMI, MOTION CONTROL

ROBOTICS, HMI, MOTION CONTROL, ERP, MES

C. INDUSTRY 4.0 in education = CREATIVITY ENVIRONMENT

We know what is Creativity – Knowledge-of-Ideas, Freedom of Thinking, Open Space in Mind,

USER’s MANUAL, THEORY MANUAL, EXERCISES MANUAL (STUDENT and PROFESSOR)

18.05.2017, SMC Industrial Automation Bulgaria EOOD

© International Scientific Conference on Engineering, Technologies and Systems

AUTOMATION AND CONTROL SYSTEMS