GDANSK UNIVERSITY OF TECHNOLOGY

FACULTY OF OCEAN ENGINEERING AND SHIP TECHNOLOGY

SECTION OF TRANSPORT TECHNICAL MEANS
OF TRANSPORT COMMITEE OF POLISH ACADEMY OF SCIENCES
UTILITY FOUNDATIONS SECTION
OF MECHANICAL ENGINEERING COMMITTEE OF POLISH ACADEMY OF SCIENCE

ISSN 1231 – 3998

ISBN 83 – 900666 – 2 – 9

Journal of

POLISH CIMAC

SELECTED PROBLEMS
OF DESIGNING
AND OPERATING
TECHNICAL SYSTEMS

Vol. 9 No. 3

Gdansk, 2014

Science publication of Editorial Advisory Board of POLISH CIMAC

 Editor in Chief: Jerzy Girtler Editorial Office Secretary: Ryszard Zadrąg

Editorial Advisory Board

J. Girtler (President) - Gdansk University of Z. Matuszak - Maritime Academy of Szczecin

Technology J. Merkisz – Poznan Unversity of Technology
A. Adamkiewicz - Maritime Academy of Szczecin Y. Ohta - Nagoya Institute of Technology
J. Báachnio - Air Force Institute of Technology M. Orkisz - Rzeszow University of Technology
C. Behrendt - Maritime Academy of Szczecin S. Radkowski - President of the Board of PTDT
P. Bielawski - Maritime Academy of Szczecin Y. Sato - National Traffic Safety and Environment
T. Chmielniak - Silesian Technical University Laboratory, Japan
R. Cwilewicz - Maritime Academy of Gdynia A. Soudarev - Russian Academy of Engineering
T. Dąbrowski - WAT Military University of Sciences
Technology Z. Stelmasiak - Bielsko-Biala Technology-
M. Dzida - Gdansk University of Technology Humanistic Academy
V. Hlavna - University of Žilina, Slovak Republic Z. Smalko - Warsaw University of Technology
M. Idzior – Poznan University of Technology M. ĝlĊzak - Warsaw Motor Transport Institute
A. Iskra - Poznan University of Technology W. Wasilewicz Szczagin - Kaliningrad State
J. JaĨwiĔski - Air Force Institute of Technology Technology Institute
J. KiciĔski - President of SEF MEC PAS, member of J. Wajand – Lodz University of Technology
MEC E. Wiederuh - Fachhochschule Giessen Friedberg
L. Ignatiewicz Kowalczuk - Baltic State Maritime M. WyszyĔski - The University of Birmingham,
Academy in Kaliningrad United Kingdom
J. Lewitowicz - Air Force Institute of Technology S. ĩmudzki - West Pomeranian University of
J. Macek - Czech Technical University in Prague Technology in Szczecin
J. ĩurek - Air Force Institute of Technology

Editorial Office:
GDANSK UNIVERSITY OF TECHNOLOGY
Faculty of Ocean Engineering and Ship Technology
Department of Ship Power Plants
G. Narutowicza 11/12 80-233 GDANSK POLAND
tel. +48 58 347 29 73, e – mail: sek4oce@pg.gda.pl
www.polishcimac.pl
This journal is devoted to designing of diesel engines, gas turbines and ships’ power transmission systems containing these
engines and also machines and other appliances necessary to keep these engines in movement with special regard to their
energetic and pro-ecological properties and also their durability, reliability, diagnostics and safety of their work and
operation of diesel engines, gas turbines and also machines and other appliances necessary to keep these engines in
movement with special regard to their energetic and pro-ecological properties, their durability, reliability, diagnostics and safety
of their work, and, above all, rational (and optimal) control of the processes of their operation and specially rational service
works (including control and diagnosing systems), analysing of properties and treatment of liquid fuels and lubricating oils,
etc.

All papers have been reviewed

@Copyright by Faculty of Ocean Engineering and Ship Technology Gdansk University of Technology
All rights reserved
ISSN 1231 – 3998
ISBN 83 – 900666 – 2 – 9

Printed in Poland
 CONTENTS

KarpG.:CHANGESINWATERTRANSPORTSTANDARDREQUIREMENTSANDTHEIREFFECT
ONECOLOGY………………………………………………………………………………………………………………………… 7
KotlarzW.,KrólikM.,KolasaR.:INFLUENCEOFMODULARF100ͲPWͲ229ENGINE
CONSTRUCTIONUPONTHEFLEETOFFͲ16AIRCRAFTEXPLOITATION…………………………………… 13
KulaM.:ACOUSTICEMISSIONASATOOLFACILITATINGTECHNICALCONDITION
INSPECTIONSOFTANKERSCARGOTANKS…………………………………………………………………………… 19
LandowskiB.:ANEXAMPLEOFATECHNICALOBJECTOPERATIONPROCESSMODEL
DESCRIBINGINFLUENCEOFENGINEDAMAGESONTHEOPERATIONPROCESSCOURSE……… 33
BosiewiczZ.:CONDITIONSOFUSAGEANALISYSOFTHETWOͲSTROKE,SLOWͲRUNNING,
HIGHPOWERRATING,MAINENGINESANDTHECAUSESANDEFFECTSOFTHE
TRIBOILOGICALSYSTEMSDAMAGE……………………………………………………………………………………… 43
BukasiewiczM.,KaųaczyŷskiT.,LissM.,KanigowskiJ.:THELMSVIRTUAL.LABAPPLICATION
INMACHINESTECHNICALSTATEANALYSIS…………………………………………………………………………… 59
NanowskiD.:THEINFLUENCEOFCONDENSINGPRESSURESOFTHEETHYLENECARRIER
CASCADECYCLEONITSREFRIGERATIONCAPACITY……………………………………………………………… 67
RosųanowskiJ.:VALUATIONOFROTODYNAMICPUMPSOPERATION,BYMEANSOF,
DIMENSIONALANALYSIS……………………………………………………………………………………………………… 73
SadowskiA.,‚óųtowskiB.,KuliƑE.:APPLICATIONOFENERGYDISSIPATIONANALYSISFOR
IDENTIFICATIONOFMACHINECOMPONENTCONDITION…………………………………………………… 81
SzturomskiB.,BohnM.:THESTRESSSTATEANALYSISOFRESCUESEATAREAOFSUBMARINE
KOBBENCLASSDURINGRESCUEVEHICLELANDING……………………………………………………………… 91
SzulistN.:MODELLINGOFSOMESTEALTHFEATURESFORASMALLNAVYSHIPATTHE
CONCEPTDESIGNSTAGEͲPARTII………………………………………………………………………………………… 101
WilczarskaJ.:APPLICATIONOFINFORMATIONTECHNOLOGIESINOPTIMIZATIONOFTHE
MACHINESTATEEVALUATIONPROCESS……………………………………………………………………………… 111

‚yczkowskiM.:MICROCOMPASSFORSAFESEANAVIGATION……………………………………………… 121
 CHANGES IN WATER TRANSPORT STANDARD REQUIREMENTS
AND THEIR EFFECT ON ECOLOGY
*UDĪ\QD.DUS

The Technology Institute of Air Force

35 A Ostroroga Street
01-163 Warsaw, Poland
phone/fax: +48 22 685308, e-mail: grazyna.karp@itwl.pl

Abstract

In recent years, there is tendency to adjust domestic petroleum products to European and global standards that
establish high requirements regarding environment protection. Meanwhile it’s tendency to use more biocompoents
resulting in decrease of harmful effect on environment. One of the goals of European Directives is to introduce
ecological trends in water, inland and sea transport. The intensity of fuel „ecologisation” concept introduction can
be watched basing on example of change in standard requirements, regarding mostly the reduction of carbon dioxide
emission from combustion of specific liquid petroleum fuels. This is achieved by employing acceptable levels of
sulphur content in such fuels as the condition of their use on state territory, territorial sea waters, exclusive economic
zones, and pollution control zones of Member Countries.

Keywords: environment protection; biocomponents; biofuels - alternate fuels; greenhouse effect; gas emission
1. ,QWURGXFWLRQ

There were a lot of changes that took place in refinery industry, enforced by increasing
requirements regarding fuels, especially in terms of ecological aspects.
Development trends aimed at harmful emission to the environment have been noticed.
There were several ways to achieve this goal.
One of them is to ensure relevant high quality of fuels. It could be achieved by the following:
x chemical composition optimization aimed at ensuring proper combustion conditions while
using as few as possible toxic components (e.g. aromatic compounds content limiting, using
oxygenates etc.);
x discarding environment harmful components (e.g. benzene, lead, sulphur, PNA);
x searching alternate energy sources;

Recent use of fuels in Poland is the result of this occurrence in Europe. The EU has
determined relevant requirements in EU directives regarding, among others, quality of fuels used
in water transport (e.g. 1999/32/EU, replaced by 2005/33/EU, and the next 2009/30/EU). The aim
of these Directives is to reduce sulphur dioxide from combustion of some liquid fuels, thus
reduction of harmful effect of such emission on human and environment.

7
2. :DWHUWUDQVSRUW(LQODQGDQGVHD)

The sea transport results in c.a. 4 % of global carbon dioxide emission due to human
activity. This transport supports over 70 % of international trade. There were no significant
environment losses until the first tankers have appeared. Increasing demand for crude oil and
petroleum products resulted in sudden demand for oil tankers. Vast quantities of crude oils, oils
and other harmful materials enter waters causing huge losses in ecosystems. Vast areas of sea,
especially coastal ones near harbors, become ”dead waters”. Inland water transport also causes
great losses. Water life in rivers and other inland water bodies is destroyed by harmful chemicals
leaking. Sea transport is the most friendly to environment in terms of greenhouse gases emission.
In order to prevent climate change as well as threat to human health, it’s very important to reduce
pollution caused by sea transport, and other sources. In comparison to other means of transport,
the sea transport has the lowest carbon diooxide emission coefficient related to one tonn-
kilometer, however it’s expected that greenhouse gases emission caused by this type of transport
will increase from current 1 Gt by 150-200 % in next 40 years. The goals for Europe 2020
strategy cover obligation to lower greenhouse gases emission by at least 20 % in comparison to
1990 or, in case of favourable circumstances, even by 30 %. The scope of this obligation is
described in EU regulations regarding climate and energy. The water transport is also the source
of air pollution. It was expected that in case of not taking relevant measures, taking into
consideration estimated increase of sulphur dioxide emission by 10-20 %, air pollution at main
sea routes could increase in recent two years (2010-2012). The sulphur content in fuels used in
sea transport is high, in the range from 10000 ppm for Sulphur Emission Control Areas (SECA)
to 27000 ppm for world average. New regulations regarding naval fuels significantly lower air
pollution and its effect on human health. Air pollution caused by sea transport is transferred at big
distances, resulting in increasing problems regarding air quality in many cities. In case of no
countermeasures, sulphur emission from EU sea vessels would exceed emission level from all
land sources by year 2020. Altered regulations lower such tendency, lower not onlu sulphur
emission, but also particulate emission, that means significant progress in terms of human health
and environment protection. The Directive that took effect in December 2012 basing on standards
developed by International Marine Organisation (IMO), would take to gradual reduction of max.
sulphur content in naval fuels from 3,5 % to 0,5 % by year 2020. In case of very sensitive
ecosystems (thanks to agreement within International Marine Organisation), vessels sailing on
Sulphur Emission Control Areas, such as Baltic Sea and North Sea including English Channel,
would have to switch to fuel containing only 0,1 % sulphur by year 2015.

3. (FRORJLFDODQGHFRQRPLFDODVSHFWRIIXHOIRUZDWHUWUDQVSRUW

The possibility of choice of equivalent methods for fulfilling the requirements, such as
exhaust purifying systems or vessels powered by liquefied natural gas (LNG) is an alternative to
low sulphur fuel. According to EC, the LNG is the most prospective alternate fuel for water
transport.

The EC, in document connected with „clear energy for transport” package, has noticed
LNG values, both regarding environment and economy. The environment value is connected
mostly with lower emission of greenhouse gases but also with sulphur oxides emission. Decision
of European Parliament dated September 2012, regarding this issue, establishes sulphur content
limit in fuels for all vessels sailing baltic Sea, North Sea and English Channel, from current 1 %
to
0,1 %, and from 3,5 % to 0,5 % in case of other EU waters.

8
 The new standards that will be in effect from 2015 in practice will eliminate the most
popular fuel so far, such as heavy fuel oil (mazout) generating large amounts of fuel dust and
soot. The economical changes are equally vital. The cost of LNG is much lower than fuel oil,
especially the one with low sulphur content. Therefore it’s expected that in the long run, ship
owners will trade at least the part of fleet for vessels powered by LNG. Proecological projects
regarding marine area are currently funded by TEN-T and Marco Polo programs and European
Bank of Investment (EBI). Moreover, the European Commission has initiated activities regarding
use the LNG as naval fuel. Medium and long-term means aimed at supporting ecological
technology of vessels, alternate fuels and development of „green” transport infrastructure in the
context of setting tool of balanced water transport still will be implemented, with cooperation of
business and member nations. Vessels sailing all other European areas will have to use fuel with
sulphur content below 0,50 % starting from 2020.
The intensity of the „ecologisation” concept implementation could be traced basing on
standard requirements changes regarding mostly sulphur dioxide emission reduction from
combustion of specific petroleum liquid fuels which is achieved by implementing allowable
sulphur content in such fuels as a condition of their use in Member Countries territory, territorial
waters, exclusive economic zones, and pollution control zones.

4. 7UHQGVLQFKDQJHRIUHTXLUHPHQWVIRUIXHORLO

The most noticeable changes took place in relation to sulphur content) (Fig. 1).
After the World War II Polish Standard PN/C-96048 dated 1949 required the sulphur
content at maximum level of 1 % (m/m). In 1955 the requirements were changed for oils of
category I (for high-speed engines) were changed to 0,1 % (m/m) max., 0,2 % (m/m) max. for
category II,
and 0,5 % (m/m) max. for category III. In 1967 the standard PN-67/C96048 appeared. It replaced
two specifications, PN-61/C-96047 and PN-55/C-96048 for fuel oils DS and DZ and fuel oils
respectively. It has been allowed to increase sulphur content up to 0,6 % (m/m) in light fuels from
crude oils of medium sulphur content, and up 1 % (m/m) from high-sulphur crude oils. The reason
is unknown at the moment. In case of winter fuels, sulphur content r5anges from 0,2 % (m/m) to
0,6 % (m/m). Sulphur content in all types of fuel oil was limited to 0,3 % (m/m) in 1992
(PN-92/C-96051). Like in case of gasoline, the requirements basing on European standards were
introduced in 1999. PN-EN-590 (implementation of EN 590:1993) allowed sulphur content at the
level of just 0,20 % (m/m). Before implementation the sulphur content at the level of 350 ppm
(0,350 % m/m), like in EN 590:1999, The Minister of Economy issued regulation (March 08
2002) restricting sulphur content to 500 ppm. As a matter of fact the fuel for CI engines with
sulphur content up to 50 ppm already appeared in 2000. Fuel with sulphur content of 25 ppm and
lower is also produced though PN-EN 590:2003 standard allowed 350 ppm until the end of 2004
and 50 ppm from the beginning of 2005. In order to market the fuels with sulphur content up to
10 ppm from 2009 the Directive 2003/17/EU demanded governments to take appropriate action
aimed at implementation fuels with sulphur content of 10 ppm.

9
 11000

10000

9000

8000
Sulphur content, ppm

7000

6000

5000

4000

3000

2000

1000

0
1949 1954 1959 1964 1969 1974 1979 1984 1989 1994 1999 2004 2009 2014
Ye ars

Fig. 1 Sulphur content in fuel oil according to national requirements in years 1949-2012

5. 0DULQHIXHOV

The first specification describing requirements for fuels for low-speed marine engines
(Class II acc. to PN-67/C-96048) was developed in 1967, and demanded sulphur content max. 1,2
% (m/m). It allowed even 2 % (m/m) of sulphur content, after agreement with contracting party.
Sulphur content in DS fuels (high-speed engines) was max. 0,2 % (m/m). The above specification
was replaced by PN-74/C-96049 standard in 1974.
This standard allowed the following, higher sulphur content:
x class I for fuels from low-sulphur crude oils: max. 0,5 % (m/m),
for fuels from sulphur crude oils: max. 1,5 % (m/m),
x class II for fuels from sulphur crude oils: max. 2,8 % (m/m),
x class III for fuels from sulphur crude oils: max. 3 % (m/m),

The standard sulphur content was lowered in the range from 0,5 to 1,5 % (m/m) for light
fuel distillates and to 1,5 % for residual fuels just in 1993. In 1994, the company Rafineria
*GDĔVN6$VWDUWHGWRSURGXFH'6IXHOZLWKVXOSKXUFRQWHQWRI m/m). In case of marine
fuels, decreasing the sulphur content is much slower. It’s because of specific design requirements
of engines.

10
 30000

25000
Sulphur content, ppm

20000

low-speed engines
15000
high-speed engines

10000

5000

0
1967 1972 1977 1982 1987 1992 1997 2002 2007 2012 2017
Years

Fig. 2 Sulphur content in marine according to national requirements in years 1967-2012

6. 6XPPDU\

x The trend towards permanent sulphur content in fuels lowering has been noticed both in
refinery industry and on the fuel market. This is especially noticeable in recent years, after
implementing international agreements.
x Apart from obvious advantages, significant sulphur content lowering enforces manufacturers
to modify arrangement of additives. It consists mostly in necessity to ensure proper lubricity
which has been achieved earlier by use of polar sulphur compounds.
x The trend towards sulphur voiding from marine fuels is noticeable slower. It’s, among others,
due to requirements set by marine engines and their environment of use.

5HIHUHQFHV

>@ PN/C-96048 z 1949 r. Przetwory naftowe - 2OHMHQDSĊGRZH JD]RZH - Warunki techniczne

>2@ PN-55/C-96048 Przetwory naftowe - 2OHMHQDSĊGRZH
>3@ PN-61/C-96047 Przetwory naftowe - 2OHMHQDSĊGRZH'6L'=
>4@ PN-67/C-96048 Przetwory naftowe - 2OHMHQDSĊGRZH
>5@ PN-74/C-96049 Przetwory naftowe - 2OHMHQDSĊGRZHGRVLOQLNyZRNUĊWRZ\FK
>6@ PN-92/C-96051 Przetwory naftowe - Oleje QDSĊGRZHOHNNLH
>7@ PN-93/C-96049 Przetwory naftowe - 3DOLZDĪHJOXJRZH
>8@ WT-3/94 Warunki techniczne RGSA - 2OHMQDSĊGRZ\'6
>9@ PN-EN 590: 2006 3DOLZDGRSRMD]GyZVDPRFKRGRZ\FK- 2OHMHQDSĊGRZH- Wymagania
LPHWRG\EDGDĔ
>0@ PN-EN 590: 2009 3DOLZDGRSRMD]GyZVDPRFKRGRZ\FK- 2OHMHQDSĊGRZH- Wymagania
LPHWRG\EDGDĔ
>1@ PN-(1$3DOLZDGRSRMD]GyZVDPRFKRGRZ\FK- 2OHMHQDSĊGRZH- Wymagania
LPHWRG\EDGDĔ
>2@ PN-(1$3DOLZDGRSRMD]GyZVDPRFKRGRZ\FK- OlejHQDSĊGRZH- Wymagania
LPHWRG\EDGDĔ

11
>3@ 5R]SRU]ąG]HQLH0LQLVWUD*RVSRGDUNL]
>4@ 5R]SRU]ąG]HQLH0LQLVWUD*RVSRGDUNL3UDF\L3ROLW\NL6RFMDOQHMGRXVWDZ\]U
RV\VWHPLHPRQLWRURZDQLDLNRQWURORZDQLDMDNRĞFLSDOLZFLHNá\FK ']8z 04.02.2003)
>5@ Dyrektywa 98/70/EU
>6@ Dyrektywa 87/219/EEC
>7@ Dyrektywa 2003/17/EC
>8@ Dyrektywa Rady 1999/32/WE
>9@ Dyrektywa 2005/33/WE
>20@ Dyrektywa Parlamentu Europejskiego i Rady 2009/30/WE
>2@ ']LĊJLHOHZVNL:.DUS*7HQGHQFLHVWRZDUGV changes of sulphur content (as a selected
„ecological” parameter) in fuels for land and marine engines. Journal of EXPLO-DIESEL & GAS
TURBINE'

12
INFLUENCE OF MODULAR F100-PW-229 ENGINE CONSTRUCTION
UPON THE FLEET OF F-16 AIRCRAFT EXPLOITATION
:RMFLHFK.RWODU]0LHF]\VáDZ.UyOLN5DIDá.RODVD
The Polish Air Force Academy
ul. Dywizjonu 303 nr 35
08-'ĊEOLQ3RODQG

Introduction

Due to high operational costs of aircraft turbine engines, back in the 1970s, users of this
type of engines – i.e. airlines and the military – began to search for ways of lowering the costs
considerably. In consequence jet engines were replaced with jet turbine engines that were far
more economical, particularly in case of their fuel economy. Another significant aspect was
maintenance costs, in particular the necessity to conduct frequent repairs and overhauls,
which largely increased and complicated the performance of the engines, especially for
airlines, due to the fact that at that time turbine engines had to be serviced having flown
several dozen hours and upgraded every few hundred working hours in specialist maintenance
shops. The activities resulted in “grounding” of the aircraft for the period of maintenance and
repair, instead of flying. The attempts of the largest engine users, that is large airlines and the
US Air Force, caused that the manufacturers of aircraft engines (Pratt&Whitney, General
Electric and Rolls Royce) initiated research and development work, which in the next years,
led to the construction of engines that were capable of working for several thousand hours,
without the necessity for dismantling the airframe. The new engines were not exploited in
accordance with the old and reliable service life strategy, but with their technical condition.
The immediate consequence was the time shortening as well as lowering repair and
maintenance costs, although the engines had to undergo more detailed checkup and overhaul.
In order to conduct the work more efficiently and smoothly, a new class of engines started
to be designed, in accordance with the so-called modular conception, which made aircraft
engines flexible in terms of their exploitation, maintenance and diagnostics.
Modular construction engines enable:
- easy access to the components, systems and engine modules;
- easy adjustment
- maintainability
- safety of conducting maintenance tasks
- easy replacement of engine components, systems and modules
- unification and normalizing of engine components and subsystems. [4]
The engines, powering the F-16C/D aircraft and exploited in Poland since 2006, are F100-
PW-229 manufactured by Pratt&Whitney company.

13
 F100-PW-229 engine construction

Towards the end of the 1980s, through another modernization, the Pratt&Whitney
company constructed the F100-PW-229 engine. Its design used the latest advances of engine
technology, implemented from the F119 and F135 engines. The engine F100-PW-229 is an
afterburning turbo-fan (twin turbine), equipped with a 13 stage axial compressor, annular
combustor, four stage turbine, afterburning chamber with a mixer and an adjustable
convergent-divergent nozzle.

Fig. 1 F100-PW-229 engine [3]

The basic technical data of the engine are as follows:

- engine mass 1732 kg
- maximum thrust: 7918 daN
- maximum thrust with full afterburning 12945 daN;
- length of engine 5.283 m;
- inlet diameter 0.88 m;
- maximum diameter 1.18 m;
- bypass ratio 0.36;
- air mass flow 120 kg/s
- overall pressure ratio 32
- specific fuel consumption (maximum) approx. 0.7 kg (daN x h);
- specific fuel consumption (full afterburning) approx. 2 kg/(daN x h).
The F100-PW-229 consists of five modules:
- inlet fan;
- core engine;
- fan drive turbine;
- augmentor and nozzle;
- engine gearbox.

14
 Apart from the above-mentioned modules, the engine comprises a number of components
which are not integral module parts and therefore can be easily replaced. They are the so-
called non-modular parts. [3]

Fig. 2 Modular construction of F100-PW-229 engine [2]

The fan is a three-stage axial compressor with pressure ratio measuring four. The
compressed air, behind the fan, is divided into two separate ducts. Part of the air reaches the
external duct (so-called bypass air), whereas the rest flows into the high pressure compressor.
The fan module in its rear part is connected to the intermediate case, constituting an integral
part of the core engine. The fan is the part of the engine which is particularly prone to damage
caused by foreign objects that may enter the engine via air inlets. The fan module comprises a
number-one bearing.
The aim of the core engine is to produce hot gases and direct them at the fan drive turbine.
The gases which leave the combustor transfer some of its energy to the high-pressure turbine,
which powers the high pressure compressor and the engine gearbox. In terms of its
construction, the core engine is made up with four systems – intermediate case, high-pressure
compressor, combustion chamber and high pressure turbine.
The low-pressure turbine (fan drive turbine) transforms part of the internal gases energy
into mechanical work, essential for the fan rotor. This module includes a long turbine shaft
and number-five bearing.
The augmentor with an adjustable nozzle enables to diffuse the gases stream with the
external air flow. It also directs this mixture towards the exhaust nozzle, where it
decompresses to atmospheric pressure levels, accompanied by significant stream acceleration
(in its maximum value and after-burning, the stream reaches supersonic speeds). If the engine
operates on after-burning, in its after-burning section, additional amounts of fuel are
combusted, which in turn leads to a temperature increase and raising gases pressure – this is
when gases are exhausted from the exhaust nozzle at increased velocity, consequently
translating to higher engine thrust.
The engine gearbox is used to power engine components (fuel pump, oil pump and engine
generator). It also enables to forward the drive between the engine and the airframe. The
engine gearbox takes the power from the high-pressure rotor by means of the vertical shaft,
which is later transferred to the engine drive gearbox of the airframe (ADG) through the
horizontal PTO shaft – during regular engine running, or in the reverse direction, while
starting the engine. The gearbox is attached in three points to the bottom side of the
intermediate case.

15
 ,QIOXHQFHRIPRGXODUFRQVWUXFWLRQXSRQWKHVFKHGXOLQJRI)-DLUFUDIWoperation

The achievements in the field of commercialization and modularization of drives exploited

in civilian aviation have made a noticeable impact on the construction and manners of F100-
PW-229 engines exploitation. This type of engines requires extremely well-qualified
maintenance personnel, who are capable of immediate controlling and monitoring the
engines’ technical condition, as well as repairing minor and major malfunctions and testing
repaired engines. Therefore, the personnel of maintenance and engineering service in air force
bases had to be increased by admitting engine specialists in air and technical squadrons. In the
first place, during daily operation of aircraft, engine specialists are responsible for the current
checkup of the engine technical condition, removal of minor malfunctions and failures,
assembly and dismantling of engines from the airframe. In technical squadrons, engine
specialists perform periodical engine maintenance and repair of major engine failures. The
repair and maintenance team in Krzesiny deliver repairs that are connected with engine
dismantling into modules (exchanging or repairing compressor blades and turbines,
exchanging faulty bearings, injectors, seals, etc.); they also test engines on engine test facility
on completion of repair.
The presence of engine specialists during each and every flight causes that engines are
quickly and efficiently diagnosed after each flight and the discovered malfunctions are rapidly
removed. In case of more serious failure or malfunction, which cannot be repaired in a short
period of time, the engines are changed for an new one within one day. Meanwhile, the faulty
engine is being repaired in the most optimal way. During the last seven years of F-16s
operations in Poland, only five additional engines and one set of extra engine modules have
been able to fully secure the exploitation of 48 aircrafts in two air force bases. Several dozen
engine specialists delivered several dozen dismantles and assemblies of engines, several
engine repairs, exchange of faulty components, where the engines had to be dismantled into
modules. The repairs were performed either through exchanging faulty items (such as the
removal of over one hundred compressor and turbine blades) or repairing them (for instance,
approximately one hundred blades were repaired by blending). If an aircraft is to be fully
operational after its repair, the manufacturer requires that it undergoes testing on a stationary
engine test facility. Therefore, for a number of years, a specialist engine test facility called
“hush house” has been in use in Krzesiny air force base.

,PSDFWRIPRGXODUFRQVWUXFWLRQXSRQ)-DLUFUDIWRSHUDWLRQDOFRVWV

The modular engine construction as well as operations in accordance with its technical
condition led to extension of the service life of the F100-PW-229 engine in comparison with
the Russian engines, mounted on the Su-22 or the Mig-29. The operations of $à-21F3 and
RD-33 engines are based on their service life. The number of engine operating hours equals
1,600 hrs, whereas engine life between overhauls are held every 400 hrs. F100-PW-229
engines have their durability specified in the so-called TAC (Total Accumulated Cycles). In
case of an engine used in a civilian (commercial) aircraft, usually one TAC is an equivalent to
one flight. If a flight lasts three hours, the engine consumes approximately 1 TAC. F100-PW-
229 power combat aircraft, whose flight characteristics considerably differ from the one of a
civilian aircraft. A pilot of a combat aircraft during take-off and climb frequently uses the
afterburner; while performing maneuvers, the engine often changes rotational speed, which
consequently leads to the changes in gas temperatures in the engine, and ultimately in
components of the combustion chamber, afterburner and the exhaust nozzle. Therefore, for
military aircraft the average TAC consumption has been assumed as 1 flight hour = 2.5 TAC.
In case of the Polish F-16s, after seven years of their operation, the above-mentioned value
proves to be genuine.

16
 The F100-PW-229 engine manufacturer specified their service life between overhauls at
4,300 TAC. This value is identical for all engine modules, except the engine gearbox. Having
assumed that one hour of flight consumes 2.5 TAC, it appears that 4,300 TAC translates to
1,700 flying hours. After this time, the aircraft are eligible for an overhaul. Between 2007 and
the end of 2013, the most frequently operated F-16 aircraft, in Poland, flew 1,000 hours.
Provided that the intensity of operations is sustained in the following years, it is no earlier
than the year 2018, which is after eighteen years of their operations, that the first engines will
be scheduled for an overhaul. (Fig. 3)

Fig. 3 Current and estimated flying hours of the most frequently exploited F-16 aircraft

While comparing the F100-PW-229 operations with the RD-33 operation, it becomes evident
that in the first 12 years:
- the F100-PW-229 engine will not require major maintenance and overhaul costs;
- one RD-33 engine will undergo three overhauls and a new one must also be
purchased.

Fig. 4 Comparison of F100-PW-229 and RD-33 engine overhauls

Procurement and maintenance of a modern combat aircraft is a huge expense for the state, and
therefore it must be exploited for a period of at least 30 years. The operational costs are
largely connected with the power unit – its durability, current maintenance and overhaul costs.
In commercial aviation, it is assumed that the power unit consumes 35-40% total aircraft
maintenance and repair costs. [1] In military aviation, this share is quite similar. Figure 5
depicts a comparison of the frequency of overhauls and procurements in new aircraft, in the
period of 30 years, for F-16 and MiG-29 aircraft, providing that the number of annual flying
hours remains on the same level.

17
 Fig. 5 Comparison of the frequency of overhauls and procurements in F-16s and MiG-29s

The diagram shows that

- in order to secure the maintenance of one F-16C/D aircraft, in the span of 30 years, it
is enough to possess one engine, which will undergo two overhauls;
- in order to secure the maintenance of one Mig-29 aircraft, in the span of 30 years, it is
enough to possess six engines, which will undergo eighteen overhauls.
Concluding, it seems that in case of the Polish F-16s which are to conduct air operations
for the period of at least 30 years, two cycles are sufficient, around the year 2018 and 2030,
when each engine will undergo an overhaul. Due to this relatively low number of overhauls
(maximum 96 overhauls), it is not economically feasible to schedule engine overhauls in
Poland. All the engines will, therefore, be overhauled outside of Poland.

Conclusion

Based on the above-mentioned facts and comparisons of modular construction F100-PW-

229 engines, exploited in Poland in accordance with their technical state, it is possible to
observe a number of positive aspects of their use, such as:
- reducing the time when the engine is not operational;
- in case of malfunctioning on one engine element, it is possible, within a short period of time,
to replace only the engine module, which includes a particular component, without
stopping the operation of the whole engine;
- high qualifications of the maintenance personnel enable to solve almost all failures on site,
without involving the headquarters and technical institutions;
- running maintenance shops, even on an average level, means that it is not necessary to send
engines for emergency repairs, which significantly affects lowering the costs of operating
the system of engine exploitation;
- a modern engine with modular construction may boast of prolonged exploitation, and in
consequence significantly reduce total costs of service life of the F-16 product.

Readings

1. Ackert S.: Engine Maintenance Concepts for Financiers. Elements of Engine Shop
Maintenance Costs, 2011
2. F100-PW-229 Power plant – Customer training. Lockheed Martin, 2006.
3. Królik M.: 3áDWRZLHFLVLOQLNVDPRORWX)-16, 'ĊEOLQ.
4. 3U]\E\áHN 3 .RPRUHN $.: Modularyzacja w budowie silników lotniczych wojskowych
statków powietrznych, 2009

18
 ACOUSTIC EMISSION AS A TOOL FACILITATING TECHNICAL
CONDITION INSPECTIONS OF TANKERS CARGO TANKS
Marek Kula

*GDĔVN8QLYHUVLW\RI7HFKQRORJ\
)DFXOW\RI2FHDQ(QJLQHHULQJDQG6KLS7HFKQRORJ\

7KLV DUWLFOH GLVFXVVHV VHYHUDO DWWHPSWV DW XVLQJ $( DV D GLDJQRVWLF VLJQDO DQG D PHWKRG IDFLOLWDWLQJ WHFKQLFDO
FRQGLWLRQLQVSHFWLRQVRIWDQNHUVFDUJRWDQNV

KeywordsDFRXVWLFHPLVVLRQFRUURVLRQFDUJRWDQNVWDQNHUV

1. Introduction

Anywhere in the world there is a large number of ships carrying crude oil. Tankers are
ships especially designed for holding as much liquid cargo as possible. Despite regulations
aimed at increasing the safety of crew and cargo, maritime disasters still occur. The currently
used methods provide a relatively good overview on the diagnosed structure, however, they
certainly can be improved.

Tankers characteristics.

&UXGHFDUULHUV are used solely for transporting crude oil. They are usually built with one
or two longitudinal bulkheads and a relatively small number of cargo tanks – 15-20, no more
than 36.
2UH-RLO FDUULHUV are adapted to transporting ore in central bins. The main difference
between them and conventional tankers is that central bins are located over the double bottom
and are enclosed from the top with large and heavy hold covers. [1]
%XON-RLOFDUULHUV. The main difference between them and the above-described carriers is
the fact that they do not carry oil in lateral tanks, which are used solely for ballast. Their
central bins have a greater width, so their lateral tanks are narrower than in typical ships. Such
vessels usually have between 50 000 and 250 000 DWT.[1]
3URGXFWFDUULHUV are built considering maximum operational flexibility in view of using
their full capacity. They usually contain two longitudinal bulkheads and a separate ballast
system. Their system of cargo pipelines is also more complex than in the case of crude
carriers [1].

Issues of hull and hold damages.

The four basic mechanisms leading to the damaging of the hull are:

19
- ship overload that causes deformations when the material of which the hull is made is
subjected to stress exceeding it local strength,
- various cracks caused by material fragility, due to poor condition of the structure or
severe corrosion,
- material fatigue, fatigue caused by corrosion, due to recurrent hull overloads
resulting from operation,

Corrosion damages

Metal corrosion may be generally classified into chemical and electrochemical

corrosion. Chemical corrosion takes place when there is a direct chemical reaction of metal
with its surroundings, while electrochemical corrosion occurs in electrolyte solutions and in
the atmosphere. Most materials used in shipbuilding are subjected to electrochemical
corrosion due to sea water and the atmosphere.
The course of a corrosion processes may be represented schematically, as shown in
Figure 1 below.

)LJ&RXUVHRIHOHFWURFKHPLFDOFRUURVLRQRI
LURQLQDQHOHFWURO\WHGURSGXHWRDHUDWLRQ>@

Typical areas especially exposed to corrosion in oil tankers are: [4]

- upper section of lateral and longitudinal bulkheads,

- upper section of lateral deck bracing,
- longitudinal deck bracing,
- longitudinal bracing on internal plating and longitudinal bulkheads,
- openings for bracing crossings,
- transverse girder web plate,
- longitudinal bracing shelves in bottom and floor structures.

Factors affecting the development of pitting corrosion are:

- adjacent, heated tanks,

- local loss of corrosion protection as a result of poor quality of hull construction,
- incorrect cargo distribution,
- excessive local load,
- areas with increasing flow, i.e. hatches and openings.

2. Acoustic emission (AE) as a supplementary method for diagnosing the technical

condition of cargo tanks.

Strong AE signals occur most of all in the case of:[5]

20
ƒ partition cracking at higher stresses and rapid spreading of cracking,
ƒ non-uniform spreading of cracks due to loads,
ƒ high pulsating loads close to the ultimate strength,
ƒ cracking as a result of stress corrosion,

)LJXUH*UDSKLFLOOXVWUDWLRQRIDFRXVWLFHPLVVLRQPHWKRG>@
Instrumentation and software

Modern AE analysers comprise the following functional elements:

 converter of signals into an electrical signal,

 main amplifier with regulated gain,
 low- and high-pass filters system,
 noise cut-off system,
 signal energy or amplitude processing system,
 systems enabling the connection of AE measured signals loggers,

The AMSY5 system is a multi-sensor AE system comprising parallel measurement

channels. Each channel is connected to an analogue measurement section, digital signal
processing.
A complete AMSY5 system includes:

- main unit (MC2, MC6, MC15, etc.),

- one or more AE channels,
- basic software package – SWAT or SWBN. [6]

The components of an assembly used for analysing AE are listed below (fig. 4-9).

21
 )LJ$06<HTXLSPHQW
)LJ$(3SUHDPSOLILHU JDLQXSWRG% 

)LJ6HQVRUVIRUWHVWV

)LJ0$*0$ORJJHU

)LJ0DJQHWLFKROGHUV
)LJ0$*0$3&0&$SURFHVVLQJXQLW

Tests

The AMSY-5 equipment, described in the previous section, was successfully used for
diagnosing the technical condition of the following objects:

- "Icarus" tanker
- Polish navy fleet support ship – 253%DáW\N
- tank for storing brine of the capacity of 3000 m3

The application of measurement instrumentation on board of "Icarus" in 2005 was the

first case of the tests conducted in marine conditions. The aim of the tests was to measure the

22
acoustic background of MT IKARUS II at the anchorage by the "Baltic Beta" platform during
PDQRHXYUHV LQYROYHG LQ LWV ILUVW YR\DJH WR WKH SRUW LQ *GDĔVN 3UHSDUDWLRQ IRU WKH WHVWV
involved the construction of cable routes and a special seal for under the cover of the ballast
tank hatch.
Large amounts of flammable and explosive gases are released from the tanks of M/T
IKARUS II while extracting crude oil from the B-3 deposit.

Technical characteristics of M/T IKARUS II.

M/T IKARUS II was constructed in 1975 in Japan by KANASASHI SHIP-BUILDING.

The ship is currently being used as a resource base for the production centre located on the
"Baltic Beta" platform and as a carrier for oil extracted from the B-J deposit to the port in
*GDĔVN

Capacity – GRT – 21 576; Main engine type - Kawasaki M.A.N K

Capacity – DWT – 36 194; 7SZ 70/125,
Total length – Lc = 182.03 [m], Main engine power - M = 79 650 [kW],
Length on waterline – Lwp = 171 [m], Speed - 10 [Mm/h] at 145
Width - B = 27 [m], [RPM],
Draught - T = 17 [17],

)LJ07,.$586,,DWVHD

Tests on board of M/T IKARUS II were conducted in the following manner.

Selection of tanks filled with oil from the B-3 deposit for analysis.

Considering safety, the following tanks were selected for the tests:

- Tank No. 1 of the capacity of 6 710 m3, located in the centre between wing tanks,
between frames 72 and 80;
- Wing tank No. 1 of the capacity of 2 767 m3, located on the port side, between frames
72 and 80.

The tanks were filled with oil from the B-3 deposit up to 98%

23
 The figures below show the places and the method of AE sensors distribution.

)LJ0DUNHGDQGFOHDQHGDUHDVIRUILWWLQJVHQVRUVDW&PHDVXUHPHQWSRLQWDQGPDUNLQJRIVHQVRUW\SHVDQG
QXPEHUVRQWKHVXUIDFHRIDZLQJWDQNEXONKHDG

- VS 30-V and VS 150 RIC sensors with measurement cables at "D" measurement point on
the surface of the ship sides plating
- VS 30-V and VS 75-V RIC sensors with measurement cables at "E" measurement point on the
surface of the central tank bulkhead

An assembly by Vallen, type AMSY5, was used for the measurement of the acoustic background,
containing:

- 16-channel measurement system;

- laptop docking station;
- laptop;
- 8 VS 30-V sensors with a preamplifier and magnetic holders;
- 4 VS 75-V sensors with a preamplifier and magnetic holders;
- 4 nagar magnetic holders;
- 16 BMC measurement cables of the length of 125 m.

)LJ  -FKDQQHOPHDVXUHPHQWV\VWHP DQGDGRFNLQJVWDWLRQZLWKDODSWRSLQDERZURRP

Having connected measurement cables, sensors were calibrated and the preliminary value of
discrimination on individual channels was determined.

24
- value of discrete emission, AE events count [HIT] – a basic parameter describing the source
activity.
- Average intensity of continuous emission signals [RMS] – a basic parameter for specifying
continuous emission.
- Amount of oscillations above the discrimination level, i.e. exceeding the discrimination level,
and their rate in a unit of time,
- Amplitude of a discrete emission signal, the best parameter for specifying attenuation.
- Power (energy) of the discrete emission signal to various sensors,

With the recording of elevated parameters, the difference between the discrete emission signal travel
times was also recorded (area under the AE signal breakdown).

Sample results of background tests are presented below.

)LJ6DPSOHUHVXOWVRIVLJQDOVGHULYLQJIURPFRUURVLRQDUHSUHVHQWHGEHORZ

25
 )LJ7KHUHVXOWVRIWHVWVHQDEOHGWKHGLVFRYHU\DQGORFDWLRQRIDUHDVPRVWFRYHUHGZLWKFRUURVLRQ'HVSLWH
QRLVHDQGPDQ\LQWHUIHUHQFHVWKHUHVXOWVZHUHXQHTXLYRFDO

Another test object was a Polish navy fleet support ship – 253%DáW\N". The aim of the
tests was to analyse below-deck corrosion, test the corrosion tester and the GUT fatigue tester
on the ship.
7HVWVRQERDUGRI253Ä%DáW\N´LQFOXGHG$(PHDVXUHPHQWVLQ9WDQNDQGFRIIHUGDPV
and their location with the interferences of the acoustic background from:

- operating auxiliary mechanisms, generators, main engine, screw propeller and others,
- commercial and navy ships travelling near the ship during its stoppage at the quay and anchorage
LQ*GDĔVN%D\
- sea waves and wind during manoeuvres at sea and on the anchorage,

253%DáW\NZDVFRQVWUXFWHGLQWKHVKLS\DUGRIWKH3ROLVK1DY\LQ*G\QLD2NV\ZLH
Its task was to provide fuel, oils and fresh water, as well as collect fuel and waste.

26
 7DE253%DáW\NWHFKQLFDOGDWD

Displacement D = 2984 t
Load carrying capacity 1276 t
Length L = 84.75m
Width

B = 13.07 m
Draught T = 4.80 m
Engines power N = 2u1480 kW
Speed V = 15 Mm/h
)LJ253%DáW\NDWVHD

Equipment used for the tests:

- Valen AMSY5 with ASIP2 filters

- New ISAS75 sensors,
- SE25-P sensors,
- Corrosion source – GUT TESTER, and fatigue GUT tester,
- Corrosion solution (2 L of water, 0.3 L of 65% HNO 3 , 0.25 L of 98% H 2 SO 4 , 0.07 kg of NaCl)

)LJ&DUJRGHFNRI253%DáW\NZLWKVHQVRUVDUUDQJHGIRUWHVWV

)LJ  9LVXDOL]DWLRQRIWHVWVUHVXOWV9LVLEOHVLJQDORIFRUURVLRQ>@

27
Tests results

The conducted tests were successful. Corrosion sources were recorded and located,
without having to empty the tested tanks. The use of acoustic emission as a diagnostic signal
supplements the currently used methods.
One of the overground objects covered by the tests was a brine tank in a lye facility in
Kosakowo. The aim of the tests was to record AE signals in the tank structure side surface
during its operation.

)LJ%ULQHWDQN 9 P
Instruments by Vallen Systeme GmbHn were used to measure AE signals:

- 16-channel AMSY5 system,

- VS75-V, VS30-V measurement sensors (7 + 7 pieces),
- AEP preamplifiers,
- magnetic holders and cabling

AE SYSTEM ASSEMBLY AND PREPARATION FOR TESTS

Monitoring of the tank condition during the tests of its operation using acoustic
emission (AE) was conducted using standard procedures.
Distribution of sensors on the tank plating was preliminarily calculated and the sensors
were fitted onto the tank plating:

1. The first 7 VS30-V sensors were evenly distributed every L1 = 6.88 m along the
circumference, at the distance of h1 = 0.6 m from the tank base. Sensors from 1 to 7 were
numbered (clockwise).
2. Another 7 VS75-V sensors were evenly distributed every L1 = 6.88 m along the
circumference, at the distance of h2 = 4.1 m from the tank base. Sensors from 8 to 14 were
numbered (clockwise).
3. Sensors were installed with AEP4 preamplifiers in the designated places on the tank
plating using magnetic holders. Then they were connected using cables with AMSY-5
instrumentation.

Arrangement of sensors on the first and second day of measurements

28
 )LJ$UUDQJHPHQWRIVHQVRUV
Procedure for preparing the AE system for the tank tests was as follows for both days of
tests:

1. Sensors calibration test and measurement system check.

2. Acoustic background level specification.
3. Acoustic background measurements during the tank operation for two variants – with
water and brine inflow and drainage systems in on and off states.
4. AE signals measurement using a system of sensors.

TESTS RESULTS

The level of acoustic background specified for the measurements on the first day of tests
equalled 22 dB. This value remained unchanged for the second day of tests. Weather was
changeable with rainfall spells during the preliminary works on the first day.
Representative results of the AE signals tests in the tank and on its side surface – during
normal operation and with water and brine inflow and drainage systems in an off state – are
presented below.

)LJ$FRXVWLFEDFNJURXQG

29
 )LJ$FWLYLW\RILQGLYLGXDOFKDQQHOV

3. SUMMARY AND CONCLUSIONS

As part of preparations for the tests, measurement instrumentation and sensors were
checked, measurement system was dimensioned, and the places of measurement sensors
distribution were included in a coordinate system in the VISUAL AE system. Preliminary
tests were conducted, and the THRESHOLD value was determined to be 22 dB. This enabled
proper collection of measurement signals providing for the assessment of the tested
phenomenon.
Several measurement tests were performed. The most representative ones were presented in this
article. The tests commenced with the analysis of acoustic background, followed by a series of tests,
analysing the quality and nature of the collected AE signals. Final conclusions drawn from the
performed tests of the brine tank structure using the acoustic emission method are as follows:
· The tested brine tank is tight (no leaks), and its side walls (cylinder side walls) are
devoid of active corrosion centres.
· Energy of the collected AE signals emitted during the tests is relatively low due to
very strong attenuation by individual layers on the tank (paint, membrane).
· The obtained results of AE signals analysis (using the Visual AE module in the Vallen
software) enable the correct assessment of the processes taking place and the proper
verification of the condition of the tested tank structure.
The usefulness of a new generation of measurement sensors for tests was confirmed.
Those new sensors may be installed even in poor weather conditions without the risk of
damaging the measurement system or corrupting the tests results.

The tests were conducted as part of CORFAT project (Contract No. SCP7-GA-2008-218637 – Cost
effective corrosion and fatigue monitoring for transport products) performed as part of the European
Union FP7 framework program.

4. REFERENCES

1. R.J. Beck, Supply Management By Major Exporters Key To Market In Years Second Half, Oil
Gas, 1998.

30
2. www.ABS.com.
3. P.A. Caridis, Inspecion, Repair and Maintenance of Ships Structures, Witherby & Co Ltd.,
London, 2002.
4. Oli Companies International Marine Forum, September 1997.
5. $ 'RPDĔVNL - %LUQ .RUR]MD RNUĊWyZ L MHM ]DSRELHJDQLH :\GDZQLFWZR 0RUVNLH *GDĔVN
1970
6. ,0DáHFNL-5DQDFKRZVNL(PLVMD$NXVW\F]QDħUyGáD0HWRG\=DVWRVRZDQLD:DUVDZ
7. - 6LHGODF]HN (PLVMD DNXVW\F]QD Z\EUDQ\FK PHWDOL L VWRSyZ Z SURFHVLH GHIRUPDFML
SODVW\F]QHMLSURSDJDFMLSĊNQLĊFLD]PĊF]HQLRZHJR3$1:DUVDZ
8. F. Mansfeld, P.J. Stocker, Acoustic Emission from Corroding Electrodes, 1973
9. Specyfikacja aparatury pomiarowej AMSY-4. Vallen – Systeme GmbH, 2001.
10. CORROSION TESTING OF SHIPS : No evg1-CT-2002-00067. Badania laboratoryjne emisji
DNXVW\F]QHM SURFHVyZ NRUR]\MQ\FK LQLFMRZDQ\FK Z SURFHVLH NRUR]ML HOHNWURFKHPLF]QHM QD SUyENDFK
Z\NRQDQ\FK]HVWDOLNDGáXERZ\FKAuthors: +%XJáDFNL:'DUVNL3(LFKHUW*GDĔVN0DUFK

31
32
 AN EXAMPLE OF A TECHNICAL OBJECT OPERATION PROCESS
MODEL DESCRIBING INFLUENCE OF ENGINE DAMAGES ON THE
OPERATION PROCESS COURSE

Bogdan Landowski

University of Technology and Life Sciences

Faculty of Mechanical Engineering
ul. Prof. S. Kaliskiego 7, 85–791 Bydgoszcz, Poland
tel.+48 52 3408208, fax: +48 52 3408462
e-mail: lbogdan@utp.edu.pl

Abstract
The authors of this paper present, a method for building a model of technical objects operation whose sequence of
successive operation states, their duration times, incomes and costs connected with the objects being in these states
and the way a given state is reached depend not only on the object current state but also on other factors.
The model describes, among others, influence of damages of the powertrain, being a compression-ignition
internal combustion piston engine, of the analyzed object on the operation process of that object. The analysed
technical object is a means of transport (a vehicle).
A simplified computing model has been presented in order to illustrate the discussion.
The model of operation process was built basing on the analysis of spaces of states and operation events
concerning technical objects used in a real transportation system. In result of identification of the analyzed system and
its multi-state process of technical objects operation, operation states and possible transitions between these states,
significant for this research, have been determined.
Sets of source data indispensable for the model assessment and its initial verification were obtained on the basis
of experimental tests with the use of passive experiment method from a real research object.
In order to perform mathematical modelling of the technical object operation process the Markov decision
process was applied.
Keywords: operation process, Markov decision process, state of object, operation state, damage

1. Introduction
The authors of this paper present, a method for building a model of technical objects operation
whose sequence of successive operation states, their duration times, incomes and costs connected
with the objects being in these states and the way a given state is reached depend not only on the
object current state but also on other factors. A simplified computing model has been presented in
order to illustrate the discussion.

33
 A natural model of a vehicle (technical object) operation is a random process with a finite state
space S and a set of parameters R+ (subset of natural numbers t 0) [5, 11, 12]. Homogenous
stochastic processes, including Markov and semi-Markov processes [10, 11, 12, 13] are commonly
used for modeling of operation state changes. This is an oversimplification of real processes. In
result of identification of the process of urban bus transportation it was found to be a
nonhomogeneous process. Also, due to the research objective and the need to model a sequence of
the object states, whose changes depend not only on the previous states but also on other factors,
the operation process was analyzed using the theory of Markov decision processes.
Stochastic model {X t , D t , tT}, tt0,being a specific case of Markov decision process, was
used. In practical applications it must be decided whether there are reasons to reject the
assumptions connected with the mathematical apparatus.
It has been assumed that each of the operated and maintained technical objects, at the specific
moment in time t, t >0, may be in only one of the distinguished states, forming a finite set of the
operation and maintenance states of an object.
The model of the process of changes of the maintenance states of technical objects is the
stochastic process {X t , tT}, t t 0 with finite set (space) of states S ^1,2,3,..., n`.
Probability of changing the state in a single process step {X t , tT} from the state i, iS to the
state j, jS, provided that the condition a, aA is met is denoted as p ija , ¦ p ija 1 , p ija t 0 , for
jS

i, j  S and aA.
The sequence of the process states {X t , tT} is a non-homogeneous Markov chain.
The stochastic process described that way is a special event of a non-stationary Markov
decision process.
Due to the character of this paper, only selected assumptions of the simplified vehicle
operation process model have been presented. The values of indices characterizing the analyzed
process are determined with the use of computer simulation of Markov decision process, being a
mathematical model of technical objects operation process.

2. Mathematical model of the operation process

In the next section the main assumptions accepted for the description of stochastic process
{X t , D t , tT}, tt0,being a mathematical model of the analyzed operation system, are discussed.
It is assumed that the process of operation state changes can be described by a stochastic
process

{X t , tT}, t t 0, (1)
with a finite state space:
S = {1, 2, ..., n}, nN, (2)
where:
N – set of natural numbers ,
T – subset of real numbers.
For the needs of this research, it is accepted that the states of the discussed stochastic process
correspond to operation states of the technical object (vehicle).
Alternative k, accepted upon entering state i, is denoted as (iS, kN). Finite set A i , of
alternatives (decisions) corresponds to each state i, i  S.
Sets of alternatives have to be equal for each state in terms of quantity. Also states, for which
the set of alternatives is a single-elements set, can occur. States, for which there is no possibility to
choose an alternative, are referred to as non-decision states. Some literature sources refer to non-
decision states as to such for which the set of alternatives in s an empty set [4]. It appears that

34
acceptance of one-element sets of alternatives is more consistent for a description of the analyzed
process {X t , D t , tT}, tt0.
It is accepted that elements of set A i , iS are elements a ik , (iS, kN), that is:
A i = {a i,1 , a i,2 , ..., a i, i }, (3)
where:
i - denotes capacity of set A i .
The set of all the subsets of alternatives is denoted as A, that is:
A  Ai . (4)
iS

Generally, application of a given alternative upon process {X t , tT} entering i  S State, can
have an influence on the process successive state jS and the state features (time of being in the
state – type of distribution and its parameters, costs or profits obtained by the systems in this
particular state, etc.).
The alternatives can represent given modes of operation, events, decisions, etc. which can be
assigned to the state of the modeled process [6, 7, 8, 9, 12, 13, 14]. In a real system of operation
there can be different ways of maintenance, repair, surveys, operation modes and scopes, e.g.
different transportation routes, on which the vehicle is used. Acceptance of a given alternative can
affect: costs, income, frequency and kinds of failures, times of operation states, sequences of
states, etc. [12, 13, 14.]
Stochastic process:
{D t , tT}, t t 0, (5)
with a finite set of alternatives A describes the selection method of aA alternatives.
Change of process{D t , tT} state occurs in times t, of process{X t , tT} state changes, In
times t n Qࣅ1RISURFHVV^;t, tT} state changes alternative aA is chosen. If in time t n , state i is
the state of process {X t , tT} then a  A i .
Process ^X t , D t ` , t t 0 with a finite state set S and finite set of alternatives A, is called a
stochastic decision process. In result of this process, a sequence of states and decisions is obtained
from initial time t 0 to time t n :
h t n {i t 0 , a t 0 , i t1 , a t1 , ... , i t n , a t n ) , (6)
which is called the process history until time t n .
Further, it is assumed that the analyzed stochastic decision process:
^ `
X tn , D tn n 1 ,
f
(7)
is Markov decision process. A set of possible implementations of Markov decision process is set
W ^S u A` . It is also assumed that the probability of choosing alternative a t n  A depends
f

merely on state i t n 1  S , and does not depend on the process history h t n 1 .

In this case the sequence of process {X t , tT} states is a nonhomogeneous Markov
chain [1, 2, 3].
In order to define the analyzed stochastic decision process {X t , D t }, tt0 it is also necessary to
define:
- method for choice of alternatives for process {D t , tT}.
- initial distribution of {X t , tT},
- conditional probabilities of process {X t , tT} state changes,
- random variables of process {X t , tT} states times.
Simplifying, the rule used for determination of choice of alternative aA, upon entering state
iS, is referred to as a strategy. The manner of choosing the alternative upon entering the process
state can be of random or determined character.

35
Formula:
p >p1, p2 ,..., pn @ , ¦ p i 1 , pi t 0 , i  S , (8)
iS
is used for determination of {X t , tT}. Giving values of p i of elements of the initial distribution
vector p, determines the probability that, in time t, process {X t , tT} will be in state i.
The probability, that upon entering state i, process {X t , tT} will, in one step, change its state
from state i, iS into j, jS, for accepted alternative aA i , has been denoted as p ija .
Condition:
¦ paij 1 , paij t 0 , i, j  S , a  Ai ,
jS
(9)

is satisfied.
Stochastic matrix P i ,a , defining conditional probability of transition paij , can be assigned to
each state i  S and alternative a A i . The set of matrixes assigned to state i  S has capacity equal
to i (capacity of set A i ).
Matrix P i , a is a matrix made up of stochastic verses, which determine probabilities of
transition from a state marked with the verse number to all the remaining states. Element paij ,
situated on an intersection of a verse with number i and a column with number j in matrix P i , a , is
the probability of transition from state i  S to state j S as long as alternative a  A i has been
used upon entering state i  S .
Random variable, denoting duration time of state i  S of process {X t , tT}, when the
successive state is j  S , and when upon entering state i decision a  A i is made about distribution
defined by distribution function Fija t , is denoted as Tija .
For the purpose of simplification it was assumed that:
Fija t = Fia t = Fia t , i,jS, a  A i . (10)
This means that the duration time of state iS does not depend on the process successive state.
Function Fia t is a function of state i  S duration time distribution on condition that, decision a is
made upon entering this state.
Random variable denoting duration time of state i  S , with distribution defined by distribution
function Fia t , is denoted as T ia .
In order to evaluate economic aspects involved in the modeled operation process it is necessary
to additionally determine appropriate values of economic categories connected with the manner of
the process states entering and staying in them.
Moreover, it has been assumed that the model of changes of the maintenance states of the set n
of homogenous, from the point of view of the purpose of the bus investigations, are independent
processes {X t , tT}. The random vector X(t) = [X 1 (t), X 2 (t), …, X n (t)] describes the process of
changes of the maintenance states of the bus set [2, 3].
A computer program has been developed that facilitates simulation of execution of the
stochastic process described that way. When executing the simulation, values of the selected sets
of indicators enabling to analyse the modelled process of changes of states of the objects are
determined.

3. Investigation object
The paper analyses a real maintenance system of buses in an urban transport system in
a selected urban agglomeration. The essential purpose of the operation of the analysed system is
performance of the effective (in terms of technical and economic criteria) and safe passenger

36
transports by the operated and maintained means of transport in the determined quantitative and
territorial scope.
The scope of the performed repair of a bus is decisive for directing the bus to a diagnostic stand
in order to perform post-repair diagnostics. The post-repair diagnostics are particularly performed
after repairing the following systems and assemblies of a bus [8, 10, 12]:
- Steering system,
- Braking system,
- Engine,
- Truss,
- Front axle suspension.
If the inspection result is negative, the bus is directed again to the current repairs stand.
The following dependencies have been ascertained in the process of identification of the
maintenance system of buses in an urban transport system and the bus maintenance process being
performed in it, as well as on the basis of the analysis of the maintenance study results:
- Distribution of the random variable describing duration of the renewal state on the type of the
damaged bus subsystem,
- Distribution of the random variable describing duration of the diagnostics state on the type of
the damaged bus subsystem,
- Repair cost on the type of the damaged bus subsystem,
- Diagnostics costs on the type of the damaged bus subsystem,
- Sequence of the next maintenance states on the type of the damaged bus subsystem.
The type of the damaged bus subsystem has a significant influence on the course of the
maintenance process. Moreover, the applied methods of proceeding and sequences of actions
aimed at restoring serviceability of technical object as well as the measures to perform them
depend on the type of the damaged subsystem of that object.
The model describes, among others, influence of damages of the powertrain, being a
compression-ignition internal combustion piston engine, of the analyzed object on the operation
process of that object.
An assumption, that the identified set of technical objects, used in the analyzed system of
operation, can be divided into n separate subsets of objects, homogeneous in terms of the research
purpose, has been accepted. So classified subsets of technical objects are called: categories of
objects [11]. In practice, a given technical object (bus) is assigned to a given category on the basis
of the following criteria [8, 10, 12]:
- type of object (make and type of bus),
- operational potential,
- operation time,
- others.
Further, in this paper, one category of objects is discussed.
In result of identification of a real urban bus transportation system and its operation process,
three finite sets of states and operation-related events significant for an analysis of the system
[6, 8, 10], have been distinguished. In the computing model, selected subsets of these states and
events are analyzed.

4. Computing model
In order to illustrate the discussion, the following operation states of a bus have been analyzed:
S 1 -state connected with provision of transport services, that is a state in which a bus and its
operator perform the transport task;
S 2 - state of corrective servicing (restoring serviceability state);
S 3 -state of waiting for performance of transport tasks, that is standby of vehicles on the territory
of a bus depot while not being in operation;

37
S 4 -state of post–repair servicing, that is a state in which the object, after being repaired undergoes
control of the object condition (the so called post-repair diagnosing) and quality of performed
corrective servicing of the damaged subsystems.
In order to illustrate the discussion, influence of the type of bus subassembly damage on the
change of probability of transition between states and duration times of the states as well as costs
connected with the vehicle being in a particular state, has been modeled as well. An example of a
vehicle system, in the research object, the damage of which affects the sequence of the successive
operation states and their features is the powertrain. Damages to the engine fuel injection system
requiring adjustment of the injection pump occur with relatively high frequency. The procedure of
carrying out corrective servicing for that system, being applied in the enterprise under analysis,
provides for the necessity to perform post-repair diagnostics. The following major assumptions
have been accepted:
- After completion of corrective servicing of the vehicle subassemblies it is necessary to control
the object state, that is, state S 4 is possible to reach only after a repair of the vehicle identified
damage;
- If in result of control of the object state (state S 4 ), it is still found to be unable to perform its
transport tasks or carrying out the working process would be ineffective (e.g. excessive fuel
consumption due to incorrect adjustment of the fuel injection system) it is a subject to repeated
corrective servicing processes performed at the station where control of its state was performed
(it remains in state S 4 );
- The repair cost per time unit (state S 2 ) is connected with the type of failure;
- The repair duration time (state S 2 ) depends on the type of the damaged vehicle subassembly;
- Duration time of post-repair object (state S 4 ) depends of the bus state control result, that is
whether the vehicle is allowed to perform its transport tasks or needs to undergo another repair;
- The cost per time unit related to post-repair servicing (state S4) of the vehicle is related to the
result of the control of the vehicle condition;
Buses used in the research objects were decomposed into systems. For the needs of the
simplified computing model the vehicle systems (their elements damage) were divided into the
following subsets:
- Systems (their elements damage) characterized by low repair labor consumption, the repair
time is relatively short (frequently the process of repair is performed outside a bus depot by
units of the so called standby service) denoted by code U 1 ,
- Systems (their elements damage) which need post-repair service denoted by code U 2 ,
- Systems (their elements damage) which do not need post-repair service characterized by high
labor consumption, whose average time of repair is relatively long (multiple of system denoted
by U 1 code), denoted by code U 3 ,
- Engine fuel injection system (damages to the engine fuel injection system requiring adjustment
of the injection pump) for which post-repair servicing denoted by code U4 must be performed.
The data obtained by analysing results of preliminary tests performed in the considered
enterprise are used in the computing model.
In the analyzed example states i  S of process {X t , tT} correspond to the identified
operation states S i , i = 1, 2, 3, 4. States i = 1, 3 of process {X t , tT} are non-decision states, that
is, subsets of alternatives A 1 i A 3 are one-element subsets. In the same states, alternatives are of
only formal character (maintaining the notation consistence) and they have no influence on the
analyzed process {X t , D t }, t t 0 course. In state i = 2 of alternatives a  A 2 of {D t , tT} process
correspond to the codes of the bus distinguished damaged systems and represent their failures
(elements). Set A 2 contains the following elements A 2 = {a 2,1 , a 2,2 , a 2,3 , a 2,4 } . Interpretation of
entrance to state i = 2 of process {X t , tT}, in time t and occurrence of alternative a 2,3 of process
{D t , tT} is as follows; a bus failure occurred and its repair, and the damaged system is the
system denoted by code U 3 . In state i = 4 alternatives aA 4 of process {D t , tT} correspond to

38
the codes of the vehicle state control results. Set A 4 contains the following elements
A 4 = {a 4,1 , a 4,2 }. a 4,1 and a 4,2 were used to denote respectively the control which finished with
admitting the vehicle to perform transport tasks and recommending a post repair.
The rule for choice of alternatives a in state i is determined by distribution of probability of the
analyzed alternatives occurrence. It was assumed that q ik , i  S, k  N means probability of
alternative a ik occurrence upon entering state i.
Expression
q i >q i1 , q i 2 ,..., q ik @ , q ik t 0 , iS, kN, (11)
is used for denotation of vector of alternatives occurrence distribution in state i. Elements of vector
q i meet the condition:

iS, kN
¦ q ik 1 . (12)

In the considered example, elements of this vector for i = 2 denote probabilities of a given
system failure occurrence.

Fig. 1. Window used to enter the data for simulation – parameter values of state duration times depending on the
selected alternative
For the discussed assumptions, simulation of the process of a single technical object operation
involves simulating the described stochastic decision process, being a model of the process of
operation state changes of vehicles used in the analyzed research object.
A program enabling simulation of the stochastic decision process has been developed.
In order to perform the simulation it is necessary to use data indispensable for determination of
the described process {X t , D t } t t 0.
Simulation experiments have been performed to illustrate of the discussion.
The Fig. 1 shows one of the windows used to enter the data required to perform simulation
experiments.
The assumption that random variables T ia for i  S, a  A have gamma distributions with
different parameters, and their being in the process states is connected with gaining profits (state 1)
and bearing costs by the operation system, has been accepted for the needs of the simulations. The
values defining conditional probability of stochastic matrix P i , a i, j  S, a  A transition paij , have
been estimated on the basis of initial experimental tests.

39
 Values of probabilities q ik , i = 2, k = 1, 2, 3, 4 of alternative a ik occurrence have been
determined basing on data concerning the bus failures. For the description simplification,
alternatives a 2k are further denoted by code k.
The remaining values of parameters used for simulation experiments have been estimated on
the basis of results of initial tests performed in the research object.
It is necessary to accept that the parameters values of random variables T ia i S, aA, used in
the model, are of hypothetical character. The parameters values were estimated basing on a small
size data set.
Calculations for one category of objects consisting of 100 vehicles were performed, over a
period of 100 days. Selected calculation results are presented in the Figures 2 to 5.

Fig. 2. A part of a text file generated during simulation containing selected experiment results

2000

1500

1000
PLN

500

0
0 1 2 3 4 5 6 7

-500
time [days]

Fig. 3. An example of simulating the financial result obtained by the enterprise during the consecutive days of
carrying out the transport tasks

40
 120

100

number of vehicles
80

60

40 state code S1

20

0
0 1 2 3 4 5 6 7 8 9 10
time [days]

Fig. 4. Relationship between the number of the technical objects (vehicles) in the state S 1 and the time

State S2
2500,00
2075,00
Number of entries to the state

2000,00

1500,00

1000,00

500,00
244,00 235,00
68,00
0,00
1 2 3 4
Alternative code

Fig. 5. Number of entries to the state S 2

5. Conclusions
Presentation of the complete set of the input data required for simulating the analysed
stochastic process and the method to estimate the parameter values on the basis of the results of the
real operational data goes beyond the scope of this paper. Due to the nature of the paper and
extensive scope of the results generated only selected research results are presented for your
reference.
The purpose of the considerations was, among other things, to present possibilities of applying
the Markov decision processes for mathematical modelling of the system and process of vehicle
maintenance. A possibility of using models of this type to analyse a transport system and support
decision makers in that system, e.g. by forecasting behaviour of the vehicle maintenance system
after changing the control requests has been presented.
The results of the simulation experiments performed let us state that the model is susceptible to
a change of the value of its input parameters.
The analysis of the results of the simulation experiments shows a significant stability of the
calculation results obtained (for the same data).
Mathematical models of the maintenance processes, performed in complex systems, are
intrinsically a significant simplification of the real processes. The consequence of the above is
a necessity to carefully formulate conclusions resulting from investigations of those models
[5, 11]. However, it seems that the analysis of the results of the investigations of those models, for

41
the values of the model parameters, determined on the basis of the maintenance studies performed
in a real transport system, makes it possible to formulate both qualitative and (to the limited
extent) quantitative conclusions and opinions.
The set of the indices that may be determined includes subsets of the indices concerning:
readiness, repair times, effectiveness of performance of the transport tasks, costs and others.
Performance of the research experiments for various categories of objects makes it also
possible to evaluate the selected technical and economic aspects of replacing objects with new
ones and evaluation of usefulness of various types of objects in the specific maintenance system.

References
[1] Bernaciak K., Zastosowanie decyzyjnych procesów semimarkowskich do optymalizacji
HNVSORDWDFMLSHZQHJRXU]ąG]HQLD Zagadnienia Eksploatacji Maszyn, Zeszyt 3 (75), 1988.
[2] Buslenko, 1 .DáDV]QLNRZ W., Kowalenko, I., Theory of complex systems, PWN,
Warsaw 1979.
[3] Chrzan P., àDĔFXFK\ GHF\]\MQH 0DUNRZD L LFK ]DVWRVRZDQLD Z HNRQRPLL, Akademia
ekonomiczna, Katowice 1990.
[4] Kowalenko I. N., Kuzniecow N. J., Szurienkow W. M., Stochastic processes, Guide, PWN,
Warsaw 1989.
[5] Landowski B., Applying the Markov decision processes to model changes to the maintenance
states of an object, Journal of KONES Vol. 17, No. 3, 2010.
[6] Landowski B., Example of applying Markov decision process to model vehicle maintenance
process, Journal of KONES Powertrain and Transport, Vol. 20, No. 4, 2013, European
Science Society of Powertrain and Transport Publication, Warsaw 2013.
[7] Landowski B., Example of Markov decision process use for modelling of operation and
maintenance process. Interdisciplinary Integration of Science in Technology, Education and
(FRQRP\ 0RQRJUDSK HGLWHG E\ 6KDODSNR - DQG ĩyáWRZVNL % .KPHOQ\WVN\ – Jaremche
2013, ISBN 978-617-70-94-07-3.
[8] Landowski B., Method of determination values of the chosen decision variables to control
rationally the operation and maintenance process in the transport system, Doctoral thesis,
Academy of Technology and Agriculture, Bydgoszcz 1999.
[9] Landowski B., Simulation model of the means of transport maintenance process, Journal of
KONES Vol. 17, No. 4, 2010.
[10] Landowski B., Woropay M., Neubauer A., Controlling reliability in the transport systems,
Library of Maintenance Problems, Maintenance Technology Institute, Bydgoszcz-
Radom 2004.
[11] Woropay M., Grabski F., Landowski B., Semi-Markov model of the vehicle maintenance
processes in an urban transport system, Scientific Publishers of the Polish Scientific
Association of Automotive Engineering, Archives of Automotive Engineering Vol. 7, No 3,
2004.
[12] Woropay M., Knopik L., Landowski B., Modelling maintenance processes in a transport
system, Library of Maintenance Problems, Publishers and Printing Department of the Institute
of Technology and Maintenance, Bydgoszcz – Radom 2001.
[13] Woropay M., Landowski B., Neubauer A., Applying semi-Markov decision processes to
model and simulate the bus operation and maintenance processes, Scientific Publishers of the
Polish Scientific Association of Automotive Engineering, Archives of Automotive
Engineering Vol. 7, No. 1, 2004.
[14] Woropay M., Landowski B., Neubauer A., Simulation of the method to optimise the operation
and maintenance process of the buses in an urban transport system, International Scientific
Conference "Transport of the 21st Century ", section5, Warsaw 2004.

42
 CONDITIONS OF USAGE ANALISYS OF THE TWO-STROKE, SLOW-
RUNNING, HIGH POWER RATING, MAIN ENGINES AND THE CAUSES
AND EFFECTS OF THE TRIBOILOGICAL SYSTEMS DAMAGE
=ELJQLHZàRVLHZLF]

West Pomeranian University of Technology Szczecin

Faculty of Maritime Technology and Transport
Chair of Safety Engineering and Power Engineering
tel. 600 275871
e-mail: horn.losiewicz@wp.pl

Abstract

This work justifies the need for further research, both laboratory and real-time usage, on improving currently
operating diagnostic systems adapted to monitor ship's two-stroke, slow-running high power rating main engines in
order to prevent the damage of the main tribological systems of the engines, at least these mentioned in the paper. The
randomness of the variability of the load of the engines is highlighted. It is pointed that the load should be taken under
consideration during the period of usage of the engines as random processes, but the optional period of time of usage
as random variables. The models of these processes which are necessary to the research are offered to be elaborated
with the use of Semi-Markov model theory. The rationality of the proposition is pointed out with two hypotheses. The
first explains the relation between the mechanical load Q M (t) and the thermal load Q C (t) of the engines. The second -
the existence of the stochastic linear relation between the load Q M (t) and Q C (t) of the engines and the wear Z(t) of
their main tribological systems in such relation, that the value of the coralation coefficient r qz =1. In order to simplify
the actions made to improve the currently used diagnostic systems of the engines mentioned; the most relevant
description according to the author of chosen tribological system damage is given. The conditions of usage which
favor the described damage are also characterized.

Keywords: hypothesis, load, two-stroke engine, tribological system, damage, conditions of usage, wear

,QWURGXFWLRQ

The conditions of usage of the main ship engines, particularly two-stroke, slow-running, high-
power engines are more diversified in relation to other piston diesel engines in the main ship's
drives. The conditions are dependent not only on hydro meteorological sea conditions shaped by
factors such as [3]: the windforce and its direction, the state of the sea resulting from the height,
length and direction of the waves, the direction of the currents, the depth and the width of the
water region in which the ship is moving, the precipitation (rain, snow, hail), the icing, deck
flooding, etc. They are also dependent on maneuvering conditions in the ports and navigation
waterways which forces the reduction of rotary speed along with the engine power. This
negatively affects the load of its tribological systems such as piston-ring-cylinder tubes, main and
crankshaft bearings and crosshead bearings. The conditions change in a random manner, so the
load of the main engines, both thermal and mechanical have random properties. This is the cause

43
to consider the load when the engines are running as random processes and their research should
be conducted with the use of scholastic processes theory - especially the Semi-Markov theory [4,
5]. Accordingly, the load given in a specific moment of the ship's engines usage is considered
random variable. This means, that the values of the load cannot be precisely predicted, but their
probability of occurrence may be estimated. The load affects the tribological engine systems
negatively making the processes random . The processes lead to boundary wearing (superficial or
volumetric) of the engines and in result to their damage. These are the reasons to pursue the
recognition of the properties of engine load and its effects such as excessive wearing and damage
to the specific tribological systems of the main engines, which are the most important elements
when considering the durability and reliability of such type of engines.
This is taken under consideration in further parts of the paper, starting from the analysis of the
conditions of usage of the two-stroke, slow-running high power engines.

2 7KH DQDO\VLV V RI WKH FRQGLWLRQV RI XVDJH RI WKH  WZR-VWURNH VORZ-UXQQLQJ KLJK SRZHU
HQJLQHV

Recent research shows that some of the values considered as parameters (indicators) of work of
the piston diesel engines such as: average usable pressure (p e ), average piston speed (c ĞU ) [1, 6, 7,
8, 9], are characterizing both thermal and mechanical load. It is obvious that there is a relationship
between the mechanical load and thermal load. Due to the fact that the loads occurring in time are
random processes there is a conclusion that there has to be a stochastic relation between them.
Therefore a following hypothesis forms H1 [2] There is a stochastic relation between the
mechanical load Q M (t) and thermal load Q C (t) because specific versions of the mechanical load
Q M (t) accompany various versions of thermal load Q C. The conclusion is that the relation
between these load processes Q M (t) and Q C (t) cannot be described as a result of use of casual
algebraic equation [13]. The relation is influenced by many immeasurable factors [14, 16, 19, 21,
22]. Therefore the level of relation between Q M (t) and Q C (t) may vary. In conclusion, there is a
need of taking into consideration the intensity (force) of the stochastic relation between Q M and
Q C which requires long-term and costly statistical research. The load causes the wearing of the
tribological systems of every engine. So the course of wearing is also a random process which is
irreversible and its valuation is conditioned by the load of the engine [1, 7, 8, 9].
Therefore one can assume a hypothesis (H 2 ) [2] when rationally using a diesel engine, the
wearing (Z) of each tribological system and its load ( Q ) are strictly connected random
variables, because one can find that there is a linear stochastic relation such as: the value of the
corelation coefficient r T] =1. In conclusion, with rational usage the increase in load causes
increased wear and vice versa - decreased load casues less wear of the tribological systems in the
same time range. Moreover, due to the stochastic properties of the load of the tribologic systems
also the process of wearing of these systems is a random variable. This means that the appearance
of the boundary usage and damage connected to it is a random event. This is the main cause these
events cannot be correctly predicted but only estimated in terms of probability of their occurrence.
Diversified (random) loads cause irregular usage of specific tribological systems along with
random speed of diesel engine wear. This makes it important to identify the features of wear of the
tribological systems in practical operation.
Given that two-stroke, slow-running, high-power engine of the main ship's drive should be adapted
to work in an unlimited water region, one needs to analyze the factors influencing its usage
process. These factors are:
1. Engine working conditions - randomly varied, such as:
- random sea weather conditions (waves, wind, heels, water and air temperatures) subject to
constant change in time and intensity,

44
- random load caused by: port maneuvers, river and channel navigation, hull fouling, changing the
course in relation to wave and wind direction, weight of the load and changing conditions of
usage.
2. Qualifications of the engine's users - picked at random:
- the crew is certified and IMO approved (convention STCW 95) but this doesn't guarantee high
qualification (this is dependent on the level of training and examination in IMO approved centers).
These centers often do not meet expectations escpecially in terms of maintance prophylaxis or
engine damage. The crew is not working properly even if the shipowner takes care of the ship by
providing spare parts and not abuses the engines.
- the crew comes from different countries and has different technical culture,
- the crew's physical condition is various and dependent on their health. It is also affected by long-
term voyages, the intensity of their work, the ship's technical state, physical exhaustion, family
problems and the "atmosphere" onboard.
3. Legislation, with specific instructions to follow. They arise from valid rules, restrictions, air and
water pollution limits. The ship is used on international waters or territorial waters so it has to
meet all legal expectations of the waters. The main legislative bodies are:
- International Maritime Organization (IMO) as main international organization, the United
Nations branch, which states the worldwide rules.
- countries' administration - in their legal acts they accept IMO rules, and very often rise the
standards in relation to their territorial waters.
4. Binding law and its regulations such as:
- main legal regulations - IMO, conventions SOLAS 74, MARPOL 73/78,
- local regulations - i.e. HELCOM (Baltic Sea and UE organization) - Helsinki Convention, US
Coast Guard regulations, Canadian Ministry of Transport regulations,
- usage assumptions - the needs and requirements of the ship-owner
- additional regulations extending the safety - classification companies rules.
During the usage of the main ship's engine (OSG) when at sea, the weather conditions cause often
wear of the tribological systems leading to their damage. Unfortunately, the damage appears
despite the use of diagnostic systems (SDG) on the diagnosed systems (SDN), which are the two-
stroke, slow-running, high power engines. Therefore the SDG have to be improved in order to
prevent the damage described in the next part of the paper.

37KHGHVFULSWLRQRIWKHGDPDJHWRWKHWULERORJLFDOV\VWHPVRIWKHWZR-VWURNHVORZ-UXQQLQJ
KLJKSRZHUPDLQHQJLQHV

During the usage of the OSG both the manufacturer and the user gather information about its
damage, which consists of its working duration (until damaged), the cause of the damage and its
effects, the maintenance leading to its repair. etc. This makes it possible to describe the damage to
the tribological systems of the OSG, which occurred despite the use of diagnostic systems (SDG)
on the diagnosed systems (SDN).
The research shows, that (drawing 4.1) one can distinguish three zones (I,II,III) of the engine,
in which the main damage to the tribological systems and other parts and components of the OSG
is the influence of factors dominating in the processes (chemical and physical) that occur during
conversion and transmission of energy to the propeller [27]. The factors are the forces creating
friction and related to it grinding use, stresses, which wear the material leading to fractures and
relatively high temperature favoring the adhesive wear and finally chemically aggressive
environment favoring corrosive wear.
Zone I - in this zone the dominant process is combustion in the combustor at the time which
the main factors are high temperature and aggressive chemical emission. They lead to wear of the
pistons, cylinder sleeves, and rings of the engine.

45
 Zone II - here the dominant process is the transmission of the mechanical energy leading to
damage of pistons, rings, cylinder sleeves, main bearings, crankshaft bearings and crosshead
bearings.
Zone III - here the dominant processes are the vibrations, which lead to wear and damage to
main bearings, crankshaft bearings and crosshead bearings.
However it is important to know that it is basically impossible to determine the soul cause of
the damage caused to the tribologic system of the engine. It is possible that sometimes incorrect
assembly or imprecise quality of the element is the reason to generate more destructive factors
such as, i.e. vibrations, noise, technical dry friction and other events.

Fig. 1. Cross-section of the structural SG and its damaged zones by the dominating factors [14]:
Zone 1: 1- exhaust valve basket, 2 - fuel injection valve, 3 -exhaust valve head, 4 - piston, 5 - cylinder sleeve,
Zone 2: 5 -cylinder sleeve, 6 - piston rod, 7 - piston shank choke, 8 - crosshead bearing, 10 - crankshaft bearing, 11 -
main bearing,
Zone 3: 9 - engine frame

Zone I (Fig.1) is a part of construction structure, in which the main factor of previously described
damage of the SG is thermal energy generated in the process of combustion in the combustor.
In the second picture (Fig. 2) one can notice a fracture of the piston casing's collar where the
piston's head, casing and shank are combined

46
 Fig. 2. The deformation of the piston's head and casing due to combustion gas force activity [17]:
1 - gas forces, 2 - deformation of the piston caused by heat and gas forces, 3 - stress causing deformation, 4 -
relatively high deformation in the vicinity of the telescopic tube and friction of the piston casing's collar

The friction occurred on the outside of the collar which joins the head with the piston casing. It
took place after several thousand working hours (failing to comply with the limits anticipated
between the overhauls TBO (Time Between Overhauls), due to material fatigue. The reason for
that was the use of constructional solutions of the older generation of engines in the newer types,
in which higher thermal load and higher stress occurred. The cause of the load/stress were superior
pressure combustion gases.

Fig. 3. a) piston casing's collar surface influenced by fretting corrosion, b)parts of the piston mutually co-operating
with the cylinder sleeve: 1 - head, 2 - piston casing, 3 - piston shank, 4 - linking bolt of the head, casing and shank
[18]

Fig. 3 describes the wear of the piston casing's collar due to frictional corrosion of the planes
contacting the piston casing's collar with the head and the shank of the piston. The cause of such
damage is the pressure occurring due to the combustion gases of greater pressure than anticipated
in design. The gases cause deformation of the piston, friction of the piston collar and piston part
displacement leading to frictional corrosion. The corrosion causes material loss of ca. 0,01
mm/1000h and lowering of the elastic pressure in the bolt which links the parts.
Fig. 4 shows the damage to the grooves of the piston rings inside the head of the piston. The
reason of the damage is the loss of material caused by frictional corrosion leading to piston groove
size alteration.

47
 Fig. 4. Head of the piston with the piston grooves plated with chromium layer.
The cause of such damage are normal processes of the material loss due to the piston ring work
in the grooves (happening in shifting sea weather conditions) and additional factors causing
chemical-frictional corrosion . Improper use of the engine leads to thermal pressure increase,
piston material overheating, carbon deposit occurrence and high-pressure corrosion caused by
aggressive sulfur compounds created when residual fuel is combusted.
Fig. 5 and 6 show the damage to the surface of the piston head's bottom. Fig. 6 displays three
types of the bottom's surface damage. In the picture are shown:
- corroded material layer on the wide area of the bottom (a),
- local loss of material caused by erosion (b),
- deep loss of material caused by burnup (c).
The causes of such damage are;
- corrosion of the piston bottom's surface (known as "sett block structure" or "elephant skin"),
which is caused by high-temperature corrosion and by sulfur, vanadium and sodium compounds in
the residual fuel (a) in the Fig. 6; the layer of the corrosion products if thin (<0,2 mm/1000working
hours), uniform and coherent then has protective properties; when damaged, in the place of
damage often occur (invisible for the human eye) corrosion pits.
- high-temperature corrosion and erosion, happening on the areas of the piston's bottom, which is
subject to flame dynamics and streams of hot combustion gas, which leads to loss of the bottom's
material (b) in the Fig. 6,
- improperly sprayed and leaking fuel (from the damaged injection valve), which percolates into
the carbon deposit, burns on the surface of the piston's bottom creating deep pinholes (c) in the
Fig. 6.

Fig. 5. The loss of material on the surface of the piston bottom [20]

48
 Fig. 6. Types of damage to the surface of the piston bottom for the types of engines with [21]:
I - two injection valves, II - three injection valves, III - four injection valves, a - corrosion layer, b - local, erosive
material loss, c - deep material loss caused by burnout

The above mentioned phenomena occurrence and the meaning of the consequence of their
occurrence is influenced by:
- increased temperature of the surface of the piston bottom caused by release of carbon buildup on
the surface of the cooling oil holes inside the piston which render the cooling difficult (1 mm thick
buildup causes the increase of temperature by 200 oC),
- insufficient air charging caused by impurities in filter, compressor or turbo-compressor turbine,
air cooler, utilization pot or the silencer,
- correctness of the using process
On the other hand, the Fig. 7 displays the examples of piston bottom damage, which are the result
of occurrence of the causes described above.

Fig. 7. Examples of extreme piston bottom damage [25]:

a) burnout hole caused by bad long-term fuel spraying and piston surface burning, b) damage caused by
improper piston cooling

Fig. 8 shows damage to the piston rings installed in the head of the piston.

49
 Fig. 8. Piston rings are abraded, they crack and break [25]

The damage shown in Fig. 8 points out the loss of piston rings. The reasons of such damage are
construction errors coming from the use of construction solutions from the older types of engines
in newer ones (which have different operating parameters). The damage occurs also due to DTR
incompatible (technical-movement documentation) process of grinding-in [15, 16], using of non-
original spare parts, improper piston-cooling media selection. Moreover, the damage is caused by
improperly prepared cooling media and keeping improper parameters of work of the media [15,
16], invalidly prepared fuel [77], damage to the injection valve, wrong shape of the injection valve
(count, size, direction of the sprayed fuel) [19].
In the Fig. 9 one can observe a damage to the inner surface of the cylinder sleeve.

Fig. 9. Damage to the surface of the inner cylinder sleeve [26]:

1 - mechanical damage by particulate solids, 2 - frictional corrosion, 3 - pinhole corrosion

The damage displayed in the Fig. 9 is the shape change of the sleeve caused by electro-chemical,
frictional, pinhole corrosion. Therefore the causes of the damage are:
- electro-chemical corrosion caused by sulfur acid created from high amount of sulfur in the
residual fuel and high amount of water from the air pressure charging which liquefies on cool
sleeve walls.
- frictional corrosion caused by particulate solids from the outside - from the process of
combustion, friction and grinding of the piston parts, rings and sleeves.
- pinhole corrosion caused by improper lubrication of the sleeve surface (wrong or badly prepared
oil, misplaced lubrication holes, or bad honing of the sleeve), when highly loaded - inadequately
running injection apparatus.
- corrosion of the lower part of the sleeve caused by too-low pressure charging temperature, or too
much water in it (bad work of the dehumidifier in the below-piston area).

50
Fig. 10 shows damage to the connecting rod's head. The damage is caused by alteration of the
head's structure on the contact surface with the bottom bearing liner. The bearing overheats from
excessive wear of sliding surface.

Fig. 10. Connecting-rod [22, 24]: a) damage to the connecting rod's head where the lower bearing liner is adjoined,
b) traces of overheating of the contact surface of crosshead's lower bearing liner, c) destroyed lower bearing liner

In the Fig. 11 one may observe the damage of the sliding coating of the bearing liner of the
main engine bearing, which bears the marks of seizing. The cause of the damage is too long period
between overhauls, along with improper smear oil processing, which leads to particulate solids
layering. Also chemical constitution of the oil affects the coating damage ("used oil"), and external
pressure due to bearing misuse. One may assume the marks of seizing are the result of oil film
stoppage.

Fig. 11. Bearing liner with the marks of extensive wear [23]

Fig.12 displays the damage to the crank-pin, which bears the marks of corrosive wear. The
reason of the damage is inadequate installation of the bearing causing extensive wear, low smear
oil pressure and high temperature leading to bearing liner overheating.

51
 Fig. 12. Visible marks of corrosive wear caused by bearing overheating

Fig. 13 and 14 show damage to the engine crankshaft, which is at the place of joining of its
elements.

Fig. 13. Crankshaft with marked crank-pin [22]

Fig. 14. Crack in the crankshaft at the joining of its elements [24]

Zone II (Fig. 14.), that is the zone, in which the dominating process is mechanical energy
transmission, and tribological processes significantly influence the damage to the engine's
tribological systems.
In the Fig. 15 one can notice the damage to the working surface with piston shank stuffing-box.
The damage is the loss of material of the hardened face of the piston shank, local wear caused by

52
frictional corrosion and the occurrence of deep longitudinal scratches being the consequence of
ridging and microslicing.

Fig. 15. The damage to the coating of the piston shank surface [12, 14, 22]; a) damaged surface, b) important
dimensions: 1 - piston's diameter, 2 - thickness of the hardened coating, 3 - length of the hardened coating

The reasons of the damage are:

- excessive frictional wear caused by inadequate lubrication,
- frictional corrosion caused by water occurrence in the pressure charging and creation of water-oil
emulsion, aggressive chemicals occurrence as effect of combustion (they get to the below-piston
space during cold starting or due to poor condition of the piston-cylinder sleeve system),
- deep longitudinal scratches caused by particulate solids coming from damaged piston, sleeve or
piston shank's stuffing-box elements or the piston shank itself.
Picture 16 displays the damage to the piston shank's stuffing-box. The damage is the wear in the
process of engine use, leading to section ovality occurrence and mechanical damage to the casing
elements and segmental elements of the skimming-tightening rings.
The reasons of the damage are:
- frictional wear, visualizing as uniform loss of material of the tightening rings caused by ring
friction against the tightened surface of the piston's shank,
- DTR incompatible: "dry-running" of the piston's shank, crosshead bearing clearance, stuffing-
box installation, contracting spring rupture, contamination of the tightening rings by the products
of combustion leading to section ovality occurrence.
- installation errors (impurities of the elements, skew, ring assembly from different elements of the
rings, "up-side-down" setup of the ring), corrosive damage, which may cause ring cracking [13],
- aggressive chemicals created during incomplete combustion in changing load, especially during
maneuvers at the time which inadequate media parameters are kept or when the piston-cylinder
system is damaged, which lead to corrosive damage of the casing and stuffing-box rings.

53
Fig. 16. Stuffing-box of the piston's shank [11, 13]: a) set on the shank [22], b) section ovality of the stuffing-box: D1
- allowable diameter, D2 - excessive wear diameter

Zone III (Fig. 17), that is the zone in which the main factor of damage are vibrations.
The damage and their causes in the zone will not be taken under consideration because they
apply to the engine hull damage, such as cracks in the cross-corners of the welded joints of the
engine's hull pillar, cracks of the welded joint of the main plate and ribbing plate, cracks of the
ribbing plate between the technological holes, cracking of the resistance bearing column, etc.
Some damage occurring in the zone influence the proper work of the main bearings. In example,
the cracking of the column of the resistance bearing (being a part of the hull's pillar) (Fig. 17) may
lead to main bearings' damage.

Fig. 17. Engine's hull pillar [23]: a) welded pillar multi-element construct , b)resistance bearing column

An example of resistance bearing crack along the fusion weld of the column and the main plate
is described in the Fig. 18.
The reasons for such damage are:
- constructional factors - joining of two heavy column elements with light ribbed element
constructions,
- installation factors - welding errors such as improperly prepared edges of the welded elements,
bad weld penetration, improper joint face of the fusion weld.
- running factors - changing pressure caused by periodical fuel combustion, increase in vibrations
caused by improper running of the light lateral elements, vibration dampeners, using of the shock
methods of tightening on the resistive bolts fastening the main bearing.

54
 Fig. 18. Pillar of the engine's hull with a crack in the resistance bearing column along the fusion weld of the column
and the main plate [10]

 $QDO\VLV RI WKH GDPDJH FDXVHV RI WKH PDLQ WULERORJLFDO V\VWHPV RI WKH WZR-VWURNH VORZ-
UXQQLQJKLJK-SRZHUPDLQHQJLQHV

The research on the subject of wear and damage including the main tribological systems failure
has shown many difficulties both in the aspect of gathering the data and its analysis. In most cases
of the damage to the systems mentioned in the research, there is a break-down related maintenance
documentation. However, there is no data on the causes of the damage, even the widespread ones -
called "break-downs", rich source of photographic data is available, gathered within the private
collections of mechanics and vast unpublished expert knowledge in collections f engine
manufacturers, ship-owners, classification and insurance companies. The data is accessible only in
cases of spectacular and media-advertised break-downs and is displayed in professional literature
such as; Marine Chamber or Marine Board bulletins in expert approved versions. The paper shows
and analyses given material made available by ship-owners, mechanics and unpublished bulletins
of the engine manufacturers and classification companies. Described facts demonstrate the
complexity of genesis of the damage to the tribological systems and its dependence of the
randomly occurring factors and events, which can be divided in course of accepted criteria such as:
1. Constructional structure of the engine's tribological systems - complex of specific element
shapes and sizes,
2. Geometric dimensions of the engine's tribological systems elements - dependent on the assumed
power, planned period of load or type of work (i.e. dynamics of load change)of the engine, which
are dependent on the ships assignation (main dimensions are the piston diameter, piston stroke,
and related: size of the crankshaft curves and sizes of crank-pins and frame bearing liners and
crank bearings),
Geometric shapes and dimensions of the piston shank and stuffing-box of the below-piston space
are dependent on the amount of energy transmitted to the crankshaft.
3. Material structure - the main criteria here is (strictly related to the function of the tribological
system elements) endurance and durability, i.e. piston's material structure: different head material,
coating material, sieve wall material, piston ring material (covered with aluminum, chromium, and
ceramic material).
The focus however should be on the strict hierarchic relation of the co-working elements in the
tribological systems i.e. more durable (crankshaft) crank-pin (of high size, mass, difficult and
time-consuming disassembly), faster wearing surface of the bearing (smaller size, mass, easy to
disassemble, low material cost).

55
An important feature of material is the resistance to high temperature and aggressive chemical
waste gas environment, especially of the pistons, their rings and cylinder sleeves (which make the
combustor)
4. Physical and chemical phenomena - every identified damage occurring at the time of using of
the engine's tribological systems is explained in accordance to scientific(and sometimes non-
scientific) knowledge, and experience of the users and designers. Therefore, research explaining
basic damage causes with the use of corresponding scientific methods would prove useful.
During practical use fields of pressure, temperature and stress are created. Their values are
determined at the time of work in normal conditions, but maximal values and security coefficients
are known. The values are verified during lab-trials (engine test house trials) in the manufactories
and while engine using. The changes in the temperature, pressure and stress fields are dependent
on the conditions of use, including the choice of proper preparation of used media (mainly fuel and
smear oil). The quality of the process of combustion is related to the calorific value and physical
properties of fuel. The chemical composition of the fuel and smear oil (i.e. sulfur or/and vanadium
content) affects the damage by aggressive corrosive compounds.
5. Human factor - crew actions and other clerical employees of the ship-owner affect:
- the main two-stroke slow-running high-power engine adjustment as diagnostic systems (SDN),
especially their tribological systems and diagnostic systems (SDG) used for diagnostics (the choice
of sensors and data processing apparatus including its precision class),
- the choice of the crew and its rotation (having influence on the quality of work of the crew and -
so called by engine manufacturers - "emotional bonding between the machine and the man"),
- the management of the spare parts and media (choice of manufacturers, and spare part sellers,
media, shipping logistics and storage of the spare parts and media, the regularity in the media
quality checkups)
- the quality of maintenance, both damage-related and planned servicing called preventive (the
choice of repair shipyard, ship's crew, inspecting services, servicing firms etc.)
6. External factors - the conditions specific for the water region, the weather - currents, waving
(the length, steepness and height of the waves), rainfall (air humidity), frequency of occurrence,
direction and force of the winds, water salinity (the amount of salt in the water-mist sucked in with
the air).
In reality, every displayed factor influences the quality of use of the two-stroke slow-running
main engines and their main tribological systems. Often accumulation of several different factors
leads to extensive damage of engine's every main tribological system, thus the engine itself.
The described damage , including extensive ones (break-downs) were the caused by
accumulation of several factors. It is important to have in mind that the faults in the constructional
structure may not be disclosed during designer assumed (DTR) prophylactic maintenance of
tribological systems. Whereas the accumulation of operational faults leads to short-time damage
disclosure in their SDG.
An opposite situation occurs in cases when engine's structural construct is correct, and the errors
are made and long unnoticed and the losses accumulate.
Picture 1 shows zones in which the main damage to tribological systems and their elements is
caused by factors dominating in physical and chemical processes during energy processing and
transmission. The factors are forces, stress, high temperature and aggressive chemical
environment.
- the dominating process in zone I is combustion in the combustor, at the time which the most
important factor is high temperature and aggresive chemical waste gas.
- in zone II the dominant is mechanical energy transmission, at the time which the most important
factors are tribological processes.
- in zone III the dominating processes are vibrations.

56
 After thorough analysis of engine failure, one can isolate such causes of the damage: structural
errors, hidden defects, or structural material damage during heat treatment, improper assembly,
incorrect using (engine abuse, unfit supervision etc.), bad choice and preparation of media/utilities
(fuel and smear oil), improper mechanical processing, insufficient filtering, spinning,
homogenizing, lack of supervision and bad chemical processing of energy factors (fuel
composition, smear oils, cooling water), inadequate values of thermal parameters (too high or too
low temperature of diesel fuel and smear fuel, improper temperature of cooling water cooling the
pistons, sleeves, headings, escape valves etc.), lack of supervision and time exceeding between all
maintenance.

&RPPHQWDU\DQG&RQFOXVLRQ

Although, diagnostic systems (SDG) are implemented to diagnose the engine running; during
the use of two-stroke, slow-running high-power main engines one can distinguish several different
damage of their tribological systems. SDG's are more and more computerized because engine
manufacturers want to improve and implement SDG to engine use. An example of such action is
the use of CoCoS systems (Computer Controlled Surveillance System) in the MAN B&W Diesel
Group manufactured engines.
The use of such systems however, cannot prevent the damage. One of main causes of damage
occurrence is the random attribute of thermal and mechanical load of such engines and in
consequence - wear. Such damage to the tribological systems is unpredictable by its randomness
but can only be estimated in terms of probability of occurrence in the process of use.
The load values of the tribological systems of the two-stroke, slow-running, high power main
engines and their relation to mentioned system wear have been explained in forms of two
hypotheses: the firs t clarifies the relation between the mechanical load Q M (t) and thermal load
Q C (t) of the engines, and the other- the existence of stochastic linear relation in which the value of
corelation coefficient R qz =1.

5HIHUHQFHV

[1] %UXQ 5 6]\ENRELHĪQH VLOQLNL Z\VRNRSUĊĪQH :.Là :DUV]DZD  'DQH R RU\JLQDOH
Science et Technique du Moteur Diesel Industriel et de Transport. Copyright by Societe des
Editions Technip et Institut Francais du Petrole, Paris 1967.(2).
[2] Girtler J.: Issue of changes in technical states of a diesel engine as the result of wear of its
tribological systems. Zeszyty Naukowe Akademii Morskiej w Szczecinie, 31 (103), 2012., s.
77-82.(OMUP).
[3] *LUWOHU-.XV]PLGHU63OHZLĔVNL/:\EUDQH]DJDGQLHQLDHNVSORDWDFMLVWDWNyZPRUVNLFK
ZDVSHNFLHEH]SLHF]HĔVWZDĪHJOXJLMonografia. WSM, Szczecin 2003.
[4] Girtler J.: Necessity for and possibility of application of the theory of semi-Markov
processes to determine reliability of diagnosing systems. Journal of POLISH CIMAC. - Vol.
7, No. 2 (2012), s. 45-54.
[5] Girtler J.: The semi-Markov model of energy state changes of the main marine internal
combustion engine and method for evaluating its operation during ships voyage // Polish
Maritime Research. - Vol. 18, nr 4, 2011, s. 36-42.
[6] 3LRWURZVNL,2NUĊWRZHVLOQLNLVSDOLQRZH:0*GDĔVN  .
[7] 3LRWURZVNL ,:LWNRZVNL.(NVSORDWDFMDRNUĊWRZ\FKVLOQLNyZVSDOLQRZ\FK$0*G\QLD
2002.(16)
[8] :DMDQG -$ :DMDQG -7 7áRNRZH VLOQLNL VSDOLQRZH ĞUHGQLR- i szybkoobrotowe. WNT,
Warszawa 2005.(21).

57
[9] :áRGDUVNL -. 7áRNRZH VLOQLNL VSDOLQRZH 3URFHV\ WULERORJLF]QH :.Là :DUV]DZD
1981.(22)
[10] Cracks in Column, Service Biulletin No RTA-46. Wartsila NSD Corporation Ltd,
Winterthur 1999.(12).
[11] Piston Rod Gland Package. Wartsila Corporation, Waasa 2004.(64).
[12] Reconditioning of Piston Rods RTA-“8 Series” Engines. Service Biulletin No RTA-37.1,
Wartsila NSD Corporation, Winterthur, 2001.(72)
[13] Retrofit for Piston Rod Stuffing Boxes RTA”-2 and 2U Series” Engines. Service Biulletin
No RTA-35.1, Wartsila NSD Corporation, Winterthur, 2001.(73)
[14] Running-in of Cylinder Liner and piston Rings. Service Biulletin No RTA-18.1, Wartsila
NSD Corporation, Winterthur, 1998.(82)
[15] Running-in of Cylinder Liner and piston Rings. Service Biulletin No RTA-18.2, Wartsila
NSD Corporation, Winterthur, 2002.(83)
[16] S 80ME-C Project Guide Electronically Controlled Two-stroke Engines. MAN B&W,
Kopenhaga 2002.(84)
[17] RTA 58/68/76/84 Cracks on Piston SkirtsType Engines. Sulzer Information Biulletin No
DUQ1-008E, Diesel United Sulzer, 1991.(75)
[18] RTA 58/68/76/84 Checks of elastic studs on piston for tightening. Service Information
Biulletin No DUQ1-009E, Diesel United Sulzer, 1991.(76)
[19] RTA Engine, Corrosive Burn-Off Pocket on Cylinder Cover. Service Information Biulletin
No DUQ1-017E, Diesel United Sulzer, 1990.(77)
[20] RTA Engine, Corrosion in exhaust gas passage of exhaust valve cage. Service Information
Biulletin No DUQ1-016E, Diesel United Sulzer, 2003.(81)
[21] Sulzer RTA Engines, RTA 2–2U Series, Piston Crowns Loss of Material on Combustion
Side. Service Biulletin no RTA-38, Wartsila NSD Corporation, Winterthur, 1998.(95)
[22] Sulzer RT-flex 60C, Wartsila Corporation, Helsinki 2004.(96)
[23] Transportation Safety Board of Canada. Marine Investigation Report M00L0034, Main
Engine fires, 2001.(104)
[24] Transportation Safety Board of Canada. Marine Investigation Report M00L0034, Main
Engine Failure, 2000.(105)
[25] Warsash Maritime Centre. The Learning Resource for Maine Engineers, Southampton
Solent University, Southampton 2005.(107).
[26] RTA Type Engines, Characteristics of Cylinder Liner and Piston Ring Wear. Service
Information Biulletin No DUQ1-02E, Diesel United Sulzer, 1991.
[27] 6SUDZR]GDQLH ] UHDOL]DFML HWDSX SURMHNWX EDGDZF]HJR ZáDVQHJR SW ÄAnaliza wáDVQRĞFL
PRGHOX GLDJQRVW\F]QHJR XNáDGyZ NRUERZR-WáRNRZ\FK VLOQLNyZ SU]\VSRVRELRQHJR GR SRWU]HE
UDFMRQDOQHJR VWHURZDQLD SURFHVHP HNVSORDWDFML W\FK VLOQLNyZ Z DVSHNFLH EH]SLHF]HĔVWZD
VWDWNyZ RUD] PRUVNLHJR ĞURGRZLVND QDWXUDOQHJR´ ]UHDOL]RZDQHJR Z UDPDFK SURMektu
EDGDZF]HJR ZáDVQHJR finansowanego przez MNiSW Nr N N509 494638 SW ÄDecyzyjne
VWHURZDQLH SURFHVHP HNVSORDWDFML XNáDGyZ NRUERZR-WáRNRZ\FK VLOQLNyZ QDSĊGX JáyZQHJR
VWDWNyZ PRUVNLFK ] ]DVWRVRZDQLHP GLDJQRVW\NL WHFKQLF]QHM RUD] XZ]JOĊGQLHQLHP
EH]SLHF]HĔVWZDLRFKURQ\ĞURGRZLVND´$XWRU]\VSUDZR]GDQLD*LUWOHU-àRVLHZLF]=Prace
badawcze nr 06/10 /PB :\G]LDá 2FHDQRWHFKQLNL L 2NUĊWRZQLFWZD 3ROLWHFKQLNL *GDĔVNLHM
*GDĔVN

58
 THE LMS VIRTUAL.LAB APPLICATION
IN MACHINES TECHNICAL STATE ANALYSIS
0DUFLQàXNDVLHZLF]7RPDV].DáDF]\ĔVNL0LFKDá/LVV-DNXE.DQLJRZVNL

University of Technology and Life Science

S. Kaliskiego 7, 85-789 Bydgoszcz, Poland
tel.: +48 52 3408262
e-mail:mlukas@utp.edu.pl

Abstract

The proper work of turbocharger has a very important aspect for vehicles combustion engines charging, it directly
influence on: the vitality of the engine, the issues of harmful substances to surroundings and safety, the turbocharger
damage could be dangerous and causes to the vehicle passage in the emergency mode or finally brought to the road
incident. This paper introduces an example of way of machine engines chosen units modelling and simulation analysis
on the introduced example of LMS Virtual.Lab application usage. The LMS Virtual.Lab was used for strain and
deformation analysis of KKK BV39B-0050 turbocharger.

Keywords: turbocharger simulation, degradation processes analysis, vehicle, Computer Aided Design

1. ,QWURGXFWLRQ

Development of three-dimensional modelling together with computers computational

possibilities spread influence on significant growth of engineering software to assist in the
creation, modification, analysis, or optimization of a design. Applying this software to designing
process helps engineers fully replaced and accelerated work in the comparison with traditional
drawing technics. The traditional methods of projecting required from the engineers huge precision
while sketching. Thanks to virtual sketching process we could faster plotting possible changes and
correction in design projects.
Computer – Aided Design (CAD) is an important design software that gives engineers the
direct access to the set of engineering standards, codes of practise or actually instructions for given
industrial branch of production which in the considerable way accelerate designing process with
support of newest legal requirements [1,3].
The first version of CAD systems allows only the replacement of traditional ways of sketching
and in significant way lowering the number of sketch mistakes during designing process.
Introduction of three-dimensional objects (3D) during sketching to systems CAD just made
possible of projecting the machine engines and devices units and their spatial visualization. The
improvement of CAD systems in 3D technics facilitated many actions during designing process
and also allowed to the correct understanding of technical drawings by non-engineer persons (e.g.
management) [1,3,4].
The next step of design development of new objects was CAD systems improvement by
implementation to software new modules that will realize simulations of sketched objects in
different conditions. Thanks to these solutions we could reduce the costs of new prototypes

59
creation, costs of their investigations and also have an influence on new laboratory positions
creation used for experimental audits. The computer programmes allowing to the simulations of
the objects behaviours saves many time additionally and make possible the quick referral to the
production of projected units [1,3,4].
In this paper authors introduced the simulation investigation realised with usage of LMS
Virtual.Lab used for strain and deformation analysis of chosen elements of KKK BV39B-0050
turbocharger, which were realised basis on real object modelling and simulations.

2. 7KH /069LUWXDO/DE V\VWHP

One of the CAD software that helping designing process is the LMS International system called
LMS Virtual.Lab. The LMS Virtual.Lab is an integrated suite of 3D finite elements and multibody
simulation software which simulates and optimizes the performance of mechanical systems for
structural integrity, noise and vibration, system dynamics and durability for real objects. This
software offers a unique approach to simulations – input data that are given from real object
measures are connecting with virtual data of simulated object.
The samples of virtual simulation of technical object behaviour that are realized in the
Virtual.Lab software were introduced on figure 1.

Fig.1. The sample of simulation realised in LMS Virtual.Lab [5]

LMS Virtual.Lab allows tracing all critical steps of designing process. It offer special tools and
technologies that will allow to assess the whole designing process of technical unit with analysis
of all key aspects long before accession to expensive creation and prototype testing process [3,4].
In the aim of the created patternel analysis execution and his individual working unit’s
dependence audits is the necessity of model virtual geometry creation. The virtual patternel should
have some parameters – the degrees of freedom should be defined or the places and values of
external forces should be estimated. Finally we receive the ready virtual prototype which we could
thanks to the suitable possibilities of the programme examine under various regards after creating
patternel e.g. strain and deformation analysis for chosen unit, the virtual analysis of cooperating

60
units, the comparison of work parameters of the same mechanical unit made by different structural
materials [3,4].
The software has a large base of finished elements that improve modelling and designing
process in many range of knowledge. In range of motor industry, the LMS Virtual.Lab offer a
ready-made libraries of the chosen units and aggregates of vehicles which include: suspension,
drive transmission arrangements, combustion engines units, tooth gear transmission which are
ready to implementation by engineer to created patternel and analysis of virtual model on this
stage of mechanical unit creation [3,4].

3. MODELLING OF TURBOCHARGER

The motorization is the one of disciplines that using modern systems of virtual modelling and
simulations. In vehicles exists many parts and components which are put forward in various kinds
of burdens. Thanks to integrated virtual systems that aided design we could very strictly define and
examine the work parameters of chosen technical objects before they will be produced.one of
those elements is turbocharger aggregate, which during his normal exploitation is put forward of
high and changing value of burdens booth thermal and mechanical.
Designing the turbocharger assembly if we know the work conditions we could use the LMS
Virtual.Lab to analysis and simulations of behaviour of turbocharger during exploitation. As an
effect of this simulation we could model simulate and analysis parameters as e.g.:
- deformation schedule,
- strain schedule,
- critical velocity,
- thermal influence on partial elements of investigated combustion engine turbocharger unit.
In this paper authors conducted the computer simulations of chosen burdens for chosen
elements of turbocharger aggregate no KKK BV39B-0050, which were used in diesel combustion
engines by Ford and Volkswagen AG. The simulations were conducted for combustion idle run
(800 rpm) and full speed of diesel engine (4500 rpm) [2].
The design process of turbocharger elements enclose given steps:
a) in the first step during modelling - virtual creation of turbocharger all elements and then
their assembly in the general patternel of turbocharger. The example of virtual design of
compressor wheel were introduced on figure 2;

Fig.2. Virtual model of compressor wheel [own source]

61
 b) in the second stage of work in LMS Virtual.Lab was proper attributing of turbocharger
chosen units the material properties;
c) in the third stage of work designer put on the elements: compressor and turbine wheel,
turbocharger shaft virtual mesh of finite elements (MESH). This mesh will allow defining
burdens and conducting rotor simulations in chosen points of the structure. The virtual
model of turbocharger rotor was introduced on figure 3.

Fig.3. Virtual model of turbocharger rotor with MESH [own source]

4. THE ANALYSIS OF CHOSEN TURBOCHARGER BURDENS

After design of turbocharger virtual patternel in the next stage of investigations was simulated
the pressure influence in turbine and compressor chambers on the turbocharger rotor. During
investigations authors establish pressure for idle run of engine: 0,1 MPa for compressor and 0,115
MPa for turbine chambers. For maximum speed of engine authors establish pressure: 0,210 MPa
for compressor and 0,255 MPa for turbine [2].
The pressure force was introduced on the virtual models as yellow arrows on the compressor
and turbine wheels. The sample of virtual burdens was introduced on figure 4.

a) b)

Fig.4. Virtual model of wheels with pressure force a) compression wheel, b) turbine wheel [own source]

62
 The final stage of the simulation depended on the calculation and the visualization for so the
received burdens of turbocharger assembly wheels and shaft the values of strain and deformation
for given engine speed.

5. RESULTS OF SIMULATIONS

The results of simulations in LMS Virtual.Lab were received in the graphics figure. As a
results system puts on the mesh fields of colours – each colour has an estimated value of studied
quantity. During simulation process of strain schedule their highest value for compressor and
turbine wheels were placed at the basis of blades, were step out stress concentration called out the
notch, meanwhile the farther part of blade come under to the insignificant values of strains. The
similar phenomena stepped out in case of the rotor shaft, where the largest vales of strain step out
in place of the shaft diameter change. The results of virtual simulations of strain schedule were
introduced in table 1.

Table 1. The strain schedule of turbocharger elements [2]

Minimum strain value Maximum strain value

Unit Rotation speed
[N/mm2] [N/mm2]
800 rpm 3 46
Compressor
4500 rpm 5 101
800 rpm 7 106
Turbine
4500 rpm 15 276
800 rpm 4 119
Shaft
4500 rpm 10 263

In case of deformation simulations the highest values step out on the end of compressor and
turbine blades, because the forces coming from the gases pressure in turbine and compressor
chambers worked out on the longest arm of a force. The similar phenomena stepped out in case of
the rotor shaft, where the highest values of deformations were placed on the shaft ending points.
The results of virtual simulations of deformation schedule were introduced in table 2.

Table 2. The deformation schedule of turbocharger elements [2]

Minimum deformation Maximum deformation

Unit Rotation speed
[mm] [mm]
800 rpm 0,002 mm 0,029 mm
Compressor
4500 rpm 0,005 mm 0,074 mm
800 rpm 0,002 mm 0,072 mm
Turbine
4500 rpm 0,005 mm 0,157 mm
800 rpm 0,001 mm 0,013 mm
Shaft
4500 rpm 0,001 mm 0,029 mm

The graphical results of strain schedule simulations of turbocharger aggregate chosen elements
were introduced on figure 5, where as an example of strain schedule were introduced course of
strain along the line between point A and B. AB line is placed on the turbine blade and simulations
results are conducted for maximum rotation speed of engine. The graphical results of strain

63
schedule simulations for compressor blade along the line between point A and B conducted for
engine maximum rotation speed were introduced on figure 6. The graphical results of strain
schedule simulations for rotor shaft along the line between point A and D conducted for engine
maximum rotation speed were introduced on figure 7.

Fig.5. Strain schedule course on the turbine blade along line A-B for maximum rotation speed of engine [own source]

Fig.6. Strain schedule course on the compressor blade along line A-B for maximum rotation speed of engine
[own source]

Fig.7. Strain schedule course on the rotor shaft along line A-D for maximum rotation speed of engine [own source]

64
6. CONCLUSIONS

In this paper were introduced the possibilities of usage the LMS Virtual.Lab design application
in virtual modelling process of strain and deformation schedule on the example of real technical
object – turbocharger aggregate. Basis on the virtual patternel were marked out the dangerous
work states and precisely define the conditions which should be fulfilled for real object so that
turbocharger worked out unfailingly according his destination. The investigations were conducted
to mark out the analysis of changes course for chosen parts of turbocharger on the results of strain
and deformations influence.
Marked out during simulation values of turbine and compressor blades strain raised up together
with the distance decreasing to the centre of the rotor turn. However the largest values of strains
achieved at the basis of the blade. During designing process of these elements we should pay a
special attention into appointed critical places and we should to conduct suitable constructional
materials selection, geometrical dimensions and pay attention on the ways of rotor processing.
The highest values of blades deformations marked out in the terminal point of blade and
lowering in direction of rotor rotation centre. The critical blades deflection of rotor wheels is very
dangerous because the turbocharger aggregate are designed and produced with very small value of
linear tolerance. Each deformation could case to whirling units point of the contact with casing and
finally damage and immobilize the turbocharger assembly.
The highest values of rotor shaft strain were marked out in points B and C (fig. 7) – places
where the shaft diameter value changes. Such strain schedule in these points forces on the
designers to put special attention during designing process of these elements. Designers should
choose suitable constructional materials, proper shape of rotor shaft and predict suitable shaft
processing to this unit damage avoidance.
The highest values of rotor shaft deformation were marked out in shaft pins where the rotors
wheel is placed. Such phenomena could be predicted early on the shaft projecting stage, because
these places are deformation susceptible – the places with the smallest geometrical value.
Similarly as in rotor wheels for shaft, to high value of deformation could damage and immobilize
the turbocharger assembly. The main reason of rotor shaft deformation was caused his bending
and torsion. It was caused by unequal schedule of forces on the rotors wheel circuit. The
inequalities of forces reigning in the turbine and compressor chambers are called out by reducing
chambers diameter what causes gases pressure increasing and also from the angular frequency of
the admission expiratory gases.
The most important parameter during exploitation on turbocharger assembly in vehicles is the
value of combustion engine rotation speed. During simulations estimated strain and deformations
values depend on engine rotation speed. During exploitation of turbocharger assembly, the violent
change of combustion engine rotation speed influence on estimated values of strain and
deformations growth twice what was introduced in this studies.

5HIHUHQFHV

[1] &HPSHO&]ĩyáWRZVNL%,QĪ\QLHULDGLDJQRVW\NL PDV]\Q, Polskie Towarzystwo Diagnostyki

Technicznej, Warszawa, Bydgoszcz, Radom 2004.
[2] Kanigowski J.: $QDOL]D REFLąĪHĔ G\QDPLF]Q\FK ]HVSRáX WXUERVSUĊĪDUNRZHJR, Praca
LQĪ\QLHUVND873%\GJRV]F]
[3] àXNDVLHZLF] 0  Vibration measure as information on machine technical condition,
Studies & Proceedings of Polish Association for Knowledge Management 35, ISSN 1732-
324X.

65
[4] àXNDVLHZLF] 0 .DáDF]\ĔVNL 7 .DVSURZLF] 7 Rola innowacyjnych programów
NRPSXWHURZ\FK ZVSRPDJDMąF\FK SURMHNWRZDQLH Z RGQLHVLHQLX GR PLQLPDOL]DFML VNXWNyZ
RGG]LDá\ZDQLDQDĞUodowisko, s. 377 - 383, ISBN: 978-83-923256-9-7, BSW Bydgoszcz 2012.

:HEVLWH:
[5] http://www.ects.pl/?page=PStructure&id=19&cid=120 (20.12.2012).

66
 THE INFLUENCE OF CONDENSING PRESSURES OF THE
ETHYLENE CARRIER CASCADE CYCLE ON ITS REFRIGERATION
CAPACITY

Dariusz Nanowski

Gdynia Maritime Univeristy

ul. Morska 81-87, 81-226 Gdynia, Poland
tel.: +48 586901449
e-mail: d.nanowski@wm.am.gdynia.pl

Abstract

Ethylene, Propane and others liquefied gases as a cargo are transported at sea by temperature below 0°C. It
requires sufficient efficiency of reliquefaction plant. Short description of that gas plant is shown including
cascade system and its cooling processes are described. Based on Mollier diagrams the refrigeration capacity
investigation of gas plant is done. This analysis is focused on condensing pressure which takes place in the
condenser. Theoretical cycles are taken into account and real condensers of cargo compressor and screw
compressor of refrigerant cycle are considered. Thermostatic expansion valves are employed in cascade cycles
and their operation parameters are taken into account as well. Conclusions include some guidelines on
condensing pressure and related operation parameters to maintain the gas plant with high efficiency of
refrigeration capacity.

Keywords: refrigeration cycle, reliquefaction plant, condensing pressure, Ethylene gas carrier

1. Introduction. The gas plan of Ethylene carrier

Boiling point of Ethylene at atmospheric pressure is 103.8°C means that cooling

processes are not belong to cryogenics e.g. below 111.1 K when at the same pressure Methane
has boiling point. However temperatures below 100°C achieved in Ethylene carriers cargo
tanks require using cascade systems, because cooling down the cargo between 60°C till
100°C with multi-stages cycles is very difficult or even impossible [4]. Of course, Ethylene
is not only one grade of cargo for these ships [4].
General principles of cargo gas plant operation are being described by using as an
example simplified layout of reliquefaction and cascade systems, whereas processes
parameters were shown in previous papers [5,6].
Reliquefaction system with reciprocating cargo compressor is shown on Fig. 1. The
characteristic points of system and processes are marked by figures 1 to 10. When Ethylene
vapour is sucked from the cargo tank by first stage of cargo compressor (point 1), vapour has
already left suction drum, where liquid phase could be separated from compressor suction line
(to avoid liquid hammering). After discharging by interstage pressure (point 2), before
compressing in the compressor second stage, vapour is cooled down by mixing with saturated
vapour after its vaporization (from point 9 to 10) in the cargo economizer. By this way

67
vapour temperature in compressor second stage suction is decreased from t 2 to t 3 = 50°C.
Second stage compression increases vapour temperature up to t 4 = 130°C, by pressure p 4 =
18.5 bar and Ethylene is directed to LPG condenser.
There is no any Ethylene condensing process in this heat exchanger, but only cooling
down of vapour from t 4 to temperature t 5 = 30°C by using sea water as a cooling medium.
Condensing process 5-6 takes place in Ethylene condenser.

Fig. 1. Layout of reliquefaction plant [6]

In this heat exchanger are connected two systems: reliquefaction (Fig. 1) and refrigerant (Fig.
2), because Ethylene condensing process is carried out by vaporization of refrigerant R404A
or Propylene in Ethylene condenser, common heat exchanger of both systems: reliquefaction

68
and refrigerant, as a one cascade system. By high gauge pressure p 4 = 18.5 bar is possible to
condense Ethylene and cooling down to temperature t 6 = 30°C by means of R404 of which
evaporating temperature is 40°C. In next step, flowing through the cargo economizer coil
Ethylene condensate is subcooled in process 6-7 to temperature t 7 = 63°C with the use of
Ethylene, which evaporates (process 9-10) by interstage pressure p 2 = 5 bar and temperature
t 9 = 72°C in the cargo economizer.

Fig. 2. Layout of refrigerant system[6]

Isenthalpic expansion of refrigerant 6-7 (Fig.2) is carried out by means of three TEVs,
which operate as a controllers for supplying R404A to Ethylene condenser. More details
about this cycle were described in previous papers [6].

69
 2. The cargo compressor theoretical Mollier pressure-enthalpy diagram

Condensing pressure of counter clockwise cycle is one of the most important parameters
in order to maintain high efficiency ratio of refrigeration capacity. But sometime this ratio is
higher when condensing pressure is higher, sometimes when it is as low as possible.

R290 Ref :W.C.Rey nolds: Thermody namic Properties in SI

20,00
DTU, Department of Energy Engineering

50
VLQ>N- NJ. @YLQ>PANJ@7LQ>ú&@
M.J. Skovrup & H.J.H Knudsen. 14-09-19
50
3' 2'

40
40

0 0060
30

3 0,0 v= 0
, 30
2
08
10,00
10 15 20
0,0
v=

0,0 0,0 v=
v= v=
20

9,00
20
,030

,70
,60
8,00
Pressure [Bar]

s = 2,40
v= 0

s= 2
s = 2,5

s= 2
7,00
10

,040 10
v= 0
6,00

,060
0

5,00
v= 0 0

,080
4,00 v= 0
-10

,10
v= 0 -10

3,00
4 1
5
4' v= 0,1

x = 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,80 0,90 0 20 40 60

s = 1,00 1,20 1,40 1,60 1,80 2,00 2,20
150 200 250 300 350 400 450 500 550 600 650 700
Enthalpy [kJ/kg]

Fig. 3. Mollier diagram of Propane [2]

Above on Fig. 3 is shown the difference of Propane specific refrigeration capacity caused
by different sea water temperature as a coolant of condensers. The reliquefaction plant shown
on Fig. 1 operates as two stage without intercooling. Cycle 1-2-3-4-1 operates with sea water
temperature 28°C and specific refrigeration capacity h 1 -h 4 , whereas 1-2’-3’-4’-1 with
temperature 45°C and reduced specific refrigeration capacity h 1 -h 4’ . According to data of
Burckchardt cargo compressor 2K160-2H volume capacity, it can be calculated that by 3 bars
absolute pressure in suction line total refrigeration capacity is:

Q 1 = m 1* (h 1 -h 4 )= 554[kW]. (1)

This equation is for a cycle 1-2-3-4-1 where m 1 is mass flow of cargo through
compressing process 1-2 and:

Q 2 = m 2* (h 1 -h 4’ )= 432[kW]. (2)

for a cycle 1-2’-3’-4’-1 with Propane mass flow m 2 in process 1-2’. It means because of
higher temperature of sea water refrigeration capacity of reliquefaction unit is less approx.
22%. Mass flow m 1 is higher than m 2 because this calculation takes into account less volume
capacity of the compressor caused by higher discharge pressure [1,3]. Of course the same loss
of refrigeration capacity is if there is higher pressure of condensing because the condenser is
not properly clean.

70
 3. Refrigerant cycle

On Fig.4 are shown two cycles of refrigerant system (Fig.2) with different condensing
pressure. First one operates with 8 bar abs and second cycle 15 bars abs condensing pressure.
Analogous to cycles described on Fig.3 it seems that cycle 1-A-B-C-D-E-1 has higher
refrigeration capacity than cycle 1-2-3-4-5-6-1 because of higher specific refrigeration
capacity as a difference between enthalpies:

h 1 -h E > h 1 -h 6 ) [kJ/kg]. (3)

R1270 Ref :W.C.Rey nolds: Thermody namic properties in SI

20,00 0,030
DTU, Department of Energy Engineering

40
VLQ>N- NJ. @YLQ>PANJ@7LQ>ú&@
40
M.J. Skovrup & H.J.H Knudsen. 14-09-20
5 4
30

30
060
0,0
20

v= 080 20

,40
,0 0 10
10,00
v= 0,0 C

s= 2
9,00 D v= 15 20
10

0,0 0,0
8,00 v= v= 10
,030
7,00
v= 0 3 2
0

6,00 0
,040

,50

2,90
,60
v= 0

,70

,80
s= 2
Pressure [Bar]

s= 2

s= 2

s= 2

s=
5,00
-10

-10
4,00 v= 0
,060 A
B
-20

,080
3,00
v= 0 -20

,10
v= 0
-30

-30
5
2,00 v= 0,1
6 v= 0,2
0
E
-40

-40

0
1
v= 0,3

1,00
x = 0,10 0,20 0,30 0,40 0,50 0,60 0,70 0,80 0,90 -40 -20 0 20 40 60 80 100
s = 0,80 1,00 1,20 1,40 1,60 1,80 2,00 2,20 2,40
50 100 150 200 250 300 350 400 450 500 550 600 650 700 750
Enthalpy [kJ/kg]
Fig. 4. Mollier diagram of refrigeration cycle [2]

In this plant with old design of TEVs it is not true. According to manufacturer manual
condensing pressure should be kept approx. 14.0 bar abs in order to achieve required mass
flow of refrigerant through thermostatic expansion valves (TEV). Tab.1 shows refrigeration
capacity of thermostatic valves TES 12 (R404A) related to drop of refrigerant pressure
during a flow through valve.

Tab. 1. TEV refrigeration capacity kW related with refrigerant pressure drop

Pressure drop through expansion

Thermostatic expansion valve valve [bar]
type
6 8 10 12 14 16

Danfoss TES 12-13.4

21,6 23,8 25,3 26,1 26,3 26,5
Catalouge year 2009
Danfoss TES 12-20
29,6 29,3 28,3 26,9 25,2 23,5
Catalouge year 2014

71
 It can be noted that difference between inlet and outlet pressure of refrigerant (before
expansion valve and after) is important for old TEV`s constructions (year 2009). Increase of
refrigerant capacity may change from 21,6 kW to 26,5 kW – approx. 20% depends on
refrigerant pressure difference. It means that condensing pressure is an essential parameter of
refrigerant plant in order to achieve high refrigeration efficiency.

4. Conclusions

Refrigeration cycles used at sea are not only as a provision plant, air conditioning systems
or reefers. This refrigeration system is used on LPG and Ethylene carriers, where very often
total refrigeration capacity of reliquefaction plants are higher than 1000 kW or 2000 kW and
20% drop od refrigeration efficiency means delays in cargo transportation and lost money.
Very easy is to notice when refrigeration system operates with improper, to high
condensing pressure, but it is important to know that sometimes too low pressure is
disadvantage as well. As was shown in Tab.1 especially old TEV`s constructions require
accurate condensing pressure to provide high efficiency of plants. High difference between
TEV’s inlet and outlet pressure provides required refrigerant mass flow and oil back to the
compressor.
New solutions employed in these systems:
- new design of thermostatic expansion valves
- electrically operated expansion valves
- valves with PLC controlers
These devices operate properly with wide range of condensing pressure in refrigeration
systems.

5. References

[1] Bohdal, T., Charun, H., Czapp, M., 8U]ąG]HQLD FKáRGQLF]H VSUĊĪDUNRZH SDURZH
Wydawnictwo Naukowo-Techniczne, Warsaw 2003.
[2] Coolpack 1.49 – IPU& Department of Mechanical Engineering Technical University of
Denmark.
[3] Królicki, Z., 7HUPRG\QDPLF]QH SRGVWDZ\ REQLĪDQLD WHPSHUDWXU\, Oficyna Wydawnicza
3ROLWHFKQLNL:URFáDZVNLHM, :URFáDZ.
[4] McGuire and White, Liquefied gas principles on ships and in terminals, Witherby &Co,
London 2000.
[5] Nanowski, D., Wybrane parametry procesów termodynamicznych rzeczywistego obiegu
kaskadowego wykorzystywanego do morskiego transport etylenu, 7HFKQLND &KáRGQLF]D L
.OLPDW\]DF\MQD, ʋ4(194), 2012, pp.182-185.
[6]Nanowski, D., Gas plant of Ethylene gas carrier and two stages compression optimization
of Ethylene as a cargo based on thermodynamic analysis, Journal of Polish CIMAC,
2012, vol 7, pp183-190

72
 VALUATION OF ROTODYNAMIC PUMPS OPERATION, BY MEANS OF,
DIMENSIONAL ANALYSIS

Jan Roslanowski Ph.D.Ch. (Eng.)

Gdynia Maritime Academy

Faculty of Marine Engineering
81-87 Morska str.
81-225 Gdynia Poland
e-mail: rosa@am.gdynia.pl

Abstract

The following article introduces determination method of rotodynamic pumps operation. The pumps have been
installed in installations, basing on their work parameters, by means of dimensional analysis. Operation of
rotodynamic pumps, as energetic devices, according to J. Girtler, has been treated as a new physical quantity of
dimension, Joul multiplied by second [J·s]. It expresses transformation of energy supplied to liquid, being forced
through by the pump. This quantity can be determined, on the basis of algebraic diagram of dimensional analysis,
constructed by S. Drobot. This diagram allowed us to control correctness of inference rules, in relation to
mathematical aspect, used in numerical functions of rotodynamic pumps operation, fitted in instalations.

Key words: operation of rotodynamic pump, algebraic diagram of dimensional analysis, physical quantity of Joul
multiplied by second dimension, dimensional analysis

1. Introduction

Operation of rotodynamic pump consists in transportation of mechanical energy from any

external source into the liquid which passes through the pump. This energy is used up on gain in
angular momentary or liquid circulation and also overcoming of hydraulic resistance in pressure
conduit.
It causes an increase of all energy components, such as: position pressure and kinetic energy
in flowing liquid. An increase of liquid energy in rotodynamic pump is determined by
constructional parameters of rotor and its rotational speed. On the other hand, pressure increase
depends on delivery of rotodynamic pump.
According to the above, it makes sense to investigate the operation at the same time, by
energy supplied to it, during its transfer to the liquid being just forced through. Rotodynamic
pumps being passive working machines, belong to energetic devices. Operation valuation of
energetic devices is proposed by J. Girtler in works [3,4,6]. He equates it to dimensional physical
quantity of measure unit called Joul second. Such operation can be interpreted as transfer of
mechanical energy to the liquid that is forced through by the pump, in a definite time.
Operation of rotodynamic pump, in which transfer of energy takes place, can be the carrier

73
of information about its technical condition.
Parameters of rotodynamic operation depend on characteristic of installation in which the
pump takes place.
In building of rotodynamic pumps, one can observe a tendency to increase rotational speed
of the rotor and its higher speed, in order to diminish the size of the pump and of the electric motor
and thus lower the cost of investment and exploitation [5,8].
It is the reason why in pumps of newer generation , one could observe destructive force of
cavitation, worse than in pumps of older generation. Durability and reliability of pump operation,
to a great extent, depend on cavitation character.

2. Rotodynamic pump operation in dimensional space

Rotodynamic pump operation makes the liquid molecules rotate together with the rotor, and
at the same time move along the blades from the centre towards the circuit. It decreases the
pressure at the rotor inlet, creating a phenomenon of liquid sucking into the pump’s interior.
Molecule movement is generated by the influence of circumferential forces of friction and radial
centrifugal forces.
Water flow through the pump takes place in the field of gravity forces. Field of current
changes according to energy position. During liquid flow through the rotor, its kinetic energy
increases, which is caused by an increase of floating speed of liquid molecules in tis circular
movement around the rotor axis. Further change of energy takes place under influence of decrease
of relative speed of the liquid, owing to its flow through channels between stator blades of
increasing transverse intersection or in diffuser. It means that revolving rotor of the pump
generates energy increase in flowing liquid. This energy is equal to work done by the torque of
pump’s propulsion engine. The amount of energy transferred to liquid by the rotor, can be
calculated, on the basis of second principle of dynamics, for rotary motion. According to this
principle, the torque transferred by rotating rotor, causes the change of its angular momentum of
liquid. The work done by the rotor of the pump causes an increase of pressure energy, overcoming
of altitude difference, surmounting of pressure difference and friction resistance between inlet
intersection and the outlet one of the pump [2,8]. Thus it causes energy increase of the liquid
flowing through the rotor. So, the operation of rotodynamic pump can be defined by the following
formula:
t
D ³ U ˜ Q ˜ 'E W
0
˜ W ˜ dW (1)

where:
ȡ– density of pumped liquid in [ kgm-3],
Q- output of the pump (volumetric intensity of the flow) in [ m3 s -1 ],
ǻE(IJ) – total increase of proper liquid energy in time function, during the flow through the pump,
according to mass unit in [ m2s-2 ],
IJ- time of liquid flow through the pump in [ s ],
D– operation of rotodynamic pump in [ Js].

Operation of the pump is characterized by dimensional quantities which possess number and
dimension. They create dimensional space, in which there are product and involution of
dimensional numbers of real exponent. Results of these operations, together with real exponent
belong to dimensional space, which means that its elements are dimensional and non-dimensional
quantities, i.e. numbers. Therefore pump operation can be expressed as dimensional function of
many variables. Such operation of the pump presented by the above formula (2) is in dimensional
space ʌ whose arguments are also elements of this space. It is not a numerical function and

74
therefore it must fulfil additional conditions of invariance and dimensional homogeneity. These
conditions do not limit numerical functions, i.e.: such, whose both, arguments and the function
value are non-dimensional quantities [1,3,6,7,9].
In this way we can obtain general physical equation of liquid flow through the pump rotor of
the following form:

D m ˜ g ˜ H ˜W ) U , n, d , Q,Q ,W (2)
where:
m – liquid mass flowing through the pump in [kg],
g – acceleration of gravity in [ ms -2 ],
H – height of pump hoisting in [ m ],
d- nominal diameter of pump rotor in [ m ],
n - rotational speed of the rotor in [ s -1 ],
Ȟ - coefficient of kinematic viscosity of the liquid in [ m2s-1],
- remaining denotations as in formula (1).

It is worth mentioning that condition of invariance results from a description possibility of

physical dimensional quantities in different unit arrangements. In an accepted system of basic units
of measure SI, matrix of involution exponents of function arguments (2) is of the third order. It
means that in case of dimensional function (2) three quantities out of six arguments are
dimensionally dependent on these three.
Therefore, from dimensional function (2) one can choose arguments dimensionally
independent and separate them from the remaining ones, in other words, one can accept
dimensional base of this function in twenty different ways (combination 63 =20 ).
Not all accidentally accepted dimensional bases will be proper. This fact results from
dimensional independence of units system: kg, m, s. Out of twenty random ways of choosing
dimensional function bases of rotodynamic pump operation, only 8 of them are mathematically
correct, which are given in table 1.
According to works [1,6] one can accept that the best base of dimensional function (2) will
be the following dimensionally independent quantities: ȡ d, n.
Applying Buckingham theorem [1,6] to function (1), after accepting the above dimensional
base, we obtain the following equation:

D m ˜ g ˜ H ˜W f1 IQ ,IQ ,IW ˜ U ˜ n ˜ d 5 (3)

where:
f1 - numerical function,
Q
IQ - efficiency discriminant (dimensionless argument of numerical function f 1 ),
n˜d3
Q
IQ - discriminant of viscosity resistance (interior friction),
n˜d2
IW n ˜W - discriminant of pump operation time,
-the remaining denotations as above.

Formula (3) can be rearranged to the following form:

௠ή௚ήுήఛ ொ ఔ
ߔ஽ ൌ ఘή௡ήௗఱ
ൌ ݂ଵ ቀ௡ήௗయ ǡ ௡ήௗమ ǡ ݊ ή ߬ቁ (4)

75
where:
ߔ஽ - dimensionless discriminant of rotodynamic pump operation,
-remaining denotation as above.

Table1. Choice possibilities of arguments dimensionally independent , so called, dimensional bases in function of
rotodynamic pump operation D ) U , n, d , Q,Q ,W .

o number Form of dimensional function Dimensional base Remarks

D f1 IQ , IQ , IW ˜ U ˜ n ˜ d 5
1
Q Q
IQ ,˜ ˜ ˜ ˜ IQ ,˜ ˜ ˜ ˜ IW n ˜W ʌ͕Ŷ͕Ě
n˜d3 n˜d2
Q5
D f 2 Id , IQ , IW ˜ U ˜ 3
n2
2 ʌ͕Ŷ͕Y
n Q
Id d ˜ 3 ,˜ ˜ ˜ ˜ IQ ,˜ ˜ ˜ ˜ IW n ˜W
Q 3
n ˜ Q2
Q5
D f 3 Id , IQ , IW ˜ U ˜
3
n3 ʌ͕Ŷ͕ʆ
n n
Id d˜ ,˜ ˜ ˜ ˜ IQ Q˜ ,˜ ˜ ˜ ˜IW n ˜W
Q Q3
D f 4 I n , IQ , IW ˜ U v ˜ Q ˜ d 5
4 ʌ͕Y͕Ě
n˜d 3 Q ˜d W ˜Q
In ,˜ ˜ ˜ ˜ IQ ,˜ ˜ ˜ ˜ IW
Q Q d3
Q9
D f 5 In ,Id ,IW ˜ U ˜
Q 11
5 ʌ͕Y͕ʆ
n ˜Q2 d ˜Q W ˜Q 3
In ,˜ ˜ ˜ ˜ Id ,˜ ˜ ˜ ˜ IW
Q3 Q Q2
D f 6 In ,Id ,IQ ˜ U ˜ 3 Q 5 ˜W 2
6 d W ʌ͕Y͕ʏ
In n ˜W ,˜ ˜ ˜ ˜ Id ,˜ ˜ ˜ ˜ IQ Q ˜3
Q ˜W
2
3 Q
D f 7 I n , I d , IQ ˜ U ˜ Q ˜ W 5 3

7
d Q
In n ˜W ,˜ ˜ ˜ ˜ Id ,˜ ˜ ˜ ˜ IQ ʌ͕ʆ͕ʏ
Q ˜W Q 3 ˜W
U ˜d5
D f 8 In , IQ , IQ ˜
W
8 ʌ͕ʏ͕Ě
Q ˜W Q ˜W
In n ˜ W ,˜ ˜ ˜ ˜ IQ ,˜ ˜ ˜ ˜ IQ
d3 d2

Discriminant of rotodynamic pump operation (4) can be expressed as product of

dimensionless discriminants of hoisting pump height and the volume of flowing liquid through its

76
rotor, in the following form:

ொ ఔ
ߔ஽ ൌ ߔ௏ ή ߔு ή ߔఛ ൌ ݂ଵ ቀ௡ήௗయ ǡ ௡ήௗమ ǡ ݊ ή ߬ቁ (5)
where:
௠ ௠ ௏
ߔ௏ ൌ ఘήௗయ ൌ ೘ ൌ ௗయ - dimensionless discriminant of liquid volume flowing through the rotor
ήௗయ
ೇ
of rotodynamic pump,
௚ήு
ߔு ൌ ௡మ ήௗమ  -discriminant of pump hoisting height ,
- the remaining denotation as above.

It is worth noticing that in discriminant of pump hoisting height ߔு in the formula (5) liquid
density does not appear, therefore, one can draw conclusion that the height of pump hoisting does
not depend on liquid density.
Also, from the above dependence it appears that the height of pump hoisting of rotor
diameter d , changes proportionally to rotational velocity square. Besides, the inverse of height
discriminant of pump hoisting, constitutes one of the forms of Froud number, and the inverse of
discriminant is defined by the following formula:
1 d 2 ˜u2 u2
Fr | (6)
)H g˜H g ˜ H th
where:-
u - FLUFXPIHUHQWLDOYHORFLW\RISXPSLQ>UDGÂV– 1 ],
H th - theoretical height of pump hoist in,
- remaining denotations as above.

The formula (6) shows that Froud’s number determines the condition of forces similarity
necessary to hoist the liquid to height H th .
The resistance of liquid flow caused by its viscosity is determined by Reynold’s number,
expressed by the following formula:
1 d2 ˜n d ˜n
Re | (7)
)Q Q Q
where:-
- denotations as above.

The value of Reynold’s number, in relation to a characteristic dimension d and to

circumferential speed u prevailing on the circumference of the rotor with a diameter d , determines
character of liquid flow through the pump. It should be mentioned here, it was experimentally
proved that by heavy flows of liquid through the pump’s rotor, Reynold’s numbers are
characterized by great values, which means that the influence of liquid viscosity on flow character
is minimal [7,9].
Therefore in equation (3) we can ignore the discriminant of viscosity resistance IQ , and thus
dimensional function of pump operation is expressed by means of numerical function of two
variables presented in the following form:
D m ˜ g ˜ H ˜W f1 IQ ,IW ˜ U ˜ n ˜ d 5 (8)
where:
- denotation as in formula (3).

Operation of the pump can also be expressed , by means of power absorbed by it, in the form

77
of general physical equation of the following dimensional function:

‫ ܦ‬ൌ ܰ ή ߬ ଶ ൌ Ȳሺߩǡ ݀ǡ ݊ǡ ܳǡ ߥǡ ߬ሻ (9)

where:-
N –power absorber by pump in [ m 2 kg s -3 ],
- denotation as in formulas (1) and (2).

Dimensional function Ȳ, determined by formula (9) can be in similar way, as in formula (1)
and adopting the same dimensional base, transformed into numerical function of the following
form:
D N ˜ W 2 f 2 [ Q , [Q , [W ˜ U ˜ n ˜ d 5 (10)
where:-
Q Q
[Q , [Q , [W n ˜ W - independent dimensionless arguments of numerical function
n˜d3 n˜d2
f 2 , so called, similarity invariants,
- denotations of the remaining quantities, as in previous
formulas.

From the above formula it appears that if in dimensional function, the same dimensional
quantities independent of the remaining ones were accepted, then adequate variable arguments of
their numerical functions will be equal, i.e. IQ = [Q . Hence their name of similarity invariants.
In the simplest case, the forms of dimensional numerical functions, under discussion, can be
approximated by linear functions. Such approximations have been given in table 2.

Table 2. Linear estimations of numerical functions of rotodynamic pump in three – dimensional space of
dimensional base ȡ, d, n

Dimensional
On. Linear estimations of dimensional function
base
1. D A0 ˜ U ˜ Q ˜ d 2  B0 ˜ U ˜Q ˜ d 3  C0 ˜ U ˜ n 2 ˜ d 5 ˜W  D0 ˜ U ˜ n ˜ d 5 ʌ͕Ě͕Ŷ

2. D A1 ˜ U ˜ Q ˜ d 2  C1 ˜ U ˜ n 2 ˜ d 5 ˜ W  D1 ˜ U ˜ n ˜ d 5 ʌ͕Ě͕Ŷ

Dimensionless numbers, so called, similarity invariants

Q Q
MQ ; MQ ; IW n ˜W
d 3n d 2n
A i ; B i ; C i ; D i – real numbers determined on the basis of measurements (i = 0,1).

They should be treated as a rough definition of dependences between quantities which

describe the influence of rotodynamic pump on the liquid flowing through the pump.

3. Diagnostic properties of dimensional functions

From dimensional analysis it appears: delivery of a pump Q, operation of rotodynamic pump
D and rotational speed n , they constitute basic hydraulic parameters of the pump.
They determine each state of pump operation and for this reason, they can be used as
diagnostic symptoms of technical condition.

78
 Besides, its technical state can be determined by similarity invariants, called as
discriminants: delivery of a pump MQ , operation of rotodynamic pump ߔ஽ and time of
rotodynamic pump operation IW . These discriminants have constant values, determined by the
values of geometric parameters, kinematic and dynamic ones of the pump.
It means, that on the basis of these values deviations of the pump under investigation, and
comparing it with a model one, completely efficient, it will be possible to make a diagnosis about
its technical condition.
Taking into consideration the expression (2) and ignoring viscosity of the liquid flowing
through rotodynamic pump, we can determine the hydraulic performance of rotodynamic pump,
by means of the following formula:-

f1 M Q ; MQ ; MW
Kh (11)
f 0 MQ
where:
- denotations as in previous formulas.

From the formula (11) it follows directly that hydraulic efficiency of the pump depends on
the discriminant of working capacity, Reynold’s number and time of rotodynamic pump operation.
On the basis of the size of relative energetic losses, taking place in the pump, one can
evaluate its general technical condition. Therefore operation of rotodynamic pump is a diagnostic
symptom [4]. Total energetic losses in rotodynamic pump can be determined on the basis of the
definition in the following way:-

'H He f1 M Q ;MQ ;MW § 1 ·

1 1 K h 1 K  f 3 ¨¨ M Q ; ;MW ¸¸ (12)
H th H th f 0 MQ © M Q ¹
where:
ǻH – height corresponding with the loss of mechanical energy during flow of the liquid through
the pump,
K – experimentally fixed constant value for the pump of a given type,
H e – effective hoisting value the pump,
H th – theoretical hoisting value,
ǻH/ H th –relative energetic loss in the pump,
- the remaining denotations as in previous formulas.

Variability testing of relative energetic losses or hydraulic efficiency Ș h leads to determining

of the function f 3 .
In order to determine analytical form of numerical function (12) one should carry out
diagnostic testing of rotodynamic pump in specified work conditions. Basing on these
measurements, one can assume some conditions concerning the form of numerical function.

4. Recapitulation
Analyzing procedures of creating dimensional functions and their transformation into
numerical functions, one can state:
1. Diagnostic testing is carried out in order to make a diagnosis about technical state of the pump;
2. It is necessary for a dimensional function to fulfil interpretation rules accepted together with
notions determining working processes of the pump;
3. Selected dimensional quantities interfere in specification of technical condition of the pump, in

79
 an essential way, and limit considerably a possibility of its specification;
4. Technical condition of the rotodynamic pump can be made a diagnosis of, by means of such
parameters of its work , as:
 operation of rotodynamic pump D,
 capacity (volumetric capacity of the flow) Q ,
 diameter of pump’s rotor d ,
 density of flowing liquid through the pump U ,
 rotational speed of rotodynamic pump rotor n,
 time of pump operation W ,
 coefficient of kinematic viscosity of the liquid flowing through the pump Q .
5. On the basis of the knowledge of linear estimations of numerical functions, presented in table
1, it will be possible to make a diagnosis of technical condition of the pump;
6. Measurements results in the form of dimensional functions of rotodynamic pump operation, in
its important technical states, can be collected, in so called information bank.

5. Bibliography

[1] Drobot S.,On the foundation of dimensional analysis. Studia Mathematica, T XIV, 1954.
[2] Drobot S., Warmus W., Dimensional analysis in sampling inspection of merchandise.
Mathematica, vol. 5, 1954.
[3] Ksprzak W., and others, Formal rules of building mathematical model in objects identification
Archives of Automatics and Telemechanics 2.4. TXV 1970.
[4] àD]DUNLHZLF]67URVNRODĔVNL$7Rotodynamic pumps WNT Warsaw 1968.
[5] Müller L., Application of dimensional analysis in model investigation PWN Warsaw 1983.
[6] Collective work , Manual of mechanics engineer Volume II WNT Warsaw 1969.
[7] Roslanowski J., Modelling of ship movement by means of dimensional function Radom
Uniwersity of Technology,Transport No 3(23), pp 443-448,2005.
[8] Roslanowski J., The methodology of energetic process model construction in ship propulsion
systems by means of dimensional analysis defining their dynamical features. International
Conference Technical, economic and environmental aspects of combined cycle Power plants
*GDĔVN8QLYHUVLW\RI7HFKQRORJ\SS-66.
[9] Roslanowski J., Methodology of building energetic processes models, taking place in vessel
propulsion units, by means of dimensional analysis, in order to determine their dynamic
features Scientific brochure WSM Gdynia 1982.
[10] Siedow L., S., Dimensional analysis and theory of probability in mechanics WNT Warsaw
1978.
[11] 6WĊSQLHZVNL0Pumps WNT Warsaw 1978.
[12] Zierep I., Criteria of probability and rules of modeling in liquid mechanics PWN Warsaw
1978.

80
 APPLICATION OF ENERGY DISSIPATION ANALYSIS FOR
IDENTIFICATION OF MACHINE COMPONENT CONDITION

Andrzej Sadowski, %RJGDQĩyáWRZVNL.XOLĞ(ZD

Uniwersytet Technologiczno - Przyrodniczy

LP-DQDL-ĊGU]HMDĝQLDGHFNLFKw Bydgoszczy
:\G]LDá,QĪ\QLHULL0HFKDQLF]QHM
$OSURI6.DOLVNLHJR-%\GJRV]F]
WHO 
e-mail: DVDGRZVNL#XWSHGXSO

Abstract

Energy transformation in power transmission systems of machines is accompanied by RFFXUUHQFHRIDVHULHVRI

prRFHVVHV ZKLFKFDXVHHQHUJ\ ORVVHV$QDO\VLVRIWKHSKHQRPHQDRFFXUULQJLQWKHSURFHVVRIHQHUJ\WUDQVIRUPDWLRQ
DQG  HQHUJ\ WUDQVPLVVLRQ DQG WKH PHWKRGV XVHG IRU Getermination of its dissipation paths is WKH PDMRU WDVN RI WKLV
VWXG\6WUXFWXUDOFRPSRQHQWVRIDWKUHH-axle vehicle with all-axle drive were XVHGIRUa description and analysis of
degradation mechanisms involved in energy dissipation. Practical aspects of the proposed methodology allow to
define precisely the sites and the PDMRUIDFWRUVDIIHFWLQJRFFXUUHQFHRIORVVHVLQWKe power transmission systems of
vehicles. Demand for methods and means enabling eIIHFWLYH LQVLJKW LQWR WKH FDXVHV DQG HIIHFWV RI HQHUJ\ ORVVHV LQ
PDFKLQHFRPSRQHQWVDUHLVVXHVWKDWQHHGWREHDGGUHVVHGXVLQJ available means and in a possibly short time.

Key words: energy dissipationefficiencyenergy losses power transmission system

1. INTRODUCTION

Energy distribution in a power transmission system of machines is directly connected with

factors such as: the object technical condition, mechanical efficiency of particular subassemblies,
maintenance, strategy of operation, environment of operation, behavior of the operator and many
others. All these factors contribute to rational, effective and safe operation of machines and
is reflected by different factors affecting a machine state including energy losses in its particular
subassemblies.
A machine is referred to as a mechanism or set of mechanisms, powered by a stimulus induced
or applied by a human, designed to perform the assigned work connected with the process of
production, energy transformation. The main part of a machine is the operating component which
is directly involved in energy transformation, power transmission, transformation of movements,
transfer of loads [12]. The power transmission system of machine is an example of such
an operating component which transforms and transfers energy onto other components making
up a set of mechanisms connected with each other.
Members of this system take part in transforming and transferring energy from drive links
onto passive ones, that is, loaded by resistant forces. However, not the whole work of active forces
is utilized for the intended useful goals. Part of the energy is used to overcome friction resistance
accompanying movements, and dissipates into the environment in the form of heat while a portion

81
of the energy accumulates in the mechanism as kinetic energy and sometimes as potential energy
[8]. The aim of this study is: analysis of the phenomena occurring in the process of energy
transformation and transfer and presentation of the method for identification of its dissipation on
the basis of tests carried out with the use of an off-road truck.

2. DEGRADATION OF THE MACHINE ACCORDING TO ENERGY BASED

APPROACH
Degradation of a mechanical system involves loss of structural and functional properties
of their structural components, and in effect of the whole system. This loss will result in change
in mechanical or geometrical properties of the system exceeding the specified tolerances. These
changes appear in effect of functioning in a given environment and external and internal energetic
interactions [2].
According to the general theory of systems, operating systems are open systems characterized
by mass, energy and information flow. Thus, they are systems transforming energy with
its inseparable external and internal dissipation, which is presented in figure 1. Hence, the input
mass flow (material), energy and information are transformed into two output flows, useful energy
in its other desired form or a product being the designed target of a given object. The other flow
includes energy dissipated, partly exported into the environment or the system meta, and partly
accumulated in the object in effect of different wear processes occurring during a machine
operation [2,13].

Fig.1. Machine as a system of energy transformation and its diagnostics possibilities

All possibilities of diagnosing can be classified into three basic groups, one that accounts for
advisability of research on assessment of the quality of condition or the product.

82
The first of them involves diagnosing through observation of operational processes by monitoring
their parameters in a continuous manner, or at special stands carrying out tests of machine
efficiency (power, torque, speed, pressure, etc,) Tests of this kind have a future due to sensors,
microprocessors etc. increasingly used in machines, which requires to be familiar with the model
of the object functioning.
The second method of machine diagnostics includes the product quality tests, consistence
of dimensions, joints, fittings etc. as the better the machine technical condition is the higher
its production quality.
The third possibility of diagnostics involves observation of residual processes, using different
physicochemical processes included in the machine output processes, and being the source
of many attractive machine diagnosis methods [13].
Damage is one the most significant events that occur in the process of machine use,
determining its reliability, efficiency, the process of maintenance and the needs for technical
diagnostics.
Most generally speaking, the concept of machine damage can be defined as an event involving
transition of a machine (system, component) from the state of usability to the state of being
unusable. The state of usability is referred to as such a state in which a machine can perform the
functions specified in its technical documentation. Whereas, the state of being unusable
is understood as such that makes it impossible to perform one of the functions specified in the
technical documentation.
Due to the environmental impact and performance of the assigned tasks the object initial
properties can undergo change, which will result in a change of initial measurable values of the
object properties and possibly in a change of its immeasurable properties.
Machine damage during operation and maintenance can occur:
- in effect of slow, irreversible aging and wear processes , occurring in a machine;
- in effect of reversible processes of variable intensity, caused by temporal exceeding
permissible values of one or more forcing factors;
- in a step-like manner, involving discontinuous transition of one or more properties beyond
the boundaries accepted as permissible for a given machine [12,13].

3. ENERGY DISSIPATION IN THE POWER TRANSMISSION SYSTEM

Energy dissipation is directly connected with such issues as: energy balance, mechanical
efficiency of the discussed object, energy consumption and energy losses. It plays the role
of quantitative presentation of energy transmitted and lost in particular components of the power
transmission system.
The energy contained in fuel in a chemical form called supply which is transformed into
mechanical energy is accepted to be the system feed energy. The amount of energy obtained
in result of mechanical energy transformation depends on the engine efficiency and ranges from
35 to 55% of the whole energy supplied. It is further transmitted by the power transmission
system, being then limited by the quantity of losses connected with energy transfer to wheels.
Mechanisms used for transmission of this energy are called power transmission mechanisms and
the whole system is referred to as a power transmission system. Torque is applied to the axles
of powered wheels in effect of which a tangent reaction of the surface occurs and the vehicle starts
moving. The stream of power flowing to the engine is at some point divided into smaller streams
flowing to particular wheels [1,4,5].
The power stream diminishes while flowing through each mechanism of the power
transmission system which is caused by occurring resistances, which is presented in fig. 2.

83
 The analyzed energy losses can be described by mechanical efficiency of particular
components of the power transmission system as well as by overall efficiency of the power
transmission system.

FigEnergy transformations in power transmission system [own research]

Mechanical efficiency is referred to as the ratio of utilized energy to the energy supplied in the same
time. Improving efficiency involves decreasing energy use. A method for determination of energy
dissipation in the power transmission system is dependent on the way particular structural components are
joined. [4,9,12]. If machine is composed of n mechanisms series connected, this is useful work
L U of each proceeding mechanism work L D supplied to the next mechanism [3,6,8,9,11],
is expressed by dependence (1):
LU 1 LU 2 LUn
K1 ;K 2 ...K n (1)
LD LU 1 LU ( n 1)
After transformation of the above formula we obtain:
K K1 ˜K 2 ˜ ...K n
(2)
In the below figure 3 there are schemes of series connected machine components:

Fig. 3. Series connected machine components [@

In the case of a parallel connection of n machine components, the largest influence on the
system efficiency is on the part of the energy stream division itself, we obtain the following
formula defining efficiency:

84
 LU LD ˜ ¦ Ni ˜Ki
K
LD LD (3)
where:
k i – coefficient of power dissipation.

Fig. 4. Parallel connection of machine components [@

The above presented dependence 3 and figure 4 prove that the overall efficiency of a machine
with parallel connection of mechanisms largely depends on the energy stream division in the
machine [8,9]. It should be mentioned that the presented case of parallel connection
of mechanisms is hardly ever used in machines and most commonly mixed connections are used.

4. METHOD FOR MEASURERING EFFICIENCY OF A SUBASSEMBLY

Energy flowing through the system of power transmission is torque and rotational speed
obtained in effect of processes occurring in an internal combustion engine and connected with
transformation of energy contained in the chemical form in fuel into mechanical energy. In result
of measuring and comparison of energy obtained on the output to the energy used to supply
a given subassembly we obtain its efficiency coefficient defining the size of energy losses. For
quantitative determination of losses, a method for determination of the system efficiency involving
determination of the value of transmitted torque in particular points of the power transmission
system, has been proposed. Measurements of the transmitted torque were performed in a non-
contact manner in service conditions, without a necessity of structural changes in the power
transmission system, on the basis of shaft strain, by means of resistant tensometry [7,10].
The measurement method is based on voltage value measurement on the tensometric bridge
output, and proportional to the strain occurred on the drive shaft due being loaded by torque on the
basis of dependence 4 and 5.

2'U
H (4)
DN8 o

where:

85
ǻU – voltage increase (measured voltage) [V]
a – signal enhancement
U o – tensometer induced voltage [V]
k – tensomeer constant [mV/V]

Fig. . MeasXrement points in the power transmission system ZLWKPHDVXULQJHTXLSPHQW>RZQUHVHDUFK]

1
Mo ˜ H ˜ G ˜ S ˜ (D 3  d 3 ) (5)
8
where:
ʌ – pi number
G – Kirchoff module [Pa]
ᖡ – shaft deformation (measured )
D – shaft external diameter [m]
d – shaft internal diameter [m]

The chosen localization of measurement points aims at minimal intervention into the structures
of power transmission system and simultaneously providing the possibility of verification of the
measurement signal from particular sections of the studied system which is presented in figure 5.
Measurement equipment with telemetric measurement signal transmission enabling wireless
measurement of torque with acquisition of measurement data in the form of voltage signal.

5. RESEARCH
The research object of the performed experiment was a drive system of an off-road truck Star
model 266M2, presented in figure 6. The vehicle was modernized in 2012 in AUTOBOX
Starachowice from version 266 to 266M2.
The basic changes involved the process of modernization connected with the power transmission
system included:
‡ montage of IVECO type F4AE0481 with power 125 kW (170KM) and maximal torque 560
Nm [date obtained from producer].
‡ montage of 6 –gear Eaton gear box,
‡ all the assemblies (including drive axle, transfer case, drive shafts) were overhauled.

86
 Fig5HVHDUFKREMHFW>RZQUHVHDUFK@
3UHSDUDWLRQRIWKHREMHFWIRUWHVWV

Preparation of the object for tests involved preparing the surface of drive shafts for
tensometers to be glued , for this purpose the paint had to be removed from the shafts.
The tensometer was glued by means of cyanogen acrylic glue and next it was dried in higher
temperature.

Fig. . 0HDVXUHPQWV\VWHPZKLOHEHLQJPRXQWHGRQWKHGULYHVKDIW>RZQUHVHDUFK@
Encoder was secured together with an antenna by a ferromagnetic tape eliminating interference
of the measurement signal, the montage way is presented in figure 7. As the measurement system
was provided with additional power source, the distance between the transmission and reception
antennas was increased from 50 to 150 mm. It allowed to perform measurements under service
conditions of the power transmission system and suspension system.

87
 7HVWVFRQGLWLRQV
The tests were carried out outdoor on a surface strengthened with hexagonal concrete slabs
of Trylinka type. During the test the air temperature was 21C°, with the air relative humidity 88%.
Atmospheric pressure: 1-14hPa. Speed of wind: 2km/h. The tests were performed for a fixed
rotational speed of the engine in the amount of 1512 rev/min and in third gear which corresponds
to gear box ratio 1:2.80.
3URFHVVLQJRIPHDVXUHPHQWGDWD
The measurements were performed during six tests on a straight section of a road which
provided the values of voltage. The sampling frequency was established at the level of 41Hz.
He results were recorded and reduced to 1 Hz in result of equalization of measurement blocks.
On the basis the presented mathematical dependencies, the strain of shafts was determined,
according
to dependence (4), for which the following data was accepted:
ǻU – voltage increase (voltage was measured ) [V]
a – signal increase 8000
U o – voltage of tensometer excitation 4V
k – tensometer constant 2,075
Next, the provided values of strain were used in dependence (5) determining torque, basing on:
shape deformability module of Kirchoff module – 80 [GPa],
external diameter – 70 [mm]
internal diameter - 64 [mm]
In result of calculations, the values of torque were provided in established measurement points.
These values are included in figure 8 showing the torque time history.

Fig. . 7RUTXHWLPHKLVWRU\IRUPHDVXUHPHQWSRLQWV>RZQUHVHDUFK@

6. CONCLUSIONS
Analysis of the phenomena occurring in the process of energy transformation and transfer
and the method for identification of its dissipation paths, enable determination of the
efficiency of particular subassemblies of the power transmission system. In result of the

88
 carried out experiment a torque of the transfer box reduced by the value of energy losses was
reported, which enabled determination of the subassembly efficiency. On the basis of the
provided measurement data it was found that the transfer box efficiency was 91.6%. The
proposed method can find application in the process of machine condition identification,
using data from machine efficiency tests (power, torque, peed, etc.) carried out during
its operation and maintenance.

7. BIBLIOGRAPHY

[1] Arcz\ĔVNL60HFKDQLNDUXFKXVDPRFKRGX, WNT, Warszawa 1994.

[2] Cempel C., +ROLVW\F]QHPRGHOHSURFHVyZGHJUDGDFMLZV\VWHPDFKPHFKDQLF]Q\FK– SU]HJOąG,
Diagnostyka nr. 29, 2003.
[3] 'ĊELFNL07HRULDVDPRFKRGXWHRULDQDSĊGX WNT, Warszawa 1976.
[4] -HGOLĔVNL 5 PodwozLD 6DPRFKRGyZ 3RGVWDZ\ 7HRULL L .RQVWUXNFML, WUUTP Bydgoszcz
2007.
[5] Kolator B.: 6WXGLXP GRVNRQDOHQLD SURFHVX IXQNFMRQRZDQLD DJUHJDWX PDV]\QRZHJR Z
ZDUXQNDFKSRORZ\FK WUUTP Bydgoszcz 2010.
[6] Lisowski M., 7HRULD UXFKX VDPRFKRGX – 7HRULD QDSĊGX, Wyd. Uczelniane Politechniki
6]F]HFLĔVNLHM6]F]HFLQ
[7] 0LáHN 0 0HWURORJLD HOHNWU\F]QD ZLHONRĞFL QLHHOHNWU\F]Q\FK, Oficyna Wydawnicza
Uniwersytetu Zielonogórskiego, Zielona Góra 2006.
[8] Miller S., 7HRULD PDV]\Q L PHFKDQL]PyZ DQDOL]D XNáDGyZ NLQHPDW\F]Q\FK, Oficyna Wyd.
3ROLWHFKQLNL:URFáDZVNLHM:URFáDZ
[9] 6DGRZVNL$ĩyáWRZVNL%, BDGDQLDVSUDZQRĞFL]áRĪRQ\FKXNáDGyZQDSĊGRZ\FK, ,QĪ\QLHULD
i Aparatura Chemiczna 2012.
[10] Sadowski A., ĩyáWRZVNL % %H]LQZD]\MQH PHWRG\ SRPLDUX PRPHQWX REURWRZHJR
LOGISTYKA.net.pl 6/2012.
[11] 6LáND:, (QHUJRFKáRQQRĞüUXFKXVDPRFKRGX WNT, Warszawa 1997.
[12] ĩyáWRZVNL % ûZLN = /HNV\NRQ 'LDJQRVW\NL 7HFKQLF]QHM Wydawnictwo Uczelniane
Akademii Techniczno-5ROQLF]HM, Bydgoszcz 1996.
[13] ĩyáWRZVNL%3RGVWDZ\GLDJQRVW\NL0DV]\Q WUUTP Bydgoszcz 2012.

89
90
 THE STRESS STATE ANALYSIS OF RESCUE SEAT AREA OF
SUBMARINE KOBBEN CLASS DURING RESCUE VEHICLE LANDING

Bogdan Szturomski, Marek Bohn

NAVAL ACADEMY in Gdynia, Poland

Mechanical and Electrical Engineering Department,
Institute of Bases Machines Construction
XOĝPLGRZLF]D-*G\QLD
EV]WXURPVNL#DPZJG\QLDSO, PERKQ#DPZJG\QLDSO,

Abstract

This article includes calculating stamina of seat area construction (stress states) during rescue vehicle type
SRC or DSRV GRFNLQJ WDNLQJ LQWR DFFRXQW the depth of immersion and sea current in CAE program, ZKLFK
are the basis for determining the maximum safe depth for use the emergency system. Lists the documents that
contain guidelines for the preparation of the design model using Finite Elements Methods (FEM) [5], reflect the
geometry of the object and its GLVFUHWL]DWLRn, a description of the material, boundary conditions and loads. Posted
examples of the results of the stress state in the design of the seat area of submarine Kobben class rescue obtained
from simulation FEM IRUGHHSP.

Keywords: rescue seat area of submarine, DSRV (Deep Submergence Rescue Vehicle), SRC (Submarine Rescue
Chamber), NSRS (NATO Submarine Rescue System), FEM (Finite Elements Method), CAE (Computer Aided
Engineering), Kobben class, NAVSEA (Naval Sea Systems Command).

1. Introduction:

The object of the research is a rescue seat area of submarine Kobben class, which is a part of
submarine rescue system (Pic. 1.). Rescue seat area enables landing of rescue vehicles and crew
evacuation from damaged submarine by dry method. Rescue seat area was inspected and
certificated and results from recent years shows that its dimensions and required thickness
reached values below limit. In this situation it is necessary to conduct stamina calculations for
current geometry, in order to determine maximum and safe depth for using rescue system.

91
 Pic. . Rescue seat area of submarine Kobben class (ORP 6RNyá) [2]

Nowadays NATO`s equipment is NSRS – (NATO Submarine Rescue System) (pic. 2.) and
phase out DSRV (Deep Submergence Rescue Vehicle) and SRC (Submarine Rescue Chamber).

Pic. 2. Rescue vehicles: old DSRV, SRC and modern NSRS []

Stamina calculations of seat area construction during rescue vehicle landing in function of the
immersion depth and sea current were conducted in CAE program [3]. Rescue seat area stamina
analysis were conducted on basis:
x Defense Standard NO-42-A207:2001, submarines – rescue seat areas. Requirements;
x STANAG 1297 Requirements for a NATO Common Rescue Seat;
x ATP-57(B) The Submarine Search and Rescue Manual;
x International Submarine Rescue Requirements Manual and Instruction Manual for Mating
with U.S. Navy Rescue Assets” (NAVSEA SS700-AA-INS-010).

92
Stamina calculations were conducted in CAE program in the following parts:
1. Measurement of rescue seat area construction, preparing auxiliary drafts and
photographic documentation of rescue seat area with segment of strong hulk;
2. Imitation of rescue seat area solid geometry in CAD programs;
3. Rescue seat area discretization and verification of solid geometry;
4. Determination of rescue seat area burdens in draught function and sea current speed;
5. Execution of mating seat stress and deformation calculations using CAE programs;
6. Analysis of results;

2. Define the real dimensions of rescue seat area

Measurements of seat area elements was conducted in Gdynia harbor during submarine
standstill. Measurements was conducted with use certified measuring equipment with precision
to 1 mm. Measure of thickness was conducted certified supersonic measures made by Metrison
with precision to 0,1 mm. Supporting construction was also measured together with element of
strong hulk. Elements not having influence on stamina construction were skipped – e.g. devices
installed inside on the submarine hulk. Drafts and auxiliary drawings were created on basis of
measuring results (pic. 3).

Pic. . Pictures of seat area construction during standstill in harbor [2]

3. Image of rescue seat area solid geometry and discretization:

Solid geometry of rescue seat area construction was imitated in CAD program. On base of
prepared drafts, drawings and pictures all seat area elements were create, and then they were
integrated (pic. 4).

93
 Pic. 4. *HRPHWU\RIVHDWDUHDZLWKDSDUWRIVWURQJKXON [2]
Solid geometry of seat area was discretized by 125224 linear coating elements, designated in
space by 125442 nodes (Pic. 5.), what constitutes a task with 752652 degrees of freedom.
Irregular elements net was used, thickened in area of seat area construction, where the distance
between nodes carries out 5 mm. Considering strong hulk, outside seat area construction,
elements net were diluted, assumed distance between nodes up to 100 mm.

Pic. 5. 6ROLGJHRPHWU\GLVFUHWL]DWLRQRIUHVFXHVHDWDUHD

4. Stamina analysis of rescue seat area:

Stamina calculations of rescue seat area were conducted by licensed software CAE according
to NAVSEA SS700-AA-INS-010 requirements.

Seat area deformation caused by the pressure of docking rescue vehicle is determined by
solving the finite element equation

94
 ۹൫ઽ௣௟ ൯‫ ܃‬ൌ ۴ሾ‫݌‬௖ ሺ݄ሻǡ ‫݌‬௟ ሺ‫ݒ‬ሻሿ
(1)

where:
۹൫ઽ௣௟ ൯ – stiffness matrix takes into account the structure of plastic material model;
ઽ௣௟ – vector of plastic strain;
U – displacement vector;
F – burden vector;
‫݌‬௖ ሺ݄ሻ – emphasis of the docking vehicle ring depends on the depth of immersion;
‫݌‬௟ ሺ‫ݒ‬ሻ – emphasis of the docking vehicle ring depends on the sea current speed;
h – depth of immersion, m;
‫ – ݒ‬speed of the sea current.

To determine material data requires „cut-out” samples from seat area construction. On base
of literature data, NATO submarines from years ‘60 –‘70 last century were made with steel
HY 80 or HY100. Taking into account data for steel HY 100 is suggested by instruction
NAVSEA SS700-AA-INS-010. In General Engineering Laboratory1 precisely tested steel with
similar properties marked 10GHMBA. Therefore steel data 10GHMBA was taken to calculations
with following properties. It is high-quality steel used for responsible marine structures.
Mechanical properties [7]:
density U = 7830 kg/m
3
plasticity limit R e = 695 MPa
Poisson’s Q = 0.3 stamina limit R m = 759 MPa
5
Young’s module E = 2.09˜10 MPa

Plastic characteristics of material are based on the so-called nominal characteristics H nom - V nom
obtained from the testing machine, according to algorithms described in the literature [3, 5]
(pic. 6).

stress, MPa
900

850

800

750

700 plasticy
nominal
650

600
0,00 0,05 0,10 0,15
strain
Pic. . Nominal characteristic H nom - V nom and plastic H pl - V true

Taking as a model material

௡
ߪ௣௟ ൌ ‫ ܣ‬൅ ‫ߝܤ‬௣௟
(2)

1
General Engineering Laboratory, Institute of Bases Machines Construction, Naval Academy in
Gdynia

95
coefficient values for steel 10GHMBA are as follows:
A = R e = 700 MPa;
B = 640 MPa;
n = 0,6.

Seat area was burden by pressure which is exert by surface (pic. 8, 9), the point of contact with
rescue vehicle while landing on the depths from 100 to 300 meters regarding sea currents,
according to NAVSEA SS700-AA-INS-010. During rescue vehicle landing, after sealing and
water pump out, it causes pressure on seat area construction by column of water pressure on
limited surface by outside diameter of rescue vehicle collar Sea current (side) were assumed as a
linear variable pressure p l counting r 2 MPa(pic. 7). The burden of contact plate for depth of
immersion including sea current running is given by model:
‫ ݌‬ൌ ‫݌‬௖ ൅‫݌‬௟ MPa
( 3)

Pic. 7. %XUGHQRIVHDWDUHDDFFRUGLQJWR1$96($66-AA-INS- >@

Pic. . Collar intersection of landing DSRV >@ Pic. . ,OOXVWUDWHGFRQWDFWRIYHKLFOHULQJZLWKVHDWDUHD>]

Next, analyzed section of strong hulk together with seat area construction were supported
along cut edge of hulk, taking away to all laying there nodes possibility of movement in normal
direction to the cut plane and rotation in other direction. For example, pressure exerted on seat
area surface created by stress of rescue vehicle on the depth 250 m with regard side undercurrent
carries out out 8.36 y 12.35 MPa (pic. 10).

96
 Pic. . Seat area burden p c +p l RQWKHGHSWKP

Analyzed section of strong hulk together with seat area construction were supported along
cut edge of hulk, taking away to all laying there nodes possibility of movement in normal
direction to the cut plane and rotation in other direction (pic. 11).

Pic. . Boundary conditions – seat area support

5. State of stress and displacement on the depth 250 m:

Maximum main stress on the depth 250 m, are locate in interval 0 y 613 MPa (pic. 12).
Reduced stress according to hypothesis HMH (Hubera - Misesa - Hencky’ego) achieving values
651 MPa (pic. 13). Local concentration of stress appear on connection seat area postument with
hulk. Displacements of seat area surface in vertical direction are located in interval -10.6 y -
13.75 mm (pic. 14), maintain flatness.

97
 Pic. . Scheme of maximum main stress inside seat area on the depth 25P [2]

Pic. . Scheme of reduced stress according to HMH hypothesis on the depth 25P

Pic. . State of displacement in vertical direction on the depth 25P [2]
Extreme values of calculated stress and displacements, generated in seat area construction
during rescue vehicle landing at assumed depth are presented in schedule 1.

98
 Schedule Summary of numerical simulation results [2]
maximum
Landing maximum Vertical
reduced stress
depth main stress displacements
according to HMH
m MPa MPa mm
100 241 277 -2.46 y -5.43
200 491 514 -6.47 y -9.44
250 613 651 -10.6 y -13.75
300 741 762 -10.2 y -13.63

6. Conclusions:

Calculated values of main and reduced stress according to hypothesis HMH for depth 250 m
locally reach yield point, on depth 300 m reached stamina limit what means that it is limit depth
on which damage of seat area construction can occur during landing a rescue vehicle. Therefore
calculations for larger depth of immersion are pointless. In case of landing DSRV, SRC,
SRDRS, NSRS, URF and similar rescue vehicle on submarine Kobben class rescue seat area on
the depth 300 m, in area where seat area postument is connected to strong hulk, local and
permanent deformation will occur what will lead to strong hulk crack and unseal.

Based on the results received from CAE simulations states that, landing of DSRV, SRC,
SRDRS, NSRS, URF and similar rescue vehicles on analyzed seat area is safe to the depth 250
m. In case of landing on the depth 300 m in area where seat area postument is connected to
strong hulk, local and permanent deformation will occur what will lead to strong hulk crack.
Maximum immersion for submarine Kobben class is assumed 200 m, therefore border
immersion is estimated at approximately 300 m (permissible x 1.5).

7. Bibliography

[ 1] %RKQ03URMHNWSLHUĞFLHQLDGRGRNRZDQLDSRMD]GXUDWRZQLF]HJRQDSRZLHU]FKQLSU]\OJQLRNUĊWX
SRGZRGQHJRSUDFDLQĪ\QLHUVNDSRGNLHURZQLFWZHP6]WXURPVNL%*G\QLD$0:U
[ 2] -XUF]DN:ĝZLąWHN K., Bohn M., Szturomski B., Report No: 12/6RNyá/LPT/2012, Expert opinion
about rescue seat area submarine 294 ORP„6RNyá”, Part I i II, Naval Academy in Gdynia, 2012 r.;
[ 3] Abaqus 6.12, PDF Documentation, Theory Manual, Simulia, Dassault Systems 2012;
[ 4] %RKQ0:Sá\ZJHRPHWULLSU]\OJQLRNUĊWXSRGZRGQHJRW\SXNREEHQQDUR]NáDGQDSUĊĪHĔSUDFD
PDJLVWHUVNDSRGNLHURZQLFWZHP6]WXURPVNL%*G\QLD$0:U
[ 5] 6]WXURPVNL % ,QĪ\QLHUVNLH ]DVWRVRZDQLH 0(6 Z SUREOHPDFK PHFKDQLNL FLDáD VWDáHJR QD
SU]\NáDG]LHSURJUDPX$%$486:\GDZQLFWZR$NDGHPLFNLH$0:- Gdynia 2013;
[ 6] International Submarine Rescue Requirements Manual and Instruction Manual for Mating
with U.S. Navy Rescue Assets” (NAVSEA SS700-AA-INS-010);
[ 7] )OLV / 6SHUVNL 0 %DGDQLD RGSRUQRĞFL EDOLVW\F]QHM SDQFHU]\ ]H VWDOL JKPED QD RVWU]Dá
SRFLVNDPLPP=HV]\W\1DXNRZH$NDGHPLL0DU\QDUNL:RMHQQHM52./,,15  
[ 8] http://en.wikipedia.org/wiki/McCann_Rescue_Chamber (28.05.2014);
[ 9] http://www.adventurersclub.org/images/DSRV-1.jpg (28.05.2014);
[ 10] http://www.meretmarine.com/fr/content/les-europeens-ont-enfin-leur-sous-marin-de-sauvetage
(28.05.2014);

99
100
 MODELLING OF SOME STEALTH FEATURES FOR A SMALL NAVY
SHIP AT THE CONCEPT DESIGN STAGE - PART II
Natalia Szulist

*GDĔVN8QLYHUVLW\RI7HFKQRORJ\
1DUXWRZLF]D6WU
80-*GDĔVN3RODQG

Abstract

,QWKHSDSHUDIHZSUREOHPVDVVRFLDWHGZLWKPRGHOOLQJWKHEDVLFVWHDOWKIHDWXUHVIRUDVPDOOVKLSDWWKHFRQFHSW
GHVLJQVWDJHDUHLQWURGXFHG2QHSUREOHPFRQFHUQVWKHPRGLILFDWLRQRIWKHLPPHUVHGVKLSKXOOXVLQJWKHUDSLG
FKDQJHRIWKHVKLSORDGLQJFRQGLWLRQ7KHVHFRQGLVDVVRFLDWHGZLWKWKHPRGLILFDWLRQRIWKHVKLSERXQGDU\OD\HU
E\WKHKXOOVNLQFRYHU7KHRWKHUVWHDOWKIHDWXUHVRIWKHVKLSDUHQRWSUHVHQWHGLQWKLVSDSHU7KHUHVHDUFKPHWKRG
LVEDVHGRQWKHULVNDVVHVVPHQWDQGDVVHVVPHQWRIWKHVKLSSHUIRUPDQFH7KHULVNPD\EHHVWLPDWHGDFFRUGLQJWR
D SURSRVHG ULVN PRGHO 7KH ULVN DVVHVVPHQW FDQ EH GRQH DFFRUGLQJ WR WKH ULVN PDWUL[ FULWHULD 7KH VKLS
SHUIRUPDQFHDVVHVVPHQWLVEDVHGRQPRGLILFDWLRQRIWKHVKLSKXOOSRVLWLRQRQWKHZDWHUVXUIDFHDQGPRGLILFDWLRQ
RIWKHVKLSERXQGDU\OD\HU6RPHHOHPHQWVRIWKHUHVHDUFKPHWKRGDQGPDWKHPDWLFDOPRGHOIRUHVWLPDWLQJWKH
VKLSSHUIRUPDQFHDUHLQWURGXFHGLQWKHSDSHU*HQHUDOO\WKHUHVHDUFKLVDVVRFLDWHGZLWKWKHLQWHUGLVFLSOLQDU\
LQYHVWLJDWLRQV DQG PXOWL-FULWHULD VKLS GHVLJQ 7KH SUREOHPV SUHVHQWHG LQ WKH SDSHU IROORZ IURP WKH D 3K'
UHVHDUFK FRQGXFWHG DW WKH )DFXOW\ RI 2FHDQ (QJLQHHULQJ DQG 6KLS 7HFKQRORJ\ *GDĔVN 8QLYHUVLW\ RI
7HFKQRORJ\

Keywords: VPDOOQDY\VKLSVWHDOWKWHFKQRORJ\VKLSK\GURPHFKDQLFVVKLSSHUIRUPDQFHULVNDVVHVVPHQW

1. Introduction
The navy operations at sea often require to apply more and more advanced multi-task
ships. The following paper regards the problems associated with modelling some stealth
features for a multi-task small navy ship at the concept design stage. The ship should enable
to perform the general tasks, reconnaissance and combat operations, to support the safe and
rescue operations, to launch the water drones, etc. Despite of the size the multi-task navy
ships may be the platforms for the flying drones and unmanned water vehicles including the
autonomous water drones.
The main objective of the research is to work out a functional model of the multi-task
small navy ship having some stealth features. To obtain the stealth features the modification
of the immersed ship hull form and modification of the ship boundary layer may be used.
The research method combines the navy ship risk assessment and assessment of the ship
performance.
The risk assessment requires first of all to estimate the ship susceptibility, vulnerability
and killability and consequences. Then the risk may be assessed using the risk matrix criteria.
The risk assessment is associated with the ship killability which may be understand in the
opposite way as the ship survivability. According to the risk estimation for the data scenario
development the safety assessment of the ship may be done for the given operational
conditions.

101
 The ship performance assessment requires to define the operational conditions first of
all and next to estimate the ship floatability, stability, resistance and propulsion characteristics.
Before the ship performance may be started the ship weight should be estimated according to
the data operational condition. At this stage of design the operational conditions concern
mainly the ship loading condition than the sea state. The seakeeping of the ship will be
considered at the preliminary stage of design.

2. The research problem

The aim of this paper is to outline a study into a feasibility of an advanced multi-task
small navy vessel having some stealth features enabling to The main objective of the research
is to work out a functional model of the multi-task small navy ship having some stealth
features. The ship should be able to move on the water surface in two different operational
conditions with the range of speed up to 30 knots. When the operational conditions have to be
changed the ship speed may be substantially decreased.
To obtain the stealth features the modification of the immersed ship hull form by the
relatively rapid change of the loading condition and modification of the ship boundary layer
by the hull skin cover may be used.
It is proposed that the modification of the immersed ship hull form is done by the
sequential rapid change of the loading condition. The loading condition very much depend on
the quantity of water in the ballast tanks. It has been decided to take into account two loading
conditions which have the impact on the ship stealth features. The first operational loading
condition is called as the surface loading condition when the ship has no water or having the
remaining quantity of water in the ballast tanks. The second is called as the immersed loading
condition when the ballast tanks are full of water. The process of ballasting the ship should be
rather fast or rapid in time. Therefore the dynamical stability of the ship is a very important
problem to be solved during the research. The ship speed should be decreased during the ship
ballasting.
It is proposed that the modification of the ship boundary layer by the ship skin cover
would increase the stealth effect.
The influence of the ship hull immersion and ship boundary layer on the ship resistance
and propulsion it may be investigated by estimating the flow around the ship hull. This may
be checked using either the computational fluid dynamics technique or investigations using
the circulating channel.
The study has been driven by the tactical advantages that such the ship could offer to the
military authorities and for the commercial users as well.
The major assumptions concerning the multi-task small stealth navy ship have been
defined as follows:
- main dimensions (length x breadth x height): (60 meters) x (10 meters) x 8.4
(meters),
- displacement up to 900 tons,
- operational speed up to 30 knots.
The ship stealth domain has been defined as a combination of the ship features as
follows:
- feature 1: ship hull form enabling to change the efficiency of the ship detection in
the air and in the water,
- feature 2: materials including the ship hull skin cover enabling to change the
efficiency of the ship detection in the air and in the water,
- feature 3: radiation of the noise and vibrations,
- etc.
The research problems to be solved are as follows:

102
 - development of a model for estimating the influence of the ship hull form on the
efficiency of the ship detection in the air and in the water depending on the loading
condition,
- development of a model for estimation the influence of the ship hull skin cover on
the boundary layer, wake and ship resistance.
The challenges associated with the research are significant. It follows from the fact as
an example that the loading condition and boundary layer (wake and resistance) have
divergent design requirements in general.

3. The research method

The research method is a kind of risk-based performance-oriented method which
enables to assess the safety and performance of a ship at the design stage and in operation,
Gerigk (2010). The method enables to take into account the influence of factors following
from different sources including the design, operational and safety management related
factors. The risk assessment and performance assessment may be done for all the possible
design scenarios and sequences of events. All the above describe the so-called holistic
approach to ship safety.
Within the method the risk assessment is based on application of the risk model which
may be modified according to the scenarios of events under consideration. The risk
assessment requires first of all to estimate the ship susceptibility, vulnerability and killability
and consequences. Then the risk may be assessed using the risk matrix criteria. The risk
assessment is associated with the ship killability which may be understand in the opposite
way as the ship survivability. According to the risk estimation for the data scenario
development the safety assessment of the ship may be done for the given operational
conditions.
The risk assessment. The risk model associated with the different hazards and scenarios
regarding a multi-task small navy ship should be estimated according to the well known
general formulae, Gerigk (2010):

R i = Pi x Ci (1)

where:
P i - probability of occurrence of a given hazard;
C i - consequences following the occurrence of the data hazard and scenario
development, in terms of fatalities, injuries, property losses and damage to the
environment.
In the case of the multi-task small navy ship depending on a hazard occurred a general
formulae for estimating the risk may be presented as follows:

R i = P HO P SD/HO (1-P SNSS ) C (2)

where:
P HO - probability of the data hazard occurrence,
P SD/HO - probability of the data scenario development conditional on the data hazard
occurrence,
P SNSS - probability of small navy ship survivability conditional on the data scenario
development and conditional on the data hazard occurrence,
C - consequences regarding the fatalities C HF (HF - Human Factor), property (ship)
C SNS (SNS - Small Navy Ship as a whole), environment C E (E - Environment),
estimated at each stage of an accident (catastrophe).

103
 A major hazard taken into account during the research is the killability of the small navy
ship under consideration. The probability of the data hazard occurrence, probability of the
ship killability may be written as follows:

P HO = P K = P S P K/S (3)

where:
P K - probability of the ship killability,
P S - probability of the ship susceptibility,
P K/S - conditional probability vulnerability in the case of the ship damage.
The probability of the ship susceptibility P S is the measure of the ship inability to
intercept any of the threats: detecting, classifying, targeting, attacking or hitting. The key
issue from the interception point of view are the stealth features the multi-task small navy
ship may possess.
The probability P SNSS may be estimated using the following methods, Gerigk (2010):
- binary method;
- method based on definition of the ship hydromechanic characteristics,
- method based on definition of the ship performance including the ship dynamics in waves.
In the case of the last method the sway, heave, roll and pitch functions in time domain
have been anticipated as the major characteristics enabling the risk assessment, Gerigk
(2010). The risk analysis requires to estimate the conditional probabilities concerning the
major events (initial events, hazards), intermediate events and final events (consequences).
The typical additional events may concern the water on deck, air cushions, cargo leakage,
additional heeling moments, etc. The above risk model should be prepared in such a way to
enable to consider many of the possible scenarios using the event tree analysis ET
The performance assessment. The performance assessment may be done using either the
physical model (towing tank or circulating channel investigations) or numerical simulation
techniques. It has been assumed that the computer simulation will be used during the research.
Within the research the ship performance require to estimate the ship floatability, stability,
dynamical stability and resistance and propulsion.
The ship hydromechanic characteristics associated with the above features are the base
for estimation how much the modification of the immersed ship hull form by the sequential
rapid change of the loading condition affects the ship stealth characteristics like the ship
detection in the air and in the water. The latest concerns the modification of the ship boundary
layer by the ship skin cover as well.
The ship performance assessment requires to define the operational conditions first of
all and next to estimate the ship floatability, stability, resistance and propulsion characteristics.
Before the ship performance may be started the ship weight should be estimated according to
the data operational condition. At this stage of design the operational conditions concern
mainly the ship loading condition than the sea state. The seakeeping of the ship will be
considered at the preliminary stage of design.
It is proposed that the stealth features of the ship will be delivered by the modification
of the immersed ship hull form and modification of the ship boundary layer.
How big is the influence of the ship hull immersion and ship boundary layer on the ship
resistance and propulsion is the may aim of the research.
From the formal point of view the aim of research is to work out a method for analyzing
two basic features deciding if a ship may be considered as a stealth ship. The first feature is
the modification of the immersed ship hull form by a relatively rapid change of the loading
condition. The second feature is the modification of the ship boundary layer by the ship hull
skin cover. Both the features are closely connected with the ship hydromechanics. The

104
research contains the method, models, numerical computations, validation and verification of
results obtained and conclusions.
The problems associated with the ship performance should be precisely considered for
the multi-task small navy ship under consideration during the Ph.D. studies carried out at the
FacuOW\ RI 2FHDQ (QJLQHHULQJ DQG 6KLS 7HFKQRORJ\ *GDĔVN 8QLYHUVLW\ RI 7HFKQRORJ\ The
problems regarding the ship maneuverability and seakeeping will be investigated according to
a methodology of another research project.
As it was mentioned before the research method combines the navy ship risk assessment
and assessment of the ship performance.
The method. The method is based on the following main steps:
- setting up the requirements, criteria, limitations, safety objectives,
- defining the ship including the hull form, arrangement of internal spaces, ballasting,
weights and centre of gravity, loading condition,
- defining the environment,
- identifying the hazards and sequences of events (scenarios),
- assessing the ship performance including the ship floatability, stability, dynamic
stability (research task1: modification of the immersed ship hull form is done by the
sequential rapid change of the loading condition), ship flow and resistance (research
task 2: modification of the ship boundary layer by the ship skin cover),
- estimating the risk according to the event tree analysis ETA and risk model (risk value
is estimated for each scenario separately),
- assessing the risk according to the risk acceptance criteria (risk matrix) and safety
objectives, (Abramowicz-Gerigk 2006, Abramowicz-Gerigk 2008b, Abramowicz-
Gerigk & Burciu 2013, Arangio 2012, Burciu & Grabski 2011, Gerigk 2004, Gerigk
2005, Gerigk 2006, Gerigk 2008, Gerigk 2010, Gerigk 2012, Nowakowski &
:HUELĔND).
The structure of the method is presented in Figure 1, Gerigk (2010).
The criteria within the method is to achieve an adequate level of risk using the risk
acceptance criteria, risk matrix, Gerigk (2010). Providing a sufficient level of safety based on
the risk assessment is the main objective. It is either the design, operational or organizational
objective. Safety is the design objective between the other objectives. The measure of safety
of the object is the risk (level of risk). The key drivers during the ship design and during the
research on the multi-task small navy ship are the research task 1 and research task 2.

105
 Start

Safety objectives Hull form

Standard design objectives
Design requirements, criteria, limitations Arrangement of
Risk acceptance criteria Ship/aircraft internal spaces
Limitations concerning the costs and benefits
Design/operation Loading conditions

Natural Wind
Aircraft/ship and environment environment
definition
Waves

Hazard identification Accident categories

Modification of design or Hazard assessment Another

operational procedure
Stranding
Identification of accident scenarios

Risk Ranking the Collision

control hazards, Estimation of the
options: Risk acceptance probability of hazard
-prevention criteria occurence Pi
-reduction
-mitigation
Estimation of Assessment of aircraft/ship
accident performance in:
consequences Ci - undamaged conditions
- damaged conditions
Risk assessment:
Is risk tolerable? Risk estimation
Ri = Pi * Ci

No Yes
Models of risk

Costs/benefits analysis

Yes Are costs too high ? No

Choice of optimal design or

operational procedure
Pi, Ci, Ri – concerns the
iterations in respect to
all the possible events System of making the decisions on the object
accident scenarios safety in undamaged/damaged conditions

End

)LJXUH7KHVWUXFWXUHRIWKHUHVHDUFKPHWKRGFRPELQLQJWKHULVNDVVHVVPHQWDQGSHUIRUPDQFHDVVHVVPHQWRIWKH
PXOWL-WDVNVPDOOQDY\VKLS
From the design point of view the main research tasks associated with development of
the small stealth ship concept are as follows:
- development of ship hull form and arrangement of internal spaces,
- development of ship structure and ship hull skin nano-cover,
- selection of ship devices and subsystems,
- estimation of ship mass and centre of gravity,
- investigations of ship performance,
- investigations of selected problems associated with the ship performance during the
towing tank or circulating channel investigations,
- investigations of influence of the ship hull skin nano-cover on the boundary layer

106
 using both the computer simulation and physical model investigations.

4. Basic information on the model for modification of immersed hull form by

sequential rapid change of loading condition
I has been assumed that modification of the immersed ship hull form should be
achieved by the sequential rapid change of the loading condition. The buoyancy of the ship
increases and decreases according to the efficiency of the ballasting pumps. Two options have
been considered regarding the speed changing the loading condition form the floating loading
condition to the immersed loading condition and in the opposite way. It is connected with the
rate of flow of the ballasting pumps. The maximum quantity of the ballast water (volume of
the ballast tanks) is of about 200 tons. Two pumps of the rate of flow 10 cubic meters per
minute or two pumps of the rate of flow 25 cubic meters per minute may be applied. It means
that the ballast tanks may be full of water in 10 minutes in the first case and in four minutes
using the pump of efficiency 25 cubic meters per minute.
During the ballasting process the dynamic stability of the ship should be permanently
controlled. It is very important to the location of the centre of gravity and ship centre of
buoyancy in each step time.
Thus the restoring moments M R (transversal or longitudinal) at small changes of heel is
composed of the moment due to the ship buoyancy and moment due to the ship weight:

MR = MB + MW (4)
where:
M B - moment due to the ship buoyancy, M B = V ȡg GZ quasi-static ,
V - immersed buoyancy of the ship,
ȡ- density of water,
g - gravity acceleration,
GZ quasi-static - righting arm of buoyancy for the quasi-static condition at each buoyancy
increase and time step,
M B - moment due to the ship weight, M W ™ ȡg ǻV Ti ) r i ,
ǻV Ti - volume of ballast water in the ballast tank under consideration,
r i - heeling arm following from the ballast water in the data ballast tank.

The scheme of modification of immersed ship hull by sequential rapid change of

loading condition is presented in Figure 2.

)LJXUH7KHVFKHPHRIPRGLILFDWLRQRILPPHUVHGVKLSKXOOE\VHTXHQWLDOUDSLGFKDQJHRIORDGLQJFRQGLWLRQ

107
 5. Some remarks on modification of the ship boundary layer by the ship skin cover
It is prepared that the simulation of the ship flow will be based on the RANSE
(Reynolds Averaged Navier-Stokes Equations) flow model, implemented in the STAR-CCM+
solver. It has been assumed that the simulation in the calm sea condition will be considered
first of all. Later the problems of the ship motion in waves will be analyzed.
The simulation of the ship motion includes the following problems:
- taking into account the free surface of water;
- solving the equations of motion for the floating ship in calm water and in waves;
- preparing the dynamic (i.e. moving) mesh; the floating ship is moving during the
simulation; it means that the nodes of mesh in which the flow is being solved should
move as well.
The Volume of Fluid (VOF) model will be used to capture the flow of two fluids (air
and water) on the free surface of water. It means that the fluids (air and water) filling the
domain are treated as single fluid. An additional variable is introduced for each of the
components: “volume fraction” FL . The parameters of the resulting multiphase fluid (e.g.
density U ) are then computed as follows, by Kraskowski, Gerigk et al. (2012): U ¦UL
L ˜ FL ,

where FL and U L are the volume fraction and density for each of the components,
respectively. Note that ¦F
L
L 1 in each point of the domain.

Solving the motion of the floating ship it is necessary to integrate the Newton equations
using the forces due to weight of the structure and forces exerted by the fluids (these are
updated in each time step). It has been assumed that the sway, heave, roll and pitch motions
are the crucial characteristics from the ship performance point of view. A simplification is
used: only the motion in the mentioned degrees of freedom are considered. In the case of
using the “sliding mesh” approach introducing such a simplification greatly simplifies the
mesh generation, by Kraskowski, Gerigk et al (2012).
As mentioned above, solving the flow in the domain with moving boundaries (in the
considered case – the boundaries of the floating object), requires using the dynamic mesh.
Different approaches are possible here: rigid mesh (no relative motion between nodes),
deforming mesh, overlapping mesh and sliding mesh, by Kraskowski, Gerigk et al (2012).
Such an approach provides good compromise between the accuracy, robustness and
computational time. The ship hull is closed into the cylindrical sub-domain located in larger
rectangular domain. The inner sub-domain undergoes angular and translational motion, while
the outer domain undergoes only translational motion, by Kraskowski, Gerigk et al. (2012).
For the presented analysis the sliding mesh approach was selected and its simplified
example is presented in Figure 3.



)LJXUH7KHVLPSOLILHGH[DPSOHKRZ
WKHVOLGLQJPHVKDSSURDFKZLOOEH
DSSOLHGIRUWKHPRWLRQDQDO\VLVRIWKH
VPDOOQDY\VKLSXQGHUFRQVLGHUDWLRQ E\
0.UDVNRZVNL6KLS5HVHDUFKDQG
6KLSEXLOGLQJ&HQWUH 

6. A concept of a small stealth ship

The primary aim of the research is to work out a functional model of functional model

108
of the multi-task small stealth navy ship which should be able to move on the water surface
with two different operational (loading) conditions with the range of speed up to 30 knots.
The novel solutions have been applied regarding the hull form, arrangement of internal
spaces, materials and propulsion system. The ship is defined as a mono-hull having an
unconventional arrangement of internal spaces.
The major factors to obtain the stealth features of the ship are: hull form, hull skin
cover, limited boundary layer (nano-skin) and wake, limited vibration and acoustic emission.
A visualization of the second version (2nd design version) of the multi-task small navy
ship MTSNS under consideration is presented in Figure 4.

)LJXUH$YLVXDOL]DWLRQRIWKHVHFRQGYHUVLRQ QGGHVLJQYHUVLRQ RIWKHRIWKHPXOWL-WDVNVPDOOQDY\VKLS

07616

The MTSNS has the following main dimensions:

- Length L = 60 m,
- Breadth B = 8 m,
- Maximum speed v = 30 knots.
The main technical data of the MTSNS are as follows:
- a multi-purpose vessel for the combat or/and logistic tasks,
- innovative solutions applied: multi-fuel main engine, lift for launching the water drones
and air drones, stealth technology features.

7. Conclusions
The problems presented in the paper are connected with the research on the modelling
of the stealth features for the multi-task small navy ship MTSNS at the concept design stage.
The research itself seems concerns the application of an advanced approach to the stealth
technology problems presented in the paper.
The major research problems from the stealth technology application point of view have
been defined as follows:
- research task1: modification of the immersed ship hull form is done by the sequential
rapid change of the loading condition, ship flow and resistance,
- research task 2: modification of the ship boundary layer by the ship skin cover.
The research presented in the paper is divided into a few parts. One of them is
associated with submitting the projects to be supported by the NCBiR Ministry in Warsaw.
The whole research is performed by a team of Ph.D. students.

8. References
[1] Dudziak J., 7HRULDRNUĊWX 7KHRU\RI6KLSV , Publisher: )XQGDFMD3URPRFML3U]HP\VáX
2NUĊWRZHJRL *RVSRGDUNL0RUVNLHM*GDĔVN 2008.

109
[2] Faltinsen O.M., 6HD /RDGV RQ 6KLSV DQG 2IIVKRUH 6WUXFWXUHV, &DPEULGJH 8QLYHUVLW\
Press, 1990.
[3] *HULJN0.RPSOHNVRZDPHWRGDRFHQ\EH]SLHF]HĔVWZDVWDWNXZVWDQLHXV]NRG]RQ\P
z XZ]JOĊGQLHQLHP DQDOL]\ U\]\ND '6F WKHVLV LQ 3ROLVK  0RQRJUDphy:
Monografie 101, Publisher: :\GDZQLFWZR3ROLWHFKQLNL*GDĔVNLHM*GDĔVN
[4] Grabowski M., Merrick J. R. W., Harrald J. R., Mazzuchi T. A., Rene van Dorp J., 5LVN
PRGHOLQJLQ GLVWULEXWHG ODUJH-VFDOH V\VWHPV, “IEEE Transactions on Systems, Man, and
Cybernetics – part A: Systems and Humans”, Vol. 30, No. 6, November 2000.
[5] Jasionowski A., Vassalos D., &RQFHSWXDOLVLQJ 5LVN, Proceedings of the 9th
International Conference on Stability of Ships and Ocean Vehicles STAB 2006, Rio de
Janeiro, 25-29 September 2006.
[6] Gerigk M., Kraskowski M. Safety assessment of ships in damaged conditions by the
risk-based method. Assessment of damaged ship performance using the CFD modeling.
Joint 19th International Conference on Hydrodynamics in Ship Design and 4th International
6\PSRVLXPRQ6KLS0DQRHXYULQJ,áDZD Poland, 19th - 21st September 2012, ISBN 978-83-
9229-352-1, pp. 100-110.
[7] .UĊĪHOHZVNL0, +\GURPHFKDQLNDRJyOQDLRNUĊWRZD (General ship hydromechanics),
F]ĊĞü, (part I), Skrypt Politechniki GdDĔVNLHM*GDĔVN
[8] .UĊĪHOHZVNL0, +\GURPHFKDQLNDRJyOQDLRNUĊWRZD(General ship hydromechanics),
F]ĊĞü,, (part II), 6NU\SW3ROLWHFKQLNL*GDĔVNLHM*GDĔVN
[9] Pillay A., Wang J., 7HFKQRORJ\ DQG 6DIHW\ RI 0DULQH 6\VWHPV, Elsevier Ocean
Engineering Book Series, Volume 7, Elsevier 2003.
[10] Sii H.S., Ruxton T., Wang J., 1RYHOULVNDVVHVVPHQWDQGGHFLVLRQVXSSRUWWHFKQLTXHV
IRUVDIHW\ PDQDJHPHQWV\VWHPV, Journal of Marine Engineering and Technology, No. A1,
2002.
[11] Skjong R., Vanem E., Rusas S., Olufsen O., +ROLVWLF DQG  5LVN %DVHG $SSURDFK WR
&ROOLVLRQ'DPDJH 6WDELOLW\ RI 3DVVHQJHU 6KLSV, Proceedings of the 9th International
Conference on Stability of Ships and Ocean Vehicles STAB 2006, Rio de Janeiro, 25-
29 September 2006.

110
 APPLICATION OF INFORMATION TECHNOLOGIES IN
OPTIMIZATION OF THE MACHINE STATE EVALUATION PROCESS
Joanna Wilczarska

University of Technology and Life Sciences

ul. Kordeckiego 20, 85-225 Bydgoszcz, Poland
tel.: +48 52 3408283, fax: +48 52 3408286
e-mail: asiulazol@utp.edu.pl
Abstract

The use of information technologies in machine diagnostics allows to optimize the process of the machine
state evaluation. Information technologies are tools used to support the machine technical state. There is a need
to do more research into these technologies and continue to develop them as they provide methods for the
process optimization. Multiple criteria optimization methods often find application in selection of a set of
diagnostic parameters to be further used for assessment of a machine state. This study deals with the properties
and possibilities of OPTIMUS software used for the optimization process.

Keywords: technical state of a machine , multiple criteria optimization, information technologies

1. Process of the machine state evaluation

The methodology for the machine state evaluation process consists of a few stages
involving the following diagnostic procedures (fig.1) [1]:
a) diagnostics – a process involving determination of the machine state in time 4 b ;
b) prognostics - a process involving predicting the machine future condition, enabling for
example, to schedule the date of next servicing 4 d .
c) genesis – a recovery process of a machine condition history in order to evaluate the
machine past performance;
which in turn makes it possible to:
a) determine the machine actual technical condition on the basis of diagnostic tests
results. It enables control of the machine condition and sites of damage when the
machine is not usable.
b) predict the machine future state on the basis of an incomplete history of diagnostic
tests results. It enables estimation of the machine correct operation time or work to be
performed by the machine in the future.
The main tasks to be formulated in order to solve the problems connected with evaluation
of the machine state include [2]:
- formulation of the goal of the machine state diagnosis, prognosis and genesis;
- change of the machine states during operation;
- description of the machine state by means of its characteristics and the dependence
between the characteristics and diagnostic parameters.

111
 - solution of the problem connected with the machine state diagnosing;
- solution of the problem connected with the machine state prognosis;
- solution of the problem connected with the machine state genesis.

Fig.1. Scheme of the machine state evaluation in the system of maintenance [1]

Number of states which a machine can undergo results from its constructional –functional
structure. Three states can be distinguished for a complete machine:
9 usable machine;
unusable machine;
9 machine usable to perform tasks.
The state referred to as ‘machine usable to perform tasks’ refers to a situation when there
are failures impairing its general technical condition though the required quality of its
operation is maintained. In this case, according to the theory of technical diagnostics, though
being unusable, the machine can successfully perform the assigned tasks. The machine
technical condition should be considered together with the processes that are involved in its
operation because the operation quality is a function of its technical condition. In practice, it is
not necessary or possible to distinguish too many states a machine goes through as the
probability of occurrence of simultaneous failure of more than one component is very low.
Thus, with the assumption of a single failure occurrence, the number of states in which a
machine is unusable equals the number of its K elements [2].

2. Multi-criteria optimization in assessment of the machine condition

In optimization of one criterion all the information necessary for finding a solution should
be collected before calculations, whereas the decision process is often of iterative and multi-
stage nature involving research affecting the solution. It often happens that during the
research there emerges a need to apply a new criterion. Sometimes it is not possible to
account for all the significant factors, or the optimization leads to a breach of constraints
difficult to formalize or model, hence it appears that the formal apparatus for multiple criteria

112
optimization which reflects complicated decision problems is more useful. Optimization of
the machine state evaluation process should include the following stages:
a) determination of a set of permissible solutions X,
b) description of criteria based function F, whose elements are an exemplification of
qualitative and structural requirements resulting from the properties of the process of
machine maintenance and the operation ( degree of assemblies fatigue wear,
operational factors, structural-qualitative factors, character of external forcing,
postulates connected with the formulated genesis goal);
c) solution of the problem of multiple criteria optimization and determination of the
‘best’ set of diagnostic parameters and the ‘best’ method of a machine condition
evaluation;
d) assessment of the quality of an optimal solution through identification of sensitivities
of the obtained procedures to some factors involved in the machine operation process
(which have not been considered as elements of a criteria based function F).
The set of diagnostic parameters is to be specified from among the set of output
parameters. On the basis of tests results and findings of research works whose aim was to
confirm certain proposals concerning reduction of diagnostic information, it was found that to
determine a set of diagnostic parameters for the machine state evaluation it is necessary to
take into account [1]:
a) ability to represent the machine state changes during its operation;
b) some amount of information about the machine state;
c) appropriate variability of diagnostic parameters values during the machine operation
and maintenance.
In most cases, identification of a set of diagnostic parameters from the set of output
parameters Y  Y wy involves application of the minimum error criterion. Parameters which
are characterized by a minimum diagnosis error are included here, as well as a procedure for
selection of diagnostic parameters according to the minimum diagnosis error.
The essence of this method is to define the error of diagnosis D, that is, the ‘coverage’ area of
function of parameter \Mࣅ Y conditional probability density, defined by Serdakow [12.20] as
dependence:
§S · §S ·
D P¨¨ 1 ¸¸ ˜ Q1  P¨¨ 2 ¸¸ ˜ Q2 , (1)
© yj ¹ © yj ¹

whereas, the probability of type I error Q involving classification of a machine which is in

usability state S0=S1 as being unusable S1= S2:
f
§y ·
Q1 ³
y gr
f ¨ j ¸ dy j ,
© S1 ¹
(2)

and error probability of type II error Q 2 involving classification of a machine being

unusable S1= S2 as being in usability state S0=S 1 :

§ yj ·
y gr

Q2 ³ f ¨© S 2¹
¸ dy j , (3)
f

Next, there comes the choice of the best parameter y*Y through minimization of the
diagnosis error.
y * min D j , (4)
j

113
The choice of diagnostic parameters according to the presented method, is then reduced to:
1. Qualitative analysis of parameters involving [1,3]:
a) investigation of significance of diagnostic parameters value changes for the machine
technical state change,
b) determination and estimation of boundary values y gr according to Bayes minimum
risk criterion with the assumption of type I and II errors value of costs:
2. Quantitative analysis which involves choosing parameters according to criteria that would
be able to reflect the machine state changes during its operation and maintenance, provide
adequate information on the machine state and variability of diagnostic parameters value
changes during the machine operation. Appropriate methods accounting for these postulates
are presented below:

Method of diagnostic parameter maximal relative change

This method involves a choice of a diagnostic parameter whose kj index is of the highest
value. It accounts for the average speed of parameters change in time interval (4 1 , 4 b ). It is
defined according to dependence [2]:

bj
kj = m
,
¦b
j =1
j

K y j (4 i +1 )  y j (4 i )
bj = 1 ¦ , (5)
K i =1 (4 i 1  4 i ) y j (41 )  y j , g

where:
K- population of elements in a time array in interval (4 1 , 4 b ).

Method for correlation of diagnostic parameters values with the machine state
This method involves investigating correlation of diagnostic parameters values with the
machine state r j =r(W, y j ) (possibly with operation time, (r j = r((4, y j )) [2]:

¦ (4
k 1
k  4)( y j ,k  y j )
rj = , (6)
K K

¦ (4
k 1
k  4) 2
¦(y k 1
j ,k  yj) 2

K K
1 1
4
K
¦4 k 1
k , yj
K
¦y
k 1
j ,k , (7)

where:
r j = r(S, y j ); j = 1,..., m – correlation coefficient between variables S and y j ,
r jn = r(y j , y n ); j,n = 1,..., m; jzn – correlation coefficient between variables yj and yn .

In case there is no data from set S, it is being replaced with the machine operation time,
with the assumption that determination of procedures for evaluation of the machine state is
performed within the interval of normal wear.
Then, r j = r(4, y j ); j=1,..., m; k=1,...,K (r j – coefficient of correlation between variables
4 k (4 1 , 4 b ) (4 k – machine operation and maintenance i y j ).

114
Method of diagnostic parameter maximum information capacity
The method involves choosing a parameter providing the largest amount of information
about the machine state. A diagnostic parameter is the more valuable for evaluation of a
machine state change the more correlated it is with the state and the weaker correlated it is
with other diagnostic parameters.
This dependence is presented in the form of an index of diagnostic parameter hj
information capacity which is a modification of an index referring to a set of variables
accounting for econometric model [2]:
r j2
hj = m
, (8)
1  ¦ r j ,n
j , n 1, j z n

¦(y
k 1
j ,k  y j )( y n ,k  y n )
r j,n = , (9)
K K

¦(y
k 1
j ,k  yj) 2
¦(y
k 1
n ,k  yn ) 2

K K
1 1
yj
K
¦yk 1
j ,k ; yn
K
¦y
k 1
n ,k , (10)

where:
r j = r(S, yj ); j = 1,..., m – coefficient of correlation between variables S and yj ,
r jn = r(y j , yn ); j,n = 1,..., m; jzn – coefficient of correlation between variables yj and yn .

When there is no data from set S it is replaced with the machine operation and
maintenance time, with the assumption that determination of the procedures for evaluation of
a machine state is performed within the time of its normal wear.
An advantage of the above presented methods is the fact that they allow to choose one-
element and multi-element sets of diagnostic parameters. A one-element set refers to the case
when a machine is decomposed into subassemblies and it is necessary to choose one
diagnostic parameter. A multi-element set can be obtained when the presented procedures
provide less strict constraints involving acceptance of those parameters whose index values
are higher (lower) than, the accepted for the method, high (low) positive numbers.
The above presented discussion, formulated in the form of an algorithm for determination
of a set of diagnostic parameters includes the following stages [1]:
1. Data acquisition:
a) set of diagnostic parameters values in the machine operation and maintenance time,
{yj (4 k )}, obtained during performance of a passive-active experiment, where
4 k (4 1 , 4 b );
b) set of diagnostic parameters values : {y j (4 1 )} – nominal values, {yjg }- boundary
values, j=1, …, m;
c) set of the machine states {4 k : {s i }, k=1, …, K; i=1,…, I}defined during passive-
active experiment, where 4 k (4 1 , 4 b );
d) cost of diagnostic parameters c(yj ) = const.
2. Optimization of s set of diagnostic parameter values (only in case of high quantity of set Y,
np. m > 10). The set of diagnostic parameters is determined by means of:

115
 a) diagnostic parameters value correlation with the machine state (with operation and
maintenance time , r j = r(S, y j ), (r j = r((4, yj ));
b) method of amount of diagnostic parameters information about the machine state h j .
In order to choose a set of diagnostic parameters the value of weights is used [3]:
a) standardized calculation weights w 1j [2]:
wj
w 1j = m , (11)

j 1
¦ jw

1
wj = , dj = (1  r j* ) 2  (1  h *j ) 2 , (12)
dj
rj hj
r j* , h *j , (13)
max r j max h j
max(w1j) was accepted as a criterion for selection of a diagnostic parameter (diagnostic
parameters) and the choice of diagnostic parameters was performed according to this
criterion.

3.Description and optimization possibilities of OPTIMUS

OPTIMUS is a one-environment tool which is used to perform an automatic visualization

of the problem and provides the possibility of critical approach to the project dynamics.
OPTIMUS Noesus Solutions software can find an optimal point in the space of the created
simulation design with the use of a combination of gradient methods, through a fast analysis
– or genetic algorithms. In this way it is possible to automatize and control the whole process,
eliminating errors which are inseparably connected with the design process.
OPTIMUS refers to different optimization methods, both the old and the newest ones.
Then, it is possible to work faster and more efficiently without the necessity of restarting
work with new tools and data. Software provides simple solutions to complex problems.
Optimus differs from the available optimization tools as it is open to collaboration with any
product of CAE software [4].
Operation of OPIMUS optimization program involves simulation with the use of data
which put into the operating space in the form of an input file and obtaining output results
through an analysis of codes by means of data analysis. The software enables observation and
modification of each element starting with input data, throughout each indirect code, analyses,
conditions up to the output data, that is, the intended final effect. It is possible when a user
indicates a random graphic point in an interactive graphic grid of OPTIMUS operating space.
When the boundaries of parametric changes are defined, Optimus automatically generates,
analyzes examines and follows alternative solutions – it starts a series of ‘experiments’ in a
‘virtual test laboratory’ through a systematic assessment of answers for each virtual object.

Distinctive features of OPTIMUS software [4]:

ƒ possibility of using CAE codes of all engineering disciplines,
ƒ creation of ‘virtual experiments’ on the basis of DOE techniques,
ƒ methods: Taguchiego aries, Placketta - Burmana, Box- Behnkena, Taylora
multinominal,
ƒ vast Modeling of the response surface,
ƒ powerful post-processing through modeling of the response surface,
ƒ stochastic interpolations (kriging, RBF),
ƒ optimization of numerical procedures,

116
 ƒ self-adaptive and differentia evolution of genetic algorithms,
ƒ dynamic presentation of the design space,
ƒ assessment of reliability and optimization,
ƒ possiblity of using parallel computer networks.

Fig.2. starting window of Optimus software [own research]

The software provides the possibility of creating simulation designs. While doing this it is
necessary to indicate information that will flow between different processed objects. The
elements to be accepted include:
x three project inputs,
x two simulation programs,
x one input file,
x two output files,
x four project outputs,
x three output vectors.

More significant stages in OPTIMUS software include:

x creation of connections between appropriate groups (elements):
x determination of characteristics of an input file for simulation design
x introduction of appropriate commands necessary to perform a simulation,
x determination of a design constraints
x performance o calculations using methods available in the program (fig.5),
x visualization of results through creation of charts (fig.6).
x visualization of results by creation of diagrams (fig.6).

117
 Fig.3. Specification of first parameters [own source]

Input data introduced to the program can be appropriately grouped.

Fig.4. Grouping of input data [own source]

Fig.5. Graphic presentation of correlation method] own source

118
 Fig.6. 3D chart of data [own source]
Advantages of OPTIMUS software application:
x possibility to perform simulation in a unique interface,
x use of many simulation tools in one environment,
x fast configuration of a model in a working space,
x direct interface to a world class software CAD/CAE,
x analysis and database parametrization stencils,
x possibility to perform simultaneous parallel calculations on an infinite number of
processors,
x performance of a series of simulations inside a non-homogenous simulation
environment,
x reduction of calculation time in order to obtain maximal efficiency
x automatic simulation using many processors without intervention of a user.

4. Conclusions

A system for evaluation of a machine state should meet requirements such as:
a) functions to be performed (determination of the next date of servicing), providing
genesis of diagnostic parameters values ( determination of the cause of a machine
inability to perform its functions in the time of its operation)
b) properties: structural, functional, reliability and safety, economic and others.)
Multiple criteria optimization is the most genuine and natural concept of decision making,
hence it seems to be necessary to use its tools including information technologies.
Optimization software OPTIMUS allows to perform different types of optimizations in a
very short time. This program is the next step forward in the field of the most complex tools
to be used for the process integration and optimization of design, with operation dedicated
for applications in Qualitative Techniques.
The main advantages of the software:
x systematic approach to optimization of the design process and simulation definition as
well as work flow automation,
x formalization of the design assessment process,
x improvement of a product efficiency through automation of assessment of alternative
solutions of the design,
x increasing work efficiency,

119
 x response to the real engineering challenge, which design is the best to meet the
specified goals.

References

[1]Wilczarska J., Genezowanie stanu technicznego w procesie eksploatacji maszyn. Rozprawa

doktorska. Bydgoszcz, 2008.
>@ĩyáWRZVNL%Podstawy diagnostyki maszyn. Wyd. uczelniane UTP, Bydgoszcz, 2012.
>@ĩyáWRZVNL%7\OLFNL+Elementy diagnostyki technicznej maszyn, Wydawnictwo PWSZ,
3LáD
[4] http://noesissolutions.com.

120
 MICROCOMPASS FOR SAFE SEA NAVIGATION
0DUFLQĩ\F]NRZVNL
*GDĔVN8QLYHUVLW\RI7HFKQRORJ\
)DFXOW\RI2FHDQ(QJLQHHULQJDQG6KLS7HFKQRORJ\
'HSDUWPHQWRI0DULQH0HFKDWURQLFV
WHO  -20-
H-PDLOPDU]\F]N#SJJGDSO

Abstract
7KLV SDSHU SUHVHQWV WKH UHVXOWV RI GHVLJQ PLFURFRPSDVV IRU VDIH VHD QDYLJDWLRQ ZKHUH QDYLJDWLRQ HOHFWURQLF
SURJUDPPLQJ NQRZOHGJH ZDV MRLQHG WR PDNH PRUH VDIH VHD QDYLJDWLRQ ,Q ILUVW SDUW RI DUWLFOH SUHVHQWV
QDYLJDWLRQ WRROV ZKLFK XVLQJ LQ SUDFWLVH $UWLFOH XQGHUOLQHV WKDW PRVWO\ VDWHOOLWH V\VWHP LV XVHG WR PDNH
QDYLJDWLRQ EXW UHFRPPHQGV WUDGLWLRQDO QDYLJDWLRQ PHWKRG OLNH DOWHUQDWLYH WR GHWHUPLQDWH SRVLWLRQ LQ FDVH RI
EUHDNGRZQRUDQ\DFFLGHQWZKLFKUHGXFHUHOLDELOLW\RIVDWHOOLWHV\VWHP
1H[WSDUWRIDUWLFOHLQWURGXFHVGHFOLQDWLRQDQGGHYLDWLRQFRPSDVVNQRZOHGJH$UWLFOHXQGHUOLQHVZKDWSUREOHP
VKRXOGEHFRQVLGHUHGGXULQJGHVLJQPLFURFRPSDVVIRUVDIHVHDQDYLJDWLRQ$UWLFOHSUHVHQWVSURMHFWRQGLDJUDP
RI LQWHUPRGXOH FRPPXQLFDWLRQ GHVFULEHV PDLQ IXQFWLRQV DQG UHTXLUHPHQWV DQG PDLQ DLP ZKLFK LV DOWHUQDWLYH
WRROWRD*36V\VWHP,QDUWLFOHLQQH[WSDUWSUHVHQWVWUDQVIRUPWHFKQLTXHRIUDZGDWDIRUPPDJQHWRPHWHUDQG
QH[WVWHSZKLFKLVFDOLEUDWLRQSURFHVV5HVXOWRIPLFURFRPSDVVWHVWLVFRPSDULVRQZLWK FRQYHQWLRQDOPDJQHWLF
FRPSDVVZKLFKLVSUHVHQWHGLQDUWLFOH/DVWSDUWRIDUWLFOHLQWURGXFHWRH[SDQVLRQRSSRUWXQLWLHV

KeywordsQDYLJDWLRQHOHFWURQLFFRPSDVVSURJUDPPLQJPDJQHWRPHWHU

,QWURGXFWLRQ
The main purpose of navigation is to keep to a voyage schedule. The most efficient
method for meeting this goal is to navigate along a route specified on a map. Such a route
comprises multiple waypoints whose positions are marked on the map, and which
systematically bring the ship closer to its destination. A straight line is determined between
adjacent waypoints that is the voyage direction. Figure 1 below shows a route with waypoints
(WP).
A person in charge of navigation – the navigator – must not only posses a map, but also
tools enabling the determination of the ship's position and direction. This rule has been
applied for hundreds of years and the method for efficient navigation has not changed yet. The
most common method for determining a ship's position is to read data from the GPS. After
two positions are read, direction is also determined. When all the waypoints of the planned
route are input into the GPS receiver memory, we are dealing with semi-automatic navigation.
The receiver informs the navigator of the approaching waypoints and suggests a change of
course. It must be noted, however, that the American military satellite system, GPS, is not
autonomous and as a result of political decisions it may be switched off or access to it may be
restricted. The latter took place during military actions in the Persian Gulf in the nineties,
when the GPS signal in this area was limited. That is why a magnetic compass and a map are
required for sea navigation and each ship approved for sea navigation by the international

121
IMO regulations shall be equipped with them.


)LJ15RXWHRQDQDYLJDWLRQPDS
A magnetic compass is a basic navigation tool whose principle of operation has not
changed for hundreds of years. This principle involves the determination of direction on the
basis of analysing the impact of the Earth's magnetic field. The first mention of a magnetic
needle used for navigation, floating in a container of water, was found in a Chinese book –
Zhu Yu Pingzhou Ke Tan from 1117. In our part of the world the first mention of a magnetic
needle and its use in navigation was found in De naturis rerum by Alexander Neckam, wrote
around 1190 in Paris. A device measuring the Earth's magnetic field called the magnetometer
was invented in 1833 by a German mathematician and physicist, Carl Friedrich Gauss. The
determination of direction using the magnetic field is not perfect, however, in view of its
independence and wide availability to navigators it is treated as equal to the GPS satellite
system. Furthermore, a navigator performing continuous navigation with the map is able to
approximately locate the ship's position without the GPS. A map and a magnetic compass are
sufficient for safe open-sea navigation. In inshore navigation, position and direction are
controlled using visual and radar observations on the basis of an up-to-date map of the
navigation area.
This article presents a concept of a design and construction of a microcompass for the
purposes of navigation safety. The advantage of the compass is its simplicity and low cost of
construction and operation.

'LUHFWLRQPHDVXUHPHQW
Direction measurement is the deviation of the compass needle from a given reference
surface at a given angle (Figure 2). Which instrument or method we will be using for direction
measurement depends on the name and the value of direction. There is a true, magnetic,
compass and gyro direction, each assigned a different value.
The reference plane for a true direction is the plane of the geographic meridian of the
observer that determines North and South. Therefore, all geographic meridians on the Earth
determine the direction N-S. This method assumes that Earth is a sphere and the axis crossing
the centre of the Earth determines geographic poles P n and P S . In the case of a magnetic
direction, the great circle crossing the magnetic north pole and the magnetic south pole is the
reference plane. The Earth is a giant, natural magnet. Geomagnetism is directly related with
the structure of the Earth and its rotations. The Earth magnetic field is caused by vortical
electric currents flowing in its core. This field may be compared to a magnetic dipole field
inside the Earth.

122
 )LJ'LUHFWLRQPHDVXUHPHQW
The extended axis of a symbolic dipole (substitute magnet) crosses the surface of the
Earth, giving rise to magnetic poles: MN – north and MS – south. The axis connecting both
poles is inclined to the Earth rotation axis by approximately 11.5º (Figure 2). The difference
between true direction TD, and magnetic direction MD is called declination, TD = MD + d.
A compass direction is different from a magnetic direction by the deviation that is
related with ferromagnetic elements of which a ship is made and the direct vicinity of the
external compass. Deviation on a ship is not very dynamic, and does not require frequent
corrections during the compass operation. In the proposed mobile microcompass such a
deviation correction would change depending on the changing level of the battery supplying
the microcompass and its current location. In this case the formula accounting for deviation is
0' &'įZKHUHįLVGHYLDWLRQ and CD is compass direction.

)LJ*HRPDJQHWLVP>@
A gyro direction is determined by a device tracking the Earth's rotations. The process of
determining the reference plane, i.e. the true meridian, is relatively time consuming and
sensitive to additional external interferences, as well as burdened with a correction called tgc
(total gyrocompass correction). The dependence of true direction TD on GD is GD=TD+ tgc.

123
Table1 summarizes the methods for determining directions in sea navigation.
7DE0HWKRGVIRUGHWHUPLQLQJGLUHFWLRQVLQQDYLJDWLRQ
'LUHFWLRQ 'HVLJQDWLRQ &RUUHFWLRQ &DOFXODWLRQV $SSOLFDWLRQDQGGHVFULSWLRQ
true TD None TD Theoretical, assumes that the Earth is a
sphere. The primary direction in
navigation. Navigation maps and the
GPS system both specify this direction.
magnetic MD Declination TD=MD + d Magnetic compass, the main tool for
determining directions, using
geomagnetism, correction results from
varying distribution of geomagnetism
impact
compass CD Deviation 7' &'Gį Magnetic compass, accounts for
corrections resulting from ferromagnetic
interferences of the observer's
surroundings
gyro GD Gyrocompass GD=TD+tgc Gyrocompass. Device using the Earth's
correction rotations, correction results from the
device imperfections

0LFURFRPSDVVGHVLJQ
The proposed digital microcompass operates on the basis of the Earth's magnetic field
measurement, however, it also considers necessary corrections and presents a true direction.
Declination that arises from varying distribution of the geomagnetism impact, as well as
deviation that arises from ferromagnetic interferences in the observer's surroundings were
both considered while programming the compass. In the long term of the compass operation,
deviation and declination corrections would be input from various geographical positions or
the current battery level in the form of an online database, and as a correction input by
external devices. The navigator may find the information on declination in navigation
documentation (a navigation map, etc.), and information on deviation – in the ship's
documentation (a deviation table for a given ship). External devices cooperating with the
microcompass are gyrocompass, or the navigator. It may also cooperate with the GPS,
however, in such a case it is no longer an autonomous device. Figure 4 shows a logical design
of the microcompass, with relevant modules performing certain functions and certain tasks.
The primary element of the microcompass is a microcontroller supplied from a proper voltage
source, that combines all other modules. Raw data will be taken from the magnetometer. The
data is useless without proper software and calibration. That is why raw data are transferred
into the programming module. After data are transformed and calibrated, they are sent to the
microcontroller as a true direction. While operating the microcompass it is possible to input
correction updates from external devices. The desired result will be displayed. Its format
depends on particular requirements, it may be programmed and displayed in a complete
mode, point mode or other.

124
 )LJ'LDJUDPRILQWHUPRGXOHFRPPXQLFDWLRQLQWKHPLFURFRPSDVV
The design assumptions for the microcompass are given in table 2:
7DE'HVLJQDVVXPSWLRQV
Aim:
To create a tool for the navigator that would constitute an alternative to a GPS system as regards direction
measurement, in view of a possible lack of the GPS signal
Functions:
- to make a conventional magnetic compass more attractive to the user through mobility, semi-automatic
updates, clear display
- equipped with the possibility to connect with other tools in view of correction updates
- equipped with the possibility to add other modules that would perform other functions
Requirements:
Energy savings – the microcompass is built in such a manner that it consumes little electric energy while
performing the basic functions of a compass
Low production cost – the microcompass is built using widely available and inexpensive electronic and
navigation components
The centre of the designed system, through which all data pass, is the microprocessor.
Its task is to control the entire process: collection of information from the magnetometer,
proper data transformation, calibration, correction update and display of the compass
indication in a clear manner. For the purposes of the current design, an ATMEL AVR
processor by ATMEL was selected: megaAVR – Atmega 8. Its characteristics include the
possibility of supply with voltage between 4.5 and 5.5 V, 28 pins, speed of up to 16 MHz,
power consumption at 4 MHz – 3.6 mA. Microsoft Visual Studio based ATMEL STUDIO
AVR software is available in Atmel series that enables the development of a programme, its
compilation, as well as initiation of the processor operation. Processors may be programmed
in several languages, including C++, in which the programme for this particular
microcompass was written.
Another designed element was the display. The initial design assumed the use of a
relevant number of LEDs for displaying information. This, however, posed difficulties in
constructing the system. In order to increase simplicity, as well as the accuracy of data
presentation, a display was used instead. The two most common solutions are alphanumerical
or a graphical display. Ultimately, an alphanumerical, single-line display without backlight
was chosen, displaying up to 16 characters in one line. The main reason behind this was
power consumption and the fact that the presentation of elements other than characters was
not necessary. What was also important was the fact that displays based on the HD44780
controller are well documented and many libraries are available for their operation using the
ATMEL AVR processor.
A proper programmator was necessary for the construction of the microprocessor. The
most common programmator for AVR processors was used – USBasp. The programmator is
connected to a USB interface and it provides for not only the programming of the processor,

125
but also the supply of the system with 500 mA power fed from the computer USB port. It is
not equipped with a debugging function, but considering the simplicity of the programme, it
was decided that only functional testing will be used. It is worth noting that the programmator
is able to programme many AVT processors, including ATtiny2313 and ATmega8.
A device reading raw data – LSM303DLHC by Pololu – was used as a magnetometer. It
has complete documentation and manufacturer's support as regards the system operation. Its
characteristics are: supply voltage of 2.5 – 5.5 V, power consumption of 10 mA, output data
I2C – magnetometer and accelerometer, measuring range for accelerometer: ±2G, ±4G, ±8G,
±11 G, for magnetometer: ±1.3, ±1.9, ±2.5, ±4.0, ±4.7, ±5.6, ±8.1 gauss, dimensions – 13 x 23
x 3 mm, pins arrangement: VDD – output 3.3 V, VIN – power supply, GND – ground, SCL –
clock line I2C, SDA – data line I2C, DRD-Y – magnetometer readiness indication, INT1,
INT2 – configurable interruptions.
The system power supply should enable both long term operation of the device and its
easy replacement and regeneration. The required voltage of +-5 V had to be as easily
achievable as possible, i.e. preferably without using additional voltage increase or decrease
systems that are always associated with costs due to lost energy and often the generation of
additional heat. Therefore, battery supply was selected (AAA battery). The advantage of the
battery is that it is lightweight, small, low-cost and has the capacity of 900-1155 mAh. This
selection is in accordance with the design requirements (Table 2).
Having selected the components, all that is left is to develop a programme in Atmel
Studio using available libraries that facilitate communication between the modules (Figure 5).
Table 4 shows types of intermodule communication, libraries used and functions responsible
for communication, as well as tasks performed and the course of intermodule communciation
process:
7DE,QWHUPRGXOHFRPPXQLFDWLRQ
Type Tasks Libraries used Performed Commends
functions
Magnetometer – - making raw data Library of 1)i2c_init()- Initiates I2C i2c bus: Control
Microprocessor from the I2C bus interface process – i2c
magnetometer #include 3)i2c_stop()- Stops initiation +
available to the "i2cmaster.h" communication on i2c switching of the
microprocessor Library of 2)i2c_write(compass compass into
through i2c control: address)- Writes into writing mode +
#include compass register selection of a
4)i2c_start_wait(compass register
<delay.h>
address+1)- Switches the responsible for
compass into writing mode data
and awaits writing, when + transmission +
0 is read-out input into the
5)i2c_start(compass address register of the
+1) - Switches the compass value that
into writing mode, when + 0 stimulates the
is read-out compass to send
data + receipt of
6) i2c_rep_start(compass
data in the form
address + 1)- Again,
switches the compass into of 8 bits to the
writing mode, when + 0 is board and
awaiting other
read-out
data (repeated
7)i2c_readAck(void )-
three times) +
Collects data from the
receipt of 8 bits
compass and demands other
and reception
data
completion. Data
8)i2c_readNak (void)- are included in
Collects data from the the character
compass and ends reception table.
9)delay_ms(5000) – delays

126
 transmission
Microprocessor - conversion of Auxiliary 1)int przelicz(int x, int y) - A link in the
+ Programming the character libraries: Converts the collected form of a
table into #include magnetometer data into programmator is
magnetometer <math.h> degrees necessary for
vector variables #include 2)char * rumby(int deg)- this type of
- calculation of <stdio.h> Converts degrees into points communication
direction using #include 3)void – for this design
vector variables <string.h> odchylenie_gora_dol(int k) - Usbasp was
- conversion of Displays information on used. The
degrees into incorrect deviation of the microcompass
points compass may be
- calculation of programmed
compass when the new
deviation programme code
- data is entered into
transmission onto the
the display microcontroller
- compass using the link.
calibration
Microprocessor Corrections Libraries of Depends on the capacity of One PORTD
– External updates external devices external devices input in the
External control microcontroller
is intended for
additional
signals from
external devices
Microprocessor - Display of a Display control 1)lcd_init(LCD_DISP_ON)- Total process
display message received library Initiates the display control by the
from the #include "lcd.h" 2)lcd_clrscr()- Display clear microprocessor
microprocessor command
3)lcd_puts("char")-
Transmission of text onto
the display

)XQFWLRQDOWHVWV
Thanks to the use of I2C interface and a proper library for its operation, data flow can
always be monitored on the display. In order to evaluate the correctness of operation, results
generated by the microcompass without a correction, with reference to the GPS, were
analysed. Such raw data from the magnetometer are shown in Diagram 1.

'LDJ1'DWDIURPWKH PDJQHWRPHWHUZLWKRXWWUDQVIRUPDWLRQ
Apart from the monitoring of raw data, the manufacturer also provided information on
the necessary data transformation. The magnetometer provides three 16-bit variables. X and Y

127
are used for determining the direction, while Z is the information on the device deflection.
Data in the form as in Diagram 1 are raw and useless – in order to be able to determine the
direction using those data, they must be subjected to a relevant transformation. In accordance
with documentation, LSM303DLHC requires additional calFXODWLRQV RI ǻ;  PLQ ;  
PD[ ; ǻ<  PLQ < PD[ <  ZKHUHPLQ ;  -290, max(X1)= 220, ǻ; -
35, min (Y1) = -460, PD[ <  ǻ< -170. The transformation is shown in Table 5. The
transformation of data collected from the magnetometer is shown in Diagram2.
7DE7UDQVIRUPDWLRQLQOLQHZLWKWKHGRFXPHQWDWLRQRIPDJQHWRPHWHU/60'/+&
Direction
X1 Y1 X=X1- ǻ; Y= Y1- ǻ<
Directions In accordance with the formula
before trans. before trans. after trans. after trans.
from Table 7
0 81 86 116 256 65.6
30 176 -12 211 158 36.8
60 220 -150 255 20 4.5
90 203 -294 238 -124 332.5
120 126 -409 161 -239 304.0
150 -4 -460 31 -290 276.1
180 -130 -436 -95 -266 250.3
210 -244 -330 -209 -160 217.4
240 -290 -190 -255 -20 184.5
270 -260 -40 -225 130 150.0
300 -177 70 -142 240 120.6
330 -50 120 -15 290 93.0

7DE7UDQVIRUPDWLRQRIGDWDFROOHFWHGIURPWKHPDJQHWRPHWHU
L.p. Data X Data Y Direction
1 X >0 Y >=0 arctan(Y/X)
2 X<0 180 + arctan(Y/X)
3 X>0 Y<0 360 + arctan (Y/X)

The results of the transformation are given in Diagram 2 below:

'LDJ'DWDREWDLQHGIURPWKHPDJQHWRPHWHUIROORZLQJWUDQVIRUPDWLRQ
The red and blue lines are two sinusoids used as arguments for the mathematical
function – tangens. The green line is the determined direction that does not coincide with 0
due to the lack of calibration with the true North, however, it may be observed that it is
displaced by a constant. The green line should start with 360, then drop to 0, and rapidly

128
increase to 360. Due to abrupt measurements, the line looks broken, while in fact it is a
continuous function. Calibration was performed during programming, so that 0 coincides with
the true North. The declination and deviation correction is 65.6.
In order to evaluate the microcompass operation quality, a comparative method was used,
where the microcompass operation was compared with a conventional magnetic compass.
Table 9 below contains measurements from four different locations.
7DE'LIIHUHQFHVLQPHDVXUHPHQWVXVLQJDFRQYHQWLRQDOPDJQHWLFFRPSDVV&&DQGWKHFRPSDVVIURPWKH
SURMHFW3&
CC PC PC PC PC Average Difference
Measurement 1 Measurement Measurement Measurement 4 Average KP
*GDĔVN 2 3 *GDĔVN
University of Beach, %HDFK*GDĔVN University of
Technology *GDĔVN %U]HĨQR Technology
Jelitkowo
[°] [°] [°] [°] [°] [°] [°]
0 355 356 355 7 358.25 1.75
90 94 88 96 96 93.5 -3.5
180 210 180 187 187 191 -11
270 256 269 273 268 266.5 3.5
Average 4.9
deviation

As results from table 9, deviations remain within permissible limits. The result of the
comparison with a conventional compass is +- 5 °. The microcompass was programmed
considering corrections, i.e. it indicates a direction close to the true direction. A conventional
magnetic compass indicates a compass direction. Declination in the area where the tests were
SHUIRUPHG LV DSSUR[LPDWHO\    ƒ DUHDV RI EHDFKHV LQ *GDĔVN %U]HĨQR DQG *GDĔVN
-HOLWNRZRDVZHOODVWKH*GDĔVN8QLYHUVLW\RI7HFKQRORJ\ 'HYLDWLons may be omitted due
to the lack of ferromagnetic elements in a needle compass). Therefore, measurements may be
deemed satisfactory.

([SDQVLRQRSSRUWXQLWLHV
The microcompass was proposed in accordance with the design assumptions given in
Table 2 (aim, delivered functionality and requirements). Further development is still possible.
It is possible to introduce an automatic declination and deviation correction using the
indications of a GPS receiver or another external device (such as a gyrocompass) or a
navigator. This would result in a more precise measurement that would be comparable to a
GPS receiver. Another idea is to create a database of declination corrections. Up-to-date
declination corrections would be saved into the database with the record of the current
position and time from the GPS. After approaching the position in which the declination
correction was saved, the microcompass would be able to download this correction. An
external device may be used for this functionality, so that the performance and economy of
the microcompass operation are not compromised. This function may only be activated when
demanded by the microcompass, as a "check correction from the declination corrections
database" function. Another idea is to use a marine magnetic compass mechanism that reacts
in case of ship listing and enables operation even in difficult sea conditions. The
microcompass may also calibrate its horizontal and vertical deflection on the basis of a read-
out from the accelerometer, however, this requires additional calculations and the replacement
of the current microcontroller with a different one with larger memory, which in turn
increases power consumption and data reading interferences. Currently, the microcompass
provides information on incorrect placement and suggests a change with + and -. Another
solution is to design a common interface for the GPS system and the microcompass that

129
would continuously provide information on differences in direction reading. Such a solution
would reduce power consumption by the microcompass (it would not be equipped with a
display), as well as provide the navigator with additional information. Another development
step would be to use Wi-Fi technology that would enable the installation of a mobile repeater
that repeats read-out from the microcompass, for example in a hand watch or a mobile phone.
As a result, continuous direction control by the navigator would be possible even after his
watch is over.
The design is based on a modular structure that makes it possible to add new elements,
expanding the system with new functions and tasks. For example, an information module that
could provide warnings on the change of course, due to an approaching danger or due to a
deviation from the route planned on the map. This in turn would provide for the centralization
of the navigator's information and increase in navigation safety.

REFERENCES
1. 0LURVáDZ -XUG]LĔVNL ³3RGVWDZ\ QDZLJDFML PRUVNLHM´, Fundacja Rozwoju WSM w
Gdyni, 2002,Gdynia
2. )UDQFLV]HN :UyEHO ³9DGHPHFXP RILFHUD ZDFKWRZHJR´, Trademar w Gdyni, 2013,
Gdynia
3. 3DZHá %RUNRZVNL ³0LNURNRQWUROHU\ 3,& Z SUDNW\F]Q\FK ]DVWRVRZDQLDFK´, Helion,
2009, Gliwice
4. &H]DU\6SHFKW³6\VWHP*36´, Wydawnictwo Bernardinum w Pelplinie, 2007, Pelplin
5. -DFHN -DQXV]HZVNL ³6\VWHP *36 L LQQH V\VWHP\ VDWHOLWDUQH Z QDZLJDFML PRUVNLHM´,
Akademia Morska w Gdyni, 2008
6. -DFREVRQ&OLII³0DSD.RPSDV*36´, Eremis, 2007
7. Tomasz )UDQFX] ³-Ċ]\N & GOD PLNURNRQWUROHUyZ $95 2G SRGVWDZ GR
]DDZDQVRZDQ\FKDSOLNDFML´, Helion, 2010, Gliwice
8. .DUGDĞ0LURVáDZ³0LNURNRQWUROHU\$95MĊ]\N&SRGVWDZ\SURJUDPRZDQLD´, Atnel
2013, Szczecin
9. .DUGDĞ 0LURVáDZ ³-Ċ]\N & 3DVMD SURJUDPRZDQLD PLNURNRQWUROHUyZ -ELWRZ\FK´,
Atnel 2012, Szczecin
10. :ROIJDQJ7UDPSHUW³$95-5,6&0LNURFRQWUROOHU´, Franzis,2002,Germany
11. http://magnetic-declination.com/
12. http://www.gps.gov/applications/marine/
13. http://inventors.about.com/od/cstartinventions/a/Compass.htm
14. http://www.compassadjustment.com/
15. http://www.visualmicro.com/page/Arduino-for-Atmel-Studio.aspx
16. http://www.physicalgeography.net/fundamentals/6h.html
17. http://www.microchip.com/ParamChartSearch/chart.aspx?branchID=1009&mid=10&l
ang=en&pageId=74
18. http://www.microchip.com/ParamChartSearch/Chart.aspx?branchID=211
19. http://www.atmel.com/images/atmel-2486-8-bit-avr-microcontroller-
atmega8_l_datasheet.pdf

130