Final Book

Alexandria University
Faculty of Engineering
Mechanical Department
Natural Gas Network Optimization

Power generation & P2G applications
Under The supervision of:

Prof. Dr. Hassan Warda
Eng. Sherif Haddara
Prof.Dr. Essam Wahba
This page is intentionally left blank
About the team

The strength of Engineers for development team stems from the combined
expertise in both academic and technical areas.
The team has worked together energetically since 2015 resulting in strong
chemistry between the members.
The complimentary strength of the team and its high efficiency is evident by the
level of success Engineers for development team has had since inception.
Everyone was aware of the importance of the role of each individual in the
group.
The trust between members is a key of the team cohesion, everyone
appreciates the others and shows respect and support.
Following is a list of the team members with a brief description about their work
and contacts:
Ahmed Bahaa El-Deen Negmy
+201120840171
ahmedbahaa3519@gmail.com
Natural Gas network

study.
Optimization study.
Optimization
implementation
(Coding).
Synergi Script (Coding).
Documentation
Team Leader
Ahmed Tarek Mohamed Ahmed
+201200707769
ad_tk_2008@yahoo.com
Solar systems.
Desalination systems.
Electrolysis.
Hydrogen blending.
Techno-economic
analysis.
Eslam Mahmoud Mohamed Ibrahim

+201110169144
Esl939@yahoo.com
Samar Samy Anwar Zakarya Matter

+201200477272
engsamarsamy@yahoo.com
Hydraulic modeling
using Synergi Gas.
Regulator station.
Unsteady state
modeling.
Solar systems.
Desalination systems.
Electrolysis.
Hydrogen blending.
Techno-economic
analysis.
Shaza Abd El-Hady Abd El-Kader

Power station
+201111967235
shaza_hady1993@yahoo.com
Mohamed Abd Elnaby Shams Eldyn

+201019942248
+201150507898
Engmohamedabnaby@gmail.com
Hydraulic modeling
using Synergi Gas.
Compressor station
modeling using
Synergi Gas
Mohamed Mostafa Ali Mohamed

Compressor station.
Compressor sizing.
+201096722029
mohamedmostafa9310@gmail.com
Mohamed Moustafa Mahmod

+201284598934
mouhamed.moustafa458@gmail.com
Mohamed Mamdouh Mohamed

+201116638242
medoamer36@gmail.com
Mohamed Youssef Omar

+201005754868
mohamedyoussefomar@gmail.com
Natural gas network

Schematization using
ArcGIS.
Network components
study.
Natural Gas network

study.
Optimization study.
Optimization
implementation
(Coding).
Synergi Script
(Coding).
Power Stations
Gas Turbine Power
plant operation
Thermodynamic
analysis of GT
Coding of GT
thermodynamics
Documentation
Team (HR)
Acknowledgment
First, Thank to Allah, Most Merciful and Most Compassionate. We take this
opportunity to express our deepest and sincere gratitude to our supervisors
Prof. Dr. Hassan Warda, Eng. Sherif Hadara and Prof.Dr. Essam Wahba for
their insightful advice, motivating suggestions, invaluable guidance, help and
support in successful completion of this project and also for their constant
encouragement and advice throughout this work. We feel greatly honored to
have been working under their supervision.
Our deepest thanks are further extended to Eng/ ahmed abd elmoety from
natgas, Eng/ hesham Khalil from Gasco , Gasser Elfaham from Gasco, Dr.Ahmed
Hegazy, Dr.Osama El-Masry, Eng/ Ayman El-Morshedy from Sidi Krir and abd ElMoneam.
We would like to convey our thanks to the teaching staff of the Mechanical
Engineering Department, for their invaluable help and support throughout the
period of Bachelors Degree.
Finally, yet more importantly, we would like to express our deep appreciation to
our families for their prompt help and continuous care which has a great
influence throughout our lives.
Preface
Natural gas is one of the most widely used forms of energy today, a cleanerburning fuel than either gasoline or diesel, used by industrial, residential,
electricity generation, commercial, and transportation sectors in varying
amounts.
Egypts proved reserves of natural gas are the third largest in Africa, after Nigeria
and Algeria, despite that Egypt faces a shortage in Natural gas.
in recent years. For power generation, old power stations which used to run on
oil crude has been converted to natural gas increasingly, this produces less
carbon dioxide than crude oil. Its also more economic than crude oil.
Due to increasing demand of natural gas markets it became a must to optimize
natural gas network operation and to find new energy sources to balance the
network.
Our project target is to optimize the Natural Gas National Grid (NGNG)
Operation. And Also to find new power generation application power to gas.
In this book all the algorithms procedures and considerations will be discussed,
also the final results and recommendations will be stated.
The screenshots from a licensed software package which was used to perform
the Modelling and results will be presented.
Table of contents
Abstract ........................................................................................ 18
Natural Gas Network Components .................................................... 20
1.1 Introduction ............................................................................. 21
1.2Components of network .............................................................. 24
1.2.1 Transmission pipelines ......................................................... 24
1.2.2 Compressor Stations: .......................................................... 50
1.2.3 Regulation and Gas Metering stations..................................... 50
1.2.4 City Gate Stations ............................................................... 51
1.2.5 Valves ................................................................................ 52
1.2.6 Flow Measurements ............................................................. 56
1.2.7 Remote Control System & Telecommunications ....................... 59
Power to Gas ................................................................................. 61
2.1 Introduction: ............................................................................ 62
2.1.1 What is the problem of energy in Egypt? ................................ 62
2.1.2 energy sources available in Egypt:......................................... 62
2.1.3 Power to Gas: ..................................................................... 62
2.2 What is PtG? ............................................................................ 63
2.2.1 Definition: .......................................................................... 63
2.2.2 Advantages of P2G: ............................................................. 63
2.2.3 Components of P2G plant ..................................................... 64
2.3 Solar systems ........................................................................... 64
2.3.1 Solar system in Egypt .......................................................... 64
2.3.2 History ............................................................................... 65
2.3.3 Places of solar plants in Egypt ............................................... 65
2.3.4 Types of solar system .......................................................... 67
2.4 Desalination process .................................................................. 80
2.4.1 Water Classification: ............................................................ 81
2.4.2 Desalination techniques: ...................................................... 81
2.5 Electrolysis ............................................................................... 94
2.5.1 Theory of water electrolysis operation: ................................... 94
2.5.2 Factors affecting on electrical efficiency: ................................ 95
2.5.3 Alkaline electrolysis ............................................................. 96
10
2.5.4 Polymer electrolyte membrane (PEM) electrolysis .................. 100

2.5.5 Solid Oxide Electrolysis Cell (SOEC) ..................................... 103
2.5.6 Electrolysis Calculations: .................................................... 105
2.5.7 Comparison between the different types of electrolysis: ......... 107
2.6 Blending or Injection of Hydrogen into the Natural Gas grid ......... 109
2.6.1 General order of the restrictions on Hydrogen blends: ............ 109
2.6.2 Leakage of the gas: ........................................................... 111
2.6.3 Benefits of Hydrogen blending: ........................................... 112
2.7 Implementation of Power to Gas (P2G) around the world: ............ 112
2.7.1 Germany experience: ......................................................... 112
2.7.2 Austria experience:............................................................ 113
2.7.3 Denmark experience: ......................................................... 113
2.8 Techno-Economic Analysis ....................................................... 115
2.8.1 Technical analysis: ............................................................ 115
2.8.2 Technical Comparison between the four alternatives: ............. 121
2.8.3 Economic Analysis: ............................................................ 121
2.8.4 Economic study results: ..................................................... 126
2.8.5 Conclusion:....................................................................... 134
Power Stations ............................................................................. 142
3.1 Abstract ................................................................................. 143
3.2 Introduction to Gas Turbines .................................................... 143
3.3 Applications of Gas Turbine ...................................................... 144
3.3.1 Electric utility central station power generation ..................... 144
3.3.2 Combined heat and power .................................................. 144
3.3.3 Mechanical drive................................................................ 144
3.3.4 Distributed power-only ....................................................... 144
3.4 Classifications of Gas Turbine Power Plants ................................ 145
3.4.1 Classification based on Cycle ............................................... 145
3.4.2 Classification based on Application ....................................... 148
3.4.3 Classification based on Shaft Configuration ........................... 149
3.5 Gas Turbine Design Considerations ........................................... 149
3.6 Part-Load Performance ............................................................ 149
3.7 Thermodynamic Analysis of Gas Turbines ................................... 150
11
3.8 Cycle Improvements ............................................................... 151

3.8.1 Regeneration .................................................................... 151
3.8.2 Intercooling ...................................................................... 151
3.8.3 Reheat ............................................................................. 152
3.8.5 Combined Cycle ................................................................ 152
3.9 Parameters that affect gas turbine performance .......................... 153
3.9.1 Effect of ambient temperature............................................. 153
3.9.2 Effect of compression ratio ................................................. 154
3.9.3 Effect of compressor and turbine efficiencies ......................... 155
3.9.4 Conclusion ........................................................................ 155
3.10 Gas turbine control and auxiliary systems ................................ 156
3.10.1 Governor control ............................................................. 156
3.10.2 Lubrication system ........................................................... 159
3.10.3 Fuel system .................................................................... 163
3.10.4 Starting system ............................................................... 167
3.10.5 Cooling systems .............................................................. 168
3.10.6 Instrumentation and control .............................................. 168
3.11 Protection............................................................................. 169
3.12 Turbine Performance ............................................................. 169
3.13 Gas Turbine Preventive Maintenance........................................ 169
3.14 Startup and Shutdown ........................................................... 172
3.14.1 Preparation for Startup ..................................................... 173
3.14.2 Start Initiation................................................................. 174
3.14.3 Crank and Light off .......................................................... 175
3.14.4 Warm-up ........................................................................ 176
3.14.5 Loading .......................................................................... 176
3.14.6 Normal Shutdown ............................................................ 177
3.14.7 Fast Shutdown ................................................................ 177
3.15 Interface between gas network and power stations ................... 177
3.15.1 Gas Pressure Trend .......................................................... 179
3.15.2 Gas flowrate Trend .......................................................... 179
3.15.3 Load Trend ..................................................................... 180
Compressor Stations ..................................................................... 181
12
4.1 Introduction ........................................................................... 182

4.1.1 Definition ......................................................................... 182
4.1.2 Components ..................................................................... 182
4.2 Compressor ............................................................................ 182
4.2.1 Introduction ...................................................................... 182
4.2.2 General Description of a CENTRIFUGAL Compressor............... 184
4.2.3 Centrifugal Compressors Types ............................................ 188
4.2.4 Basic Theoretical Aspects................................................... 191
4.2.5 Performance of compressor stages ...................................... 195
4.3 Cooling system ....................................................................... 204
4.3.1 Introduction ...................................................................... 204
4.3.2Cooler classification ............................................................ 204
4.3.3 Pre- cooler........................................................................ 204
4.3.4 Inter cooler....................................................................... 204
4.3.5 After cooler....................................................................... 207
4.4 Compressor Prime Mover ......................................................... 207
4.4.1 Introduction ...................................................................... 207
4.4.2 Prime Mover Types ............................................................ 207
4.4.3 Prime Mover Selection ....................................................... 209
4.4.4 Gas turbine as Prime mover ................................................ 209
4.5 Scrubber ................................................................................ 216
4.5.1 Definition ......................................................................... 216
4.5.2 Applications ...................................................................... 217
4.5.3 Selection criteria for gas/liquid separators ............................ 217
4.5.4 Components of Scrubber .................................................... 218
4.5.5 Types of separators used as scrubber .................................. 218
4.5.6 Scrubber design requirements............................................. 219
4.6 Flare System .......................................................................... 220
4.6.1 Introduction ...................................................................... 220
4.6.2 Flare system components ................................................... 221
4.6.3 Operation ......................................................................... 225
4.7 Operation Examples ................................................................ 228
4.7.1 Rashpetco compressor station operation............................... 228
13
4.7.2 WDGC compressor station .................................................. 232

4.7.3 Dahshour compressor station .............................................. 237
Modeling Using ArcGIS .................................................................. 243
5.1 introduction: .......................................................................... 244
5.2 Geographic information system: ............................................... 244
5.2.1 What is GIS? ..................................................................... 244
5.2.2 COMPONENTS OF A GIS: .................................................... 245
5.2.4 GIS application: (specific functionality) ................................ 246
5.3 Coordinate systems: ............................................................... 248
5.3.1 What is a coordinate system?.............................................. 248
5.3.2 Types of coordinate systems: .............................................. 248
5.3.3 Map projection: ................................................................. 250
5.4 Project Design: ....................................................................... 252
5.4.1 Steps of Project Design: ..................................................... 252
5.5 Mapping: ............................................................................... 254
5.6 What is ArcMap? ..................................................................... 255
5.6.1 Typical tasks performed in ArcMap: ..................................... 255
5.6.2 ArcMap documents: ........................................................... 256
5.6.3 Views in ArcMap: ............................................................... 256
5.6.4 Map layers: ...................................................................... 257
5.6.5 Working with basemap as a layer: ....................................... 258
5.7.1 Steps of the project: .......................................................... 258
5.8 Drawing the Natural Gas network using ArcMap: ......................... 259
5.9 Interaction between ArcGIS and SynerGEE: ............................... 262
5.9.1 Exporting map to SynerGEE:............................................... 262
5.9.2 Importing hydraulic data: ................................................... 262
Modeling Using Synergi ................................................................. 265
6.1 introduction. ........................................................................... 266
6.1.1 about synergi ...................................................................... 266
6.1.2 Synergi capabilities .............................................................. 266
6.1.3 Solution techniques .............................................................. 266
6.1.4 Rules for unknowns .............................................................. 267
6.1.5 Target of using Synergi Gas ................................................... 268
14
6.1.6 Synergi modules .................................................................. 268

6.1.7 script .................................................................................. 277
6.2 Building up the model .............................................................. 278
6.2.1 Import from GIS .................................................................. 278
6.2.2 Node Naming ....................................................................... 279
6.2.3 Model Builder Module ............................................................ 281
6.2.4 Inserting Data to the model ................................................... 286
6.3 Model Components .................................................................. 296
6.3.1 Nodes ................................................................................. 296
6.3.2 Model facilities in Synergi ...................................................... 297
6.3.3 Pipes .................................................................................. 298
6.3.4 valves ................................................................................. 299
6.3.5 Regulators and regulating stations ......................................... 300
6.3.6 Compressor Station .............................................................. 304
6.4 Output From Synergi ............................................................... 314
6.4.1 Reports ............................................................................... 314
6.4.2 Charts ................................................................................ 317
6.4.3 steady state analysis ............................................................ 318
6.4.4 Unstady state analysis .......................................................... 343
Optimization ................................................................................ 352
7.1 Introduction .......................................................................... 353
7.2 Corner Stones of any optimization Model ................................... 357
7.2.1 Objectives ........................................................................ 357
7.2.2 Constraints ....................................................................... 362
7.3 Optimization Problem Types ..................................................... 367
7.3.1 Linear Programming .......................................................... 367
7.3.2 Linear Integer Programming ............................................... 374
7.3.3 Quadratic programming ....................................................... 377
7.3.4 Non-Linear Programming .................................................... 378
7.3.5 Dynamic Programming ....................................................... 382
7.4 Modern Methods for Solving Optimization Problems ..................... 386
7.4.1 Introduction to Evolutionary Algorithms ............................... 386
7.4.2 Simulated Annealing (SA) ................................................... 387
15
7.4.3 Ant Colony Optimization (ACO) ........................................... 393

7.4.4 Fuzzy Optimization ............................................................ 396
7.4.5 Neural Optimization ........................................................... 399
7.4.6 Genetic Algorithm Optimization ........................................... 402
7.5 Matlab Optimization Toolbox..................................................... 416
7.5.1 GA toolbox: ...................................................................... 416
7.5.2 GA toolbox components: .................................................... 416
7.5.3 How the genetic algorithm works? ....................................... 417
7.5.4 Population ........................................................................ 417
7.5.5 Crossover ......................................................................... 417
7.5.6 Mutation ........................................................................... 419
7.5.8 Main GA code .................................................................... 421
7.5.9 Running the Genetic Algorithm with the Default Options ......... 422
7.6 The graphical user interface ..................................................... 424
7.6.1 Opening the Genetic Algorithm Tool: .................................... 424
7.6.2 Defining a Problem in the Genetic Algorithm Tool .................. 424
7.6.3 Running the Genetic Algorithm ............................................ 424
7.6.4 Pausing and Stopping the Algorithm .................................... 425
7.6.5 Displaying Plots ................................................................. 426
7.6.6 Reproducing Your Results ................................................... 426
7.6.7 Setting Options for the Genetic Algorithm ............................. 426
7.6.8.1 Exporting Options and Problems ....................................... 427
7.6.9 Genetic Algorithm Options .................................................. 428
7.7 Instructions to write the code ................................................... 433
7.8 Optimization Implementation ................................................... 434
7.8.1 Optimization Techniques: ................................................... 434
7.8.2 Optimization loop between MATLAB and Synergi ................... 435
7.9 Case studies ........................................................................... 437
7.9.1 Travelling salesman problem ............................................... 437
7.9.2 Example ........................................................................... 437
7.9.3 Methods to solve the traveling salesman problem .................. 437
7.9.4 optimum pipe diameter ...................................................... 440
7.9.5 National grid ..................................................................... 442
16
Conclusion ................................................................................... 455

References .................................................................................. 457
Appendix ..................................................................................... 459
Compressors code steady state ................................................... 459
Compressors code unsteady state ................................................ 463
Functions for mutation and crossover ........................................... 465
P2G alternatives codes ............................................................... 467
Power stations code ................................................................... 471
17
Chapter 1 | Natural Gas Network Components
flame stability, gas turbine governors, the

interface between power plants and the
natural gas network and how to improve the
gas turbine performance.
Abstract
Egypt has large potential of natural gas. The
consumption has grown unsteadily in recent
years until it has reached the first largest in
Africa which led to the increase in the gap
between production and consumption. One
of the main demands is power station which
is the major consumer of natural gas in Egypt
as they consume about 57 % of the natural
gas produced in Egypt, and 85 % of the
energy consumption in power stations is
from natural gas. Natural gas shortage and
low network pressure are the major
problems which need to be solved and this
can be achieved by studying the gas network
components and power stations and their
effect on the network and searching for new
solutions to operate the network at high
performance.
In chapter four, compressor stations and

their components will be studied, also
illustration of the operation of the
compressor station with examples of
Western Desert Gas Complex and Dahshur
compressor station will be mentioned.
In order to achieve the goal of our project,
which is the optimization of Egyptian natural
gas network operation, it is inevitably to
build a model of the natural gas network.
So in chapter five, an Important study about
the Geographic Information System, its
components, GIS benefits, GIS applications,
coordinate systems, types of coordinate
systems, map projection, steps of project
design using ArcGIS and mapping using
ArcMap, then we will illustrate how the
Egyptian natural gas network has been
drawn by GIS.
In chapter one, a study of natural gas history,

composition, sources, phase behavior,
benefits from using natural gas and the
processing of natural gas will be discussed.
Also the components of natural gas network
such as transmission pipelines, Compressor
Station, regulation stations, gas metering
station, valves, flow measurements, control
system, remote control system and
telecommunications will be mentioned.
Synergi Gas was used to model the network,

and as a consequence checking the
feasibility of the number of operation
scenarios, which is the output of our
optimization code to find the best operation
scenario. We used the Synergi Gas because
Synergi Gas is the leading natural gas
network modeling software. It is the
commercial package that is used by GASCO
and NATA for modeling the national grid of
natural gas, Synergi Gas models and analyzes
closed conduit networks of pipes, regulators,
valves, compressors, storage fields, and
production wells. Synergi Gas 4.8 software is
provided with five modules Model builder
module, Unsteady state module, Time
varying module, Optimization module and
Transient optimization module.
In chapter two a solution to reduce the

shortage in the national grid is presented.
This solution is based on power to gas
concept, this enables us getting a renewable
source of natural gas is a must instead of
importing from other countries.
In chapter three, an Important study of gas
power plant working principle and
operation, thermodynamic performance
analysis of gas-turbine power-plant, several
parameters that effects on gas turbine
performance, fuel gas system of gas turbine,
18
So in chapter six, Synergi Gas in details is

introduced and each module. To build our
model, ArcGIS software is used to build the
model with the specific pipe lengths and
nodes co-ordinate, interface between
Synergi and ArcGIS and inserting network
facilities in details (valves, regulator stations
and compressor stations). After modeling
the network on Synergi, data validation of
the network from actual data from power
stations, natural gas distribution companies
and assumption data based on the annual
report from the Egyptian electricity holding
company to insert hydraulic data to the
model and to run the model of natural gas
network by the steady state analysis to
predict the network behavior before
converting the model to unsteady state
analysis as the real case using demand
profiles. After completing the analysis,
model calibration to our model output from
Synergi for the model reliability by total
average of natural gas distributed through
the national gas grid and gas consumption by
each sector within the natural gas grid.
operate the network at a specific period and

a specific conditions, this operating scenario
includes the demands flow rate reduction
plan, the compressor stations set pressure
and the regulator stations set pressure, to
ensure the maximum performance of the gas
network operation.
On the other hand, we will discuss how to
overcome the pressure deterioration in the
gas network by adding a new compressor
stations to the gas network to operate at a
required high pressure.
In chapter seven, Optimization of the natural

gas network operation is discussed, it is
important to have knowledge about Genetic
science, the study of the mating and
development of living organisms and the
mutations that occurs. We will explain how
the scientists use these operations to
develop the optimization algorithms, in this
case called the Genetic Algorithms (GAs).
We will discuss the basics of the genetic
algorithms and its operators types, and how
they work and we will discuss how to
implement the optimization study of the
Genetic Algorithms (GAs) to optimize the
Egyptian national gas network operation, by
finding the optimum operating scenario to
19
Chapter 1
Natural Gas Network
Components
20
known natural gas well in 211 BC. In Europe,

natural gas was unknown until it was
discovered in Great Britain in 1659, although
it was not commercialized until about 1790.
In 1821 in Fredonia, United States, residents
observed gas bubbles rising to the surface
from a creek. William Hart, considered as
Americas father of natural gas, dug the first
natural gas well in North America (Speight,
1993). In 1885, Robert Bunsen invented a
burner that mixed air with natural gas. The
Bunsen burner could be used to provide
heat for cooking and warming buildings.
1.1 Introduction
Natural gas, like oil and coal, is a fossil fuel
energy source, situated deep under the
earth. The main component of natural gas is
methane (CH4), a colorless, odorless,
although flammable gas. When natural gas is
distributed commercially, it is supplemented
with an artificial odor so as to prevent leaks.
Natural gas is different than the well-known
cooking gas (known as LPG Liquefied
Petroleum Gas), whereas LPG is a product of
petroleum distillation, natural gas is pumped
from the underground, undergoes cleaning
and treatment and then is ready for use.
Similarly, to other fuels and energy sources,
like oil, diesel, fuel oil, coal and others,
natural gas also constitutes a primary energy
source, from which electricity may be
manufactured and mechanical energy and
heat produced. Natural gas is considered as
an environmentally friendly clean fuel,
offering important environmental benefits
when compared to other fossil fuels. The
superior environmental qualities over coal or
crude oil are that emissions of sulfur dioxide
are negligible or that the levels of nitrous
oxide and carbon dioxide emissions are
lower. This helps to reduce problems of acid
rain, ozone layer, or greenhouse gases.
Natural gas is also a very safe source of
energy when transported, stored, and used.
1.1.2 Natural gas Composition:

Natural gas is a complex mixture of
hydrocarbon
and
non-hydrocarbon
constituents and exists as a gas under
atmospheric conditions. Raw natural gas
typically consists primarily of methane (CH4),
the shortest and lightest hydrocarbon
molecule.
It also contains varying amounts of:
1.1.1 Natural gas history

The discovery of natural gas dates from
ancient times in the Middle East. Thousands
of years ago, it was noticed that natural gas
seeps ignited by lightning created burning
springs. In Persia, Greece, or India, people
built temples around these eternal flames
for their religious practices. However, the
energy value of natural gas was not
recognized until approximately 900 BC in
China, and the Chinese drilled the first
21
Heavier gaseous hydrocarbons: ethane

(C2H6), propane (C3H8), normal butane
(nC4H10), iso-butane (i-C4H10), pentanes
and even higher molecular weight
hydrocarbons. When processed and
purified into finished byproducts, all of
these are collectively referred to NGL
(Natural Gas Liquids).
Acid gases: carbon dioxide (CO2),
hydrogen sulfide (H2S) and mercaptans
such as methanethiol (CH3SH) and
ethanethiol (C2H5SH).
Other gases: nitrogen (N2) and helium
(He).
Water: water vapor and liquid water.
Liquid hydrocarbons: perhaps some
natural gas condensate (also referred to
as casing- head gasoline or natural
gasoline) and/or crude oil.
dissolved under high pressure and at high

temperatures in brines located deep
beneath the Earths surface; gas hydrates,
which are ice-like structures of water and gas
located under the permafrost; and deep
gas, which is found at levels much deeper
than conventional gas. Although there is no
scientific consensus, some believe deep gas
originated from inorganic sources and that it
exists everywhere as a result of the
geological processes that formed the earth.
Of the unconventional gas sources, the one
most important to the gas transportation
industry is coal bed methane.
Mercury: very small amounts of mercury

primarily in elementary form, but
chlorides and other species are possibly
present.
Table (1.1) outlines the typical makeup of
natural gas before it is refined. Natural
gas is considered dry when it is almost
pure methane, having had most of the
other
commonly
associated
hydrocarbons removed. When other
hydrocarbons are present, the natural
gas is wet. The composition of natural
gas varies depending on the field,
formation, or reservoir from which it is
extracted.
1.1.2.2 Non-associated Gas

Non-associated gas (sometimes called gas
well gas) is produced from geological
formations that typically do not contain
much, if any, higher boiling hydrocarbons
(gas liquids) than methane. Non-associated
gas can contain nonhydrocarbon gases such
as carbon dioxide and hydrogen sulfide. Nonassociated gas is directly controllable by the
producer; one just turns the valves. The gas
flows up the well under its own energy,
through the wellhead control valves, and
along the flow line to the treatment plant.
Treatment requires the temperature of the
gas to be reduced to a point dependent upon
the pressure in the pipeline so that all liquids
that would exist at pipeline temperature and
pressure condense and are removed.
Table 1-1
1.1.2.1 Gas sources

Natural gas produced from geological
formations comes in a wide array of
compositions. The varieties of gas
compositions can be broadly categorized
into three distinct groups: (1) non-associated
gas that occurs in conventional gas fields, (2)
associated gas that occurs in conventional oil
fields,
and
(3)
continuous
(or
unconventional) gas. Some types of
unconventional gas resources include tight
gas or tight sands gas, which is found in
low-permeability rock; coalbed methane
(CBM), which is natural gas that has been
formed along with the geological processes
that formed coal; natural gas from geopressurized aquifers, which refers to gas
1.1.2.3 Associated Gas

Associated gas is produced during crude oil
production and is the gas that is associated
with crude oil. Crude oil cannot be produced
without producing some of its associated
gas, which comes out of solution as the
pressure is reduced on the way to and on the
surface. Properly designed crude oil well
completions
and
good
reservoir
management are used to minimize the
22
production of associated gas so as to retain

the maximum energy in the reservoir and
thus increase ultimate crude oil recovery.
Crude oil in the reservoir with minimal or no
dissolved associated gas is rare and as dead
crude oil is often difficult to produce as there
is little energy to drive it. In the case of
associated gas, crude oil may be assisted up
the well bore by gas lift (Speight, 1993).
Thus, gas is compressed into the annulus of
the well and then injected by means of a gas
lift valve near the bottom of the well into the
crude oil column in the tubing. At the top of
the well the crude oil and gas mixture passes
into a separation plant that drops the
pressure down to nearly atmospheric in two
stages. The crude oil and water exit the
bottom of the lower pressure separator,
from where it is pumped to tanks for
separation of the crude oil and water. The
gas produced in the separators
and the gas that comes out of solution with
the produced crude oil.
some of the needed processing can be

accomplished at or near the wellhead (field
processing), the complete processing of
natural gas takes place at a processing plant,
usually located in a natural gas producing
region. The extracted natural gas is
transported to these processing plants
through a network of gathering pipelines,
which are small-diameter, low pressure
pipes. The actual practice of processing
natural gas to pipeline dry gas quality levels
can be quite complex, but usually involves
four main processes to remove the various
impurities:
Oil and Condensate Removal

Water Removal
Separation of Natural Gas Liquids
Sulfur and Carbon Dioxide Removal
1.1.3.2 Oil and Condensate Removal

In order to process and transport associated
dissolved natural gas, it must be separated
from the oil in which it is dissolved. This
separation of natural gas from oil is most
often done using equipment installed at or
near the wellhead. The actual process used
to separate oil from natural gas, as well as
the equipment that is used, can vary widely.
Although dry pipeline quality natural gas is
virtually
identical
across
different
geographic areas, raw natural gas from
different regions may have different
compositions and separation requirements.
In many instances, natural gas is dissolved in
oil underground primarily due to the
pressure that the formation is under. When
this natural gas and oil is produced, it is
possible that it will separate on its own,
simply due to decreased pressure; much like
opening a can of soda pop allows the release
of dissolved carbon dioxide. In these cases,
separation of oil and gas is relatively easy,
and the two hydrocarbons are sent separate
1.1.3 Processing of Natural Gas

1.1.3.1 Basic concepts of natural gas
processing
Natural gas goes through a major
transformation before reaching consumers.
Natural gas processing plants are used to
purify the raw natural gas extracted from
underground gas fields and brought up to
the surface by gas wells. The processed
natural gas, used as fuel by residential,
commercial and industrial consumers, is
almost pure methane (CH4) and is very much
different from the raw natural gas. Natural
gas processing consists of separating all of
the various hydrocarbons and fluids from the
pure natural gas, to produce what is known
as 'pipeline quality' dry natural gas. Due to
vigorous standards, natural gas must be
purified before it can be transported. While
23
ways for further processing. The most basic

type of separator is known as a conventional
separator. It consists of a simple closed tank,
where the force of gravity serves to separate
the heavier liquids like oil, and the lighter
gases, like natural gas.
1.1.3.5 Sulfur and Carbon Dioxide Removal

The removal of acid gases (CO2, H2S and
other sulfur components) from natural gas is
often referred to as gas sweetening process.
Carbon dioxide present in the natural gas
need to be removed in order to increase the
heating value of the gas, prevent corrosion
of pipeline and gas process equipment and
crystallization of CO2 during cryogenic
process (liquefaction process). The removal
of carbon dioxide can be accomplished in a
numbers of ways. Varieties of processes and
(improvement of each) have been developed
over the years to treat certain types of gas
with the aim of optimizing capital cost and
operating cost, meet Gas specifications and
for environmental purpose (Tennyson et .al
1977).
1.1.3.3 Water Removal

In addition to separating oil and some
condensate from the wet gas stream, it is
necessary to remove most of the associated
water. Most of the liquid, free water
associated with extracted natural gas is
removed by simple separation methods at or
near the wellhead. However, the removal of
the water vapor that exists in solution in
natural gas requires a more complex
treatment. This treatment consists of
dehydrating the natural gas, which usually
involves one of two processes: either
absorption, or adsorption. Absorption occurs
when the water vapor is taken out by a
dehydrating agent. Adsorption occurs when
the water vapor is condensed and collected
on the surface.
1.2Components of network
The natural gas pipeline network is a system
that moves gas from the areas of production
to the areas of consumption. Natural gas
pipelines provide three transportation
functions,
gathering,
transmission
(interstate and intrastate), and distribution.
As a system, the pipeline network comprises
a number of components that help move the
gas from production to consumption.
1.1.3.4 Separation of Natural Gas Liquids

Natural gas coming directly from a well
contains many natural gas liquids that are
commonly removed. In most instances,
natural gas liquids (NGLs) have a higher value
as separate products, and it is thus
economical to remove them from the gas
stream. The removal of natural gas liquids
usually takes place in a relatively centralized
processing plant, and uses techniques similar
to those used to dehydrate natural gas.
There are two basic steps to the treatment
of natural gas liquids in the natural gas
stream. First, the liquids must be extracted
from the natural gas. Second, these natural
gas liquids must be separated themselves,
down to their base components.
1.2.1 Transmission pipelines

Transmission pipelines are made of steel and
generally operate at pressures ranging from
500 to 1,400 pounds per square inch gauge
(psig). Pipelines can measure anywhere from
6 to 48 inches in diameter, although certain
component pipe sections can consist of small
diameter (ASTM), and American National
Standards Institute (ANSI) pipe that is as
small as 0.5 inch in diameter. However, this
small-diameter pipe is usually used only in
gathering and distribution systems, although
some is used for control-line or gauge-line
24
purposes. Mainline pipes, the principal

pipeline in a given system, are usually
between 16 and 48 inches in diameter.
Lateral pipelines, which deliver natural gas to
or from the mainline, are typically between
6 and 16 inches in diameter. Most major
interstate pipelines are between 24 and 36
inches in diameter. The actual pipeline itself,
commonly called line pipe, consists of a
strong carbon steel material engineered to
meet standards set by the American
Petroleum Institute (API), American Society
of Testing and Materials (ASTM), and
American National Standards Institute
(ANSI). Line pipe is produced in steel mills,
which are sometimes specialized to produce
only pipeline-quality components. There are
two different production techniques, one for
small-diameter pipes and one for largediameter pipes. For large-diameter pipes,
from 20 to 42 inches in diameter, the pipes
are produced from sheets of metal that are
folded into a tube shape, and the ends are
welded or fused together to form a pipe
section. Small-diameter pipe, on the other
hand, can be produced seamlessly. This
technique involves heating a metal bar to
very high temperatures and then punching a
hole through the middle of the bar to
produce a hollow tube. In either case, the
pipe is tested before being shipped from the
steel mill to ensure that it can meet the
pressure and strength standards for
transporting natural gas. Pipelines are
usually buried underground; the burial depth
varies depending on the local geography
along the pipeline route. Normal depth
requirements are 2 to 4 feet to top of pipe.
Many transmission pipeline companies
transport gas to metropolitan markets.
the installation of a typical transmission pipe

in a remote location.
Figure 1.1 installation of a typical transmission pipe in a

remote location.
1.2.1.1 single barrel and looped lines

Most pipelines are originally constructed as
single barrel systems, meaning they have
only a single line between the origin and
destination. As more capacity is needed,
these single barrel lines may be looped,
meaning a parallel line is installed along part
of the route with both ends connected to the
original pipe. As more capacity is needed, the
loop can be extended or another one added.
Splitting flow in this manner decreases flow
velocity thereby reducing friction loss and
increasing flow rate at the same operating
pressure.
Lines may include multiple loops. Adding one
or more loops is a common means of adding
capacity to natural gas transmission
pipelines. Carbon steel pipe is the standard
material for transmission pipelines due to its
high strength and extremely long life.
1.2.1.2 Line pack
The amount of gas in the pipe is called the
"line pack." By raising and lowering the
pressure on any pipeline segment, a pipeline
company can use the segment to store gas
during periods when there is less demand at
the end of the pipeline. Using line pack in
this way allows pipeline operators to handle
25
hourly fluctuations
efficiently.
in
demand
very
system is used to read pressures along the

pipeline on a continuous basis.
Leak detection methods may be divided into
two categories: direct and inferential. Direct
methods detect leaking commodity outside
the pipeline. Inferential methods deduce a
leak by measuring and comparing the
amount of product moving through various
points on a line.
1.2.1.3 Pipe-Coating Materials

Coating mills apply pipe coatings to ensure
that the pipe does not corrode once placed
in the ground. Often, the coating mill is
located adjacent to the pipe mill, so line pipe
moves directly from the pipe manufacturer
to the coating facility. The purpose of the
coating is to protect the pipe from moisture,
corrosive soils, and construction-induced
defects, which cause corrosion and rusting.
There are a number of different coating
techniques. In the past, pipelines were
coated with a specialized coal tar enamel.
Today, pipes are often protected with a
fusion-bond
epoxy
or
extruded
polyethylene, both of which give the pipes a
noticeable light yellow color. In addition,
cathodic protection is often used, which is a
technique that involves inducing an electric
current through the pipe toward off
corrosion and rusting (AGA 2004).
Corrosion in pipelines is a common

phenomenon, and must be controlled
effectively to prevent pipeline leaks or
structural problems. Although modern pipes
are constructed of high-quality steel, this
material will nevertheless corrode over time.
Corrosion occurs when an electrical current
flows naturally from a pipe into the
surrounding soil, causing metal loss, or
corrosion.
A cathodic protection system would be
installed to protect all underground and
submerged pipeline facilities that are
constructed of metallic materials from
external, internal, and atmospheric
corrosion.
To prepare for fusion-bond epoxy coating,

the external surface of the pipe is thoroughly
cleaned with a shot-blast process. The pipe is
then heated to a prescribed temperature,
and an epoxy powder is applied. The powder
melts onto the heated pipe and forms a
water-tight barrier. The mill tests the coated
pipe (called jeeping) with high voltage to
evaluate
the
coatings
insulating
effectiveness prior to transporting the pipe
to the job site.
Rectifiers and anode ground beds are

installed at strategic points along the
pipeline to further protect the pipeline from
corrosion. Ground beds provide cathodic
protection by inducing a very small electrical
current into the soil, impeding the flow of
electrons to the pipe. The rectifier that
induces the current into the ground bed is
checked regularly by pipeline personnel,
who ensure that the system is applying
sufficient current to maintain cathodic
protection to the pipeline. A single 200-foot
ground bed can protect as much as 50 miles
of pipeline, but the low voltages used do not
harm animals or plants in the vicinity.
Bonding occurs between separated pipelines
and companies within a common ROW (the
1.2.1.4 Pipeline operation, monitoring and

maintenance:
Once the natural gas pipeline is in service,
the pipelines control center electronically
monitors the operations 24 hours a day, 365
days a year. A computerized gas monitoring
26
current for all pipelines operates at the same

level).
1.2.1.5 Pig Launching/Receiving Facilities
Pigging facilities consist of pig launching or
receiving equipment and allow the pipeline
to accommodate a high-resolution internal
inspection tool, Pigs are devices that are
placed into a pipeline to perform certain
functions. Some are used to clean the inside
of the pipeline or to monitor its internal and
external condition. Launchers and receivers
are facilities that enable pigs to be inserted
into or removed from the pipeline.
Figure 1.31 Pig Launching/Receiving Facilities
1.2.1.6 Natural Gas Pipeline Release Events

Immediately following a guillotine-type
break in a natural gas pipeline, the gas begins
flowing rapidly through the break and into
the surrounding atmosphere. The escaping
natural gas creates a highly turbulent
mushroom shaped vapor cloud that
increases in height above the release point
due to the source momentum and buoyancy.
Initially, the natural gas flow from each
broken pipeline segment is balanced, and
the natural gas escapes to the atmosphere in
the form of jets that depend on the
alignment of the line pipe ends. Natural gas
will not burn unless the gas-to-air ratio is
between 4% and 15%. Noise produced by the
escaping natural gas is normally audible for a
long distance.
Figure 1.2 Pig Launching/Receiving Facilities
A pigging facility is generally smaller than a

typical compressor station site, but is
typically twice the size of a valve site. It may
be equipped with an office, workshops, a
small permanent camp (in remote locations),
power generators, storage areas, and a
helipad. It is normally unmanned except
during periods of manual pigging and during
routing service or repair.
For buried pipelines, the escaping natural

gas ejects the overlying soil forming a crater
of a size and shape which influences the
behavior of the released gas. the crater
produced by the natural gas pipeline rupture
that occurred near Carlsbad, New Mexico. As
the release continues, the natural gas jet
feeds the vapor cloud and entrains air that
may contain ejected soil particles. Without
an ignition source, the vapor cloud and the
escaping gas disperse into the atmosphere.
27
associated with each phase are described

below.
1.2.1.6.1 Phase 1 Detection:
The detection phase begins immediately
after the pipeline ruptures, t0, and continues
until the leak is detected by any method and
recognized by the Pipeline Operator, td. The
volume of natural gas discharged during the
detection phase depends on the duration of
this phase, td - t0, and is influenced by
factors such as the size, shape, and location
of the rupture; the performance
characteristics of the compressors; the
pipeline pressure at the time of the release;
and the effectiveness of the leak detection
system. In theory, the entire length of the
pipeline and its branch lines contribute to
the release during the detection phase
because the compressors are operating and
the block valves are open.
Figure 1.4 Crater resulting from natural gas pipeline

release near Carlsbad, New Mexico
If ignition of the released natural gas occurs

immediately, or shortly after, the guillotinetype break, a transient fireball will occur. The
fireball, which is the result of combustion of
the mushroom-shaped vapor cloud, typically
lasts 30 seconds or less leaving a quasisteady-state fire that continues to burn until
all of the escaping natural gas is consumed.
the fireball produced by the natural gas
pipeline rupture that occurred near
Carlsbad, New Mexico.
1.2.1.6.2 Phase 2 Block Valve Closure:

The block valve closure phase begins after
the leak is detected and corrective actions
are initiated to mitigate the consequences of
the release, td, and continues until the
upstream and downstream block valves are
closed, isolating the line section with the
break, ts. During the block valve closure
phase, natural gas continues to flow from
the break. The compressors may continue to
operate after the block valves are closed, but
their operation does not further affect the
gas release. The duration of this phase can
vary from a few minutes for systems with
remotely operated block valves to an hour or
more for manually operated equipment
located in remote areas. The volume of
natural gas discharged during the block valve
closure phase, ts- td, depends on the
duration of this phase and is influenced by
factors such as the type of equipment
controls (automatically, remotely, or
Figure 1.5 the fireball produced by the natural gas pipeline

rupture that occurred near Carlsbad, New Mexico.
Although the volume of natural gas released

depends on many factors, natural gas
releases are subdivided into three sequential
phases Phase 1: Detection, Phase 2: Block
Valve Closure, and Phase 3: Blowdown. The
total discharge volume equals the sum of the
volumes released during each phase. Events
28
manually operated) and personnel travel

time to shut down manually operated
equipment. The volume of natural gas
discharged during the block valve closure
phase is affected by the swiftness of block
valve closure.
1.2.1.8 Properties of Natural Gas through

pipelines
We will discuss the properties of gases that
influence gas flow through a pipeline. We
will explore the relationship of pressure,
volume, and temperature of a gas and how
the gas properties such as density, viscosity,
and compressibility change with the
temperature and pressure. Starting with the
ideal or perfect gases that obey the ideal gas
equation, we will examine how real gases
differ from ideal gases. The concept of
compressibility factor, or gas deviation
factor, will be introduced and methods of
calculating the compressibility factor using
some popular graphical correlation and
calculation methods explained.
1.2.1.6.3 Phase 3 Blowdown:

The Blowdown phase begins when the
portion of the pipeline that includes the
break is isolated by closure of upstream and
downstream block valves. This phase ends
when the natural gas remaining in the
isolated portions of the upstream and
downstream pipeline segments flows from
the break and burns, reducing the line
pressure to one atmosphere. The volume of
natural gas discharged during the Blowdown
phase depends on the duration of the
previous phases and is influenced by the line
pipe diameter and the distances from the
break to the nearest upstream and
downstream block valves.
The properties of a mixture of gases will be

discussed, and how these are calculated will
be covered. Understanding the gas
properties is an important first step toward
analysis of gas pipeline hydraulics. A fluid can
be a liquid or a gas. Liquids are generally
considered almost incompressible.
1.2.1.7 Block Valve Effects on a Natural Gas

Pipeline Release:
A gas is classified as a homogenous fluid with

low density and viscosity. It expands to fill
the vessel that contains the gas. The
molecules that constitute the gas are spaced
farther apart in comparison with a liquid and,
therefore, a slight change in pressure affects
the density of gas more than that of a liquid.
Gases,
therefore,
have
higher
compressibility than liquids. This implies that
gas properties such as density, viscosity, and
compressibility factor change with pressure.
Block valves have no influence on the

volume of natural gas released during the
detection phase because the block valves are
open and the compressors are operating
when natural gas begins escaping from the
break. However, rapid detection of the leak
and implementation of corrective actions
including closing block valves to isolate the
line section with the break reduce the total
volume of natural gas released. The
effectiveness of block valve closure in
mitigating the consequences of a natural gas
pipeline release decreases as the duration of
the detection and block valve closure phases
increase because thermal radiation effects
on buildings and humans are a function of
radiant heat flux and exposure duration.
1.2.1.8.1 Mass & weight

Mass is the quantity of matter in a
substance.
It
is
sometimes
used
interchangeably with weight. Strictly
speaking, mass is a scalar quantity, whereas
weight is a force and, therefore, a vector
29
quantity. Mass is independent of the

geographic location, whereas weight
depends upon the acceleration due to
gravity and, therefore, varies slightly with
geographic location. Mass is measured in
slugs in the U.S. Customary System (USCS) of
units and kilograms (kg) in System
International (SI) units.
Similarly, if the temperature is reduced, gas

pressure will also reduce, provided volume
remains constant. Charless law states that
for constant volume, the pressure of a fixed
mass of gas will vary directly with the
temperature. Thus, if temperature increases
by 20%, the pressure will also rise by 20%.
Similarly, if pressure is maintained constant,
the volume will increase in direct proportion
with temperature. Charless law, Boyles law,
and other gas laws will be discussed in detail
later in this chapter. Volume of gas is
measured in ft3 in USCS units and m3 in SI
units. Other units for volume include
thousand ft3 (Mft3) and million ft3 (MMft3) in
USCS units and thousand m3 (km) and million
m3 (Mm3) in SI units. When referred to
standard conditions (also called base
conditions) of temperature and pressure
(60F and 14.7 psia in USCS units), the
volume is stated as standard volume and,
therefore, Measured in standard ft (SCF) or
million standard ft3 (MMSCF). It must be
noted that in the USCS units, the practice has
been to use M to represent a thousand, and
therefore MM refers to a million. This goes
back to the Roman days of numerals, when
M represented a thousand. In SI units, a
more logical step is followed. For thousand,
the letter k (for kilo) is used and the letter M
(for Mega) is used for a million. Therefore,
500 MSCFD in USCS units refers to 500
thousand standard cubic feet per day
(500,000 ft3/day), whereas 15 Mm3/day
means 15 million cubic meters per day in SI
units. This distinction in the use of the letter
M to denote a thousand in USCS units and M
for a million in SI units must be clearly
understood. Volume flow rate of gas is
measured per unit time and can be
expressed as ft3 /min, ft3/h, ft3 /day, SCFD,
MMSCFD, etc. in USCS units. In SI units, gas
flow rate is expressed in m3/h or Mm3/day.
However, for most purposes, we say that a

10-lb mass has a weight of 10 lb. The pound
(lb) is a more convenient unit for mass, and
to distinguish between mass and Weight, the
terms pound mass (lbm) and pound force
(lbf) are sometimes used. A slug is equal to
approximately 32.2 lb. If some gas is
contained in a certain volume and the
temperature and pressure change, the mass
will remain constant unless some gas is taken
out or added to the container. This is known
as the principle of conservation of mass.
Weight is measured in pounds (lb) in USCS
units and in Newton (N) in SI units.
Sometimes we talk about mass flow rate
through a pipeline rather than volume flow
rate. Mass flow rate is measured in lb/hr in
USCS units or kg/hr in SI units.
1.2.1.8.2 Volume
Volume of a gas is the space a given mass of
gas occupies at a particular temperature and
pressure. Since gas is compressible, it will
expand to fill the available space. Therefore,
the gas volume will vary with temperature
and pressure. Hence, a certain volume of a
given mass of gas at some temperature and
pressure will decrease in volume as the
pressure is increased and vice versa.
Suppose a quantity of gas is contained in a
volume of 100 ft3 at a temperature of 80F
and a pressure of 200 psi. If the temperature
is increased to 100 F, keeping the volume
constant, the pressure will also increase.
30
temperature and pressure, its specific

volume its specific volume is 16.67 ft3/lb.
1.2.1.8.3 Density, Specific Weight, and

Specific Volume
1.2.1.8.4 Specific Gravity
Density represents the amount of gas that

can be packed in a given volume. Therefore,
it is measured in terms of mass per unit
volume. If 5 lb of a gas is contained in 100 ft3
of volume at some temperature and
pressure, we say that the gas density is
5/100=0.05 lb/ft3.Strictly speaking, in USCS
units density must be expressed as slug/ft3
since mass is customarily referred to in slug.
Thus,
Specific gravity of a gas, sometimes called

Gravity, is a measure of how heavy the gas is
compared to air at a particular temperature.
It might also be called relative density,
expressed as the ratio of the gas density to
the density of air. Because specific gravity is
a ratio, it is a dimensionless quantity.
(1.1)
(1.2)
Where:
Where
G= gas gravity, dimensionless
=density of gas
g= density of gas
m=mass of gas
air= density of air
V=volume of gas
Both densities in Equation 1.2 must be in the

same units and measured at the same
temperature. For example, natural gas has a
specific gravity of 0.60 (air= 1.00) at 60F.
This means that the gas is 60% as heavy as
air. If we know the molecular weight of a
particular gas, we can calculate its gravity by
dividing the molecular weight by the
molecular weight of air, as follows.
Density is expressed in slug/ft3 or lb/ft3 in

USCS units and kg/m3 in SI units. A
companion term called specific weight is also
used when referring to the density of gas.
Specific weight, represented by the symbol
g, is the weight of gas per unit volume
measured in lb/ft3 in USCS units, and is,
therefore, contrasted with density, which is
measured in slug/ft3. In SI units, the specific
weight is expressed in Newton per m3
(N/m3). Quite often, density is also referred
to in lb/ft3 in USCS units.
29
(1. 1)
Rounding off the molecular weight of air to

29
The reciprocal of the specific weight is

known as the specific volume. By definition,
therefore, specific volume represents the
volume occupied by a unit weight of gas. It is
measured in ft3 /lb in USCS units and m3/N
in SI units. If the specific weight of a
particular gas is 0.06 lb/ft3 at some
Where G= specific gravity of gas

Mg = molecular weight of gas
Mair = molecular weight of air=
28.9625
Since natural gas consists of a mixture of
several gases (methane, ethane, etc.), the
molecular weight Mg in Equation 1.3 is
31
referred to as the apparent molecular weight

of the gas mixture. When the molecular
weight and the percentage or mole fractions
of the individual components of a natural gas
mixture are known, we can calculate the
molecular weight of the gas mixture by using
a weighted average method. Thus, a natural
gas Mixture consisting of 90% methane, 8%
ethane, and 2% propane will have a specific
gravity of
(.91)+(.082)+(.023)
29
simply the absolute viscosity divided by the

density. The two viscosities are related as
follows:
(1.5)
Where, in USCS units,

= kinematic viscosity, ft2/s
= dynamic viscosity, lb/ft-s
= density, lb/ft3 and in SI units
(1.4)
= kinematic viscosity, St
Where:
=dynamic viscosity, P
M1, M2, and M3 are the molecular weights

of methane, ethane, and propane,
respectively, and 29 represent the molecular
weight of air. Table 1.2 lists the molecular
weights and other properties of several
hydrocarbon gases.
= density, kg/m3
Kinematic viscosity is expressed in ft2/s in
USCS units and Stokes (St) in SI units. Other
units of viscosity in SI units include
centipoise (cp) for dynamic viscosity and
centistokes (cst) for kinematic viscosity. The
viscosity of a gas depends on its temperature
and pressure. Unlike liquids, the viscosity of
a gas increases with increase in temperature.
Since viscosity represents resistance to flow,
as the gas temperature increases, the
quantity of gas flow through a pipeline will
decrease; hence, more throughput is
possible in a gas pipeline at lower
temperatures.
1.2.1.8.5 Viscosity
The viscosity of a fluid represents its
resistance to flow. The higher the viscosity,
the more difficult it is to flow. Lowerviscosity fluids flow easily in pipes and cause
less pressure drop. Liquids have much larger
values of viscosity compared to gases. For
example, water has a viscosity of 1.0
centipoise (cp), whereas viscosity of natural
gas is approximately 0.0008 cp. Even though
the gas viscosity is a small number, it has an
important function in determining the type
of flow in pipelines.
1.2.1.8.6 Real Gases

When dealing with real gases, we can apply
the ideal gas equation discussed in the
preceding sections and get reasonably
accurate results only when the pressures are
close to the atmospheric pressure.
The Reynolds Number is a dimensionless

parameter that is used to classify flow rate in
pipelines. It depends on the gas viscosity,
flow rate, pipe diameter, Temperature, and
pressure. The absolute viscosity, also called
the dynamic viscosity, is expressed in lb/ft-s
in USCS units and Poise (P) in SI units. A
related term is the kinematic viscosity. This is
When pressures are higher, the ideal gas

equation will not be accurate for most real
gases. The error in calculations at high
32
pressures using the ideal gas equation may

be as high as 500% in some instances.
Where
P = absolute pressure of gas, psia
This compares with errors of 2 to 3% at low

pressures. At higher temperatures and
pressures, the equation of state that
relates pressure, volume, and temperature is
used to calculate the properties of gases.
Many of these correlations require a
computer program to get accurate results in
a reasonable amount of time.
V = volume of gas, ft3

Z = gas compressibility factor, dimensionless
T = absolute temperature of gas, R.
n = number of lb moles.
R = universal gas constant, 10.73 psia ft3/lb
mole R.
However, we can modify the ideal gas

equation and obtain reasonably accurate
results fairly quickly using manual
calculations. Two terms called critical
temperature and critical pressure need to be
defined.
The theorem known as corresponding states

says that the extent of deviation of a real gas
from the ideal gas equation is the same for
all real gases when the gases are at the same
corresponding state.
The corresponding state can be represented
by the two parameters called reduced
temperature and reduced pressure.
The critical temperature of a pure gas is

defined as the temperature above which a
gas cannot be compressed to form a liquid,
regardless of the pressure. The critical
pressure is defined as the minimum pressure
that is required at the critical temperature to
compress a gas into a liquid. Real gases can
be considered to follow a modified form of
the ideal gas law
The reduced temperature is the ratio of the

temperature of the gas to its critical
temperature. Similarly, the reduced pressure
is the ratio of the gas pressure to its critical
pressure as indicated in the following
equations:
The modifying factor is included in the gas

property known as the compressibility factor
Z. This is also called the gas deviation factor.
It can be defined as the ratio of the gas
volume at a given temperature and pressure
to the volume the gas would occupy if it were
an ideal gas at the same temperature and
pressure. Z is a dimensionless number less
than 1.0 and it varies with temperature,
pressure, and composition of the gas.
(1.13)
Where
P = absolute pressure of gas, psia
T = absolute temperature of gas, R
Tr = reduced temperature, dimensionless
Using the compressibility factor Z, the ideal

gas equation is modified for real gas as
follows:
(1.12)
PR = reduced pressure, dimensionless

Tc = critical temperature, R
(1.2)
Pc = critical pressure, psia
33
Another term, the super compressibility

factor, Fpv, which is related to the
compressibility factor Z,
1.2.1.8.7 Pseudo-Critical Properties from

Gas Gravity
If the percentages of the various
components in the natural gas mixture are
not available, we can calculate approximate
values of the pseudo-critical properties of
the gas mixture if we know the gas gravity.
The
pseudo-critical
properties
are
calculated, approximately, from the
following equations:
= 170.491 + 307.344 (1.14)

= 709.604 58.718 (1.15)
Where
G = gas gravity (air = 1.00)
Tpc = pseudo-critical temperature, R
Figure 1.6 Compressibility factor chart for natural

gases.
Ppc = pseudo-critical pressure, psia

1.2.1.8.8 Compressibility Factor
Is defined as follows:
The compressibility factor, or gas deviation

factor, was briefly mentioned in Section
3.1.6. It is a measure of how close a real gas
is to an ideal gas. The compressibility factor
is defined as the ratio of the gas volume at a
given temperature and pressure to the
volume the gas would occupy if it were an
ideal gas at the same temperature and
pressure. The compressibility factor is a
dimensionless number close to 1.00 and is a
function of the gas gravity, gas temperature,
gas pressure, and the critical properties of
the gas. As an example, a particular natural
gas mixture can have a compressibility factor
equal to 0.87 at 1000 psia and 80F.Charts
have been constructed that depict the
variation of Z with the reduced temperature
and reduced pressure.
(1.16)
There are several approaches to calculating

the compressibility factor for a particular gas
temperature T and pressure P. One method
uses the critical temperature and critical
pressure of the gas mixture. First, the
reduced temperature, Tr, and reduced
pressure, Pr, are calculated from the given
gas temperature and gas pressure and the
critical temperature and critical pressure
using Equation 1.12 and Equation 1.13.
Once we know the values of Tr and Pr, the
compressibility factor can be found from
charts similar to the Standing-Katz chart. This
will be illustrated using an example.
The following methods are available to
calculate the compressibility factor:
a. Standing-Katz method
34
b. Dranchuk, Purvis, and Robinson method

c. AGA method
d. CNGA method
Although the Standing-Katz method is the
most popular, we will discuss this as well as
the AGA and CNGA methods.
1.2.1.8.9 Pressure Drop Due to Friction
We will discuss the various methods of
calculating the pressure drop due to friction
in a gas pipeline. The pipeline throughput
(flow rate) will depend upon the gas
properties, pipe diameter and length, initial
gas pressure and temperature, and the
pressure drop due to friction. Commonly
used formulas will be reviewed and
illustrated using examples. The impact of
internal conditions of the pipe on the pipe
capacity will also be explored.
Figure 1.7 Energy of flow of a fluid.
Hf represents the total frictional pressure

loss between points A and B.
Starting with the basic energy Equation
(1.18)
Applying gas laws, and after some
simplification, various formulas were
developed over the years to predict the
performance of a pipeline transporting gas.
These formulas are intended to show the
1.2.1.8.9.1 Bernoullis Equation

As gas flows through a pipeline, the total
energy of the gas at various points consists
of energy due to pressure, energy due to
velocity, and energy due to position or
elevation above an established datum.
Bernoullis equation simply connects these
components of the energy of the flowing
fluid to form an energy conservation
equation. Bernoullis equation is stated
afollows, considering two points, 1 and 2
2
2
2
2
Relationship between the gas properties,

such as gravity and compressibility factor,
with the flow rate, pipe diameter and length,
and the pressures along the pipeline.
Thus, for a given pipe size and length, we can
predict the flow rate possible through a
pipeline based upon an inlet pressure and an
outlet pressure of a pipe segment.
Simplifications are sometimes introduced,
such as uniform gas temperature and no
heat transfer between the gas and the
surrounding soil in a buried pipeline, in order
to adopt these equations for manual
calculations. With
the
advent
of
microcomputers, we are able to introduce
heat transfer effects and, therefore, more
accurately model gas pipelines, taking into
consideration gas flow temperatures,
soil
temperatures,
and
thermal
conductivities of pipe material, insulation,
+ = +
(1.18)
Where
Hp is the equivalent head added to the fluid
by a compressor at A and
35
and soil. In this chapter we will concentrate

on steady-state isothermal flow of gas in
pipelines.
1.2.1.8.9.2 Flow Equations
= 1.1494 103 ( )
12 22
.5
2.5
For an explanation of symbols u0sed.
Several equations are available that relate

the gas flow rate with gas properties, pipe
diameter and length, and upstream and
downstream pressures. These equations are
listed as follows:
1. General Flow equation
Where
Q=gas flow rate, measured at standard
conditions, m3/day
f=friction factor, dimensionless
2. Colebrook-White equation
Pb=base pressure, kPa
3. Modified Colebrook-White equation
Tb=base temperature, K (273+C)
4. AGA equation
P1=upstream pressure, kPa
5. Weymouth equation
P2=downstream pressure, kPa
6. Panhandle A equation
G=gas gravity (air=1.00)
7. Panhandle B equation
8. IGT equation
Tf=average gas flowing temperature, K

(273+C)
9. Spitzglass equation
L=pipe segment length, km
10. Mueller equation
Z=gas compressibility factor at the flowing

temperature, dimensionless
11. Fritzsche equation
D=pipe inside diameter, mm
We will discuss each of these equations,

their limitations, and their applicability to
compressible fluids, such as natural gas.
1.2.1.8.9.3 GENERAL FLOW EQUATION
The General Flow equation, also called the
Fundamental Flow equation, for the steadystate isothermal flow in a gas pipeline is the
basic equation for relating the pressure drop
with flow rate. The most common form of
this equation in the U.S. Customary System
(USCS) of units is given in terms of the pipe
diameter, gas properties, pressures,
temperatures, and flow rate as follows
properties, pressures, temperatures, and
flow rate as follows
Figure 1.8 Steady flow in gas pipeline.
Due to the nature of Equation 1.19, the

pressures can also be in MPa or bar, as long
as the same consistent unit is used.
Equation 1.19 relates the capacity (flow rate
or throughput) of a pipe segment of length L,
based on an upstream pressure of P1 and a
downstream pressure of P2. It is assumed
that there is no elevation difference
36
between the upstream and downstream

points; therefore, the pipe segment is
horizontal. Upon examining the General
Flow Equation, we see that for a pipe
segment of length L and diameter D, the gas
flow rate Q (at standard conditions) depends
on several factors. Q depends on gas
properties represented by the gravity G and
the Compressibility factor Z. If the gas gravity
is increased (heavier gas), the flow rate will
decrease. Similarly, as the compressibility
factor Z increases, the flow rate will
decrease. Also, as the gas flowing
temperature Tf increases, throughput will
decrease.
upstream pressure P1 constant, the flow rate

will increase.
It is obvious that when there is no flow rate,
P1 is equal to P2. It is due to friction between
the gas and pipe walls that the pressure drop
(P1P2) occurs from the upstream point 1 to
downstream point 2. The friction factor f
depends on the internal condition of the
pipe as well as the type of flow (laminar or
turbulent). Sometimes the General Flow
equation is represented in terms of the
transmission factor
F instead of the friction factor f
This form of the equation is as follows.
Thus, the hotter the gas, the lower the flow

rate will be. Therefore, to increase the flow
rate, it helps to keep the gas temperature
low. The impact of pipe length and inside
diameter is also clear. As the pipe segment
length increases for given pressure P1 and P2
= 5.747 104 ( )
12 22
.5
) 2.5 (1.20)
Where the transmission factor F and friction

factor f are related by
It is obvious that when there is no flow rate,

P1 is equal to P2. It is due to friction between
the gas and pipe walls that the pressure drop
(P1P2) occurs from the upstream point 1 to
downstream point 2. The friction factor f
depends on the internal condition of the
pipe as well as the type of flow (laminar or
turbulent). Sometimes the General Flow
equation is represented in terms of the
transmission factor
(1.21)
1.2.1.8.10 Effect of Pipe Elevations

When elevation difference between the
ends of a pipe segment is included, the
General Flow equation is modified as
follows:
= 5.747 104 ( )
F instead of the friction factor f
12 22
This form of the equation is as follows.
.5
) 2.5 (1.22)
Where
The flow rate will decrease. On the other

hand, the larger the diameter, the larger the
flow rate will be. The term 12 22
represents the driving force that causes the
flow rate from the upstream end to the
downstream end. As the downstream
pressure P2 is reduced, keeping the
( 1)
(1.23)
The equivalent length, Le, and the term ^

take into account the elevation difference
between the upstream and downstream
37
ends of the pipe segment. The parameter s

depends upon the gas gravity, gas
compressibility
factor,
the
flowing
temperature, and the elevation difference.
In SI units, the elevation adjustment
parameter S is defined as follows:
= .0684 (
2 1

accordance with Equation, from the pipeline

inlet to the end of each segment. In the
subsequent sections of this chapter, we will
discuss how the friction factor and
transmission factor are calculated using
various equations such as Colebrook- White
and AGA. It is important to note that the
General Flow equation is the most
commonly used equation to calculate the
flow rate and pressure in a gas pipeline.
) (1.24)
Where
s= elevation
dimensionless
adjustment
In order to apply it correctly, we must use

the correct friction factor or transmission
factor. The Colebrook equation, AGA
equation, and other empirical equations are
used to calculate the friction factor to be
used in the General Flow equation. Several
other equations, such as Panhandle A,
Panhandle B, and Weymouth, calculate the
flow rate for a given pressure without using
a friction factor or transmission factor.
However, an equivalent friction factor (or
transmission factor) can be calculated using
these methods as well.
parameter,
H1= upstream elevation, ft H2=downstream

elevation, ft
e=base of natural logarithms (e=2.718)
In the calculation of Le in Equation 1.2.23, we
have assumed that there is a single slope
between the upstream point 1 and the
downstream point 2. If, however, the pipe
segment of length L has a series of slopes,
then we introduce a parameter j as follows
for each individual pipe sub segment that
constitutes the pipe length from point 1 to
point 2.
(1.25)
The parameter j is calculated for each slope

of each pipe sub segment of length L1, L2,
etc. that make up the total length L.
The equivalent length term Le is calculated
by summing the individual slopes as defined
below.
= 1 1+ 2 2 +
(1.26)
The terms j1, j2, etc. for each rise or fall in

the elevations of individual pipe
Sub segments are calculated for the
parameters s1, s2, etc. for each segment in
38
A and B are related by the following

equation:
1.2.1.8.11 Velocity of Gas in a Pipeline

The velocity of gas flow in a pipeline
represents the speed at which the gas
molecules move from one point to another.
Unlike a liquid pipeline, due to
compressibility, the gas velocity depends
upon the pressure and, hence, will vary along
the pipeline even if the pipe diameter is
constant.
U11 = U22 (1.30)

If inlet conditions are represented by point A
and the volume flow rate Q at standard
conditions of 60F and 14.7 psia are known,
we can calculate the velocity at any point
along the pipeline at which the pressure and
temperature of the gas are P and T,
respectively.
The highest velocity will be at the

downstream end, where the pressure is the
least. Correspondingly, the least velocity will
be at the upstream end, where the pressure
is higher.
The velocity of gas at section 1 is related to

the flow rate Q1 at section 1 and pipe crosssectional area A as follows from Equation
1.31
Consider a pipe transporting gas from point

A to point B. Under steady state flow, at A,
the mass flow rate of gas is designated as M
and will be the same as the mass flow rate at
point B, if between A and B there is no
injection or delivery of gas. The mass being
the product of volume and density, we can
write the following relationship for point A:
The mass flow rate M at section 1 and 2 is the

same for steady-state flow.
Therefore,
M=Q11 = Q22= Qb (1.31)

Applying the gas law Equation, we get:
(1.27)
The volume rate Q can be expressed in terms

of the flow velocity u and pipe cross sectional
area A as follows:
= Z1RT1 (1.32)
Since Zb = 1.00, approximately, we can

simplify this to
1 = ( ) (1) 1
(1.28)
(1.33)
Therefore, combining Equation 1.2.28 and

Equation 1.27 and applying the conservation
of mass to points A and B, we get
In SI units, the gas velocity at any point in a

gas pipeline is given by:
M1=U1A11 = M2= U2A22 (1.29)
= 14.7349 ( 2 ) ( ) ( ) (1.34)
Where subscripts 1 and 2 refer to points A

and B, respectively. If the pipe is of uniform
cross section between A and B, then A1 = A2
= A. Therefore, the area term in Equation. At
Where
u = gas velocity, m/s

Qb = gas flow rate, measured at standard
conditions, m3/day
D = pipe inside diameter, mm
39
Pb = base pressure, kPa
velocity Equation can be rewritten in (USCS

units) as
Tb = base temperature, K (273 + C)
P = pressure, kPa
= 10029
T = average gas flowing temperature, K (273

+ C)
(1.36)
Z = gas compressibility factor at the flowing

temperature, dimensionless.
Where
1.2.1.8.11.1 Erosional Velocity
Z = compressibility factor of gas,

dimensionless
We have seen from the preceding section

that the gas velocity is directly related to the
flow rate. As flow rate increases, so does the
gas velocity. How high can the gas velocity be
in a pipeline?
R = gas constant = 10.73 ft3 psia/lb-mole R

T = gas temperature, R
P = gas pressure, psia
As the velocity increases, vibration and

noise are evident.
Usually, an acceptable operational velocity

is 50% of the above.
In addition, higher velocities will cause

erosion of the pipe interior over a long
period of time. The upper limit of the gas
velocity is usually calculated approximately
from the following equation:
100
1.2.1.8.11.2 Reynolds Number of Flow

An important parameter in flow of fluids in a
pipe is the non-dimensional term Reynolds
number. The Reynolds number is used to
characterize the type of flow in a pipe, such
as laminar, turbulent, or critical flow. It is
also used to calculate the friction factor in
pipe flow. We will first outline the calculation
of the Reynolds number based upon the
properties of the gas and pipe diameter and
then discuss the range of Reynolds number
for the various types of flow and how to
calculate the friction factor. The Reynolds
number is a function of the gas flow rate,
pipe inside diameter, and the gas density and
viscosity and is calculated from the following
equation in (USCS units):
(1.35)
Where
max = maximum or erosional velocity, ft/s
= gas density at flowing temperature,
lb/ft3
Since the gas density r may be expressed in
terms of pressure and temperature,
usingthe gas law Equation, the maximum
(1.37)
Where
Re = Reynolds number, dimensionless
40
u = average velocity of gas in pipe, ft/s
approximately 2000. Turbulent flow occurs

when the Reynolds number is greater than
4000. For Reynolds numbers between 2000
and 4000, the flow is undefined and is
referred to as critical flow. Thus,
D = inside diameter of pipe, ft

= gas density, lb/ft3
= gas viscosity, lb/ft-s
For laminar flow, Re 2000
In gas pipeline hydraulics, using customary

units, a more suitable equation for the
Reynolds number is as follows in (USCS
units):
For turbulent flow, Re > 4000

For critical flow, Re > 2000 and Re 4000
Most natural gas pipelines operate in the
turbulent flow region. Therefore, the
Reynolds number is greater than 4000.
Turbulent flow is further divided into three
regions known as smooth pipe flow, fully
rough pipe flow, and transition flow.
= .0004778 ( ) ( ) (1.39)
Where
Pb = base pressure, psia
1.2.1.8.12 Friction Factor
Tb = base temperature, R (460 + F)
In order to calculate the pressure drop in a

pipeline at a given flow rate, we must first
understand the concept of friction factor.
The term friction factor is a dimensionless
parameter that depends upon the Reynolds
number of flow. In engineering literature, we
find two different friction factors mentioned.
The Darcy friction factor is more common
and will be used throughout this book.
Another friction factor known as the Fanning
friction factor is preferred by some
engineers. The Fanning friction factor is
numerically equal to one-fourth the Darcy
friction factor as below.
G = specific gravity of gas (air = 1.0)

Q = gas flow rate, standard ft3/day (SCFD)
D = pipe inside diameter, in.
= viscosity of gas, lb/ft-s
In SI units, the Reynolds number is:
= .5134 ( ) ( ) (1.40)
Where
(1.41)
G = specific gravity of gas (air = 1.0)

Q = gas flow rate, m3/day (standard
conditions)
Where
=Fanning friction factor
D = pipe inside diameter, mm
=Darcy friction factor
= viscosity of gas, Poise
To avoid confusion, in subsequent

discussions, the Darcy friction factor is used
Laminar flow occurs in a pipeline when the

Reynolds number is below a value of
41
and will be represented by the symbol f. For

laminar flow, the friction factor is inversely
proportional to the Reynolds number, as
indicated below.
64
simply a dimensionless parameter obtained

by dividing the absolute (or internal) pipe
roughness by the pipe inside diameter as
follows:
(1.42)
(1.43)
For turbulent flow, the friction factor is a

function of the Reynolds number, pipe inside
diameter, and internal roughness of the
pipe. Many empirical relationships for
calculating f have been put forth by
researchers. The more popular correlations
include
Where
e = absolute or internal roughness of pipe,
in.
1.2.1.8.12.1 Colebrook-White Equation
The Colebrook-White equation, sometimes
referred to simply as the Colebrook
equation, is a relationship between the
friction factor and the Reynolds number,
pipe roughness, and inside diameter of pipe.
The following form of the Colebrook
equation is used to calculate the friction
factor in gas pipelines in turbulent flow.
The Colebrook-White and AGA equations.

Before we discuss the equations for
calculating the friction factor in turbulent
flow, it is appropriate to analyze the
turbulent flow regime. Turbulent flow in
pipes
(Re > 4000) is subdivided into three
separate regions as follows:
1. Turbulent flow in smooth pipes
2. Turbulent flow in fully rough pipes
(1.44)
3. Transition flow between smooth pipes

and rough pipes.
Where
f = friction factor, dimensionless
For turbulent flow in smooth pipes, the

friction factor f depends only on the
Reynolds number. For fully rough pipes, f
depends more on the pipe internal
roughness and less on the Reynolds number.
In the transition zone between smooth pipe
flow and flow in fully rough pipes, f depends
on the pipe roughness, pipe inside diameter,
and the Reynolds number. The various flow
regimes are depicted in the Moody diagram,
The Moody diagram is a graphic plot of the
variation of the friction factor with the
Reynolds number for various values of
relative pipe roughness. The latter term is

e = absolute pipe roughness, in.
Re = Reynolds number of flow,
dimensionless.
Since Re and f are dimensionless, as long as
consistent units are used for both e and D,
the Colebrook equation is the same
regardless of the units employed.
In order to calculate the friction factor f, we
must use a trial-and-error approach.
42
It is an implicit equation in f, since f appears

on both sides of the equation. We first
assume a value of f (such as 0.01) and
substitute it in the right-hand side of the
Therefore, for smooth pipe flow, the friction

factor equation reduces to:
Figure 2 Moody chart
equation. This will yield a second

approximation for f, which can then be used
to calculate a better value of f, and so on.
Generally, 3 to 4 iterations are sufficient to
converge on a reasonably good value of the
friction factor. It can be seen from the
Colebrook Equation, for turbulent flow in
smooth pipes, the first term within the
square brackets is negligible compared to
the second term, since pipe roughness e is
very small.
(1.45)
For turbulent flow in smooth pipes similarly,
for turbulent flow in fully rough pipes, with
Re being a large number, f depends mostly
on the roughness e and, therefore, the
friction factor equation reduces to
(1.46)
43
For turbulent flow in fully rough pipes Table

2.1.3 lists typical values of pipe internal
roughness used to calculate the friction
factor.
F = transmission factor
Table 1.4 Pipe Internal Roughness
The Colebrook-White equation discussed in

the preceding section has been in use for
many years in both liquid flow and gas flow.
The U.S. Bureau of Mines, in 1956, published
a report that introduced a modified form of
the Colebrook-White equation. The
modification results in a higher friction factor
and, hence, a smaller value of the
transmission factor. Because of this, a
conservative value of flow rate is obtained
due to the higher friction and pressure drop.
The modified version of the ColebrookWhite equation for turbulent flow is as
follows:
1.2.1.8.13.1 Modified Colebrook- White

Equation
1.2.1.8.13 Transmission Factor

The transmission factor F is considered the
opposite of the friction factor f. whereas the
friction factor indicates how difficult it is to
move a certain quantity of gas through a
pipeline, the transmission factor is a direct
measure of how much gas can be
transported through the pipeline.
(1.49)
Since Re, f, and F are dimensionless, as long
as consistent units are used for both e and D,
the modified Colebrook equation is the
same, regardless of the units employed.
As the friction factor increases, the

transmission
factor
decreases
and,
therefore, the gas flow rate also decreases.
Conversely, the higher the transmission
factor, the lower the friction factor and,
therefore, the higher the flow rate will be.
1.2.1.8.13.2 American Gas Association

(AGA) Equation
The transmission factor F is related to the

friction factor f as follows:
In 1964 and 1965, the American Gas

Association (AGA) published a report on how
to calculate the transmission factor for gas
pipelines to be used in the General Flow
equation. This is sometimes referred to as
the AGA NB- 13 method. Using the method
outlined in this report, the transmission
factor F is calculated using two different
equations. First, F is calculated for the rough
pipe law (referred to as the fully turbulent
zone).
(1.47)
Therefore,
(1.48)
Where
f = friction factor
Next, F is calculated based on the smooth

pipe law (referred to as the partially
44
turbulent zone). Finally, the smaller of the

two values of the transmission factor is used
in the General Flow Equation for pressure
drop calculation. Even though the AGA
method uses the transmission factor F
instead of the friction factor f, we can still
calculate the friction factor using the
relationship shown in Equation.
length of the
consideration.
pipe
section
under
(1.53)
Note: The drag factors above are based on
40-ft joints of pipelines and mainline valves
at 10-mile spacing.
For the fully turbulent zone, AGA

recommends using the following formula for
F, based on relative roughness e/D and
independent of the Reynolds number:
Table 1.5 Bend Index and Drag Factor
(1.50)
Equation 1.2.50 is also known as the Von
Karman rough pipe flow equation.
For the partially turbulent zone, F is
calculated from the following equations
using the Reynolds number, a parameter Df
known as the pipe drag factor, and the Von
Karman smooth pipe transmission factor Ft:
The value of Df is generally chosen from

Table 1.5. For further discussion on the bend
index and drag factor, the reader is referred
to Steady Flow in Gas Pipelines listed in the
Reference section.
(1.51)
And,
1.2.1.8.13.3 Weymouth Equation
Where
The Weymouth equation is used for high

pressure, high flow rate, and large diameter
gas gathering systems. This formula directly
calculates the flow rate through a pipeline
for given values of gas gravity,
compressibility, inlet and outlet pressures,
pipe diameter, and length. In USCS units, the
Weymouth equation is stated as follows:
(1.52)
Ft = Von Karman smooth pipe transmission

factor
Df = pipe drag factor that depends on the
Bend Index (BI) of the pipe the pipe drag
factor Df is a parameter that takes into
account the number of bends, fittings, etc.
Its value ranges from 0.90 to 0.99. The Bend
index is the sum of all the angles and bends
in the pipe segment, divided by the total
= 433.5 ( )
45
12 .22
0.961 .
) . 2.667 (1.54)
Where
1.2.1.8.13.4 Panhandle A Equation
Q = volume flow rate, standard ft3/day

(SCFD)
The Panhandle an Equation was developed

for use in natural gas pipelines, incorporating
an efficiency factor for Reynolds numbers in
the range of 5 to 11 million. In this equation,
the pipe roughness is not used. The general
form of the Panhandle A equation is
expressed in USCS units as follows:
E = pipeline efficiency, a decimal value less

than or equal to 1.0
1.0788 2
(1 22 )
( )
= 435.87 0.8539
. .
2.2.618
P1 = upstream pressure, psia

P2 = downstream pressure, psia
(1.56)
Tf = average gas flow temperature, R (460 +

F)
Where
(SCFD)
Le = equivalent length of pipe segment, mi


than 1.0
In SI units, the Weymouth equation is as

follows:

( ) (12 22 )
= 3.7435 103
. .
2.667

(1.55)

F)
Where
Q = gas flow rate, standard m3/day
Le = equivalent length of pipe
Tf = average gas flow temperature, K (273 +

C)
Other symbols are as defined previously.

In SI units, the Panhandle A equation is:
P1 = upstream pressure, kPa

P2 = downstream pressure, kPa
Le = equivalent length of pipe segment, km
46
= 4.5969 103
1.0788 2
(1 22 )
( )
0.8539 . .
2.2.618

than 1.0
(1.57)
Where

than 1.0

F)

C)

Other symbols are as defined previously. In
SI units, the Panhandle B equation is
P1 = upstream pressure, kPa (absolute)
1.02
= 1.002 102 ( )
0.51
12 . 22
( 0.961
)
. 2.53
. .
P2 = downstream pressure, kPa (absolute)

other symbols are as defined previously.
1.2.1.8.13.5 Panhandle B Equation
(1.59)
The Panhandle B equation, also known as the

revised Panhandle equation, is used in large
diameter, high pressure transmission lines.
In fully turbulent flow, it is found to be
accurate for values of Reynolds number in
the range of 4 to 40 million. This equation in
USCS units is as follows:
Where
than 1.0
1.02
= 737 ( )

0.51
12 . 22
( 0.961
)
. .

C)
. 2.53
(1.58)
Where

(SCFD)

Z = gas compressibility factor at the flowing
temperature, dimensionless.
47
1.2.1.8.13.6 Institute Of Gas Technology
(IGT) Equation
Tb
= 1.28822 103 E ( )
Pb
0.555
2
s 2
P1 e . P2
( 0.961
)
. D2.667
G
. Tf . Le .2
The IGT equation proposed by the Institute

of Gas Technology is also known as the IGT
distribution equation and is stated as follows
for USCS units:
(1.61)
Where
than 1.0
C)
= gas viscosity, Poise
Tb
= 136.9 E ( )
Pb
0.555
P12 es . P22
( 0.961
)
G
. Tf . LeZ
. D2.667
(1.60)
Where
(SCFD)
than 1.0
1.2.1.8.14 Effect of Pipe Roughness

In the preceding sections, we used the pipe
roughness as a parameter in the friction
factor and transmission factor calculations.
Both the AGA and Colebrook-White
equations use the pipe roughness, whereas
the Panhandle and Weymouth equations do
not use the pipe roughness directly in the
calculations.
Instead, these equations use a pipeline
efficiency to compensate for the internal
conditions and age of the pipe. Therefore,
when comparing the predicted flow rates or
pressures using the AGA or Colebrook-White
equations with the Panhandle or Weymouth
equations, we can adjust the pipeline
efficiency to correlate with the pipe
roughness used in the former equations.
Since most gas pipelines operate in the
turbulent zone, the laminar flow friction
factor, which is independent of pipe
roughness, is of little interest to us.
Concentrating, therefore, on turbulent flow,
we see that Colebrook-White affected by
variation in pipe internal roughness.

P1 = u P2 = downstream pressure, psia
F)
= gas viscosity, lb/ft-s
In SI units, the IGT equation is expressed as
follows: upstream pressure, psia
48
Thus, the impact of pipe roughness is less in

the smooth pipe zone or for a lower
Reynolds number. A similar comparison can
be made using the AGA equation.
the effect of pipe roughness on the pipeline
flow rate considering the AGA and
Colebrook-White equations. The graph is
based on NPS 20 pipe, 0.500 in. wall
thickness, 120 miles long, with 1200 psig
upstream pressure and 800 psig
downstream
pressure.
The
flowing
temperature of gas is 70F. It can be seen
that as the pipe roughness is increased from
200 to 800 in., the flow rate decreases
from 224 MMSCFD to 206 MMSCFD for the
Colebrook-White equation and from 220
MMSCFD to 196 MMSCFD for the AGA
equation.
1.2.1.8.15 Comparison of Flow Equations

In the preceding sections, we calculated the
flow rates and pressures in gas pipelines
using the various flow equations. Each
equation is slightly different from the other,
and some equations consider the pipeline
efficiency while others use an internal pipe
roughness value. How do these equations
compare when predicting flow rates through
a given pipe size when the upstream or
downstream pressure is held constant?
Obviously, some equations will predict
higher flow rates for the same pressures
than others.
Similarly, if we start with a fixed upstream
pressure in a pipe segment at a given flow
rate, these equations will predict different
downstream pressures. This indicates that
some equations calculate higher pressure
drops for the same flow rate than others.
some of these comparisons when using the
AGA, Colebrook-White, Panhandle, and
Weymouth equations. we consider a
pipeline 100 mi long, NPS 16 with 0.250 in.
wall thickness, operating at a flow rate of 100
MMSCFD. The gas flowing temperature is
80F. With the upstream pressure fixed at
1400 psig, the downstream pressure was
calculated using the different flow
equations. It is clear that the highest
pressure drop is predicted by the Weymouth
equation and the lowest pressure drop is
predicted by the Panhandle B equation. It
must be noted that we used a pipe
roughness of 700 in. For both the AGA and
Colebrook equations, whereas a pipeline
efficiency of 0.95 was used in the Panhandle
and Weymouth equations.
Figure 1.10 Effect of pipe roughness
We can therefore conclude that decreasing

the pipe roughness directly results in a
throughput increase in a pipeline. However,
the cost of internally coating a pipe to reduce
the pipe roughness must be weighed against
the revenue increase due to enhanced flow
rate.
comparison of the flow equations from a

different perspective. In this case, we
calculated the upstream pressure required
for an NPS 30 pipeline, 100 miles long,
49
holding the delivery pressure constant at 800

psig. The upstream pressure required for
various flow rates, ranging from 200 to 600
MMSCFD, was calculated using the five flow
equations.
1.2.3 Regulation and Gas Metering

stations
The main purpose of the reduction station in
the power plants is to maintain the upstream
pressure of the fuel gas compressor constant
as Electric power generation stations
consume fuel directly from the local
reducing stations or by utilizing dedicated
fuel gas compressors to boost the fuel gas
pressure to their needs. Fuel gas compressor
suction pressure needs steady stable
pressure downstream the nearby dedicated
fuel reducing station.
Again it can be seen that the Weymouth

equation predicts the highest upstream
pressure at any flow rate, whereas the
Panhandle A equation calculates the least
pressure. We therefore conclude that the
most conservative flow equation that
predicts the highest pressure drop is the
Weymouth equation and the least
conservative flow equation is Panhandle A
Meter/regulator stations are generally

constructed adjacent to the cleared pipeline
right-of-way (ROW) at each of the receipt
and interconnect points to meter the flow
and adjust the pressure of natural gas
received from or delivered to those systems.
A meter/regulator station typically includes
meter and regulator equipment, a filter
separator, odorant equipment, and a control
building housed within a fenced perimeter.
1.2.2 Compressor Stations:

Compressor station is one of the main
components of the natural gas national grid,
as metering stations, pipes, valves, and
reduction station. Compressor stations have
two main purposes, they can act as a lifting
station
or
booster
station.
Figure 1.11 Compressor Stations
Compressor stations are powered by

compressors that are each rated at several
thousand horsepower (hp). The stations
contain valves, pipes, and control systems
that monitor the functioning and operating
parameters of the system. Most compressor
stations are fully automated.
Figure 1.12 Regulation and Gas Metering stations
The Metering and/or Regulating Stations

reduce and regulate the downstream
pressure, measure the quantity of energy
transferred from the natural gas
transmission system to the medium pressure
Compressor station will be discussed in

details in next chapter.
50
networks or to consumers connected

directly to the transmission system and add
a distinctive odour to the natural gas for
prompt detection of any gas leakage (when
required by the national and international
regulations) For this reason the Metering
and/or Regulating Stations are equipped
with:
downstream. When the regulator is repaired

the slam shut-off valve is to be opened
manually by pulling the knob. Once the flow
starts through the slam shut-off valve it
remains open till the outlet pressure of the
regulator is within limits.
Regulating valves which measure and

regulate continuously the operating
pressure
of
the
distribution
networks;
Slam shut valves for protection of the
downstream systems from any
violation of the operating pressure
limits (for example, after failure of
the regulating valve)
Modern metering systems for the
continuous measurement of the
quantity and the quality of the
supplied natural gas.
Modern automation, remote control,
invoicing and telecommunication
systems; and
Facilities to add the distinctive odour.
Figure 1.13 Slam Shut-Off Valve
1.2.4 City Gate Stations

The natural gas for most distribution systems
is received from transmission pipelines and
fed through one or more city gate stations,
sometimes called town border or tap
stations. The basic function of these stations
is to meter the gas and reduce its pressure
from that of the pipeline to that of the
distribution system. The latter operates at a
much lower pressure (reduced from
approximately 5001,400 psig to about
0.25300 psig). a city gate station that would
typically be covered with a fiberglass
enclosure or metal building to protect it from
the weather.
Again lower network pressure or floating

pressures will lead to shutdown of the
electric power station. Also reduction
station include:
Gas Filtration
Flow metering
Heating of the fuel
Reduction the fuel pressure
1.2.3.1 Slam Shut-Off Valve

Slam shut-off valve is installed immediately
after the filter and prior to the Regulator. It
normally remains open, in case the outlet
pressure of the regulator exceeds the
permissible limits, the slam shut-off valve
senses it through the impulse line and
immediately shut-off the flow to
Figure 1.14 City Gate Stations
51
Most city gate stations measure the gas flow

with metering devices and reduce its
pressure with pressure regulators. These
devices control the rate of gas flow and/or
pressure through the station and maintain
the desired pressure or flow level in the
distribution system. Gas received at city gate
stations may or may not contain odorant, the
compound that gives odorless natural gas its
distinctive smell. Odorant must be added to
the gas if it is received with insufficient or no
odorant before the gas can leave the city
gate station (the odorant concentration is
normally three-quarters of a pound per 1
million cubic feet of natural gas).
maintenance, to direct the fluid from one

location to another, to shut down flow
through pipe sections, and to protect pipe
and prevent loss of fluid in the event of a
rupture. On long-distance pipelines
transporting natural gas and other
compressible fluids, design codes and
regulatory requirements dictate that
sections of pipeline be isolated by installing
mainline block valves at certain fixed
spacing.
1.2.5 Valves
Valves and flow measurements used on gas
pipelines. The design and construction codes
for valves, materials of construction, and
application of the different types of valves
and their performance characteristics will be
explained. The importance of flow
measurement in a gas pipeline and the
accuracy of available instruments, codes,
and standards used will be discussed.
Various American National Standards
Institute (ANSI), American Petroleum
Institute (API), and American Gas Association
(AGA) formulas used in connection with
orifice meters will be reviewed. Since a small
error in mea surement in gas flow in a
pipeline can translate to several thousand
dollars of loss of revenue, it is important that
industry strives to improve upon
measurement methods. Accordingly, gas
transportation companies and related
industries have been researching better
ways to improve flow measurement
accuracy.
Figure 1.15 Mainline valve installation
1.2.5.2 TYPES OF VALVES

The various types of valves used in the gas
pipeline industry include the following:
1.2.5.1 PURPOSE OF VALVES

Valves are installed on pipelines and piping
systems to isolate sections of piping for
52
Gate valve
Ball valve
Plug valve
Butterfly valve
Globe valve
Check valve
Control valve
Relief valve
Pressure regulating valve
Each of these valves listed will be discussed

in detail in the following sections. Valves can
be of screwed design, welded ends, or
flanged ends. In the gas industry, large valves
are generally of the welded type, in which
the valve is attached to the pipe on either
side by a welded joint to prevent gas leakage
to the atmosphere.
in Table, This ratio represents the resistance

of the valve.
The gate valves used in the main lines
carrying oil or gas must be of full bore or
through conduit design to enable smooth
passage of scrapers or pigs used for cleaning
or monitoring pipelines. Such gate valves are
referred to as full bore or through conduit
gate valves.
Figure 1.16 mainline block valve.
1.2.5.2.1 GATE VALVE

A gate valve is generally used to completely
shut off fluid flow or, in the fully open
position, provide full flow in a pipeline. Thus,
it is used either in the fully closed or fully
open position. A gate valve consists of a
valve body, seat, and disc; a spindle; a gland;
and a wheel for operating the valve. The seat
and the gate together perform the function
of shutting off the flow of fluid.
Gate valves are generally not suitable for
regulating flow or pressure or operating in a
partially open condition. For this service, a
plug valve or control valve should be used. It
must be noted that, due to the type of
construction, a gate valve requires many
turns of the hand wheel to completely open
or close the valve. When fully opened, gate
valves offer little resistance to flow, and their
equivalent length to diameter ratio (L/D) is
approximately 8. The equivalent L/Ds for
commonly used valves and fittings are listed
Figure 1.17 Typical gate valve.
53
Table 1.6 the equivalent L/Ds
Figure 1.18 Typical ball valve
1.2.5.2.3 PLUG VALVE

The plug valve traces its origin to the
beginnings of the valve industry. It is a simple
device for shutting off or allowing the flow of
a fluid in a pipe by a simple quarter turn of
the handle. In this sense, it is similar to the
ball valve. Plug valves are generally used in
screwed piping and in small pipe sizes. Plug
valves can be hand wheel operated or
operated using a wrench or gearing
mechanism. The L/D ratio for this type of
valve ranges from 18 to 90, depending upon
the design.
1.2.5.2.2 BALL VALVE

A ball valve consists of a valve body in which
a large sphere with a central hole equal to
the inside diameter of the pipe is mounted.
As the ball is rotated, in the fully open
position the valve provides the through
conduit or full bore required for unrestricted
flow of the fluid and scrapers or pigs.
Compared to a gate valve, a ball valve has
very little resistance to flow in the fully open
position. When fully open, the L/D ratio for a
ball valve is approximately 3.0. The ball
valve, like the gate valve, is generally used in
the fully open or fully closed position.
Unlike a gate valve, a ball valve requires onequarter turn of the hand wheel to go from
the fully open to the fully closed position.
Such quick opening and closing of a ball valve
can be important in some installations where
isolating pipe sections quickly is needed in
the event of emergency.
Figure 1.19 Typical plug valve
1.2.5.2.4 BUTTERFLY VALVE

The butterfly valve was originally used where
a tight closure was not absolutely necessary.
54
However, over the years, this valve has been

manufactured with fairly tight seals made of
rubber or elastomeric materials that provide
good shutoff similar to other types of valves.
Butterfly valves are used where space is
limited. Unlike gate valves, butterfly valves
can be used for throttling or regulating flow,
as well as in the full open and fully closed
positions. The pressure loss through a
butterfly valve is higher in comparison with
the gate valve. The L/D ratio for this type of
valve is approximately 45. Butterfly valves
are used in large and small sizes. They can be
hand wheel operated or operated using a
wrench or gearing mechanism.
open position. This is due to the fact that the

flow of fluid changes direction as it goes
through the valve. The L/ D ratio for this type
of valve is approximately 340. Globe valves
are manufactured in sizes up to NPS 16. They
are generally hand wheel operated.
Figure 1.21 Typical globe valve
1.2.5.2.6 CHECK VALVE

Check valves are normally in the closed
position and are open when the fluid flows
through them. They also have the capability
of shutting off the flow in the event the
pressure downstream exceeds the upstream
pressure. In this respect, they are used for
flow in one direction only. Thus, they
prevent back flow through the valve. Since
flow of the fluid through these valves is
allowed to be in one direction only, check
valves must be installed properly by noting
the normal direction of flow. An arrow
stamped on the outside of the valve body
indicates the direction of flow. Check valves
can be classified as swing check valves and
lift check valves. The L/D ratios for check
Figure 1.20 Typical butterfly valve
1.2.5.2.5 GLOBE VALVE

Globe valves, so called because of their
outside shape, are widely used in plant
piping. They are suitable for manual and
automatic operation. Unlike the gate valve,
globe valves can be used for regulating flow
or pressures as well as complete shutoff of
flow. They can also be used as pressure relief
valves or as check valves. Compared to a gate
valve or ball valve, the globe valve has
considerably higher pressure loss in the fully
55
valves range from 50 for the swing check

valve to as high as 600 for lift check valves.
to the lateral pipeline is 800 psig, but the
Figure 1.23 PRESSURE CONTROL VALVE
customers piping is limited to 600 psig, a

pressure regulator is used to reduce the
pressure by 200 psig.
Figure 1.22 Typical check valves
1.2.5.2.7 PRESSURE CONTROL VALVE

A pressure control valve is used to
automatically control the pressure at a
certain point in a pipeline. In this respect it is
similar to a pressure regulator discussed
next. Whereas the pressure regulator is
generally used to maintain a constant
downstream pressure, a pressure control
valve is used to control the upstream
pressure. The upstream and downstream are
relative to the location of the valve on the
pipeline. Generally, a bypass piping system
around the control valve is installed to
isolate the control valve in the event of an
emergency or for maintenance work on the
control valve.
Figure 1.24 Pressure regulator
1.2.6 Flow Measurements
1.2.5.3 PRESSURE REGULATOR
Gas flow measurement in a pipeline is

necessary for properly accounting for the
amount of gas transported from one point to
another along a gas pipeline. The owner of
the gas and the customer who purchases the
gas both require that the correct amount of
gas be delivered for the agreed-upon price.
A pressure regulator is a valve that is similar

to a pressure control valve. Its function is to
control or regulate the pressure in a certain
section of a pipeline system. For example, on
a lateral piping that comes off a main
pipeline, used for delivering gas to a
customer, a lower pressure might be
required on the customer side. If the main
pipeline pressure at the point of connection
56
Figure 1.25 Flow Measurements

Figure 1.27 Different types of orifice meters
1.2.6.1 Orifice Meter

The orifice meter is a flat steel plate that has
a concentric machined hole with a sharp
edge and is positioned inside the pipe.
1.2.6.2 VENTURI METER

The venturi meter is based upon Bernoullis
equation. It consists of a smooth gradual
contraction from the main pipe size to a
reduced section known as the throat, finally
expanding back gradually to the original pipe
diameter.
Figure 1.26 Orifice meter
As the gas flows through the pipeline and

then through the orifice plate, due to the
reduction in cross-sectional area as the gas
approaches the orifice, the velocity of flow
increases and, correspondingly, the pressure
drops. After the orifice, the cross-sectional
area increases again back to the full pipe
diameter, which results in expansion of gas
and decrease in flow velocity. This process of
accelerating flow through the orifice and
subsequent expansion forms a vena
contracta, or a throat, immediately past the
orifice.
Figure 1.28 Venturi meter.
The gas pressure in the main pipe section is

represented by P1and that at the throat is
represented by P2. As gas flows through a
venturi meter, it increases in flow velocity in
the narrow throat section. Correspondingly,
the pressure reduces in the throat section
according to Bernoullis equation. After gas
leaves the throat section, it reduces in flow
velocity due to the increase in pipe crosssectional area, and it reaches the original
flow velocity.
57
Figure 1.29 Operation of Venturi meter

Figure 1.30 Flow nozzle
1.2.6.3 FLOW NOZZLE
1.2.6.4 Ultrasonic flow meters
The flow nozzle is another device for

measuring flow rate. It consists of a main
pipe section, followed by a gradual reduction
in cross-section area and a short cylindrical
section, ending in a gradual expansion to the
original pipe size. The discharge coefficient C
for a flow nozzle is approximately 0.99 for
Reynolds numbers greater than 106. At lower
Reynolds numbers, due to greater energy
loss subsequent to the nozzle throat, C
values are lower.
During the past several years, the use of

ultrasonic flow meters for natural gas
custody transfer applications has grown
significantly. The publication of AGA Report
No. 9, Measurement of Gas by Multipath
Ultrasonic Meters [Ref 1] in June 1998, has
further accelerated the installation of
ultrasonic flow meters (USMs). Today
virtually every transmission and many
distribution companies are using this
technology fiscal or for operational
applications.
The discharge coefficient C depends on the

beta ratio and Reynolds number. It is
calculated using the following equation;
Since the mid-1990s the installed base of

USMs has grown by approximately 50% per
year. There are many reasons why ultrasonic
metering is enjoying such healthy sales.
Some of the benefits of this technology
include the following:
58
Accuracy: Can be calibrated to

<0.1%.
Large Turndown: Typically >50:1.
Naturally Bi-directional: Measures
volumes in both directions with
comparable performance.
Tolerant of Wet Gas: Important for
production applications.
Non-Intrusive: No pressure drop.
the velocity and the cross-sectional area,

uncorrected volume can be computed. Let us
review the equations needed to compute
flow.
Low Maintenance: No moving parts

means reduced maintenance.
Fault Tolerance: Meters remain
relatively accurate even if sensor(s)
should fail.
Integral Diagnostics: Data for
determining a meters health is
readily available.
The transit time (T12) of an ultrasonic signal

traveling with the flow is measured from
Transducer 1 to Transducer 2. When this
measurement is completed, the transit time
(T21) of an ultrasonic signal traveling against
the flow is measured (from Transducer 2 to
Transducer 1). The transit time of the signal
traveling with the flow will be less than that
of the signal traveling against the flow due to
the velocity of the gas within the meter.
It is clear that there are many benefits to

using USMs. Although the first several
benefits are important, the most significant
may turn out to be the ability to diagnose the
meters health. The primary purpose of this
paper is to discuss basic gas ultrasonic meter
operation,
diagnostics,
review
the
fundamentals of field maintenance, discuss
some test results and provide the reader
with an examples of good and not-so-good
piping designs.
1.2.7 Remote Control System &

Telecommunications
The Remote Control and Communications
(RCC) system has primary target:
to ensure that the National Natural Gas
Transmission System will supply the natural
gas market with sufficient quantities in all
foreseeable conditions, normal and
abnormal
And secondary targets:
The remote supervision of the natural gas
stations, the management of the alarms and
the acquisition of technical and statistical
data to the departments of DESFA such that
the transmission of natural gas is performed
in a safe, reliable and cost-effective manner
1.2.6.4.1 ULTRASONIC METER BASICS

The basic operation of an ultrasonic meter is
relatively simple. Even though there are
several designs of ultrasonic meters on the
market today, the principle of operation
remains the same.
It consists of the following subsystems:
Figure 1.31 ULTRASONIC METER
Ultrasonic meters are velocity meters by

nature. That is, they measure the velocity of
the gas within the meter body. By knowing
59
Fiber Optic Cable (FOC), installed

parallel to the high pressure gas
pipeline, being the channel of all
internal
voice
and
data
communications of DESFA;
Time Division Multiplexing (TDM)
system, installed in every station of
the
National
Natural
Gas
Transmission System and used for
the communication of voice and data

through the fiber optic cable;
Supervisory Control and Data
Acquisition (SCADA) system, used for
the remote supervision and control
of all the Metering and/or Regulating
Stations, the Line Valve Stations and
the Remote Communication Stations
from the Gas Control and Dispatching
Centers;
Private Automatic Branch exchange
(PABX) network, installed in the
Operation and Maintenance Centers,
the LNG and the Border Metering
Stations; and
Land Mobile Radio (LMR) privatelyowned system, providing radio voice
communication
between
maintenance field personnel and
additionally between shift operators
of the Gas Control and Dispatching
Centers, covering an area of 10 km at
both sides of the high pressure gas
pipeline.
Figure 1.32 Remote Control System
60
Chapter 2| Power to Gas Application
Chapter 2
Power to Gas
61
generating potential of 73.6 Peta-watt

hour (PWh).
2.1 Introduction:
2.1.1 What is the problem of energy
in Egypt?
In Egypt, as everywhere, energy plays a
substantial role in countrys economic
development.
On the other hand, Egypt as a developing
fast growing country suffers from rapid
annual population growth currently at a
rate of 1.68%. Egypts population is
expected to reach 110 million by 2030 and
128 million by 2050.
Such a fast population growth along with
other environmental challenges is overstraining the limited energy resources of
the country.
Figure 2.1 Annual Direct Normal Irradiation in Egypt
A huge increase in natural gas

consumption compared with production,
that is affect network line packing and gas
network pressure begins to decrease.
All that renewable sources should be

utilized to overcome the problem of Egypt.
Hence, the researchers investigate about a
new renewable source of substitute
Natural Gas, which is Power to Gas (PtG).
So getting a renewable source of natural

gas is a must instead of importing from
other countries.
2.1.3 Power to Gas:

Power-to-Gas (PtG) is one of many
flexibility options that can provide
flexibility on the demand side by
converting electricity into gas during times
of high renewable energy sources (RES)
supply.
This renewable gas is a versatile energy
carrier that can be used in the electricity,
heating or transport sector as well as for
industrial processes.
2.1.2 energy sources available in

Egypt:
Egypt has many power sources like: solar
power plants and wind power plants.
Egypt has many water sources like: the
Nile, Mediterranean-sea and Red sea.
Egypt has many sources of CO2 like: wells
and biomass.
Egypt belongs to the global sun-belt. The
country's geographical position is
advantageous
with
solar
energy.
In 1991 solar atlas for Egypt indicated that
the country enjoys 2900-3200 hours of
sunshine annually with annual direct
normal energy density 1970-3200 kWh/m2
and technical solar-thermal electricity
Thus, the use of renewable gas could help

to reduce greenhouse gas emissions in
these sectors.
Power to gas describes the process of
converting electricity into hydrogen via
electrolysis and an optional further
62
conversion step into synthetic natural gas

via methanation.
2.2 What is PtG?

2.2.1 Definition:
Global warming, local air pollution and

external energy dependency have
intensied the development of renewable
and clean energy sources and fuels along
last decades.
Power to gas or P2G is an energy

conversion and it also considered an
energy storage technology. Surplus power
converted into substitute natural gas.
This process takes several steps:
Singular efforts have been made in the use

of hydrogen as energy carrier and energy
storage. Hydrogen is a clean fuel able to
reduce local CO2 and pollutants emissions.
However, to be a true zero-emission
energy carrier, hydrogen must be
produced through low-carbon processes.
1- Getting electricity from solar plants.

2- Using electricity in separating water
into hydrogen and oxygen by the
electrolysis.
3- Blending hydrogen in national natural
gas
grid.
OR using any source of CO2 to make a
methanation process and get synthetic
natural gas and blend it into the gas
grid.
Electrolysis systems have a great potential

for CO2-free hydrogen production due to
the simplicity of the process and because it
can be easily connected with renewable
electrical energy systems like wind and
solar.
Electricity cannot be stored easily but

hydrogen can be stored easily in the gas
grid.
Due to the fluctuation phenomenon, an
energy storage is needed to compensate
the low and high consumption energy.
More natural gas is needed so we try to get
a synthetic natural gas which used in
several industries and mainly used in
power stations. Power station efficiency
can be increased by increasing the calorific
value of the fuel by bleeding hydrogen in
the gas network.
Additionally, hydrogen production through

electrolysis creates a bridge between
power and fuel networks. Thus, hydrogen
could be used as a temporary energy
storage contributing to grid stabilization,
so the penetration of renewable energies
can be increased.
One major drawback to this approach is
the significant energy loss involved. The
conversion of electricity into methane
occurs with an efficiency of only 60
percent. If the methane is later used in a
natural gas power plant to produce
electricity, the efficiency falls to 36
percent.
2.2.2 Advantages of P2G:

P2G is expected to play an increasingly vital
role in the way energy systems develop.
The issue with renewable energy sources
like solar and wind power is intermittency.
With the P2G concept, temporal surpluses
can be reduced by converting renewable
energy into gas. Another point that speaks
for the P2G concept is that the gas
infrastructure is able to accommodate
63
larger volumes of electricity converted into

gas than what the grid alone can support.
then chemically converting it to

methaneor synthetic natural gasfor
storage in existing gas grids.
Hence, the advantages of PtG are:
So, Power to Gas technology is based on

many components which are solar system
(Photovoltaics to get electricity or/and
parabolic trough collectors to get steam
and electricity) to obtain heat and
electrical energies which are needed in
getting fresh water from sea water used in
desalination and in hydrogen production
process via electrolysis, respectively and
finally, blending that hydrogen produced in
natural gas grid or using it with Carbon
Dioxide which obtained from such a source
to get substitute natural gas which is
blended into the gas grid.
Overcome the importation of natural

gas.
Provide big storage of natural gas and
hydrogen.
Use renewable source through wind or
solar.
Can use the existing infrastructure to
store the gas.
Reduce the Greenhouse Gas emissions
2.2.3 Components of P2G plant

Power-to-gas, a fairly new type of energy
storage, involves converting renewable
power to hydrogen, using electrolysis, and
Figure 2.2 Power to gas technology component
2.3.1 Solar system in Egypt
2.3 Solar systems
Recently, world has become more

conscious about the environment, changes
occurring in climate and in earth in general.
It started to pay more attention to the
impact of the technological and industrial
revolution on the ecology and human
health, which resulted in directing all the
new researches towards renewable energy
and recycling materials.
Solar power is the conversion of sunlight

into electricity directly or indirectly.
Solar power plants are divided into two
types:
Photovoltaic plants.
Concentrating solar power plants (CSP)
(Parabolic trough).
64
In the development of energy sources in

Egypt of the 21st century, it is necessary to
view the use of solar energy in all
applications as one of the most promising
new and renewable energy sources.
about 200 GWh/year (thermal) of solar

heat. Parabolic trough collectors
spread over an area of 624,438 square
meters.
KURAYMAT operates at night as a
combined cycle gas power plant
producing roughly 110 MW of natural
gas-fueled energy. During the day, a
1,900 square meter solar thermal field
filled with parabolic trough technology
converts energy captured from the sun
into steam-powered electricity.
It is found that providing electricity to a

family house in a rural zone using
photovoltaic systems are very beneficial
and competitive with the other types of
conventional energy sources, especially
considering the decreasing prices of these
systems and their increasing efficiencies
and reliability. They have also the
advantage of maintaining a clean
environment.
The solar field has a capacity of 150

MW during peak hours. This plant is
supplied with water from the Nile.
On average the KURAYMAT site
provides more than 2,400 kilowatt
hours of solar irradiation per square
meter per year. The plant supplies
nearly 2 million people with power.
In Egypt, there are many new projects such

as those carried out in the new valley in the
western desert of Egypt and those on the
northern coast of Egypt, which will be
accompanied by a new population who
require energy for principle life
requirements.
2.3.2 History
The first solar power plant in Egypt was
built
in
MAADI
in
1913!
it is not a hoax because if you check the
history of the solar energy you will find that
solar energy is much older than we think, it
is not a 21th century idea after. It is 20th
century idea.
2.3.3 Places of solar plants in Egypt

1) KURAYMAT
it comprises two gas turbines of about
40 MWe each, one steam turbine of
about 70 MWe and one parabolic
trough solar field capable to generate
Figure 2.3 Kuraymat Integrated Solar Combined

Cycle Power Plant
65
Figure 2.4 Combined cycle power plant with a solar field
2) NORTH COAST
running from 2012-2017 designed to

establish the Egypt as one of the top
generators of solar energy in North
Africa.
Solar water heating is currently used in

residential commercial and tourist
hotel buildings, especially those
located in touristy resorts in Sinai,
along the red sea coast and Egypts
North Coast (south Mediterranean
coast).
This plant will be a 200 MW solar

thermal
(concentrated
parabolic
trough) power plant with (4 6) hour
thermal storage and will be composed
of:
3) TIBA-QENA
1. Solar field Consist of concentrated

solar mirrors arrays (parabolic
trough) with total land area 3.5
k2 .
2. Heat transfer fluid loops (oil loop)
The temperature of special oil in
the receiver is about 400 C and
pressure 12 bar, then the oil is
pumped to heat exchanger to
generate superheated steam, this
steam is pushed to the steam
turbine to generate electricity.
3. Thermal storage By using molten
salt as storage material with tank
Along 4km of the TIBA-QENA desert

road. The launch comes besides the
launching of a solar energy plant with
capacity of 90 KW in new TIBA city.
4) KOM OMBO
The solar power project at KOM
OMBO, near the ASWAN high dam
hydro-electric plant, will be financed by
a number of international institutions,
including the African Development
Fund
and
the
World
Bank.
The project is part of a five-year plan
66
technology (hot & cold), we can

store solar thermal energy during
the day in the hot tank and then we
can use thermal heat from the hot
tank to generate enough steam to
generate electricity at night from 46 hours to share in cover the peak
load at night.
4. Power blocks It consists of steam
turbine,
Mechanical
parts,
electrical parts and the generator.
The power block is fixed in building
with 20 m maximum Height for
generated unit of 200 MW. The
electric energy generated will be
385 Million KWh annually, saving
about 90,000 Ton Oil Equivalent
(T.O.E) and reduce emission about
200,000 tons C2 .
in newly reclaimed lands, in rural

electrification, refrigeration, etc.
Figure 2.5 PV Solar panels field
The process of energy conversion is clean

and simple and it does not involve injecting
any kind of smoke or harmful chemicals
into the atmosphere. The modern form of
the solar cell was invented in 1954 at Bell
Telephone Laboratories. Today, PV is one
of the fastest growing renewable energy
technologies.
2.3.4 Types of solar system

2.3.4.1 Photovoltaic (PV)
System specific factors such as the local

irradiance, weather, soil, wind, and
topography must be taken into account for
the design. A lot of companies have
decided to have grids containing
photovoltaic cells installed on their
premises so that they can derive the
maximum amount of energy from them
and also reduce their dependence on the
traditional forms of electrical energy.
Photovoltaic, also called solar cell, are

electronic devices that convert sunlight
directly into electricity.
The electricity they produce is DC (direct
current) and can either be used directly as
DC power; converted to AC (alternating
current) power; or stored for later use. The
basic element of a photovoltaic system is
the solar cell that is made of a
semiconductor material, typically silicon.
2.3.4.1.1 Components of
photovoltaic plant
1) Solar cell: is the basic component of
any photovoltaic, or PV, system. A
device like a calculator requires only a
single cell, but most devices or
commercial
applications
needs
multiple cells.
A solar cell is a solid-state electrical
device (p-n junction) that converts the
energy of light directly into electricity
(DC) using the photovoltaic effect.
PV systems are considered one of the most

appropriate applications for remote areas
away from national grid and targeted to be
applicable over all Egypt. According to New
and Renewable Energy Authority (NREA)
photovoltaic technologies are in use for
lighting, commercial advertisements,
wireless communications and cell phone
networks, in water pumping for irrigation
67
The process of conversion first requires a

material which absorbs the solar energy
(photon), and then raises an electron to a
higher energy state, and then the flow of
this high-energy electron to an external
circuit. Solar cells are most commonly
made of silicon, and come in two common
varieties, crystalline and thin-film cells.
charge on the p-side. The negative charge

on the p-side restrict the movement of the
additional electrons from the n-side to the
p-side, while the movement of the
additional electrons from the p-side to the
n-side is made easier because of the
positive charge at the junction on the nside. This restriction makes the p-n
junction behave like a diode. This diode
character of a p-n junction is made use of
in solar photovoltaic cells.
(p-n) junctions:
An n-type material has some impurity

atoms with more electrons than the rest of
the semi-conductor ions. If those excess
electrons are removed, the impurity atoms
will fit more uniformly in the structure
formed by the main semi-conductor
atoms, however the atoms will be left with
positive charges. P-type material has some
impurity atoms with fewer electrons than
the rest of the semi-conductor atoms.
Therefore, these atoms seem to have holes
that could accommodate excess electrons
even though the atoms are electrically
neutral. If additional electrons could be
brought to fill the holes, the impurity
atoms would fit more uniformly in the
structure formed by the main semiconductor atoms however the atoms will
be negatively charged.
Depletion region:
A depletion region forms instantaneously
across a p-n junction. Electrons and holes
diffuse into regions with lower
concentrations of electrons and holes,
much as ink diffuses into water until it is
uniformly distributed.
By definition, N-type semi-conductor has
an excess of free electrons compared to
the P-type region, and P-type has an excess
of holes compared to the N-type region.
When N-doped and P-doped pieces of
semiconductor are placed together to
form a junction, electrons migrate into the
P-side and holes migrate into the N-side.
Departure of an electron from the N-side
to the P-side leaves a positive donor ion
behind on the N-side, and likewise the hole
leaves a negative acceptor ion on the Pside.
Figure 2.6 The (p-n) junction

Figure 2.7 Depletion region
As soon as the two materials are joined,

excess electrons from the n-layer jump
to fill the holes in the p-layer. Therefore,
close to the junction, the material has
positive charge on the n-side and negative
68
Photovoltaic effect:
When a photon of light is absorbed by
a valence electron of an atom, the
energy of the electron is increased by
the amount of energy of the photon. If
the energy of the photon is equal to the

band gap of the semi-conductors, the
electron with excess energy will jump
into the conduction band where it can
move freely. If, however, the photon
energy is less than the band gap, the
electron will not have sufficient energy
to jump into the conduction band. If
the absorbed photon had more energy
than the band gap, in this case, the
excess energy of the electrons is
converted to excess kinetic energy of
the electron which manifests in
increased temperature.
Figure 2.8 Moving the electrons from n-type to p-type

through an external circuit
If the n-layer is made extremely thin,

the movement of the electrons and the
probability of recombination within
the n-layer is greatly reduced unless
the external circuit is open.
It must be noted that a photon can free

up only one electron even if the photon
energy is a lot higher than the band
gap. This fact is a big reason for the low
conversion efficiency of photovoltaic
device. The key to using the
photovoltaic effect for generating
useful power is to channel the free
electrons through an external
resistance before the recombine with
the holes. This is achieved with the help
of the p-n junction.
If the external circuit is open, the

electrons generated by the action of
photons eventually recombine with the
holes, resulting in an increase in the
temperature of the device.
2) Module
Multiple electrically connected solar
cells that are mounted in a support
structure, or frame, make up a
photovoltaic
module.
Multiple
modules combined together are
known as panels. Modules are
designed to supply electricity at a
certain voltage, like the common 12
volt system, for example. The current
produced depends directly on how
much light strikes the module.
3) Array
As free electrons are generated in the

n-layer by the action of photons, they
can either pass through an external
circuit, recombine with positive holes
in the lateral direction or move toward
p-layer. This negative charges in the player at the p-n junction restrict their
movement in that direction.
Solar farms are spread across a large

area. Building a single panel to span an
area is economically, technologically,
and operationally sub-optimal. For that
reason, multiple panels are connected
together to form an array. Cells,
modules, and arrays produce direct
current.
69
Figure2.9 Arrangement of solar system

2.3.4.1.2 Surface area
requirements and latitude
Open space system modules are usually
installed at a flat angle (20 to 30 degrees)
to the earths surface in order to capture as
much sunlight as possible. This orientation
is a compromise, because they should be
more vertical in the morning, and flatter at
noon.
The greater the angle, the more distance is
required between each module row, so
that no or minimal self-induced shadowing
occurs. In order to use the available space
as best as possible, the module rows must
be close together. On the other hand, they
should have the greatest possible distance
from one to another, in order to ensure
that they do not cast a shadow in the
mornings and evenings.
4) Tracking system
To increase solar yield and electricity
production, solar trackers and
concentrator systems are sometimes
used. Trackers direct solar panels or
modules toward the sun.
These devices change their orientation
throughout the day to follow the suns
path to maximize energy capture.
In photovoltaic systems, trackers help
to minimize the angle of incidence (the
angle that a ray of light makes with a
line perpendicular to the surface)
between the incoming light and the
panel, which increases the amount of
energy the installation produces.
A compromise must be found between

these two requirements, which can be
determined by using an optimization
formula, and which depends upon the
latitude.
The angle is largest in northern Europe,
and lowest at the Equator; there it makes
theoretically the most sense to install the
modules closest to one another at the
flattest angle.
Figure 2.10 Conventional active PV panel tracking system
70
2.3.4.1.3 PV technologies
There are a wide range of PV cell
technologies on the market today, using
different types of materials, and an even
larger number will be available in the
future.
PV cell technologies are usually classified
into three generations, depending on the
basic material used and the level of
commercial maturity:
wafer-based c-Si PV cells and modules

dominate the current market. Trina Solar
(TSL), Yingli Green Energy (YGE), and
Canadian Solar (CSIQ) are some of the
largest North American producers of
crystalline cells.
The efficiency of crystalline silicon modules
ranges from 14% to 19%. While a mature
technology, continued cost reductions are
possible through improvements in
materials and manufacturing processes.
Commercial production of c-Si modules
began in 1963 when Sharp Corporation of
Japan started producing commercial PV
modules and installed a 242 Watt (W) PV
module on a lighthouse, the worlds largest
commercial PV installation at the time.
Crystalline silicon technologies accounted
for about 87% of global PV sales in 2010.
- First-generation PV systems (fully

commercial) use the wafer-based
crystalline silicon (c-Si) technology, either
single crystalline (sc-Si) (mono crystalline)
or
multi-crystalline
(mc-Si)(poly
crystalline).
Silicon is one of the most abundant
elements in the earths crust. It is a
semiconductor material suitable for PV
applications.
- Second-generation PV systems (early

market deployment) are based on thin-film
PV technologies and generally include
three main families:
1) amorphous (a-Si) and micro morph
silicon (a-Si/c-Si).
-Amorphous silicon PV module efficiencies
are in the range 4% to 8%.
The main disadvantage of amorphous
silicon solar cells is that they suffer from a
significant reduction in power output over
time (15% to 35%), as the sun degrades
their performance.
Figure 2.11 Mono-Crystalline PV panel
Figure 2.12 Poly-Crystalline PV panel
Crystalline silicon is the material most

commonly used in the PV industry, and
71
module efficiencies are in the range of

7% to 16%, but efficiencies of up to
20.3% have been achieved in the
laboratory.
Figure 2.13 Thin Film solar panel (amorphous silicon)
2) Cadmium-Telluride (CdTe).
Figure
(CIGS)
They have lower production costs and

higher cell efficiencies (up to 16.7%
[Green, 2011]) than other thin-film
technologies. Cadmium is a by-product of
zinc mining and tellurium is a by-product of
copper processing. A potential problem is
that tellurium is produced in far lower
quantities than cadmium.
2.15 Copper-Indium
Gallium-Diselenide
Thin-film solar cells could potentially

provide lower cost electricity than c-Si
wafer-based solar cells. Thin-film solar cells
are comprised of successive thin layers,
just 1 to 4 m thick. They require a lot less
semiconductor material to manufacture in
order to absorb the same amount of
sunlight (up to 99% less material than
crystalline solar cells). Hanergy and First
Solar (FSLR) are leaders in thin-film cell
production.
- Third-generation PV systems include
technologies, such as concentrating PV
(CPV) and organic PV cells that are still
under demonstration or have not yet been
widely commercialized.
Figure 2.14 Cadmium Telluride PV panel
3) Copper Indium-Selenide (CIS) and

Copper-Indium Gallium Diselenide
(CIGS).
They offer the highest efficiencies of all
thin-film PV technologies. Current
72
Table 2.1 Comparison between cell material types
Table 2.2 Comparison between 1st and 2nd generations
Table 2.3.Prices of variety of PV technologies
73
2.3.4.1.4 How to clean PV panels?

It really depends on weather conditions.
Solar panels installed in windy and dusty
areas may require more frequent
inspection and cleaning. Some areas need
a weekly washing, while other areas can go
every other month.
Solar panel conversion efficiency, typically

in the 20 % range, is reduced by dust,
grime, pollen, and other particulates that
accumulate on the solar panel. "A dirty
solar panel can reduce its power
capabilities by up to 30% in high
dust/pollen or desert areas" <says Seamus
Curran, associate professor of physics at
the University of Houston and director of
the Institute for Nano Energy>.
2.3.4.1.5 Advantages of PV
system
1. PV panels provide clean, green energy.
During electricity generation with PV
panels there is no harmful greenhouse
gas emissions thus solar PV is
environmentally friendly.
2. Solar energy is energy supplied by
nature, so its free and abundant. Solar
energy can be made anywhere there is
sunlight.
3. Solar Panels cost is currently on a fast
reducing track and is expected to
continue reducing for the next years.
Solar PV panels has a highly promising
future both for economic viability and
environmental sustainability.
4. Compared to other renewable green
energy technologies, PV systems
require minimum maintenance; only
minor checking of cable connections
and a basic-regular cleaning of the
panel surfaces is adequate to keep
them operational for several years.
5. PV panels have no mechanically
moving parts, except in cases of
sun-tracking
mechanical
bases;
consequently they have far less
breakages or require less maintenance
than other renewable energy systems
(e.g. wind turbines).
6. PV panels are totally silent, producing
no noise at all; consequently, they are
- Methods of cleaning:
Solar Panel Cleaning by yourself.

Use a Solar Panel Cleaning Service.
Use a Cleaning Robot.
Automated Self-Cleaning Technology.
- cleaning warning:
Use a non-abrasive cleaning sponge or

towel to wash the glass. Get the glass
clean and clear as possible dont
scratch it.
Remove remaining water with a high
quality squeegee to avoid hard water
mineral build up.
Use a lint free cloth to dry the edges of
the panel.
Each solar panel is different, so make

sure to review the instructions from
the solar manufacturer first before
cleaning the panels; some panels are
sensitive to certain chemicals, and
others may have fragile components
that you need to be aware of.
Inspect the solar panels on a periodic
basis. Before washing, make sure that
there arent any cracks and ensure all
connections are tight. If there are
damaged wires or cracks, it might be
time to call an electrical technician.
Remove any debris on the solar panel.
Trim back trees or plants that may be
blocking the suns energy to the solar
panels.
74
a perfect solution for urban areas and

for residential applications.
7. An important advantage of PV systems,
inherited to all solar power
technologies, is that they have peak
production of energy when energy
demand is at maximum levels (mainly
during the summer). This attribute
facilitates towards peak-shaving and
smoothing of the load curve thus
reduce the possibility of a power blackout.
8. Residential solar panels are easy to
install on rooftops or on the ground
without any interference to residential
lifestyle.
2.3.4.1.6 Disadvantages of PV
system
1. As in all renewable energy sources,
solar energy has intermittency issues;
not shining at night but also during
daytime there may be cloudy or rainy
weather. Consequently, intermittency
and unpredictability of solar energy
makes solar energy panels a less
reliable solution.
2. Solar energy panels require additional
equipment (inverters) to convert direct
electricity (DC) to alternating electricity
(AC) in order to be used on the power
network.
But it is not needed in the case of
power to gas.
3. For a continuous supply of electric
power,
especially
for
on-grid
connections, Photovoltaic panels
require not only Inverters but also
storage batteries; thus considerably
increasing the investment cost for PV
panels.
and also the batteries are not needed
in the case of PtG as the gas network is
4.
5.
6.
7.
considered as a storage of the

electricity in the form of gas.
In case of land-mounted PV panel
installations, they require relatively
large areas for deployment; usually the
land space is committed for this
purpose for a period of 15-20 years or
even longer.
Solar panels efficiency levels are
relatively low (between 14%-25%)
compared to the efficiency levels of
other renewable energy systems.
Though PV
panels have no
considerable
maintenance
or
operating costs, they are fragile and
can be damaged relatively easily;
additional insurance costs are
therefore of ultimate importance to
safeguard a PV investment.
the basic disadvantage of solar
photovoltaic systems is their cost
which remains relatively high. Of
course this is relative to the
alternatives that PV systems are being
compared
to.
For example, the electrical energy
produced by a photovoltaic system is
estimated to cost a lot more than the
cost of energy produced by a system of
Wind
turbines
or
from
biomass. Similarly, the cost of PV
energy is a lot more than that of energy
produced from conventional, nonrenewable energy sources (fossil fuels),
such as oil or natural gas.
2.3.4.1.7 Photovoltaics
calculations:
Power calculations
The global formula to estimate the
electricity generated in output of a
photovoltaic system is:
E = A. r. H. PR
Where,
75
E: output energy from PV system

(KWh)
A: Total solar panel area (m2)
TCF: Temperature correction factor (If

the cell temperature is assumed to
reach 60 , then the temperature
correction factor (TCF) will be 0.8)
r: Solar panel yield (%)
To calculate the number of PV panels

for
the
system:
given by the ratio of electrical power

(in kWp) of one solar panel divided by
the area of one panel.
( )
= (
H: Annual average solar radiation on

tilted panels (shadings not included)
2.3.4.2 Parabolic trough concentrators
PR: Performance ratio, coefficient for

losses (range between 0.5 and 0.9
default value = 0.75)
A parabolic trough collector (PTC) is made

by bending a sheet of reflective material
into
a
parabolic
shape.
A metal black pipe, covered with a glass
tube to reduce heat losses, is placed along
the focal line of the receiver.
To calculate the total power of

modules from the power needed from
PV panels:
total power of modules
= power needed from pv panel lost
When the parabola is pointed towards the

sun, the parallel rays incident on the
reflector are reflected and focused onto
the receiver tube. The concentrated
radiation reaching the receiver tube heats
the fluid that circulates through it, thus
transforming the solar radiation into useful
heat.
The Power in (watt.hour/day) energy

lost in the system = 1.3
to calculate the total Watt-peak rating
needed
for
PV
modules:
( ) =
Where, panel generation factor = 3.43
PV array sizing:
To calculate the PV area:
PV Area =
Where,
EL: Average
(KWh/day)
daily
load
energy
Figure 2.16 Concentrated parabolic trough collector
Concentration is specified by the

concentration ratio (CR), the concentration
ratio of a concentration collector is the
H: Average solar energy input /day

(KWh/m2/day)
: photovoltaics panel efficiency
76
ratio of area of sunlight intercepted to the

area onto which it is concentrated.
aperture always faces the sun at noon. A

dedicated computer-operated tracking
mechanism is responsible for keeping the
collector focused at all times.
The receiver of a parabolic trough collector

is linear. Usually a tube is placed along the
focal line to form an external surface
receiver. The surface of the receiver is
typically plated with selective coating that
has a high absorbance for solar irradiation
but a low emittance for thermal radiation.
A glass cover tube is usually placed around
the receiver tube to reduce the convective
heat loss from the receiver, thereby
further reducing the heat loss coefficient.
The collector fluid used in the tests is

water. A high temperature heat transfer
fluid (HTF) such as synthetic oil with a
relatively low freezing point, low vapor
pressure, and the capability to retain heat
absorbs the thermal energy in the HCE as it
flows through the receiver tubes.
In one system the collector can be washed
automatically thus reducing drastically the
maintenance cost, which is the mostly
used process required.
A disadvantage, resulting from the use of

the glass cover tube, is that the reflected
light from the concentrator must pass
through the glass to reach the receiver,
adding a transmittance loss of about 0.9,
when the glass is clean. The glass envelope
usually has an anti-reflective coating to
improve transmissivity.
2.3.4.2.1. CSP plant components

A Solar Collector Assembly (SCA) is
primarily composed of parabolic mirror
facets with backing plates, heat collection
elements with supports, a hydraulic
actuator drive, and a sun tracking sensor;
all supported by a metal support structure.
The interconnecting piping system (except
ball joints) and collector foundations are
typically not considered part of the SCA
manufacturers supply; but are engineered
and constructed by the general contractor.
One way to further reduce convective heat

loss from the receiver tube and thereby
increase the performance of the collector,
particularly
for
high
temperature
applications, is to evacuate the space
between the glass cover tube and the
receiver.
The collector field is made up of multiple

rows of SCAs, which are the basic, modular
building blocks of the solar energy
collection system. The solar collectors will
rotate slowly east to west during each day
to track the suns movement and to keep
the suns rays focused on the absorber
tube. This reference plant uses hydraulic
actuators which are driven by a ~1 hp
(0.746 kW) electric pump motor.
The total receiver tube length of PTCs is

usually from 25 m to 150 m. The collector
installed has a length of 12.2 m and
consists of galvanized steel mounts,
lightweight, stiff and precise parabolic
reflector panels manufactured from
reinforced
polymeric
material,
a
structurally efficient galvanized steel
torque tube, a tubular receiver and an
accurate solar tracking system.
The advantages of this tracking mode are
that very little collector adjustment is
required during the day and the full
77
1. Heat Collection Element (HCE).

The heat collection element absorbs
heat into the HTF. This element
consists of a coated stainless steel tube
surrounded by a glass tube insulator
holding a vacuum between the pipe
and the glass, also known as evacuated
glass (Figure 7). The coating on the
steel tube decreases reflectivity and
increases
its
heat
absorption
capabilities. The outer glass cylinder
has anti-reflective coating on the inner
and outer surfaces to reduce reflective
losses, increasing the transmission of
solar energy.
Figure 2.18 Parabolic mirror facets
3. Heat Transfer Fluid System (HTF).

Heat transfer fluid is the link between
the suns radiation energy and the
steam energy utilized in the power
block. In a parabolic trough plant, HTF
travels through a series of processes
such as pumping, expanding, heating,
cooling, and purifying. Circulation,
expansion, and freeze protection are
the primary HTF sub-systems.
The HTF selected for the reference
plant is a high-temperature synthetic
oil that has a proven operating history
and is used in operating CSP trough
plants and other heat transfer
processes. It has a maximum operating
temperature of 750F (399C) and a
freezing temperature of 54F (12.2C).
Figure 2.17 Heat Collection Element
2. Mirrors.
The parabolic mirrors that reflect solar
energy to the HCE use in this study will
be low-iron laminated or monolithic
glass mirrors. These mirrors are known
to be reliable components that show
no long-term degradation in reflective
quality. Reflectivity of the mirrors is
closely monitored to determine the
frequency and method of mirror
washing. Regular mirror washing has
been proven to maintain mirror
reflectivity at installed conditions.
These properties are critical factors in

the design of the HTF system. Because
the HTF maximum temperature is the
primary factor limiting the steam cycle
efficiency, proper HTF selection is
important in actual plant design.
78
2.3.4.2.2 Configurations of
parabolic trough systems
Using only parabolic trough panels to
get heat energy, some of that heat
energy is used in desalination purposes

and the rest steam is used in getting
power by coupling with a heat recovery
boiler, steam turbines and condenser.
Figure 2.19 PTC solar plant
Typical CSP parabolic trough plants have a

reheat Rankine power cycle driven by oilto-water
heat
transfer.
As described above, hot HTF flows from
the solar field through a series of heat
exchangers in the power block, which
ultimately provides superheated high
pressure steam to drive the steam turbine
generator.
circulating water from a mechanical draft

cooling tower, where the condensate is
collected in the hot well, from which the
condensate pumps take suction.
Condensate is pumped through a series of

low pressure closed feedwater heaters and
into an open de-aerating type feedwater
Using both photovoltaics and parabolic

trough panels (hybrid system) to get heat
energy from parabolic trough panels which
is used in the desalination process and
getting power from photovoltaic panels
which is used in electrolysis and other
purposes such as pumps.
2.3.4.2.3 Advantages of parabolic
trough systems
- No Fuel Cost, Solar Thermal Energy does
not require any fuel like most other
sources
of
renewable
energy.
This is a huge advantage over other fossil
fuels whose costs are increasing every
year.
heater used to remove oxygen from the

system. The feedwater pump discharge is
sent through the economizer, evaporator
and superheater. The steam from the
steam generator train is admitted to the
high pressure (HP) turbine. HP turbine
exhaust is reheated in the oil-to-steam
reheater where lower pressure steam is
admitted to the low pressure (LP) turbine.
LP turbine exhaust is then condensed in a
wet surface condenser, using cool
- Solar Thermal Energy can generate power

24 hours a day (by using storage in hot
tanks). This is made possible as solar
79
thermal power plants store the energy in

the form of molten salts. Other forms of
Renewable Energy like Solar PV and Wind
Energy are intermittent in nature. The
electricity supply is much more uniform
and
reliable.
- No Pollution and Global Warming Effects,
Solar Thermal Energy does not cause
pollution which is one of the biggest
advantages.
- Using Existing Industrial Base, Solar
Thermal Energy uses equipment like solar
thermal mirrors and turbines which is
made in large scale at low cost by the
existing Industrial Base and requires no
major changes in equipment and materials
unlike new technologies such as CIGs
Panels (copper indium gallium selenide
solar cell ).
populated areas. Solar Thermal Energy also

can only be built in large sizes which are at
least 50 MW in size to be economical.
- Long Gestation Time Leading to Cost
Over-runs, it take 5-7 years to develop a
concentrated solar thermal power plant.
Compare this to 6 months for a small wind
farm or 3 months for a Solar PV plant.
2.3.4.2.5 Calculations
- Energy input from solar rays to Parabolic
collector:
= .
Where,
Aa: the aperature area
Gb: Solar beam radiation and it equals to
GDN cos, where GDN is the direct normal
solar irradiance and is the solar angle of
incidence.
2.3.4.2.4 Disadvantages of
parabolic trough systems
- High Costs, Solar Thermal Energy costs at
least 3.5 Euro /watt. However, these costs
are too high as Solar PV already costs 2.5
Euro /watt and it will have its costs
reduced by 5% in the next 10 years making
it attain half the cost of Solar Thermal
Technology
by
2020.
- Water Issue, Solar Thermal Plants use lots
of Water which is Major Problem in Desert
Areas which raises the cost of CSP projects
(concentrating solar power) too much.
While using Sea Water has been proposed
it. Remains to be seen if it possible to
implement this solution as this would
imply building Plants very near the Coastline.
- Limited Locations and Size Limitations,
Solar thermal energy can only be built in
places which have the high amount of solar
radiation. They can be built in deserts
mostly and require a large land area. This
means it is not possible to build them in
- To calculate the energy gain:

Qu = FR [Gb Aa - Ar UL (Ti Ta)]
where,
FR: Heat removal factor
: Collector optical efficiency
Ar: Receiver area
UL: Overall heat loss coefficient
Ti: Collector inlet temperature
Ta: Ambient temperature
- To calculate the collector efficiency:
=
- To calculate the oil output temperature:

Qu = m . Cp . (Tout Tin)
2.4 Desalination process

The provision of fresh water is becoming
an increasingly important issue in many
areas of the world. In dry areas, potable
80
water is very scarce and the establishment

of a human habitat in these areas strongly
depends on how such water can be made
available.
Water is one of the most abundant
resources on earth, covering three-fourths
of the planets surface. About 97% of the
earths water is salt water in the oceans
and 3% is fresh water contained in the
poles (in the form of ice), ground water,
lakes and rivers, which supply most of
human and animal needs. Nearly, 70%
from this tiny 3% of the worlds fresh water
is frozen. 30% of all fresh water is
underground hard to reach.
Table 2.4 Water Classification based on salinity content
2.4.2 Desalination techniques:

A solar distillation plant may consist of
separated or integrated systems for the
solar collector and the distiller. Integrated
systems are often referred to as direct
solar desalination and usually involve
different types of solar stills. Separated
systems are known as indirect solar
desalination, in which the thermal energy
from the sun is used for either heating the
seawater or generating steam in
conventional distillation plants, using for
example Multi-Effect distillation (MED),
Multi-Stage Flash (MSF) or Reverse
Osmosis (RO) systems.
The only nearly inexhaustible sources of

water are the oceans. Their main drawback
is their high salinity. Therefore, it would be
attractive to tackle the water shortage
problem with desalination of this water.
The regions in most need of additional
fresh water are also the regions with the
most intense solar radiation. For this
reason thermal solar energy in
desalination processes should be the most
promising application of renewable
energies to seawater desalination.
Thus, desalination techniques may be

classified into the following categories:
1- membrane or single phase processes
(RO)
2- Phase change or thermal processes
(MSF
&
MED)
The main problems with the use of solar

thermal energy in large scale desalination
plants are the relatively low productivity
rate, the low thermal efficiency and the
considerable land area required.
All processes require a chemical pretreatment of raw seawater to avoid

scaling, foaming, corrosion and fouling and
also require a chemical post-treatment.
2.4.1 Water Classification:

Classification of various types of water is
based on the purpose for which the water
is used.
In our case, we need to the desalination

process because the electrolyser feed in
water should be fresh water to protect the
electrodes.
Solar energy can be used for seawater
desalination either by producing the
81
thermal energy required to drive the

phase-change processes or by producing
electricity required to drive the membrane
processes.
To be "selective", this membrane should

not allow large molecules or ions through
the bores (holes), but should allow smaller
components of the solution (such as the
solvent) to pass freely.
2.4.2.1 Reverse Osmosis
2.4.2.1.1 Theory of operation:

The RO system depends on the properties
of semi-permeable membranes which,
when used to separate water from a salt
solution, allow fresh water to pass into the
brine compartment under the influence of
osmotic pressure. If a pressure in excess of
this value is applied to the salty solution,
fresh water will pass from the brine into
the water compartment. Theoretically, the
only energy
Reverse osmosis (RO) is a water

purification technology that uses a semipermeable membrane to remove larger
particles
from
water.
The fluid has a tendency to move from low
to high solute concentrations for chemical
potential equilibrium. Reverse osmosis is
the process of forcing a solvent from a
region of high solute concentration
through a semipermeable membrane to a
region of low solute concentration by
applying a pressure.
requirement is to pump the feed water at

a pressure above the osmotic pressure. In
practice, higher pressures must be used,
typically 5080 atm, in order to have a
sufficient amount of water pass through a
unit area of membrane.
The operating principle of membrane

processes leads to direct production of
electricity from solar energy, which is used
to drive the plant. Energy consumption is
usually expressed in KWhe /m3.
Solar energy can be used with RO systems

as a prime mover source driving the
pumps.
Reverse osmosis can remove many types

of molecules and ions from solutions,
including bacteria, and is used in both
industrial processes and the production
of potable water. The result is that
the solute is saved on the pressurized side
of the membrane and the pure solvent is
allowed to pass to the other side.
Reverse osmosis is sometimes used to
prevent formation of minerals on the
surface of electrodes. The largest and most
important application of reverse osmosis is
the separation of pure water from
seawater and brackish waters; seawater or
brackish water is pressurized against one
surface of the membrane, causing
transport of salt-depleted water across the
membrane and emergence of potable
drinking water from the low-pressure side.
82
Figure 2.20 Principle of operation of a reverse osmosis system
2.4.2.1.2 RO feed-in water pretreatment

The output of RO systems is about 500 1500 liter per day per square meter of
membrane, depending on the amount of
salts in the raw water and the condition of
the
membrane.
The membranes are in effect very fine
filters, and are very sensitive to both
biological and non-biological fouling. To
avoid fouling, careful pre-treatment of the
feed is necessary before it is allowed to
come in contact with the membrane
surface.
2.4.2.1.3 Membrane modules

There are four main types of modules:
plate-and-frame, tubular, spiral wound,
and
hollow
fiber.
The plate-and-frame module is the
simplest configuration, consisting of two
end plates, the flat sheet membrane, and
spacers.
In tubular modules, the
membrane is often on the inside of a tube,
and the feed solution is pumped through
the tube. The most popular module in
industry for nano-filtration or
reverse osmosis membranes is the spiral
wound module. This module has a flat
sheet membrane wrapped around a
perforated permeate collection tube. The
feed flows on one side of the membrane.
Permeate is collected on the other side of
the membrane and spirals in towards the
center collection tube.
Hollow fiber
modules used for seawater desalination
consist of bundles of hollow fibers in a
pressure vessel. They can have a shell-side
feed configuration where the feed passes
along the outside of the fibers and exits the
fiber ends. Hollow fiber modules can also
be used in a bore-side feed configuration
where the feed is circulated through the
fibers.
One method used recently for the pretreatment of seawater before directed to
RO modules is nano-filtration (NF). NF is
primarily developed as a membrane
softening process which offers an
alternative to chemical softening.
The main objectives of NF pre-treatment
are:
1. Minimize particulate fouling of the RO
membranes by removal of turbidity and
bacteria.
2. Prevent scaling by removal of the
hardness
ions.
3. Lower the operating pressure of the RO
process by reducing the feed-water total
dissolved solids (TDS) concentration.
83
Figure. 2.21. Membrane module types, a) plate and frame, b) tubular, c) spiral wound and d) hollow fiber modules
2.4.2.1.4 RO calculations
- To determine the rate of water transport
across the membrane:
: the membrane thickness.
NAw = L ( )
- For an ideal solution, with complete

dissociation of salt ions, osmotic pressure
is defined as:
Where,
= CRT
NAw: the water flux through the membrane
where,
p:
the
difference
: the osmotic pressure
transmembrane
pressure
C: the salt ion concentration
: the difference in osmotic pressure

between the feed and the permeate
R: the ideal gas constant

T: is the solution temperature.
L: a constant describing the physical

characteristics of the membrane itself.
The salt ion concentration, C, is given by

the number of ions in solution per gram of
water divided by the specific volume of
water.
- L is given by:
L=
where,
D: the water diffusivity in the membrane
S: the water solubility in the membrane
V: the molar volume of water
R: the ideal gas constant
T: the ambient temperature
84
number of stages increases, the terminal

temperature difference over the heat
exchangers increases and hence less heat
transfer area is required with obvious
savings in plant capital cost.
Table 2.5.Typical osmotic pressure values for solutions at
25C
2.4.2.2 Phase change or thermal

processes (MSF & MED)
The operating principle of phase change
processes requires reusing the latent heat
of evaporation to preheat the feed while at
the same time condensing steam to
produce fresh water. The energy
requirements of these systems are
traditionally defined in terms of units of
distillate produced per unit mass of steam
per 2326 KJ (1000 Btu) heat input which
corresponds to the latent heat of
vaporization at 73 . This dimensional
ratio in Kg/2326 KJ or lb/1000 Btu is known
as the performance ratio (PR).
Figure 2.22 View of a typical multistage flash desalination

plant
2.4.2.2.1.1 Advantages of MSF:

The main reason for the wide use of MSF
plants is the high capacity output that they
provide at relatively higher thermal
efficiency and reliability, which leads to
high performance and lower production
costs. In addition, they are easy to operate
and have almost the same characteristics
as a power plant with which they are
mostly combined.
Phase change or thermal processes are

divided into the multi stage flashing
desalination (MSF) and the multi effect
distillation (MED).
2.4.2.2.1.2 Disadvantages of
MSF:
A disadvantage of MSF is that precise
pressure levels are required in the
different stages and therefore some
transient time is required to establish the
normal running operation of the plant.
2.4.2.2.1 Multi Stage Flash (MSF)

The MSF process is the workhorse of the
desalination industry in the Gulf States as
well as several other countries across the
world.
The system is divided into heat-recovery
and heat-rejection sections. Seawater is
fed through the heat-rejection section,
which rejects thermal energy from the
plant and discharges the product and brine
at the lowest possible temperature.
2.4.2.2.1.3 Configurations of
MSF:
The MSF desalination plants are divided
into two models, Once-through MSF (MSFOT) and Brine recirculation MSF designs
(MSF-BC).
Current commercial installations are

designed with 10 30 stages with
2 temperature drop per stage. As the
85
2.4.2.2.1.3.1 Once-through MSF

configuration:
The Once-through MSF plants consist of an
evaporation section (heat recovery
section) and a brine heater and the
condenser tubes arrangement will let the

process be either Long-Tube (LT)- once
through, or Cross-Tube (CT)- once through.
Figure 2.23.Multi-stage Flash Desalination Once through process. (MSF-OT)
Process details are described in the

following points:
formation.
- The feed seawater (Mf) enters the
flashing stages. The flashing process takes
place due to a decrease in the stage
saturation temperature that causes the
reduction in the stage pressure.
- In each stage, the flashed off vapor
condenses on the outside surface of the
condenser tubes, where the feed seawater
(Mf) flows inside the tubes from the cold to
the
hot
side
of
the
plant.
- The condensed fresh water vapor outside
the condenser tubes accumulates across
the stages and forms the distillate product
stream (Md). This stream cascades in the
same direction of the flashing brine from
stage to stage and is withdrawn from the
last
stage.
- The vapor flashed off in the flash chamber
is separated from the condenser by a
demister which intercepts brine droplets
entrained in the flashing vapor resulting in
- The feed seawater (Mf) is de-aerated and

chemically
treated
before
being
introduced into the condenser tubes of the
last flashing stage in the heat recovery
section.
- As the seawater flows across the tubes
from stage (n) to stage (1), its temperature
increases due to absorption of the latent
heat of the condensing fresh water vapor.
- The feed seawater (Mf) enters the brine
heater tubes, where the heating steam
(Ms) is condensed on the outside surface of
the tubes. The feed
seawater (Mf) temperature increases to its
maximum design value known as the top
brine
temperature
(To).
This value depends on the nature of
chemicals used to control the scale
86
reduction of the product salinity to values

below 10 ppm.
At such conditions, air and other dissolved

gases in the flashing brine are noncondensable and their presence in the
system may result in severe reduction in
the heat transfer rates within the chamber,
increase of the tendency for corrosion, and
reduction of the flashing rates. This
condition necessitates proper venting of
the flashing stages to enhance the flashing
process and to improve the system
efficiency
2.4.2.2.1.3.2 Brine circulation
MSF configuration:
The Brine recirculation MSF consists of
brine heater, heat recovery section, and
heat rejection section. The role of the
rejection section is to remove the surplus
thermal energy from the plant, thus
cooling the distillate product and the
concentrated brine to the lowest possible
temperature. Also the condenser tubes
arrangement will let the process to be
either Long-Tube (LT)-brine recirculation,
or Cross-Tube (CT)-brine recirculation.
Figure 2.24.Inside view of the process
- The flashing process and vapor formation

is limited by the increase in the specific
vapor volume at lower temperatures and
difficulties encountered for operation at
low pressures. Common practice limits the
temperature of the last stage to range of
30 to 40 , for winter and summer
operation,
respectively.
- In MSF, most of flashing stages operating
at temperatures below 100 have
vacuum pressure. This increases the
possibilities of leakage of the outside air
into the vessel.
Figure 2.25.Multi-stage Flash Desalination brine circulation process (MSF-BC)
87
MSF-BC is a system where the brine is

recycled as much more common. Details of
the MSF process are described below:
the increase in the feed seawater

temperature.
- The condensed fresh water vapor outside
the condenser tubes accumulates across
the stages and forms the distillate product
stream
(Md).
- Treatment of the intake seawater
(Mf+Mcw) is limited to simple screening and
filtration. On the other hand, treatment of
the feed seawater stream is more
extensive and it includes de-aeration and
addition of chemicals to control scaling,
foaming, and corrosion.
- The intake seawater stream (Mf+Mcw) is

introduced into the condenser tubes of the
heat
rejection
section.
- The warm stream of intake seawater is
divided
into
two
parts:
the first is the cooling seawater (Mcw),
which is rejected back to the sea.
The second is the feed seawater (Mf),
which is de-aerated, chemically treated
and then mixed in the brine pool of the
last flashing stage in the heat rejection
section.
- The brine recycle stream (Mr) is extracted
from the brine pool of the last stage in the
heat rejection section and is introduced
into the condenser tubes of the last stage
in
the
heat
recovery
section.
- The brine recycle stream (Mr) enters the
brine heater tubes, where the heating
steam (Ms) is condensed on the outside
surface of the tubes. The brine stream
temperature increases to its maximum
design value known as the top brine
temperature (To). Its value depends on the
nature of chemicals used to control the
scale
formation.
- The hot brine enters the flashing stages in
the heat recovery section and then in the
heat rejection section. The flashing process
takes place due to decrease in the stage
saturation temperature and causes the
reduction in the stage pressure.
- In each stage of the heat recovery section,
the flashed off vapor condenses on the
outside surface of the condenser tubes,
where the brine recycle stream (Mr) flows
inside the tube from the cold to the hot
side of the plant. This heat recovery
improves the process efficiency because of
2.4.2.2.1.3.3 Comparison
between MSF processes:
The most common multistage flash
desalination process is the brine recycle
system
(MSF-BC).
As for the once through system (MSF-OT)
its use is found on a very limited scale.
88
This is because the (MSF-BC) has the

following
advantages:
- In MSF-OT there is no control of the feed
sea
water
temperature.
In MSF-BC control of the sea water
temperature takes place in the last stage of
the
heat
rejection
section.
- The flow rate ratio of the feed to product
in the MSF-OT process is approximately
10:1. On the other hand, this ratio in the
brine
circulation
MSF
is
2.5:1.
This requires use of larger equipment for
chemical treatment and higher dosing
rates of various treatment chemicals.
- The gain output ratio or performance
ratio (amount of distillate produced per
unit mass of steam feed to the brine
heater) of MSF-BC remains constant as the
seawater temperature drops from 25
(summer) to 15 (winter) and that is due
to the winter loop operation which
controls the intake seawater temperature

before entering the plant. While MSF-OT
reduces the gain output ratio to low values
of 3. That is because more steam will be
required to heat the seawater to the same
top
brine
temperature.
- Low temperature of the reject brine in the
MSF-BC due to the mixing in the stage and
high temperature of the reject brine in the
MSF-OT.
On the other hand, the MSF-OT use can be
useful especially in equatorial regions,
where the seawater temperature remains
nearly constant throughout the year. The
MSF-OT system has the following
advantages over the MSF-BC system such
as:
- Reduction in the number of pumping
units (MSF-OT has pumps for feed
seawater, distillate product, and brine
reject) and (MSF has additional pumps for
brine recycle and cooling seawater) and
that would reduce the number plant trips
caused to the pumping units (50% of all
MSF
plant
trips)
- The salinity of the brine recycle stream is
much higher than that of the intake sea
water. Therefore the tendency towards
scale formation would be higher in the
MSF-BC system than in MSF-OT.
-Low brine salinity 36,000-45,000 ppm in

MSF-OT will reduce the consumption rate
of the anti-scalent. High brine salinity
60,000-75,000 ppm in the MSF-BC will
increase the consumption rate of the antiscalent
Other factors to be taken into
considerations:
- The gain output ratio of the MSF-OT is
similar
to
the
MSF-BC.
- The capacity of the MSF-OT is similar to
the
MSF-BC.
- The specific heat transfer area (defined as
the total heat transfer area per unit
product flow rate) of MSF-OT is similar to
MSF-BC.
2.4.2.2.1.4 Multi-stage flash
calculations:
Physical
Model
Description
It is easier to describe a single flash stage if
it is split up into four control volumes,
which can be treated separately.
The four control volumes are flash
chamber, vapor space, product tray, and
condenser
tube
bundle.
Each fluid stream communicating with the
individual stage has four characterizing
variables, flow rate, temperature,
pressure, and salt concentration.
Figure 2.26.Block diagram of a repeated stage
89
Stage Model:
- The mass conservation equation for
the brine in the flashing pool is given
by:
()
() ()
()
= () () ()
+ () ()
() () ()
= (1) () ()
Where U(i) is the heat transfer coefficient,

AC(i) is the heat transfer area and T(i) is the
log mean temperature difference (LMTD),
given by:
, () = () () ()
- The balance of the dissolved solids
reads:
()
()
= (1) (1)
() ()
- The energy balance for brine pool is
given by:
()
() ()
= (1) (1) (1)
()
(() () ) (() (+1) )
=
(() () )
ln[
]
(() (+1) )
- The mass balance for the condenser tubes
can be written as:
() () () () ()
() = (+1)
- The mass conservation equation for the

product in the tray is:
- The energy balance for the condenser

tubes yields:
()
= (1) + () ()
()
() ()
()
= (+1) (+1) (+1)

() () () + () () ()
- The energy balance for the product tray

is:
() ()
()
= (1) (1) (1)

+ () ()
() () ()
Last Stage Model:

- The brine mass conservation equation
is:
()
= (1) + ()
()
() ()
- The mass balance for the vapor space
produces:
- The balance of dissolved salt is:

()
()
= (1) (1)
+
() ()

()
= () + () ()
()
- The energy balance for the vapor space is
as follows:
90
- The energy balance for the brine is:

()
()
= (1) (1) (1) +
system according to the feed seawater

arrangements include the forward,
backward and parallel/cross feed.
2.4.2.2.2.1 Theory of operation:
An MED system mainly consists of number
of single effect evaporators in series and an
end condenser. It is also provided with
venting system, distillate ashing boxes
and seawater feeding and rejecting
facilities as additional auxiliary equipment.
Fresh water is obtained as a result of the
evaporation of water in each evaporator.
In each evaporator, there are a horizontal
heat transfer tube bundle, demister and
vapor space.
() () ()
() ()
Brine Heater Model:
- The mass and energy balance
equations are given by:

( ) (1) ((0) (0)

(1) (1) )
=
(1) = (0)
= and =
(1) = (0)
=
Where, (0) refers to the 0th stage.
2.4.2.2.2 Multi Effect Desalination
Multi-effect distillation (MED) is also
known as long-tube vertical distillation
(LTV) and is in principle similar to multistage flash evaporation except the steam is
used to heat up the seawater in the first
stage and the resulting vapor is used in
subsequent stages to evaporate the water.
And the seawater / brine is used to cool
and condense the vapor in each successive
stage to that the temperature gradually
falls across each stage of the process. As in
the case of multi-stage flash evaporation,
many stages are used in commercial
plants. MED process has different types
and arrangements. On the basis of tube
arrangements inside the evaporator, MED
system is divided into three congurations:
horizontal tube arrangement, vertical tube
arrangement and vertically stacked tube
bundles. There are different arrangements
for supplying the feed seawater to the
evaporators in the MED system.
Commonly used congurations of the MED
Figure 2.27.Multi effect desalination process
There are different arrangements for

supplying feed seawater to the
evaporators including forward, backward
and parallel or cross feed schemes. These
congurations differ in the ow direction
of evaporating brine and heating steam. In
each scheme, saturated steam from an
external source enters into the tube side of
the rst effect as heating medium to heat
the feed seawater. The condensate of the
rst effect is returned to the external
source such as boiler.
91
Figure 2.28. Configurations of multi effect desalination system
- In the forward feed arrangement, feed

water is supplied to the rst effect. Part of
the feed is evaporated in the rst effect
while the remaining unevaporated part
known as brine, and enters into the second
effect. In the second effect, some part of
brine evaporates while the remnant enters
as a feed source to the next effect and this

process continues up to the last effect. The
cooling seawater enters into the
condenser where it exchanges heat with
the vapor exiting the last effect. So in
forward feed conguration, feed and
vapor enter the effects and ow in the
92
same direction. Forward feed arrangement

is also called co-current arrangement. The
main advantage of the forward feed
conguration is its ability to be operated at
high top brine temperatures.
because brine ows from higher to lower

pressure effects. Parallel or cross feed MED
system is most commonly used on the
industrial scale.
2.4.2.2.2.2 Advantages of MED:
Multi-effect distillation has a number of
distinguished features, in particular, lower
primary energy consumption, higher gain
output ratio, lower heat transfer area and
the product water is of a high quality. MED
process is less susceptible to corrosion and
scaling than the MSF process. The MED
requires only a very simple seawater
treatment when compared to MSF. The
MED plant can handle normal levels of
biological or suspended matters.
- In the backward feed conguration, feed

seawater after passing through the end
condenser, enters into the last effect
where the temperature and pressure is the
lowest within the system. The brine leaving
the last effect is directed through each
effect towards the rst effect. The brine
leaving the rst effect is sent back to the
sea as rejected brine. Temperature and
pressure increases as the ow moves from
last effect to the rst effect. In this layout,
feed and vapor enter the effects have
opposite ow directions, which is called
counter-current cascade. In the backward
feed
arrangement,
highest
brine
concentration is subjected to the effect
which has highest temperature in the
system, which represents a major
drawback of backward feed conguration.
It can cause severe scaling problem in the
evaporator. Another disadvantage of this
conguration is the increase in pumping
power requirement and maintenance cost.
2.4.2.2.2.3 Disadvantages of
MED:
- They are expensive to build and operate
as energy consumption is particularly high.
- The product water is at an elevated
temperature and can require cooling
before using as a potable water.
- The recovery ratio is low, although not as
low as for MSF.
2.4.2.2.3 Comparison between
MSF and MED:
MSF and MED are both thermal
desalination
processes.
The
way
evaporation and heat transfer take place in
these processes are completely different.
In MED process, vapor is originated from
the falling lm of seawater which brought
in contact with the heat transfer surface. In
MSF process, seawater is rstly heated
within the tubes by convection then vapor
is initiated by ashing as it streams in each
stage of reduced pressure. MSF system
normally consists of three sections: heat
recovery, heat rejection and brine heater.
- In the parallel or cross conguration, feed

is distributed almost equally to all effects.
In order to utilize the energy of brine
leaving the rst effect, it is directed to the
next effect and so on it goes to the last
effect at a lower pressure than the
previous effect. Feed and vapor enter the
effects and ow in same direction in this
conguration. Flow pattern is same as in
the forward feed conguration. Parallel or
cross feed conguration has simple design
as compared to the other two layouts.
Another feature of this arrangement is
that, no pumps are needed for moving the
unevaporated brine between the effects
93
electrolyte membrane (PEM) electrolysis is

currently discussed and described to be in
the pre-commercial state. While alkaline
electrolysis is already used in large-scale
operations and shows advantages like long
durability and low costs, it also has lower
flexibility with a minimum load
requirement of 20 % to 40 %. These
characteristics could lead to problems with
fast start-up requirements for power to gas
applications. However, improvements in
start-up and ramp characteristics can be
expected
2.5.1 Theory of water electrolysis

operation:
The core of an electrolysis unit is an
electrochemical cell, which is filled with
pure water and has two electrodes
connected with an external power supply.
At a certain voltage, which is called critical
voltage, between both electrodes, the
electrodes start to produce hydrogen gas
at the negatively biased electrode and
oxygen gas at the positively biased
electrode. The amount of gases produced
per unit time is directly related to the
current that passes through the
electrochemical cell.
Table 2.6. Comparison of MSF and MEE desalination

processes.
2.5 Electrolysis
Water electrolysis is currently one of the
most promising methods for hydrogen
production. Hydrogen is produced via
electrolysis by passing direct current (DC)
through two electrodes in water. The
water molecule is split, producing oxygen
at the anode (positive electrode) and
hydrogen at the cathode (negative
electrode). The technique can be used to
make hydrogen fuel and breathable
oxygen; though currently most industrial
methods make hydrogen fuel from natural
gas instead.
Electrolysis reactions:
Cathode Electrode reaction:
2 2 + 2 2 + 2
Anode Electrode reaction:
2 2 + 2 +
Three different electrolysis technologies

are of interest for PtG process chains:
alkaline electrolysis (AEL), polymer
electrolyte membranes (PEM), and solid
oxide
electrolysis
(SOEC).
Alkaline electrolysis is a mature technology
and commercially available. Also polymer
Hence,
the
2 2 2 +
total
1
2
1

2 2
reaction
is:
2 + 2
Through this reaction, two molecules of

water are decomposed and hydrogen
evolves in the cathode. In the anode,
94
oxygen evolves and at the same time one

molecule of water is regenerated. As a
result, one molecule of water is
decomposed and another molecule of
water moves to the anode. The most
important parameters of the electrolysis

technologies are efciency, exibility, and
lifetime. Electricity expense constitutes the
largest fraction of hydrogen production
costs.
Figure 2.29.Electrolysis plant principle
It is forbidden to mix the produced

hydrogen and oxygen with each other
which may lead to an explosion especially
if there is a spark. The separator porous
diaphragm between 2 & 2 was made by
asbestos, Now it is made by polymers to
prevent 2 from going towards the
cathode again and forming water (by
reduction) which increases the current
consumption
(reducing
efficiency).
An alkaline or acidic material is added to
the solution with a certain concentration
which increases the specific conductivity to
a maximum value (minimum resistance).
An
Massive current densities cause a

noticeable value of unwanted ohmic
voltage drop between the electrodes. As a
result, formation of unwanted electrical
power loss and less process efficiency
values are inevitable. Many efforts are
made in order to force an electrolysis cell
to reach a certain current level by applying
lower voltage levels to it.
2.5.2 Factors affecting on electrical

efficiency:
- Electrolyte quality: Bases and acids are
known to change the nonconductive
nature of pure water. These compounds
have a great reducing effect on the
overvoltage value of an electrolyser as
they improve the ionic conductivity
aqueous electrolyte compounds. However,
the concentration level of acidic and alkali
solutions are limited in practice due to the
highly corrosive behavior of such
electrolyzer is usually subjected to massive

current values in order to break the water
molecules into oxygen and hydrogen.
Obviously, for very large current levels, any
slight increase of the cell voltage can cause
a drastic raise in the power demand.
95
materials. A 25% to 30% KOH aqueous

solution is reported to have a wide use in
electrolysers.
- Temperature: Temperature is known to
be one of the most effective variables on
the electric power demand of an
electrolysis cell. Electrolysis process is
much
more
efficient
at
raised
temperatures as water splitting reaction
potential reduces as the temperature
increases. Moreover, ionic conductivity
and surface reaction of an electrolyte rise
directly
with
temperature.
High
temperature water electrolysis requires
less energy to reach any given current
density in analogy with a low temperature
process.
- Pressure: high pressure electrolytes will
consume less power in the process of
electrolytic decomposition. The main
reason was stated to be the shrinking
effect of pressure on the gas bubbles which
cause the ohmic voltage drop and power
dissipation to reduce. Moreover, high
pressure electrolysis has less power
demand for the phase of product
compression.
- Electrical resistance of the electrolyte:
Electrical resistance of an object is an
evaluation of its opposition to the passage
of electric current.
- Separator material: Placing a separator

plate in a cell, blocks the free movement of
mass and ions to some extent. Moreover,
the presence of such barrier increases the
void fracture by further accumulating of
gas bubbles in the electrolyte. In addition,
the effective electrical resistance of a
separator plate is frequently calculated to
be as large as three to five times of those
of
the
electrolyte
solutions.
- Applied voltage waveform: It is almost
common for electrolysis systems to use a
steady or smooth DC voltage to
decompose an electrolyte. Their goal was
to reach higher cell power (increase the
gas production rate) without reducing the
efficiency.
2.5.3 Alkaline electrolysis

Despite the fact that the discovery of
electrolytical water decomposing was first
observed in acidic water, in industrial
plants the alkaline medium is preferred,
because corrosion is more easily controlled
and cheaper construction materials can be
used compared to acidic electrolysis
technology.
The anode and cathode materials are
typically made of nickel-plated steel and
steel, respectively. Alkaline electrolysers
are very well capable of flexible operation
with a power load ranging from 5% - 100%,
being able to ramp within this range within
seconds. By increasing the operational
pressure or temperature for alkaline
electrolysis, water molecules can be split
with lower currents (so higher splitting
efficiency).
- Electrode material: A wide range of

materials are being used as electrodes.
Each metal has a different level of activity,
electrical resistance and corrosion
resistivity. Platinum and gold are known to
be two of the best choices for being used
as electrodes. However, high prices limit
their usage in industrial and commercial
electrolysers. Aluminum, Nickel, Raney
nickel and cobalt are the most common
electrode materials for being used in
alkaline
electrolytic
baths.
96
and parallel to one another, diaphrams

being
placed
between
alternate
electrodes. Each electrode is bipolar that is
operates as an anode on one side and a
cathode on the other.
2.5.3.1 Configurations of Alkaline

Electrolysis:
Whereas pressure-less electrolysers are
mainly
built
in
the
monopolar
arrangement, which is the simplest
construction principle, pressurized species
are manufactured in the so-called bipolar
filter press arrangement.
2.5.3.1.1 Monopolar Type:
It is also called the Tank electrolyser, as a
series of electrodes , anode and cathode
alternately , is suspended (vertical and
parallel to one another) in a tank .
Alternate electrodes are surrounded by a
diaphragm and the whole assembly is
hung from a series of bells which serve to
collect the evolved gases .
Figure 2.31.Bipolar configuration
Filter press electrolysers are more compact

than tank electrolysers of the same
capacity. The whole unit tends to be more
expensive than a similar tank electrolyser.
Units often operate across quite high
voltages.The main reasons for the bipolar
arrangement are the savings in space and
in electrical bus bars and housing material,
which are essential if large pressurized
plants need to be built. A small
disadvantage of the bipolar type when
compared to the monopolar, is the
occurrence of shunt currents inside the
electrolyte ducts of the bipolar
electrolyser. In order to face this
disadvantage, a more sophisticated design
of the electrolyte and gas ducts and the
application of a small protection current
during operational shut-down times are
necessary in bipolar water electrolysers. To
meet the demands of the future energy
market, in which electrolysers will have to
serve as filling stations for vehicles in the
environment of a town, rigorous
reconstruction, optimization and safety
implementation, is necessary. Pressurized
alkaline electrolysers have a lower
efciency and produce a lower purity
Figure 2.230. Monopolar configuration
All the anodes and cathodes in one tank

are connected to a pair of common
conductors so that one tank operates
across a 2 2.5 V DC supply. A series of
tanks is therefore connected electrically so
as to give the optimum current / voltage
relationship. The tank electrolyser is
simple to assemble and maintain, and the
component parts are fabricated cheaply. It
tends to occupy a large floor area.
2.5.3.1.2 Bipolar type:
It is also called a Filter Press electrolyser. A
series of electrodes is arranged, vertical
97
product than atmospheric electrolyzer.

The foremost advantage of pressurized
electrolyzer compared to atmospheric
electrolyzer is that it produces compressed
hydrogen with less additional energy
input. This is a result of the fact that the
reduction in electric efciency of the
electrolysis with increased pressure is
lower than the energy needed to compress
the produced hydrogen.
hydrogen and two hydroxyl ions at the

cathode. The hydrogen escapes from the
surface of the cathode recombined in a
gaseous form and the hydroxyl ions
migrate under the influence of the
electrical field between cathode and
anode through the porous diaphragm to
the anode, where they are discharged to
1 molecule of oxygen and one molecule
2
of water. The oxygen recombines at the
electrode surface and escapes as hydrogen
does, as a gas.
2.5.3.2 Principle of operation:

Through this reaction, two molecules of
water are reduced to one molecule of
Figure 2.32. Principle of Alkaline electrolyser
Oxygen evolution reaction (OER):
The enthalpy change H to electrolyze

water
is
shown
as
follows:
= +
2 0.5 2 + 2 + 2
Hydrogen evolution reaction (HER):
2 2 + 2
The decomposition of water by electrolysis

is an endothermic reaction, where energy
corresponding to must be supplied in
the form of electricity and the rest ()
by
heat.
Cell voltage corresponding to is
reversible voltage, about 1.2 V, and
(heat absorption potential) about 0.28 V.
In actual electrolysis, cell voltage is higher
than reversible potential and the
difference is converted into heat. Since the
2 + 2 ()
In industrial alkaline water electrolysis, 20

40 % sodium hydroxide or potassium
hydroxide aqueous solution is used for
electrolyte instead of pure water. This is
used because of KOHs high conductivity,
and because the oxygen evolution reaction
has the least energy loss in this solution.
pure water is highly resistive to electricity.
98
reaction is endothermic, heat thus caused

is absorbed by the reaction until total cell
voltage exceeds 1.48 V (the sum of
reversible voltage and heat absorption
potential). At this cell voltage, all heat
generated by overvoltage and ohmic loss is
used by the reaction and there is no heat
generation or absorption to and from
outside of the system. Therefore, this
voltage is called thermo-neutral voltage
where all electric energy used for
electrolysis is converted into heat content
of evolved hydrogen gas.
for alkaline water electrolysis, it is not

practical to expect an electrolyzer to work
below the thermo-neutral voltage. The
thermo-neutral voltage is very important
in designing electrolyzers, since all
electricity supplied beyond this point is
converted into heat, which must be
removed to maintain the temperature of
cells. It is desirable to design the heat
balance of electrolyzers so that the heat
generated at the operating cell voltage is
equal to the heat loss at the operating
temperature. The reversible voltage
decreases with increased temperature, but
thermo-neutral voltage does not show
much change, since this corresponds to the
energy of hydrogen generated.
Figure 2.33. Distribution of cell voltage
Figure 2.34. Change of cell voltage by temperature
The thermo-neutral cell voltage is the

voltage at which a perfectly efficient cell
would operate if electricity provided the
entire energy requirement. Under these
conditions, the cell does not generate any
waste heat, nor does it require any heat
input, the cell has no net heat exchange,
hence it is then said to be thermo-neutral.
No electrolysis is possible at cell voltage
under reversible voltage, but it is possible
(at least theoretically) to electrolyze water
under thermo-neutral voltage. High
temperature steam electrolysis uses this
principle and by supplying part of the
energy with heat, it reduces cell voltage. As
The energy efficiency of a water

electrolyzer is defined as the quotient of
higher heat value of generated hydrogen
by electric energy supplied to the system.
In water electrolysis, the current efficiency
is usually very high, more than 95 %, and
therefore the energy efficiency is almost
equal to the voltage efficiency.
2.5.3.3 Advantages of alkaline:
-
99
Alkaline
electrolysis
is
the
cheapest
and
most
reliable
technology.
Alkaline
electrolyser
can be operated between 20 and
100% of the design capacity, and
overload operation up to 150% is

possible. This makes AEL a good
choice for systems such as P2G
which
are
coupled
with
a
uctuating and intermittent power
supply.
- Alkaline water electrolyzers with
anion-conducting
membranes
hold
many advantages in comparison to
other systems. As cheaper, nonnoble metals are stable in alkaline
media, the relatively low cost of
the electrode materials is one of
the main advantages of alkaline
systems over electrolyzers using
proton-conducting
electrolytes.
-Low
investment
electrolyzer.
Large
stack
size.
- Extremely low hydrogen impurity
(0.001%).
- A general overhaul of the system is
necessary every 7-12 years. According to,
the expected lifetime for an alkaline
electrolyser is currently about 30 years,
which is high compared to the other
considered electrolyser types. In recent
years, AEL has been optimised, especially
with respect to efciency and total
investment.
a shutdown, so we need a storage

technology in PV panels to avoid
this
problem.
Limited
cost
reduction
and
efficiency
improvement.
High
maintenance
intensity.
- Stacks < 250 KW require unusual
AC/DC
converters.
- Corrosive electrolyte deteriorates
when not operating nominally.
2.5.4 Polymer electrolyte

membrane (PEM) electrolysis
It is also called Proton Exchange
Membrane (PEM). PEM electrolysis is a
promising alternative for energy storage
coupled with renewable energy sources.
PEM electroyzer was introduced to
overcome the issues of Partial load, low
current density and low pressure
operation,
plaguing
the
alkaline
electrolyzer. The heart of a PEM
electrolyser is a proton exchange
membrane (or solid polymer) electrolyte.
2.5.4.1 Historical facts
The first solid polymer electrolysers (SPE)
were developed by the General Electric
Company as fuel cells for the NASA space
program (project Gemini). Subsequently,
small-scale SPE water electrolysers were
used for military and space applications in
the early 1970s. In 1975, a development
program was established to scale-up the
technology targeted towards large-scale,
commercial applications such as energy
storage.
2.5.3.4 Disadvantages of Alkaline:

-
Alkaline water electrolysis uses

fresh water with low salt content,
and
hence
additional
treatment
and desalination systems add to
the cost of hydrogen produced.
Alkaline
electrolysis
uses
potassium hydroxide (KOH) water
solution,
which
would
require
transport and storage of KOH and
is therefore not considered as
feasible
either.
- One problem is that it takes 30 60
min to restart the system following
2.5.4.2 Theory of operation of PEM:

In this unit the electrolyte is a solid ion
conducting
membrane
and
the
electrolyzer therefore is fed with pure
100
water (This is in contrast to the KOH

aqueous solution in the alkaline
electrolyzers). Polymer is a hydrocarbon
chain, we add an acidic material (S3H) to
it or adding an alkaline material (N 2 )
which will be reduces to ions:
R S 3 H
R S 3
+ +
OR
R N 2 + 2 O
R N 3 OH R N 3+ + O
The formation of negative ions (for
example) prevents the oxide ions to go
through the membrane because they will
repel each other and vice versa. The
traditional membrane is Nafion and
consists of a Teflon like polymer with
attached sulfonic-acid groups. The
membrane allows the H+ ion to transfer
from the anode side of the membrane,
where oxygen is produced, to the cathode
side where it forms hydrogen. The PEM
electrolyser membrane also serves to
separate the hydrogen and oxygen gases.
This effectively is an acid environment,
therefore significant precious metal (Pt, Ir,
Ru) loadings are used. PEM electrolyzers
typically are configured in the bipolar
mode.
Figure 2.35. Polymer Electrolyte Membrane principle
The actual value for open circuit voltage of

an operating electrolyzer will lie between
the 1.23 V and 1.48 V depending on how
the cell/stack design utilizes the thermal
energy inputs. This is however quite
difficult to determine or measure because
an operating electrolyzer also experiences
other
voltage
losses
from
internal electrical resistances, proton
conductivity, mass transport through the
cell and catalyst utilization.
Anode reaction:
The half reaction taking place on the anode
side of a PEM electrolyzer is commonly
referred to as the Oxygen Evolution
Reaction (OER). Here the liquid water
reactant is supplied to catalyst where the
supplied water is oxidized to oxygen,
protons
and
electrons.
101
Cathode reaction:
- Greater safety and reliability are expected

since no caustic electrolyte is circulated in
the cell stack.
The half reaction taking place on the

cathode side of a PEM electrolyzer is
commonly referred to as the Hydrogen
Evolution Reaction (HER). Here the
supplied electrons and the protons that
have conducted through the membrane
are combined to create gaseous hydrogen.
- Previous tests made on bare membranes

demonstrated that some materials could
sustain high differential pressure without
damage and were efficient in preventing
gas mixing.
- High current densities, this can result in
reduced operational costs.
- The minimum load is reported to be 5%
- Very fast response time.
- Compact design.
- 6590 % efficiency.
- Large operational ranges.
- Very high gas purities (99.999%).
Figure 2.36.PEM electrolysis components
- Overcome Partial load.

2.5.4.3 Requirements for the membrane
- High protonic conductivity for the

transport of ions.
High pressure
12000 psi).
operation
(300-
2.5.4.5 Disadvantages of PEM electrolysis
- High oxidative stability.
- Low permeability for gases to prevent

mixing of the gases produced.
- This technology is currently more

expensive than Alkaline electrolysis
systems (due to the costs for the
membrane and the use of a noble metal
catalyst).
- Limited lifetime.
- Electric insulator to prevent short circuits.
- Requires high water purity.
- Chemical and mechanical stability to

withstand the harsh conditions in a PEM
electrolysis cell.
- One of the PEM drawback that it is not

available in Egypt and only exists in
producing polyvinylchloride companies.
- Sufficient mechanical and thermal

stability.
2.5.4.4 Advantages of PEM electrolysis

- The main advantages of PEM include
faster cold start, higher exibility, and
better coupling with dynamic and
intermittent systems.
102
Cathode reaction:
2.5.5 Solid Oxide Electrolysis Cell

(SOEC)
2 +2 = 2 + 2-
One of the major problems of conventional

electrolysers is their high electricity
consumption. Steam electrolysis is a
technology that reaches higher total
energy efficiency compared to alkaline and
proton exchange membrane electrolysis.
From the thermodynamic viewpoint of
water decomposition, it is more
advantageous to electrolyze water at high
temperature (800-1000 ) because the
energy is supplied in mixed form of
electricity and heat. The high temperature
system employs oxygen ion conducting
ceramics as the electrolyte. The fluid to be
dissociated is 200 steam which after
being further heated to 800-1000
enters on the cathode side. After the
steam is split to hydrogen gas and O-- ions,
the oxygen ions are transported through
the ceramic material to the anode, where
they discharge and form oxygen gas.
Despite this high efficiency with respect to
electricity, the high temperature system
still produces hydrogen at about four times
the cost of the steam reformed hydrogen.
As a result of the reaction, oxygen ions

enter the solid electrolyte and pass
through to the opposite electrode, the
anode. Hydrogen is released from the
cathode side of the SOEC, the moisture
level of which is inversely proportional to
the amount of water used. From the
anode side of the SOEC, oxygen is released:
Anode reaction:
2- = 2 + 2
Figure 2.37.Solid Oxide electrolysis operation
2.5.5.1 Theory of operation:
In conventional steam electrolysers, the

gas supplied to the cathode side (where
water is decomposed) is usually a mixture
of steam and hydrogen, while the gas
supplied to the anode is usually air. At zero
current, the system has an open-circuit
voltage of 0.8 to 0.9 V, depending on the
hydrogen/steam ratio and on operating
temperatures. This open-circuit voltage is
even higher for water electrolysis. In order
to electrolyze water, a voltage that
opposes and is higher than the open circuit
voltage must be applied in order to pump
oxygen from the steam side to the air side.
Clearly, much of the electricity used, 60 to
70% of the total electrical power, is wasted
Solid Cell is developing HTE technology

that will operate at 700C - 900C, based on
high temperature electrochemistry in
solid-oxides. This type of HTE system is
called a solid-oxide electrolyzer (SOE). The
basic building block of the system is a solidoxide electrolytic cell (SOEC). The SOEC is
built around a gas-tight oxygen ion
conducting electrolyte. The
opposite
surfaces of the electrolyte are in contact
with porous electrodes. One of the
electrodes, the cathode, is supplied with
steam (water vapor):
103
forcing the electrolyser to operate against

the high chemical potential gradient for
oxygen. In addition, the liberation of
oxygen coming from the decomposition of
water into the air stream at the anode side
is clearly a waste.
electric energy input. The deficient energy

for water decomposition is supplied in the
form of heat.
2.5.5.2 The difference of SOEC from the
Most of the current SOEC cell

configurations are developed based on the
SOFC technologies. The characteristic of
the solid-state components enables SOEC
to be manufactured in many different
configurations. Same with SOFC, the SOEC
stacks can be built based on the planar
design, tubular design, the segmentedcells-in-series design and the monolithic
design. Among these configurations, the
tubular cells and planar cells are widely
accepted as the appropriate design for
SOEC commercialization.
2.5.5.3 Solid Oxide Electrolysis Cell

configurations:
other types:
In the best low-temperature electrolytic
cells operating under normal pressure, the
average cell voltage is 1.7 V. It
corresponds to a power input of 4.07 KWh
/ Nm3H2. However, this number does not
reflect all the power inputs for hydrogen
production
in
a
low-temperature
electrolytic cell. Electrolyzers based on
low temperature electrolytic cells require
very pure water (distilled or even bidistilled), so for a proper analysis of energy
consumption of such devices it is necessary
to take into account the power inputs for
preparation of pure water. To prepare a
sufficient amount of water for production
of 1 m3 of hydrogen, it is necessary to
spend no less than 0.73 kWh in the case of
one step distillation and about 1.5 kWh in
the case of double distillation. Electrical
energy is generally used for water
distillation, so, the real power input in a
low temperature electrolytic cell is 4.8 to
6 KWh / Nm3H2.
2.5.5.3.1 Planar type:

Each flat single cell is stacked in series and
connected by the metallic or ceramic
interconnect
plates.
The
planar
configuration has the advantages of high
volumetric gas production rate, low
complexity
in
manufacturing and
fabrication, and the relatively easier
current collecting. However, the sealing
and thermal cycling performances are big
concerns of planar cells.
Since the enthalpy change of the reaction

corresponds to a power input of 3.54 KWh
/ Nm3H2, the best low temperature
electrolyzers require a specific energy
input of more than 15% higher, even
without taking into account the energy
inputs for water distillation. This energy is
released as waste heat. The process of
water decomposition in a SOEC runs at an
acceptable rate under an average cell
voltage of 1.1 V, which corresponds to the
specific power inputs 2.63 KWh / Nm3H2.
This is less than 75% of the theoretical
Figure 2.38. Planar SOEC construction
104
2.5.5.3.2 Tubular type:
the electrolyser is a mixture of hydrogen

and steam, and requires additional
processing; thereby further increasing the
capital costs. Moreover, SOEC systems are
not stable against uctuating and
intermittent power sources.
Tubular SOEC has the advantages of

relatively easier sealing, higher thermal
shock resistance enabling better thermal
cycling performance, high mechanical
strength, and soon, and it has the
disadvantages
of
manufacturing
difficulties, low gas volumetric production
rate, and the difficulties in current
collecting. Tubular SOEC is considered to
be one of the promising configurations for
SOEC commercialization applications.
2.5.6 Electrolysis Calculations:

The basic electrolysis reaction is:
1
1 H2O + Electricity = 1 H2 + 2 O2
This reaction is endothermic and the
required energy by the process is:
= +
where, (J/mol) is the change of
enthalpy, (J/mol) is the Gibbs free
energy and represents the minimum
electrical energy and TS (

) is the
process temperature and the entropy
change and represents the minimum
thermal energy demand.
Figure 2.39. Tubular SOEC type
2.5.5.4 Advantages of SOEC:
- Based on the required cell voltage in

the electrolyser can be determined with:
The main advantage of the SOEC is the low

electricity demand as a substantial part of
the energy needed for the electrolysis
process is added as heat, which is much
cheaper than electric energy. In addition,
the high temperature accelerates the
reaction kinetics, reducing the energy loss
due to electrode polarization, thus
increasing the overall system efficiency.
Typical high temperature electrolyser
system achieves 92% electrical efficiency
while low temperature electrolysers can
reach at most 85% efficiency.
where, 2 is the number of electrons and F

is the Faraday constant which is F =
96485.3365
is usually the higher heating value

(HHV) of hydrogen
- The reversible potential (Er) (theoretical
voltage of electrolysis) is given by the
following equation:
= 0
2.5.5.5 Disadvantages of SOEC:
ln
2
Where,
0 : standard equilibrium potential
The biggest challenges for SOEC systems

are the fast material degradation and the
limited long term stability, both of which
are due to high temperature operation
[29]. Furthermore, the high temperature
level means that the product stream from
R: the gas constant

T: absolute temperature
0 and P: vapor pressure of pure water and
electrolyte respectively.
105
- To calculate the equilibrium potential:

0 =
Therefore electrodes with highly active

electrocatalysts
can
reduce
the
overvoltage.
0
2
Ohmic loss is mainly caused by electric

resistance of electrolyte, which can be
reduced by shortening the distance
between anode and cathode. Both
overvoltage and ohmic loss increase with
the increasing current density (current per
unit area of electrode), hence increase of
cell voltage and increase of electric power
to make hydrogen. The water electrolyzers
currently used in industry work at a cell
voltage of 1.8 2.2 volts corresponding to
4.3 5.3 KWh per 3 of hydrogen (at
standard conditions of temperature and
pressure).
Where,
0 is increment of Gibbs free energy
F is Faraday constant (96519.4
coulomb/mol)
EO is 1.226 V under 298 K and 1 atm. This
is slightly higher than the decomposition
voltage of pure water under the same
condition.
Since is theoretical equilibrium
potential, actual cell voltage needed to
continue the reaction is higher with the
addition of ohmic loss of electrolyte and
diaphragm and overvoltage by electrode
reaction.
- The important difference that the SOEC

electrolyser can operate at a cell voltage
below minimum cell voltage of the higher
heating value of hydrogen which
- Actual cell voltage E is calculated

according to the following equation:
= + +
_ =
Where,
141.86 (
= 1.4819 V
E is electrolysis voltage
and in some cases even below the cell

voltage of the lower heating value
which
is
is reversible voltage
is overvoltage
_ =
is ohmic loss.
119.93 (
1.253
where M =2.015 is the molar mass
Reversible potential is theoretically

decided according to the reaction
condition, but overvoltage and ohmic loss
vary with activity of electrodes and cell
design. Overvoltage is loss due to
resistance by the chemical reaction rate.
of hydrogen.
- The amount of hydrogen produced
can be calculated by:
= ( )
2
where p (MWh) is the available energy
and is the electrolyser efficiency
2 =
To drive the chemical reaction of

electrolysis, extra energy is required in
addition to the reversible potential which
corresponds to a zero reaction rate. In
water electrolysis, overvoltage is decided
by the reaction rate at the electrodes.
- The amount of feed in water needed

for producing a certain rate of
106
hydrogen
2 =
is
calculated
by:
PEM electrolyser is a pressurized

operating.
So, the energy needed for compression
is calculated by:
where ( ) is the production

rate of hydrogen, M is the molar mass
of hydrogen (2.015 ) , is the

efficiency of water use and 2 is the
molar mass of water (18.015 )
W=
2 ()
Where, n is the molar amount of

hydrogen, R=8.314 (
- For the SOEC the water is preceded

into an electric steam boiler. The
energy needed for steam generation
can
be
determined
with
=
nRTln( 2 )
) is the
.
2
universal gas constant, ln( ) is the

1
logarithm of the ratio of pressure after

and before compression and is the
compressor efficiency. Compressor
efficiency is around 70 %
(MW)
where h is the enthalpy change for

water between ~100 C and the desired
temperature and is the boiler
efficiency that can be up to 99 %
- SOEC and AEL electrolysers are
operating at atmospheric pressure, but
2.5.7 Comparison between the different types of electrolysis:
Figure 2.40. The different types of electrolysis
107
Figure 2.41. Technical features of electrolysis types
Table 2.7. Efficiencies and specifications of electrolysis types
108
Table 2.8 Summary of the key operational parameters of Alkaline, PEM, and Solid Oxide Electrolysis.
2.6.1 General order of the

restrictions on Hydrogen blends:
2.6 Blending or Injection of

Hydrogen into the Natural
Gas grid
Blending hydrogen into the existing natural
gas pipeline network has been proposed as
a means of increasing the output of
renewable energy systems. Blending
renewable hydrogen with natural gas can
improve the carbon intensity and
sustainability of the final natural gas
product delivered to consumers. If
hydrogen blending in natural gas with
downstream extraction proves to be
economically viable during the early
market growth phase, it could prove to be
viable in the long term as an additional
mode of delivery.
Figure 2.42 Restrictions of hydrogen blending
2.6.1.1 End-use appliances and Safety

concerns:
The conditions determining a maximum
hydrogen blend level that does not
adversely influence appliance operation or
safety vary significantly according to the
composition of the natural gas, the type of
appliance (or engine), and the age of the
appliance. Adding low concentrations of
hydrogen to existing natural gas pipeline
systems, at volumes of 20% or less, results
in a minor increase in the risk of ignition. In
109
instances where natural gas leaks result in

explosions, inclusion of 20% or less
hydrogen would result in minor increases
in the severity of the explosion. The Gas
Technology Institute (GTI) assessment
presented here is based upon data specific
to the U.S. natural gas supply system. They
suggested that higher concentrations of
hydrogen in distribution mains, up to 50%,
present a minor increase in overall risk.
Risks associated with service lines are
different because service lines are often
found in confined spaces where leaked gas
would be more likely to accumulate. If
hydrogen concentrations exceed 20% in
service lines, the increase in overall risk is
more significant than for distribution
mains. For both distribution mains and
service lines, proper risk management
practices, such as the installation of
monitoring devices, reduces overall risk.
However, adding more than 50% hydrogen
to either distribution mains or service lines
results in a significant increase in overall
risk. The study concluded that hydrogen
concentrations up to 28% may safely be

used with the existing domestic
appliances. But, long-term (more than 15
years from now) material compatibility of
domestic appliances with hydrogen and
natural gas mixtures is uncertain. End-use
requirements are generally the most
restrictive conditions on increasing
hydrogen blend levels in natural gas. For
poorly adjusted appliances, no hydrogen
blends would be acceptable. If
implemented
with
relatively
low
concentrations, less than 5%15%
hydrogen by volume, this strategy of
storing and delivering renewable energy to
markets appears to be viable without
significantly increasing risks associated
with utilization of the gas blend in end-use
devices (such as household appliances),
overall public safety, or the durability and
integrity of the existing natural gas pipeline
network.
Figure 2.43. Supply chain schematicz
Four general types of transmission lines

are indicated in the supply chain
schematic, Gathering lines bring natural
gas from various sources to processing
plants, typically high-volume and long-
distance transmission lines deliver gas to

the city gate, and two types of distribution
lines (mains and service lines) deliver it to
local consumers. Underground storage
facilities and large industrial consumers
110
are connected directly to transmission

lines. In terms of supply capacity, the U.S.
transmission line system is supplemented
by natural gas storage capacity to meet
peak demand during the winter heating
season, and the distribution pipeline
system is sized for this peak demand.
and transmission pipelines is location with

respect to populated areas. The level of
hydrogen that is acceptable for
transmission pipelines may need to be
reassessed for distribution systems in
terms of the frequency and severity of fire
or explosion in a highly populated area. In
addition, the hazards arising from gas
leakage in a distribution system may be
more severe than in transmission
pipelines, especially in a confined service
area. The integrity management for
distribution systems under hydrogen
services may require a leak detector or a
monitoring device or sensor. The
maintenance costs for distribution systems
under hydrogen service likely will increase
because these systems will need to be
inspected more frequently and likely will
require additional leak detection systems.
2.6.1.2 Material Durability and Integrity

Management
The durability of some metal pipes can
decreases when they are exposed to
hydrogen over long periods, particularly
with hydrogen in high concentrations and
at high pressures. This effect may be of
concern when hydrogen is injected at high
concentrations into existing high-pressure
natural gas transmission lines. The effect is
highly dependent on the type of steel.
However, metallic pipes in U.S. distribution
systems are primarily made of lowstrength steel, typically API 5L A, B, X42 and
X46, and these are generally not
susceptible
to
hydrogen-induced
embrittlement under normal operating
conditions. There is no major concern
about the hydrogen aging effect on
polyethylene (PE) or polyvinylchloride
(PVC) pipe materials. Most of the
elastomeric materials used in distribution
systems are also suitable to hydrogen. In
most research programs, the focus of
integrity management has been on
transmission pipelines because of
concerns at high operating pressures, up to
2,000 psi (139 bar), and the pipeline steels
that are subject to hydrogen-induced
cracking. Natural gas distribution systems
are very different from transmission
pipelines, and the integrity program for
transmission pipelines does not apply to
distribution systems. One important
difference between distribution systems
2.6.2 Leakage of the gas:

Hydrogen is more mobile than methane in
many polymer materials, including the
plastic pipes and elastomeric seals used in
natural gas distribution systems. The
permeation coefficient of hydrogen is
higher through most elastomeric sealing
materials than through plastic pipe
materials. However, pipes have much
larger surface areas than seals, so leaks
through plastic pipe walls would account
for the majority of gas losses. Permeation
rates for hydrogen are about 4 to 5 times
faster than for methane in typical polymer
pipes used in the U.S. natural gas
distribution system. In general, hydrogen
blends would slightly reduce natural gas
leakage due to the higher mobility of
hydrogen molecules, resulting in a net
reduction in the greenhouse gas impact
due to leakage. Though gas loss from
service lines is economically negligible,
leakage into confined spaces may pose a
111
safety risk. Gas leakage from elastomeric

seals at joints in service lines may also
increase the risk in confined spaces.
Further investigation into specific pipe and
seal materials and systems can provide a
basis to estimate gas leakage more
accurately. This basis can be used to
determine whether leakage in confined
spaces might present a safety risk over
time and the degree to which detection
and monitoring devices may be required to
manage risks.
2.7 Implementation of
Power to Gas (P2G) around
the world:
2.7.1 Germany experience:
Hydrogen production (70 percent energy
conversion efficiency) is just the first step
in a process which will result in the
production of synthetic gas (i.e. methane 55 to 65 percent energy conversion
efficiency) which can replace natural gas.
Existing energy storage technologies like
compressed-air energy Storage (CAES) and
pumped storage hydroelectricity certainly
have a role to play in the future, but alone
they are not able to meet the expected
surpluses of solar and wind energy or
expected energy shortfalls in periods of
low sunshine and wind (future renewable
energy-based domestic power supply
requires a storage capacity between 20 to
40 terawatt hours, current energy storage
capacity is approximately 0.04 terawatt
hours). At present, a compensating
capacity level of 17 GW is already required
to maintain domestic supply and demand
balance. This figure is set to rise to 28GW
in 2025. To date, balancing power has
largely been provided by hydro pumped
storage
and
gas power
plants.
2.6.3 Benefits of Hydrogen

blending:
- Significant reduction of greenhouse gas
emissions if hydrogen is produced from
renewable sources.
- Potential benefits from reducing
petroleum consumption in automotive
applications and improving air quality by
reducing sulfur dioxide, oxides of nitrogen,
and particulate emissions.
- Increasing natural gas amount, when a
hydrogen/natural gas mixture is used in
existing appliances for heat and electricity
generation. This benefit is similar
to increasing the mix of renewable
generation on the electricity grid in
that it does not require significant
changes in end-use equipment.
- Blending hydrogen into natural gas
pipeline networks has also been proposed
as a means of delivering pure hydrogen to
markets, using separation and purification
technologies downstream to extract
hydrogen from the natural gas blend close
to the point of end use.
112
purposes. Model Projects in Germany:

Center for Solar Energy and Hydrogen
Research Baden-Wuerttemberg (ZSW),
SolarFuel GmbH and Fraunhofer Institute
for Wind Energy (IWES). The Center for
Solar Energy and Hydrogen Research
Baden-Wuerttemberg
(ZSW),
in
partnership with SolarFuel GmBH and the
Fraunhofer Institute for Wind Energy and
Energy System Technology (IWES) has
been operating a 25 kilowatt power-to-gas
pilot plant since 2009 (a second research
facility, with 250 KW of electric power, is
scheduled for completion by the end of
2012).
2.7.2 Austria experience:

With increasing amounts of energy supply
by renewable energy sources (wind and
solar), the demand for storage possibilities
increases to compensate for seasonal
fluctuations. Surpluses
from
the
production of renewable energy will be
converted by electrolysis into hydrogen
and thus stored in the natural gas system.
The gas storage (Only 7 bcm
correspondingly 77,000 GWh in Austria)
could with their huge storage volumes to
be repositioned in the energy system of
the future and serve as a balancing storage
for renewable energy.
Figure 2.44. P2G implementation in Germany
2.7.1.1 Germany's Gas Network

Infrastructure:
Germany's generous gas network reserves
more than 400,000 km of pipeline
connecting natural gas reservoirs with a
total storage volume of 23.5 billion cubic
meters and a further 15.2 billion cubic
meters in planning, allow for provision of
approximately one sixth of annual
domestic electricity generation. But
another way, Germany's extant gas
network provides energy storage capacity
of approximately 220 terawatt hours.
Power-to-gas represents a major energy
storage opportunity. Germany is unique in
the fact that hydrogen can be fed into the
gas grid in significant quantities (up to 30
percent grid capacity in some regions).
Germany
enjoys
another
unique
advantage: the presence of salt caves in
wind-intensive regions which are already
being used for natural gas storage
2.7.3 Denmark experience:

Denmark has set ambitious goals for the
de-carbonisation of its economy. The large
and growing share of wind energy will lead
to more frequent and prolonged periods of
electricity supply, creating the need for
bulk energy storage. Denmark is aiming at
procuring 50% of its total power
consumption from domestic wind energy
by 2020 and 80% by 2050. A large share of
intermittent renewables will cause more
frequent and prolonged periods in which
113
electricity supply surpasses electricity

demand. Since supply and demand in the
power grid must be balanced at all time,
any excess electricity needs to either be
exported to neighboring countries or
stored domestically for later use. Powerto-gas allows Denmark to use its state-ofthe-art natural gas infrastructure to store
wind energy in a practically unlimited
manner. P2G is a novel production route
for renewable gas that can be used in
transportation and heating applications.
Denmark is looking to also decarbonize
other sectors of its economy. For instance,
the Danish gas transmission grid operator
has made a public commitment to fully
decarbonize the gas sector by 2050. In
sum, Denmarks unique natural resources
and ambitious renewable targets create a
great environment for the development
and early deployment of the power-to-gas
technology.
hydrogen and oxygen. Both gases are

immediately separated and leave the
electrolyser in separate pipes. The
hydrogen is then
delivered to the methanation reactor, while
the oxygen is supplied to the activated
sludge basins for wastewater treatment.
Due to its fast response time, the
electrolyser is able to provide frequency
regulation services to the local power grid.
It does so by adjusting its power intake (i.e.
increasing or decreasing electricity
consumption) in response to an automated
signal. The signal is triggered whenever the
frequency in the grid approaches the upper
or lower frequency boundary.
In Electro-chaeas biological methanation
system, the hydrogen from the
electrolyser is combined with carbon
dioxide, and this gas mixture is then
introduced to a liquid phase methanation
reactor. Over the course of the
experimental testing phase in the BioCat
Project, two sources of carbon dioxide will
be
used:
(i) Raw biogas from an adjacent anaerobic
digester with a composition of
approximately 60% methane and 40%
carbon
dioxide.
(ii) A pure stream of CO2 supplied by an onsite biogas upgrading system. Inside the
bioreactor, a culture of methanogenic
archaea metabolizes the hydrogen and
carbon dioxide to methane. The specific
strain used by Electrochaeas methanation
system has been selectively evolved (not
genetically modified) in the laboratory of
Prof. Laurens Mets at the University of
Chicago and exhibits properties ideal for
applications in industrial environments.
Those properties include:
2.7.3.1 Denmark project:

Project Name: Power-to-Gas via Biological
Catalysis
(P2G-BioCat)
Size of System: Input: 1 MW of electrical
power
(at
electrolyser)
Duration of Project: February 2014
October 2016
Project Budget: Total cost is DKK 49.9
million (EUR 6.7 million)
Key Components of Power-to-Gas Facility:
The BioCat Project uses an advanced
alkaline electrolyser and an optimized
biological methanation system to generate
renewable gas for the direct injection into a
local 3-bar gas distribution grid. The
electrolyzer is connected to the local power
grid and draws electricity in times of low
electricity prices, which is an indication of a
supply/demand imbalance. In four
different cell stacks, the electricity splits
water into its molecular components
114
Low energy metabolic pathway from

carbon dioxide to methane
Very high carbon mass conversion
efficiency
CNG-enabled vehicles, or as a heat source

in household and industrial applications.
High tolerance to contamination

(oxygen, hydrogen sulfide, particulates)
Moderate operating temperatures (6065C)
High selectivity in product gas (100%
methane, no intermediate products)
Low catalyst operating costs enabled
by self-maintenance
and
selfreplication
Very high responsiveness (ability to
cycle on/off within seconds)
2.8 Techno-Economic
Analysis
2.8.1 Technical analysis:
Based on the Egyptian national natural gas
network data, the capacity of NG is
1083000 cubic meter per day in WDGC,
and a 3% Hydrogen blending is determined
as it is a safe injection blending into the gas
network. Hence, the rate of hydrogen
production needed from the P2G is 0.03
multiplied by 1083000 which equals to
32490 m3/day.
Gas Injection System:

The renewable gas produced by the BioCat
facility will be analyzed for gridspecification (molecular composition,
heating value, Wobbe number, etc)
downstream of the methanation reactor
before it is injected into a 3-bar gas
distribution grid. Being the principal
component of natural gas, methane can be
injected into the natural gas network in a
practically unlimited manner. Once in the
grid, the gas can be used for electricity
production, as a transportation fuel in
The rate of potable water produced from

the desalination which will be sold to
consumers is assumed to be 4 times of
hydrogen production rate.
The rate of brine output from the
desalination process is 30 % of the total
water capacity input to the desalination
process.
2.8.1.1 First Alternative:
Figure 2.45. First alternative construction
115
Mass Balance:
T: Thermodynamic temperature (K)
- For Alkaline electrolyser:
: Van't Hoff factor
2 =
R: gas constant, 8.314 J/(molK)
1
2
2 .
: pump efficiency 80 %
Where,
2 : The molar mass of hydrogen (2.015
).
Q: water flow rate (3 )
. : The efficiency of the electrolyser

74%.
2 : The molar mass of water (18.015
).
Table 2.9. Osmotic pressures based on concentrations
Hence, for sea water = 397 psi

2737218.64413 Pa
27.37 bar
So, we can consider the pressure which
should provide by the RO pump equal to
28 or 30 bar.
- For Reverse Osmosis:

5 = 2 + 3 + 4
,
2 + 3 = 0.7 5
, 3 = 4 1 ,and 4 = 0.3 5
- Seawater pump power
(E8) =
Energy Balance:
1 2
2 .
, Input power (E9) =
where:
: Cell voltage 2.1 Volt
and
panels
Where:
E=
Energy
(kWh
/year)
A = Total solar panel Area (m)
H = Annual average solar radiation on
tilted panels (3200 KWh /m2. year)
: Photovoltaics panel efficiency
: assumed be 0.765
- For Reverse Osmosis:
Output energy from PV

= A * TCF* H * *
F: Faraday constant = 96,485.3375 ()
system:
- To calculate the area of solar system:
2: Number of electrons
power (E6) =
For
Photovoltaic
output from PV = E6 + E7+ E8
- For Alkaline electrolyser:

7 = power (7) =
where:
: The osmotic pressure of solution (Pa)
C: The concentration of solution
(mole/3 )
TCF: Temperature correction factor,

assumed to be 0.8
116
2.8.1.2 Second alternative:
Figure 2.48. Second alternative construction
Figure 2.49. Stand-alone parabolic trough collector solar system components
117
- output Distilled water energy =

Mass Balance:
For
PEM
electrolyser:
1
2 =
2
2 .
. 78 %
For
MSF
process:
5 = 2 + 3 + 4
- For CSP system:
= +
- Energy balance for MSF block:
+ 5 5 5 = 4 4 4
+ 3 3 3 + 2 2 2
- For CSP system:

= 6 + 7 + 8
= + +
9 =
Energy Balance:
For
PEM
Input power (E8) =
For
, assume =0.23
7 =
( )
electrolyser:
2
2
2 .
Assume = 2 volt
-
=
MSF
process:
seawater pump power (E6) =

Q=
- Input Electricity = 6 + 8
and, assume P= 10 bar
118
2.8.1.3 Third alternative:
Figure 2.52. Third alternative construction
- For photovoltaics panels:
Mass Balance:
- For PEM electrolyser:
0
2 =
2
2 .
Output power = 1 + 8
8 =
5 = 2 + 3 + 4
2 + 3 = 0.7 5
- To Calculate the PV panels area:

Output energy from PV panels
= A * TCF* H * * For CSP
Energy Balance:
- For PEM electrolyser:
=
& =
Input power =
- For MSF process:
For CSP panels system:
0 2
2 .
Input
Assume = 2 volt
power
Assume = 0.23
- For MSF process:
Output power ( 6 ) =
(
)
Input steam energy (6 ) =

.
Assume = 300
= 240
- Energy balance for MSF block:

2 2 2 + 3 3 3 + 4 4 4
= 5 5 5 +
( )
- To calculate the CSP panels area:Output

energy from CSP panels (6 ) =
119
2.8.1.4 Forth alternative:
Figure 2.55. Forth alternative construction
Mass Balance:
- For SOEC electrolyser:
0
1 =
2
2 .
.
For
MSF
4 = 1 + 2 + 3
1 + 2 = 0.7 4
Energy Balance:
- For MSF process: Input steam energy (5 )
=
.
-
54
Energy balance for MSF block:

1 1 1 + 2 2 2 + 3 3 3
= 4 4 4 + 5
- For photovoltaics panels:
%
process:
Output power = 7 + 8
7 =
- For SOEC electrolyser:

8 =
Input power =
0 2
& =
2 .
- To Calculate the PV panels area:
Assume = 0.96 Volt
Output energy from PV

= A * TCF* H * * For
- Input steam energy to SOEC:
10
=
()
10
panels
- For CSP system:
= hS/H hsat.
Input power =
hS/H = heat enthalpy of superheated steam

at temperature of 160
Output power ( 5 ) =
(
)
hsat. = heat enthalpy of saturated steam at

temperature of 100
- To Calculate the CSP panels area:

Output energy from CSP panels = E5 + E10
99 %
Output energy from CSP panels =

120
2.8.2 Technical Comparison between the four alternatives:
For production of 0.8113 Kg/sec of Hydrogen:-
Table 2.10. Comparison results between the four alternatives
All proposals are seem equal, so the

economic analysis is needed.
2.8.3.1 Payback Period (P.P):

The payback period is the length of time
required to recover the cost of an
investment. The payback period of a given
investment or project is an important
determinant of whether to undertake the
position or project, as longer payback
periods are typically not desirable for
investment positions. Payback period is
calculated as:
2.8.3 Economic Analysis:

An economic study should be made to
compare between the four scenarios as
the technical study results were similarly
equal, so the decision about which
proposal will be applied depends on the
economic results.
Economic study is made by the finance
accountants to determine if that project is
profitable or not.
P.P =
All other things being equal, the better
investment is the one with the shorter
payback period. For example, if a project
costs $100,000 and is expected to return
$20,000 annually, the payback period will
be $100,000/$20,000, or five years.
That study can be made by many ways such

as: Payback Period (P.P), Discounted
payback Period (D.P.P), Net Present Value
(NPV) and Internal Rate of Return (IRR).
Any profit can be calculated by abstracting
the total expenditures from the total
revenue of the whole project.
There are two main problems with the

payback period method:
The taxes and loan (if exists) should be

taken into consideration.
1. It
121
ignores any benefits that

occur after the payback period
and,
therefore,
does
not
measure profitability.
2. It ignores the time value of money.
2.8.3.3 Net Present Value (NPV):

Net Present Value (NPV) is the difference
between the present value of cash inflows
and the present value of cash outflows.
NPV is used in capital budgeting to analyze
the profitability of a projected investment
or project.
Because of these reasons, other

methods of capital budgeting, like net
present value, internal rate of
return or discounted cash flow, are
generally preferred.
The following is the formula for calculating

NPV:
2.8.3.2 Discounted Payback Period (D.P.P):
A capital budgeting procedure used to

determine the profitability of a project. In
contrast to a NPV analysis, which provides
the overall value of a project, a discounted
payback period gives the number of years
it takes to break even from undertaking the
initial expenditure. Future cash flows are
considered are discounted to time "zero."
NPV = =1 (1+)

Where,
Ct = net cash inflow during the period t
Co = total initial investment costs
r
=
discount
rate
t = number of time periods
A positive net present value indicates that
the projected earnings generated by a
project or investment (in present dollars)
exceeds the anticipated costs (also in
present dollars).
This procedure is similar to a payback

period; however, the payback period only
measure how long it takes for the initial
cash outflow to be paid back, ignoring
the time value of money.
Generally, an investment with a positive

NPV will be a profitable one and one with
a negative NPV will result in a net loss. This
concept is the basis for the net present
value rule, which dictates that the only
investments that should be made are
those with positive NPV values.
Capital budgeting is the process in which a

business determines and evaluates
potential expenses or investments that are
large in nature. These expenditures and
investments include projects such as
building a new plant or investing in a longterm venture. Often times, a prospective
project's lifetime cash inflows and
outflows are assessed in order to
determine whether the potential returns
generated meet a sufficient target
benchmark,
also
known
as
"investment appraisal."
This way of economic study can be used
when the investment likes to be risky if its
continued for a long time, so the investor
wants to know when his money will be paid
back with the profit to go out as fast as
possible from that investment.
Determining the value of a project is

challenging because there are different
ways to measure the value of future cash
flows. Because of the time value of
money (TVM), money in the present is
worth more than the same amount in the
future. This is both because of earnings
that could potentially be made using the
money during the intervening time and
because of inflation. In other words, a
dollar earned in the future wont be worth
as much as one earned in the present.
122
One primary issue with gauging an

investments profitability with NPV is that
NPV relies heavily upon multiple
assumptions and estimates, so there can
be substantial room for error. Estimated
factors include investment costs, discount
rate and projected returns.
To calculate IRR using the formula, one

would set NPV equal to zero and solve for
the discount rate r, which is here the IRR.
Because of the nature of the formula,
however, IRR cannot be calculated
analytically, and must instead be
calculated either through trial-and-error or
using software programmed to calculate
IRR.
A project may often require unforeseen

expenditures to get off the ground or may
require additional expenditure at the
projects end.
Generally speaking, the higher a project's

internal rate of return, the more desirable
it is to undertake the project. IRR is
uniform for investments of varying types
and, as such, IRR can be used to rank
multiple prospective projects a firm is
considering on a relatively even basis.
Additionally, discount rates and cash

inflow estimates may not inherently
account for risk associated with the project
and may assume the maximum possible
cash inflows over an investment period.
This may occur as a means of artificially
increasing investor confidence. As such,
these factors may need to be adjusted to
account for unexpected costs or losses or
for overly optimistic cash inflow
projections.
Assuming the costs of investment are

equal among the various projects, the
project with the highest IRR would
probably be considered the best and
undertaken first.
You can think of IRR as the rate of growth a
project is expected to generate. While the
actual rate of return that a given project
ends up generating will often differ from its
estimated IRR rate, a project with a
substantially higher IRR value than other
available options would still provide a
much better chance of strong growth.
Projects that have a negative net present

value will not have a discounted payback
period, because the initial outlay will never
be fully repaid. This is in contrast to
a payback period where the gross inflow of
future cash flows could be greater than the
initial outflow, but when the inflows are
discounted, the NPV is negative.
One popular use of IRR is in comparing the

profitability of establishing new operations
with that of expanding old ones. For
example, an energy company may use IRR
in deciding whether to open a new power
plant or to renovate and expand a
previously existing one. While both
projects are likely to add value to the
company, it is likely that one will be the
more logical decision as prescribed by IRR.
2.8.3.4 Internal Rate of Return (IRR):

Internal rate of return (IRR) is used
in capital budgeting measuring the
profitability of potential investments.
Internal rate of return is a discount
rate that makes the net present value
(NPV) of all cash flows from a
particular project equal to zero. IRR
calculations rely on the same formula as
NPV does.
In theory, any project with an IRR greater

than its cost of capital is a profitable one,
and thus it is in a companys interest to
123
undertake such projects. In planning

investment projects, firms will often
establish a required rate of return (RRR) to
determine the minimum acceptable return
percentage that the investment in
question must earn in order to be
worthwhile.
2.8.3.6 Inflation:
A sustained, rapid increase in prices, as
measured by some broad index (such
as Consumer Price Index) over months or
years, and mirrored in the correspondingly
decreasing purchasing
power of
the currency. When a dollar buys less than
you would expect it to, we call that
inflation. Inflation is caused by a variety
of factors, but most of them are related to
interest and debt. When the Federal
Reserve bank raises interest rates, it
causes the dollar to inflate. There is
more money in the system, so every dollar
is worth just a little bit less.
Any project with an IRR that exceeds the

RRR will likely be deemed a profitable one,
although companies will not necessarily
pursue a project on this basis alone.
Rather, they will likely pursue projects with
the highest difference between IRR and
RRR, as chances are these will be the most
profitable.
2.8.3.5 Hurdle Rate:
2.8.3.7 Discount Rate:

The discount rate also refers to the interest
rate used in discounted cash flow
(DCF) analysis to determine the present
value of future cash flows. The discount
rate in DCF analysis takes into account not
just the time value of money, but also the
risk or uncertainty of future cash flows; the
greater the uncertainty of future cash
flows, the higher the discount rate.
Present Value= future value / (1+J)n
Where, J is the discount rate and n is
number of years.
A hurdle rate is the minimum rate of

return on a project or investment required
by a manager or investor. In order to
compensate for risk, the riskier the project,
the higher the hurdle rate.
In capital
budgeting,
projects
are
evaluated either by discounting future
cash flows to the present by the hurdle
rate, so as to ascertain the net present
value of the project, or by computing
the internal rate of return (IRR) on the
project and comparing this to the hurdle
rate.
If the IRR exceeds the hurdle rate, the
project would most likely go ahead.
2.8.3.8 Interest Rate:

Interest rate is the amount charged,
expressed as a percentage of principal, by
a lender to a borrower for the use
of assets. Interest rates are typically noted
on an annual basis, known as the annual
percentage rate (APR). The assets
borrowed could include, cash, consumer
goods, large assets, such as a vehicle or
building. Interest is essentially a rental, or
leasing charge to the borrower, for the
asset's use. In the case of a large asset, like
a vehicle or building, the interest rate is
sometimes known as the "lease rate".
For example, a company with a hurdle rate

of 10% for acceptable projects, would most
likely accept a project if it has an internal
rate of return of 14% and does not have a
significantly higher degree of risk.
Alternately, discounting the future cash
flows of this project by the hurdle rate of
10% would lead to a large and positive net
present value, which would also lead to the
project's acceptance.
124
When the borrower is a low-risk party,

they will usually be charged a low interest
rate; if the borrower is considered high
risk, the interest rate that they are charged
will be higher.
2.8.3.10 Capital Expenditures (CAPEX):

Capital expenditures are funds used by a
company to acquire or upgrade physical
assets such as property, industrial
buildings or equipment. It is often used to
undertake new projects or investments by
the firm.
Future Value= Present value * (1+J)n

Where, J is the interest rate and n is
number of years.
This type of outlay is also made by

companies to maintain or increase
the scope of their operations. These
expenditures can include everything from
repairing a roof to building, to purchasing
a piece of equipment, or building a brand
new factory.
2.8.3.9 What is Revenue?

Revenue is the amount of money that a
company actually receives during a specific
period. It is the "gross income" figure from
which costs are subtracted to determine
net
income.
Revenue is calculated by multiplying the
price at which goods or services are sold by
the number of units or amount sold. A
company's revenue may be subdivided
according to the divisions that generate it.
For example, a recreational vehicles
department might have a financing
division, which could be as a separate
source of revenue. Revenue can also be
divided into "operating revenue," or sales
from a company's core business, and "nonoperating revenue," which derives from
other, secondary sources.
2.8.3.11 Operating Expenditures (OPEX):

Operating expenses is a broad line item
that represents the costs of performing
day-to-day business activities. In times of
financial difficulty, operating expenses
become a very important figure and a
major target of cost control.
Operating expenses include most costs
that are recognized even when no sales
occur, such as advertising costs, rent,
interest payments on debt, administrative
salaries, etc.
The difference between a company's

earnings and its revenue is that revenue is
the top line amount of money the
company makes by selling its product or
service, and earnings are the bottom line
profit the company makes after all costs of
goods sold, expenses, taxes, depreciation,
amortization and interest are paid.
2.8.3.12 Selling, General and

Administrative Expenses:
Most companies do not want to externally
report the specific details of employee
salaries, pensions, insurance, marketing
costs, depreciated assets or other business
expenses. To comply with reporting
requirements, accountants sum up all of
these numbers and include them as a
single line item in a figure titled "selling,
general and administrative.
Revenue in our case study is divided into

revenue from the hydrogen production
and another from the potable water that is
sold to consumers.
125
In many cases, the operating expenses of a

company and the selling, general and
administrative expenses, or SG&A, of a
company represent the same thing.
finance
government
activities.
In economics, taxes fall on whomever pays
the burden of the tax, whether this is the
entity being taxed, like a business, or the
end consumers of the business's goods.
Other times, SG&A is listed as a

subcategory of operating expenses on
the income statement.
2.8.4 Economic study results:

A sensitivity study should be done for all
proposals to know the effect of each
component cost on the whole system
internal rate of return.
Both operating expenses and SG&A are

used in financial statements to represent
the expenditures of business incurred
through normal operations. These are
aggregated in contrast with cost of goods
sold, or COGS, which are business
expenditures incurred through the
production process of a saleable item.
2.8.4.1 For First alternative:

Sensitivity of Alkaline electrolyser price:
In case of existence of PV panel with price
of 0.5 $ / watt and land cost of 1 $ / m2.
year
2.8.3.13 Depreciation:
Depreciation is the process by which a
company allocates an asset's cost over the
duration of its useful life. Each time a
company prepares its financial statements,
it records a depreciation expense to
allocate a portion of the cost of the
buildings, machines or equipment it has
purchased to the current fiscal year. The
purpose of recording depreciation as an
expense is to spread the initial price of the
asset over its useful life.
2.8.3.14 Loan or Debt:
Any investment project takes a loan from a
bank which afford the total cost of the
project with the investor. Generally, that
loan is a percentage of the total cost and
that percentage is used to be 70 %.
Table 2.11. Effect of electrolyser price
2.8.3.15 Taxes:
Sensitivity of Hydrogen price per 1 Kg:

In case of existence of PV panel with price
year
Taxes are generally an involuntary fee

levied on individuals or corporations that is
enforced by a government entity, whether
local, regional or national in order to
126
Table 2.14. Effect of PV price at 500 M$ electrolyser price

Table 2.12. Effect of Hydrogen price
When the price of alkaline electrolyser is

400,000,000 $
Sensitivity of PV Panel price:

800,000,000 $
Table 2.13. Effect of PV panels price at 800 M$

electrolyser price

500,000,000 $

electrolyser price

200,000,000 $
127

In case of eliminating the PV panel and land
cost
The electricity price is 2.5 Cent / KWhe

electrolyser price
Sensitivity of Alkaline electrolyser price:

In case of eliminating the PV panel and land
cost
Table 2.18. Effect of hydrogen price when the PV panels

and land are eliminated
2.8.4.2 For Second alternative:

Sensitivity of PEM electrolyser price:
In case of existence of CSP panel with price
year
Table 2.17. Effect of electrolyser price when the PV

panels and land are eliminated
128

year
When the price of PEM electrolyser is

800,000,000 $
Table 2.22. Effect of CSP panels price at 800 M$

electrolyser price
Table 2.20. Effect of hydrogen price

600,000,000 $
Sensitivity
of
CSP
Panel
price:
1,000,000,000 $

electrolyser price

electrolyser price
129

400,000,000 $

In case of eliminating the CSP panel and
land
cost.
Table 2.26. Effect of hydrogen price when CSP panels


electrolyser price
2.8.4.3 For Third alternative:

of 3.5 $ / watt, PV panel with price of 0.5 $
/
watt,
and
land
cost
of
2
1$ / m . year
In case of eliminating the CSP panel and

land cost.
Table 2.25. Effect of electrolyser price when CSP panels

130

/
watt,
and
land
cost
of
2
1 $ / m . year
CSP Panels price dont have a significant

effect, hence assume the worst case of its
price which is 3.5 $ / watt.
800,000,000 $

electrolyser price
Sensitivity of PV price:



600,000,000 $

1,000,000,000
$
Table 2.31. Effect of PV panels at 600 M$ electrolyser

price

electrolyser price
131


In case of eliminating the solar panels and
land
cost.

400,000,000 $
Table 2.34. Effect of hydrogen price when Solar

systems and land are eliminated
2.8.4.4 For Forth alternative:

Table 2.32. Effect of PV panels at 400 M$ electrolyser
price
Sensitivity of SOEC electrolyser price:

/
watt,
and
land
cost
of
2
1 $ / m . year

land
cost.
Table 2.33. Effect of electrolyser price when Solar

132


/
watt,
and
land
cost
of
2
1 $ / m . year
When the price of SOEC electrolyser is

600,000,000 $

electrolyser price

Sensitivity of PV price:

500,000,000 $

800,000,000 $

electrolyser price

electrolyser price
133


land
cost.

400,000,000 $
Table 2.42. Effect of hydrogen price when Solar systems

electrolyser price
2.8.5 Conclusion:
The most effective parameters of the
Internal Rate of Return (IRR) are the
electrolyser price and the hydrogen price
per Kg.
Sensitivity of SOEC electrolyser price:

land cost.
In proposal one, for the same Alkaline

electrolyser price, the existence of
Photovoltaics (PV) solar panels is better
than using the grid electricity which has a
price of 2.5 cent / KWhe
The price of PV panels is an effective

parameter but not as much as the price of
the electrolyser
For the same price of Alkaline (AEL) and
Polymer Electrolyte Membrane (PEM)
electrolysers (in proposal one and two),
proposal one has a much lower hydrogen
cost
Table 2.41. Effect of electrolyser price when Solar
134
In proposal two, eliminating the

Concentrated Solar Panels (CSP) is better
than their existence
Using the PEM electrolyser is better than

using a Solid Oxide Electrolysis Cell (SOEC)
Eliminating the hybrid system when using
the SOEC electrolyser is better than its
existence
i.e. using the grid electricity with price of

2.5 cent / KWhe produces hydrogen with
lower cost than using CSP panels with price
of 3.5 $ / watt
For the same electrolyser and solar prices,

Alkaline and PEM are better than SOEC
In the hybrid systems (i.e. using both of PV

and CSP), the CSP panels are not a much
effective parameter, so we assume the
worst price of CSP panels which is 3.5 $ /
watt
For the same electrolyser price, First

alternative is best choice
These sensitivity tables are no too obvious
as they too much table to be compared.
Hence, existing of a chart that view these
sensitivities is a must, so we made a
tornado and spider charts for each
proposal
For the same PEM electrolyser price, using

a hybrid system is better than using only
CSP panels
Using the hybrid system is better than
using the grid electricity for the same
electrolyser price in the third proposal
2.8.5.1 For first alternative:
Table 2.43. Tornado table for first alternative
Sensit - Sensitivity Analysis - Tornado

Alkaline
Hydrogen price
110
$20
$80
PV Price
-20%
Net Cash Flow
800
$3
-10%
0%
$0.5
10%
20%
Figure 2.58. Tornado chart for first alternative
135
30%
40%
50%
Table 2.44. Spider table for first alternative
Sensit - Sensitivity Analysis - Spider

0.5
0.4
Alkaline
Hydrogen price
PV cost
Net Cash Flow Value
0.3
0.2
0.1
-0.1
-0.2
40%
60%
80%
100%
120%
140%
% Change in Input Value
Figure 2.59. Spider chart for first alternative
136
160%
180%
2.8.5.2 For Second alternative:
Table 2.45. Tornado table for second alternative
Net Cash Flow

PEM
Hydrogen price
1000
$50
$120
CSP
-20%
300
$3.5
-10%
0%
10%
$1.0
20%
Figure 2.60. Tornado chart for second alternative
Table 2.46. Spider table for second alternative
137
30%
40%
50%

0.5
0.4
PEM
Hydrogen price
Net Cash Flow Value
0.3
CSP
0.2
0.1
-0.1
-0.2
40%
60%
80%
100%
120%
140%
160%
180%
Figure 2.61. Spider chart for second alternative
2.8.5.3 For Third alternative:
Table 2.47. Tornado table for third alternative

Net Cash Flow

PEM
Hydrogen price
1000
$44
$95
PV
-20%
$3
-10%
0%
$0.5
10%
20%
Figure 2.62. Tornado chart for third alternative
138
30%
40%
50%
Table 2.48. Spider table for third alternative

0.5
0.4
PEM
Hydrogen price
Net Cash Flow Value
0.3
PV
0.2
0.1
-0.1
-0.2
40%
60%
80%
100%
120%
140%
Figure 2.63. Spider chart for third alternative
2.8.5.4 For Forth alternative:
Table 2.49. Tornado table for forth alternative
139
160%
180%
Net Cash Flow

SOEC
800
Hydrogen price
$50
280
$95
PV
$3
-20%
-10%
0%
$0.5
10%
20%
30%
40%
Figure 2.64. Tornado chart for forth alternative
Table 2.50. Spider table for forth alternative

0.5
0.4
PEM
Hydrogen price
PV
Net Cash Flow Value
0.3
0.2
0.1
-0.1
-0.2
40%
60%
80%
100%
120%
140%
Figure 2.65. Spider chart for forth alternative
140
160%
180%
50%
Figure 2.66. Overall Power to Gas Application scheme
141
Chapter 3 | Power Stations
Chapter 3
Power Stations
142
network pressure on power station

operation.
3.1 Abstract
Egypt has been an industrialized nation.
But, energy shortages have been a
challenge to Egyptian industry.
It is necessary to discuss the gas turbine

cycles and how they operate, the
parameters directly affected by the gas
pressure, the fuel system and the gas
turbine control system.
In the past years, Natural gas industries are

becoming of a great Importance as the
demand of natural gas is increasing in a
huge rate while the production cannot
cover the gap with this huge demand.
3.2 Introduction to Gas

Turbines
This huge demand affects the line pack of

the network resulting in decrease in gas
network pressure.
Gas turbines are used in a wide range of

services. They power aircraft of all types
and drive mechanical equipment such as
pumps, compressors, and generators in
electric utilities. They also generate power
for peak loads and base-load duties.
Studies have found that more than 60% of

the natural gas production in Egypt goes to
industries specially power stations.
In order to optimize the natural gas
network, we have to study the main
consumer of natural gas and its effect on
the performance of the network.
Recently, the interest in gas turbines has

grown significantly in combined-cycle
plants.
The deep study of the gas turbine plant

operation which is directly affected by the
gas provides us with the knowledge
needed to identify the effect of natural gas
These plants use combinations of gas and

steam turbines in various configurations of
turbines, heat recovery steam generators
and regenerators.
Fig. 3-1 Gas Turbine

Figure 0-1
Gas Turbine Construction
143
3.3.4 Distributed power-only
3.3 Applications of Gas

Turbine
Gas turbines are used for distributed

power generation at remote oilfield
facilities and can be used by industry and
utilities alike for portable power
generation. Large industrial facilities install
simple-cycle gas turbines without heat
recovery to provide peaking power in
capacity constrained areas, and utilities
often place gas turbines in the 5 to 40 MW
size range at substations to provide
incremental capacity and grid support. A
number of turbine manufacturers and
packagers offer mobile turbine generator
units in the 5-40 MW size range that can be
used in one location during a period of
peak demand and then transported to
another location for the following season.
Gas turbines are ideally suited for CHP
applications
because
their
high
temperature exhaust can be used to
generate process steam at conditions as
high as 1,200 pounds per square inch
gauge (psig) and 900 F or used directly in
industrial processes for heating or drying.
A typical industrial CHP application for gas
turbines is a chemical plant with a 25 MW
simple cycle gas turbine supplying baseload power to the plant with an unfired
heat recovery steam generator (HRSG) on
the exhaust. This gas turbine CHP system
will produce approximately 29 MWth of
steam for process use within the plant.
Gas turbines are used in a variety of

stationary applications:
3.3.1 Electric utility central station

power generation
Gas turbines are used widely by the electric
utility industry. Combined cycle turbine
plants contribute to base-load power
needs, and simple cycle turbines are used
for meeting peak-load. Today, gas turbines
comprise 32 percent, or 315 GW, of central
station power plant capacity.
3.3.2 Combined heat and power

Gas turbines are used extensively for CHP
applications providing efficient, economic,
and reliable service. Gas turbines comprise
63 percent, or 51.5 GW, of total installed
CHP capacity in the U.S. 38 Close to 80
percent of this gas turbine CHP capacity is
in large combined cycle plants that
maximize the export of power to the
electrical grid 39 The remaining GT CHP
capacity is made up of simple-cycle gas
turbine based CHP systems, typically less
than 40 MW.
3.3.3 Mechanical drive

Oil and gas production, processing,
transmission, and some process industries
use
gas
turbines
for
pumping,
compression, and other mechanical drive
operations.
In
mechanical
power
applications, the turbine shaft power is
used directly. There is no electrical
generator, though there can be heat
recovery, and such systems can be
classified as CHP.
A typical commercial/institutional CHP

application for gas turbines is a college or
university campus with a 5 MW simple
cycle gas turbine. Approximately 8 MWth
of 150 psig to 400 psig steam (or hot water)
is produced in an unfired heat recovery
steam generator and sent to a central
thermal loop for campus space heating
during the winter, or to single effect
144
absorption chillers to provide cooling

during the summer.
Heat is added to the air in combustion

chamber by burning the fuel and raises its
temperature.
3.4 Classifications of Gas

Turbine Power Plants
The combustion products (hot gases)

coming out of combustion chamber are
then passed to the turbine where it
expands doing mechanical work.
Gas turbine power plants are mainly

classified based on:
Part of the power developed by the turbine

is utilized in driving the compressor and
other accessories and remaining is used for
power generation. Since ambient air
enters into the compressor and gases
coming out of turbine are exhausted into
the atmosphere, the working medium
must be replaced continuously.
1. Cycle.
2. Application.
3. Shaft Configuration.
3.4.1 Classification based on Cycle

The gas turbine power plants which are
used in electric power industry are
classified into two groups as per the cycle
of operation.
Advantages of open cycle gas

turbine
1. Open cycle gas turbine.

2. Closed cycle gas turbine.
1. Warm-up time: Once the turbine is

brought up to the rated speed by
the starting motor and the fuel is
ignited, the gas turbine will be
accelerated from cold start to full
load without warm-up time.
2. Low weight and size: The weight in
kg per kW developed is less.
3. Fuel
Flexible:
Almost
any
hydrocarbon fuel from high-octane
gasoline to heavy diesel oils can be
used in the combustion chamber.
4. Open
cycle
plants
occupy
comparatively little space.
5. The stipulation of a quick start and
take-up of load frequently are the
points in favor of open cycle plant
when the plant is used as peak load
plant.
6. Component
or
auxiliary
refinements can usually be varied
to improve the thermal efficiency
and give the most economical
overall cost for the plant load
3.4.1.1 Open Cycle Gas Turbine

A simple open cycle gas turbine consists of
a compressor, combustion chamber and a
turbine as shown in Fig. 3.1.
Fig. 3-2 Open cycle gas turbine
The compressor takes in ambient air and

raises its pressure. No heat is added in
compressor section but compression
process
increases
the
discharge
temperature of the compressor.
145
factors and other operating

conditions envisaged.
7. Open-cycle gas turbine power
plant, except those having an
intercooler, does not require
cooling water. Therefore, the plant
is independent of cooling medium
and becomes self-contained.
3.41.2 Closed Cycle Gas Turbine

Closed cycle gas turbine plant was
originated and developed in Switzerland.
In the year 1935, J. Ackeret and C. Keller
first proposed this type of machine and
first plant was completed in Zurich in 1944.
Disadvantages of open cycle gas

turbine
1. The part load efficiency of the open
cycle plant decreases rapidly as the
considerable percentage of power
developed by the turbine is used to
drive the compressor.
2. The system is sensitive to the
component efficiency; particularly
that of compressor. The open cycle
plant is sensitive to changes in the
atmospheric air temperature,
pressure and humidity.
3. The open-cycle gas turbine plant
has high air rate compared to the
other cycles, therefore, it results in
increased loss of heat in the
exhaust gases and large diameter
ductwork is necessary.
4. It is essential that the dust should
be prevented from entering into
the compressor in order to
minimize erosion and depositions
on the blades and passages of the
compressor and turbine and so
impairing their profile and
efficiency. The deposition of the
carbon and ash on the turbine
blades is not at all desirable as it
also reduces the efficiency of the
turbine.
Fig. 3-3 Close Cycle Gas Turbine
It used air as working medium and had a

useful output of 2 MW. Since then, a
number of closed cycle gas turbine plants
have been built all over the world and
largest of 17 MW capacity is at
Gelsenkirchen, Germany and has been
successfully operating since 1967.
In closed cycle gas turbine plant, the
working fluid (air or any other suitable gas)
coming out from compressor is heated in a
heater by an external source at constant
pressure. The high temperature and highpressure air coming out from the external
heater is passed through the gas turbine.
The fluid coming out from the turbine is
cooled to its original temperature in the
cooler using external cooling source before
passing to the compressor.
The working fluid is continuously used in
the system without its change of phase and
146
the required heat is given to the working

fluid in the heat exchanger.
velocities remain almost constant.

This result in velocity ratio in the
compressor
and
turbine
independent of the load and full
load thermal efficiency maintained
over the full range of operating
loads.
Advantages of Closed Cycle Gas

Turbine
1. The inherent disadvantage of open
cycle gas turbine is the atmospheric
backpressure at the turbine
exhaust. With closed cycle gas
turbine plants, the backpressure
can be increased. Due to the
control on backpressure, unit
rating can be increased about in
proportion to the backpressure.
Therefore, the machine can be
smaller and cheaper than the
machine used to develop the same
power using open cycle plant.
5. The density of the working medium

can be maintained high by
increasing internal pressure range,
therefore, the compressor and
turbine are smaller for their rated
output. The high density of the
working fluid further increases the
heat transfer properties in the heat
exchanger.
6. As indirect heating is used in closed
cycle plant, the inferior oil or solid
fuel can be used in the furnace and
these fuels can be used more
economically because these are
available in abundance.
2. The closed cycle avoids erosion of

the turbine blades due to the
contaminated gases and fouling of
compressor blades due to dust.
Therefore, it is practically free from
deterioration of efficiency in
service. The absence of corrosion
and abrasion of the interiors of the
compressor and turbine extends
the life of the plant and maintains
the efficiency of the plant constant
throughout its life as they are kept
free from the products of
combustion.
7. The closed cycle opens the new

field for the use of working medium
(other than air as argon, CO,
helium) having more desirable
properties. The ratio of the
working fluid plays an important
role
in
determining
the
performance of the gas turbine
plant. An increase in from 1.4 to
1.67 (for argon) can bring about a
large increase in output per kg of
fluid circulated and thermal
efficiency of the plant.
3. The need for filtration of the

incoming air which is a severe
problem in open cycle plant is
completely eliminated.
8. The maintenance cost is low and

reliability is high due to longer
useful life.
4. Load variation is usually obtained

by varying the absolute pressure
and mass flow of the circulating
medium, while the pressure ratio,
the temperatures and the air
9. The thermal efficiency increases as

the pressure ratio decreases.
Therefore, appreciable higher
thermal efficiencies are obtainable
147
with closed cycle for the same

maximum
and
minimum
temperature limits as with the
open cycle plant.
3.2.2.2.2 Aeroderivative gas turbines

As the name indicates, are power
generation units that have origin in
the aerospace industry as the
prime mover of aircraft. These
units have been adapted to the
electrical generation industry by
removing the bypass fans and
adding a power turbine at their
exhaust. The power of these units,
ranges from about 2.5 to 50 MW.
The efficiencies of these units can
range from 35% to 45%.
10. Starting of plane is simplified by

reducing
the
pressure
to
atmospheric or even below
atmosphere so that the power
required for starting purposes is
reduced considerably.
Disadvantages of Closed Cycle Gas

Turbine
3.4.2.3 Industrial-type gas turbines
1. The
system is dependent on
external means as considerable
quantity of cooling water is
required in the pre-cooler.
2. Higher internal pressures involve
complicated
design
of
all
components and high quality
material is required which
increases the cost of the plant.
3. The response to the load variations
is poor compared to the open-cycle
plant.
4. It requires very big heat exchangers
as the heating of workings fluid is
done indirectly.
These turbines vary in range from

about 2.5 to 15 MW. These are
used
extensively
in
many
petrochemical
plants
for
compressor drive trains. The
efficiencies of these units are in the
low 30s.
3.4.2.4 Small gas turbines
These gas turbines are in the range
from about 0.5 to 2.5 MW. They
often have centrifugal compressors
and radial-inflow turbines. The
efficiencies of the simple-cycle
applications vary from 15% to 25%.
3.4.2 Classification based on

Application
The simple-cycle gas Turbine
classified into six broad groups:
is
3.4.2.5 Microturbines
3.4.2.1 Frame type heavy-duty gas
These turbines are in the range

from 20 to 350 kW. The growth of
these turbines has been dramatic
from the late 1990s, as there is an
upsurge in the distributed
generation market.
turbines
The frame units are the large power
generation units ranging from 3 to 480
MW in a simple-cycle configuration,
with efficiencies ranging from 30% to
48%.
148
power turbine shaft; A split shaft

design is advantageous where the
driven equipment has a wide
speed range or a high starting
torque. The air compressor is able
to run at its most efficient speed
while the power turbine speed
varies with the driven equipment.
The split shaft design allows a
much smaller starting system
since only the air compressor
shaft is accelerated during the
start cycle.
3.4.2.6 Vehicular gas turbines

These turbines have ranged from
300 to 1,500 HP. The first vehicular
turbine was built in 1954 by
Chrysler Corporation and followed
by the Ford Motor Companys truck
engine.
3.4.3 Classification based on Shaft

Configuration
3.4.3.1 Single shaft
In a single shaft design, all rotating
components of the gas turbine are
mounted on one shaft; The single
shaft design is simpler, requiring
fewer bearings, and is generally
used where the speed range of the
driven equipment is narrow or
fixed (as in generator sets). It
requires a powerful starting
system since all the rotating
components (including the driven
equipment) must be accelerated
to idle speed during the start
cycle.
Fig. 3-5 Split shaft gas turbine
3.5 Gas Turbine Design

Considerations
1. High efficiency.
2. High reliability and thus high
availability.
3. Ease of service.
4. Ease
of
installation
and
commission.
5. Conformance with environmental
standards.
6. Incorporation of auxiliary and
control systems that have a high
degree of reliability.
7. Flexibility to meet various service
and fuel needs.
Fig. 3-4 Single shaft gas turbine
3.4.3.2 Split shaft

In a split shaft design, the air
compressor rotating components
are mounted on one shaft, and the
power
turbine
rotating
components are mounted on
another shaft. The driven
equipment is connected to the
3.6 Part-Load Performance

When less than full power is required from
a gas turbine, the output is reduced by
lowering the turbine inlet temperature. In
addition to reducing power, this change in
149
operating conditions also reduces

efficiency. Figure 3-6 shows a typical part
load debate curve. Emissions are generally
increased at part load conditions,
especially at half load and below.
Fig. 3-7 T-S diagram of Brayton cycle
(1-2) Isentropic compression process

02 = 12 01
Fig. 3-6 Part load performance
02
01
3.7 Thermodynamic
Analysis of Gas Turbines
=(
02 1
01
Considering
efficiency:
Gas Turbines basically operate on Brayton

Cycle which is made up of four internally
reversible processes:
02 = 01 +
= ( )
compressor
isentropic
02s 01
= 12 (02 -01 )
(1-2) Isentropic compression in the

compressor section.
(2-3) Constant-pressure heat addition in
combustion chamber.
(2-3) Constant-pressure heat addition
03 =
(3-4) Isentropic expansion in the turbine

section.
(4-1) Constant-pressure heat rejection
(Exhaust)
To simplify the analysis of the gas turbine,
air-standard assumptions should be
considered.
. + . 23 . 02
. 34
34 03 23 02
.
= 34 03 23 02
(3-4) Isentropic expansion process
= ( )
= ( )
Considering turbine isentropic efficiency:

03 04
04 = 03
150
= g 34 (03-04 )
=
(4-1) Constant-pressure heat rejection
= 41 (04 01 )
Fig. 3-8 Brayton cycle with Regeneration
Cycle Efficiency:
=
This approach, which is becoming less

common, allows the efficiency of the gas
turbine to be improved by 15-20%.
Disadvantages are the increased capital
cost and the fact that there are increased
pressure losses with the newer highpressure ratio compressors. Instead, many
installations now use the exhaust heat for
combined cycle and regeneration.
3.8 Cycle Improvements

In order to improve the efficiency of the
basic gas turbine cycle, three methods, or
a combination of them, can be
implemented. As simple cycle gas turbines
are improving in efficiency, these cycle
improvements
are
becoming
less
necessary as combined cycle, utilizing
waste heat for other purposes, is becoming
more prevalent. Since these cycle
improvements are generally not applied to
aircraft engines because of the weight
penalty involved, aero-derivative engines
are usually not able to use them.
3.8.1 Regeneration
The most common cycle improvement is
regeneration where exhaust heat is used
to increase the temperature of
compressed air before combustion. This is
accomplished by installing a heat
exchanger in the exhaust to preheat the air
between the compressor and the
combustors.
Fig. 3-9 P-V diagram of Brayton cycle Regenration
3.8.2 Intercooling
In some gas turbine arrangements, the
compression of the inlet air is done in two
stages with the air being cooled between
the stages in a heat exchanger
(intercooler).
Since
isothermal
compression (compression without an
increase in temperature) takes less work
than adiabatic compression (compression
where no heat is removed so that the air
temperature increases), more of the
151
turbine power will be available for the

output load.
The effect of this reheating is to increase

the energy content of the gases and thus
improve the thermal efficiency of the
cycle.
As a result, less air must be compressed to
do the same amount of work. The
reheating is done by burning fuel in a
second combustion chamber using the
excess oxygen content of the gases from
the high-pressure turbine for combustion.
3.8.5 Combined Cycle

Gas turbines can be integrated into a wide
variety of combined cycle or cogeneration
systems. Many of these are in use today,
and their utilization is increasing. These
systems usually extract the exhaust heat
from the gas turbine by means of a heat
exchanger, usually a type of boiler. Steam
is produced, which can be used to drive a
steam turbine, or to provide steam for
process plant purposes or other heating
applications. The steam turbine can be
connected to the same generator as the
gas turbine, another generator, a
compressor or some other mechanical
drive. An example is shown in Fig.
Fig. 3-10 Intercooling
The beneficial effect of intercooling

decreases as pressure ratio increases. A
high-pressure ratio also means that losses
through the intercooler become more
significant. Using an intercooler is more
beneficial if it is combined with
regeneration as more of the exhaust heat
will be recovered and this will improve the
overall cycle efficiency.
The intercooler is a shell and tube heat
exchanger similar in construction to the
regenerator. Cooling water passes through
the tubes while the air passes over the
outside of the tubes. In some cases, the air
may pass through tubes surrounded by
water.
3.8.3 Reheat
In addition to compressing the air in two
stages and intercooling between these
stages, the gas turbine plant may also be
arranged to expand the hot gases in two
stages, with the gases being reheated
between the stages. The gases are
expanded first in a high-pressure turbine
and then reheated before entering a lowpressure turbine.
Fig. 3-11 Combined Cycle Diagram
152
increases with decrease of the ambient

temperature. So, the fuel mass flow rate
will increase, since air to fuel ratio is kept
constant. The power increase is less than
that of the inlet compressor air mass flow
rate; therefore, the specific fuel
consumption increases with the increase
of ambient temperature this occurs
because of increased losses due to the
increased amount of flue gases.
3.9 Parameters that affect

gas turbine performance
3.9.1 Effect of ambient temperature
Figure 3-12 shows that the gas turbine
thermal efficiency is Affected by ambient
temperature due to the change of air
density and compressor work.
Fig. 3-12 Effect of compressor inlet temperature and

compression ratio on thermal efficiency
A lower ambient temperature leads to a

higher air density and a lower Compressor
work that in turn gives a higher gas turbine
output power as shown in Figure 3-13.
Fig. 3-14 Effect of ambient temperature and air to fuel

ratio on thermal efficiency
However,
increased
the
ambient
temperature an increased the air to fuel
ratio caused increased of specific fuel
consumption and heat rate as shown in
Figures 3-15 and 3-16.

ratio on power (KW)
Figure 3-14 shows that when the ambient

temperature increases the thermal
efficiency decreases. This is because, the
air mass flow rate inlet to compressor

ratio on SFC (Kg/KW)
153

ratio on heat rate
Fig. 3-18 Effect of compressor inlet temperature and

compression ratio on compressor work
3.9.2 Effect of compression ratio
Figure 3-19 shows the effect of

compression ratio on the thermal
efficiency with variation isentropic
compressor efficiency. Note that the
thermal efficiency is increased with
compression
ratio
and
isentropic
compressor efficiency. Also the increased
in compression ratio and isentropic turbine
efficiency caused increased in the thermal
efficiency as shown in Figure 3-20.
Figure 3-17 represents a relation between

the gas turbine cycle thermal efficiency
versus compression ratios for different
turbine inlet temperature. It can be seen
that the thermal efficiency increases with
compression ratio at higher turbine inlet
temperature. The deviation of thermal
efficiency at lower compression ratio is not
significant while the variation at higher
compression ratio is vital for thermal
efficiency.
Fig. 3-19 Effect of compression ratio and efficiency of

compressor on thermal efficiency
Fig. 3-17 Effect of compression ratio and turbine inlet
temperature on thermal efficiency
Figure 3-18 present the effect of

compression ratio on compressor work,
this work increase with increase the
compression ratio and the Ambient
temperature.
154
Figure 3-22 Effect of isentropic turbine efficiency air to

fuel ratio on thermal efficiency
Fig. 3-20 Effect of compression ratio and efficiency of

turbine on thermal efficiency
3.9.4 Conclusion
The simulation result from the modeling of
the influence of parameter showed that
compression ratio, ambient temperature,
air to fuel ratio and turbine inlet
temperature effect on performance of gas
turbine power plant. The results were
summarized as follows:
3.9.3 Effect of compressor and

turbine efficiencies
Figures 3-21 and 3-22 highlight the effect
of the compressor and turbine isentropic
efficiencies on the thermal efficiency for
different air fuel ratio. The thermal
efficiency increase with increase the
compressor and turbine isentropic
efficiencies, this is mean the thermal losses
have been reduced in both compressor
and turbine respectively, this lead to
increased power output.
The
compression
ratios,
ambient
temperature, air to fuel ratio as well as the
isentropic efficiencies are strongly
influence on the thermal efficiency of the
gas turbine power plant
The variation of thermal efficiency at
higher compression ratio, turbine inlet
temperature and ambient temperature are
very important.
The thermal efficiency and power output
decreases linearly with increase of
ambient temperature as well as the air to
fuel ratio.
The specific fuel consumption and heat
rate increase linearly with increase of both
ambient temperature and air to fuel ratio.
Fig. 3-21 Effect of isentropic compressor efficiency air to

fuel ratio on thermal efficiency
The peak efficiency, power and specific

fuel consumption occur at higher
compression ratio with low ambient
temperature.
155
2. Programmable
3.10 Gas turbine control

and auxiliary systems
(PLC)
3. Distribution control system (DCS)
The gas turbine control system determines

what happens during all operating
conditions including start-up, run and
shutdown. The turbine must also have
auxiliary support system that keeps the
keep the turbine operating. Auxiliary
systems and control system can become
extremely complex. Every operation and
function of the turbine is monitored and
will have a shutdown sequence if there is a
malfunction in the system. These
protection
systems
monitor
the
performance of the turbine. Any
malfunctions are detected engine will
shutdown by the appropriate system.
Most of the controls systems now are

Programmable logic control system (PLC)
because it very accurate, more control
could be achieved, and the facility to
change the programs any time if needed.
The Distribution control system (DCS) is
commonly used in a large plants which
have multi deferent operating units, each
one of them perform deferent functions
and the DSC is used to indicate, control and
maintain the operation synchronization
between all of these units.
The relay logic control is an old fashion
control system however it is very reliable
and it is not so complicated like PLC or DCS,
also it is often used as backup control
system in case of main control system
failure.
Auxiliary systems help the turbine run

more efficiently within its design
specifications.
The main systems are:

1.
2.
3.
4.
5.
6.
Governor control
Lubrication system
Fuel system
Starting system
Cooling system
Instrumentation and control
3.10.1.2 Control System function

The control system provides automatic
operation of the turbo-machinery systems
to meet the user demand. It allows
starting, stopping, loading, and unloading
either manually or automatically, and from
local or remote locations. The function
provided by the control system is:
3.10.1 Governor control

The governors purpose is to automatically
control the turbine speed under varying
loads. The governor provides automatic
regulation of fuel and compressor vanes
(where applicable) for both starting and
normal operation of the gas turbine.
1.
Sequencing
2. Control
3. Protection
4. Display
There many types of control systems, the

most common types are :
1.
logic control system
Relay logic control system (RLC)
156
Initially, the operator at the loading station

sets the desired engine speed. This
establishes a speed reference setting in the
reset mechanism. If engine speed is above
or below the setting, a proportional error
signal is transmitted to the metering valve
positioner. The valve is repositioned until
the error is eliminated and the power
turbine is at the desired output.
3.10.1.3 The component of the control

system
The control system includes the
instrumentation and control elements
installed in the system, the main control
consol, and complete electrical system; all
of these components could be electrical,
electronics, electromechanical, or electrohydro-mechanical.
Various override trims, such as over

temperature and over speed protection for
the gas producer, are applied on the
governor reset mechanism through the
loading station to produce an artificial
error signal until the adverse condition is
overcome.
For example, the microprocessor is an

electronic component, also the solenoid
valve is an electrometrical device.
3.10.1.4 Functional Description
Control of a gas turbine is affected by
varying the fuel flow to the fuel nozzles
according to operating requirements and
conditions. Although the final control
system configuration varies according to
the application and type of fuel used, the
block diagram, Fig. 3-23, shows some of
the basic control units common to both
liquid and gaseous fuel systems. The
supply of pressurized fuel may come from
a liquid fuel reservoir and engine-driven
gear pumps incorporating a bypass system,
or in the case of gaseous systems, directly
from a gas supply line through a pressure
regulator.
To avoid compressor surge during

acceleration, a separate signal, applied
directly to the metering valve positioner,
overrides the governor error signal. The
acceleration fuel flow schedule is based on
the rise in compressor outlet pressure.
An automatic emergency trip system
closes the main fuel shutoff valve if:
1.
The lubricating oil pressure drops

to an unsafe value
2. The
exhaust temperature exceeds

the allowable limit
The gas turbine unit shown in the block

diagram is a dual shaft machine with
separate compressor drive (or gas
producing) turbine and power turbine.
3. The
rotor speed of Gas Generator

(NGG) or Power Turbine (NPT)
becomes excessive
All-speed governing of the power turbine is

used to control the speed and power
output of the unit. A reset mechanism in
the governor determines the position of
the fuel-metering valve and hence,
controls the engine fuel flow.
4. Fire
occurred
enclosure
in
the
turbine
5. Failure of the main control processor
157
Fig. 3-23 Gas Turbine Control System

Manual shutdown is also affected through
this circuit by means of a manual hydraulic
valve.
turbine stage or in between the

engine turbine and power turbine
6. Vibration,
using accelerometers
mounted on the engine case if antifriction bearings are used (applies
to most aero-derivatives)
3.10.1.5 Instrumentation
The following instrumentation is normally
provided for control and monitoring of a
gas turbine:
1.
7. Vibration
using
eddy-current
displacement probes if journal or
tilt-pad bearings are used (applies
to most heavy-duty gas turbines)
Rotor speed for each shaft, in RPM

(revolution per minutes)
2. Air
inlet temperature in C,
preferably after the intake filter but
often ambient air temperature.
8. Bearing temperatures in C if
journal
or tilt-pad bearings are used
(applies to most heavy-duty gas
turbines)
3. Differential
pressure across the

intake filters
9. Fuel
gas flow,
temperature
4. Compressor
discharge pressure
measured at the exit of the
compressor
and
before
combustion, in kPa
5. Exhaust
gas temperatures in C,
usually measured at multiple
circumferential points and as an
average after the first or second
158
pressure
pressures
and
10.
Oil
system
temperatures
and
11.
Generator output, in kW, or

compressor shaft power
3.10.1.6 Operator Interface
3.10.2.1 Lube Oil System
With every control system, there is an

operator interface (sometimes called an
MMI Man Machine Interface or HMI
Human Machine Interface). It allows an
operator to:
Lube oil system ensures that the gas

turbine receives clean and filtered oil
under certain conditions of pressure and
temperature.
1.
Almost all gas turbines have a lube oil

system that lubricates the bearings that
support the rotor or rotors.
Startup or shutdown the gas

turbine
Aero-derivative gas turbines normally use

antifriction bearings which require only a
small lube oil system.
2. Control the turbines speed

3. Modify the control system logic, with
special access only
Heavy-duty gas turbines use radial

bearings that necessitate a larger lube oil
system.
4. Monitor measured parameters
3.10.1.7 Startup and Shutdown
Micro turbines are an exception and,

because of their small size, are able to
operate with air bearings that do not
require a lube oil system.
Sequencing
The startup and shutdown of a gas turbine
may occur automatically if predetermined
conditions occur. For example, a backup
power generation unit may start if there is
an increase in demand and a compressor
may start if the pressure drops in a process.
Often, operators monitoring the overall
process will initiate a manual start. Once
the start or shutdown is initiated, the
sequencing is almost always automatic.
Gas turbine installations may have more

than one lube oil system. These are the
major configurations:
Some heavy-duty gas turbines have a
single integrated lube oil system that
serves the gas turbine, power turbine,
gearbox and driven equipment
(compressor or generator)
2. Other heavy-duty gas turbines have a
lube oil system for the gas turbine and
power turbine but incorporate a
separate lube oil system for the load
device
3. Most
aero-derivatives
require
separate lube oil systems for the
engine and for the power turbine and
load
1.
3.10.2 Lubrication system

The lubrication system supplies oil to the
bearings for lubrication and cooling.
Depending on the turbine size, the lube oil
may be used for hydraulic operation of its
control.
turbine because of their high quality and

better resistance to high temperatures and
oxidation than petroleum based oils.
3.10.2.2 Lubricants
Synthetic oils with low viscosity and high
viscosity index are recommended for gas
3.10.2.3 Functions lube oil systems
1.
2.
Lubricate sliding surfaces in the

bearings
159
Cool the bearings especially those

located close to the combustion and
turbine sections of the gas turbine
3.10.2.4 Components of lube oil system
end of the compressor, before the turbine

and after the power turbine. Thrust
bearings will be positioned at the front end
of the compressor and next to the power
turbine bearings (one for each shaft)
An oil reservoir to ensure an

adequate supply of oil
2. Filters to ensure the oil is clean
3. Pumps to provide pressure
3.10.2.5 Bearings
1.
Gas turbines use two different types of

bearings:
1. Antifriction
(roller and/or ball)

bearings which are common in
aero-derivative
gas
turbines
because rotors are light enough
2. Radial
(journal or tilt-pad) bearings

for heavy-duty gas turbines that
have heavier rotors.
Fig 3-25 Typical Radial Tilt-Pad Bearing
Fig 3-24 Antifriction bearing
The Antifriction Bearing has two shafts

(rotors) and uses a total of 5 bearings two
thrust (ball) bearings and three radial loads
(roller) bearings.
Figure 3-26 Typical Tilt-Pad Thrust Bearing
This oil sysem is divided into two sections:

a supply system and a scavenge system.
The scavenge system returns the oil from
the bearings to the supply and treating
equipment. All piping, fittings and
reservoir are Type 304 Stainless Steel to
prevent corrosion. The system uses
synthetic
oil
suitable
for
high
temperatures.
Heavy-duty gas turbines require bearings,

which can take higher loads. Although
standard journal bearings have been used
in the past, the most common type of
radial bearing used today is the tilt-pad
bearing. Fig. 3-30 shows a bearing, which
has five tilting pads on individual pivot
pins. A tilt-pad thrust bearing can be seen
in Fig. 3-31. On a typical dual-shaft heavyduty gas turbine, bearings will be located
on the front end of the compressor, at the
The oil reservoir contains approximately

500 litres in a 568 litre tank. Protection
devices are fitted against low oil level and
low oil temperature. A thermostatically
160
controlled heater is included and ensures a

minimum temperature is maintained while
the unit is not operating, to facilitate easy
starting.
the required pressure to the bearings.

After the pump, the oil is filtered by a
duplex, full flow filter that allows filter
change out while running. High oil
temperature, low oil pressure and high
filter differential pressure switches protect
the oil supply.
A positive displacement pump driven by an

auxiliary gearbox on the engine provides
Figure 3-27 Typical Lube Oil System
The oil flows through the bearings and

accumulates in the bearing sumps. The oil
temperature is measured at each scavenge
line in case of bearing problems.
duplex, water-cooled coolers that are

thermostatically controlled. The oil then
flows back to the reservoir.
Chip detectors are often located in the

sumps to detect metal particles from the
bearings. If a bearing becomes damaged,
metal particles will become entrained in
the oil. The chip detector is basically a
magnet that attracts these metallic
particles and detects when they
accumulate on the magnet. Upon alarm,
the detector is removed and inspected to
diagnose the type and extent of bearing
damage.
Heavy-duty Gas Turbine Lube Oil System

The lube oil system shown in Fig. 38 is
typical for a heavy-duty gas turbine with
one integrated oil system for the gas
turbine, gearbox and driven equipment.
The oil reservoir is much larger than for
aero-derivative gas turbines. It normally
contains mineral oil, which does not have
as high a temperature capability as
synthetic oil, but is more economic.
Generally, oil temperatures are not as high
in heavy-duty gas turbines since the oil
flow is greater. Oil heating may be supplied
if required.
Scavenge pumps, also driven by the

auxiliary gearbox of the turbine, provide
the pressure for the oil to flow through
another set of filters and then through
161
Oil pressure is supplied during normal

operation by a main lube oil pump driven
from the accessory drive mounted on the
front of the compressor shaft. Prior to
startup and on shutdown, a motor driven
pre/post lube oil pump supplies oil
pressure. This pump runs for a period of
time after shutdown to cool and lubricate
the bearings and prevent damage. A third
pump using another source of energy (for
example a direct current motor supplied by
batteries) is available as backup in case of
power loss or pre/post lube oil pump
failure.
There is a differential pressure alarm and

pressure gauge. The oil is then cooled prior
to entering the bearings by either an air or
water cooler. At the lube oil header, there
is protection against high oil temperature
and low oil pressure.
The oil drains back into the oil reservoir
using gravity. The oil temperature is also
normally measured in the drains to
monitor bearing health.
The oil is cleaned with duplex filters that

allow filter replacement during operation.
Figure 3-28 Lube Oil System for a Typical Gas Turbine
162
Hydraulic System
7. Taking
oil samples regularly for

analysis and replacing oil when
required
A hydraulic pump is sometimes provided

after the main lube oil pump to supply high
pressure oil for the actuation of the main
turbine instrumentation for controlling
speed and load. The location of the
connection for this point is shown in Fig. 5
by an asterisk (*), located just after a filter
branching off of the main lube oil line,
immediately downstream of the main lube
pump. The hydraulic system is typically the
medium through which the turbine speed
sensor sends its signal to the speed control
valve that modulates the fuel flow to the
turbine, as well as to the variable inlet and
stator vanes, and bleed valves.
8. Calibrating
instrumentation
testing protective devices
System temperatures and pressures,

and the status of pumps, filters and
coolers, are usually monitored and
displayed on computer screens. This
type of interface may also be used to
start or stop the lubrication and
hydraulic system pumps.
3.10.3 Fuel system

In gas turbines there are three main fuel
systems; fuel gas system, liquid fuel system
and dual fuel system.
3.10.2.6 Oil System Maintenance

Oil systems are relatively maintenance
free and automatic protection is usually
provided against common problems.
Maintenance consists of:
1.
3.10.3.1 Fuels for gas turbines

3.10.3.1.1Solid fuels
(experimental)
1. Coal
2. Wood
3. Peat
Checking for oil leaks (usually

daily)
2. Monitoring
oil pressures
temperatures (usually daily)
3. Checking
and
3.10.3.1.2 Natural gas fuel

(Methane and some Ethane)
Natural gas is the ideal fuel for gas turbine
but not practical for mobile applications.
The gas is very clean so no contaminants to
cause blade deposits, also no vaporization
is required.
chip detectors when they
alarm
4. Topping
up the oil reservoir or

secondary lube oil tank. Often a
second supply tank with an
automatic slow fill valve and level
control keeps the reservoir filled.
5. Changing
oil filters when

differential pressure alarms
and
3.10.3.1.3 LPG (Propane)

LPG has excellent burning characteristics
like natural gas, however the cost is more
than the natural gas
the
3.10.3.1.4 Liquid fuels

1.
2.
3.
4.
5.
6.
6. Cleaning the cooler externally
163
Alcohol
Kerosene
Gasoline
Stove Oil
Diesel
Heavy residual oil
7. Blend of diesel and gas.
out through the tubes. Impurities being

heavier fall down & collected in drain area.
Then gas now has two paths either goes to
set of regulators directly or goes to two fire
tube boiler (set to be one at service and
one is standby) to raise the temperature of
gas.
3.10.3.2 Fuel Gas System

Fuel gas system controls the fuel gas flow
by regulating fuel gas control valves in
accordance with fuel demand from gas
turbine controller.
Note: The fire tube boilers are not used in

the gas regulating station because the gas
compressor
already
raises
the
temperature of gas so both of the fire tube
at out of service.
So, the gas mainly goes to set of regulators
three regulators at three parallel lines and
the three regulators are at service at the
same time to adjust the pressure of gas at
reasonable value around 26 bar.
Also it ensures precise fuel gas distribution

control DLN Combustor (Dry Low NOx
Combustor).
Case Study Sedi Krier Power Plant

The gas comes from GASCO with high
pressure (it used to be 40 bar but now
about 30 bar) then gas pass through gas
reduction area to reduce the pressure to
reasonable value (around 26 bar) which is
suitable for Gas compressors.
Gas compressor compresses the gas
pressure to 37 bar then it divided into the
two gas turbines then it passes through the
TCA area to increase the gas temperature
then it separates into pilot line and main
line then pass through set of valves
(control and pressure) then divided into 20
lines for each combustor.
There is a newly added line called bypass,

it is a manual valve that bypasses the
regulating station, it is opened when the
gas pressure from supplier (GASCO) is low.
Then gas passes through three flow meters
(two at service and one standby) then it
goes to the gas compressor area.
3.10.3.2.1 Gas Regulating Station

The gas comes from GASCO and passes
through manual isolation valve then a
hydraulic valve which is usually closed and
the hydraulic valve has a bypass valve
which is open. Then the gas pass through
two knockout drums (one at service and
the second is in standby).
Knockout Drum (KOD) is used to remove
solid and liquid particles. It consists of tube
bundles assembled with multiple small
diameter cyclone tubes. Gas enters
through the tubes causing a circular
motion of tube bundles, which in turn
throw solid and liquid particles against wall
of cyclone tubes. Clean gas rises
164
Fig. 3-29 Fuel gas System
3.10.3.2.2 Fuel valves

After the gas pass through the regulating
unit, it goes to the compressors area but it
passes through set of filters and knock out
before it goes directly to the compressors.
separators utilize replaceable, high

efficiency and coalescing filter elements.
Solid particles are trapped within the
filtration media, while the liquids are
coalesced by filter element from small
droplets into large droplets. By gravity, the
droplets settle in the collection sump and
then discharged into drain tank by means
of control valves.
Each filter has the following:
1. Pressure indicator.
2. Differential pressure indicator
attached with differential pressure
switch (The alarm will be triggered
if p=certain value).
3. Level indicator.
4. Two pneumatic control valves at
the collection sump and each
control valve has inlet/outlet
manual isolation valves and one
manual bypass valve used when
there is a problem with the control
valve.
5. Level switch (if the alarm is
triggered it means that the level
increased and the control valve
didnt open so it might be a
First, gas passes through a shut off valve

used to isolate the gas to the compressors.
NOTE: This valve is pneumatic and it has no
control from DCS so, if we want to close
this valve we should cut (close) the
Instrument air to this valve (shut off valve
is normally closed) and there is a knife
valve exist on the skid of the shut off valve
(if we want to close the shut off valve we
close the knife valve to cut (close) the
instrument air to this valve.
Second, the gas passes through two gas
filters (one at service and one standby).
3.10.3.2.3 Gas filtration skid
Two fuel gas filter separators are installed
to remove solid particles and liquid
droplets from fuel gas. The filter
165
contain sulfur and other metals, which may

corrode the internal parts of the turbine.
problem with the control valve so

bypass valve should be opened).
6. Safety valve on each filter (set
point=30bar).
3.10.3.4 Dual Fuel System

Some gas turbine installations feature dual
fuel capability so that the operator can
switch to a less expensive fuel or as a
backup. Liquid fuel is entering at the top
left of the system, and fuel gas at the lower
right.
3.10.3.2.4 Gas Compressors

Third, gas goes to three gas compressors.
The gas compressors, two of them are in
operation and one is standby.
Gas compressors are controlled by Recycle
control valve (RCV), when the RCV is fully
closed the compressor is at full load.
The lead compressor is the one that its
recycle is fully close at full load while the
lag compressor is the one that its recycle is
not fully close at full load. If the
compressor inlet pressure is decreased to
21.5-22 bar the inlet bypass valve will open
to increase the inlet pressure, then close
again.
If the compressor inlet pressure is
increased to 30 bar, the vent valve will
open until the pressure is decreased then
the vent valve close.
Gas compressor is controlled using two
controllers; supply press control (also
called feedback controller) and fuel gas
flow rate (also called feed forward
controller).
The electronic control unit in the center of

the system selects which metering and
control system (liquid or gas) will be in
operation, and therefore which fuel will be
sent to the burners. A special fuel nozzle is
required and the control system is more
complex to manage the two types of fuels
and to accommodate the switchover
between them. With some systems, a
mixture of gaseous and liquid fuels can be
burned simultaneously.
Compressor data
Type: centrifugal compressor
Actual inlet flow: 364 to 4338 m3/h
Inlet temperature: 300C
Discharge: 37.5 bar
Main motor speed: 2980 RPM
Maximum speed of pinion shaft: 27000 RPM
Lube system: 2.2 bar & 540C
Oil filling: 1250 liter
Cooling water supply pressure: 5 bar
Fig. 3-30 Dual Burner
3.10.3.5 Fuel system maintenance

Routine Maintenance
Fuel gas systems are relatively
maintenance free and automatic
protection is usually provided against
common problems. Maintenance consists
of:
3.10.3.3 Liquid Fuel System

For liquid fuel a storage tank is required,
fuel pump, and filters. The filters are
required to remove solids in the fuel as
they could damage the blades. Some fuels
1. Checking for fuel gas leaks (usually

daily)
166
2. Monitoring
pressures
and
temperatures (usually daily)
3. Changing fuel filters at the required
differential pressure
4. Calibrating
instrumentation
and
testing protective devices
Fig. 3-31 and 3-32 illustrate a pneumatic

starter system. In this case a dual starter
configuration is used. The user has the
option of using air or gas depending on the
most convenient source. The exhaust of
the starter motor is normally vented to the
atmosphere. The starter transmits power
to the front end of the gas turbine rotor by
means of an overrunning clutch. The
shutoff valve is also pneumatically
operated.
3.10.4 Starting system

A gas turbine is not self-starting, thus a
starting motor in an essential auxiliary item.
Gas turbines must have a source of power
(other than the compressor or load
turbines) to provide the initial compression
needed for ignition. This power source is
used to rotate the large mass of the
compressor, and bring it up to the speed
necessary to supply combustion air to the
combustor. The starting system engages
the compressor shaft at the beginning of
the start-up. Once ignition has been
obtained, this system is disengaged from
the compressor shaft. A variety of devices
can be used to provide the initial rotation
of the main compressor, including:
Fig 3-31 Pneumatic Starter System Diagram
Pneumatic starters using compressed

air or gas
Electric motors
Small diesel engines
Steam turbine expanders
Of these, pneumatic starters and electric

motors are the most common.
3.10.4.1 Pneumatic Starters
High-pressure air or gas can be expanded
in a small turbine (called a starter motor)
to drive the main gas turbine compressor
up to ignition conditions. This is a
particularly convenient system in remote
gas pipeline applications (where the gas
turbine itself is used to drive large
transmission compressors). The highpressure pipeline gas is used as the starting
gas supply.
Figure 3-32 Pneumatic Starter Installation
3.10.4.2 Electric Starters

Electric starters consist of an ac motor and
a variable frequency drive as shown in
Fig. 3-33 and 3-34, but otherwise operate
in the same fashion as a pneumatic starter.
The variable frequency drive enables the
167
motor speed to be controlled while still

maintaining motor efficiency.
rotor becomes bowed, the rotor may lock

and prevent startup or it may result in
high vibration until the bow slowly
disappears.
To prevent this, a special turning motor is
provided to slowly turn the shaft for a
number of hours after shutdown. This is
sometimes a hydraulic ratchet or a slow
turning electric motor. The motor and the
turbine shaft have corresponding gears.
The turning motor gear can be engaged or
disengaged (normal turbine operation)
with the shaft. A similar type of device is
used on large steam turbines for their
warm-up and cool-down periods.
Figure 3-33 Electric Starter Diagram
3.10.5 Cooling systems

Cooling system will keep the hot sections
and lube oil in safe operating range. This
system can be air cooled or liquid cooled,
the internal blading and discs will be
cooled by the secondary air.
Figure 3-34 Electric Starter Installation
On larger gas turbines, especially of the

heavy-duty type, the rotors are quite
heavy and they develop a sag or bow when
they cool down after shutdown. If a rotor
becomes bowed, the rotor may lock and
prevent startup or it may result in high
vibration until the bow slowly disappears.
3.10.6 Instrumentation and control

Protective devices, such as pressure &
temperature switches, pressure &
temperature sensors, vibration detectors,
and speed protection devices is often used
to protect the gas turbine from any
abnormal conditions.
To prevent this, a special turning motor is

provided to slowly turn the shaft for a
number of hours after shutdown. This is
sometimes a hydraulic ratchet or a slow
turning electric motor. The motor and the
turbine shaft have corresponding gears.
The turning motor gear can be engaged or
disengaged (normal turbine operation)
with the shaft. A similar type of device is
used on large steam turbines for their
warm-up and cool-down periods.
Other instrumentation is used for display

and indications of some operating
parameters such as:
1.
2.
3.
4.
3.10.4.3 Turning Gear
5.
6.
7.
8.
9.
On larger gas turbines, especially of the

heavy-duty type, the rotors are quite
heavy and they develop a sag or bow
when they cool down after shutdown. If a
168
RPM of the turbine

RPM of the GG (Gas generator)
Inlet air temperature
Gas temperature before the
turbine
Gas temperature after the turbine
Exhaust temperature
Lube oil pressure
Lube oil temperature
Vibration levels
3.11 Protection
3.13 Gas Turbine

Preventive Maintenance
A number of protective devices will first

alarm and then shutdown the engine if
certain levels are exceeded. They normally
protect against:
1.
The Most Important Item in The

Periodical Maintenance is The Air
Compressor Wash
Power turbine over speed
The major cause of deterioration in gas

turbine performance is fouling of the
compressor blading. Fouling results in a
decrease in compressor efficiency which
reduces overall thermal efficiency and
maximum power. It will also result in
compressor surging and acceleration
problems.
2. Exhaust
gas temperatures higher

than the maximum limit
3. Excessive vibration
4. Excessive
oil
temperatures
and
bearing
5. Fuel gas pressure too low or too high
The source of contamination is usually

dust, salt and other airborne particles that
are not trapped by the intake filters.
Contamination can also come from other
machinery close to the gas turbine or even
the gas turbine exhaust being re-ingested
under certain wind conditions. Sometimes,
a compressor front bearing oil leak will
make the problem worse.
6. Excessive
air intake filter differential

pressure
3.12 Turbine Performance

Maximum Power
The maximum power that can be achieved
is also a function of the life expected of the
hot gas path components. Vendors usually
specify a maximum power that will permit
the engine to achieve a reasonable life.
They also specify a maximum peak power
limit that can be applied if a user is willing
to incur higher maintenance costs. This
tradeoff is sometimes used for peak power
generation where gas turbines operate for
only short periods of time to satisfy peak
load conditions. For base load operation,
this is not desirable because of the
increased maintenance costs and the risk
of failure.
Compressor cleaning can be accomplished

by using either a liquid or an abrasive
material. In the past, it was quite common
for walnut shells or even rice (or other
abrasive materials sometimes called
carbo-blast) to be injected into the intake
to abrasively clean the compressor
blading. This is done while the unit is
running and the materials are burnt up in
the combustion section and then pass
through the engine. Since it is not as
effective as the water wash method, it is
not utilized as often any more. It also has
the disadvantage of plugging up cooling
passages in the compressor and cooling
holes in the turbine blades.
The most effective method of compressor
cleaning is the off-line water wash. This
method consists of stopping the unit,
169
injecting water wash fluids into the intake

of the compressor while running on the
starter and then restarting. It is also
referred to as the crank-soak method.
held sprayer. The waterwash ring can be

installed on the intake bell mouth as
shown in Fig. 3-35 or on the intake volute.
On-line water washing is not as effective as

off-line although it is still a viable
alternative if downtime is not acceptable.
Water Wash Fluids
The water that is used must be very clean
and must conform to quality standards
specified by the gas turbine vendor. Using
hard water or water contaminated with
sodium, potassium, vanadium or other
chemicals can cause further fouling and
increased corrosion.
To remove oily substances, additional
cleaning agents and solvents need to be
used. These are mixed with water and gas
turbine vendors often specify acceptable
cleaners. However, the most effective
cleaning agents are also the most toxic and
require special handling.
Fig.3-35 Waterwash Connections
In some cases, operators will actually go

into the intake to manually scrub and wash
the first few stages with a brush if deposits
are especially stubborn.
The waterwash fluids are contained in a
special tank and the spray pressure is
provided by compressed air applied to the
tank. This tank may be permanently
installed or located on a special cart such
as the one shown in Fig. 3-36
If the temperature is less than 4C, a 1:1

mixture of water and ethylene glycol is
recommended to prevent icing. Again the
gas turbine vendor has to be consulted
since commercial and automotive antifreeze products are usually not acceptable.
Once all of the fluids have been sprayed

into the intake, all disconnected lines are
reconnected and the unit restarted.
Off-Line Water wash

To perform an off-line waterwash, the
engine is stopped and allowed to cool first.
Some vendors specify that various
instrumentation, bleed and drain lines be
disconnected prior to water washing. The
engine is then run at maximum crank
speed with the fuel valve and igniters
deactivated.
If the waterwash has been successful, the

operator should be able to see an increase
in compressor discharge pressure and a
decrease in turbine inlet temperature.
The waterwash fluid can be injected either

through a pre-installed waterwash ring or
provided manually by means of a hand-
170
6. Bore
scope for the internal parts of

the engine
Gas Turbine Overhauls

The manufacturer of the gas turbine
identify the intervals between overhauling
of the engine according to various
parameters such as the engine running
hours, working load (full load or partial
load) and the No. of starts of the engines.
Fig. 3-36 Waterwash Cart
On-Line Waterwash
In general the overhaul of the engine

includes the following:
To perform an on-line wash, the gas

turbine is brought to idle and allowed to
cool. The waterwash fluid or abrasive
material is then injected and burnt in the
combustion chamber, after which the
engine is returned to the required
operating condition. This method is
sometimes used to reduce the time
between off-line washes, /however this
method is not recommended with some
gas turbine manufacture
1.
2. Repair /change the combustors

3. Repair
/change the nozzles and

rotary blades of the turbines
4. Perform
non destructive test for the

shaft, rotor disks and the casings to
discover the thin cracks which can
not be seen by visual check ,repair
if required
Other periodical maintenance is six

monthly and annual Maintenance
The manufacturer of the gas turbine
identify the check list of those types of
maintenance, in general it contains the
following as example:
1.
Repair the air

compressor
stationary and rotary blades
5. Cleaning all parts and cooling paths

6. Check
the seals and bearings,

repair/change if required
Instrument checks like pressure

and temp switches, transmitters
and indicators
7. Check
shaft unbalance and correct if

required
2. Cleaning the fuel injectors
TROUBLE SHOOTING
3. Inspection of the thermocouples
When you complete this objective you will

be able to
4. Calibrating pressure safety valves
Study some of trouble shooting during

startup or normal operation of the gas
turbine
5. Check
alignment between gas

turbine shaft and the driven
equipment shaft
1.
171
Low oil pressure alarm
Check
the
pressure
instrumentation (gauge or
transmitter)
o The
pressure regulator
need adjusting
o The lube oil filter need
replacement due to hi
deferential pressure
o There is blockage in main
lube oil header
o There is major leak in the
lube oil line
o The lube oil pump is not
pumping
due
to
(cavitations- failure- pump
gear box failure)
2. High oil temp. alarm
o Check
the
temp.
Instrumentation
(gaugeRTD)
o Oil cooler is not operating
o Oil cooler is not clean
o The temp. Control valve is
internal leaking (by-passing
the cooler)
o Problem in one or more
lube oil bearing
o The lube oil needs to be
replaced due to aging
3. Ignition failure (fail to ignite - flame
failure)
o The fuel pressure is not
enough
o The igniter is bad or not
cleaned
o The flame detector is not
working
o Bad fuel metering valve
4. The engine fails to reach the self
sustain speed
o The fuel pressure is not
enough
o The fuel metering valve is
not working good
The speed pick-up is not

working
o The starter motor is not
engaged or is weak
o The clutch of the starter is
not working good (slipping
disengaged)
o The gas turbine itself is not
good
o The air bleed valve or the
IGV in not-correct position
5. Fail to load the engine
o Not enough fuel pressure
o Fuel metering valve is not
good
o Air bleed valve or IGV is not
working good
o Air intake filter is not clean
o Bad engine performance
6. Hi engine temp. (Exhaust temp.)
o Sudden increase in fuel
pressure
o One tip or more of
thermocouples are bad
o Bad fuel metering valve
o Bad fuel contents (liquids in
the fuel)
o Bad air intake filters
o Bad engine combustors or
injectors
3.14 Startup and Shutdown

Gas Turbine Start-Up Procedure
The basic steps in starting a gas turbine
are:
172
1.
Preparation for startup
2.
Start initiation
3.
Crank and light off
4.
Warm-up
5.
Loading
These steps must occur in a specific

sequence and at certain time intervals.
They are usually managed by the control
system and the operator often has no role
except to watch the process. If certain
conditions occur or specific requirements
are not met at some point in the startup
sequence, the startup will be aborted and
the unit stopped. The progress of the
startup is displayed on a control panel such
as the one shown in Fig. 3-37
activated and energized. These systems

will usually be active and will only need to
be turned on if the equipment has been
shut down for maintenance or an extended
period of no demand. In remote
applications, startup normally occurs
automatically without human participation
and intervention unless an abnormal
situation requires response.
The operator may need to reset the system
if a previous malfunction or abnormal
condition has occurred. Pressing a reset
switches on the panel, or on a computer
screen does this.
There are also a number of per missives
that need to be satisfied before the start
sequence will be allowed by the control
system. Some of these will pertain to the
gas turbine, such as a minimum oil
reservoir temperature, and others are
related to the requirements of the
particular load being driven by the turbine
(typically an electrical generator or
process/transmission gas compressor). Fig
3-38 shows an example of a startup screen
for a gas turbine driving a gas compressor.
Fig. 3-37 Typical Control Panel Used to Monitor Startup,

Shutdown and Operation
3.14.1 Preparation for Startup

When a start is initiated, it is assumed that
all
electrical,
pneumatic,
air,
instrumentation and control systems are
173
Fig. 3-38 Startup Screen with Permissives
3.14.2 Start Initiation

During the startup sequence, a number of
operating conditions must be met as
determined
by
various
pressure,
temperature and status switches. Timers
are used to ensure these conditions occur
in the expected time period, or else the
startup will be aborted. An example of a
startup sequence screen is given in Fig. 339 Notice the sequence of steps and the
descriptions on the left of the screen.
A gas turbine is normally in one of two

modes of operation: remote or local. The
mode of operation is set either by a switch
on the control panel or a selection box on
a computer screen. When in remote, a
higher level process control system has the
ability to initiate a start. When in local, the
start can only be initiated from the control
panel.
174
Fig. 3-39 Startup Screen with Sequencing Status
When the start button is pressed (locally

or remotely), the following occurs:
1.
The ventilation fans starts which

vents the building or enclosure.
2.
Pre-lubrication occurs. Depending

on the design of the lube oil
system, the backup pump will start,
and if adequate pressure is
achieved within a certain time
period, the pre-lube pump will start
and the pre-lube timer resets to
ensure adequate pressure.
3.
There may be a check of the fuel

gas system to ensure that the fuel
valves are operating properly and
adequate fuel pressure is available.
3.14.3 Crank and Light off

Once these steps have been completed,
the starter begins to rotate the gas turbine
rotor (see Fig. 3-40). The first portion of
this process purges the gas turbine for
several minutes in case explosive vapors
are still present. The rotor then coasts
down to a speed appropriate for light off.
175
Fig. 3-40 Startup of a Typical Gas Turbine
Fuel is then admitted to the combustion

chambers, the igniters are energized and
light off occurs. This results in a rapid
increase in speed. The overrunning clutch
disengages the starter, which then shuts
down, and the igniters are de-energized.
3.14.5 Loading
3.14.4 Warm-up
If the gas turbine drives a compressor, the

compressor will have been pressurized
prior to the purge crank, and the suction
and discharge valves will have been
opened. The compressor is started in the
unloaded position, with the recycle valve
open. Loading is accomplished by slowly
closing in on the recycle valve. The actual
operating point will be determined by the
control system.
After the warm-up is finished, the fuel flow

is increased and the load is applied. For a
generator, this will require synchronizing
the speed, phase and voltage, and then
closing the breaker.
Once idle speed is reached, the engine is

allowed to warm up. For backup power
generation, especially for small gas
turbines, this may be very short. When
time is not as critical, it is best to permit the
engine to warm up slowly. Heavy-duty gas
turbines take longer to start and warm up
than aero-derivatives.
On start initiation, the bleed valve(s) will
be open and the inlet and variable guide
vanes will be in their closed position. The
bleed valves will close either at a certain
speed or over a specified range of speeds.
The guide vanes will open to their
optimum position over a range of speeds
as designated by a specified schedule
(based on a control program that relates
guide vane position to turbine speed).
The acceleration and deceleration of the

gas turbine is limited to a certain rate.
Sudden increases in speed will cause rapid
increases in turbine temperature that
could easily be above the limit. A rapid
decrease in speed could cause combustion
to be interrupted and any re-lighting,
without going through the required startup procedure, would be catastrophic.
176
of the shutdown sequence is the same as

for a normal shutdown.
3.14.6 Normal Shutdown

An operator most often initiates shutdown
of a gas turbine although some systems do
have an automatic shutdown when the gas
turbine is no longer required. To a large
extent, a shutdown is the reverse
procedure of a startup.
This type of shutdown increases the wear

on the gas turbine because of the rapid
cooldown it entails and is reserved for
emergency conditions only.
List of most important malfunction
shutdown during normal operation
The first step in a normal shutdown is to

reduce speed to idle so the gas turbine can
cool down. In this step opening the breaker
that connects the generator to the power
system or opening the recycle valve of the
compressor unloads the power turbine, for
example. The gas turbine cools down as
much as possible to minimize the negative
thermal effects. During this time, the
engine can easily be restarted, according
to the specified procedure.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Low lube oil pressure S/D

High lube oil temp. S/D
GG over speed
P.T over Speed
Backup P.T over speed
High vibration for any bearing
High exhaust engine temp.
High deferential exhaust temp.
High fuel gas pressure
Low lube oil tank level
3.15 Interface between gas

network and power
stations
When the cooldown timer times out, the

fuel valve is closed which extinguishes
combustion. The rotor speed decreases
until the rotor stops.
Natural Gas reaches the power plant with

pressure range 35:37 (bar) then Reduction
Station (3 regulators) with pressure 26
(bar).
As the speed drops, the main lube oil pump

(if driven off the rotor) will lose pressure.
At a specified point, usually based on oil
pressure, the post lube pump starts up and
continues to lubricate and cool the
bearings for a specified time period.
Regulator is used only when gas reaches

the power plant with high pressure (37
bar) and low required load and high
pressure is useless
The enclosure or building fans shut off.

On some gas turbines, the turning gear
activates once the rotor stops.
By-pass valve used to completely remove

the regulating effect when required high
load or when the N.G pressure is too low.
3.14.7 Fast Shutdown

In certain situations, a fast shutdown will
be initiated. This occurs when a protective
device detects an abnormal condition such
as high vibration, or when an operator
initiates an emergency stop. In this case,
the cooldown period is eliminated and the
fuel valve is closed immediately. The rest
Natural gas is then compressed in gas

compressor (3-compressors) into 37 (bar)
After compressing and cooling to reach the
required gas pressure (37bar)
The combustion chamber pressure is 19
bar (air and fuel mixture)
177
Hours
Gas Flowrate (Nm3/hr)
Load (MW)
Gas Pressure (bar)
44133
149
27
43479
150
27
43966
150
28
43868
149
28
43786
150
28
43431
149
27
43255
150
27
44219
149
27
44127
150
27
10
44380
149
26
11
44105
149
26
12
43629
149
26
13
43960
149
27
14
44283
148
27
15
44101
149
27
16
43951
149
27
17
44463
149
27
18
44103
149
27
19
44255
149
26
20
59435
219
25
21
59690
218
25
22
59417
219
25
23
59129
218
26
24
59248
219
26
Table 3-1 Unit 1 Operation data
178
3.15.1 Gas Pressure Trend

Gas Pressure (bar)
30
Gas Pressure
25
20
15
10
5
0
0
10
15
20
25
Hour
Fig. 3-41 gas pressure (bar) with time (hours)
3.15.2 Gas flowrate Trend

Gas Flowrate (Nm3/hr)
70000
60000
50000
40000
30000
20000
10000
0
0
10
15
Fig. 3-42 gas flowrate (Nm3/hr) with time (hours)
179
20
25
3.15.3 Load Trend

Load (MW)
250
200
150
100
50
0
0
10
15
Fig. 3-43 Load (MW) with time (hours)
180
20
25
Chapter 4 | Compressors
Chapter 4
Compressor Stations
181
pipelines inner coating and increase its

transmission efficiency.
4) Piping &Valves: It moves natural gas
between the pipeline and compressor
station.
5) Gas turbine as a prime mover to the
station.
6) Flare system as safe emergency release
system in a refinery or chemical plant
7) Scrubber: It removes any solids or
liquids from the natural gas that enters the
compressor station.
8) Mechanical seal used to prevent gas
leakage from compressor.
4.1 Introduction
Compressor station is one of the main
components of the natural gas national
grid, as metering stations, pipes, valves,
and reduction station. Compressor
stations have two main purposes, they can
act as a lifting station or booster station.
The lifting station is located near the fields
e.g. Rashpetco compressor station, or it is
located at the processing plants e.g. WDGC
(western desert gas complex), the lifting
stations act as a source of gas to the
national grid.
The boosting compressor station is located
through the network e.g. Dahshour, or it is
located at the consumers e.g. Al-Akaba
compressor station to rise the gas pressure
in order to provide the gas to the
consumers with the required pressure.
4.2 Compressor
4.2.1 Introduction
A compressor is a device used to increase
the pressure of a compressible fluid. The
inlet pressure level can be any value from
a deep vacuum to a high positive pressure.
The discharge pressure can range from sub
atmospheric levels to high values in the
tens of thousands of pounds per square
inch. The inlet and outlet pressures are
related, corresponding with the type of
compressor and its configuration. The fluid
can be any compressible fluid, either gas or
vapor, and can have a wide molecular
weight range. Recorded molecular weights
of compressed gases range from 2 for
hydrogen to 352 for uranium hexafluoride.
Applications of compressed gas vary from
consumer products, such as home
refrigerators,
to
large
complex
petrochemical plant installations.
4.1.1 Definition
A compressor station is a facility which
provides the motive power for the
transportation of natural gas from one
location to another. Natural gas, while
being transported through a gas pipeline,
needs to be constantly pressurized at
certain distance intervals (usually
compressor stations are located 40 to 100
miles apart along the pipeline, but are
sometimes even farther apart). The gas
enters the compressor station with low
pressure and velocity, looking for an
energy source to move it further down the
pipeline.
4.1.2 Components
1) Compressor: Re-pressurize the volume
of gas flowing through the pipeline.
2) Lubrication system: Lubricate, cool and
protect the moving parts.
3) Cooling system: When natural gas is
compressed, its pressure and temperature
increase. The gas may be cooled before its
return to the pipeline to protect the
182
Compression Methods
Figure 4-2 compares the characteristic

curves of the dynamic compressors (axial
and
centrifugal)
with
positive
displacement compressors.
Compressors have numerous forms, their

exact configurations being based on the
application. For comparison, the different
types of compressors can be subdivided
into two broad groups based on the
compression mode, either intermittent or
continuous. The intermittent
mode of compression is cyclic in nature: A
specific quantity of gas is ingested by the
compressor, acted upon, and discharged
before the cycle is repeated. In the
continuous compression mode the gas is
moved into the compressor, acted upon,
moved through the compressor, and
discharged without interruption of the
flow at any point in the process.
Figure 4.2 General performance curve for axial- flow,

centrifugal, and positive displacement compressors.
Compressors using the intermittent

compression mode are referred to as
positive displacement compressors, of
which there are two distinct types:
reciprocating and rotary. Continuousmode compressors are also characterized
by two fundamental types: dynamic and
ejector.
Figure 4-1 shows
compressors.
the
family
Figure 4-3 shows the typical application

range of each compressor.
of
Figure 4.3 Typical application ranges of compressor types.
In our project we will focus on centrifugal

compressor that achieve the proper
pressure and flow rate .
Figure 4.1 Family of compressors
183
4.2.2 General Description of a CENTRIFUGAL Compressor
Figure 4.4 Sectional view of centrifugal compressor schematic
A centrifugal compressor is a dynamic

machine. It has a continuous flow of fluid
which receives energy from integral shaft
impellers. This energy is transformed into
pressurepartly across the impellers and
partly in the stator section, i.e., in the
diffusers. This type of machine is
composed (see Fig. 4.4) of an outer casing
(A) which contains a stator part, called a
diaphragm bundle (B), and of a rotor
formed by a shaft (C), one or more
impellers (D), a balance drum (E), and
thrust collar (F).
Gas is drawn into the compressor through

a suction nozzle and enters an annular
chamber (inlet volute), flowing from it
towards the center from all directions in a
uniform radial pattern (see Fig. 4.5). At the
opposite side of the chamber from the
suction nozzle is a fin to prevent gas
vortices.
The gas flows into the suction diaphragm
and is then picked up by the first impeller
(see Fig. 4.6). The impellers consist of two
discs, referred to as the disc and shroud,
connected by blades which are shrunk
onto the shaft and held by either one or
two keys. The impeller pushes the gas
outwards raising its velocity and pressure;
the outlet velocity will have a radial and a
tangential component (see section 4.10 for
further details). On the disc side, the
impeller is exposed to discharge pressure
The rotor is driven by means of a hub (G)

and is held in position axially by a thrust
bearing (I), while rotating on journal
bearings (H). The rotor is fitted with
labyrinth seals (L) and, if necessary, oil film
end seals (M).
184
(see Fig. 4.7) and on the other side partly

to the same pressure and partly to suction
pressure. Thus a thrust force is created
towards suction. The gas next flows
through a circular chamber (diffuser),
following a spiral path where it loses
velocity and increases pressure (similar to
fluid flow through conduits). The gas then
flows along the return channel; this is a
circular chamber
rings made in two or more parts. The last

impeller of a stage (the term stage refers
to the area of compression between two
consecutive nozzles) sends the gas into a
diffuser which leads to an annular chamber
called a discharge volute (see Fig. 3.6).
Figure 4.5 Qualitative view of the flow in the volute.
bounded by two rings that form the

intermediate diaphragm, which is fitted
with blades (see Fig. 4.8) to direct the gas
toward the inlet of the next impeller. The
blades are arranged to straighten the spiral
gas flow in order to obtain a radial outlet
and axial inlet to the following impeller.
The gas path is the same for each impeller.
The discharge volute is a circular chamber

which collects the gas from the external
boundary of the diffuser and conveys it to
the discharge nozzle. Near the discharge
nozzle there is another fin which prevents
the gas from continuing to flow around the
volute and directs it to the discharge nozzle
(see Fig. 4.10).
Labyrinth seals are installed on the

diaphragms to minimize internal gas leaks
(see Fig. 4.8). These seals are formed by
The balance drum (E) is mounted on the

shaft after the end impeller (see Fig. 4.4). It
serves to balance the total thrust produced
by the impellers. Having end impeller
185
delivery pressure on one side of the drum,

compressor inlet pressure is applied to the
other by an external connection (balancing
line, see Fig. 4.11). In this way, gas
pressures at both ends of the rotor are
roughly balanced. To get even closer
pressure levels and, therefore, the same
operating conditions for the shaft
Figure 4.8 Labyrinth seals and diaphragms.
Figure 4.6 First stage sectional view.
Figure 4.9 Last impeller of a stage.
Figure 4.7 Pressure distribution on the impeller.
186
Figure 4.10 Discharge volute: qualitative view of the flow
Figure 4.11 External connection of the oil system
187
Chapter 4
compressors
end oil seals, another external connection

is made between the balancing chambers
(balancing line, see Fig. 4.11).
Multistage compressors with one

compression stage only (Fig 4.12).
Multistage compressors with two
compression stages. The two compression
stages are set in series in the same
machine. Between the two stages, cooling
of the fluid is performed in order to
increase the efficiency of compression.
The gas chambers are positioned outside

the shaft-end labyrinths. They are
connected to achieve the same pressure as
that used as reference for the oil seal
system (see Fig. 4.11 for a block diagram).
In special cases, when the seal oil and
process gas have to be kept separate, inert
gas is injected into the balancing chamber
(buffer gas system) at a pressure that
allows it to leak both inwards and
outwards forming a seal.
Multistage compressors with more than

two compression stages in a single casing.
As a rule they are used in services where
different gas flows have to be compressed
to various pressure levels, i.e., by injecting
and/or extracting gas during compression.
4.2.3 Centrifugal Compressors Types
Sometimes compression stages are

arranged in parallel in a single casing. The
fact that both stages are identical and the
delivery nozzle is positioned in the center
of the casing makes this solution the most
balanced possible. Moreover, a double
flow is created by a common central
impeller (see Fig. 4.15).
Centrifugal compressors may have

different configurations to suit specific
services and pressure ratings. They may be
classified as follows:
4.2.3.1 Compressors with

Horizontally-split Casings
Horizontally-split casings consisting of half
casings joined along the horizontal centerline are employed for operating pressures
below 60 bars.
The suction and delivery nozzles as well as
any side stream nozzles, lube oil pipes and
all other compressor-plant connections are
located in the lower casing.
With this arrangement all that is necessary
to raise the upper casing and gain access to
all internal components, such as the rotor,
diaphragms and labyrinth seals is to
remove the cover bolts along the
horizontal center-line.
Horizontally-split casing compressors may
be further identified according to the
Figure 4.12 Horizontally-split casing.
number of stages.
188
Chapter 4
compressors
4.2.3.2 Compressors with Verticallysplit Casings

Vertically-split casings are formed by a
cylinder closed by two end covers: hence
the denotation barrel, used to refer to
compressors with these casings. These
machines, which are generally multistage,
are used for high pressure services (up to
700 kg/cm2). Inside the casing, the rotor
and diaphragms are essentially the same as
those for compressors with horizontallysplit casings.
Barrel type compressors which have a
single compression stage
Barrel type compressors with two
compression stages in series in a single
casing
Figure 4.13 Multistage two phase compressor.
Figure 4.14 Multistage three phase compressor.
Figure 4.16 Barrel type compressor with one compression

phase.
Figure 4.15 Two phase compressor with a central double

flow impeller.
Figure 4.17 Barrel type compressor with two compression
phases.
189
Chapter 4
compressors
Compressors which incorporate two

compression stages in parallel in a single
casing.
4.2.3.3 Compressors with Bell

Casings
Barrel compressors for high pressures have
bell-shaped casings and are closed with
shear rings instead of bolts (see Fig. 4.18).
Figure 4.19 Pipeline compressor.
4.2.3.5 SR Compressors
These compressors are suitable for
relatively low pressure services. They have
the feature of having several shafts with
overhung impellers. The impellers are
normally open type, i.e., shroudless, to
achieve high tip speeds with low stress
levels and high pressure ratios per stage.
Each impeller inlet is coaxial whereas the
outlet is tangential. These compressors are
generally employed for air or steam
compression, geothermal applications etc.
(see Fig. 4.20).
Figure 4.18 High pressure barrel type compressor.
4.2.3.4 Pipeline Compressors

These have bell-shaped casings with a
single vertical end cover. They are
generally
used
for
natural
gas
transportation (see Fig. 4.19). They
normally have side suction and delivery
nozzles positioned opposite each other to
facilitate installation on gas pipelines.
Figure 4.20 SR type compressor.
190
Chapter 4
compressors
and fixed blading, in which the energy is

converted from one form to another.
4.2.4 Basic Theoretical Aspects

4.2.4.1 Preliminary Definitions
The term turbomachinery is used to
indicate systems in which energy is
exchanged between a fluid, evolving
continuously and in a clearly determined
quantity, and a machine equipped with
rotary blading.
4.2.4.2 The Compression Process

Consider Fig. 4.21, which represents a
generic compression process in the Mollier
plane (enthalpy-entropy) taking place in a
single compressor stage. The fluid, taken in
determined conditions p00 and T00, is
subsequently accelerated up to the inlet to
the stage where it reaches the conditions
defined by thermodynamic state 1. The
acceleration process is accompanied by
dissipation phenomena linked to the
increase in speed of the fluid. In flowing
along the rotor the fluid undergoes a
transformation that brings it to the
conditions p2 and T2. During this phase
there is an increment in potential energy
per mass unit of fluid given by:
Turbomachines can be classified as:

Process machines, in which the machine
transfers energy to the fluid.
Drivers, in which the machine receives
energy from the fluid.
An initial classification of turbomachines
may be made on the basis of the
predominant direction of the flow within
the machine:
Axial machines, in which the
predominant direction is parallel to the
axis of rotation.
Ep,1-2=h2-h1
(4.1)
and an increment in kinetic energy per

mass unit of fluid given by:
Radial machines, in which the

predominant direction is orthogonal to the
axis, although portions of the flow may
have an axial direction.
Ek,1-2 =
Mixed machines, where the situation is

intermediate between the described
above.
(4.2)
The entropy of the fluid, as it flows through

the stage, increases as a consequence of
the dissipation processes involved in
compression. In the stator part the kinetic
energy of the fluid is converted into
potential energy. The total enthalpy for
state 4 can thus be evaluated as:
Turbocompressors
(more
briefly,
compressors) constitute a special category
of process machines. They operate with
compressible fluids and are characterized
by an appreciable increase in the density of
the fluid between the first and the last
compression stages. The compression
process is frequently distributed among
several stages, a term used to indicate an
elementary system composed of mobile
blading, in which the fluid acquires energy,
h0,4 = h4 +
(4.3)
The fluid then leaves the stage in the

conditions defined by state 4, with residual
velocity C4.
191
Chapter 4
compressors
accumulated by the fluid under the form of

increment in potential energy; it is
expressed by:
HP =
(4.7)
in which the relationship between

pressure and density is expressed in the
form
p = Cos tan te
where n represents the mean exponent of

polytropic transformation between the
two states 1 and 2. Polytropic head can
thus be expressed by the following
formula:
Figure 4.21 Entropy-enthalpy diagram of a compression

process.
Hp=
Zo R Too [( )
4.2.4.3 Basic Quantities of

Compression Process
The basic quantities utilized to quantify the
exchanges of energy in compressors are
given below. Note that the quantities
indicated apply both to complete
compressors and to individual stages. It is
also assumed that the thermodynamic
characteristics of the fluid are represented
by the perfect gas model.
Isentropic Head. Isentropic head is defined

as the energy per mass unit accumulated
by the fluid subsequent to a reversible (and
thus isentropic) adiabatic transformation
between states 1 and 2. This gives the
following equation:
With
(4.10)
p = Cos tan te
(4.11)
in which constitutes the ratio between

the specific heat values of the gas.
(4.4)
H s=
Zo R Too [ ( )
We also have
HR = ( h04 h01) + QExt
Hs =
between blading and fluid per mass unit of

fluid processed:
]
(4.9)
Effective Head. The effective head HR is

defined as the effective work exchanged
HR =
(4.8)
(4.5)
(4.12)
In the hypothesis of adiabatic conditions

QEXT = 0 and we further obtain:
Polytropic Efficiency. The polytropic

efficiency is defined as the ratio between
HR = ( h04 h01)
(4.6)
polytropic head HP and effective head HR

necessary to effect compression between
Polytropic Head. The polytropic head HP is

defined as the energy per mass unit
192
Chapter 4
compressors
states 0 and 4. Applying the preceding

definition we obtain:
p =
The isentropic efficiency can be rewritten

as:
s =
[ ( )
by developing the above equation, we

obtain:
With reference to Fig. 4.22 , we may

consider a rotor belonging to a generic
turbomachine, taking into examination the
conditions existing in section 1 (inlet) and
section 2 (discharge). Utilizing the
equation of balance of momentum for the
stationary flow between two sections, it is
possible to obtain:
The polytropic head can be further

rewritten in the form
p =
() ( )
( )
(4.15)
Polytropic efficiency possesses the

important property of being dependent
only on the properties of the gas, the
pressure, and temperature ratios. It is
independent of the absolute pressure level
from which the compression process
starts.
= ( ) (4.18)
The work transferred through the blading
per mass unit of fluid processed is thus
given by:
Wx = /m = ( )
Isentropic Efficiency. The isentropic

efficiency is defined as the ratio between
isentropic head HS and effective head HR
associated with compression between
states 0 and 4. From this definition we
obtain:
s =
(4.19)
The first principle of thermodynamics
establishes that the work per mass unit is
equal to, for an adiabatic flow, the
variation in total enthalpy. We thus obtain:
, = h2 h1
(4.17)
4.2.4.4 Euler Equation for

turbomachines
(4.14)
()
( )
It may be stated that, for a compressor, the

polytropic efficiency is always greater than
the isentropic efficiency relevant to the
same transformation.
(4.13)
( )
p =
( )(/ )
[ ( )
= ( )
=
( )
(4.20)
The above equation, known as the Euler

equation, is one of the fundamentally
important equations for the study of
turbomachines. Application of the Eq. 4.20
(4.16)
193
Chapter 4
compressors
shows that in a generic stage composed of

a rotor and a stator, there is no transfer of
mechanical work outside of the rotary
parts; in particular, then, the enthalpy of
The number of dimensionless parameters

necessary and sufficient to describe the
characteristics of a stage is specified by
Buckinghams theorem, which is also
used to determine their general form. The
dimensionless parameters used to
describe the performance of axial and
centrifugal compressors are given below.
Flow Coefficient. The flow coefficient for
an axial machine is defined as the ratio
between the axial velocity at the rotor inlet
section and the tip speed of the blade
=
FIGURE 4.22 The Euler turbomachinery equation.
the fluid does not change in traversing the

stationary components, but only in
traversing the rotary ones. Consequently it
may be stated:
(4.21)
flow
(4.24)
Both of these definitions can be

interpreted as dimensionless volume flow
rate of the fluid processed by the machine.
and in the case of perfect gas:
, = Cp ( T04 T02 )
= , = Cp ( T02 T01 )
(4.23)
For centrifugal machines the

coefficient is defined as follows:
, = h04 h00 = h02 h00

=
Machine Mach Number. The Mach

number, MU, is defined as the ratio
between the machine tip speed and the
velocity of sound in the reference
conditions:
(4.22)
The total temperature is thus constant

throughout the stationary components.
4.2.4.5 Dimensionless Parameters

The behavior of a generic stage can be
characterized in terms of dimensionless
quantities which specify its operating
conditions as well as its performance. The
dimensionless representation makes it
possible to disregard the actual
dimensions of the machine and its real
operating conditions (flow rate and speed
of rotation) and is thus more general as
compared to the use of dimensional
quantities.
Mu =
(4.25)
The Mach number can be interpreted as a

dimensionless speed of rotation of the
machine.
Reynolds Number. The Reynolds number,
Re, is generally defined as the ratio
between inertial forces and viscous forces,
evaluated in relation to assigned reference
conditions, acting on a fluid particle for the
194
Chapter 4
particular
question.
compressors
fluid-dynamic
problem
in
(4.30)
The two quantities defined above can be
interpreted as dimensionless work per
In the axial machine field, a frequently

used formulation for the Reynolds number
is the following:
Re =
mass unit transferred by the blading to the

fluid.
(4.26)
Polytropic Efficiency. The same definition

given in 4.2.4.3 is applied.
For centrifugal machines the following

formulation is frequently used:
Re =
p =
(4.27)
Specific Heat Ratio. The specific heat ratio

is simply defined as the ratio between
specific heat at constant pressure and at
constant volume for the gas in question:
( )
In any one of the compressor stages, work

is transferred by the rotary blading to the
fluid in modes depending on the geometry,
the fluid-dynamic conditions and the
properties of the gas processed. The study
of these energy interactions, governed by
the Euler Eq. (4.20), calls for analysis of the
speed of the fluid in suitable sections of the
stage. This analysis is usually carried out
utilizing speed triangles determined in
suitable sections of the stage.
(4.29)
The quantity of energy absorbed by the

compressor cannot be entirely converted
into a pressure increment in the fluid due
to dissipation phenomena of various kinds
involving the machine as a whole. Among
these, the pressure drops directly
For a centrifugal machine an identical

parameter is defined, which is however
called head coefficient. It is expressed by
the following equation:

4.2.5.1 General Information
In the above formula, the Euler equation

and the kinematic equality U1 = U2, valid in
first approximation for an axial machine,
have been introduced.
[ ( )
4.2.5 Performance of compressor

stages
Coefficients of Work and of Head. The

coefficient of work for an axial machine is
defined as the ratio between the work per
mass unit transferred by the blading to the
fluid and the square of the tip speed.

This formula applies to both centrifugal

and axial machines.
The specific heat ratio is used to take

account of the thermodynamic properties
of the fluid.
(4.31)
(4.28)
195
Chapter 4
compressors
attributable to effects of aerodynamic type

will be examined here.
the blade. This phenomenon, termed

deviation, determines a reduction in the
value of Cq in respect to the value that
could be theoretically obtained in the case
of null deviation.
Knowledge of the energy transfer and

dissipation mechanisms in a stage provides
the necessary tools for understanding the
factors that determine and influence its
performance. These aspects are examined
in the following paragraphs, along with the
representations normally utilized to
describe performance.
Recalling the Euler equation and the

definition of work coefficient, we may
write:
4.2.5.2 Speed Triangles
(4.32)
In studying turbomachines the concept of

speed triangles is frequently used to
represent the kinematic conditions, for
both fluid and blade, existing at the inlet
and discharge sections of a generic fixed or
rotary blading.
With reference to the speed triangle in Fig.

4.23 and also considering the simplifying
assumption that the flow can be
considered
incompressible
between
sections 1 and 2, the formula for the work
coefficient can be rewritten as follows:
The speed triangles for an axial compressor

stage are shown in Fig. 4.23. Note that the
absolute velocity C of the fluid in a given
section of the stage is obtained
= [
( + ) ]
= [ ( + ) ]
(4.33)
The above equation shows that, in the

further hypothesis that the direction of
flow does not change from blade inlet to
outlet, the relation between flow
coefficient and work coefficient depends in
linear manner on (tg1 1 + tg2 2 ) in the
mode shown in Fig. 4.24. In the hypothesis
of compressible flow and non-constant
angles the relation is no longer linear but
the qualitative description is still valid.
FIGURE 4.23 Velocity (speed) triangles for an axial

compressor stage.
by combining a relative velocity W with a

velocity U determined by the rotation of
the blade. The absolute velocity C can be
further broken down into an axial velocity
CA and a tangential velocity Cq.
The speed triangles for a centrifugal stage

are shown in Fig. 4.25 The physical
The speed triangle on discharge from the

rotor is characterized by the fact that the
direction of the absolute velocity vector
does not exactly coincide with the
direction indicated by the trailing edge of
interpretation of the quantities is the same

as that of the axial machine, although the
meridian rather than the axial components
of the quantities represented should be
taken into consideration. In centrifugal
196
Chapter 4
machines
deviation
too
the
compressors
phenomenon
of
The dependency between the structural

angle b2 and t can be expressed in explicit
form by introducing the quantity:
(4.35)
called flow coefficient at the impeller

discharge section. A quantity , termed
slip factor, which takes account of the
imperfect guiding action of the impeller, is
also introduced ; it may be defined as:
(4.36)
where the term VS represents the

tangential velocity defect associated with
the slip effect.
Utilizing
these
definitions
and
hypothesizing inlet guide vane conditions
null (Cq1 = 0), Eq. (4.32) is rewritten as:
FIGURE 4.25 Velocity triangles for a centrifugal stage.
(4.37)
blades or those turned in the direction

opposite that of rotation up to bb2 values
of about - 60 degrees are normally used in
common applications.
The head coefficient for a centrifugal

compressor can be expressed as follows:
structural angles bb2 greater than zero (i.e.,

blades turned in the same direction as that
of rotation) are not utilized insofar as they
generate high pressure drops. Radial
conventionally termed slip, can be

observed, so that the relative velocity on
discharge from the impeller is not aligned
with the direction of the blade.
The above equation is illustrated in Fig.

4.26, which is the equivalent of the one
already given for axial machines. For
centrifugal compressors, geometries with
FIGURE 4.24 Loading coefficient vs. flow coefficient for an

axial stage.
4.2.5.3 Conventional
Representation of Pressure Drop in
Compressors
Pressure drops in compressors are

conventionally divided into two main
categories.
(4.34)
197
Chapter 4
compressors
WA= KA
1. Pressure drop due to friction

2. Pressure drop due to incidence
Pressure Drop Due to Incidence. With

reference to the classic studies on bi
dimensional wing contours, it should be
recalled that the pressure drop of a generic
contour, stated in relation to the
incidence, follows a trend of the type
shown in Fig. 4.27. This distribution can be
approximated with a parabolic law that
presents a minimum point at a certain
incidence i*.
These two phenomena are discussed in the

following paragraphs.
Pressure Drop Due to Friction. These are
dissipation terms associated with friction
phenomena between the walls of the ports
of the machine (both rotor and stationary)
and fluid flowing through it. In general, the
flow in compressors is characterized by
turbulence, so it can be considered that
the energy dissipated is proportional, in
first approximation, to the square of the
fluid velocity and thus to the square of the
volume flow in inlet conditions. This
energy is not transferred to the fluid under
the form of potential energy, but only
under
the
form
of
heat.
Although the behavior described here

applies, strictly speaking, to wing contours
alone, it may be extended with reasonable
accuracy to the blading of centrifugal
machines as well. It is thus possible to
define, for a generic turbomachine
blading, whether stationary or rotary, an
optimum incidence condition, at which the
pressure loss phenomena deriving from
incidence are minimum.
This optimum incidence value depends on
the geometry of the blade and on the
speed triangle immediately upstream of
the blade leading edge. When the speed of
rotation and the geometry have been
assigned, the speed triangle and the
incidences depend only on the volume
flow of the processed fluid.
FIGURE 4.26 Loading coefficient vs. flow coefficient for a

centrifugal stage.
Pressure drop due to incidence may

therefore be expressed in the form:
Accordingly, indicating by WA the work per

mass unit associated with dissipation due
to friction, we may write:
WA = KA Q002
(4.39)
W1 = Ku ( Q00 Q00*)2 + Ko (4.40)
(4.38)
This equation can be expressed in

dimensionless form:
where kA represents a suitable constant

that takes account of the specific fluid
dynamic characteristics of the stage in
question. In dimensionless terms this can
be expressed as:
198
Chapter 4
compressors
Curve A represents the evaluated

relationship between flow coefficient and
head, evaluated taking account of the
effective deviation phenomena that occur
in a real blading. Curve A thus represents
all of the energy per mass unit that is
transferred to the fluid and that is thus
theoretically available to be converted
under the form of pressure. This quantity is
diminished by the dissipation associated
with pressure drop due to friction (curve B)
and pressure drop due to incidence (curve
C), both expressed by parabolic equations.
Point C1 thus expresses the work which is
effectively contained in the fluid under the
form of potential energy and kinetic
energy for an assigned flow coefficient 1.
FIGURE 4.27 Typical airfoil losses distribution as a

function of incidence.
WyI = Ku ( ) +Ko (4.41)
The constants k0 and k1 are once again

associated to the particular problem
considered.
From this analysis, it can be stated that,

due to the shapes of the various curves
considered, the quantity defined above
tends to present a maximum in
coincidence with a clearly determined
value of 1.
Overall Pressure Drop. In going on to

consider the overall pressure drop of a
compressor stage, note that in general
both
of
the
above-mentioned
contributions will be present. The overall
pressure loss can thus be represented as:
WT = WI+WA
=KAQ002 + KU(Q00-Q00*)2 + Ko
(4.42)
or in dimensionless form:
WT = KA + KU (1-1*)2 + Ko
(4.43)
FIGURE 4.28 Losses breakdown as a function of the flow
coefficient.
The above equation can be given

significant graphic interpretation, as
shown in
Other Pressure Drop Contributions. The

description given in the preceding
paragraphs of subsection 4.2.5.3 provides
a substantially correct illustration of the
main effects linked to pressure drop in a
generic compressor and their influence on
Fig. 4.28.
199
Chapter 4
compressors
overall performance. However, real

compressors present dissipation effects,
which do not fall within the simplified
scheme of pressure drops due to incidence
and pressure drops due to friction. In these
cases, it is often necessary to classify the
pressure drop contributions through
detailed reference to the physical modes in
which dissipation takes place.
type linked to the generation of impact

waves and to consequent production of
entropy. In general these consist of leading
edge impact waves and port impact waves,
depending on the place where these
effects occur. These phenomena tend to
involve all types of compressors, both axial
and centrifugal, with the exception of the
totally subsonic ones.
For centrifugal machines the main effects

of additional pressure drop are linked to
the presence of the blade tip and casing
recess (in open machines), to ventilation
phenomena between the rotating and
stationary surfaces in the spaces between
hubs and diaphragms, and to the presence
of end seals and inter-stage seals.
Pressure drop due to mixing. This

consists of irreversibility associated with
transition between a non-uniform fluiddynamic state, linked for example to local
separation effects, and a uniform
condition. These phenomena take place in
the regions downstream of the stator or
rotor blade arrays and are estimated
through experimental correlations.
Further pressure drops can be attributed

to the presence of separation areas in the
impeller.
In spite of the physical diversity of the

pressure drop contributions involved, the
qualitative considerations on the overall
pressure drop curves, presented in
subsection 4.2.5.3 in the paragraph
entitled Overall Pressure Drop, remain
valid in a general sense for both axial and
centrifugal compressors.
For axial machines, the representation of

pressure drop is slightly different to take
account of the different aerodynamic
phenomena involved. One possible
distinction
Could be the following:
Contour pressure drop. This is pressure
drop deriving from the presence of
boundary layers which develop along the
blade surfaces. It can be estimated through
the methods used for calculating turbulent
boundary layers.
4.2.5.4 Operating Curve Limits:

Surge and Choking
The operating curves of the stages, both
centrifugal and axial, present limits to the
flow ranges that can be processed by the
stage itself or by the machine of which it is
a part. These limits are established by two
separate phenomena, called surge and
choking, described below.
Endwall pressure drop. Pressure drop of

this kind depends on the presence of
localized limit states on the casing surface
or the compressor rotor. These effects are
usually evaluated through experimental
correlations.
Surge. The term surge indicates a

phenomenon of instability which takes
place at low flow values and which involves
an entire system including not only the
compressor, but also the group of
Pressure drop due to impact. This term

indicates phenomena of the dissipation
200
Chapter 4
compressors
components traversed by the fluid

upstream and downstream of it. The term
separation indicates a condition in
which the boundary layer in proximity to a
solid wall presents areas of inversion of the
direction of velocity and in which the
streamlines tend to detach from the wall.
Separation is in general a phenomenon
linked to the presence of adverse
pressure gradients in respect to the main
direction of motion, which means that the
pressure to which a fluid particle is
subjected becomes increasingly higher as
the particle proceeds along a streamline.
Surge prevention is effected through

experimental tests in which pressure
pulsation at low flow rates is measured on
the individual stages. On this basis, it is
possible to identify the flow values at
which stable operation of the stage is
guaranteed.
A knowledge of the operating limits of
each stage can then be used to evaluate
the corresponding operating limits of the
machine as a whole.
Choking. Assume that a stage of assigned
geometry is operating at a fixed speed of
rotation and the flow rate of the processed
fluid is increasing. A condition will
ultimately be reached at which, in
coincidence with a port, the fluid reaches
sonic conditions. In this situation, termed
choking, no further increase in flow rate
will be possible and there will be a rapid,
abrupt decrease in the performance of the
stage.
The term stall, referring to a

turbomachine stage, describes a situation
in which, due to low flow values, the stage
pressure ratio or the head do not vary in a
stable manner with the flow rate. Stall in a
stage is generally caused by important
separation phenomena in one or more of
its components.
Surge is characterized by intense and rapid
flow and pressure fluctuation throughout
the system and is generally associated with
stall involving one or more compressor
stages. This phenomenon is generally
accompanied by strong noise and violent
vibrations which can severely damage the
machines involved.
The occurrence of choking depends not

only on the geometry and operating
conditions of the stage, but also on the
thermodynamic properties of the fluid. In
this regard, choking can be particularly
limiting for machines operating with fluids
of high molecular weight, such as coolants.
Experience has shown that surge is

particularly likely to occur in compressors
operating in conditions where the Q-H
curve of the machine has a positive slope.
Many types of compressors, including

industrial process compressors, normally
operate in conditions quite far from those
of choking. For these machines, the
maximum flow limit is frequently defined
as the flow corresponding to a prescribed
reduction in efficiency in respect to the
peak value.
Less severe instability can moreover take

place also in proximity to areas of null
slope. This depends on the presence of
rotary stall, defined as the condition in
which multiple separation cells are
generated which rotate at a fraction of the
angular velocity of the compressor.
201
Chapter 4
compressors
As previously stated for a centrifugal

compressor efficient operation at constant
N lies to the right side of a pseudoparabolic
line called surge line, approximately
falling near maximum point for pressure.
Both for a axial or centrifugal compressor
the surge line delimits the range of stable
working conditions, unstable operation
being characterized by severe oscillation of
the mass flow rate.
4.2.5.5 Antisurge Protection System

4.2.5.5.1 General
Surge control represent a regulation
system to maintain compressors inside
their stable working range, assuring a
volume flow rate at impeller inlet section,
higher than the surging rate. An efficient
control method prevents compressors and
other turbo-machines from crossing the
surge line and avoids rotating stall
conditions for compression ratios as wide
as
possible.
These
aerodynamic
instabilities are intrinsic to almost all kinds
of turbo-machines, and often represent a
strong limitation to the range of efficient
performance. Anti-surge control systems
can thus represent a useful instrument to
improve the global performance of a
compressor.
The extreme regulation line (dotted line in

the picture), should be parallel and slightly
to the right with respect to the actual surge
line.
The measure of the volumetric flow rate
processed by the turbo-machine is
necessarily the key point for any kind of
regulation system. In general the surge
control system can be chosen according to
different specific requirements and relies
Fig. 4.29 represents a typical performance

map, obtained from compressor test
results. In the figure the compression ratio
= (p2 /p1), across the whole machine
on other physical variables different from

volume flow rate, but some basic features
should be guaranteed:
is plotted against volume flow rate Q1. One

of the most striking features of a typical
a. The regulation line should be as close as

possible to the surge limit line moved
parallel to the capacity axis by an
established amount.
b. When conditions of the suction fluid
vary, the regulation line should not get any
closer to the surge line, relating to the
design conditions, than may be necessary
for proper operation of the antisurge
system.
c. The regulation system should protect
the machinery in an automatic way during
start up, standstill, and all possible off
design conditions.
FIGURE 4.29 Typical performance map for a compressor

stage.
4.2.5.5.2 Basic Regulation Theory
performance characteristic is the strong

dependence shown by the compressor on
the rotational speed (N1 and N2 in figure).
The application of general formulation of

similarity laws in a turbo-machine states:
202
Chapter 4
compressors
the working condition for two different

rotational speeds are dynamically similar if
all fluid particles, in corresponding points
within the machine, have the same
direction and velocities proportional to
blade speed.
with a calibrated orifice, through the

following relation:
Q1= CQ 2
in which h is the measured pressure loss in

the orifice, expressed in mmH2O
If points, belonging to different

dimensional
characteristic
curves,
represent dynamically similar states, then
the non-dimensional groups of variables
involved in the physical problem (ignoring
Reynolds number effects), are expected to
have the same numerical value.
CQ is a flow coefficient depending both on

the geometrical characteristics of the
orifice ( 2 / 1) and, the real flow
conditions though Cr (related to viscosity
losses and flow contraction):
CQ =
Hp = KN2
(4.45)
Once the geometry of the orifice and his

fluid-dynamics characteristics are known
the following expression for volume flow
rate can be used:
Q1 =
(4.49)
where 1 is the specific weight, expressed

in the following way for a perfect gas:
Eliminating N, from (4.44) and (4.45), the

equation for the surge curve can be
obtained:
Hp = KQ2
(4.48)
Fluid compressibility is accounted in

equation (4.47) c depending on pressure
drops in the orifice.
If similarity theory is applied to the same

turbo-machine, D is constant, so dynamic
head becomes proportional to N2 and
volume flow rate to N. In particular, for
surge points the following laws can be
applied:
(4.44)
()
The dotted line in Fig. 4.29 shows that the

set of dynamically similar states lies on a
parabola, since the non-dimensional
groups Q/(ND3) and gH/(ND)2 are in
similarity.
Q1 = K N
(4.47)
(Kg/m3)
(4.50)
Substituting in equation (4.49) the 1 value

obtained from equation (4.50), the
following final expression for the volume
flow rate can be obtained:
(4.46)
Surge points, in a single stage compressor,

for the various speeds lay on a parabola
with vertex in the origin and axis coinciding
with the axis of ordinates on the polytropic
headvolumetric flow rate plane. The
polytropic head can be obtained from
(4.8), while volume flow rate can be
experimentally derived from measures
Q1 =
(4.51)
with h1 expressed in Kg/cm2

Substituting equations (4.47) and (4.51) in
(4.46) the following relation can be
203
Chapter 4
compressors
obtained between stage compression ratio

in incipient stall condition and volumetric
flow rate measured in the calibrated
orifice:
II. Position.
1) Pre- cooler.
2) Inter cooler.
3) After cooler.
(/ )=
4.3.3 Pre- cooler.

(4.52)
It used if the suction compressor

temperature is too high and required a
large number of inter coolers to reduce it.
An economic study must be performed to
optimize between using inter cooler or pre
cooler.
4.3 Cooling system

4.3.1 Introduction
4.3.4 Inter cooler
The maximum allowable temperature for

compressor case 180 C and maximum
operating temperature 175 C. The
discharge temperature from the station in
some application like pipeline projects
must be lower than 60 C, to protect the
pipeline coating.
The inlet temperature to the second stage

compressor will depend on the coolant
temperature and permissible approach.
Usually, we try to make the inlet
temperature to the compressor to be low
as much as possible so that we can use
higher pressure ratios without exceeding
the discharge temperature limit.
Cooler function
1- Protect the equipments from
overheating due to material temperature
limitation.
Texit = Tinlet
(4.53)
In fact, the intercooling process is used for

two reasons:
2- Reduce the pressure drop along the

pipeline.
3- Reduce the power requirements used by
compressors and other equipment.
1- Decreasing the compressing station

discharge temperature to the required
limit.
4- Protect the pipeline external coating

from damage due to high discharge
temperature.
2- Reducing the overall compression

power.
Isothermal compression is presented here
to represent the upper limits of cooling
and horsepower savings.
4.3.2Cooler classification
Coolers classified according to:
I. The refrigerant Fluid.
1) Water-cooled Heat Exchangers.
2) Air-cooled Heat Exchangers.
204
Chapter 4
compressors
Where m is the mass flow rate and R gas

is the universal gas constant divided on the
gas molecular weight.
This equation is useful when evaluating the
benefit of multiple intercoolers because it
establishes the theoretical power limit that
can be achieved by cooling. So, Percent
Uncooled Horsepower is introduced in
equation:
FIGURE 4.30 Intercooling Arrangement between Two
Successive Compression.
% Uncooled Hp =

(4.56)
Therefore, the minimum %Uncooled HP
could be satisfied is:
% Uncooled Hpmin =
(4.54)
Where P is the absolute pressure and V is

the volume.
From this, a theoretical value for the power
used by the compressor can be found by
equation:
(4.57)
The figure below is a plot comparing the

effect of different numbers of intercoolers
in terms of % uncooled horsepower.
It is the equivalent of an infinite number of

intercoolers and is not achievable in the
practical types of compressors. For
isothermal process:
WI.T = m Rgas T ln ( )
Usually it is not practical to install more

than three or two intercoolers.
Figure 4.31 Intercooling Effect on T-S Diagram.
PV = constant
(4.55)
205
Chapter 4
compressors
In the real, the Px Py. And this pressure

drop should be taken in consideration.
Figure 4.32 Overall pressure Ratio.
Note the diminishing effect as the number

of coolers is increased.
Figure 4.33 Intercooler Pressure.
As a rule of thumb, the cooler pressure

drop can be approximated by using 2% of
the absolute pressure at the entrance to
the cooler. Because the percentage gives
unrealistic values at the lower pressures, a
lower limit of 2 psi should be used.
Also note that by increasing the number of

intercoolers we will increase the pressure
losses leading to consuming higher
horsepower.
This is a very important step to choose the
proper intercooling pressure Px.
The pressure drop must be made up by the

compressor by additional head, and can be
added to the first or second section
pressure ratio.
The optimum position of an intercooler is

found to positioned so that:
By applying a little experience, the

guessing can be improved. Usually the
front section has a lower inlet temperature
and is generally more efficient.
(4.58)
And for the case of two intercoolers:

()
()
So the best location for additional pressure

would be in the first section.
(4.59)
(4.60)
206
Chapter 4
compressors
2. Help reduce the pressure drop along the

pipeline because the gas will be flowing at
a colder temperature.
3. Reduce the power requirements at the
downstream compressor station when it
receives gas at a lower suction
temperature.
4.4 Compressor Prime

Mover
4.4.1 Introduction
Prime movers for mechanical drives is a
common term for machines made for
transferring mechanical energy to
compressors.
prime mover develope the required torque
at a constant speed or over a range of
speeds. The driver's energy source can be
either electrical or mechanical. Electrical
energy is used by motors, either of the
induction or synchronous type, while the
mechanical energy covers a multitude of
sources. It may be a fuel, as in internal or
external combustion engines, or it may be
a gas, such as steam or process gas used in
a turbine or expander.
Figure 4.34 Comparison of work inputs to a single-stage

compressor (1AC) and a two-stage compressor with
intercooling (1ABD).
4.4.2 Prime Mover Types

There are main four type of prime movers:
1) Electrical motors.
4.3.5 After cooler
Electric motor drives offer efficient

operation and add flexibility to the design
of petroleum refineries, petrochemical
plants, and gas processing plants. Electric
motors can be built with characteristics to
match almost any type of load. They can be
designed to operate reliably in outdoor
locations where exposed to weather and
atmospheric contaminants.
Coolers may also be used as after-coolers

(at the discharge of a compressor station)
to:
1. Protect the pipeline external coating
from damage at high temperatures
(exceeding 65C or 149 F).
207
Chapter 4
compressors
Electrical energy is used by motors, either

of the induction or synchronous type.
3) Internal combustion engines.

Internal combustion engines are classified
according to the type of fuel used and the
method of fuel ignition. Many subclassifications are used to describe engines
according to their speed, cycle
arrangements, mechanical configuration,
and other design characteristics.
Spark ignition and compression ignition are
the two methods of initiating combustion
used in reciprocating internal combustion
engines. In practice the ignition method
also defines the fuels or range of fuels used
. also there are Dual-fuel engines may
operate in one of two modes. One mode is
as an ordinary diesel engine. It may also
operate on a gaseous fuel with a pilot
injection of liquid diesel fuel for ignition.
Figure 4.35 Electric Motor
2) Steam turbines.
Mechanical drive steam turbines are prime
movers for compressor, blower, and pump
applications. Steam turbines are available
for a wide range of steam conditions,
horsepower, and speeds. Typical ranges
for each design parameter are:
Inlet Pressure, psig 30 2000
Inlet Temperature, F saturated 1000
Exhaust Pressure, psig saturated 700
Horsepower 5 100,000
Speed, rpm 1800 14,000
Figure 4.37 Internal combustion engine
4) Gas turbines.
The gas turbine was first widely used as an
aircraft power plant. However, as they
became more efficient and durable, they
were adapted to the industrial
marketplace. Over the years the gas
turbine has evolved into two basic types
for high power stationary applications: the
industrial or heavy-duty design and the
aircraft derivative design.
Figure 4.36 Three stages steam turbine
208
Chapter 4
compressors
turbine is normally specified. The industrial

gas turbine has certain advantages which
should be considered when determining
application requirements. Some of these
are:
Figure 4.38 Gas turbine
Less frequent maintenance.

Can burn a wider variety of fuels.
Available in larger horsepower
sizes.
2) the aircraft derivative design.

An aircraft derivative gas turbine is based
on an aircraft engine design which has
been adapted for industrial use. The
engine was originally designed to produce
shaft power and later as a pure jet. The
adaptation to stationary use was relatively
simple.
4.4.3 Prime Mover Selection

The main type of natural gas compressors
in the grid is the centrifugal compressor we
recommended Gas Turbine to be the main
driver for the centrifugal compressor in the
compression station because:
Some of the advantages of the aircraft

derivative gas turbines are:
1) The main reason is the high availability

of source of energy because there will be
gas in the pipe line.
2) Gas turbines are high-speed machines.

3) Large amounts of power.
4) Reliable electric grid within a short

distance from the compressor station, also
transmission lines are costly.
Higher efficiency than industrial

units.
Quick overhaul capability.
Lighter and more compact, an
asset where weight limitations
are important such as offshore
installations.
4.4.4 Gas turbine as Prime mover

4.4.4.1Gas turbine types
Gas turbine classified according to:
. Applications.
1) the industrial or heavy-duty design.
The industrial type gas turbine is designed
exclusively for stationary use. Where high
power output is required, 35,000 hp and
above, the heavy duty industrial gas
Figure 4.39 the aircraft derivative design.
209
Chapter 4
compressors
II. Shaft configuration.

1) Single shaft design.
In a single shaft design, all rotating
components of the gas turbine are
mounted on one shaft ; The single shaft
design is simpler, requiring fewer bearings,
and is generally used where the speed
range of the driven equipment is narrow or
fixed (as in generator sets). It requires a
powerful starting system since all the
rotating components (including the driven
equipment) must be accelerated to idle
speed during the start cycle.
Figure 4.40 Single shaft design and Split shaft design.
2) Split shaft design.

In a split shaft design, the air compressor
rotating components are mounted on one
shaft, and the power turbine rotating
components are mounted on another
shaft. The driven equipment is connected
to the power turbine shaft; A split shaft
design is advantageous where the driven
equipment has a wide speed range or a
high starting torque. The air compressor is
able to run at its most efficient speed while
the power turbine speed varies with the
driven equipment. The split shaft design
allows a much smaller starting system
since only the air compressor shaft is
accelerated during the start cycle.
4.4.4.2 Thermodynamic Principle

The Brayton cycle is used for gas turbines
only where both the compression and
expansion processes take place in rotating
machinery. Gas turbines usually operate
on an open cycle. Fresh air at ambient
conditions is drawn into the compressor,
where its temperature and pressure are
raised. The high pressure air proceeds into
the combustion chamber, where the fuel is
burned at constant pressure. The resulting
high-temperature gases then enter the
turbine, where they expand to the
atmospheric pressure while producing
power. The exhaust gases leaving the
turbine are thrown out (not recirculated),
causing the cycle to be classified as an
open cycle.
The open gas-turbine cycle described
above can be modeled as a closed cycle, by
utilizing the air-standard assumptions.
Here the compression and expansion
processes remain the same, but the
combustion process is replaced by a
constant-pressure heat-addition process
from an external source, and the exhaust
process is replaced by a constant pressure
210
Chapter 4
compressors
heat-rejection process to the ambient air.

The ideal cycle that the working fluid
undergoes in this closed loop is the
Brayton cycle, which is made up of four
internally reversible processes:
1-2 Isentropic
compressor)
compression
(in
2-3 Constant-pressure heat addition

3-4 Isentropic expansion (in a turbine)
4-1 Constant-pressure heat rejection
Figure 4.41 An open-cycle gas-turbine engine.

Figure 4.43 T-S and P-V diagram of an ideal Brayton
cycle.
Notice that all four processes of the

Brayton cycle are executed in steady flow
devices; thus, they should be analyzed as
steady-flow processes. When the changes
in kinetic and potential energies are
neglected, the energy balance for a steadyflow process can be expressed, on a unit
mass basis
( qin qout )+ (win wout)

= hexit - hinlet
Figure 4.42 A closed-cycle gas-turbine engine.
(4.61)
Therefore, heat transfers to and from the

working fluid are
qin= h3-h2 =Cp(T3-T2)

And
211
(4.62)
Chapter 4
compressors
qout= h4-h1 = Cp(T4-T1)

th,brayton =
= 1
= 1
Therefore, the high-pressure air leaving

the compressor can be heated by
transferring heat to it from the hot exhaust
gases in a counter-flow heat exchanger,
which is also known as a regenerator or a
recuperator.
(4.63)
( )
( )
Under the cold-air-standard assumptions,

the thermal efficiency of an ideal Brayton
cycle with regeneration is
= 1
( )
( )
(4.64)
Processes 1-2 and 3-4 are isentropic, and

P2=P3 and P4=P1 thus,
=( )
= ( )
th,regen = 1 ( ) ( )
(4.68)
(4.65)
Substituting these equations into the

thermal efficiency relation and simplifying
give:
th,brayton = 1
Where
(4.66)
Figure 4.44 A gas-turbine engine with regenerator.
(4.67)
rp is the pressure ratio and k is the specific

heat ratio. efficiency equation shows that
under the cold-air-standard assumptions,
the thermal efficiency of an ideal Brayton
cycle depends on the pressure ratio of the
gas turbine and the specific heat ratio of
the working fluid. The thermal efficiency
increases with both of these parameters,
which is also the case for actual gas
turbines.
THE
BRAYTON
REGENERATION
CYCLE
WITH
Figure 4.45 T-S diagram of a Brayton cycle with
regeneration.
In gas-turbine engines, the temperature of

the exhaust gas leaving the turbine is often
considerably higher than the temperature
of the air leaving the compressor.
Therefore, the thermal efficiency of an

ideal Brayton cycle with regeneration
212
Chapter 4
compressors
depends on the ratio of the minimum to

maximum temperatures as well as the
pressure ratio. The thermal efficiency is
plotted in Fig 4.46 for various pressure
ratios
and
minimum-to-maximum
temperature ratios. This figure shows that
regeneration is most effective at lower
pressure ratios and low minimum-tomaximum temperature ratios.
4.47 A gas-turbine engine with two-stage compression

with intercooling, two-stage expansion with reheating,
and regeneration.
Figure 4.46 Thermal efficiency of the ideal

Brayton cycle with and without regeneration.
THE
BRAYTON
CYCLE
INTERCOOLING,
REHEATING,
REGENERATION
WITH
AND
Figure 4.48 T-S diagram of an ideal gas-turbine cycle
with intercooling, reheating, and regeneration.
The net work of a gas-turbine cycle is the

difference between the turbine work
output and the compressor work input,
and it can be increased by either
decreasing the compressor work or
increasing the turbine work, or both.
It was shown that the work required to

compress a gas between two specified
pressures can be decreased by carrying out
the compression process in stages and
cooling the gas in between (Fig 4.49) that
is, using multistage compression with
intercooling. As the number of stages is
increased, the compression process
becomes nearly isothermal at the
compressor inlet temperature, and the
compression work decreases.
213
Chapter 4
compressors
4.4.4.4 Gas Turbine Components
Figure 4.51 gas turbine sections.
1) Air Inlet System.
Figure 4.49 Comparison of work inputs to a single-stage

compressor (1AC) and a two-stage compressor with
intercooling (1ABD).
The air inlet system, down-stream of the

air filtering installation, It consists of an air
duct, followed by sections of parallel
baffles silencers, then a screen system
located in an inlet elbow, and an expansion
joint after which air will reach the gas
turbine air inlet plenum. The gas turbine
inlet plenum contains the compressor inlet
casing.
4.4.4.3 Principle of operation
The silencers are of baffle-type

construction to attenuate the high
frequency noise in the air inlet, caused by
the rotating compressor blading. More
details are given in the "Auxiliary plants
and systems" volume, especially about the
filtering installation.
Figure 4.50 gas -turbine principle of operation.
Ambient air enters the air compressor and

is compressed and then passes to the
combustion chamber. This combustion
causes a sudden increase in temperature.
The hot gases from the combustion
chamber then enter the turbine. The
turbine expands the burned gases from the
combustion section and produces
mechanical power to drive the gas
compressor or any mechanical equipment.
2) Compressor Section.
In the compressor, air is confined to the
space between the rotor and stator blading
where it is compressed in stages by a series
of alternate rotating (rotor) and stationary
(stator) airfoil shaped blades. The rotor
blades supply the force needed to
compress the air in each stage and the
stator blades guide the air so that it enters
in the following rotor stage at the proper
angle. The compressed air exits through
214
Chapter 4
compressors
the compressor discharge casing to the

combustion chambers. Air is extracted
from the compressor for turbine cooling,
for bearing sealing, and during startup for
pulsation control.
bolted to the compressor case, and the

back end is attached to the combustion
section.
Since minimum clearance between rotor

and stator provides best performance in a
compressor, parts have to be made and
assembled very accurately.
Figure 4.54 gas turbine diffuser section.
Figure 4.52 gas turbine compressor section
4) Combustion chamber section.

The combustion system is of the reverseflow type, the combustion chambers
arranged around the periphery of the
compressor discharge casing. This system
also includes fuel nozzles spark plug
ignition system, flame detectors, and
crossfire tubes. Hot gases, generated from
burning fuel in the combustion chambers,
are used to drive the turbine. High
pressure air from the compressor
discharge is directed around the transition
pieces and into the combustion chambers
liners.
Figure 4.53 Compressor section.
3) Diffuser.
The diffuser is located between the
compressor section and the main
components of the combustion section.
The purpose of the diffuser is to prepare t
the air for entry into the combustion
section. The front end of the diffuser is
215
Chapter 4
compressors
6) Exhaust Section.
This air enters the combustion zone

through metering holes for proper fuel
combustion and through slots to cool the
combustion liner. Fuel is supplied to each
combustion chamber through a nozzle
designed to disperse and mix the fuel with
the proper amount of combustion air.
In the exhaust section, the gases which

have been used to power the turbine
wheels are redirected to be released to
atmosphere After leaving the exhaust
frame, the hot gases reach the diffuser,
located in the exhaust plenum. On the
exhaust plenum wall facing the exhaust
diffuser, a circular arrangement of
thermocouples permits exhaust gas
temperature
measurement.
The
thermocouples send their signals to the
gas turbine temperature control and
protection system. The exhaust plenum
configuration is that of a box open at the
top and welded to an extension of the
turbine base.
5) Turbine Section.
4.4.4.5 Factors Affecting Gas

Turbine Performance.
Figure 4.55 combustion chambers.
1- Air Temperature and Site Elevation.
The turbine section is located between the

combustion and the exhaust sections of
the engine. The purpose of the turbine
section is to convert the energy of the
expanding gases into mechanical energy to
drive the compressor, the accessories, and
the load.
2- Humidity.
3- Inlet and Exhaust Losses.
4- Fuels.
5-Diluent Injection.
6- Air Extraction.
7- Inlet Cooling.
4.5 Scrubber
4.5.1 Definition
A scrubber is a two-phase separator that is
designed to recover liquids carried over
from the gas outlets of production
separators or to catch liquids condensed
due to cooling or pressure drops. Liquid
loading is much lower than that in a
separator.
Figure 4.56 turbine section details.
216
Chapter 4
compressors
4.5.3 Selection criteria for gas/liquid

separators
I. Duty
It is often necessary to separate liquid and
gas phases in a certain stage of an
operation or process. Since both the
conditions of the wet gas stream (or more
generally the gas/liquid stream) and the
required efficiency may vary widely, care
shall be taken in selecting a separator in
order to match the specific duty. For
instance, a gas/liquid separator upstream
of a gas compressor would need to be very
efficient, whereas in other cases a simple
knock-out vessel may be sufficient if only
bulk separation of the gas and liquid
phases is required (e.g. upstream of a heat
exchanger in which condensation will take
place).
II. Orientation
Figure 4.57 Different shapes of natural gas scrubber.
For gas/liquid separation, a vertical vessel

should normally be selected for the
following reasons:
4.5.2 Applications
1) A smaller plan area is required (critical

on offshore platforms).
1) Upstream of mechanical equipment

such as compressors that could be
damaged, destroyed or
rendered
ineffective by free liquid.
2) Easier solids removal.

3) liquid removal efficiency does not vary
with liquid level (area in vessel available for
gas flow remains constant).
2) Down stream of equipment that can

cause liquids to condense from a gas
stream (such as coolers).
4) Vessel volume is generally smaller.

However, a horizontal vessel should be
chosen if:
3) Up stream of gas dehydration

equipment that would lose efficiency, be
damaged, or destroyed if contaminated
with liquid hydrocarbons.
1) large liquid
accommodated.
slugs
have
to
be
2) there is restricted head room.
4) Upstream of a vent or flare outlet.
3) A low downward liquid velocity is

required (for de-gassing purposes or for
foam breakdown).
217
Chapter 4
compressors
4.5.4 Components of Scrubber
4) axial flow (e.g. "swirltubes").
I. Feed inlet.
5) reversed flow (tangential).
This comprises the upstream piping, the

diameter of the inlet nozzle is a function of
the feed flow rate and pressure.
6) filter candles.
The criterion for nozzle sizing is that the

momentum of the feed shall not exceed
prescribed levels. The maximum allowable
inlet momentum can be increased by
fitting inlet devices.
A combination can also be used, such as:
The function of the inlet device is to initiate

gas/liquid separation and distribute the
gas flow evenly in the gas compartment of
the vessel.
3) vertical-flow vane + swirltubes (in SVS).
1) wiremesh + swirltubes (in SMS).

2) wiremesh + swirltubes + wiremesh (in
SMSM).
4) filter candles + vane pack or wiremesh

(in two-stage filter separator).
Selection depends on the required
efficiency, capacity, turndown, maximum
allowable pressure drop and fouling
tolerance.
II. Separator internals.

After passage through the feed inlet, the
gas stream will usually still contain liquid in
the form of droplets. Normally a
separation internal is required in the vessel
to complete the separation process. Only
when moderate separation efficiency is
required and the droplets are relatively
large can the separation internal be
omitted. By selection of a sufficiently large
vessel diameter the gas velocity in the
vessel can be kept low and the majority of
the droplets will settle by gravity. This
separation mechanism is used in knock-out
vessels.
III. Gas and liquid outlets

After completion of the gas/liquid
separation process the two phases will
leave the vessel via the gas and liquid
outlet respectively.
4.5.5 Types of separators used as

scrubber
Vertical scrubbers are most commonly
used. Horizontal scrubbers can be used,
but space limitations usually dictate the
use of a vertical configuration.
Separation internals are required in all

other types of gas/liquid separators. The
choice of internal depends on the required
duty.
There are many types of vertical separator

used as scrubber.
1) Vertical wiremesh demister.
Options include:
2) Vertical vane-type demister.
1) wiremesh
3)
Schoepentoeter-mistmat-swirldeck
separator
2) vane pack (either for horizontal or

vertical flow).
4) Schoepentoeter-vane pack-swirldeck
separator
3) (multi)cyclones.
218
Chapter 4
compressors
5)
Schoepentoeter-mistmat-swirldeckmistmat separator.
6) Cyclone with straight inlet.
7) Vertical separator with reversed-flow
multicyclone
bundle
(conventional
multicyclone).
Vertical wiremesh demister is the most
commonly used in compressor station
then vane-type. Others types stated above
not used in this case because they have
high pressure drop.
Figure 4.59 Horizontal gas scrubber
4.5.6 Scrubber design requirements

Separators designed for use in the gas
industry are mechanical devices, which
remove liquids from the natural gas
stream. Proper separator design must
accomplish the following:
1) Dissipate and control the energy of the
inlet gas stream.
2) Reduce gas and liquid velocities so that
initial gravitational separation will occur.
3) Allow adequate liquid retention time for
the separation of entrained gases from
hydrocarbon liquid streams.
Figure 4.58 Vertical gas scrubber.
4) Eliminate separated liquid reentrainment into the outlet gas phase.

5) Provide adequately sized connections
for inlet, gas outlet, liquid outlet, pressure
safety valve, level controls, level gauge,
pressure gauge, temperature
gauge, drains and inspection openings.
219
Chapter 4
compressors
may be necessary to flare gas during

drilling, production or processing.
4.6 Flare System

4.6.1 Introduction
To
safeguard
the
environment,
atmospheric discharges shall be limited to
nonflammable and non-toxic gases. Any
requirement to discharge hydrocarbons to
atmosphere shall be highlighted at the
earliest opportunity .The discharge point
shall be at a safe location and the flow
directed away from areas where personnel
may be present.
Flare systems are one of the most

important components at any plant.
The flare is a last line of defense in the safe
emergency release system in a refinery or
chemical plant.
What is natural gas flaring and why is it

happen?
Flaring is a volatile organic compound
(VOC) combustion control process in which
the VOCs are piped to a remote, usually
elevated, location and burned in an open
flame in the open air using a specially
designed burner tip, auxiliary fuel, and
steam or air to promote mixing for nearly
complete (> 98%)VOC destruction.
Completeness of combustion in a flare is
governed by flame temperature, residence
time in the combustion zone, turbulent
mixing of the components to complete the
oxidation reaction, and available oxygen
for free radical formation.
Figure 4.60 flare in chemical plant.
Flaring is the controlled burning of natural

gas and a common practice in oil/gas
exploration, production and processing
operations. A flare system consists of a
flare stack and pipes that feed gas to the
stack. Flare size and brightness are related
to the type and amount of gas or liquids in
the flare stack. Flares generate heat and
noise. Large flares can be quite noisy
because of the volume and velocity of the
gas going through the flare stack.
Combustion is complete if all VOCs are

converted to carbon dioxide and water.
Incomplete combustion results in some of
the VOC being unaltered or converted to
other organic compounds such as
aldehydes or acids.
The flaring process can produce some
undesirable by-products including noise,
smoke, heat radiation, light, SO , NO , CO,
and an additional source of ignition where
not desired. However, by proper design
these can be minimized. The Flare is a
major part of the pressure relieving system
for hydrocarbon processing plants. For
Because natural gas is valuable, companies

would rather capture rather than flare it.
However, there are several reasons why it
220
Chapter 4
compressors
most plants it is the last line of defense

against potential hazards. Flare design
should include thermal radiation limits,
explosion hazards, liquid carryover, noise,
temperature limits, and ground level
concentrations of combustion products.
header and transfer lines are removed by a

knock-out drum. The knock-out drum is
typically either a horizontal or vertical
vessel located at or close to the base of the
flare, or a vertical vessel located inside the
base of the flare stack. Liquid in the vent
stream can extinguish the flame or cause
irregular combustion and smoking.
4.6.2 Flare system components
Figure 4.61 flare system components.
Figure 4.62 Vertical and horizontal Knock out drum.
1-Gas transport piping
In addition, flaring liquids can generate a

spray of burning chemicals that could
reach ground level and create a safety
hazard. For a flare system designed to
handle emergency process upsets this
drum must be sized for worst-case
conditions (e.g., loss of cooling water or
total unit depressing) and is usually quite
large. For a flare system devoted only to
vent stream VOC control, the sizing of the
drum is based primarily on vent gas flow
rate with consideration given to liquid
entrainment.
Process vent streams are sent from the

facility release point to the flare location
through the gas collection header. The
piping (generally schedule 40 carbon steel)
is designed to minimize pressure drop.
Ducting is not used as it is more prone to
air leaks.
Valuing should be kept to an absolute
minimum and should be "car-sealed"
(sealed) open. Pipe layout is designed to
avoid any potential dead legs and liquid
traps. The piping is equipped for purging so
that explosive mixtures do not occur in the
flare system either on start-up or during
operation.
2- Knock out drum
Liquids that may be in the vent stream gas
or that may condense out in the collection
221
Chapter 4
compressors
the flare system and the flame front pulls

down into the stack.
Figure 4.63 knock out drum in flare system.
Figure 4.64 Liquid seal.
The liquid seal also serves to maintain a

positive pressure on the upstream system
and acts as a mechanical damper on any
explosive shock wave in the flare stack. (51
Other devices, such as flame arresters and
check valves, may sometimes replace a
liquid seal or be used in conjunction with
it. Purge gas also helps to prevent
flashback in the flare stack caused by low
vent gas flow.
Advantage of knock out drum
Disentrains liquid droplets.

Prevents flaming rain.
Reduces flare smoke.
Recovers liquids.
Design Feature
The main purpose of the liquid seal drum is

to stop flame propagation. By quenching
the flame with a barrier of water, Flare
Industries Liquid Seal design achieves this
purpose as well as providing other intrinsic
design advantages. Second, the liquid seal
acts as a large check valve so that gas
cannot travel upstream for any reason. A
third function of the liquid seal is to
Disentrains liquid droplets.
Automatic liquid level controls

Specially designed separation
internals
Inspection opening
Used in combination with liquid
seal
3- Liquid seal
Process vent streams are usually passed
through a liquid seal before going to the
flare stack. The liquid seal can be
downstream of the knockout drum or
incorporated into the same vessel. This
prevents possible flame flashbacks, caused
when air is inadvertently introduced into
Thus, the liquid seal drum provides flame

arresting capabilities, acts as a check valve,
and knocks liquids out of the waste gas.
Liquid seals can be designed as a separate
222
Chapter 4
compressors
vessel or as an integral part of the flare

riser structure.
potential. To prevent this, a gas seal is

typically installed in the flare stack.
4- Flare stack
One type of gas seal (also referred to as a

flare seal, stack seal, labyrinth seal, or gas
barrier) is located below the flare tip to
impede the flow of air back into the flare
gas network. There are also "seals" which
act as orifices in the top of the stack to
reduce the purge gas volume for a given
velocity and also interfere with the passage
of air down the stack from the upper rim.
These are known by the names "internal
gas seal, fluidic-seal, and arrestor
seal".These seals are usually proprietary in
design, and their presence reduces the
operating purge gas requirements.
For safety reasons a stack is used to elevate

the flare. The flare must he located so that
it does not present a hazard to surrounding
personnel and facilities. Elevated flares can
be self-supported (free-standing), guyed,
or structurally supported by a derrick.
Self-supporting flares are generally used
for lower flare tower heights (30-100 feet)
but can be designed for up to 250 feet. Guy
towers are designed for over 300 feet,
while derrick towers are designed for
above 200 feet. Free-standing flares
provide ideal structural support.
6- Burner tip
However, for very high units the costs

increase rapidly. In addition, the
foundation required and nature of the soil
must be considered. Derrick-supported
flares can be built as high as required since
the system load is spread over the derrick
structure. This design provides for
differential expansion between the stack,
piping, and derrick.
The burner tip, or flare tip, is designed to

give
environmentally
acceptable
combustion of the vent gas over the flare
system's capacity range. The burner tips
are normally proprietary in design.
Consideration is given to flame stability,
ignition reliability, and noise suppression.
The maximum and minimum capacity of a
flare to burn a flared gas with a stable
flame (not necessarily smokeless) is a
function of tip design. Flame stability can
be enhanced by flame holder retention
devices incorporated in the flare tip inner
circumference. Burner tips with modern
flame holder designs can have a stable
flame over a flare gas exit velocity range of
1 to 600 ft/sec. The actual maximum
capacity of a flare tip is usually limited by
the vent stream pressure available to
overcome the system pressure drop.
Derrick-supported flares are the most

expensive design for a given flare height.
The guy-supported flare is the simplest of
all the support methods. However, a
considerable amount of land is required
since the guy wires are widely spread
apart. A rule of thumb for space required
to erect a guy-supported flare is a circle on
the ground with a radius equal to the
height of the flare stack.
5- Gas Seal
7- Pilot burners
Air may tend to flow back into a flare stack

due to wind or the thermal contraction of
stack gases and create an explosion
Reliable ignition is obtained by continuous

pilot burners designed for stability and
positioned around the outer perimeter of
223
Chapter 4
compressors
the flare tip. The pilot burners are ignited

by an ignition source system, which can be
designed for either manual or automatic
actuation. Automatic systems are
generally activated by a flame detection
device using either a thermocouple, an
infra-red sensor or, more rarely, (for
ground flare applications) an ultra-violet
sensor.
hydrocarbon
molecule,
thereby
minimizing polymerization, and forms
oxygen compounds that burn at a reduced
rate and temperature not conducive to
cracking and polymerization. Another
theory claims that water vapor reacts with
the carbon particles to form CO, CO2, and
H2, thereby removing the carbon before it
cools and forms smoke. An additional
effect of the steam is to reduce the
temperature in the core of the flame and
suppress thermal cracking.
The physical limitation on the quantity of
steam that can be delivered and injected
into the flare flame determines the
smokeless capacity of the flare. Smokeless
capacity refers to the volume of gas that
can be combusted in a flare without smoke
generation. The smokeless capacity is
usually less than the stable flame capacity
of the burner tip. Significant disadvantages
of steam usage are the increased noise and
cost. Steam aggravates the flare noise
problem by producing high-frequency jet
noise. The jet noise can be reduced by the
use of small multiple steam jets and, if
necessary, by acoustical shrouding. Steam
injection is usually controlled manually
with the operator observing the flare and
adding steam as required to maintain
smokeless operation. To optimize steam
usage infrared sensors are available that
sense flare flame characteristics and adjust
the steam flow rate automatically to
maintain smokeless operation. Automatic
control, based on flare gas flow and flame
radiation, gives a faster response to the
need for steam and a better adjustment of
the quantity required. If a manual system
is used, steam metering should be installed
to
significantly
increase
operator
awareness
and
reduce
steam
consumption.
Figure 4.65 pilot burner.
8- Steam jets
A diffusion flame receives its combustion
oxygen by diffusion of air into the flame
from the surrounding atmosphere. The
high volume of fuel flow in a flare may
require more combustion air at a faster
rate than simple gas diffusion can supply.
High velocity steam injection nozzles,
positioned around the outer perimeter of
the flare tip, increase gas turbulence in the
flame boundary zones, drawing in more
combustion air and improving combustion
efficiency. For the larger flares, steam can
also be injected concentrically into the
flare tip.
The injection of steam into a flare flame
can produce other results in addition to air
entrainment and turbulence. Briefly, one
theory suggests that steam separates the
224
Chapter 4
compressors
9- Controls
4.6.3 Operation
Flare system control can be completely

automated or completely manual.
Components of a flare system which can be
controlled automatically include the
auxiliary gas, steam injection, and the
ignition system.
4.6.3.1 General operation of the

compressor station
A compressor is a machine driven by an
internal combustion engine or turbine that
creates pressure to "push" the gas through
the lines. Most compressors in the natural
gas delivery system use a small amount of
natural gas from their own lines as fuel.
Fuel gas consumption can be minimized by

continuously measuring the vent gas flow
rate and heat content (Btu/scf) and
automatically adjusting the amount of
auxiliary fuel to maintain the required
minimum of 300 Btu/scf for steam assisted
flares.
Steam consumption can likewise be
minimized by controlling flow based on
vent gas flow rate. Steam flow can also be
controlled using visual smoke monitors.
Automatic ignition panels sense the
presence of a flame with either visual or
thermal sensors and reignite the pilots
when flameouts occur.
Flare system operation monitoring
Figure 4.66 typical compressor station
1) Video camera
2) Pilot gas flow and temperature Relief
gas flow, pressure, temperature and
composition
4.6.3.2 Starting preparation
3) Knock-out drum level
2- All control, alarm and trip circuits should

be energized.
1- The whole unit should be clean and free

of all loose material.
4) Water seal level, pressure, and

temperature
3- Open stop valves on all pressure and

level instruments.
5) Steam flow to flare
4- Make sure that the settings of all

temperature, pressure and other alarm
and trip devices are correct.
6) Collection header purge gas flow

7) Flare gas recovery
5- Make sure all valves of the lube oil and

seal gas system are in their correct status.
225
Chapter 4
compressors
4.6.3.3 Starting Procedures
1. Check the oil level in the oil reservoir at

full mark, fill up if necessary and check the
condition of the oil.
2. Start the instrument air supply to the oil

guard between the dry gas seals and radial
bearings.
3.! CAUTION to prevent lube oil carry over

to the gas seal area, always start the air
supply to the labyrinths between the radial
bearings and the dry gas seals before the
oil pump is started.
From pressure indicator see at

what pressure the standby oil
pump starts.
(After testing, put the valves in
their original positions and stop
the stand by oil pump)
By start button on motor control
station
10. Check all components of the oil

systems to see that they are operating
properly.
11. Check the oil supply gauge pressure.
4. Verify that the lube oil filter and the seal

gas filter are ready to operate.
12. Make sure there is a normal flow of oil

to all components requiring lubrication by
checking the flow through the flow glasses.
5. Make sure that the lube oil cooler and

the cooling fans are ready to operate and
that the cooling system of the electric
motor driver - is ready to operate .Make
sure that all process equipment is ready to
operate.
13. Make sure that the compressor driver

is prepared for operation.
14. Make sure that the gas vent systems
are ready to accept gas.
15. Make sure that the compressor suction
and discharge valves are closed.
6. Make sure that the temperature of the

oil is not below 10 C. Operate the oil
reservoir heater if necessary.
16. Vent and pressurize the compressor

casing, to suction pressure.
7. Start the main oil pump
! WARNING
! WARNING: To prevent overloading of the

oil pump driver, the oil pump should not be
started at low oil temperature.
During venting, the anti-surge valve should

be forced to close in order to vent the
compressor casing. After venting the antisurge valve should be opened.
8. Vent the oil system thoroughly.

! NOTE: To avoid interruption of the oil
supply, the standby oil filter should be
filled with oil and vented before putting it
on line.
17. Open the drain valves of the

compressor to remove any condensate or
moisture.
9. Check the functioning of the standby oil

pump.
18. Close the drain valves of the

compressor after condensate stops
flowing.
By closing valve in the sensing line

to pressure transmitter and
opening the bleed off valve before
pressure transmitter.
19. Open the drain valves from the inboard

seal chambers to remove any condensate
from it.
226
Chapter 4
compressors
20. Close the drain valves after condensate

stops flowing.
remote area of the plant yard. Because of

the required venting, some noise can result
from compressor maintenance, activation,
or Emergency Shutdown System testing.
21. Make sure there are no trips.

The compressor permissive to start signal
will be healthy if:
When an emergency shutdown occurs?

It will be heard a very loud noise often
compared to the sound of a jet engine or a
freight train, depending on how close you
were to the station The sound will last
anywhere from one to four minutes. This
sound is the result of the release of
pressure from the compressor station
piping. You would likely see a vapor cloud
discharging into the air.
There are no trips from either the

compressor or the process.
The buffer gas supply pressure to the
labyrinths between the radial bearings
and the dry gas seals is sufficient.
The temperature of the oil at the cooler
outlet is higher than 30 C.
The level in the oil reservoir is high
enough.
The level in the lube oil rundown tank
is high enough.
When we take the decision of an

emergency shutdown?
Natural gas and flame sensors, located
inside the compressor building, constantly
monitor the station. If a problem is
detected, the emergency system will
activate automatically in order to protect
the community, station personnel and
facility.
22. Check dry gas seal leakage flow is

normal.
23. Start the driver.
24. Put the anti-surge valve on automatic
mode.
This stop is initiated by pushing the stop

button on the compressor control panel.
25. The startup procedure is completed.
Operators Stop: An operator planned stop

should be done as follows:
26. Check and record: oil pressure and

temperature, oil flow, vibrations, axial
position, and air supply.
Reduce the load of the compressors to

minimum.
Press the stop button at the compressor
control panel.
4.6.3.4 Emergency shut down

To protect the public, company personnel,
and property, all compressor stations are
equipped with several safety devices. One
of these safety systems, common to all
compressor stations, is an Emergency
Shutdown System. When activated, the
Emergency Shutdown System stops the
engines, isolates and vents the compressor
piping, and routes the gas away from the
station. During the venting process,
natural gas is released through a stack in a
To depressurize the compressor:

The Suction valve should be closed.
The discharge valve should be
closed.
227
Chapter 4
compressors
The blow down valve should be

opened.
Gas from Burullus terminal is used to

supply natural gas natural grid at 73 barg
maximum, 72 barg normal and 64 barg
minimum and ELNG at 55 bar
4.7 Operation Examples
The current arrival pressure is 83 barg but

its expected to fill to 59 barg and 31 barg
thereafter
4.7.1 Rashpetco compressor station

operation
The gas from fields is fed to RASHPETCO
compressor station to increase its pressure
to required pressure of network and ELNG
Egyptian liquefied natural gas
So to maintain gas delivery there are two

separate compressor station:
Booster compressor station: boosting the
domestic gas from 55 barg to 85 barg
Main compressor unit: it supply both
domestic and ELNG gas 26 barg to 59 barg.
Figure 4.67 RASHPETCO compressor station layout
228
Chapter 4
compressors
4.7.1.1 RASHPETCO Scenarios

There are many operation scenarios in
RASHPETCO compressor station, these
scenarios depend on the gas pressure from
fields and the network required pressure
and flow rate.
First scenario: (2B.C. &2M.c on)

In this scenario there are two booster and
two main compressors are operated as
shown in figure (4.68)
This scenario is the normal operation, the
booster compressor station takes the gas
from the main compressor in order to rise
it`s pressure and to provide the network
and ELNG with pressure 70, 55 barg
respectively.
The scenarios are
Two booster compressors and two

main compressors are on
Booster compressors are off
Booster compressors are on
Current scenario (3 main on)
Figure 4.68 first operating scenario (two booster compressor and two main compressor)
229
Chapter 4
compressors
Second
scenario:
COMPRESSOR OFF)
(BOOSTER
There are two main compressors working

and provide the ELNG with gas at pressure
(68 barg) and to the network with (65
barg).
Figure 4.69 second operating scenario (booster compressor off and three main compressors on)
Third scenario: the booster compressors

are on and the main compressors are off.
In this scenario the gas is provided to the

network with 65 barg and to the ELNG with
45 barg.
Fig. (4.70) third operating scenario (booster compressor on and main compressor off)
230
Chapter 4
compressors
Fourth scenario (current scenario)

In this scenario the booster compressors
are off and this is because it has many
problems as the maintenance intervals are
very short, it`s blade needed to be replaced
after a short time of its operating time and
its higher fuel consumption, so it is less
efficient.
Fig. (4.71) fourth operating scenario (booster compressor off and three main compressors on and two main compressor
stand by)
231
Chapter 4
compressors
And the 90% (sales gas) of inlet gas is

compressed into national grid:
4.7.2 WDGC compressor station

4.7.2.1 Schematic of operation and
processing
31 national grid pipeline with a capacity

of 649.1 MMSCFD.
Western desert gas complex (WDGC) is

considered a source node to the network
The natural gas is supplied to western
desert gas complex from two sources:
12 Dekhela pipeline with a capacity of

213.73 MMSCFD.
1. 24 Alameria LPG pipeline with a

capacity of 18.39 MMSCFD.
2. BAPETCO with capacity of 947.98
MMSCFD.
Figure 4.72 schematic of operation and processing
About 10% of inlet gas to WDGC has been

processed (85.15MMSCFD) for obtaining:
C2/C3.
Commercial propane(exporting)
LPG (local market).
Condensate (C5+).
4.7.2.2 Pressure schematic

The natural gas is supposed to enter the
WDGC with pressure equal to 70 bars, and
after processing the pressure decrease to
14 bars. Then pressure is boosting by
turbo-expander to 19 bar and finally enter
compressor station to be compressed to its
initial pressure which equal to 70 bars.
232
Chapter 4
compressors
Figure 4.73 pressure schematic of WDGC
4.7.2.3 WDGC compressor station

layout
WDGC compressor station provide the
national grid with natural gas with required
pressure. The compressor station consists
of four units supplying two pipelines:
National grid pipeline (31) compressors,
and contain two compressors each of them
is two stages 18.6 MW General Electric,
one compressor is also two stages 16 MW
siemens for maximization.
Dekhela pipeline (12) compressor which
is single stage 9 MW General Electric.
233
Chapter 4
compressors
Figure 4.74 WDGC compressor station layout
Figure 4.75 compressor station layout
234
Chapter 4
compressors
4.7.2.4 Operation scenarios of WDGC compressor station

1st Scenario-by pass valve is totally closed
(normal operation)
Figure 4.76 1st Scenario-by pass valve is totally closed
2nd Scenario-by pass valve is totally open

(Processing plant is isolated)
Figure 4.77 2nd scenario-by pass valve is totally open
235
Chapter 4
compressors
3rd Scenario-bypass valve is partially

closed (rush hour) For raising calorific
value Of sales gas.
Figure 4.78 3rd scenario-bypass valve is partially closed
4th Scenario Dekheila valve is closed for

avoiding back pressure On Amerya LPG
compressor station
Figure 4.79 4th scenario dekheila valve is closed
236
Chapter 4
compressors
5th Scenario closing one of four compressor
Figure 4.80 5th scenario closing one of four
4.7.3 Dahshour compressor station

4.7.3.1 General schematic of
compressor station
Figure 4.81 General schematic of compressor station
237
Chapter 4
compressors
The gas arrives from the national grid

through the single tie-in point TP001 (refer
to P&ID SOK0886414 sh.2).
and is
addressed to the suction piping of the four
compression stations and to the common
conditioning skid 200-X-08, which feeds
the gas turbines.
After compression the gas is sent to the
common discharge header, which Collects
the single discharges of the four units (tiein point TP004).
The compression stations are single stage,
gas turbine driven. and are provided with
an anti-surge line that permits partial or
total recycle flow in case of critical
conditions of the compressor. and an onoff hot bypass line for fast response to
avoid surge flow in the compressor.
Figure 4.82 schematic of compressor station
238
Chapter 4
compressors
4.7.3.2 Compressor suction

description
4.7.3.4Compressor suction
description (k.o drum)
Reference is made to Process gas P&ID

SOK0886415 sh.2 & 3.
The scrubber is designed to collect any

liquid droplets that may be entrained in
the gas flow (maximum water contained in
the feed gas is 74 ppm in volume).
Main components are:
1. Suction scrubber 200-V-11x.
The separation is made by gravity using a

vane pack device, which removes the
residual liquid entrained in the gas.
2. Compressor 200-K-11x.
The scrubber is provided with:
4.7.3.3 Compressor suction

description (valves)
1. High level protections LAHH-x604.
Process gas coming from tie-in point TP001

is delivered through the 200-V-11x (single
stage vertical vane-pack type scrubber).
2. Low level protections LALLx605.
Since the gas can arrive with a pressure

ranging from 25.3 to 50 bara, the throttling
valve PV-x203 is installed to limit the
pressure at 30 bara (in case of inlet
pressure lower than 30 bara, the valve will
be fully open taking the minimum pressure
drop).
4. Level gauge LG-x703.
6 vent valve PV-x204 is installed. It

operates in two different ways:
This valve regulates the liquid level on the

scrubber by means of an on/off control
(gap control). It opens when the liquid level
reaches the high set point of level gap
control (signal from LT-x606) and closes
when the low set point of level gap control
is reached.
3. Liquid level control LICA-x606.
5. And is protected from overpressure for

fire by safety valves PSV-x801A/B.
The liquid from the bottom of the scrubber
is sent to the closed drain system via the
control valve LV-x606.
1. During normal operation it opens when

the pressure reaches 32 bara to prevent
bad operation of the throttling valve.
The shutdown valve SDV-x605 is also

installed on the closed drain line in order
to prevent the gas flowing towards the
closed drain header in case of malfunction
of the liquid level control.
2. When the compression unit is in

pressurized shut down this valve is used to
assure the minimum difference pressure
between the unit and the seal gas external
source in order to guarantee a sufficient
gas flow to the sealing system of the
compressor.
Before reaching the compressor, the gas

flow is measured by the flow meter FTx107 and filtered by the T-type basket filter
200-FL-11x .
239
Chapter 4
compressors
During the normal operation all the fans

are running. Each fan is provided with
vibration switch VSHH-x901.
4.7.3.5 Compressor discharge

description
Reference is made to Process gas P&ID
SOK0886415 sh.3 and 4.
4.7.3.8 Fuel gas system description

Reference is made to Fuel gas P&ID
SOK0886416 sh.2 and 3.
1. Compressor 200-K-11x.
2. Compressor after-cooler 200-A-11x.
1. Pressure reduction station PCV-5212

A/B, PCV-5213.
Hot gas from compressor discharge

is sent to the forced draft air cooler
200-A-11x, and cooled down to 5 C
2. Scrubber 200-V-81.
3. Filters 200-FL-81 A/B.

description (valves)
4. Heaters 200-E-81 A/B.
The discharge pipe between the

compressor and the air cooler is
provided of:
5. Final filters 200-FL-91A/B, 200-FL 92A/B,

200-FL-93A/B, 200-FL-94A/B. (each train)
4.7.3.9 Fuel gas system

description(cond.skid)-pressure
reduction
1. high temperature protection TAHHx405.

2. high pressure protection PAHH-x211.
The same gas coming from the national

network is used to feed the four gas
turbines, 200-TB-11A/B/C/D, which drive
the centrifugal compressors.
3. the surge detector system PAH-x212 (to

detect the fast pressure variations).
On the same line an additional
instrumented high pressure protection
system is installed to protect the system:
The conditioning skid is common to the

turbo-compressor packages and is
designed considering the total flow due to
the contemporary operation of the four
turbines at full load.
It is activated by PT-x213A/B/C and it will

prevent the pressure increase in the
system over the maximum allowable value
by quickly cutting off the fuel gas to the gas
turbine.
1. The fuel gas is at first addressed to the

conditioning skid 200-X-08 through SDV5912 where it is reduced in pressure to
20.5 bara by PCV-5212A/B (one operating,
one in stand-by) controlled by PIC-5232.

description (cooler)
The gas-air cooler is a 3 bays / 6 bundles
exchanger provided with six 100%
autovariable pitch fans controlled by the
temperature control TIC-x409 and
protected in case of high temperature by
the high temperature trip TAHH-x408,
both installed on the line downstream of
the cooler.
2. PCV-5213 controlled by PIC-5231. This

latter is designed to manage 10% of the
total gas flow and it is used mainly during
start-up.
3. Full flow PSV-5811A/B (one operating,
one in stand-by) are installed downstream
240
Chapter 4
compressors
the pressure reduction station to protect

the system from overpressure due to
contemporary failure (set at 30
barg=pcvs5121A/B,5123 failure) .
cartridge) and very small liquid particles

(by means of an additional vane pack with
agglomerator).
Since the big amount of liquid (if present)
will be removed by the scrubber 200-V-81,
only level gauges LG-5710 & LG-5711 have
been foreseen on these filters. They have
to be periodically checked and manually
drained if full of liquid.
4.7.3.10 fuel gas system description

(cond.skid)-scrubber
After pressure reduction the gas is sent to
the scrubber 200-V-81 (vane pack type)
where the liquid particles are removed.
The switch from one filter to the other is

done manually by operating the three-way
valves mechanically linked for continuous
transfer.
The scrubber is provided with :
1. high level protections LAHH-5612

(closing of SDV-5912).
Each filter is protected from overpressure

for fire by double safety valves (one
operating, one in stand-by): PSV-5812A/B
and PSV-5813A/B.
2. low level protections LALL-5614 (closing

of SDV- 5614).
3. liquid level control LICA-5615.

(cond.skid) -heater
4. and level gauge LG-5714.
After the fine filtration, the gas is

addressed to the direct heaters 200-E81A/B. They are connected in seriesparallel and are normally operated
contemporary at 50% of total load so that
each one can be operated at 100% without
manual transfer in case of failure of the
other one. (max duty =600KW).
The liquid from the bottom of the scrubber

is sent to the closed drain system via the
control valve LV-5615.This valve regulates
the liquid level on the scrubber by means
of an on/off control.
The shutdown valve SDV-5614 is also
installed on the closed drain line in order
to prevent the gas flowing towards the
closed drain header in case of malfunction
of the liquid level control.
Heater outlet temperature is kept constant

at 27 C by means of TIC-5414 .
The common line downstream of the
heaters is provided with high temperature
alarm
TSH-5414
and very
high
temperature protection TSHH-5415, to
4.7.3.11 fuel gas system description

(cond.skid)-filter
Then the fuel gas enters the filters 200-FL81A/B (one operating, one in stand-by) to
remove solid particles (by means of filter
241
Chapter 4
compressors
Figure 4.83 Schematic of fuel gas system
protect the line from gas overheating.
Each skid is provided with :
In addition the resistance of each heater is

protected from overheating by means of
its own very high temperature protection
device (TSHH-5412 and TSHH-541 Each
heater vessel is protected from
overpressure for fire and abnormal heat
input by double safety valves (one
operating, one in stand-by): PSV-5814A/B
and PSV- 5815A/B.
1. inlet SDV-x915.

(filter skid)
94A/B), one operating and one in stand-by,

are installed.
Then the fuel gas is sent to dedicated filters

skids 200-X-09A/B/C/D, one for each gas
turbine.
The filters are foreseen for the final

removal of solid particles before entering
the gas turbine.
2. flow meter FT-x120 (Coriolis type) .

3. high and low pressure and temperature
alarms.
4. BDV-x916.
5. outlet SDV-x917(operated by a signal
from ESD unit) when p is HH in pipe
between compressor and cooler.
On each skid two filters (200-FL-91A/B,

200-FL-92A/B, 200-FL-93A/B, 200- FL-
242
Chapter 5 | Modeling Using ArcGIS
Chapter 5
Modeling Using ArcGIS
243
layers. For example, parcels can be shaded

with colors based on their zoning types, or
the size of point symbols for oil wells can
be specified based on production levels.
5.1 introduction:
We have two reasons for drawing the Gas
network model on ArcGIS, Due to the
abilities ArcGIS provides to draw accurate
map of Egyptian national grid of natural
gas we use this map as a model on
SynerGEE package to perform hydraulic
analysis and optimize its operation.
5.2.1 What is GIS?

A geographic information system (GIS) is a
computer-based tool for mapping and
analyzing geographic phenomenon that
exist, and events that occur, on Earth.
After
performing
analysis
and
optimization, generated data from
SynerGEE can be added to the attribute
table of each layer this could make us
supported with accurate map with
accurate coordinates of each feature and
hydraulic data of each point on the map to
be displayed.
GIS technology integrates common

database operations such as query and
statistical analysis with the unique
visualization and geographic analysis
benefits offered by maps. These abilities
distinguish GIS from other information
systems and make it valuable to a wide
range of public and private enterprises for
explaining events, predicting outcomes,
and planning strategies.
5.2 Geographic information

system:
Map making and geographic analysis are

not new, but a GIS performs these tasks
faster and with more sophistication than
do traditional manual methods.
Geography is the science of our world.

Coupled with GIS, geography is helping us
to better understand the earth and apply
geographic knowledge to a host of human
activities, Interactive maps provide the
main user interface for most GIS
applications.
Today, GIS is a multi-billion-dollar industry

employing hundreds of thousands of
people worldwide. GIS is taught in schools,
colleges, and universities throughout the
world.
Professionals
and
domain
specialists in every discipline are become
increasingly aware of the advantages of
using GIS technology for addressing their
unique spatial problems.
Users can point to locations and discover

new relationships, perform editing and
analysis, and effectively present results
using geographic views such as maps and
globes, GIS users pan and zoom interactive
maps, where map layers apply symbols
based on a set of attributes and perform
query and analysis operations through the
map layers, A GIS user can point to a
geographic object in an interactive map to
get information about the object.
We commonly think of a GIS as a single,

well-defined, integrated computer system.
However, this is not always the case. A GIS
can be made up of a variety of software
and hardware tools. The important factor
is the level of integration of these tools to
provide a smoothly operating, fully
functional geographic data processing
environment.
GIS users pan and zoom interactive maps,

where map layers apply symbols based on
a set of attributes and perform query and
analysis operations through the map
244
from centralized computer servers to

desktop computers used in stand-alone or
networked configurations.
5.2.2.2 Software:
GIS software provides the functions and
tools needed to store, analyze, and display
geographic information.
5.2.2.3 Data:
Figure5.1 Esri products.
Perhaps the most important component of

a GIS is the data. Geographic data and
related tabular data can be collected inhouse, compiled to custom specifications
and requirements, or occasionally
purchased from a commercial data
provider.
In general, a GIS provides facilities for data

capture,
data
management,
data
manipulation and analysis, and the
presentation of results in both graphic and
report form, with a particular emphasis
upon preserving and utilizing inherent
characteristics of spatial data.
A GIS can integrate spatial data with other

existing data resources, often stored in a
corporate DBMS. The integration of spatial
data (often proprietary to the GIS
software), and tabular data stored in a
DBMS is a key functionality afforded by
GIS.
The ability to incorporate spatial data,

manage it, analyze it, and answer spatial
questions is the distinctive characteristic of
geographic information systems.
5.2.2 COMPONENTS OF A GIS:

An operational GIS also has a series of
components that combine to make the
system work. These components are
critical to a successful GIS,
5.2.2.4 People:
GIS technology is of limited value without
the people who manage the system and
develop plans for applying it to real world
problems. GIS users range from technical
specialists who design and maintain the
system to those who use it to help them
perform their everyday work.
5.2.2.5 Procedures or methods:
A successful GIS operates according to a

well-designed implementation plan and
business rules, which are the models and
operating practices unique to each
organization.
Figure5.2 COMPONENTS OF A GIS.
5.2.2.1 Hardware:
Hardware is the computer system on
which a GIS operates. Today, GIS software
runs on a wide range of hardware types,
245
maintenance worker addressing a citizen

complaint, allowing the city to better
coordinate its utility response team. Plus,
with mobile GIS technology we can take
GIS mapping systems into the field to log
and send up-to-the-minute data in real
time.
5.2.3 Benefits from GIS:

we use geographic information systems
(GIS) mapping software to help clients with
a variety of infrastructure management
projects, from assessment district analyses
to topographic, facility and flood mapping.
We find that its a highly effective if not
essential tool for cities. GIS offers many
capabilities and benefits, but weve
outlined our top five below:
5.2.3.4 Keep better records:

At Moore, we find one of GISs most useful
capabilities is keeping track of records,
such as maintenance reports.
5.2.3.1 Save time and money:
For example, GIS reporting can help you

identify which streets repeatedly need
repair. In some cases, it may be more
efficient to completely replace a street
rather than investing in frequent
maintenance. It can also help you manage
construction permits to ensure you
maintain compliance with city, state and
federal regulations.
You can use GIS to improve your citys

maintenance schedules and daily fleet
movements. In fact, according to Esri, an
international GIS software supplier,
implementing GIS typically saves 10 to 30
percent in operational costs. Thats
because it helps decrease fuel use and staff
time, improves customer service, and
promotes more-efficient scheduling.
5.2.3.5 Predict the future:
5.2.3.2 Make better decisions:
GIS mapping helps us understand not only

what is happening, but also what will
happen, in a geographic space.
GIS mapping systems can help you make

more-informed decisions on anything
related to location.
For example, you can use GIS to map pipecorrosion patterns, allowing you to predict
risks and meet DIMP requirements. The
technology lets you view patterns and
trends to predict the future and, more
importantly, prepare for it.
For example, it can help with choosing real

estate sites, developing evacuation or
ambulance routes, estate planning or even
mapping wetlands. Every choice you make
has impacts, and GIS can help ensure
youre making the best possible decisions
for your city.
5.2.4 GIS application: (specific

functionality)
5.2.3.3 Improve communication:
Although most GIS software packages

share similar features, programs vary in
price and functionality. Some have a
number of built-in functions whereas
others
have
additional
modules/extensions that can be purchased
separately to perform specific types of
analysis.
Because GIS is digital reporting, it is highly

effective for cross-office and even crossorganizational
communication.
The
software helps everyone involved
understand situations more thoroughly.
For example, the technology can
communicate the location of a city
246
The basic functions GIS software should

offer can be grouped into data capture,
data management, data analysis and
representation:
5.2.4.3 Data analysis - ask questions:

A GIS is typically used to answer questions
related to space. For the purpose of finding
an answer, a GIS should not only offer tools
to query and retrieve data, but also
functions for data integration and
modelling.
5.2.4.1 Data capture - no data, no GIS:

Data capture functionality could either be
the importing of digital data from a disk,
network or database, or digitizing it
through the use of peripheral devices.
Data integration and conversion are only a

part of the input phase of GIS. What is
required next is the ability to interpret and
to analyze the collected information
quantitatively and qualitatively.
Data used in GIS often come from many

types, and are stored in different ways.
A GIS provides tools and a method for the
integration of different data into a format
to be compared and analyzed. Data
sources are mainly obtained from manual
digitization and scanning of aerial
photographs, paper maps, and existing
digital data sets.
For example, satellite image can assist an

agricultural scientist to project crop yield
per hectare for a particular region. For the
same region, the scientist also has the
rainfall data for the past six months
collected through weather station
observations.
Remote-sensing satellite imagery and GPS

are promising data input sources for GIS.
5.2.4.4 Visualization - make it visible:
5.2.4.2 Data management - quality is
Any analysis of data is only useful if the

final output is also understood by the
intended recipient.
essential:
One step beyond data capture is data
editing and maintenance. Editing involves
functions such as adding, duplicating and
deleting spatial and attribute data as well
as creation and management of metadata.
Using tools to detect errors and to perform
transformations is another aspect of data
management.
A GIS offers tools to present data in

various ways, ranging from displaying
tabular information, over graph/chart
production, to printing maps. Some GIS
offer advanced visualization beyond two
dimensions, such as digital elevation
models (DEM) or digital movie files for "fly
throughs". Output can also be shared by
storing it on a disk or a webserver, which
allows for dynamic views of the data.
After data are collected and integrated, the

GIS must provide facilities, which can store
and maintain data. Effective data
management has many definitions but
should include all of the following aspects:
data security, data integrity, and data
storage and retrieval, and data
maintenance abilities.
One of the most exciting aspects of GIS

technology is the variety of different ways
in which the information can be presented
once it has been processed by GIS.
Traditional methods of tabulating and
graphing data can be supplemented by
maps and three dimensional images.
247
5.3.2 Types of coordinate systems:
5.3 Coordinate systems:
Coordinate Systems are ways of splitting

up the world in order to form transferable
units (numbers) that relate to points on a
map. They are very useful, as a set of
coordinate values can be given to almost
anyone in the world and they can relate
the coordinate values to a real life place.
Different coordinate systems have
different methods of splitting up the
earth.
5.3.1 What is a coordinate system?

Coordinate systems enable geographic
datasets to use common locations for
integration.
A coordinate system is a reference system
used to represent the locations of
geographic features, imagery, and
observations, such as Global Positioning
System (GPS) locations, within a common
geographic framework.
A coordinate system enables every

location on the earth to be specified by a
set of coordinates of known location on a
grid.
Each coordinate system is defined by the

following:
Its measurement framework,

which is either geographic (in which
spherical
coordinates
are
measured from the earth's center)
or plan metric (in which the earth's
coordinates are projected onto a
two-dimensional planar surface).
Units of measurement (typically
feet or meters for projected
coordinate systems or decimal
degrees for latitude-longitude)
The definition of the map
projection for projected coordinate
systems.
Other
measurement
system
properties such as a spheroid of
reference, a datum, one or more
standard parallels, a central
meridian, and possible shifts in the
x- and y-directions.
Several hundred geographic coordinate
systems and a few thousand projected
coordinate systems are available for use. In
addition, you can define a custom
coordinate system.
There are two major global coordinate

systems, known as the Geographic
Coordinate System (i.e. latitude and
longitude) and the Universal Transverse
Mercator System (i.e. UTM coordinates).
The following are two common types of
coordinate systems used in a geographic
information system (GIS):
5.3.2.1 A global or spherical coordinate
system:
A geographic coordinate system (GCS) uses
a three-dimensional spherical surface to
define locations on the earth, a point is
referenced by its longitude and latitude
values. Longitude and latitude are angles
measured from the earth's center to a
point on the earth's surface, the angles
often are measured in degrees (or in
grads). In the spherical system, horizontal
lines, or eastwest lines, are lines of equal
latitude, or parallels.
248
The points are all in very different

locations but share either the same
longitude or latitude with one of the other
points.
The green and blue points are on same
latitude, +30 (horizontal line, parallel to
the equator). The red and blue points are
on the same longitude, +30 (vertical line,
parallel to the prime meridian).
As you can see, latitude and longitude can
have the same numerical value (+30 in this
case) but as these values in fact indicate
very different points on the landscape
Figure5.3 A global or spherical coordinate system.
The most well-known coordinate system is

the Geographic Coordinate System (GCS),
which uses measures of latitude and
longitude to determine your location.
Geographic coordinates (latitude and

longitude) are perhaps the best known
method for describing a horizontal position
on the surface of the Earth.
Lines of latitude run north and south

parallel to the equator, with positive
numbers representing the northern
hemisphere,
negative
numbers
representing the southern hemisphere,
and the equator representing 0 degrees
latitude.
Latitude and longitude are expressed in

angular units, typically degrees, minutes
and seconds (e.g., 90 45 15) or decimal
degrees (e.g., 90.75416667).
5.3.2.2 Projected coordinate systems:
A projected coordinate system is defined
on a flat, two-dimensional surface. Unlike
a geographic coordinate system, a
projected coordinate system has constant
lengths, angles, and areas across the two
dimensions.
Lines of longitude run east and west

parallel to the prime meridian, with
positive numbers representing the eastern
hemisphere and negative numbers
representing the western hemisphere,
with the prime meridian representing 0
degrees longitude. This may be
understood by looking at this grid in
relation to a map of the world (see figure
25).
A projected coordinate system is always

based on a geographic coordinate system
that is based on a sphere or spheroid.
In a projected coordinate system, locations
are identified by x,y coordinates on a grid,
with the origin at the center of the grid.
In this image you can see the globe is

divided into grid squares using lines of
longitude and latitude. It is important that
you understand the difference in longitude
and latitude and how we write locations in
longitude and latitude form.
Each position has two values that

reference it to that central location. One
specifies its horizontal position and the
other its vertical position.
There are three points on the grid map

marked in red, blue and green.
249
The two values are called the x coordinate

and y-coordinate. Using this notation, the
coordinates at the origin are x = 0 and y =
0.
most features of the earth at a

usable scale would be too large to be
useful, so we use maps.
Also imagine peeling an orange and
pressing the orange peel flat on a table the peel would crack and break as it was
flattened because it can't easily transform
from a sphere to a plane.
The same is true for the surface of the
earth and that's why we use map
projections.
Geographic coordinates are converted
from angular units (degree) to distance
units on a map (feet or meters).
Figure5.4 Projected coordinate systems.
Measurements,
computations
and
computer applications are more difficult to
manage using angular units, thus, latitude
and longitude values are commonly
converted to a rectangular coordinate
system of northings (Y-axis) and
eastings (X-axis) that can be expressed
and easily understood in linear units such
as meters or feet.
Figure5.6 Map projection.
This process inevitably distorts at least one

of the following properties: shape, area,
distance, or direction.
For small areas, such as a city or county,
the distortion will probably not be great
enough to affect your map or
measurements.
If youre working at the national,
continental, or global level, so youll want
to choose a map projection that minimizes
the distortion based on the requirements
of your specific project.
Figure5.5 Projected coordinate systems.
5.3.3 Map projection:

It is impossible to accurately represent the
spherical surface of the earth on a flat
piece of paper.
While a globe can represent the planet
accurately, a globe large enough to display
250
secant, in which case it cuts through the

earth at two standard meridians.
The Mercator and Transverse Mercator
projections are conformal projections.
In a conformal projection local angles are
preserved and shapes are represented
accurately and without distortion for small
areas.
As a result of preserving angles and
shapes, area or size of features are
distorted in these maps.
Figure5.7 A very famous projection is the Mercator

Map.
As mentioned, choosing a secant

projection and a narrow zone minimizes
the distortions in a map generated from
UTM projection.
Geradus Mercator invented his famous

projection in 1569 as an aid to navigators.
On his map, lines of latitude and longitude
intersect at right angles and thus the
direction of travel - the rhumb line - is
consistent.
The distortion of the Mercator Map
increases as you move north and south
from the equator.
On Mercator's map Antarctica appears to
be a huge continent that wraps around the
earth and Greenland appears to be just as
large as South America although
Greenland is merely one-eighth the size of
South America.
Figure5.8
Projection.
Universal Transverse Mercator (UTM)
5.3.3.1.1 UTM Zone:

UTM zones are numbered from 1 to 60
starting from 180 longitude at the
International Date Line and proceeding
eastward. Each zone is 6 longitude wide.
Therefore Zone 1 lies between 180W to
174W longitude lines, Zone 2 is between
174W and 168W longitude lines, and
Zone 60 lies between 174E and 180
(International Date Line) longitude lines.
Mercator never intended his map to be

used for purposes other than navigation
although it became one of the most
popular world map projections.
5.3.3.1 Universal Transverse Mercator
(UTM) Projection:
The Universal
Transverse
Mercator projection is based on the
cylindrical Transverse
Mercator projection. The cylinder in the
Transverse Mercator projection is tangent
along a meridian (line of longitude) or it is
Each zone is divided into horizontal bands

8 of latitude wide. The 20 bands are
labeled with letters, beginning with C and
ending with X from south to north.
Letters I and O are omitted to prevent
confusion with numbers one and zero.
251
Note that letter N designates the first

latitude band above the equator, so any
letter of alphabet after N refers to a band
in northern hemisphere and any letter
before N refers to a band in southern
hemisphere.
The central meridian of the zone lies along

15W meridian or line of longitude.
Since UTM zones are bounded by lines of
longitude and these lines converge
towards the poles, consequently the width
of each zone tapers (narrows) from the
equator towards the poles.
Letters A, B, Y and Z label polar regions in

the UPS coordinate system. To specify a
UTM grid zone, the zone number (column)
is given before the zone letter (row), such
as Zone 11U.
Zone divisions with horizontal bands, as

above, result in each division having
dimensions of 6 longitude wide and 8
latitude high.
However, there are exemptions, where
dimensions of some zones are adjusted.
The northernmost band (X) is 12 latitude
high instead of 8 and extends from 72N
to 84N latitudes.
Zone 32V covering west coast of Norway is
widened to 9 longitude instead of 6,
resulting in narrowing of Zone 31V.
Zones 33X and 35X have been widened to
12 to cover Svalbard.
Similarly zones 31X and 37X are widened to
9.
As a result zones 32X, 34X and 36X have
been omitted.
5.4 Project Design:

In any kind of research project, one of the
first steps (that is often overlooked) is to
plan the various steps that will be involved.
Taking the time to prepare a detailed work
plan will most likely save you time overall.
Figure5.9 UTM Zone.
Each zone has a central meridian which

extends north-south along the middle of
the zone (3 from the east and west zone
boundaries), splitting the zone into two
equal halves. For example, as shown in the
figure, zone 28 (northern hemisphere
shown) is bounded by 18W meridian to
the left and 12W meridian to the right.
We present here five steps that you can

follow to be better prepared to face your
own research challenges.
5.4.1 Steps of Project Design:

A GIS project, like any research project,
requires good preparation, and the time
252
spent preparing might save precious hours

along the way.
One way to work on your methodology is

to create a flowchart (or a model) of the
various steps required to answer your
question.
The realization of a GIS project can be

conceptualized in 5 main steps: definition
of the research question, the design of the
methodology, the data gathering, the
analysis, the presentation of the results.
You can think of this as the logical

flowchart of the operations required to
solve the problem.
You will need to identify the type of
operations you will need to perform
(spatial statistics, creation of new
shapefile, raster analysis, etc.) and the
expertise you will need (do you know how
to work in specific software? do you know
how to work with different projections,
which particular tool you will need?).
5.4.1.1 Define your research question:

This will be the basis of all the subsequent
steps of your project. This question should
lead you to define goals and objectives
specific to your own project.
The more interesting you find the
question, the more exciting you will find
your project.
At this moment of your project, you should

be able to identify what data you will need
in your project.
Inspiration for research questions can

come from very different, and sometimes
unusual sources.
5.4.1.3 Find the right data:
Take the time to browse on-line, find

scientific articles related to a topic you like,
explore
datasets
that
you
find
intriguing. Sometimes inspiration for your
research questions can come from your
day to day life, sometimes it can comes
from a question raised in a class.
Finding data is a crucial part of your

project. Depending on the data available
you may or not have to modify your
research objectives or your methodology.
When looking for data, starts by identifying
the 3Ws:
Take the time to seek out different options

if you are having doubts about your
interest or the feasibility of the project.
WHAT? What are the different

layers you will need: road network,
urban
plans,
deforestation,
watersheds? Are you looking
specifically for vector or raster
data?
WHERE? What spatial extent are
you looking for? A city, a country,
a continent?
WHEN? Do you want to do an
analysis about past, current, or
future conditions.
5.4.1.4 Perform the analysis:
If you think your research question and

objectives are clear and concise, and that
they fit under the umbrella of your class
project or the timeline you need to meet,
you are ready for step 2.
5.4.1.2 Design an appropriate
methodology:
Once you are settled on one research
question and the subsequent objectives,
the design of the methodology is your next
step.
Now that you have the objectives ready,

the methodology defined (and refined
253
based on data), and data in your

possession, it is time to use you skills as a
GIS analyst and apply your methodology.
When facing problems, there are many
resources available online via blogs,
university websites, and software help
documents.
Remember that this process can be filled
with hick-ups and that you might have to
start over (many times!).
Figure5.10 ArcMap visualization.
Remember too, that it is not the number

of mistakes that matters, but what you
accomplish at the end.
5.5.2 ArcGlobe:
ArcGlobe is used for seamless 3D
visualization of geographic data using a
continuous global view.
5.4.1.5 Present your results:

At the end of your project, you will need to
present your final results. Depending on
the nature of the research project,
different options will be more effective:
final
report,
poster,
PowerPoint
presentation.
This application is generally designed to be

used with very large datasets that are
displayed using varying levels of detail.
ArcGlobe is part of the optional ArcGIS 3D
Analyst extension.
Any GIS project will most likely present a

map of the final (or intermediate) results.
An effective map should be able to be read
independently
from
the
wider
presentation of the project.
Have a look to our Cartography section for
more tips and resources to create a map.
5.5 Mapping:
Figure5.11 ArcGlobe visualization.
ArcGIS for Desktop has three applications

that can be used for mapping and
visualization:
5.5.3 ArcScene:
ArcScene is used for 3D visualization of
focused scenes or areas.
5.5.1 ArcMap:
It creates a 3D scene view that you can

navigate and interact with for enclosed
areas of interest.
ArcMap is the main application used in

ArcGIS for Desktop for mapping, editing,
analysis, and data management. ArcMap is
used for all 2D mapping work and
visualization.
ArcScene is also part of the ArcGIS 3D

Analyst extension.
254
ArcMap
represents
geographic
information as a collection of layers and
other elements in a map.
Common map elements include the data
frame containing map layers for a given
extent plus a scale bar, north arrow, title,
descriptive text, a symbol legend, and so
on.
Figure5.12 ArcScene visualization.
5.6.1 Typical tasks performed in

ArcMap:
5.5.4 ArcCatalog:
ArcMap is the primary application used in

ArcGIS and is used to perform a wide range
of common GIS tasks as well as specialized,
user-specific tasks. Here is a list of some
common workflows you can perform:
ArcCatalog is used for managing your

spatial data holdings, and database
designs, and for recording and viewing
metadata.
Work with mapsyou can open and use

ArcMap
documents
to
explore
information, navigate around your map
documents, turn layers on and off, query
features to access the rich attribute data
that is behind the map, and to visualize
geographic information.
Print mapsyou can print maps, from the
simplest to very sophisticated cartography,
using ArcMap.
Compile and edit GIS datasetsArcMap
provides one of the primary ways that
users automate geodatabase datasets.
Figure5.13 ArcCatalog visualization.
ArcMap supports scalable full-function

editing. You select layers in the map
document to edit and the new and
updated features are saved in the layer's
dataset.
5.6 What is ArcMap?

ArcMap is the central application used in
ArcGIS.
ArcMap is where you display and explore
GIS datasets for your study area, where
you assign symbols, and where you create
map layouts for printing or publication.
Use geoprocessing to automate work and

perform analysisGIS is both visual and
analytical. ArcMap has the ability to
execute any geoprocessing model or script
as well as to view and work with the results
through map visualization. Geoprocessing
can be used for analysis as well as to
ArcMap is also the application you use to

create and edit datasets.
255
automate many mundane tasks such as

map book generation, repairing broken
data links in a collection of map
documents, and to perform GIS data
processing.
the ability to write software add-ins to add

new functionality, to simplify and
streamline the user interface, and to use
geoprocessing for task automation.
Organize and manage your geodatabases

and ArcGIS documentsArcMap includes
the Catalog window that enables you to
organize all of your GIS datasets and
geodatabases, your map documents and
other ArcGIS files, your geoprocessing
tools, and many other GIS information
sets. You can also set up and manage
geodatabase schemas in the Catalog
window.
When you save a map you have created in

ArcMap, it will be saved as a file on disk.
This is an ArcMap document and is
referred to as a map document or mxd
since the file name extension (.mxd) is
automatically appended to your map
document name. You can work with an
existing .mxd file by double-clicking the
document to open it. This will start an
ArcMap session for that .mxd file.
Publish map documents as map services

using ArcGIS for ServerArcGIS content is
brought to life on the web by publishing
geographic information as a series of map
services. ArcMap provides a simple user
experience for publishing your map
documents as map services.
Map documents contain display properties

of the geographic information that you
work with in the mapsuch as the
properties and definitions of your map
layers, data frames, and the map layout for
printingplus any optional customizations
and macros that you add to your map.
Share maps, layers, geoprocessing

models, and geodatabases with other
usersArcMap includes tools that make it
easy to package and share GIS datasets
with other users. This includes the ability
to share your GIS maps and data using
ArcGIS Online.
5.6.3 Views in ArcMap:
5.6.2 ArcMap documents:
ArcMap displays map contents in one of

two views:
Data view
Layout view
Each view lets you look at and interact with
the map in a specific way.
Document your geographic information

A key goal in GIS communities is to
describe your geographic information sets
to help you document your projects and
for more effective search and data sharing.
In ArcMap data view, the map is the data

frame. The active data frame is presented
as a geographic window in which map
layers are displayed and used. Within a
data frame, you work with GIS information
presented through map layers using
geographic (real-world) coordinates.
Using the Catalog window, you can

document all of your GIS contents. For
organizations who use standards-based
metadata, you can also document your
datasets using the ArcGIS metadata editor.
These will typically be ground

measurements in units such as feet,
meters, or measures of latitude-longitude
(such as decimal degrees).
Customize the user experienceArcMap

includes tools for customization, including
256
The data view hides all the map elements

on the layout, such as titles, north arrows,
and scale bars, and lets you focus on the
data in a single data frame, for instance,
editing or analysis.
for example, as a collection of

contour lines and elevation points
or as shaded relief.
When you're preparing your map's layout,

you'll want to work with your map in page
layout view.
In addition to representing geographic

information, each layer's map symbols,
colors, and labels help describe the objects
in the map.
A page layout is a collection of map

elements (such as a data frame, map title,
scale bar, north arrow, and a symbol
legend) arranged on a page.
You can interact with the layers displayed

in each data frame to query each feature
and see its attributes, perform analytical
operations, and to edit and add new
features to each dataset.
Layouts are used for composing maps for

printing or export to formats such as
Adobe PDF.
The Layout view is used to design and
author a map for printing, exporting, or
publishing. You can manage map elements
within the page space (typically, in inches
or
A layer doesn't store the actual geographic

data.
Instead, it references a dataset, such as a
feature class, an image, a grid, and so on.
Centimeters), add new map elements, and

preview what your map will look like
before exporting or printing it.
Referencing data in this way allows the

layers on a map to automatically reflect the
most up-to-date information in your GIS
database.
Common map elements include data

frames with map layers, scale bars, north
arrows, symbol legends, map titles, text,
and other graphic elements.
5.6.4.1 Layers in the table of contents:

The table of contents lets you specify
which layers are displayed on the map (by
turning them on and off).
5.6.4 Map layers:

Within the data frame, you display
geographic datasets as layers, where each
layer represents a particular dataset
overlaid in the map. Map layers help
convey information through:
Aerial photography or satellite

imagery that covers the map
extent.
In addition, the order of the layer list in the

table of contents determines the drawing
order of layers on the maplayers higher
in the table of contents are drawn on top
of those that are lower.
Discrete feature classes such as

collections of points, lines, and
polygons.
Continuous surfaces, such as
elevation,
which
can
be
represented in a number of ways
To change the drawing order, click the

table of contents List By Drawing Order
button , click and hold a layer name,
then drag it up or down in the table of
contents to a new position.
257
5.6.5 Working with basemap as a

layer:
Designing your desktop maps for highquality display while maintaining good
performance is always a key goal.
One of the ways that you can improve
display performance is by using basemap
layers.
Basemap layers are a kind of map layer
that provides a framework onto which you
display your dynamic operational
information.
Basemap
display
performance is very fast.
Figure5.14 Table of contents.
5.6.4.2 Working with layer attributes:

You can work with attribute tables for the
datasets referenced by each map layer.
Right-click a layer in the table of contents
and click Open Attribute Table.
Since basemap layers are relatively static

and do not change frequently, their display
can be computed once and reused many
times. The first time you visit an area at a
particular map scale, basemap layer
display is computed.
In the Table window, you can perform

queries, make selections, and locate
features on the map, and so on.
The display is recalled on return visits to

that area and map scale.
5.7 Natural Gas Network

Using GIS:
5.7.1 Steps of the project:
We deal with natural gas network as GIS
project so the general steps for GIS
projects will exist in the project of natural
gas network:
5.7.1.1 Identify the objective:
Figure5.15 Attribute table.
The Objective of this part of project is to

draw a full detailed map of natural gas
natural grid provides data about
coordinates, flow rate, and pressure of
each point on the network (Supply, Valves,
Demand, Utilities, process and distribution
points) .
Click the Table window's Table Options

menu to create graphs and reports,
change the font for the table, print the
table, and perform various other
operations.
When you select an item in a table or
graph, the feature is also selected on the
map (and vice versa).
258
5.7.1.2 Prepare the Data:
5.8 Drawing the Natural

Gas network using ArcMap:
After knowing the objective we can

determine the layers can be created to
reach our target of the clear,
understandable and full detailed map of
NG network.
5.8.1 Create new MXD and set data frame

coordinate system:
Create new MXD (Map executable
document), and chose the default
database.
We start to collect data from different

sources depending on the specification of
required data, but we meet a lot of
obstacles due to the political instability in
Egypt, so we exhibits another ways to
gather data about the network sometimes
by searching in the internet and collecting
data in small scale from distribution
companies.
We collected a not so dusty data and start
analysis to localize this data on a base map.
5.7.1.3 Analyze the data:

Figure5.16 Create MXD.
Data analysis is the most important step to

check the validity of the gathered data and
to make the best use of this data to reach
our scoop.
5.8.2 Set data frame coordinate

system:
Classify the data of suppliers in (western

desert, Mediterranean see and Suzie bay),
Demands (PowerStation which represent
77 % of NG consumption, Cement plants,
steel industries..), process units and
distribution points.
5.7.1.4 Present the results:
Figure5.17 Set data frame coordinate system.
The results represents different scenarios

of views show the map and related
attributes tables of each point in each layer
(Supply, Utilities and process..).
Create a feature class for each layer in the

project and set its coordinate as the
coordinate of data frame.
5.8.3 Create file geodatabase:
Creating file geodatabase which contains
feature classes representing the whole
network.
259
Figure5.18 Creating file geodatabase.
Figure5.20 Write the name and select the type of each

layer
5.8.4 Create Feature classes:

Create Feature classes like points for
supply, demand, Utilities and distribution
points, thermal power stations, gas and
combined power stations and lines for
natural gas pipelines.
Chose co-ordinate system for each feature

class
Figure5.21 Chose co-ordinate system for each feature

class
Put fields name and chose the type of

each field.
We select two types suitable for data we
had:
Figure5.19 Create feature class for all layers
Text: a series of alphanumeric symbols.
Write the name and select the type
Float: Numeric values with fractional

values within specific range.
260
Figure5.22 Put fields name and chose the type of each

field.
After creation of feature classes and MXD

add them to the map as layers. We have
seven layers:
Figure5.23 Project layers.
As well as the layer of base map:
Figure5.24 Natural gas national grid.
261
5.9 Interaction between

ArcGIS and SynerGEE:
5.9.1 Exporting map to SynerGEE:
As mentioned before the map drawn on
ArcGIS needs to be exported to SynerGEE
which convert this map into hydraulic
model (pipes & Nodes) and perform
hydraulic analysis for this model.
We export it as shapefile or as map layer.
Figure5.26 Export as Shapefile.
Figure5.27 Save Data as layer

Figure5.25 Export Data from ArcMap.
Only pipes layer are exported as SynerGEE

can only convert pipes into model and
nodes come from each pipe ends.
5.9.2 Importing hydraulic data:

After SynerGEE finish analyzing the
network a table of each point generated
contains pressure and flow rate data.
262
We add this table to each layers attribute

table.
The importance of this step lies in the ease
of changing attribute tables data on ArcGIS
map as SynerGEE output change. Using
join tables method, we achieved this goal
perfectly.
This method requires to tables at least, all
have a common column.
5.9.2.1 Open attribute table for Demand
layer:
Figure5.29 Add new field.
Figure5.30 Write the name of new field.
This steps are repeated with the rest

layers.
Figure5.28 Open attribute table for Demand layer.
5.9.2.2 Add new field:

With name of SynerName and fill this
fields in the edit session with the name of
each demand node in SynerGEE model as
shown:
5.9.2.3 Join the output excel sheet with

the attribute table:
Finally join the output excel sheet with the
attribute table of each layer, so any change
in this file will be automatically change in
ArcGIS map attribute tables.
263
Figure5.31 Join the output excel sheet with the

attribute table.
264
Chapter 6
Modeling Using Synergi Gas
Chapter 6
Modeling Using Synergi
265
Chapter 6
system analysis. Condensate loss

generated by heat transfer to the
surroundings can be incorporated into
the solution.
6.1 introduction.
6.1.1 about synergi
Synergi Gas is the leading natural gas
network modeling software and it is the
commercial package that is used by GASCO
& NATA for modeling the national grid of
natural gas.
Heating, venting, and air conditioning

(HVAC) systems. Examine ventilating
systems for power plants, mines, and
buildings for sizing of ducts, vents, and
fans, and for air quality analysis.
Synergi Gas models and analyzes closed

conduit networks of pipes, regulators,
valves, compressors, storage fields, and
production wells. Synergi can be used as a
general purpose modeling tool for piping
networks that carry substances including,
but not limited to, natural gas, propane,
steam, oxygen, carbon dioxide, nitrogen,
chlorine, and air. In addition to the main
installation, modules and utilities are
available to extend Synergis functionality.
6.1.3 Solution techniques

The model of piping system is constructed
from non-linear mathematical equations
based on the provided network
information. These equations represent
network interconnection based on
Kirchhoffs first law, which states that the
flow into or out of a node in a network
must sum to zero in order for mass to be
conserved.
6.1.2 Synergi capabilities
The equation solutions provide predictions

of pressures, flows, valve positions, pipe
diameters, compressor powers and
speeds, and storage field utilization
factors.
Model Natural gas distribution systems

serving a small town or an entire
metropolitan area.
Model Natural gas transmission
systems. Analyze steady-state pressure
flow relationships of long pipeline
systems.
A model solution is achieved when the

corrections to each of the unknowns is less
than a predefined tolerance value.
The model can be:
Can easily handle systems of 1,000,000

nodes or more
1. Solved
Steady-state hydraulic analysis with

extensive
composition
and
temperature tracing functionality
A solved model contains the solution for all

unknown variables to within their
prescribed tolerances. This does not mean
that all flow residuals are near or equal to
0.0. Rather, it is an indication that the
corrections to the unknown variables have
become very small at least as small as the
tolerances.
Model Natural gas gathering or

production systems. Model a system
with integrated gas wells, storage
fields, compressors, regulators, and
piping.
Steam systems. Solve saturated steam
flow conditions for district heating
266
Chapter 6
2. Unsolved
to be solved before Synergi can

produce the answers that we want.
A model is unsolved if the maximum

number of iterations has been reached
and the corrections to the unknown
variables are larger than the
predefined
tolerances.
If
the
corrections applied on the last iteration
are reasonably small, then the solution
probably is converging. However, all
corrections have not yet reached the
prescribed tolerances.
6.1.4.2 What are the rules for

unknowns?
All networks that can be built and
operated under known constraints can
be modeled. To assist you in setting up
a modeling problem, keep the
following rules in mind:
1. In an NN node system, there must be
NN unknowns. In other words, for
every node, at least one of the
following should be an unknown:
3. Infeasible
If an unknown variable has reached a
limiting
value
and
constraint
interchange does not apply, an
infeasible solution is reached. An
infeasible solution is not a broken
simulation or failure; often, an
infeasible solution contains results that
are meaningful. Possible infeasible
solutions are pressures below absolute
zero, or regulators closing due to sonic
conditions.
Node flow.
Node pressure.
Node pressure of an adjacent
connected node or a facility
parameter of a facility connected to
this node as a control node
2. There must be at least one unknown

node flow in the model.
3. There must be at least one known
node pressure in the model.
6.1.4 Rules for unknowns
4. For every unknown node flow and

unknown facility, there must be a
unique fixed pressure. This fixed
pressure may be anywhere in the
model, provided that it is possible to
trace a hydraulically connected path
through known facilities from the
unknown node flow or facility to the
fixed pressure node without going
through an intervening fixed pressure
node.
Synergi is based on the premise that

there must be one unknown for each
independent equation in order for the
equations to be solved for a unique
solution. The unknowns can be any
combination of node solvable.
6.1.4.1 Why does Synergi need rules

for unknowns?
It is easy to forget that underneath
Synergis crisp, easy-to-use interface is
a very complicated hydraulic analysis
program that can solve millions of
complex equations. Using the data that
we input at the nodes and facilities in
the model, Synergi builds a complex set
of simultaneous equations that needs
5. Any unknown facility parameter

must be on a path between two fixed
pressures, and any serial grouping of
unknown facility parameters must
have an intervening fixed pressure.
267
Chapter 6
The rules listed above apply for all

networks except for those that contain
processing plants, loss elements and
forced flow elements, which have
special rules.
2. Model builder module

3. Time varying module
4. Unsteady state module
6.1.6.1 Optimization module
6.1.4.3 How does Synergi apply the

rules for unknowns?
6.1.6.1.1 Introduction
Optimizer simulates elements of a gas
pipeline using Synergis modeling
techniques.
The following points listed below

provide simplified information:
The arrows at nodes that point

away from the network are
demands. The arrows pointing
towards the network are supplies.
The status of a node pressure is
unknown unless a P appears at the
node.
The node flow status is known
unless QN=? Appears.
There is zero flow if no arrow
appears.
The status of the facility is known
unless a question mark is next to
the line representing the facility.
Optimization algorithms are then used to

select the best operating mode for the
pipeline, depending on the objective
requested and the information provided.
This service is a valuable tool for the
marketing,
engineering,
operations,
supply, and transportation departments of
a pipeline company. The information you
gain from using Optimizer helps you run a
gas
transmission
operation
more
efficiently and competitively.
Synergis optimization module can be used
for two primary purposes. In Synergi 3.x,
these tasks were available in different
modules: the Energy Minimization Module
(EMM) and the Economic Transport
Module (ETM). ynergi 4 combines these
two prior modules into a single module
known as the Optimization module.
6.1.5 Target of using Synergi

Gas
In order to achieve the projects target,
which is the optimization of the
operation of the National Gas Grid,
Synergi Gas is used to model the
network, and as a consequence
checking the feasibility of the number
of operation scenarios, which is the
output of our optimization code to find
the best operation scenario for the
network.
1. Energy minimization. Synergi can

automatically adjust compressor and
regulator set points to determine the
optimum compressor and regulator
2. operating plan to minimize systemwide volumetric fuel requirements or
total fuel cost
3. Economic transport. In addition to the
energy
minimization
capabilities
described above, it also minimizes fuel
during the optimization as a secondary
goal.
6.1.6 Synergi modules

Synergi software is provided with four
modules:
1. Optimization module
268
Chapter 6
internal branch may, however,

contain regulators, provided you
have checked the Enable Prune End
Regulators check box (in the
Optimization area of the Model
Setting dialog
6.1.6.1.2 Optimization terminology
Operating point: An operating

point is a device that is optimized,
such as a compressor or regulator.
An unknown flow node: is a node

at which an analysis reconciles any
differences between system inflow
(total specified supply) and system
outflow (total specified deliveries,
overall fuel usage by compressors,
and processing plant shrinkage).
Nodes are set to an unknown flow
status by unchecking the Flow
check box in the node editor
Segments: are similar to a single

length of pipe if the pressure at
some point is raised or lowered, the
pressures at other points in the
segment rise or fall accordingly. In
a valid Optimizer model, every
segment has at least one
Figure 6- 2
Figure 6- 1
Operating loops: If the defined

area contains a flow path that can
be traced from some originating
node through two or more
operating
points
(such
as
compressors or regulators) and
back to its node of origin, that flow
path is considered an operating
loop.
Operating branch: A branch is said

to exist if a lateral from the pipe
system splits off without rejoining.
For a branch to be considered as
operational, it must end in either a
supply or delivery node and must
contain one or more optimizable
compressor and/or regulator
elements.
Internal loops: If the network

contains a flow path that can be
traced from some originating node,
through other nodes, and back to
its node of origin without passing
through any operating points, that
flow path is considered an internal
loop.
Such a loop is entirely contained
within a segment.
Internal branches: A lateral is

considered an internal branch if it
contains no compressor(s) and no
regulators having set pressures. An
269
Region nodes: A region node has one

or more contracts (specifying flows and
costs) or a variable limit attached to the
node. A region node can either be fixed
(user-specified flow) or variable where
Optimizer determines the flow rate.
Synergi optimizes the contracts or
Chapter 6
Maximum operating margin analysis
Maximum throughput analysis
6.1.6.1.5 Fuel minimization analysis

During a fuel minimization analysis,
Synergi automatically adjusts compressor
and regulator set points to determine the
operating plan to minimize system-wide
volumetric fuel requirements. Before you
perform a fuel minimization analysis, you
should make sure that you have properly
set up your model for optimization
Figure 6- 3
flows connected to the region node

depending on the objective.
6.1.6.1.3 Setting up a valid unknown
structure
Both Synergi and Optimizer use the same
set of hydraulic equations to solve steadystate gas network problems, and both are
governed by the same rules for unknowns.
1. In a system having N nodes, there

must be N unknowns.
6.1.6.1.6 Solution technique for fuel
2. A system must have at least one set

pressure.
minimization
During
a
steady-state
analysis,
assumptions must be made about the
systems operating parameters. Usually,
either the suction or discharge pressure of
a compressor is set. This pressure is
maintained for as long as the compressor
engine has the horsepower to do so When
you run an optimization analysis, the
following takes place:
3. A system must have at least one

unknown node flow.
4. Every unknown node flow or
unknown facility must have a fixed
pressure.
5. Every unknown element must lie
on a path between two known
pressure nodes.
6.1.6.1.4 SETTING UP AND PERFORMING

AN OPTIMIZATION ANALYSIS
The required steps to set up an

optimization analysis are based on the
optimization objective that you have
selected in the Model Settings dialog box:
Fuel, Cost, Margin, or Throughput
Fuel minimization analysis
Cost of fuel analysis
solution
technique
for
fuel
minimization
Performing a fuel minimization analysis
Closed valves are stripped.

The system is balanced, if possible. (It
is extremely unlikely that an
optimization can be obtained for a
model that does not balance.)
The balanced system is optimized.
6.1.6.1.7 Performing a fuel minimization

analysis
TO PERFORM A FUEL MINIMIZATION
ANALYSIS
270
From the main menu, select Analysis >

Optimize. During the analysis, a
Chapter 6
progress bar will display the result of

each iteration during the analysis. A
red diamond in the progress bar
indicates where in the analysis that the
optimum result has been achieved.
When the analysis is complete, the
solution summary displays, indicating
the initial objective, the estimated final
objective, the number of completed
iterations,
and
whether
the
optimization was successful
6.1.6.1.8 Cost of fuel analysis

During a cost of fuel analysis, Synergi
automatically adjusts compressor and
regulator set points to determine the
operating plan to minimize system-wide
total fuel cost. Your model has been
properly set up, refer to the topics listed
below for more information on performing
a cost of fuel analysis
Setting up cost versus power equations

Setting up compressor stations
Setting up optimization pricing
Performing a cost of fuel analysis
Setting up cost versus power equations

Equations are used to define functions of
cost versus power. All functions use a
piecewise linear equation. Cost functions
for individual units in the compressor
stations reference these equations.
Figure 6- 4
create an equation
edit an equation
rename an equation
delete an equation
Figure 6- 5
Figure 6- 6
271
Chapter 6
General Operational Summary. This

section displays for all optimization
objectives and displays the following
information:
- System Status
- Objective
- Objective Value
- Total Linepack
- Total Power
- Total Fuel
Figure 6- 7
- Total Deliveries
6.1.6.2 Model builder module
6.1.6.1.9 SETTING UP CONTRACTS
In this module you are able to convert the

external data into Synergi model
The direction of flow (entering or leaving

the network) is defined by the type of
contracts at the node. All flow values
defined for contracts and region nodes
must be positive. All contracts at a given
node must have the same flow direction.
Model Builder will import external data

into a Synergi model from either Middle
Link 3.x CSV files or any number of non-CSV
Data Source Definitions (DSDs), which
includes shape files, CAD files, Arc/Info
Coverages, personal geo-databases, SDE
(multi-user) geo databases, and query
layers. It is not possible to import both CSV
files and non-CSV (spatial DSD) files at the
same time.
PURCHASE and DELIVERY are for the

fixed flows that are not being optimized.
For a TRANSPORT agreement, flow at the
supply node must exactly match flow at
the delivery node. The price represents the
unit revenue for transport.
For CSV import, the entire set of CSV files

in a specified directory will be converted.
The types of CSV files in the specified
directory control the set of imported
facilities. Each set of CSV files must have at
least a node file and a pipe file (which is the
same requirement in Middle Link 3.x).
The total number of contracts cannot

exceed 2,000.
VIEWING OPTIMIZATION DATA

After performing a successful optimization
analysis, you can use a number of Synergi
tools to view the results of the
optimization, including setting up
annotation and Color By schemes in the
map display or by viewing various
optimization reports and charts
For spatial DSD import, you can select DSDs

that represent combinations of conversion
options. You can:
272
Create Synergi pipe facilities, nonpipe facilities, and isolation valves.

Create only Synergi pipe facilities.
Chapter 6
Second, the computational effort required

by the transient analysis becomes
prohibitive. This effect of exhausting the
system capacitance is a good
Create only Synergi non-pipe

facilities.
6.1.6.3 Time varying module

Synergis Time Varying Module (TVM) lets
you model almost any gas system
composed of pipelines, compressors,
valves, regulators, and storage fields. In
addition, you can control all aspects of the
modeling process and have the flexibility
to easily simulate extremely complex
cases.
indication that a steady-state technique

would be valid to represent a portion of
time.
TVM addresses this basic problem of
modeling a natural gas system for an
extended period of time.
The following list, though not conclusive,
summarizes the most predominant
applications of TVM.
During a time-varying analysis, Synergi

performs a series of consecutive steadystate analyses to represent the changes
that take place in a network over time. The
analysis is performed in increments called
time steps. Each time step represents a
planned interval when you want Synergi to
perform a steady-state analysis. The size of
the time steps, as well as the total amount
of time that you want the simulation to
run, are set before the simulation begins.
TVM is particularly useful for studying slow

time-varying effects where the system can
be assumed to be at steady-state
conditions during all points within the
analysis. Examples of such conditions are
the determination of storage field usage
during a heating season and the evaluation
of compressor fuel usage.
Steady-state analysis, by definition, does

not permit parameters to vary with time,
while traditional transient analysis
examines the problem of varying loads
over a fairly short period of time (a day for
example) to determine the short-term
capacitance of the system and how it
affects the daily operation. When longer
periods of study are required, generally
two things occur. First, the effects
associated
with
the
short-term
capacitance of the system are minimized.
273
System adequacy to meet the

anticipated loading. With TVM the
effect of different temperature
dependent loading conditions can be
studied, and the results of the various
simulations
can
be
examined
graphically. The effect of significant
swings in load due to temperature
changes can be studied to anticipate
what impact these swings may have on
the operation of the network.
Compressor fuel requirement over the
extended time period. The use of
compressor fuel over a season can be
studied and predicted ahead of time to
budget for fuel consumption. In
addition, inefficiencies over the season
can be examined and different
operating scenarios can be tried to
reduce the amount of fuel used.
Storage injection and withdrawal
schedules. The complex operation of a
storage system can be planned ahead
of time, based on predicted flows to
ensure that the entire volume in
storage can be turned over. In addition,
iterations can be performed that allow
the use of current storage levels to help
predict the behavior of the system and
Chapter 6
whether or not the demand from

storage can be met in severe weather
conditions.
Predict system behavior based on
starting conditions. Real data from the
system can be imported into Synergi
and used as starting conditions. These
starting conditions can then be used in
TVM to assist operators and planners
with running
2. Attributes and variables

Attributes are the individual pieces of
Synergi data that you can access for
reporting and simulation control.
Attributes include data such as node
pressure and flow, facility flow,
selected operating parameters, and
overall system parameters such as
linepack, total power requirement, and
total fuel requirement. Many of these
attributes are available through
reports, charts, scripts, and the Results
Plan.
What If scenarios to assist with the

decision making process about the
operation of their network.
In addition, you can create your own

custom
variables.
User-defined
variables are based on some
combination of other attributes and
variables. For example, you can define
a variable to represent the sum of the
linepack in three pipe sections, or the
sum of compressor fuel over a group of
stations. These variables offer a great
deal of flexibility when working on your
model.
Now let us review the components that

distinguish the time-varying model.
1. Profiles
A profile is a curve that defines a
functional relationship between two
variables. Many profiles in Synergi
define the relationship between a
model setting and time. For example,
in USM (Unsteady State Module), you
frequently specify a node flow profile
that simulates demand at different
times during the day, with higher
demand during peak usage hours and
lower demand at non-peak usage
hours. In a TVM simulation, while the
same can be done, it is more
appropriate to look at flow variances
over the course of many days, a
heating season, or even a period of
several years. In either case, during the
time-varying simulation, the demand
will change based on the simulation
time, and analysis calculations will be
performed based on the profiled
demand.
Other
time-dependent
profiles that you can create include
node pressure, regulator flow, valve
position, and compressor utilization.
3. Charting and reporting capabilities

Charts and reports are effective
methods to examine the results of a
time-varying analysis. You can use
Synergi charts to chart selected
variables with respect to time, or to
chart selected variables as a function of
distance along a pipeline. Any attribute
or user-defined variable may be
selected for charting.
To perform a typical time-varying
analysis:
1. Create or open a model and work on
model features and specifications.
274
Chapter 6
2. Specify profiles in your model, so

that you may simulate desired features
with respect to time.
most predominant applications of USM in

the field of gas transient analysis.
Transient network design. When designing
a gas transportation system, accounting
for the time varying flow and pressure
response is critical. Design based solely on
steady-state assumptions is often overly
conservative, since a steady-state analysis
cannot account for the effect that line pack
can have on fluctuating conditions. The
proper use of transient analysis often
allows the engineer to reduce the pipeline
size, compressor horsepower, or regulator
station capacities, therefore preventing
expensive capital improvements while
achieving a system that can operate
efficiently and economically. The basis for
reducing system facilities is often founded
on an understanding of the effect of timevarying loads and supplies.
3. Specify time simulation model

settings and options.
4. Define the variables in the Results
Plan that you may want to chart after
performing a time-varying analysis. In
addition, use the Results Plan to set up
auto labels.
5. If you desire to display a userdefined schematic diagram of the
condition of the model during a time
simulation, set up the Results Layout.
6. Perform a time-varying analysis.
7. After completing the analysis, use
the Simulation and Time Controls to
review the results of the analysis.
Transient operation analysis. Evaluating

different operating scenarios and selecting
optimum operating procedures can result
in
Generate charts and reports to further

review the results of the analysis.
6.1.6.4 Unsteady State Module

Synergis Unsteady State Module (USM)
models the time-varying pressure-flow
relationship that exists in most gas
transportation systems. In these systems,
conditions are continually changing;
therefore, the system is said to be
transient. As pipeline conditions change
(as regulators regulate, compressors start
and stop, or loads vary), the steady-state
pressure-flow
assumption
becomes
invalid. Consequently, it becomes
important to simulate the packing and
drafting of the pipeline, and to accurately
assess the impact of system changes on
pressures and flows.
increased
system
reliability
and
significantly reduced operating cost.
Transient analysis helps you to minimize
compressor fuel usage, determine proper
emergency operating procedures, and
assess system survival time under certain
conditions.
Gas blow down or lost gas simulations. If a
pipeline must be blown down, the
engineer can determine the safest and
fastest way to perform this operation. USM
is an excellent tool for designing pipeline
venting systems and determining proper
evacuation times. In addition, USM is often
used to perform lost gas calculations in
case of a pipeline rupture. How unsteadystate modeling complements steady-state
modeling
The flexibility of USM allows analysis of a

variety of gas flow problems. The following
list, though not conclusive, summarizes the
275
Chapter 6
In the natural gas network analysis

industry, there are many offerings
available for modeling compressible flow.
These tools for design, analysis, leak
detection, real-time modeling, operator
training, and optimization, and so on fall
into two basic categories of service:
steady-state and transient.
determines the time-varying pressure-flow

relationships
that
exist
in
gas
transportation systems.
Most gas transportation systems exhibit
continually changing conditions, and thus
can be said to be unsteady. When
analyzing these systems, it becomes
important to simulate the packing and
drafting of the pipeline to assess accurately
the impact of system changes in pressures
and flows.
The better known and more fully

understood of these categories is steadystate network analysis. Steady-state
analysis
makes
the
fundamental
presumption that the flow through a
system is constant, and that the sum of the
flows into the system precisely equals the
flow out of the system with the exception
of processing plant shrinkage, and
compressor fuel. This relationship ensures
the mass continuity of the system
modeled, as shown in the following
equation:
USM analyzes a wide variety of transient

gas flow problems in systems consisting of
pipelines, compressors, valves, regulators,
and storage fields, etc. Since it is
interactive, you can monitor and affect the
execution of the simulation as it
progresses.
Some typical applications of USM include:
=
Transient analyses, in contrast, are not
constrained by the Q relationship. This
category uses completely different
numeric algorithms and techniques to
solve differential equations that relate
mass flow rate, momentum, velocity,
acceleration, and friction, etc., with
respect to time and horizontal pipeline
distance.
When considering gas transportation

systems, the term transient, implying a
sudden event of brief duration, is
misleading. A much better term for gas
systems is unsteady-state, which more
appropriately
represents
the
phenomenon.
Analyzing compressor start-up or

failure and other upsets to modelingspecified load, or supply change versus
time
Examining specified pressure variation
at pressure control points
Determining system line pack
Analyzing arbitrary changes in
compression horsepower or valve
settings
Estimating gas lost in blow down or
pipeline damage situations
Scheduling equipment for more
efficient operation
USM can model the unsteady response of

natural gas transmission systems, and
specialty gas systems, such as boiler flue
gas. Because the physical properties of the
gas can be specified, USM can model many
USM provides modeling capabilities for

unsteady-state analyses. USM also
276
Chapter 6
different single-phase gas transportation

systems.
USM is capable of modeling essentially any

single-phase,
isothermal,
gas
transportation system under unsteadyflow, or transient, conditions. Both simple
and highly complex systems can be
modeled with USM.
6.1.7 script
Synergies scripts are a very powerful tool
that allows you to automate tasks that you
want to repeatedly perform in your model.
Synergi provides for several different types
of scripts, each with a distinct purpose:
You have the option of defining different

quality gases at every source. The gases
can be defined by the heat content and
specific gravity, or by giving the
composition of the gas.
Master scripts are external script files that

are designed to be run from the Opening
Assistant or from a shortcut on your
computer. You can use a master script to
load a model and then perform a series of
different tasks import and export
exchange files, run steady-state and timevarying analyses, generate charts and
reports, and so on. In fact, with a Synergi
script that is run by a desktop shortcut that
is linked to a Windows automated task,
you can set up your computer to perform
these tasks repeatedly each morning and
have key data waiting for your review
when you arrive at the office.
Pipeline friction calculation methods in

USM
You may select which flow equation you
wish to use for pipes in your model. Pipes
can use the same equation throughout
your model.
Possible pipe equations include:
Dry, saturated steam pipe equation

with condensation loss (MS)
Fundamental pipe with constant

friction factor equation (FD)
Fundamental
pipe
with
flowdependent friction equation (FM)
General pipe equation (GP)
Panhandle A equation (PA)
Panhandle B equation (PB)
Weymouth pipeline equation (WE)
Operating scripts are internal script files

that are designed to be run at each time
step during a simulation. (Operating scripts
can also be run manually from the Model
Data tab.) Operating scripts can be used to
perform specialized tasks that Synergi may
not be able do on its own. For example, you
can use an operating script to monitor the
pressure at a key delivery point in your
system. If the pressure approaches, but
does not reach, the minimum allowable
pressure, you can increase the setpoint of
a nearby regulator, with the hope that it
will keep the pressure above the true
minimum. After a set period of time, you
can then return the regulator back to its
original setpoint.
You will be unable to perform an

unsteady-state analysis if your model
contains any pipes using the following
equations:
IGT distribution pipe equation
Mueller pipe equation (MU)
Spitzglass low pressure pipe
equation (SL)
Spitzglass high pressure pipe
equation (SP)
(FS)
277
Chapter 6
The scripts can be as brief or as detailed as

you desire, and the tasks you can perform
in a script are limited only by your
imagination. Thus, the documentation
provided here is written to explain the
commands that we used in the scripting
language; provide some basic procedures
for working with script files; and provide
some very detailed descriptions on the
ways you use scripts to interact with your
Synergi models.
6.2 Building up the model

6.2.1 Import from GIS
6.2.1.1 How to import base
mapping
From the file menu Choose import then
select base mapping then choose files
The following sections provide information

on writing Synergi scripts:
1. Managing Scripts and Script Files:
provides some simple instructions on
working in Synergis script window to
create, compile, run, and save script files.
Figure 6- 8
Then choose the map of the region with

(lyr) extension
2. Understanding the Scripting Language:

provides an overview of the key concepts
you will need to understand before you
can write a Synergi script, including
variables, statements, functions, and
comments
3. Synergi Classes, Methods, and
Properties: provides detailed information
on all of the Synergi-specific actions and
data that you can interact with in your
scripts. (Classes and methods refer to
Synergi tasks you can perform on specific
model areas, such as loading a database or
performing an analysis; properties refer to
Synergi data that you can interact with,
such as a node pressure or a pipe
diameter.)
Figure 6- 9
Note: to import the base map it is required

to have an access to the- internet
connection to match the coordinates of
the pipeline network with the geographical
location of the world map
So this will allow to the model to have the
geographic map laying on it as a
background
4. Script Tips and Tricks: provides some

helpful guidance that will make the job of
writing a script much easier.
278
Chapter 6
Figure 6- 12
Figure 6- 10
6.2.1.2 How to import the shape

files
To obtain the coordinates of the nodes and
the actual lengths of pipeline it is required
to import the shape files.
By the same steps of importing the- base
mapping but choose the required shape
files instead of the base map layer
Figure 6- 13
Figure 6- 11
Then convert it from file menu then choose

import then convert base map to the
model
6.2.2 Node Naming

Manually importing the names of each
node by using the identify tool to obtain
the node names from the GIS and enter it
in the description field of the nodes in the
synergi's model.
Figure 6- 14
279
Chapter 6
6.2.2.1 How to export an exchange

file
From the file menu choose export then
choose export file
Figure 6- 17
6.2.2.2 Importing Flowrates

Then entering the flowrates manually at
the excel sheet
Figure 6- 15
Then define the name of the exchange file

and its form and the required fields
Figure 6- 18
6.2.2.3 Importing the excel sheet to

the synergi model
To reimport the same excel sheet with the
flowrates of the nodes from file menu
choose import then choose the same
exchange file
Figure 6- 16
Figure 6- 19
280
Chapter 6
geodatabases,
SDE
(multi-user)
geodatabases, and query layers. It is not
possible to import both CSV files and nonCSV (spatial DSD) files at the same time.
For CSV import, the entire set of CSV files
in a specified directory will be converted.
The types of CSV files in the specified
directory control the set of imported
facilities. Each set of CSV files must have at
least a node file and a pipe file (which is the
same requirement in MiddleLink 3.x).
For spatial DSD import, you can select DSDs
that represent combinations of conversion
options.
You can:
Create Synergi pipe facilities, non-pipe
facilities, and isolation valves.
Create only Synergi pipe facilities.
Create only Synergi non-pipe facilities.
You can also define a DSD to create Synergi
nodes. Such a DSD only has meaning in the
context of also importing one of the above
combinations of pipe and non-pipe
facilities. Any nodes remaining at the end
of the conversion process that do not have
at least one attached facility are
automatically deleted from the model. An
alternative form of a node DSD is a node
insert DSD. When imported as a node
insert DSD, Model Builder can attempt to
locate a pipe at the node points XY
coordinate, split the pipe, and insert the
node, and also to update existing nodes
with data from incoming nodes.
6.2.3 Model Builder Module

Model Builder allows you to import data
from either MiddleLink 3.x CSV files or
from a number of Esri GIS data sources,
including Shapefiles, CAD files, Arc/Info
Coverages, personal ArcGIS geodatabases,
SDE-based
enterprise
ArcGIS
geodatabases, and query layers. The
process that converts these file types
(which are referred to in this
documentation as MiddleLink CSV files
and spatial data files) involves two basic
steps:
In the configuration phase, you create
Data Source Definitions (DSDs) for each
logical unit of work associated with the
import process. For example, one or more
DSDs can be configured for importing pipe
data, while other DSDs can be configured
for importing valve data, regulator data,
etc. You can also configure a DSD to
explicitly import node data (such as the
MiddleLink node CSV file). The relationship
between DSDs and source data files are
discussed in much greater detail in
Creating Data Source Definitions
(DSDs)
In the conversion phase, you set up and
perform the actual import of external data
into
Synergi. During the conversion, you specify
which of the previously-configured DSDs
will be processed, and then set other
options that control how new nodes and
facilities are created.
6.2.3.2 Nodes and facilities
6.2.3.1 HOW MODEL BUILDER

CONVERTS EXTERNAL DATA INTO
A SYNERGI MODEL
The following list provides some general

guidelines on the conversion of nodes and
facilities:
Synergi nodes can only be created from
source data with point geometry. For any
Model
Builder conversion, you can choose as
many node DSDs (or node import DSDs) as
desired.
Model Builder will import external data

into a Synergi model from either
MiddleLink 3.x CSV files or any number of
non-CSV Data Source Definitions (DSDs),
which includes shapefiles, CAD files,
Arc/Info
Coverages,
personal
281
Chapter 6
Synergi pipe facilities can only be created

from source data with polyline geometry.
For any Model Builder conversion, you can
choose as many pipe DSDs as desired.
Synergi non-pipe facilities can be created
from source data with point geometry or
polyline geometry. For any Model Builder
conversion, you can choose as many nonpipe
DSDs as desired.
Synergi isolation valves can only be
created from source data with point
geometry. For any Model Builder
conversion, you can choose as many
isolation valve DSDs as desired.
Note that the conversion of external point
features into Synergi non-pipe facilities
and isolation valves depends on the preexistence of nodes and pipes in the Synergi
model. While there is no requirement to
import pipes on a single Model Builder
conversion, the conversion of external
point features into Synergi non-pipe
facilities or isolation valves will not be
successful if you attempt to import these
point features into a geographic area that
is not already populated with Synergi
nodes and pipe facilities.
6.2.3.4 CREATING DATA SOURCE

DEFINITIONS (DSDS)
You configure Model Builder by creating a
Data Source Definition (DSD) for each
logical unit of work associated with the
import process. For example, you will
typically create one or more
DSDs for importing pipe data, and
additional DSDs for importing valve data,
and still more
DSDs for importing regulator data. It is
even possible to configure a DSD to
explicitly import node data (such as the
MiddleLink node CSV file).
The Model Builder Configuration dialog
box, is used to create and manage DSDs.
The left side of the dialog box lists groups
that you have created to organize the
DSDs. Each group can contain an unlimited
number of DSDs. Groups can be added,
renamed, and deleted. When you delete a
group, any DSDs included in the group will
be deleted as well.
Any DSD that does not belong to a group
will be placed at the root level.
6.2.3.3 Attributes
Source file attributes can be mapped to
Synergi facility attributes, including
dynamic attributes (such as multiple flow
categories) and custom attributes. These
attribute field mappings, along with other
user selections, can be saved in a
configuration file so they can be reused in
future
Model Builder conversions.
NOTE: Custom attributes must exist in the
object model before they can be mapped
properly. If you have an existing model
with the necessary custom attributes, you
may find it easier to import them using
Model Builders supplemental data import
feature. These custom attributes must
have been assigned to model features.
Figure 6- 20
The right side of the Model Configuration

dialog box lists the DSDs that are included
in the selected group. You can add, edit,
rename, and delete any DSD, as well as
copy and paste
DSDs within the same group or to any
other group. A key column in the list of
282
Chapter 6
DSDs is the valid column. A value of Yes

indicates that there are no detected
problems with the DSD data. A value of
Invalid indicates that one of the
following problems exists:
The file or directory referred to by the
DSD no longer exists. This is most likely to
happen if a user moves files to different
directories or if one user sends a copy of
the configuration to a different user on a
different computer.
The default warehouse item specified on
the Details tab of the DSD Editor does not
exist in the currently loaded equipment
warehouse.
One or more of the warehouse item
names in the Lookups tab of the DSD Editor
do not exist in the currently loaded
equipment warehouse.
Group assignments and DSD definitions
can be saved to a configuration file
(.conf.ini file format). In the Model Builder
Configuration dialog box, you can use the
Load Settings at the top of the dialog box
to load settings from an existing file, and
then use the Save Settings area at the
bottom of the dialog box to save the
settings to either the same or a new file.
Refer to the procedures listed below for

more information on creating, editing, and
deleting DSD groups.
To create a DSD group for a MiddleLink CSV
file
1. From
the
Model
Builder
Configuration dialog box, under the
list of DSDs (located on the left side
of the dialog box), click
2. In the Select Pipe CSV File for
Import, select the name of the .PIPE
file from the group of MiddleLink
CSV files that you want to import.
3. Click OK. Synergi will automatically
create the DSD group and assign its
name as the root name of the PIPE
file. Synergi will also automatically
recognize any valid MiddleLink CSV
files that are saved in the same
directory as the .PIPE file, and
create a matching DSD for each CSV
file.
6.2.3.6 Managing DSDs

Refer to the topics listed below for more
information on creating, renaming, and
deleting DSDs.
To access the Model Builder Configuration

dialog box
To create a DSD for a MiddleLink

CSV file
From the main menu, select File > Import >

Model Builder > Configuration.
DSDs for MiddleLink CSV files are

automatically created when you create the
DSD group.
6.2.3.5 Creating and editing DSD

groups
To create a DSD for a spatial file

DSD groups allow you to organize DSDs
into different categories, such as facilitybased categories (pipes, valves, etc.), file
types (MiddleLink CSV files, personal
geodatabases, SDE geodatabases, etc.),
query layers, or other characteristics of
your data. You can create as many
DSD groups as you need, and each group
can contain an unlimited number of DSDs.
From the Model Builder Configuration

dialog box, under the list of DSDs (located
on the right side of the dialog box), click
.
To rename a DSD
283
Chapter 6
1. From the Model Builder Configuration

on the right side of the dialog box), select
the DSD that you want to rename.
dialog box are not used with MiddleLink

CSV files and therefore are grayed out.
3. Select the Details tab. Use this tab to
specify which Synergi warehouse item will
be used to create the initial data for new
facilities. Other options on this tab of the
dialog box are not used with MiddleLink
CSV files and therefore are grayed out.
4. Select the Attribute Mappings tab. Use
this tab to review how attributes from the
MiddleLink CSV file will be matched with
attributes in the Synergi model. These
matchings are read from the MiddleLink
CSV file and cannot be modified.
When a new facility is created in Synergi,
attribute values are first assigned based on
the values in the specified warehouse item
(which is selected on the Details tab of this
dialog box). Synergi will then attempt to
match any attribute values from the
external file, as specified on the Attribute
Mappings tab. If an attribute match is
accepted, the value from the external file
replaces the value from the warehouse
facility. If the attribute match is not
accepted, the value from the warehouse
facility is retained. A log message will be
generated, provided that you have
selected the Report check box for the
specified attribute.
However, the only column that you can
edit when you are processing MiddleLink
CSV files is the Report column.
5. Select the Lookups tab. Use this tab to
match pipes from the external file with
warehouse items from the current
equipment warehouse. Since most
geodatabases do not include an internal
diameter field, it is necessary to use unique
combinations of a pipes material and
nominal diameter when matching pipes
from the external file with items from the
equipment warehouse. The equipment
warehouse item includes not only the
internal diameter, but also the roughness
and many other attribute values.
2. Click
.
3. In the DSD list, type the new name for
the DSD and then press Enter.
To delete a DSD
the DSD that you want to delete.
2. Click
To copy and paste a DSD

the DSD that you want to copy.
2. Click
to copy the selected DSD.
3. Click
DSD.
to paste a duplicate copy of the
6.2.3.6 Editing DSDs

The Data Source Definition editor contains
different tabs and settings, depending on
the type of DSD that you are editing. Refer
to the procedures listed below for more
information.
To edit a MiddleLink CSV DSD
the DSD that you want to edit, then click
.
2. In the Data Source Definition Editor
dialog box, select the Definition tab. Use
this tab to modify the name of the DSD, if
required. Other options on this tab of the
284
Chapter 6
NOTE: This tab is only visible when you

have specified that the DSD includes pipes.
This is specified as the Synergi type value
on the Details tab of the dialog box. If you
are configuring the DSD for the first time, it
may be necessary to close and then reopen
the dialog box in order for the Lookups tab
to appear.
6. Click the Settings tab. Use this tab to
specify which delimiters are used in the
CSV file to represent columns and decimal
places.
7. Click OK to close the dialog box.
When a new facility is created in Synergi,

attribute values are first assigned based on
the values in the specified warehouse item
(which is selected on the Details tab of this
dialog box). Synergi will then attempt to
match any attribute values from the
external file, as specified on the Attribute
Mappings tab. If an attribute match is
accepted, the value from the external file
replaces the value from the warehouse
facility. If the attribute match is not
accepted, the value from the warehouse
item is retained. A log message will be
generated, provided that you have
selected the Report check box for the
specified attribute.
5. Select the Lookups tab. Use this tab to
match pipes from the external file with
warehouse items from the current
equipment warehouse. Since most
geodatabases do not include an internal
diameter field, it is necessary to use unique
combinations of a pipes material and
nominal diameter (or other user-specified
fields) when matching pipes from the
external file with items from the
equipment warehouse. The equipment
warehouse item includes not only the
internal diameter, but also the roughness
and many other attribute values.
NOTE: This tab is only visible when you
have specified that the DSD includes pipes.
This is specified as the Synergi type value
on the Details tab of the dialog box. If you
are configuring the DSD for the first time, it
may be necessary to close and then reopen
the dialog box in order for the Lookups tab
to appear. When specifying and retrieving
lookup values, you should be aware of how
Synergi handles NULL values. For more
information about this behavior,
6. Select the Pressure Translation tab. You
can use this tab to orient non-pipe facilities
based on the unique values in an external
attribute field that pertain to relative hightolow pressure values. If this data does not
exist in your pipe DSD source data, you
6.2.3.7 Editing a spatial data file

DSD
To edit a spatial data file DSD
the DSD that you want to edit, then click
.
2. In the Data Source Definition Editor
dialog box, make sure that the Definition
tab is selected. Perform the following steps
in the Definition tab to select a feature
class from a spatial data source to use with
the DSD. You can only select one feature
class per DSD.
3. Select the Details tab. Use this tab to
specify the Synergi facility you are creating
in this DSD, as well as to set naming
conventions and to specify how much of
the data in the external file you want to
convert.
4. Select the Attribute Mappings tab. Use
this tab to specify which attribute fields
from the external file will be imported and
assigned to Synergi attribute fields. You
must make a matching assignment for
each attribute that you want to import.
Any external attributes that are matched
to a Synergi attribute of [None] will not be
imported.
285
Chapter 6
should not enable the functionality on this

tab.
If you do have relative high-to-low
pressure values in your data, then you
should enable the functionality on this tab.
Use the grid in the upper half of the tab to
specify which attribute field in your
external data contains the relative
pressure data. Then, use the grid in the
lower half of the tab to order the pressure
values from highest (at the top of the list)
to lowest (at the bottom of the list).
When Synergi later imports non-pipe
facilities during the Model Builder import
process,
Synergi can use this data to compare the
connecting pipes on either side of the
nonpipe facility. If one pipe has a higher
pressure value than the other, then Synergi
can use that information in an attempt to
determine the correct orientation of the
non-pipe facility. If both pressure values
are equal, or if your basemap file does not
include pressure data, then Synergi will use
relative pipe lengths to determine the
direction to place the non-pipe facility. You
will need to review the non-pipe facilities
after the Model
Builder conversion is complete and reverse
the orientation of any facility that was
placed backward in the model.
The Pressure Translations tab is only visible
when you have specified that the DSD
includes pipes. This is specified as the
Synergi type value on the Details tab of the
dialog box. If you are configuring the DSD
for the first time, it may be necessary to
close and then reopen the dialog box in
order for the Pressure Translation tab to
appear.
7. Select the Facilities Naming tab. The
options on this tab are enabled when you
select
Advanced Naming as the facility naming
method on the Details tab. For more
Information.
8. Select the Nodes Naming tab. The

options on this tab are enabled when you
select
Advanced Naming as the node naming
method on the Details tab.
9. Click OK to close the dialog box.
6.2.4 Inserting Data to the

model
6.2.4.1 Introduction
The National Gas Grid is the largest and
longest in Africa and the Middle East, The
grid has reached the furthest point of the
south valley in Aswan and is extending east
to Sinai and along the coast of the Red Sea,
north to Lebanon and Syria with potential
to reach further.
National Gas Grid was expanded during the
fiscal year 2012/2013 through completion
of new gas pipelines, other pipelines are
still under implementation aiming to
achieve gas connection to the maximum
number of factories and to deliver gas to
provinces, preserving environment as a
strategic goal by reducing the consumption
of other petroleum products hydrocarbon
contaminated environment (LPG Gas Oil
Fuel Oil), contribute to establishing new
plants working with natural gas, in addition
to feed houses and factories located along
the route of these national gas grid
pipelines.
Our Transmission Capacity which has
reached 210 MMSCMD by the end of mid
year 2014 compared to 37 MMSCMD in
1997. This is due to increasing the length of
the grid to 7075 Km of Trunking lines
against 2800 Km in 1997, to secure
depreciation and pressure regions of
Egypt. Natural Gas National Grid is being
controlled by using the latest technology of
SCADA system, which controls the network
through the national network center
NATA.
286
Chapter 6
Figure 6- 21
6.2.4.2 Development of Natural Gas

Grid Quantities Distribution
Figure 6- 22
By the end of the 1st half of year 2014 the

total average of natural gas distributed
through the national gas grid reached 134
MMSCMD (48 910 MMSCM/YEAR) against

36.5 MMSCMD in 1997 achieving
development rate amounting 392%.
287
Chapter 6
6.2.4.3.2 Demands Nodes
6.2.4.3 Inserting hydraulic data

Nodes
As for providing the needed energy for

community development, a reflection of
the increase in the rate of economic
growth, mirroring a rapid increase in
consumption across all sectors of the
country. The strategy of Egyptian
petroleum sector focuses on the local
market demand and gives it absolute
priority. Total local consumption of natural
gas reached 1842 BCF, representing 88 %
of total sales gas, while the remaining 12%
of natural gas exports represents about
33% of gas revenues in order to achieve
the maximum value of natural gas without
violating the existing commitments.
Nodes in model are divided into:

Supply Nodes
Demands Nodes
6.2.4.3.1 Supply Nodes
There are many production fields in the
National Gas Grid but much of them are
located in the Mediterranean Sea and
Western Desert, in our model some feed
points are taken into consideration, these
points are defined with flow rates, such as:
table 6- 1 supply flow rates
NAME
Flow rate
(m3/h)
WDGC
934 500
Abu Qir
268 500
Badre El-Din
630 000
Rashpetco
1 085 000
UDGC
1 657 000
Abo Mady
630 000
El.Mansoura
126 000
Dahshour
123 000
Ras bakr
25000
Ras Shouqir
25 000
North Sinai
90 000
6.2.4.3.2.1 Power station sector

Electricity Sector is the largest gas consumer, where natural gas consumption
represent about 60 % of total local gas
consumption. Power stations which use
the natural gas as a fuel are classified to:
288
Steam power stations

Gas power stations
Combined power stations
Chapter 6
table 6- 2 power station flowrates
Delta
Alexandria
REGION
NAME
FLOW (m3/hr)
TYPE
Abo Qir
87 000
Steam
El Mahmoudia
68 000
Combined
Damanhour
46 000
Combined
El Seoof
50 000
Gas
Kafr El Dawar
74 000
Steam
Sidi Krir (1&2)
42 000
steam
Sidi Krir (5&6)
90 000
Combined
Intergen (3&4)
138 000
Steam
Atf new
138 000
Combined
Abo Qir new
51 000
Steam
Damietta
208 000
Combined
Talkha gas
64 000
Gas
Talkha steam
51 000
Steam
Nubaria
380 000
Combined
Talkha Comb.
125 000
Combined
New damietta
138 000
Gas
West damietta
140 000
Gas
Banha
95 000
Gas
289
Suez Canal & Sinai
Cairo
Chapter 6
Shobra El Khema
33 000
Steam
South Cairo
56 000
Combined
West Cairo
63 000
Steam
North Cairo
168 000
Combined
Wady hoof
11 000
Gas
Teppin
14 000
Steam
Korymat
127 000
Steam
Korymat (comb.)
123 000
Combined
Abou Rawash
150 000
Gas
Ataka
46 000
Steam
Abou Sultan
28 000
Steam
Port Said
14 000
Gas
El Shabab
30 000
Gas
Mousa
156 000
Steam
West Suez
153 000
Steam
East port said
180 000
steam
El Arish
19 000
Steam
The data shown in the table are collected from confirmed data sheets.
290
Chapter 6
6.2.4.3.2.2 Industrial Sector

Industrial areas are the second priority as
it represents 30 % from the total gas
consumption.
The industrial sector is divided to many

industries such as:
Petroleum Derivatives and petrochemicals.

Fertilizers industry.
Cement industry.
4Other industries.
table 6- 3 industrail flowrates
Alexandria
REGION
INDUSTRY
Petroleum ,
Derivatives
And
petrochemicals
NAME
FLOW RATE(m3/h)
Alex petroleum
19 700
Ameriya petroleum
26 600
petrochemical
27 000
Midor and Medalic
64 500
Amoic
10 500
Sidpic
7 000
Poly probelin
21 800
Misr Chemical
6 100
Carbon Black Factory
7 500
El Mogamaa for petrol
46 000
Misr petroleum
160
ANRPC
6 770
Alex for Sodium carbonat

(solvay)
5000
Star Nex
2000
Abou Qir Fertilizer
129 800
Alexandria Fertilizer
40 000
Iron & Steel
Dekheila Steel
80 000
Cement
MAX Cement
17 550
Fertilizers
291
Chapter 6
Others
Petroleum ,
Derivatives
And
petrochemicals
Fertilizers
2 720
Alex Free Zone
1 870
Alex IA
18 420
Alex sugar
31200
Nobaria Sugar
18030
Methanics
23 000
Fenosa Facilities
4 000
Uoria
72 400
Natarat
Delta
Ghazl el mahla
Others
Fertilizers
Cairo
America Cement
Cement
Others
200
19 500
Quesna(Egypt gas)
2700
Repco GAS
2800
Gastec Talkha
800
Dakhlia Sugar
26 000
Damietta pump St.
4 400
Cebaght Elmahala
2800
Helwan Fertilizer
50 000
Helwan Cement
20 000
National Cement
37 900
Ktamya cement
7 400
Kama en El Tob
75 000
Ganope Al Wadi Cement
6 000
Town gas iron & steel
35000
10th of Ram
95 000
292
Chapter 6
Suez Canal & Sinai
Petroleum ,
Derivatives
And
petrochemicals
Fertilizers
Cement
Hawmdya sugar
13 800
Fayoum Gas
22300
Nile Valley off Take
14000
EL Nasr Petroleum
7 500
Suez Petroleum
11 500
Gas Derivatives
40 000
Morgan
31 000
Gapco
7 000
Egypt for Fertilizer(2)
42 000
Suez Fertilizer
15 000
Suez Cement
1 000
Arabiya Cement
1 000
Egypt Cement
5 000
Sweedy Cement
1 000
Gastec Suez
Others
400
Sinai IA
15000
Epec
45000
Bani swaif cement
5000
Upper Egypt IA
5000
Abo Qurqas
6000
El Minia IA
1000
Qift
17700
Wadi el Nile cement
5000
El Minia
10000
Assuit
10000
293
Chapter 6
Upper
Egypt
Industries
Qena
22800
Naga hamadi
11500
Kima
185 800
Aswan
30000
Bani swaif
5000
Egypt gas Aswan
10000
The data shown in the table are collected from confirmed data sheet
6.2.4.3.2.3 Residential sector
The residential category is the less Priority as it represents 3-5 % from the total gas
consumption.
table 6- 4 residential flowrates
sector
Residential
Sector
NAME
FLOW RATE (m3/h)
Nat Gas west Alex
14400
Nat Gas El-Agamy
9920
Nat Gas Borg El Arab
34620
Nat Gas kafr eldawar
5600
Nat Gas abo homos
800
Nat Gas Damanhur
9000
Nat Gas el mahmoudia
4420
Nat Gas kom hamada
2410
Nat Gas Rashid
2810
Nat Gas wadi el natron
1970
Town Gas Alex
15000
Trans Gas-Delta
13000
Egypt Gas-Delta
40000
Egypt Gas Abo Zabal
15400
Egypt Gas Cairo
95000
294
Chapter 6
Egypt Gas Delta
40000
Nat Gas Cairo
15000
Town Gas Cairo
72000
Nat Gas 6 th Oct.
35000
Town Gas Ismailia
5000
Town Gas Port Said
16000
City Gas
124000
The data shown in the table are collected from confirmed data sheets.
295
Chapter 6
Facility flow refers to the flow as it moves

throughout the model, from one facility to
the next that make up the model. A
positive flow indicates that the fluid is
flowing from the from-node to the tonode. A negative flow indicates that the
fluid is flowing from the to-node to the
from-node. The order of the from-node
and to-node is typically not important for
pipes and valves, but it is important for
other facility types that are directional by
nature, such as a regulator or compressor.
6.3 Model Components

6.3.1 Nodes
Nodes are the starting and ending points of
facilities. Typically, nodes are located at
the
following positions:
The intersections and ends of any pipes

and other facilities
Changes in pipe material or pipe
diameter
Locations of significant demands or
supplies
Significant changes in elevation
Node flow can be specified as being either

volumetric, meaning that the demand is
specified in volume-based units such as
standard meters per hour, thousands of
cubic feet per day, etc., or thermal,
meaning that demand is specified in
heating (calorific-based) units, such as Btu
per hour, Gigajoules per day, etc. The
thermal equivalent of volumetric flow, and
vice versa, are computed by the solution
process as a function of the heating value
of the gas.
In Synergi, nodes have hydraulic properties

associated with them, such as flows and
pressures.
Nodes are created and named
automatically as you add facilities to your
model. You can also
insert nodes manually into an existing pipe
segment, thus splitting one pipe into two
pipes with
Node flow example.

Figure(6-23) shows a very simple model
that includes four pipes and five nodes.
Node A is a supply node, while nodes B, D,
and E are demand nodes. Node C is a zeroflow node because it does not have any
supplies or demands associated with it.
The total flow into the network must equal
the total flow out of the network. Since the
total system demand is -20 (-10 + -5 + -5),
the supply at node A must equal 20.
identical characteristics (except for length,

which will be determined based on the
location of the inserted node).
6.3.1.1 NODE FLOW AND FACILITY

FLOW
Node flow is the flow into or out of the
network at a node. Flow is positive if it is
entering the system (supply) and negative
if it is exiting the system (demand). If a
node has no supply or demand, it is a zeroflow node (also known as a non-demand
node). There is no reference to flow
through a node; node flow only exists if
it is entering or leaving the system.
Figure 6- 23
296
Chapter 6
6.3.1.2 INSERTING NODES.

nodes are created automatically as you
add facilities to your model. Nodes can also
be inserted manually into an existing pipe
segment, dividing the existing pipe into
two pipes with equal properties (except for
length). In addition, nodes can be inserted
manually at the intersection point of two
pipes selected in the map view. You cannot
insert nodes in non-pipe facilities or in
open spaces on the map.
To insert a node into an existing pipe
1.From the main menu, select Map > Insert
Node.
2. In the map display, move your cursor to
an area where you want to insert a node,
then click to insert the node.
6.3.1.3 EDITING NODES

edit nodes in Synergi by selecting the node
and opening the node editor. The node
editor allows you to edit all settings and
values for the node, such as its status,
coordinates, and flow and pressure values.
Figure 6- 24 node editor
Synergi facilities:
6.3.2 Model facilities in

Synergi
Facilities are components in the system
through which a fluid can flow.
Creating facilities in Synergi is as easy as
selecting the facility type in the equipment
warehouse and then clicking the points in
the map display where you want the fromnode and the to-node to be located. Once
the facility is included in the model, you
can edit, move, or delete it as required.
Pipes
Storage facilities
Processing plants
Loss elements
Forced flow elements
Valves
Regulators
Regulator stations
Compressor stations
6.3.2.1 ADDING A FACILITY

To add a facility, simply select the facility
type in the equipment warehouse, drag it
onto the map, and click the points where
you want the from-node and the to-node
to be located. The facility name will be
automatically created based on your autonaming preferences. The facility
297
Chapter 6
will also inherit any default values that are

defined in the warehouse and your model
settings.
6.3.2.2 EQUIPMENT WAREHOUSE

The equipment warehouse allows you to
create a virtual catalog of pipes, valves,
regulators, fluids, and other equipment
that you plan to use in your model. When
you add facilities to your model, you simply
select the facility that you want from your
warehouse and drag it onto your model.
The new facility in the model will include
any settings that were defined in the
warehouse, including equation types,
diameters, and other values required for
modeling.
The equipment warehouse is stored in a
separate database from your model. You
can modify and use the default warehouse
that is included with Synergi, or you can
create your own warehouse files. The
default warehouse file, saved in the
Defaults directory that is located in your
Personal
Folder,
is
named
WarehouseDefaultG.wh.mdb.
Figure 6- 25 equipment warehouse
6.3.3 Pipes
Pipes are the most basic component of any
system, connecting all of the facilities in
the system into a functional network. In
Synergi, you can select from different pipe
equations to represent the different
characteristics of each pipe in the system.
Different pipe types can be mixed together
in a single modeled system.
All warehouse files are saved with

a.WH.MDB extension.
You can also open a corporate equipment
warehouse. A corporate equipment
warehouse may be opened by any Synergi
user. However, only users with access to
the warehouse password will be able to
create, modify, and delete facilities in the
warehouse.
Synergis pipes are divided into four

categories, as discussed in the topics listed
below. Synergi recommends that you use
the fundamental equations for pipe
modeling, unless you have specific reason
to do otherwise.
298
Chapter 6
The fundamental pipe equations

included with Synergi:
Fundamental pipe with constant

friction factor (FD)
Fundamental
pipe
with
flowdependent friction (FM)
with condensate loss and superheat
(FM).
General pipe (GP)
The Traditional natural gas pipes
IGT distribution pipe (IG)

Mueller pipe (MU)
Panhandle A pipe (PA)
Panhandle B pipe (PB)
Spitzglass low pressure pipe (SL)
Spitzglass high pressure pipe (SP)
Weymouth pipe (WE)
The required data to define a pipe:
Figure 6- 26 pipe editor
Pipe diameter
Pipe length
Pipe roughness
pipe equations
6.3.4 valves
A valve is a facility that has a fixed valve
stem position with varying inlet and outlet
pressures. A valve is not a pressure
regulating or sustaining device.
Valves equations in Synergi:
In Synergi, valves are generally modeled as

one of the following:
Connections between pipelines or

equipment. Again, this is typically done
so that the model can accurately
account for pressure losses that
actually exist in a real system, and
accommodate system cross-overs
and other equipment
Closed facilities, used to isolate or
disassociate one pipeline from another
Pressure loss facilities. The valve

constant, often referred to as the Cg, is
set to a value that causes a pressure
differential across the facility. In this
way, loss facilities can be easily
modeled.
299
General valve
Check valve
Fisher valve.
Mokveld valve.
Reliance valve.
Valtek valve
Chapter 6
The valves defined by
6.3.5
Regulators
regulating stations
Maximum valve constant: Specifies the

maximum value (Cg.Max) for the valve
constant. This value must be greater
than 0 and greater than or equal to the
valve constant (Cg).
Valve constant: Specifies the valve
sizing coefficient (Cg). This value must
be greater than or equal to 0 and less
than the maximum valve constant
(Cg.Max).
and
6.3.5.1 ABOUT REGULATORS AND

REGULATOR STATIONS
A regulator is generally defined as a facility
that has an unknown valve position, with
the upstream or downstream pressure
held constant at a set value.
A regulator station is a single facility that
models a group of individual regulator
units as if they were just one regulator,
with a single unknown (Cg). It can include
up to four regulators in series and up to 50
parallel paths.
6.3.5.2 Differences between a

regulator and a regulator
station
A regulator is a single regulating facility
that exists between two nodes. It adds one
equation to the system of equations, and
therefore is associated with a single
unknown (Cg). Typically, the Cg for an
individual regulator is set as unknown.
This regulator can be modeled using a
number of different regulator equations,
which are discussed in Regulator
Equations
A regulator station, however, is
substantially different than an individual
regulator. Figure (6-28) shows a typical
regulator station. The first regulating path,
referred to in the drawing as Path 1,
contains two regulators. The second
regulating path, referred to as Path 2, also
has two regulators. In most installations
the first regulator in a path is a monitor,
and the second in a path is an operator.
Notice that the dotted lines, which
represent the control points for the
Figure 6- 27 valve editor
300
Chapter 6
regulators, all point to the downstream

side of the station
You must compensate for this new

unknown by setting a known node
pressure in the system. This node is known
as the control node. The control node is
often the regulators to-node, but it can be
any node that can be satisfied by the
regulator.
6.3.5.5 regulator stations

Advantages of modeling
with regulator stations
Figure 6- 28
6.3.5.3 REGULATORS
Adding a regulator: To add a regulator to
your model, select a regulator in your
equipment warehouse, drag it onto the
map, and click the points where you want
the from-node and the to-node to be
located. For detailed information on
adding facilities
Regulator station capacity

calculations
Regulator station operation
Note: After adding a regulator, you can

double-click it to open the regulator editor.
For more information on editing regulators
6.3.5.5.1 Advantages of modeling with
Editing a regulator: (Service state Reference flow-Minimum pressure DropInlet temperature-Travel time-Regulator
equation-Pressure
Profile-Maximum
flowrate)
regulator stations
Regulator stations provide a number of
advantages over modeling individual
regulators directly in the Synergi model.
6.3.5.4 Solving for an unknown

regulator constant (Cg)
When modeling regulators, you can specify

the regulator constant as an unknown
value. When you run an analysis, Synergi
determines the minimum appropriate
position for the regulator that satisfies
pressures and flows in the model. Since the
regulator constant is being solved for by
Synergi, this represents an additional
unknown value in the system of equations.
Include multiple regulators as a single

facility, thus adding only one unknown
to the system of equations.
Provide much more comprehensive
reporting, including intermediate
temperatures and pressures.
Converge in fewer iterations than

individual regulator facilities (although
each iteration may take longer) during
a steady-state analysis.
301
Chapter 6
failed in an open position, in which case it

rises to the next available set point
6.3.5.5.2 Regulator station capacity

calculations
6.3.5.5.4 Regulator station equations
Regulator station reporting is based on

capacity, as opposed to actual,
calculations. Regulator station capacity
calculations assume that both upstream
and downstream pressures are known, and
that valves are open if their pressure
settings permit. The total station capacity
is the sum of the capacities of each
individual path.
If you have access to data for specific

regulator brands, it is best to model them
using the actual manufacturers equations.
Manufacturer regulator equations that are
included with Synergi include, American
Meter Axial Flow, Mokveld, Valtek,
Reliance, Fisher, and Grove regulators, All
required data, however, including both the
regulator brands and the station
configurations, are not always readily
available. Therefore, a generic regulator
facility is often desirable. The Reliance
regulator equation is recommended for
generic modeling needs.
The capacity of a path of serial valves is

determined by a trial and error process, by
which
intermediate
pressure
and
temperatures are selected that balance
the flows among the serial valves. Since the
capacity of a path is limited by the lowest
flow through any valve, it may be shown
mathematically that the condition of
balanced flow does correspond to the path
capacity if the system is a monitor
regulator system.
6.3.5.5.3 Regulator station operation
The operation of two or more regulators in
a series is usually characterized in a
monitor sense. Only one regulator is
performing the pressure control function.
The additional regulator serves as a
monitor that is wide-open under normal
conditions and only regulates if the main
regulator fails in an open position. The
monitor may be either upstream or
downstream of the main regulator.
Regulator stations with upstream monitors
are modeled by setting the pressure set
point for the monitor regulator somewhat
higher than that of the main regulator,
thereby forcing it to stay open. The
resultant set point for the series path is
then the lowest set point of any regulator
within the path, unless that regulator has
Figure 6- 29 regulator equations in synergi

gas 4.8
302
Chapter 6
6.3.5.6 Editing regulators

To add regulator
6.3.5.7 Editing Regulator Station
From
Warehouse>Regulators>Sample
(fisher, General, Reliance) regulator
To
add
regulator
station
Warehouse>Regulator
station>Sample
regulator station
Setting pressure is the same as regulators
but to edit the entire station click on
Figure 6- 30 default regulators in synergi gas

4.8
To set the required pressure

From controls <Set pressure <Control type
(Pressure control, flow control, regulator
constant)
Control mode (UP, Down)
Figure 6- 32
Then add the required number and types

of regulators by the ware house
Adding path by right click on the window
Figure 6- 31 regulator editor
303
Chapter 6
units in series. Each unit must include a

driver and compressor, and can optionally
include a cooler. Each compressor can be
the same or a different type, as required.
6.3.6.2 ENHANCED COMPRESSOR

STATION ANALYSIS
The Enhanced mode features could
provide you with different results than
previous Synergi versions. While these
differences mean greater accuracy and
precision in your compressor station
modeling. Some of the new modeling
features that are only available in the
Enhanced compressor station analysis
mode are as follows:
Figure 6- 33 regulator station layout
Then set the maximum constant for each

regulator
In Enhanced mode, recycle lines can be

added to provide for unit, path, or station
recycling. In Standard mode, virtual
recycling is performed at each unit.
In Enhanced mode, cooler units can be
placed in compressor station assemblies,
at the end of compressor station paths,
and on the discharge side of the
compressor station. Path coolers can also
be shared on up to 10 compressor station
paths, or a single cooler can serve an
entire station. In Standard mode, cooler
units can only be placed in compressor
station assemblies, and the coolers cannot
be shared on multiple paths nor assigned
to an entire station.
In Enhanced mode, you can use either
Figure 6- 34
generic or aerial coolers, or a mix of both.

In Standard mode, only generic cooler
units may be used.
6.3.6 Compressor Station

In Enhanced mode, temperature limits

and pressure losses can be modeled for all
cooler types. These options are available
on the Temperature Limits and Pressure
Loss tabs of the cooler editor.
The compressor station is modeled as a

single facility with a single associated
unknown. However, the compressor
station can include as many as 50 parallel
paths, with each path including up to four
304
Chapter 6
In Enhanced mode, both flow-dependent

and constant pressure losses may be
added at the station level. In Standard
mode, only constant losses may be
specified at the station level.
General tab of the unit editor). By default,

these colors are dark blue (Available),
light blue (Running), yellow (Idle),
bright green (Online), and red
(Offline). Like most map display symbols,
you can change these colors in the
Symbology Settings dialog box.
In Enhanced mode, unified fuel treatment

can be enabled to account for different
fuel types used to operate a compressor
station.
6.3.6.3 UNIFIED FUEL TREATMENT

Unified Fuel Treatment is a method for
accounting different fuel types used to
operate a compressor station, such as
natural gas, electricity, and other fuel
types. This method is available only in
conjunction with Enhanced Compressor
Station Analysis.
Calculating fuel for gas drives (gas
turbines/engines) is a simple matter of
plugging user defined data into a formula.
Fuel in gas drives is either a function of
power, speed, or both. Expressed in
standard metric units, but easily
envisioned in any set of units desired,
when given steady-state conditions:
$
gas (
(
6.3.6.5 Compressor station capacity

calculations
Compressor station reporting is based on
capacity calculations rather than flow rate
calculations. Compressor station capacity
calculations assume that both upstream
and downstream pressures are known
(either specified by the user or solved for
by the application), and that the
compressors are running, if possible. The
total station capacity is the sum of the
capacities of each path.
)
$
gas (
) = (
Figure 6- 35
) = (
( )
electric (
(
) = 24 (
) ()
6.3.6.4 Compressor station map

display symbols and colors
The capacity of a path of series

compressors is determined by a trial and
error process of selecting intermediate
pressure and temperatures to balance the
flows among the series compressors.
The symbols and colors that you see in the

compressor station map display are
controlled through Synergis symbology
settings. One of the most noticeable color
differences that you will see is in the unit
colors. Unit colors reflect the current
status of the unit (which is set in the
In
ambiguous
situations,
Synergi
determines which path or series of paths
should be operating. For example, when
305
Chapter 6
flow demand cannot be met by operating

one path only, Synergi may specify splitting
the flow among multiple paths, running
one path at capacity and another at
reduced flow.
6.3.6.7 Editing compressors

To edit a compressor, simply double-click
the compressor in the station map display
or in any list that includes the compressor
name, then use the compressor editor to
edit the settings for the compressor, as
appropriate. The options in the
compressor editor will change depending
on the type of compressor that you are
editing. Refer to the topics listed below for
more information.
6.3.6.6 Drivers
Drivers are one of two required
components for a compressor station unit.
(The other required component is the
compressor itself).
Editing drivers: TO EDIT A DRIVER
Double-click a driver in the map display

to open the driver editor.
Use the Identification area at the top of
the driver editor to identify the name,
description, and map display symbol
for the driver.
Use the Ratings tab to specify the
minimum and maximum rated power,
the rated ambient temperature, and
the rated elevation for the driver.
Use the Fuel tab to specify the method
used to calculate fuel, specify whether
the fuel calculations are based on a
thermal or volumetric flow, and enter
the required coefficients for the
selected calculation method.
Use the De-rating tab to set (Maximum
temperature, Minimum temperature,
Temperature and Power Data)
If you are editing a centrifugal
compressor, use the Maximum Power
tab to edit (Power as a function of
compressor speed and temperature ,
Normalize using the following speed ,
Power, Speed and Temperature) Note
:The Maximum Power tab is not active
for the reciprocating, screw, or
theoretical driver types
Minimum Power tab is similar to
Maximum power tab
Click OK.
Theoretical
power/flow
(TH)
compressor
Actual power/flow theoretical form
(HP) compressor
Actual ratio/flow polynomial form (C2)
compressor
N-stage
reciprocating
(RN)
compressors
6.3.6.7.1Theoretical power/flow (TH)

compressor
The TH compressor is the most basic
compressor model used with Synergi.
This equation derivation, shown below,
follows the standard theoretical power
equation.
3.0303
[( )
306
1]
HP = compressor power ( power)

Q = flow rate through compressor
(mmscfd)
ec = compressor efficiency
(dimensionless)
Zs = gas compressibility at suction
conditions (dimensionless)
b = base pressure (psia)
Ts = gas flowing temperature at
suction conditions (R)
Tb = system base temperature (R)
Chapter 6
K = gas coefficient (dimensionless)

Pd = discharge pressure (psia)
Ps = suction pressure (psia)
HP = compressor power( power)

(mmscfd)
0 , 1 , 2 , 3, 4=Polynomials
coefficients PowerFlowrate
R
=
Compression
Ratio
Dimensionless
Z = Gas compressibility factor
Dimensionless
6.3.6.7.2 Actual power/flow theoretical
form (HP) compressor

This general-purpose equation is typically
used to model new or proposed
compression requirements, to model a
more complex existing compressor
(centrifugal or reciprocating, for example)
for which full manufacturers data is
unavailable or existing compressors of a
type not found in Synergi. The derivation
shown follows the standard theoretical
power equation.
6.3.6.7.4 N-stage reciprocating (RN)

compressors
The RN compressor models an n-stage,
single-shaft reciprocating compressor unit.
The capacity of the unit is defined as being
either the capacity of the unit to perform
work on the gas itself (power capacity), or
the capacity of the cylinders to physically
move the gas (volumetric capacity).
= [1 ( )3 2 ]( )
HP = compressor power( power)

(mmscfd)
ec = compressor
(dimensionless)
Zs = gas compressibility at suction

conditions (dimensionless)
b = base pressure (psia)
Ts = gas flowing temperature at

suction conditions (R)
efficiency
Figure 6- 36
6.3.6.7.5 N-stage reciprocating (RN)

compressors
Power capacity equation for the RN
compressor
Tb = system base temperature (R)
K = gas coefficient (dimensionless)
Pd = discharge pressure (psia)
Ps = suction pressure (psia)
[( )
= 3.0303
1
1]
6.3.6.7.3 Actual ratio/flow polynomial
Ec : Compression efficiency
form (C2) compressor
Em : Mechanical efficiency
= (0 + 1 + 2 2 + 3 3 + 4 4 )
( ) 1
Zs : supercompressibility factor at stages

suction conditions
307
Chapter 6
Zav : supercompressibility factor at stages

suction conditions at average
map display symbol for the generic

cooler
6.3.6.8 Coolers
Coolers are optional and are not required
to perform an analysis. Synergi offers two
types of coolers for compressor station
modeling: generic coolers (the traditional
cooling that has been available in Synergi)
and detailed aerial coolers. Detailed aerial
coolers are only available in Enhanced
compressor station mode when you have
selected either Aerial Cooler Only or Mixed
Cooler as the Cooler Preference in the
compression controls.
1. Cooling
method
(Delta
temperature & Target temperature
& Delta heat removed)
2. Heat removal rate or Degrees of
cooling
3. 3. Use Suction Temperature : This
option is available if you selected
Target Temperature as the Cooling
method
4. 4.
Maximum
Temperature:
Specifies the limiting output
temperature. This option is
available if you selected Target
Temperature as the Cooling
method. The temperature of the
gas discharged from an individual
compressor is given by:
In Standard compressor station mode,

generic coolers can only be added at the
discharge side of a compressor station
assembly. In Enhanced compressor station
mode, either generic or detailed aerial
coolers can be located in any of the
following positions:
Assembly coolers are located at the

discharge side of a compressor
station assembly, after the driver
and compressors.
Path coolers are located at the

station path.
Station coolers are located at the

station.
Use the General tab to select the

cooling method used by the generic
cooler:
= +
1
( ) 1
= [
]
Use the Pressure Loss tab to specify

optional pressure losses associated with
the generic cooler.
6.3.6.8.1 Editing coolers
Note: This tab is only available in the

Enhanced compressor station mode.
Generic coolers
Double-click a generic cooler in the

map display to open the generic
cooler editor.
1. Method : -Constant delta P The

pressure loss is constant based on
the following equation:
Use the Identification area at the

top of the generic cooler editor to
identify the name, description, and
Ps = Pin S drop
-Flow dependent pressure drop
308
Chapter 6
The pressure loss is flow dependent based

on the following equation:
1. Number of bays in service

2. Number of passes per bay
3. Number of fans per bay
=
= 2 +
4. Outside area correction factor
2
(. 000024 )2
5. Number of tubes per bay

6. Length
= ( , 11.2 )
7. Wall thickness
Where Cgs is consistent with:
8. Outside diameter
Use the Design Conditions tab to specify
rated conditions that help determine how
the aerial cooler will operate under
specified conditions.
= .000024 2 2
2. Pressure drop
3. Enable constraints
Note: When modeling with aerial coolers,

good design conditions data is necessary to
generate good results
4. Minimum loss
5. Maximum loss
1.
2.
3.
4.
5.
6.
7.
6. Diameter
Use the Temperature Limits tab to

set the following options
-Minimum Temperature Limit
-Use Suction Temperature
-Minimum Temperature
Use the Thermal tab to specify the

following options:
-Minimum Temperature above air

Temperature
1. Temperature correction factor
Note: This tab is only available in the

Enhanced compressor station mode and
when the Cooling Method is not set to
Gas pressure inlet

Pressure drop
Ambient air temperature
Gas Temperature Inlet
Gas Temperature outlet
Air mass
Gas mass
2. Thermal conductivity
3. Inside Heat Transfer coefficient
4 .Outside Heat Transfer coefficient
Click OK
5. Fouling Resistance gas side
6.3.6.8.2 Aerial coolers

TO EDIT AN AERIAL COOLER
6. Fouling Resistance air side
Double-click an aerial cooler in the map

display to open the aerial cooler editor
Aerial coolers
Use the Bypass tab to set the following

options:
Use the General tab to specify the

following options:
1. Method
309
Chapter 6
2. Valve constant
6.3.6.11 Compressor Station Pinch

Method
3. Maximum valve constant
Allow Pinch: Synergi presumes the

existence of a pinch valve on the suction
side of the compressor assembly. If the
necessary operating point is in the sonic
region, Synergi will begin to close this
virtual valve. This decreases the suction
flange pressure at the compressor, and
thus increases the head across the
machine, until the operating point is on the
sonic limit.
4. Maximum delta pressure

Use the Temperature Limits tab to set the
following options:
1.
2.
3.
4.
Minimum Temperature Limit

Use Suction Temperature
Minimum temperature
Minimum temperature above air
temperature
5. Minimum allowable discharge
Allow Sonic: When Allow Sonic is selected,

Synergi will linearly interpolate a line of
constant speed from the sonic line
downward to zero head, thus allowing
operation in the sonic region. Efficiencies
beyond those input are held constant.
6.3.6.9 Editing compression control

Edit > Model setting > Compression
controls
As set at individual stations: When this

option is selected, Synergi uses the station
pinch option set at individual stations.
Figure 6- 37
6.3.6.10 The difference between

Enhanced & Standard compressor
station analysis
The Enhanced mode features could
provide you with different results than
previous Synergi versions. While these
differences mean greater accuracy and
precision in your compressor station
modeling
Recycle Lines
cooler units & path coolers & type
unified fuel treatment
Figure 6- 38
Figure 6- 39
310
Chapter 6
Percent load (max rated power): If this

option is selected, Synergi calculates
centrifugal driver load based on the ratio
of a drivers utilized power to its maximum
rated power.
6.3.6.12 Cooler Preference

Aerial Cooler Only: When Aerial Cooler
Only is selected, Synergi will use only the
aerial coolers it finds at compressor
stations. It will ignore any generic coolers.
If no aerial coolers are found, no cooling
will be calculated.
Power: If this option is selected, Synergi

calculates centrifugal driver load based on
a drivers utilized power.
Generic Cooler Only: When Generic Cooler

Only is selected, Synergi will use only the
generic coolers it finds at compressor
stations. It will ignore any aerial coolers. If
no generic coolers are found, no cooling
will be calculated.
6.3.6.15 Allow Unified Fuel & Fuel

option
Mixed Coolers: When Mixed Coolers is

selected, Synergi will look at each
compressor station and assembly, and use
whatever coolers it finds there. If both
aerial and generic coolers are found, aerial
coolers will be used.
Figure 6- 40
6.3.6.13 Allow Internal Compressor

Constraint Interchange
table 6- 5
Figure 6- 41
6.3.6.16 Digitizer
How to define Compressors performance
curve?
The default compressor station is using a
theoretical compressor unit
The required compressor unit is centrifugal
not theoretical
6.3.6.14 Centrifugal driver load

calculation method
So Performance of compressor is required

to be defined by defining flow, head, and
efficiency data
Percent load (max available power): If this

option is selected, Synergi calculates
centrifugal driver load based on the ratio
of a drivers utilized power to its maximum
available power.
Digitizer is used to define flow, head, and

efficiency data
311
Chapter 6
6.3.6.16.2 Creating Project

Select the suitable image for performance
curve
Select the suitable template: Comp
template
Figure 6- 42
6.3.6.16.1 Terminologies
Figure 6- 43
Digitizer enables you to digitize data points

onto digital representation of a chart or
map, and infer the corresponding X and Y
coordinate values of those points. After
scanning a paper copy of a chart or map to
a digital image and importing that image
into Digitizer, you can easily define the
axes, variables, and series represented in
the image, and interpolate the values of
data points based on those axes, variables,
and series.
Figure 6- 44
312
Chapter 6
the project explorer. For more information

about variables in Digitizer
Figure 6- 45
Axis: An axis refers to a horizontal or

vertical line used as a frame of reference
for measurement. Axes for a Digitizer
project are defined and listed on the Axes
tab of the project explorer. Axes are
digitized on the Digitizer canvas when
points are digitized at the axis minimum
and maximum limits. For more information
about axes in Digitizer
Figure 6- 47
Series: refers to a set of data points

comprising values associated with two or
more variables one corresponding to the
X axis value, and one corresponding to the
Y axis value. Each series references two or
more variables, which can derive their
values from a number of different sources,
including an axis, a prompt, a constant
value, or a series. Digitizer supports the
definition of an unlimited number of series
in each project. Series for a Digitizer
project are defined and listed on the Series
tab of the project explorer. For more
information about series in Digitizer
Figure 6- 46
A variable: refers to an item associated

with a value and whose value may vary. A
variable in Digitizer derives its value from
one of four sources: interpolation from
specified axes, user input, constant value,
or value of a series that references the
variable. Digitizer supports the definition
of an unlimited number of variables in each
project. Variables for a Digitizer project are
defined and listed on the Variables tab of
Figure 6- 48
313
Chapter 6
6.3.6.16.3 Data
To get the points of variables
Figure 6- 51
6.3.6.16.5 Importing
Figure 6- 49
Then import the CSV file so the

performance curve of the centrifugal
compressor has been defined
6.3.6.16.4 Exporting
Export CSV file for the performance curve
and the data obtained
Figure 6- 52
6.4 Output From Synergi

In order to discuss the model output, we
should first understand the output tools of
Synergi such as reports and charts.
6.4.1 Reports
Reporting provides detailed information
about your model in a format that is easy
to read and use. Reports are generated in
either an HTML or a table-based format
and can be saved to different formats,
based on the type of report. In addition to
the different report types that can be
generated in the Report window, Synergi
can also create certain analysis based
reports, which are saved as text files in
your personal Synergi directory.
Figure 6- 50
To choose the file formula

Delimiter: System default
Options: Export column headers
Synergis reports include several visual

conventions that will help signify data that
may be of particular interest
314
Chapter 6
6.4.1.1 Using the reports windows
6.4.1.1.1 Compressor report
The Reports window is a separate tabbed

window within Synergi that allows you to
view different reports relating to your
model. The Reports window is divided into
two main areas: the display area, where
you view.
The Compressor report lists each

compressor in the model and engine
details. The data displayed is dependent on
the engine type within the Compressor
Station facility.
6.4.1.1.2 Demand Contribution report
You can create and edit Synergi reports,

they have several types:
The Demand Contribution report displays

the following data for nodes and facilities
in your model: The Demand Contribution
report displays the following data for nodes
and facilities in your model:
Compressor report
Demand Contribution report
Dew Point report
Exception report
Full report
Hydrate report
Length and Diameter report
Model report
Regulator report
Storage Field report
We will discuss them is details
Demand nodes in the model and

the demand at each node.
Contributing source nodes and the
percentage of each nodes flow
present at the demand node.
Contributing pipes and the
percentage of each pipes flow
contributing to the demand. In
addition, the flow, length,
diameter, and roughness for each
contributing pipe.
6.4.1.1.3 Dew Point report

The Dew Point report displays dew point
temperature and location information for
your model. The report is divided into
subsections that are available as tabs along
the bottom of the report. The subsections
that are included in the report are Nodes
and Pipes.
6.4.1.1.4 Exception report
The Exception report lists nodes and
facilities in the model that have the highest
and lowest values for various attributes,
such as pressures and flows. Numerous
subsections are available within the
Exception report.
Figure 6- 53 report window
315
Chapter 6
6.4.1.1.5 Full report

The Full report displays detailed
information on specified nodes and
facilities in your model. The report is
divided into subsections that are available
as tabs along the bottom of the report. The
subsections that are included in this report
are as follows:
6.4.1.1.9 Regulator report

The Regulator report displays detailed
information on specified regulators and
regulator stations in your model. Regulator
stations are displayed as hyperlinked text
in the report.
Nodes
Pipes
Looped pipes
Valves
Regulators
Compressors
Storage fields
Forced flow elements
Loss elements
Processing plants
6.4.1.1.10 Storage Field report

For each specified storage field in the
model, the Storage field report provides
the name and pressure of the observation
point (the from-node) and the wellhead
(the to-node), as well as the flow rate, the
drawdown node name and percentage,
and any violations.
6.4.1.2 Creating and editing

customized reports
6.4.1.1.6 Hydrate report

The Hydrate report displays hydrate
temperature as a function of temperature
and pressure for nodes throughout your
model. For each node in your model, the
Hydrate report displays the result pressure
and temperature after you perform an
analysis, as well as the calculated hydrate
temperature at that node.
Synergi allows you to create your own

reports with customized content based on
your specifications. For each custom
report, you can designate the specific
attribute data that you wish to display. You
can adjust decimal precision, text
alignment, sorting order, and other such
items affecting how the data is displayed in
the report. You can also specify the report
scope whether the report includes data
from the entire model or just a portion of
the model, such as from a partial set or the
Query tab. In addition, you can designate
whether to include statistics data at the
end of a report.
6.4.1.1.7 Length and Diameter report

The Length and diameter report provides
information on the length and diameter of
pipes in your system.
6.4.1.1.8 Model report
The Model report displays data for the
systems supply and demand and general
overall status, such as the number of
facilities and types in your system. The
subsections that make up the Model
report are as follows:
Demand
Supply
Fuel and Power
6.4.1.2.1 Saving reports

Synergi reports can be saved in several
different formats, including CSV, HTML,
and XML.
Status Summary
316
Chapter 6
your models connected network. This

path is specified by selecting two or more
nodes to form a node chain, which
Synergi will trace each time a distance
chart is generated. You can add multiple
variables as Y axes in the chart. For
example, you can display the pressure,
maximum pressure, and minimum
pressure for your chain of nodes, and also
the elevation for the same chain of nodes.
The resulting chart will show you how the
pressure changes over the distance of the
specified chain of nodes, how those
changes relate to elevation at the same
node points, and how close the pressures
approach your maximum and minimum
values.
6.4.2 Charts
Chart Viewer allows you to view different
charts relating to a Synergi model. Charts
can be created and viewed using the Chart
tab within Synergi and also using the
standalone Chart Viewer application.
6.4.2.1 Chart types

The types of charts available on Synergi
such as:
Distance charts
Time charts
Centrifugal compressor maps
those changes relate to elevation at the

same node points, and how close the
pressures approach your maximum and
minimum values.
6.4.2.2 Saving charts and chart data

You can save charts for use in Synergi or to
external JPG, BMP, or PNG files, and copy
charts and paste them directly in other
software products. Saved charts include
not only the chart itself, but also the title,
legend, and other formatting that you have
set up for the chart.
In addition to saving charts, you can export
chart data, including data point values and
the facilities represented in a chart, to
Microsoft Excel and to comma separated
value (CSV) files, and copy chart data to
your computers clipboard. You can also
use a charts data points to generate a
profile for use in Synergi.
Figure 6- 54 chart window
Distance charts
You can copy chart facilities and chart data

for use in another program, export
facilities and data to CSV files or to Excel,
A distance chart shows changes in

variables such as pressure or elevation
over the distance of a path traced through
317
Chapter 6
and save chart data as a profile. Refer to

the topics listed below for more
information.
Figure 6- 55 centrifugal compressor performance curve saved from synergi gas 4.8
continue the analysis, or to end the

analysis.
6.4.3 steady state analysis

During a steady-state analysis, Synergi
attempts to solve for unknown pressures,
flows, and other facility unknowns to
within set tolerances. After the analysis is
complete, Synergi displays an Analysis log
that contains simulation results. If your
model is unsolved or infeasible, you can
preview the Analysis log to identify
problems and then run another simulation.
If your model is solved, you can view
Synergi reports to view the results of your
analysis.
6.4.3.1 PERFORMING A STEADYSTATE ANALYSIS

From the main menu, select Analysis >
Steady-State.
Synergi notifies you whether a solution has
been obtained.
6.4.3.2 OUT OF TOLERANCE

ASSISTANT
When Synergi has completed the specified
number of iterations without solving all
unknown values to within their tolerance
limits, Synergi displays an Out of Tolerance
Assistant dialog box. The Out of Tolerance
Assistant displays the number of variables
that are out of tolerance
In steady-state analysis, variables are

assumed to remain constant with time,
that is, in a steady condition. Models
withvariables that are out of tolerance are
detected during the analysis. When
and prompts you to continue the analysis

with more iterations or end the analysis.
Synergi has completed the specified

number of iterations and not all variables
are solved within tolerance, Synergi
prompts you to set addition iterations and
318
Chapter 6
Figure 6- 56 out of tolerance assistant box
319
Chapter 6
Figure 6- 57 NGNG after steady state analysis
320
Chapter 6
Figure 6- 58 delta section of the NGNG
Figure 6- 59 Alexandria section of the NGNG
321
Chapter 6
Figure 6- 60 Cairo section of the NGNG
Figure 6- 61 suez section of the NGNG
322
Chapter 6
6.4.3.3 output of steady state

analysis
After modeling the natural gas network
with its facilities on Synergi and inserting
hydraulic data for nodes and facilities and
running the model steady state analysis we
can now get the output from the model.
At steady state analysis the output is:
The supply nodes pressure

The demand nodes pressure
Regulator station report
Detailed reports
6.4.3.3.1 Supply nodes pressures

table 6- 6 supply result pressure
Name
Result pressure
Ras Shouqir
37.3
North Sinai
41.1
WDGC
46.9
Dahshour
27.5
Ras bakr
37.5
LPG(badreldin)
45.1
Aboqir
35.1
Rashpetco
39.5
Abo madi
35.4
UGDC(port fouad)
44.6
Elmansoura
25.5
Ras shouqir
37.3
North sinai
41.1
323
Chapter 6
6.4.3.3.2 Demand nodes pressure:

6.4.3.3.2.1 Power stations
table 6- 7
power stations result pressure
No
Name
Flow
SM^3/h
Result pressure
ps1
talkha steam
-51000.0
9.00
ps2
shobra el khema
steam
krir 1,2 steam
-33000.0
9.00
-42000.0
9.00
-138000.0
9.00
ps5
intergen 3,4
steam
abo qir steam
-87000.0
9.00
ps6
abo qir new
-51000.0
9.00
ps7
-74000.0
9.00
ps8
kafr el dawar
steam
west cario steam
-63000.0
9.00
ps9
teppin steam
-14000.0
9.00
ps10
korymat steam
-127000.0
9.00
ps11
-180000.0
9.00
ps12
east port said

steam
mousa steam
-156000.0
9.00
ps13
ataka steam
-46000.0
9.00
ps14
west suez steam
-153000.0
9.00
ps15
al arish steam
-19000.0
9.00
ps16
abo sultan steam
-28000.0
9.00
psg1
el attf new
combined
damietta
combined
new damietta gas
-138000.0
27.00
-208000.0
18.49
-138000.0
18.62
ps3
ps4
psg2
psg3
324
Chapter 6
No
Name
Flow
SM^3/h
Result pressure
psg4
west damietta gas
-140000.0
27.00
psg5
talkha combined
-125000.0
23.32
psg6
talkha gas
-64000.0
23.70
psg7
nubaria combined
-380000.0
17.14
psg8
banha gas
-95000.0
23.31
psg9
krir 5,6 combined
-90000.0
27.00
psg10
noth cairo
combined
-168000.0
22.39
psg11
6 october gas0
-150000.0
21.79
psg12
south cairo
combined
-56000.0
23.31
psg13
wadi hoof gas
-11000.0
23.44
psg14
korymat comb
-123000.0
27.00
psg15
elshabab gas
-30000.0
27.00
psg16
port said gas
-14000.0
27.00
psg17
elsyof gas
-50000.0
27.00
-46000.0
27.00
-68000.0
27.00
psg18
psg19
damanhor
combined
el mahmoudia
combined
325
Chapter 6
6.4.3.3.2.2 Industrial and local result pressure

table 6- 8 industrial and local result pressure
No
Name
Flow
SM^3/h
Result pressure
in1
natgas borg el
arab
-34620.0
7.00
in2
natgas agamy
-9920.0
7.00
in3
natgas west alex
-14400.0
7.00
in4
town gas alex
-15000.0
35.26
in5
natgas kafr
eldawar local
-5600.0
4.00
in6
town gas portsaid
-16000.0
42.76
in7
transgas delta
-13000.0
25.23
in8
egypt gas delta(1)
-40000.0
22.54
in9
quesna(egyptgas)
-2700.0
23.60
in10
egyptgas delta(2)
-40000.0
21.46
in11
egyptgas abo
zabal
-15400.0
23.62
in12
egyptgas cairo 1
-45000.0
23.11
in13
natgas cario
-15000.0
23.93
in14
towngas cairo
-72000.0
21.39
in15
natgas 6 october
-35000.0
25.99
in16
egyptgas cairo 2
-45000.0
32.25
in17
towngas el
ismailia
-5000.0
38.78
in18
city gas
-124000.0
28.03
in19
midor and
medalic
-64500.0
38.69
in20
star nex
-2000.0
7.00
in21
misr chemical
-6100.0
7.00
in22
alex petrolum
-19700.0
7.00
in23
amoic
-10500.0
42.92
in24
anrpc
-6770.0
43.12
in25
polyprobelin
-21800.0
43.37
326
Chapter 6
No
Name
Flow
SM^3/h
Result pressure
in26
elmogammaa for
petrol
-46000.0
40.49
in27
amryia petroleum
-26600.0
43.17
in28
misr petroluem
-160.0
44.76
in29
sidpec
-7000.0
46.70
in30
petrochemical
-27000.0
45.46
in31
carbon black
factory
-7500.0
46.45
in32
fenosa facilities
-4000.0
40.11
in33
methanics
-23000.0
39.90
in34
morgan
-31000.0
36.54
in35
gapco
-7000.0
36.85
in36
suez petroluem
-11500.0
28.17
in37
elnasr petroluem
-7500.0
26.86
in38
gas derivatives
-40000.0
26.55
in39
abo qir fert
-129800.0
33.61
in40
alexandria fert
-40000.0
31.26
in41
uoria and natarat
-72600.0
24.28
in42
helwan fert
-50000.0
24.53
in43
egypt for fert2
-42000.0
33.37
in44
suez for fert
-15000.0
33.28
in45
amrya cement
-2720.0
42.14
in46
dkhela steel
-80000.0
7.00
327
Chapter 6
No
Name
Flow
SM^3/h
Result pressure
in47
max cement
-17550.0
7.00
in48
ganob elwadi
cement
-6000.0
25.10
in49
katmya cement
-7400.0
23.21
in50
helwan cement
-20000.0
23.47
in51
national cement
-37900.0
23.42
in52
kamaen el top
-75000.0
22.31
in53
towngas iron and

steel
-35000.0
24.32
in54
swedy cement
-1000.0
34.12
in55
suez cement
-1000.0
34.12
in56
arabya cement
-1000.0
33.70
in57
egypt cement
-5000.0
33.62
in58
alex free zone
-1870.0
42.21
in59
alex sugar
-31200.0
7.00
in60
nubaria sugar
-18030.0
7.00
in61
alex IA
-18420.0
32.20
in62
dakhlia sugar
-26000.0
39.14
in63
repco gas
-2800.0
24.79
in64
damietta pump
station
-4400.0
40.24
in65
gastec talkha
-800.0
25.43
in66
gazl el mahala
-19500.0
25.00
in67
cebaghet el
mahla
-2800.0
25.24
328
Chapter 6
No
Name
Flow
SM^3/h
Result pressure
in68
10 of ram
-95000.0
24.64
in69
hawamdya sugar
-13800.0
22.99
in71
fayoum gas
oftake
-22300.0
27.76
in72
nile vally oftake
-14000.0
41.93
in74
epec
-45000.0
33.23
in75
gastec suez
-400.0
27.35
-15000.0
41.08
-5000.0
41.57
in76
in78
sinai gas( cement

and IA)
wadi elnile
cement
in79
bani swaif
-5000.0
41.12
in80
upper egypt IA
-5000.0
41.10
in81
bani swaif cement
-5000.0
41.07
in82
elmania
-10000.0
37.22
in83
elmania AI
-1000.0
36.94
in84
abo qurqas
-6000.0
36.62
in85
asyout
-10000.0
33.08
in86
naga hamadi
-11500.0
27.37
in87
qena
-22800.0
24.44
in88
qift
-17700.0
21.13
in89
aswan
-30000.0
14.24
in90
egyptgas aswan
-10000.0
12.80
in91
kema fert
-185000.0
10.00
329
Chapter 6
No
Name
Flow
SM^3/h
Result pressure
in101
local abu homos
-800.0
4.00
in102
local damanhur
-9000.0
4.00
in103
local
elmahmoudia
-4420.0
6.00
in104
local kom hamada
-2410.0
7.00
in105
local rashid
-2810.0
7.00
-1970.0
7.00
-5000.0
7.00
local wadi
elnatron
sodum
carbonat(solvay)
in106
in107
6.4.3.3.3 Pipes Report

table 6- 9 pipes full report
Name
Diameter
(in)
Length
(km)
Flow
(sm^3/hr)
FromNode
Name
FromNode
Pressure
To-Node
Name
To-Node
Pressure
Pi1
Pi2
Pi3
Pi5
Pi4
Pi7
Pi6
Pi8
Pi9
Pi10
Pi12
Pi13
Pi15
Pi14
Pi16
Pi17
Pi18
Pi19
Pi20
24
24
24
24
24
24
16
8
24
8
24
24
24
8
24
10
10
24
6
14
18
13
1
1
1
1
2
2
1
3
1
9
1
11
1
1
3
1
-2720
37340
270000
138000
90000
228000
42000
64500
101840
1870
113768
373710
103848
-9920
-269862
-2000
-80000
195768
-27000
in45
No3
No5
No7
No7
No9
No9
No3
No12
No12
No14
No15
No17
No19
No15
No29
No23
No22
in30
42.14
42.18
42.22
40.85
40.85
40.93
40.93
42.18
42.21
42.21
42.6
42.41
42.54
42.5
42.41
42.65
41.79
42.8
45.46
No2
No2
No9
No329
No348
No7
No328
in19
No3
in58
No17
No5
No15
No17
No18
No20
No14
No20
No24
42.14
42.14
40.93
40.83
40.84
40.85
40.91
38.69
42.18
42.21
42.54
42.22
42.41
42.54
43.47
42.65
42.6
42.65
46.66
330
Result
Downstre
am
Velocity
(m/s)
0.1
0.8
5.6
2.9
1.9
4.8
2
12.9
2.1
0.3
2.3
7.6
2.1
1.8
5.4
0.2
9.4
3.9
8.1
Chapter 6
Name
Diameter
(in)
Length
(km)
Flow
(sm^3/hr)
FromNode
Name
FromNode
Pressure
To-Node
Name
To-Node
Pressure
Pi21
Pi22
Pi23
Pi24
Pi25
Pi26
Pi27
Pi29
Pi28
Pi30
Pi31
Pi32
Pi33
Pi34
Pi35
Pi37
Pi36
Pi39
Pi38
Pi40
Pi41
Pi42
Pi43
Pi45
Pi44
Pi47
Pi46
Pi49
Pi50
Pi51
Pi52
Pi54
Pi53
Pi55
Pi56
Pi58
Pi57
Pi59
6
18
32
8
6
24
24
32
6
6
6
6
24
24
24
18
6
18
6
18
24
6
24
18
6
32
10
18
16
8
18
18
8
16
8
8
4
28
1
1
1
1
1
1
2
17
1
1
1
1
3
1
1
12
1
2
1
10
2
1
2
1
1
39
1
10
5
9
3
4
2
4
4
9
1
8
-7000
-108304
690227
-6100
-17550
224418
206868
648727
-7500
-19700
-10500
-6770
261388
254618
244118
-181064
-160
-180904
-26600
248463
630000
21800
-531250
-154304
-46000
617527
-31200
234063
50000
-15000
184063
145043
-18420
-129800
-40000
-24020
-5600
268500
in29
d1
s14
No25
No30
No33
No32
No35
in31
No27
in23
in24
No18
No40
No38
No42
in28
No44
in27
d1
s2
d1
No18
No49
in26
No51
No28
No53
No45
in4
No45
No59
in61
in39
in40
No64
No13
s3
46.7
44.19
46.91
42.79
42.42
43.04
42.97
46.54
46.45
42.34
42.92
43.12
43.47
43.2
43.12
44.76
44.76
44.4
43.17
44.19
45.19
44.19
43.47
44.26
40.49
44.55
44.44
40.54
36.97
35.26
36.97
35.24
32.2
33.61
31.26
32.57
31.94
35.02
No26
No49
No26
No22
No32
No32
No41
No51
No35
No33
No38
No40
No40
No38
No33
s14
No42
No42
No44
No53
d1
in25
d1
No44
No49
No75
No51
No45
No316
No56
No56
No57
No64
No61
No57
No59
No64
d2
46.78
44.26
46.78
42.8
42.97
42.97
42.86
44.55
46.54
43.04
43.12
43.2
43.2
43.12
43.04
46.91
44.76
44.76
44.4
40.54
44.19
43.37
44.19
44.4
44.26
40.08
44.55
36.97
36.8
36.26
36.26
34.62
32.57
34.54
34.62
35.24
32.57
34.59
331
Result
Downstre
am
Velocity
(m/s)
2
3.7
7
1.1
5.7
4.5
4.1
7
2.2
6.4
3.3
2.1
5.2
5
4.8
6.1
0
6.2
8.4
9.3
12.1
6.9
10.4
5.3
15.6
7.4
3.4
9.7
2.6
3.3
7.8
6.4
4.5
7.5
10
5.8
5.5
4.9
Chapter 6
Name
Diameter
(in)
Length
(km)
Flow
(sm^3/hr)
FromNode
Name
FromNode
Pressure
To-Node
Name
To-Node
Pressure
Pi60
Pi61
Pi63
Pi62
Pi64
Pi67
Pi66
Pi68
Pi70
Pi69
Pi72
Pi71
Pi73
Pi74
Pi75
Pi76
Pi77
Pi79
Pi78
Pi81
Pi80
Pi82
Pi83
Pi84
Pi85
Pi86
Pi87
Pi89
Pi88
Pi90
Pi92
Pi91
Pi94
Pi93
Pi96
Pi95
Pi97
Pi99
18
18
18
16
20
16
16
28
32
10
28
16
28
30
30
42
32
28
24
32
10
24
24
28
32
32
32
32
10
32
32
10
32
24
32
6
36
24
1
3
2
1
18
1
1
17
25
5.5
14
2
6.8
26
30
165
9.4
5.4
3
59
9
2
18
28.2
40
9
110
10
7
79
23
3
6
2
38
60
12
9
-105043
-24757
-48478
-51000
114279
-138000
-87000
220022
599497
-18030
146022
-74000
-516542
677971
362572
614223
-69823
-571542
-46000
552527
-45000
-68000
210420
306122
-380000
294030
304030
293030
-1000
287030
287030
-6000
333854
14210
35492
0
298361
-196374
No61
No61
No68
No331
s4
No72
No330
d2
No75
No31
No77
No332
d6
s5
s16
d7
s5
No84
No317
No86
in16
No318
d6
d6
No346
in82
c2
No97
in83
No94
No100
in84
in71
d8
No103
No103
No103
No106
34.54
34.54
34.56
34.25
35.55
34.29
34.18
34.59
40.08
39.83
33.96
33.81
29.6
39.57
35.79
33.73
39.57
32.1
32.08
35.2
32.25
27.87
29.6
29.6
19.97
37.22
40.62
36.94
36.94
35.96
36.64
36.62
27.76
25.03
27.44
27.44
27.44
27.02
No57
No68
d2
No72
No70
No70
No72
No77
No383
No75
d7
No77
No372
d7
d7
c1
No366
No368
No84
in71
No86
No88
No374
No376
No92
No97
in82
No100
No97
No98
No94
No100
No103
No378
d9
in70
c1
in71
34.62
34.56
34.59
34.29
34.55
34.55
34.29
33.96
37.09
40.08
33.73
33.96
31.14
33.73
33.73
27.16
39.59
33.48
32.1
27.76
35.2
27.89
27.99
27.16
23.48
36.94
37.22
36.64
36.94
33.49
35.96
36.64
27.44
25.03
27.42
27.44
27.16
27.76
332
Result
Downstre
am
Velocity
(m/s)
4.7
1.1
2.2
2.9
4.1
7.8
5
4.1
7.8
2.2
2.8
4.3
11.2
11.1
6
6.5
0.9
11.4
1.2
9.8
7
2.1
6.6
7.2
9.4
3.9
4
3.9
0.1
4.2
3.9
0.8
6
0.5
0.6
0
4.3
6.4
Chapter 6
Name
Diameter
(in)
Length
(km)
Flow
(sm^3/hr)
FromNode
Name
FromNode
Pressure
To-Node
Name
To-Node
Pressure
Pi138
Pi140
Pi139
Pi142
Pi141
Pi143
Pi144
Pi145
Pi146
Pi149
Pi148
Pi151
Pi150
Pi153
Pi152
Pi155
Pi154
Pi157
Pi156
Pi159
Pi158
Pi161
Pi160
Pi163
Pi162
Pi165
Pi164
Pi167
Pi166
Pi169
Pi168
Pi171
Pi170
Pi173
Pi172
Pi175
Pi174
Pi177
18
30
24
28
10
32
16
16
24
28
24
24
10
24
24
24
12
24
16
24
24
24
10
24
10
24
10
24
24
24
10
24
12
24
12
16
8
28
159
3
1
28
6
6
14
10
10
6
3
2
10
4
2
3
3
6
2
8
4
2
1
1
2
4
4
5
3
3
2
11
4
2
4
3
8
18
68360
338030
-123000
-195488
40000
811608
4579
33000
238158
-155488
-95000
-68042
-13800
25158
168000
193158
45000
-366942
-14000
-135042
67000
-352942
-50000
-302942
-35000
-155042
-20000
56000
11000
-192942
-37900
-46842
-72000
-267942
-75000
-10821
15400
93800
No105
No146
No340
No147
No147
d11
d8
d8
d8
No154
No347
No156
in69
No157
No157
No159
No159
No162
No334
No163
No163
No166
in42
No168
in53
No170
in50
No171
No171
No174
in51
No176
in14
No178
in52
No179
No179
No182
42.2
42.11
42.09
23.8
23.8
41.24
25.03
25.03
25.03
23.64
23.57
23.46
22.99
23.39
23.39
23.66
23.66
25.66
25.65
23.49
23.49
25.09
24.53
24.87
24.32
23.85
23.47
23.44
23.44
24.09
23.42
23.37
21.39
24.36
22.31
25.02
25.02
25.43
No272
No143
No146
No377
in10
No227
No179
No327
No159
No147
No154
No163
No156
No371
No336
No157
in12
d9
No162
No170
No171
No162
No166
No166
No168
No174
No170
No338
No339
No178
No174
No183
No176
d10
No178
No367
in11
No136
37.34
41.96
42.11
24.97
21.46
40.01
25.02
24.81
23.66
23.8
23.64
23.49
23.46
23.38
23.24
23.39
23.11
27.42
25.66
23.85
23.44
25.66
25.09
25.09
24.87
24.09
23.85
23.4
23.44
24.36
24.09
23.43
23.37
24.7
24.36
25.03
23.62
25.26
333
Result
Downstre
am
Velocity
(m/s)
2.8
4.4
2.5
5.3
9.4
9.8
0.4
2.6
8.8
4.2
3.5
2.5
3
0.9
6.3
7.2
6.8
12.5
1.1
5
2.5
12.3
10.3
10.7
7.3
5.7
4.3
2.1
0.4
7
8.2
1.8
11.8
9.6
11.8
0.9
5.1
2.4
Chapter 6
Name
Diameter
(in)
Length
(km)
Flow
(sm^3/hr)
FromNode
Name
Pi176
Pi179
Pi178
Pi181
Pi180
Pi182
Pi184
Pi183
Pi185
Pi187
Pi186
Pi189
Pi188
Pi191
Pi190
Pi192
Pi193
Pi194
Pi195
Pi197
Pi196
Pi199
Pi201
Pi200
Pi203
Pi202
Pi204
Pi206
Pi205
Pi208
Pi207
Pi210
Pi209
Pi212
Pi211
Pi213
Pi214
Pi215
8
24
10
28
28
24
28
10
22
22
22
22
22
32
10
22
12
12
24
16
8
24
32
16
24
10
24
32
10
32
10
16
12
32
10
24
16
8
1
4
16
1
4
7
2
1
7
1
1
1
1
22
7
40
28
25
18
18
7
5
14
1
7
2
1
3
6
3
4
19
1
3
4
4
9
25
-800
-54242
7400
94600
126000
630000
-31400
-72600
240000
125000
64000
189000
51000
-98633
-26000
199065
41067
41067
348800
-25821
15000
348800
-238633
-140000
346000
-2800
208000
-242633
-4000
-265633
-23000
-120821
-95000
-270033
-4400
-270033
-30000
6000
in65
No183
No183
No186
s9
s6
No190
in41
d4
No192
No192
No194
No194
No197
in62
d3
d3
d3
d3
No199
No199
No202
No204
No343
No206
in63
psg3
No209
in32
No211
in33
No213
in68
No215
in64
c4
No314
No219
334
FromNode
Pressure
To-Node
Name
To-Node
Pressure
25.43
23.43
23.43
25.46
25.52
35.43
25.45
24.28
25.46
23.84
23.84
23.97
23.97
39.79
39.14
30.63
30.63
30.63
30.63
25.08
25.08
26.18
39.87
39.64
24.79
24.79
18.62
40.13
40.11
40.19
39.9
25.4
24.64
40.26
40.24
40.33
30.02
34.19
No182
No156
in49
No369
No186
d3
No186
No190
No194
No344
No345
No192
No319
No204
No197
d4
No361
No357
No202
No365
in13
No206
No209
No204
No342
No206
psg2
No211
No209
No215
No211
No218
No213
c4
No215
No360
No218
No224
25.43
23.46
23.21
25.45
25.46
30.63
25.46
25.45
23.97
23.77
23.82
23.84
23.96
39.87
39.79
25.46
27.12
27.52
26.18
25.32
23.93
24.79
40.13
39.87
22.75
24.79
18.49
40.19
40.13
40.26
40.19
30.16
25.4
40.33
40.26
40.74
30.16
33.7
Result
Downstre
am
Velocity
(m/s)
0.2
2
1.6
2.4
3.2
17.9
0.8
15.1
10.4
5.5
2.8
8.3
2.2
1.2
3.3
8.1
5.3
5.2
11.7
2
4.9
12.3
2.9
6.8
13.3
0.6
9.9
2.9
0.5
3.2
2.8
9.4
13.5
3.2
0.5
5.7
2
1.4
Chapter 6
Name
Diameter
(in)
Length
(km)
Flow
(sm^3/hr)
FromNode
Name
FromNode
Pressure
To-Node
Name
To-Node
Pressure
Pi217
Pi219
Pi218
Pi221
Pi220
Pi222
Pi223
Pi224
Pi226
Pi225
Pi227
Pi228
Pi229
Pi230
Pi231
Pi233
Pi232
Pi235
Pi234
Pi236
Pi238
Pi237
Pi240
Pi239
Pi242
Pi241
Pi244
Pi243
Pi245
Pi246
Pi248
Pi247
Pi250
Pi249
Pi251
Pi252
Pi253
Pi254
24
8
8
24
24
42
16
24
32
32
8
8
32
30
30
16
10
32
10
16
16
16
8
16
16
12
16
10
16
24
24
16
24
16
24
36
36
16
9
6
4
6
8
40
16
165
82
55
6
12
144
14
16
69
2
18
2
4
4
2
1
1
2
2
77
1
76
3
29
2
1
2
124
31
22
77
-150248
5000
-1000
119785
-119785
-1091608
96920
198439
481106
330503
1000
2000
151681
144788
135212
61920
-5000
265530
-11500
102000
82920
14000
15000
-42000
57000
-45000
66920
16000
-20685
301594
-190806
153000
-37806
0
156000
23090
100910
-20550
No222
No224
in56
No225
No225
d11
s8
s8
No227
No227
No229
No231
c5
d11
d11
No236
in17
No238
in86
No239
No241
No241
No244
in43
No246
in74
No247
No247
d13
d13
No250
No250
No252
No252
d12
d12
d12
d13
44.43
33.7
33.7
44.55
44.55
41.24
44.7
44.7
40.01
40.01
34.13
34.16
37.96
41.24
41.24
38.79
38.78
27.42
27.37
34.08
43.1
43.1
33.4
33.37
33.49
33.23
42.8
42.8
35.37
35.37
34.08
34.08
34.08
34.08
41.11
41.11
41.11
35.37
s8
in57
No224
No222
s8
s8
No241
d13
No363
No379
in54
No229
No233
d12
d12
d13
No236
No129
No238
No246
No247
No315
in44
No244
No244
No246
No236
in6
No249
No257
No249
No324
No250
ps35
No322
c3
No265
No249
44.7
33.62
33.7
44.43
44.7
44.7
43.1
35.37
33.32
38.01
34.12
34.13
36.78
41.11
41.11
35.37
38.79
26.8
27.42
33.49
42.8
43.09
33.28
33.4
33.4
33.49
38.79
42.76
35.84
34.92
35.84
33.43
34.08
34.08
36.61
41.11
41.07
35.84
335
Result
Downstre
am
Velocity
(m/s)
2.9
1.2
0.2
2.3
2.3
7.4
4.3
4.8
7
4.2
0.2
0.5
2
1.9
1.8
3.4
0.6
4.9
2.1
5.9
3.7
0.6
3.5
2.5
3.3
4.7
3.3
1.8
1.1
7.5
4.8
8.9
1
0
3.7
0.2
0.9
1.1
Chapter 6
Name
Diameter
(in)
Length
(km)
Flow
(sm^3/hr)
FromNode
Name
FromNode
Pressure
To-Node
Name
To-Node
Pressure
Pi256
Pi255
Pi258
Pi257
Pi260
Pi259
Pi261
Pi263
Pi262
Pi265
Pi264
Pi267
Pi266
Pi269
Pi268
Pi271
Pi270
Pi272
Pi273
Pi274
Pi275
Pi276
Pi277
Pi279
Pi278
Pi281
Pi280
Pi283
Pi282
Pi284
Pi286
Pi285
Pi288
Pi287
Pi290
Pi289
Pi293
Pi292
24
10
24
12
24
16
32
36
16
10
10
10
10
10
10
18
16
30
30
10
30
16
16
36
36
30
8
36
36
30
16
16
24
18
16
16
16
10
5
11
5
6
3
7
18
26
7
2
3
2
1
1
2
3
25
28
204
15
12
75
98
19
81
3
19
79
29
6
7
54
2
2
2
2
32
5
242194
59400
72194
124000
196194
46000
265530
-79090
-180000
47900
-11500
7500
40000
47500
-400
37676
41549
225030
225030
-22800
242730
-10865
-1135
34000
23090
225030
-17700
10910
90000
225030
-25000
14135
30000
195030
185030
10000
-1135
0
No257
No257
No259
No259
No261
No261
No129
No265
No321
No267
in36
No268
No268
No271
in75
No272
No272
No273
No274
in87
No263
No272
No144
No279
c3
No281
in88
No283
s12
No275
No285
No285
No287
No287
No289
No289
No293
No292
34.92
34.92
34.24
34.24
34.44
34.44
26.8
41.07
38.42
28.25
28.17
26.88
26.88
27.35
27.35
37.34
37.34
25.55
24.49
24.44
26.16
37.34
37.31
41.1
41.11
25.66
21.13
41.1
41.14
14.64
37.47
37.47
14.25
14.25
12.81
12.81
37.31
37.31
No261
No267
No219
in18
No259
No323
No263
No283
No265
No271
No267
in37
in38
No268
No271
No144
No233
No274
No275
No263
No281
No285
No293
No307
No279
No273
No281
No279
No283
No287
s18
No286
in89
No289
in91
in90
No278
No293
34.44
28.25
34.19
28.03
34.24
34.23
26.16
41.1
41.07
27.35
28.25
26.86
26.55
26.88
27.35
37.31
36.78
24.49
14.64
26.16
25.66
37.47
37.31
41.09
41.1
25.55
25.66
41.1
41.1
14.25
37.53
37.32
14.24
12.81
10
12.8
37.32
37.31
336
Result
Downstre
am
Velocity
(m/s)
6.1
10.6
1.8
15.4
5
2.6
5
0.7
9.1
8.8
2.1
1.4
7.6
8.9
0.1
1.5
2.2
5.2
8.6
4.7
5.3
0.6
0.1
0.3
0.2
4.9
6.6
0.1
0.8
8.8
1.3
0.7
1.8
23.5
35.6
1.5
0.1
0
Chapter 6
Name
Diameter
(in)
Length
(km)
Flow
(sm^3/hr)
FromNode
Name
FromNode
Pressure
To-Node
Name
To-Node
Pressure
Pi294
Pi295
Pi297
Pi296
Pi299
Pi298
Pi301
Pi300
Pi302
Pi304
Pi303
Pi306
Pi305
Pi307
Pi310
Pi309
Pi312
Pi311
Pi313
Pi314
Pi315
Pi316
Pi317
Pi318
Pi319
Pi320
Pi321
Pi322
Pi323
Pi324
Pi325
Pi326
Pi327
Pi328
Pi329
Pi330
Pi406
Pi407
16
24
16
12
16
12
16
12
16
16
12
36
16
24
36
36
36
20
24
24
32
32
18
24
28
28
16
24
8
24
24
24
24
12
17.25
36
22
24
5
4
13
4
21
3
18
4
23
31
6
8
5
45
32
220
44
208
3.3
1
1
1
5
2
31
5
25
5
4
10
2
9
115
15
1
1
87
1
0
-13000
0
0
0
0
0
0
38000
7000
31000
15000
-19000
-15000
0
0
0
0
34620
193768
656227
683227
169063
138000
-57788
322212
-150821
66194
1000
64194
-37806
-14135
0
-63209
127000
912584
-256194
64194
s13
s10
No298
in94
No300
in93
No302
ps36
s10
No304
No304
No307
No325
in76
No310
No310
No313
ps39
No2
No20
No24
No26
No56
No88
No92
No111
No218
No219
No229
No231
No239
No278
No286
d1
No105
c1
d9
No358
37.53
37.31
37.53
37.53
37.53
37.53
37.53
37.53
37.31
36.88
36.88
41.09
41.07
41.08
41.09
41.09
41.09
41.09
42.14
42.65
46.66
46.78
36.26
27.89
23.48
24.06
30.16
34.19
34.13
34.16
34.08
37.32
37.32
44.19
42.2
27.16
27.42
34.08
No300
No278
in95
No298
No302
No300
No298
No302
No304
in35
in34
No309
No307
No309
in77
No313
s11
No313
No21
No14
No35
No24
No59
No335
No140
No92
No379
No231
in55
No358
No252
No286
in73
s14
No320
No381
No105
No239
37.53
37.32
37.53
37.53
37.53
37.53
37.53
37.53
36.88
36.85
36.54
41.09
41.09
41.09
41.09
41.09
41.09
41.09
42.14
42.6
46.54
46.66
35.24
27.81
23.61
23.48
38.01
34.16
34.12
34.08
34.08
37.32
37.32
46.91
42.07
26.94
42.2
34.08
337
Result
Downstre
am
Velocity
(m/s)
0
0.3
0
0
0
0
0
0
2
0.4
2.9
0.1
0.9
0.3
0
0
0
0
0.7
3.9
6.7
7
7.4
4.3
1.6
8.8
9.8
1.7
0.2
1.6
1
0.3
0
4.9
5
13.2
9.7
1.6
Chapter 6
Name
Diameter
(in)
Length
(km)
Flow
(sm^3/hr)
FromNode
Name
FromNode
Pressure
To-Node
Name
To-Node
Pressure
Pi408
Pi412
Pi417
Pi418
Pi419
Pi420
Pi421
Pi422
Pi423
Pi424
Pi425
Pi426
Pi410
Pi411
Pi415
Pi416
Pi351
Pi352
Pi353
Pi354
Pi355
Pi356
Pi357
Pi358
Pi360
Pi361
Pi345
Pi346
Pi347
Pi348
Pi349
Pi350
Pi359
Pi362
Pi363
Pi364
Pi365
Pi366
Pi367
32
24
32
6
16
6
16
6.065
28
6.065
24
10
12
23.25
12
23.25
16
24
28
24
23.25
16
32
36
17.25
17.25
32
6.065
28
6.065
28
6.065
24
6.065
28
6.065
32
7.981
24
6
38
80
6
6
3
20
6
2
2
6
3
15
7
12
8
27
11
3
18
4
10
5
63
2
2
112.6
1
1
1
4.2
2
1
1
16.8
23
14.9
9.8
1
0
-270033
481106
0
-25821
0
-10821
0
94600
0
25158
0
41067
0
41067
0
-38811
-232041
-128159
-53119
-67329
28000
151681
912584
-114279
-476852
-72633
-2810
-572342
800
-525542
9000
206000
4420
303712
2410
597527
1970
201868
s1
No360
No363
No363
No365
No365
No367
No367
No369
No369
No371
No371
No357
No357
No361
No361
No233
No249
No377
No378
No377
No379
No379
No382
s4
s16
No366
No10
No368
No368
No372
No372
No374
No374
No376
No376
No383
No383
No41
34.08
40.74
33.32
33.32
25.32
25.32
25.03
25.03
25.45
25.45
23.38
23.38
27.52
27.52
27.12
27.12
36.78
35.84
24.97
25.03
24.97
38.01
38.01
50
35.55
35.79
39.59
39.57
33.48
33.48
31.14
31.14
27.99
27.99
27.16
27.16
37.09
37.09
42.86
No358
No222
d8
in96
No213
in97
No199
in98
No182
in99
No176
in100
d4
s19
d4
s20
No144
No233
d8
No123
No378
No326
c5
No105
s16
s5
No197
No366
d7
No1
No84
No4
No88
No6
No114
No8
No86
No11
No22
34.08
44.43
25.03
33.32
25.4
25.32
25.08
25.03
25.43
25.45
23.37
23.38
25.46
27.52
25.46
27.12
37.31
36.78
25.03
25.15
25.03
37.9
37.96
42.2
35.79
39.57
39.79
39.59
33.73
33.47
32.1
30.75
27.89
27.93
25.61
26.75
35.2
37.07
42.8
338
Result
Downstre
am
Velocity
(m/s)
0
5.7
9.5
0
2
0
0.9
0
2.4
0
0.9
0
5.6
0
5.6
0
2
5.6
3.3
1.9
2.5
1.4
1.9
8.2
5.4
22.3
0.9
1
10.9
0.3
10.8
4
6.4
2.2
7.6
1.2
8.3
0.4
4
Chapter 6
6.4.3.3.4 Regulator and Regulator Stations Report

table 6- 10 Regulator and Regulator Stations Report
339
Chapter 6
340
Chapter 6
6.4.3.3.5 compressor Report

table 6- 11 compressor Report
Suction
Pressure
(bar)
Discharge
Differential
Ratio
Power (KW)
Flow
(sm^3/h)
minimum
20
Current
26.94
Current
Maximum
50
65
Current
23.06
Maximum
45
Minimum
1.2
Current
1.82
Maximum
Current
20000
Installed
40000
Max. Available
40000
Fractional Capacity
.5931
Minimum
400000
Current
912584
Maximum
2000000
341
Chapter 6
6.4.3.3.6 Pipe Diameter Summary

table 6- 12
Actual in
6.4.3.3.8 Demand summary
pipe diameter summary report
Count
Total km
4.000
1
1.00
6.000
14
80.04
6.065
7
36.00
7.981
2
10.77
8.000
21
134.28
10.000
45
205.56
12.000
15
132.00
16.000
57
997.97
17.250
3
5.05
18.000
15
218.00
20.000
3
256.00
22.000
7
138.00
23.250
3
19.00
24.000
86
909.44
28.000
25
240.37
30.000
16
374.00
32.000
35
1235.89
36.000
15
677.96
42.000
2
205.00
6.4.3.3.7 Pipe Length Summary
6.4.3.3.9 Supply summary
Model Reliability
table 6- 13 Pipe Length Summary report
Pipes
Greater
Than Or
Equal To
0.0
20.0
40.0
60.0
80.0
100.0
120.0
140.0
160.0
200.0
220.0
After running the model, it is solved

and feasible, hence there is no
negative pressures in the model.
The result pressures for the supply
points is approximately near the
actual pressure.
Total length, and total gas
consumption is approximately near
the actual data.
Conclusion for model Reliability
Pipes
Less
Than
Count
Total
km
20.0
40.0
60.0
80.0
100.0
120.0
140.0
160.0
180.0
220.0
240.0
306
32
8
9
5
3
1
2
3
2
1
1624.60
890.18
377.00
654.98
428.00
337.57
124.00
303.00
505.00
412.00
220.00
This model is ready to be used in unsteady

state analysis
342
Chapter 6
6.4.4.2 Misconceptions regarding

unsteady-state analysis
6.4.4 Unstady state analysis

Now after the steady state analysis we
know that our model is reliable, and we
can make the unsteady state analysis
A number of misconceptions exist

regarding gas transient, or unsteady-state,
analyses. Some of the most common are as
follows:
But first there is basic terms we should

mention
6.4.4.1 Basic terms

Base length: Expressed by the symbol X,
this is the base computational increment
that USM uses for its computations. no
pipes should be shorter than this base
length.
Reach: its L/X.

Time step: Expressed by the symbol T,
this is the computational time increment,
the base length divided by the wave speed.
Alpha:
a multiplier to the inertial term of

the equation of motion, alpha ()
automatically adjusts the wave
speed in each individual pipe so
that it is an integer number of
reaches long.
It also allows USM to speed up by
adjusting the global wave speed of
the fluid, which means to take
larger time steps during a
simulation than specified by the
base time step.
There are no real numeric

differences between steady- and
unsteady-state
analyses;
an
unsteady-state analysis is just a
series of sequential steady-state
runs.
Unsteady-state analysis is just
another name for transient
analysis.
Any valid steady-state model
should be a perfectly valid
unsteady-state model.
All unsteady-state models must
begin from steady-state conditions.
Steady-state models are always
more conservative, and thus better
for design, than
are unsteady-state models.

Misconception 1: There are no
real numeric differences between
steady-state and unsteady-state
analyses
This statement is incorrect.
An unsteady-state analysis is not just a
series of sequential steady-state runs.
wave speed: Expressed by the symbol B,

this is the speed of sound in the gas being
modeled. This value is generally
somewhere near(320 m/s) for natural gas.
It is calculated by USM, using the equation
B = (gZRT)0.5, where g is the gravitational
constant, Z is the gas compressibility
factor, R is the gas constant, and T is the
gas temperature
The steady-state assumption is that the

sum of the flows entering the system is
precisely equal to the sum of the flows
leaving the system.
This is simply inadequate for a transient
study. In transient analyses, you consider
the mass balances through a system.
Consider a single long pipe. Steady state
assumptions require a continuous flow
through the pipe, and the flow into one
343
Chapter 6
Misconception 3: All unsteadystate models must begin from

steady-state
Conditions There is no rule that requires
USM to begin from a steady-state analysis.
Initial conditions; however, must be
supplied. It is certainly true that typically,
these conditions are created in a steadystate model. There are other alternatives;
however, you might instead use a capture
of SCADA information that represents the
initial conditions, or use a technique
known as transient relaxation to
determine the initial conditions. In this
technique, you make a best guess as to
the initial conditions inherent in the
system, allow the unsteady state analysis
to run until any numeric pressure
fluctuations stabilize, and then begin the
real simulation.
end is equal to the flow out the other end.

In transient analysis, the flow into the pipe
may not equal the flow through the pipe,
which may not equal the flow out of the
pipe.
Succinctly put, steady-state analyses do
not consider the effects of packing and
drafting of the volume of gas inherent in
the pipe, or pipeline system, whereas
transient models do.
Misconception 2: Any valid
steady-state model is a perfectly
valid unsteady state model
Fundamental differences exist in the
numeric methods underlying each analysis
type, and more important, in the purposes
for each analysis. Consider the simple case
of a steady-state model used to determine
the required size of a pipe in a design
scenario (solving for pipe diameter). This
obviously could not be used in an
unsteady-state analysis. Since by its very
definition an unsteady-state model has
regularly changing pressures and flows, the
solution for the pipe diameter would
change every time step. Thus your 100 mm
pipe needed at 3 AM becomes a 200 mm
pipe at 7 AM, and so on. Further, it is
generally the case that steady-state
models are much more precise than
unsteady-state and modeled to a much
higher level of detail. However, you can
move easily from a steady-state to an
unsteady-state model
You should be aware that the quality of

your initial data directly controls the
stability of your unsteady-state analysis.
Wildly inaccurate initial data can lead to
very unstable early stages of an unsteadystate analysis.
Misconception 4: Steady-state
models are always more
conservative, and thus better for
design, than are unsteady-state
models
Consider, for example, design for a powergenerating plant that exists at the end of a
pipeline. Designing the pipeline in steady
state for the absolute peak plant load
certainly ensures that a steady flow is
deliverable. In fact, the pipe diameter may
well be too large to be practical, depending
on the pressure, pack volumes, and so on.
On the other hand, since steady-state
analysis ignores pack and the timing of
events, it can be misleading. If the plant
starts very fast, it could well be that the
if you have created the model with

consideration of its ultimate analysis in
USM, and you have supplied the additional
requisite data (such as flow or profile
data).
344
Chapter 6
downstream end of the pipe runs out of

pressure long before flow is established
throughout the pipe to the plant. Thus, the
steady-state design is not conservative and
it is wrong. While it is generally true that
steady-state models are usually more
conservative, there are always exceptions,
depending on the speed of the event for
which you are designing as well as other
factors.
6.4.4.4.1 Profile types

Profiles represent functional relationships
with a finite number of points. The profile
type defines the interpolation method that
is performed to obtain dependent variable
(Y) values when the independent variable
(X) value falls between any two given
points.
Ramp profiles: With a ramp profile,
intermediate values are determined by
performing a linear interpolation between
two points on the profile.
6.4.4.3 Controlling the transient

model
You can control system flows and
pressures for an unsteady-state model
through direct facility editing and through
profiles. You can edit any node at any time,
change it from pressure to flow control,
change the pressure or flow, and assign
profiles
And editing any facilities as a Compressor
(ambient temperature, power, speed, or
utilization factor) with respect to time,
Regulator, regulator station set points and
Valve position.
Figure 6- 61 ramp profile
Step profiles: With a step profile,

intermediate values are determined by
holding constant the value of the
preceding dependent variable (Y) point on
the profile.
The practical way to control the model is

using profiles
6.4.4.4 PROFILES
A profile is a curve that defines a functional
relationship between two variables. These
variables are typically boundary condition
variables with respect to time. Each profile
is defined by sets of X and Y coordinates.
The variable represented on the X axis is
the independent variable, while the
variable represented on the Y axis is the
dependent variable.
Figure 6- 60 step profile
Synergi includes several profile categories

that you can apply to your model.
6.4.4.4.2 Attributes that can be profiled

1) Nodes:
a) Flow
345
Chapter 6
b) Pressure
g) Utilization factor
c) Gas temperature
6.4.4.4.3 create a profile

In the model explorer, click the Model
Data tab, and then select Profiles.
2) Pipes:
a) Pipeline ambient temperature
Right-click and select New Profile.
b) Elevation
3) Regulators:
a) Setpoint
b) Cgmax
c) Maximum flow rate
4) Valves:
a) Dimensionless valve position
5) Compressor Stations:
a) Ambient temperature
b) Gas cost
Figure 6- 63 creating new profile
c) Electric cost
d) Other cost
e) Set pressure
f) Maximum flow rate
Figure 6- 62 profile editor
346
Chapter 6
It is almost impossible to get the profile of

every single node in the model
So from one profile, we divided the main
flow over 24 hours by the flow in each
hour to get a dimensionless multiplier
table 6- 14 krir 1,2 flow and multiplayer over
24 h
HOURS
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
FLOW
EVERY
HOUR
43.69
43.13
44.05
42.98
43.99
43.65
34.4
35.67
44.17
38.92
38.49
38.5
39.4
38.85
39.44
40.04
39.52
39.4
42.94
42.7
44.84
45.95
44.96
45.16
FACTOR
0.996
0.983
1.004
0.980
1.003
0.995
0.784
0.813
1.007
0.887
0.877
0.878
0.898
0.886
0.899
0.913
0.901
0.898
0.979
0.973
1.022
1.047
1.025
1.029
Figure 6- 64 step profile of krir 1,2
6.4.4.5 fast transient and slow

transient analysis modes
Synergi offers two types of transient
analyses in USM. The first type is called
Fast Transient mode, which is the
traditional unsteady-state analysis
The second type is called Slow Transient
mode, which performs similar to the
traditional unsteady-state analysis, but
does not include fluid momentum in its
calculations.
6.4.4.6 considerations for deciding

to perform a fast or slow transient
analysis
1-During Slow Transient mode, the
exclusion of inertia should provide very
similar results to a Fast Transient mode
analysis, unless there are fast changes in
the simulation, in which case, the results
will be skewed. Slow Transient mode
analyses are always inappropriate for very
fast events, such as line break and gas loss
studies, power-plant start-up analyses,
rapid valve closure cases, equipment
failure studies, and the like.
2-The Fast Transient mode is essential for
fast events, as mentioned in the preceding
347
Chapter 6
point, and is also applicable to slower

events as well, as we have seen, but
provide little or no benefit for slow
changing systems.
analysis on the same structurally sound

model and arrive at a variance in results,
the fast transient analysis is the more
correct answer, because it includes the
effects of momentum
3- Slow Transient mode analyses likely will

run considerably faster than Fast Transient
mode analyses. However, slow transient
analyses will likely not yield dramatic
simulation time savings if the model is
running composition tracing, especially if
there are many short pipes in the model.
We used slow transient mode because we

dont intend to capture fast events in the
network
4-Slow Transient mode analyses do not

require the input of a base length. Instead,
you input a time step for the analysis. Be
advised that setting shorter T values for a
slow transient analysis will not ensure
automatically that it will approximate the
fast transient analysis results, because
momentum calculations are not factored
in slow transient analyses.
6.4.4.7 Output of the USM

After the slow transient analysis is done
the solved and feasible indicator turns into
USM as in figure ( 6-67 )
Figure 6- 65
Supply and demand are the same as steady

state analysis, except for power station
where we added flow profiles to indicate
the variation of flow with time
5-Slow Transient mode analyses do include

heat transfer calculations, whereas fast
transient analyses are isothermal. For
systems where heat transfer effects may
be important, you should investigate both
analysis types for their effectiveness.
6-If you perform both a Fast Transient
mode analysis and a Slow Transient mode
348
Chapter 6
Power station flow
Figure 6- 66 ps4 result flow and pressure
Total Supply flow rate
Figure 6- 67 total supply of the network over 24 h
349
Chapter 6
Total Demand flow rate
Figure 6- 68 total demand of the network over 24 h
System linepack
Figure 6- 69 total system linepack of the network over 24 h
350
Chapter 6
Regulator station flow
Figure 6- 70 regulator station 317 over 24 h
This model is ready to be used in optimization algorithms to find the best operation
scenario for the network.
351
Chapter 7 | Optimization
Chapter 7
Optimization
352
optimum seeking methods are also known

as mathematical programming techniques
and are generally studied as a part of
operations research.
7.1 Introduction
Optimization is the act of obtaining the
best result under given circumstances. In
design, construction, and maintenance of
any engineering system, engineers have to
take many technological and managerial
decisions at several stages. The ultimate
goal of all such decisions is either to
minimize the effort required or to
maximize the desired benefit.
Operations research is a branch of

mathematics
concerned
with
the
application of scientific methods and
techniques to decision making problems
and with establishing the best or optimal
solutions.
The beginnings of the subject of operations
research can be traced to the early period
of World War II.
Since the effort required or the benefit

desired in any practical situation can be
expressed as a function of certain decision
variables, optimization can be defined as
the process of finding the conditions that
give the maximum or minimum value of a
function. It can be seen from Fig. 7.1 that if
a point x corresponds to the minimum
value of function f (x), the same point also
corresponds to the maximum value of the
negative of the function, f (x). Thus
without loss of generality, optimization can
be taken to mean minimization since the
maximum of a function can be found by
seeking the minimum of the negative of
the same function.
During the war, the British military faced

the problem of allocating very scarce and
limited resources (such as ghter
airplanes, radars, and submarines) to
several
activities
(deployment
to
numerous targets and destinations).
Because there were no systematic
methods available to solve resource
allocation problems, the military called
upon a team of mathematicians to develop
methods for solving the problem in a
scientic manner. The methods developed
by the team were instrumental in the
winning of the Air Battle by Britain. These
methods, such as linear programming,
which were developed as a result of
research on (military) operations,
subsequently became known as the
methods of operations research.
In addition, the following operations on

the objective function will not change the
optimum solution x (see Fig. 7.2):
1. Multiplication (or division) of f (x) by a
positive constant c.
2. Addition (or subtraction) of a positive
constant c to (or from) f (x).
There is no single method available for
solving
all
optimization
problems
efficiently.
Hence a number of optimization methods
have been developed for solving different
types of optimization problems. The
353
It is, of course, a rare situation in which it is

possible to fully represent all the
complexities of variable interactions,
constraints, and appropriate objectives
when faced with a complex decision
problem. Thus, as with all quantitative
techniques of analysis, a particular
optimization formulation should be
regarded only as an approximation. Skill in
modeling, to capture the essential
elements of a problem, and good judgment
in the interpretation of results are required
to obtain meaningful conclusions.
Figure 7.1 Minimum of f (x) is same as

maximum of f (x).
Optimization, then, should be regarded as

a tool of conceptualization and analysis
rather than as a principle yielding the
philosophically correct solution.
Skill and good judgment, with respect to
problem formulation and interpretation of
results, is enhanced through concrete
practical experience and a thorough
understanding of relevant theory. Problem
formulation itself always involves a
tradeo between the conicting objectives
of building a mathematical model
suciently complex to accurately capture
the problem description and building a
model that is tractable. The expert model
builder is facile with both aspects of this
tradeo.
Figure 7.2 Optimum solution of f (x) or c +

f (x) same as that of f (x).
One aspiring to become such an expert

must learn to identify and capture the
important issues of a problem mainly
through example and experience; one
must learn to distinguish tractable models
from non-tractable ones through a study of
available technique and theory and by
nurturing the capability to extend existing
theory to new situations.
Mathematical programming techniques

are useful in nding the minimum of a
function of several variables under a
prescribed set of constraints. Stochastic
process techniques can be used to analyze
problems described by a set of random
variables having known probability
distributions. Statistical methods enable
one to analyze the experimental data and
build empirical models to obtain the most
accurate representation of the physical
situation.
354
As we go no in this book we will face

different types of optimization problems ,
we will lean how to distinguish between
them , classify them and to know which is
the suitable solution method or algorithm
for them .
The method was originally developed by

Kirkpatrick, Gelatt, and Vecchi.
The particle swarm optimization algorithm
mimics the behavior of social organisms
such as a colony or swarm of insects (for
example, ants, termites, bees, and wasps),
a ock of birds, and a school of sh. The
algorithm was originally proposed by
Kennedy and Eberhart in 1995. The ant
colony optimization is based on the
cooperative behavior of ant colonies,
which are able to nd the shortest path
from their nest to a food source. The
method was rst developed by Marco
Dorigo in 1992. The neural network
methods are based on the immense
computational power of the nervous
system to solve perceptional problems in
the presence of massive amount of
sensory data through its parallel
processing capability. The method was
originally used for optimization by Hopeld
and Tank in 1985.
We will see different objective functions

and problems such as : linear
programming, non-linear programming ,
quadratic functions , integer programming
, Dynamic and geometric programming ,
and finally Stochastic problem (which is
very imported in our project).
We will learn briefly about the traditional
methods of solutions like :
Simplex theory , interior point , gradient
projection , Newton Raphson , Branch and
Bound and cutting plane method .
But most importantly will be exposed to
non-traditional methods or as some called
them (state of art optimization science) ,
they have emerged as powerful and
popular methods for solving complex
engineering optimization problems in
recent years.
The fuzzy optimization methods were

developed to solve optimization problems
involving design data, objective function,
and constraints stated in imprecise form
involving vague and linguistic descriptions.
The fuzzy approaches for single and multiobjective optimization in engineering
design were rst presented by Rao in 1986.
These methods include genetic algorithms,

simulated annealing, particle swarm
optimization, ant colony optimization,
neural network-based optimization, and
fuzzy optimization.
The genetic algorithms are computerized
search and optimization algorithms based
on the mechanics of natural genetics and
natural selection.
Optimization science is used in many

industrial and research applications,
actually sometimes optimization save huge
amount of money for industries , a few
examples for this applications :
The genetic algorithms were originally

proposed by John Holland in 1975. The
simulated annealing method is based on
the mechanics of the cooling process of
molten metals through annealing.
1.Design of aircraft and aerospace

structures for minimum weight
2. Finding the optimal trajectories of space
vehicles
355
3. Design of civil engineering structures

such as frames, foundations, bridges,
towers, chimneys, and dams for minimum
cost
17. Design of optimum pipeline networks

for process industries
18. Selection of a site for an industry
19. Planning of maintenance and
replacement of equipment to reduce
operating costs
4. Minimum-weight design of structures

for earthquake, wind, and other types of
random loading
20. Inventory control
5. Design of water resources systems for

maximum benet
21. Allocation of resources or services

among several activities to maximize the
benet
6. Optimal plastic design of structures

7. Optimum design of linkages, cams,
gears, machine tools, and other
mechanical components
22. Controlling the waiting and idle times

and queuing in production lines to reduce
the costs
8. Selection of machining conditions in

metal-cutting processes for minimum production cost
23. Planning the best strategy to obtain

maximum prot in the presence of a
competitor
9. Design of material handling equipment,

such as conveyors, trucks, and cranes, for
minimum cost
24. Optimum design of control systems
10. Design of pumps, turbines, and heat

transfer equipment for maximum
efciency
11. Optimum design of electrical
machinery such as motors, generators, and
trans- formers
12. Optimum design of electrical networks
13. Shortest route taken by a salesperson
visiting various cities during one tour
14.
Optimal
production
controlling, and scheduling
planning,
15. Analysis of statistical data and building

empirical models from experimental
results to obtain the most accurate
representation
of
the
physical
phenomenon
16. Optimum design of chemical
processing equipment and plants
356
7.2 Corner Stones of any

optimization Model
Possible objectives that might

suggested for optimization by
organization are as follows:
7.2.1 Objectives
Maximize profit;
7.2.1.1 Defining Objectives
Minimize cost;
With a given set of constraints, different

objectives will probably lead to different
optimal solutions. Nevertheless, it should
not automatically be assumed that this will
always happen. It is possible that two
different objectives can lead to the same
operating pattern. As an extreme case of
this, it can happen that the objective is
irrelevant.
be
an
Maximize utility;
Maximize turnover;
Maximize return on investment;
Maximize net present value;
Maximize number of employees;
Minimize number of employees;
Minimize redundancy;
The constraints of a problem may dene a

unique solution. For example, consider the
following three constraints:
Maximize customer satisfaction;

Maximize probability of survival;
Maximize robustness of operating plan.
Many other objectives could be suggested.
It could well be that there is no desire to
optimize anything at all. Frequently, a
number of objectives will apply
simultaneously. Possibly, some of these
objectives will conflict. It is, however, our
contention
that
mathematical
programming can be relevant in any of
these situations, that is, in the case of
optimizing single objectives, multiple and
conflicting objectives or problems where
there is no optimization of the objective.
These force the solution x1=x2=1 no matter

what the objective is. Practical situations
do arise where there is no freedom of
action and only one feasible solution is
possible.
If the model for such a situation is at all
complicated, this property may not be
apparent. Should different objective
functions always yield the same optimal
solution the property may be suspected
and should be investigated as a greater
understanding of what is being modelled
must surely result. In fact, such a discovery
would result in there being no further need
for linear programming.
7.2.1.2 Single objectives

Most practical mathematical programming
models used in operational research
involve either maximizing profit or
minimizing cost. The profit that is
maximized would usually be more
accurately referred to as contribution to
profit or the cost as variable cost. In a cost
minimization, the cost incorporated in an
We now assume, however, that we have a

problem, where the definition of a suitable
objective is of genuine importance.
357
objective function would normally only be

a variable cost. For example, suppose each
unit of a product produced cost C. It
would only be valid to assume that x units
would cost Cx if C were a marginal cost,
that is, the extra cost incurred for each
extra unit produced. It would normally be
incorrect to add in fixed costs such as
administration or capital costs of
equipment.
or something analogous, where xij is the

quantity of product i produced by process
j and Di is the demand for product i .
Should such constraints be left out
inadvertently, as sometimes happens, the
minimum cost solution will often turn out
to produce nothing! On the other hand, if
a profit maximization model is built, it
allows one to be more ambitious.
Instead of specifying constant demands Di
, it is possible to allow the model to
determine optimal production levels for
each product.
An exception to this does arise with some

integer programming models when we
allow the model itself to decide whether or
not a certain fixed cost should be incurred;
for example, if we do not make anything
we incur no fixed cost, but if we make
anything at all we do incur such a cost.
Normally, however, with standard linear
programming models, we are interested
only in variable costs. Indeed, a common
mistake in building a model is to use
average costs rather than marginal costs.
Similarly, when profit coefficients are
calculated, it is only normally correct to
subtract variable costs from incomes. As a
result of this, the term profit contribution
might be more suitable.
The quantities Di would then become

variables di representing these production
levels.
In order that the model can determine
optimal values for the variable di, they
must be given suitable unit profit
contribution coefficients Pi in the objective
function. The model would then be able to
weigh up the profits resulting from
different production plans in comparison
with the costs incurred and determine the
optimal level of operations. Clearly, such a
model is doing rather more than the
simpler cost minimization model. In
practice, it may be better to start with a
cost minimization model and get it
accepted as a planning tool before
extending the model to be one of profit
maximization.
In the common situation where a linear

programming model is being used to
allocate productive capacity in some sort
of optimal manner, there is often a choice
to be made over whether simply to
minimize cost or maximize profit.
Normally, a cost minimization involves
allocating productive capacity to meet
some specified known demand at
minimum cost. Such models will probably
contain constraints such as
In a profit maximization model, the unit

profit contribution figure Pi may itself
depend on the value of the variable di . The
term Pi di in the objective function would
no longer be linear. If Pi could be expressed
as a function of di , a non-linear model
could be constructed.
358
An example of this idea is described by

McDonald et al . (1974) in a resource
allocation model for the health service.
Objectives and constraints can often be

interchanged, for example, we may wish to
pursue some desirable social objective so
long as costs do not exceed a specified
level, or alternatively, we may wish to
minimize
cost
using
the
social
consideration as a constraint on our
operations.
A complication may arise in defining a

monetary objective when the model
represents activities taking place over a
period of time. Some method has to be
found for valuing profits or costs in the
future in comparison with the present.
7.2.1.3 Multiple and conflicting objectives
This interplay between objectives and

constraints is a feature of many
mathematical programming models, which
is far too rarely recognized. Once a model
has been built, it is extremely easy to
convert an objective into a constraint or
vice versa. The proper use for such a model
is to solve it a number of times, making
such changes.
A mathematical programming model

involves a single objective function that is
to be maximized or minimized. This does
not, however, imply that problems with
multiple objectives cannot be tackled.
An examination and discussion of the

resultant solutions should lead to an
understanding of the operating options
available in the practical situation, which
has been modelled.
Various modelling techniques and solution

strategies can be applied to such problems.
We therefore have one method of coping

with multiple objectives through treating
all but one objective as a constraint.
Experiments can then be carried out by
varying the objective to be optimized and
the right-hand side values of the
objective/constraints.
The most usual technique is to discount

future money at some rate of interest.
Objective coefficients corresponding to
the future will then be suitably reduced.
A first approach to a problem with multiple

objectives is to solve the model a number
of times with each objective in turn.
The comparison of the different results
may suggest a satisfactory solution to the
problem or indicate further investigations.
Another way of tackling multiple objectives

is to take a suitable linear combination of
all the objective functions and optimize
that. It is clearly necessary to attach
relative weightings or utilities to the
different objectives.
A fairly obvious example of two objectives,

each of which can be optimized in turn, is
given by the MANPOWER PLANNING
problem
Here, either cost or redundancy can be
minimized. The computational task of
using different objectives in turn is eased if
each solution is used as a starting solution
to a run with a new objective
What these weightings should be may

often be a matter of personal judgment.
Again, it will probably be necessary to
359
experiment with different such composite

objectives in order to map out a series of
possible solutions. Such solutions can then
be presented to the decision makers as
policy options.
There is no one obvious way of dealing

with
multiple
objectives
through
mathematical programming. Some or all of
the above approaches should be used in
particular circumstances.
Most commercial package sallow the user

to define and vary the weightings of
composite objective functions very easily.
It should not be thought that this approach
to multiple objectives is completely
distinct from the approach of treating all
but one of the objectives as a constraint.
When a linear programming model is
solved, each constraint has associated with
it a value known as a shadow price. These
values have considerable economic
importance .
Given a situation with multiple objectives

in which there are no clearly defined
weightings for the objectives no cut-anddried approach can ever be possible.
Rather than being a cause for regret, this is
a healthy situation. It would be desirable if
alternative approaches were adopted
more often in the case of single objective
models rather than a once only solution
being obtained and implemented.
7.2.1.4 Minimax objectives
If these shadow prices were to be used as

the weightings to be given to the
objectives/constraints in the composite
approach that we have described in this
section, then the same optimal solution
could be obtained.
The following type of objective arises in

some
situations:
When objectives are in conflict, as multiple

objectives frequently will be to some
extent, any of the above approaches can
be adopted. Care must be taken, however,
when objectives are replaced by
constraints not to model conflicting
constraints as such. The resultant model
would clearly be infeasible.
Subjected
to
constraints.
conventional
linear
This can be converted into a conventional

linear programming form by introducing
a variable z to represent the above
objective. In addition to the original
constraints, we express the transformed
model as
Conflicting constraints necessitate a

relaxation in some or all of these
constraints. Constraints become goals,
which may, or may not, be achieved. The
degree of over, or under, achievement is
represented in the objective function.
This way of allowing the model itself to
determine how to allow a certain degree of
and is sometimes known as goal
programming.
The new constraints guarantee that z will

be greater than or equal to each
360
Of j aijxj for all i . By minimizing z , it will

be driven down to the maximum of these
expressions.
Of course, a maximin objective can easily
be dealt with similarly. It should, however,
be pointed out that a maximax (or
minimin) objective cannot be dealt with by
linear programming and requires integer
programming.
Rather surprisingly, the resultant model

can be converted into a linear
programming form by the following
transformations:
(i) Replace the expression
Example. Captain Hook and Captain

Sparrow have found some treasure. they
cant agree on how to split
their newly found riches,
and so they have asked you
for help.
by a variable t .
(ii) Represent the products xj t by variables
wj The objective now becomes
To start, you ask them to assign value

points to all of the items under
consideration, so that the points add up to
100. Their valuations are shown below:
(iii)
Introduce
constraint
in order to satisfy condition (i).

Convert the original constraints of the
form
Assume that the items can be divided
fractionally. For example, Captain Hook
can receive 60% of the gold coins, while
Captain Sparrow can receive the remaining
40%. In this case, Captain Hook gets 0.6
40 = 24 value points, and Captain Sparrow
gets 0.4 30 = 12 value points.
It must be pointed out that this

transformation is only valid if the
denominator j bjxj is always of the same
sign and non-zero. If necessary (and it is
valid), an extra constraint must be
introduced to ensure this. If j bj xj always
be negative the directions of the
inequalities in the constraints above must,
of course, be reversed.
Formulate a linear program to determine

how to allocate the treasure between
Captain Hook and Captain Sparrow, in a
way that maximizes the minimum total
value points of any captain.
7.2.1.5 Ratio objectives
In some applications, the following nonlinear objective arises:
361
Once the transformed model is solved, the

values of the xj variables can be found by
dividing wj by t .
Solving a mathematical programming

model of the situation with an arbitrary
objective function will at least enable one
to find a feasible solution if it exists, that is,
a solution that satisfies the constraints.
The last remark is often very relevant to
certain integer programming models,
where a very complex set of constraints
may exist.
An interesting application involving this

type of objective arises in devising
performance measures for certain
organizations where, for example, there is
no profit criterion that can be used. This is
described by Charnes et al . (1978).
It is, however, sometimes relevant to the

constraints of a conventional linear
programming model. The use of a
(contrived) objective is also of value
beyond simply enabling one to create a
well-defined mathematical programming
model.
The objective arises as a ratio of weighted

inputs and outputs of the organization
variables represent the weighting factors.
Each organization is allowed to choose
the weighting factors so as to maximize its
performance ratio subject to certain
constraints. This subject is now known as
data envelopment analysis (DEA
By optimizing an objective, or a series of

objectives, in turn, extreme solutions
satisfying the constraints are obtained.
These extreme solutions can be of great
value in indicating the accuracy or
otherwise of the model.
7.2.1.6 Non-existent and non-optimizable

objectives
The phrase non-optimizable objectives
might be regarded as self-contradictory.
Should any of these solutions be

unacceptable from a practical point of
view, then the model is incorrect and
should be modified if possible. As stated
before, the validation of a model in this
way is often as valuable, if not more
valuable, an activity as that of obtaining
solutions to be used in decision making.
We will, however, take the word

objective, in this phrase, in the nontechnical sense. In many practical
situations, there is no wish to optimize
anything. Even if an organization has
certain objectives (such as survival), there
may be no question of optimizing them or
possibly any meaning to be attached to the
word optimization when so applied.
Mathematical programming is sometimes
dismissed rather peremptorily as being
concerned with optimization when many
practical
problems
involve
no
optimization.
7.2.2 Constraints
7.2.2.1 Productive capacity constraints
If resources to be used in a productive
operation are in limited supply, then the
relationship between this supply and the
possible uses made of it by the different
activities give rise to such a constraint.
Such a dismissal of the technique is

premature. If the problem involves
constraints, finding a solution that satisfies
the constraints may be by no means
straightforward.
The resources to be considered could be

processing capacity or manpower.
362
balance constraints such as this will usually

be of the form:
7.2.2.2 Raw material availabilities

If certain activities (such as producing
products) make use of raw materials that
are in limited supply, then this clearly
should result in constraints.
7.2.2.3 Marketing demands and
showing that the total quantity (weight or

volume) represented by the xj variables
must be the same as the total quantity
represented by the yk variables.
limitations
If there is a limitation on the amount of a
product that can be sold, which could well
result in less of the product being
manufactured than would be allowed by
the other constraints, this should be
modelled. Such constraints may be of the
form
Sometimes, some coefficients in such a

constraint will not be unity, but some value
representing a loss or gain of weight or
volume in a particular process.
7.2.2.5 Quality stipulations
xM
These constraints usually arise in blending

problems where certain ingredients have
certain measurable qualities associated
with them.
where x is the variable representing the

quantity to be made and M is the market
limitation.
If it is necessary that the products of

blending have qualities within certain
limits, then constraints will result. Such
constraints may involve, for example,
quantities of nutrients in foods, octane
values for petrols and strengths of
materials.
Minimum marketing limitations might also

be imposed if it were necessary to make at
least a certain amount of a product to
satisfy some demand.
Such constraints would be of the form:
xL
Sometimes Inequalities might be =
constraints instead if some demand had to
be met exactly.
We now turn our attention to a number of

more abstract considerations concerning
constraints that we feel a model builder
should be aware of.
7.2.2.4 Material balance (continuity)
7.2.2.6 Hard and soft constraints
constraints
A linear programming constraint such as
It is often necessary to represent the fact

that the sum total of the quantities going
into some process equals the sum total
coming out. For example, the weight of the
final product should be equal to the total
weight of the ingredients. Such conditions
are often easily overlooked. Material
obviously rules out any solutions in which

the sum over j exceeds the quantity b.
363
There are some situations where this is

unrealistic. For example, if pervious
equation represented a productive
capacity limitation or a raw material
availability, there might be practical
circumstances in which this limitation
would be overruled.
If Inequality were a constraint, an

analogous effect could be achieved by a
slack variable. If Inequality be an equality
constraint, it would be possible to allow
the right-hand side coefficient b to be
overreached
or
underreached
by
modelling it as
It might sometimes be worthwhile or

necessary to buy in extra capacity or raw
materials at a high price. In such
circumstances, this equation would be an
unrealistic representation of the situation.
and giving u and v appropriately weighted

coefficients in the objective function.
Other circumstances exist in which it might

be impossible to violate it. For example,
the equation might be a technological
constraint imposed by the capacity of a
pipe whose cross section could not be
expanded.
It should be apparent that either u or v

must be zero in the optimal solution.
Any solution in which u and v came out
positive could be adjusted to produce a
better solution by subtracting the smaller
of u and v from both of them.
Constraints such as that, which cannot be

violated, are sometimes known as hard
constraints. It is often argued that what we
need are soft constraints that can be
violated at a certain cost. If the equation
were to be rewritten as
7.2.2.7 Conflicting constraints

It sometimes happens that a problem
involves a number of constraints, not all of
which can be satisfied simultaneously. A
conventional model would obviously be
infeasible. The objective in such a case is
sometimes stipulated to be to satisfy all
the constraints as nearly as possible.
and u were given a suitable positive

(negative) cost coefficient c for a
minimization (maximization) problem, we
would have achieved our desired effect.
The type of model that this situation gives

rise to is sometimes known as a goal
programming model. This term was
invented by Charnes and Cooper but the
type of model that results is still a linear
programming model.
The variable b would represent a capacity

or raw material availability that could be
expanded to b +u at a cost cu if the
optimization of the model found this to be
desirable. Possibly the surplus variable u
would be given a simple upper bound as
well to prevent the increase exceeding a
specified amount.
Each constraint is regarded as a goal that

must be satisfied as nearly as
possible. For example, if we wished to
impose the following constraints:
364
This variable represents the maximum

deviation.
We, therefore, have to impose extra
constraints.
but wanted to allow the possibility of them
not all being satisfied exactly, we would
replace them by the soft constraints
z ui 0, for all i,
z vi 0, for all i.
We are clearly using the same device

described before.
The objective function to be minimized is

simply the variable z . Clearly, the optimal
value of z will be no greater than the
maximum of ui and vi by virtue of
optimality.
Our objective would be to make sure that

each such constraint is as nearly satisfied
as possible. There are a number of
alternative ways of making such an
objective specific. Two possibilities are as
follows:
Nor can it be smaller than any of ui or vi by

virtue of the constraints. Therefore, the
optimal value of z will be both as small as
possible and exactly equal to the maximum
deviation of j aijxj its corresponding righthand side bi
1. Minimize the sum total of the deviations

of the row activities j aijxj
7.2.2.8 Simple and generalized upper
from the right-hand side values bi .
It was pointed out that marketing

constraints often take the particularly
simple forms of In-equality. Simple bounds
on a variable such as these are more
efficiently dealt with through a
modification of the simplex algorithm.
Most commercial package programs use
this modification. Such bounds are not,
therefore, specified as conventional
constraints but specified as simple bounds
for the appropriate variables. A
generalization of simple bounds known as
generalized upper bounds (GUBs) also
proves to be of great computational value
in solving a model and therefore worth
recognizing. Such constraints are usually
bounds
2. Minimize the maximum such deviation

over the constraints.
For many practical problems which of the
above objectives is used is not very
important.
Objective 1 can be dealt with by defining
an objective function consisting of the sum
of the slack (u) and surplus (v) variables in
the constraints
Possibly, these variables might be
weighted in the objective with non-unit
coefficients to reflect the relative
importance of different constraints
Objective 2 is slightly more complicated to
deal with but can surprisingly still be
accomplished in a linear programming
model. An extra variable z is introduced.
365
written as
belong to the GUB set, if it can belong to

no others. If variables are specified as
belonging to GUB sets, it is not necessary
to specify the corresponding constraints
such as the Equation
A further modification of the simplex
algorithm copes with the implied
constraint in an analogous way to simple
bounding constraints.
The set of variables xj is said to have a GUB

of b. This means that the sum of
these variables must be b. If Equation were
a constraint instead of an equality, the
There is normally only virtue in using the

GUB modification of the simplex algorithm
if a large number of GUB sets can be
isolated.
For
example,
in
the
transportation problem, it can be seen that
at least half of the constraints can be
regarded as GUB sets.
addition of a slack variable would

obviously convert it to the form of the
equation
The fact that the coefficients of the
variables in Equation are unity is not very
important, as scaling can always convert
any constraint with all non-negative
coefficients into this form. What is,
however, important is that when a
number of constraints such as the
Equation exist, the variables in them form
exclusive sets. A set of variables is said to
The computational advantages of

recognizing GUB constraints can be very
great indeed with large models. A way of
detecting a large number of such sets if
they exist in a model is described by
Brearley et al . (1975).
Optimizaiton
Problem types
Linear
Quadratic
Integer
Stochastic
Dynamic
366
Non-linear
It is important for the reader to appreciate,

at the outset, that the "programming" in
Linear Programming is of a different flavor
than the "programming" in Computer
Programming. In the former case, it means
to plan and organize as in "Get with the
program!", it programs you by its
solution. While in the latter case, it means
to write codes for performing calculations.
Training in one kind of programming has
very little direct relevance to the other. In
fact, the term "linear programming" was
coined before the word "programming"
became closely associated with computer
software. This confusion is sometimes
avoided by using the term linear
optimization as a synonym for linear
programming.
7.3 Optimization Problem

Types
7.3.1 Linear Programming
Linear
Programming
(LP)
is
a
mathematical procedure for determining
optimal allocation of scarce resources. LP is
a procedure that has found practical
application in almost all facets of business,
from advertising to production planning.
Transportation,
distribution,
and
aggregate production planning problems
are the most typical objects of LP analysis.
In the petroleum industry, for example a
data processing manager at a large oil
company recently estimated that from 5 to
10 percent of the firm's computer time was
devoted to the processing of LP and LP-like
models.
Any LP problem consists of an objective

function and a set of constraints. In most
cases, constraints come from the
environment in which you work to achieve
your objective. When you want to achieve
the desirable objective, you will realize
that the environment is setting some
constraints
(i.e.,
the
difficulties,
restrictions) in fulfilling your desire or
objective. This is why religions such as
Buddhism, among others, prescribe living
an abstemious life. No desire, no pain. Can
you take this advice with respect to your
business objective?
Linear programming deals with a class of

programming problems where both the
objective function to be optimized is linear
and all relations among the variables
corresponding to resources are linear. This
problem was first formulated and solved in
the late 1940's. Rarely has a new
mathematical technique found such a wide
range of practical business, commerce, and
industrial applications and simultaneously
received so thorough a theoretical
development, in such a short period of
time. Today, this theory is being
successfully applied to problems of capital
budgeting, design of diets, conservation of
resources, games of strategy, economic
growth prediction, and transportation
systems. In very recent times, linear
programming theory has also helped
resolve and unify many outstanding
applications.
What is a function: A function is a thing

that does something. For example, a coffee
grinding machine is a function that
transform the coffee beans into powder.
The (objective) function maps and
translates the input domain (called the
feasible region) into output range, with the
two end-values called the maximum and
the minimum values.
367
When you formulate a decision-making

problem as a linear program, you must
check the following conditions:
steel", "produce stainless steel", and

"produce high carbon steel". Each activity
consumes or possibly contributes
additional amounts of the resources. There
must be an objective function, i.e. a way to
tell bad from good, from an even better
decision. The problem is to determine the
best combination of activity levels, which
do not use more resources than are
actually available. Many managers are
faced with this task everyday. Fortunately,
when a well-formulated model is input,
linear programming software helps to
determine the best combination.
1. The objective function must

be linear. That is, check if all
variables have power of 1 and they
are added or subtracted (not
divided or multiplied)
2. The objective must be either
maximization or minimization of a
linear function. The objective must
represent the goal of the decisionmaker
The Simplex method is a widely used

solution algorithm for solving linear
programs. An algorithm is a series of steps
that will accomplish a certain task.
3. The constraints must also be linear.

Moreover, the constraint must be
of the following forms ( , , or =,
that is, the LP-constraints are
always closed).
For example, the following problem is not
an LP: Max X, subject to X < 1. This very
simple problem has no solution.
7.3.1.2 LP Problem Formulation

Any linear program consists of four parts: a
set of decision variables, the parameters,
the objective function, and a set of
constraints. In formulating a given decision
problem in mathematical form, you should
practice understanding the problem (i.e.,
formulating a mental model) by carefully
reading and re-reading the problem
statement. While trying to understand the
problem, ask yourself the following
general questions:
As always, one must be careful in

categorizing an optimization problem as an
LP problem. Here is a question for you. Is
the following problem an LP problem?
Max
subject
X1
X12 - 4 0
X2
to:
X2 0
Although the second constraint looks "as

if" it is a nonlinear constraint, this
constraint can equivalently be written as:
X1 -2,
and
X2 2.
Therefore, the above problem is indeed an
LP problem.
1. What are the decision variables?

That is, what are controllable
inputs? Define the decision
variables
precisely,
using
descriptive names. Remember that
the controllable inputs are also
known as controllable activities,
decision variables, and decision
activities.
For most LP problems one can think of two

important classes of objects: The first is
limited resources such as land, plant
capacity, or sales force size; the second, is
activities such as "produce low carbon
368
2. What are the parameters? That is,

what are the uncontrollable
inputs? These are usually the given
constant numerical values. Define
the parameters precisely, using
descriptive names.
7.3.1.3 Geometry of Linear Programming

7.3.1.3.1 Polyhedra and convex sets
In this section, we introduce some
important concepts that will be used to
study the geometry of linear programming,
including a discussion of convexity.
3. What is the objective? What is the

objective function? Also, what does
the owner of the problem want?
How the objective is related to his
decision variables? Is it a
maximization or minimization
problem? The objective represents
the goal of the decision-maker.
Hyperplanes, halfspaces, and polyhedra

We start with the formal definition of a
polyhedron.
Definition :A polyhedron is a set that
can be described in the
form {x lRn I Ax b}, where A is a
m x n matrix and b is a vector in lRm .
4. What are the constraints? That is,

what requirements must be met?
Should I use inequality or equality
type of constraint? What are the
connections among variables?
Write them out in words before
putting them in mathematical
form.
the feasible set of any linear programming

problem can be described by inequality
constraints of the form Ax b, and i s
therefore a polyhedron. I n particular, a set
o f the form {x Rn I Ax = b, x o} is also a
polyhedron and will be referred to as a
polyhedron in standard form.
Learn that the feasible region has nothing

or little to do with the objective function
(min or max). These two parts in any LP
formulation come mostly from two distinct
and different sources. The objective
function is set up to fulfill the decisionmaker's desire (objective), whereas the
constraints which shape the feasible
region usually comes from the decisionmaker's environment putting some
restrictions/conditions
on
achieving
his/her objective.
A polyhedron can either "extend to

infinity," or can be confined in a finite
region. The definition that follows refers to
this distinction.
Definition : A set S |Rn is bounded if
there exists a constant
K such that the absolute value of every
component of every element of S is
less than or equal to K.
The next definition deals with polyhedra
determined by a single linear constraint .
369
Convex Sets
We now define the important notion of a
convex set .
Figure 7.4: The convex hull of seven points in R2

Note that if A [0, 1] ' then Ax + ( 1 - A)y is
a weighted average of the vectors x, y, and
therefore belongs to the line segment
joining x and y.
Extreme points, vertices , and basic
Our next definition refers to weighted

averages o f a finite number of vectors; see
Figure 7 . 4 .
We observed before that an optimal

solution to a linear programming problem
tends to occur at a "corner" of the
polyhedron over which we are optimizing.
In this section, we suggest three different
ways of defining the concept of a "corner"
and then show that all three definitions are
equivalent .
Figure 7.3: The set S is convex, but the set Q is not
Our first definition defines an extreme

point of a polyhedron as a point that
cannot be expressed as a convex
combination of two other elements of
the polyhedron, and is illustrated in
Figure 7 .5. Notice that this definition is
entirely geometric and does not refer to
a specific representation of a
polyhedron in terms of linear constraints.
Thus, a set is convex if the segment joining

any two of its elements is contained in the
set ; see Figure 7 . 3 .
, because the segment joining x and y is not contained

in Q .
370
Since the total cost is the sum of the fixed

cost (F) and the variable cost per unit
multiplied by the number of units
produced. Therefore, the decision problem
is to find X1 and X2 such that:
Maximize 9X1 + 6X2 [(1.5X1 + X2) +
(2.5X1 + 2X2) + F1 + F2],
where X1 and X2 stand for the number of
tables and chairs; the cost terms in the
brackets are the raw material, and labor
costs respectively. F1 and F2 are the fixed
costs for the two products respectively.
Without loss of generality, and any impact
on optimal solution, let us set F1 = 0, and
F2 = 0. The objective function reduces to
the following net profit function:
Figure 7.5:
The vector w is not an extreme point

because it is a convex combination of v and u. The vector x
is an extreme point
7.3.1.4 Linear Programming Examples

The
Carpenter's
Problem:
Allocating Scarce Resources Among
Competitive Means
Maximize 5X1 + 3X2
During a couple of brain-storming

sessions with a carpenter (our client), he
told us that he, solely, makes tables and
chairs, sells all tables and chairs he makes
at a market place, however, does not have
a stable income, and wishes to do his best.
That is, the net incomes (say, in dollars, or

tens of dollars) from selling X1 tables and
X2 chairs.
The
constraining
factors
which,
usually come from outside, are the
limitations on labors (this limitation comes
from his family) and raw material
resources (this limitation comes from
scheduled delivery). Production times
required for a table and a chair are
measured at different times of day, and
estimated to be 2 hours and 1 hour,
respectively. Total labor hours per week
are only 40 hrs. Raw materials required for
a table and a chair are 1, and 2 units
respectively. Total supply of raw material is
50 units per week. Therefore, the LP
formulation is:
The objective is to find out how many

tables and chairs he should make
to maximize net income. We begin by
focusing on a time frame, i.e., planning
time-horizon, to revise our solution weekly
if needed. To learn more about his
problem, we must go to his shop
and observe what is going on and
measure what we need to formulate (to
give a Form, to make a model) of his
problem. We must confirm that his
objective is to maximize net income. We
must communicate with the client.
Maximize 5 X1 + 3 X2
The carpenter's problem deals with finding

out how many tables and chairs to make
per week; but first an objective function
must be established:
Subject
to:
2 X1 + X2 40 labor constraint
X1 + 2 X2 50 material constraint
and both X1, X2 are non-negative.
371
This is a mathematical model for the

carpenter's
problem.
The decision
variables, i.e., controllable inputs are X1,
and X2. The output for this model is the
total net income 5 X1 + 3 X2. All functions
used in this model are linear (the decision
variable have power equal to 1). The
coefficients of these constraints are
called Technological Factors (matrix). The
review period is one week, an appropriate
period within which the uncontrollable
inputs (all parameters such as 5, 50, 2,..)
are less likely to change (fluctuate). Even
for such a short planning time-horizon, we
must perform the what-if analysis to react
to any changes in these inputs in order
to control the problem, to update the
prescribed solution.
and effective methodologies, known as

the Linear
Programming
Solution
Techniques are based on simultaneous
thinking are commercially available in over
400 different software packages from all
over the world.
The optimal solution , optimal strategy, is
to make X1 = 10 tables, and X2 = 20 chairs.
We may program the carpenter's weekly
activities to make 10 tables and 20 chairs.
With this (optimal) strategy, the net
income is $110. This prescribed
solution was a surprise for the carpenter
since, because of more net income of
selling a table ($5), he used to make more
tables than chairs!
Hire or Not? Suppose the carpenter can
hire someone to help at a cost of $2 per
hour. This is, in addition, hourly-based
wage he/she is currently paying; otherwise
$2 is much lower than the current
minimum wage in US. Should the
carpenter hire and if yes then for how may
hours?
Notice that since the carpenter is not going

out of business at the end of the planning
horizon, we added the conditions that both
X1, X2 must be non-negative instead of the
requirements that X1, and X2 must be
positive integers. The non-negativity
conditions are also known as "implied
constraints." Again, a Linear Program
would be fine for this problem if the
carpenter were going to continue to
manufacture these products. The partial
items would simply be counted as work in
progress and would eventually become
finished goods say, in the next week.
Let X3 be the number of extra hours, then

the modified problem is:
Maximize 5 X1 + 3 X2 - 2 X3
Subject
to:
2 X1 + X2 40 + X3 labor constraint with
unknown
additional
hours
X1 + 2 X2 50 material constraint
We may try to solve for X1 and X2 by listing

possible solutions for each and selecting
the pair (X1, X2) that maximize 5X1 + 3X2
(the net income). However, it is too time
consuming to list all possible alternatives
and if the alternatives are not exhaustively
listed, we cannot be sure that the pair we
select (as a solution) is the best of all
alternatives. This way of solving a problem
is known as "sequential thinking" versus
"simultaneous thinking". More efficient
Under this new condition, we will see that

the optimal solution is X1 = 50, X2 = 0, X3 =
60, with optimal net income of $130.
Therefore, the carpenter should be hired
for 60 hours. What about only hiring 40
hours?
A Product-Replacement Problem
A price-taking firm sells S units of its
product at the market price of p.
372
Management's policy is to replace

defective units at no additional charge, on
the first-come, first-served basis, while
replacement units are available. Because
management does not want to risk making
the same mistake twice, it produces the
units that it sells to the market on one
machine. Moreover, it produces the
replacement units, denoted X, on a second,
higher-quality machine. The fixed cost
associated with operating both machines,
the variable cost, and replacement cost are
given is the short-run cost function C(X) =
100 + 20S + 30X.
nutritional and cost information in the

following table:
The problem is to find a diet (a choice of
the numbers of units of the two foods) that
meets
all
minimum
nutritional
requirements at minimal cost.
a. Formulate the problem in terms of
linear inequalities and an objective
function.
b. Solve the problem geometrically.
c. Explain how the 2:1 cost ratio
(steak to potatoes) dictates that
the solution must be where you
said it is.
The exact probability that a unit will be

defective is r. Acting out of caution,
however,
management
always
underestimate the reliability of its product.
Nonetheless, it imposes the condition that
X r.S. The demand for the firm's product
is given by S(r) = 10000e-0.2r. Hence the
decision problem is to maximize the net
profit function P(X):
d. Find a cost ratio that would move

the optimal solution to a different
choice of numbers of food units,
but that would still require buying
both steak and potatoes.
e. Find a cost ratio that would dictate
buying only one of the two foods in
order to minimize cost.
Maximize P(X) = 100000p e- 0.2r - 100 - 20S

30X
subject to: X r.S.
a) We begin by setting the constraints for

the problem. The first constraint
represents the minimum requirement for
carbohydrates, which is 8 units per some
unknown amount of time. 3 units can be
consumed per unit of potatoes and 1 unit
can be consumed per unit of steak. The
second
constraint
represents
the
minimum requirement for vitamins, which
is 19 units. 4 units can be consumed per
unit of potatoes and 3 units can be
consumed per unit of steak. The third
constraint represents the minimum
requirement for proteins, which is 7 units.
1 unit can be consumed per unit of
potatoes and 3 units can be consumed per
unit of steak. The fourth and fifth
constraints represent the fact that all
As we will learn, the solutions to the LP

problems are at the vertices of the feasible
region. Therefore, the net profit P(X) will
be maximized if the management set X =
r.S.
A Diet Problem
Suppose the only foods available in your
local store are potatoes and steak. The
decision about how much of each food to
buy is to made entirely on dietary and
economic considerations. We have the
373
feasible solutions must be nonnegative

because we can't buy negative quantities.
unit time to meet the minimum nutritional

requirements.
constraints:
{3X1 + X2 8, 4X1+ 3X2 19, X1+ 3X2 7,
X1 0, X2 0};
Next we plot the solution set of the
inequalities to produce a feasible region of
possibilities.
c) The 2:1 cost ratio of steak to potatoes
dictates that the solution must be here
since, as a whole, we can see that one unit
of steak is slightly less nutritious than one
unit of potatoes. Plus, in the one category
where steak beats potatoes in healthiness
(proteins), only 7 total units are necessary.
Thus it is easier to fulfill these units without
buying a significant amout of steak. Since
steak is more expensive, buying more
potatoes to fulfill these nutritional
requirements is more logical.
7.3.2 Linear Integer Programming

7.3.2.1 Explanation
problem is a mathematical optimization or
feasibility program in which some or all of
the variables are restricted to be integers.
In many settings the term refers to integer
linear programming (ILP), in which the
objective function and the constraints
(other than the integer constraints) are
linear.
d) Now we choose a new cost ratio that will

move the optimal solution to a different
choice of numbers of food units. Both
steak and potatoes will still be purchased,
but a different solution will be found. Let's
try a 5:2 cost ratio.
Standard Form
An integer linear program in canonical
form
is
expressed
as:[

and an ILP in standard form is expressed as
Thus, the optimal solution for this cost

ratio is buying 8 steaks and no potatoes per
374
value of 2.8. Note that if the solution of the

relaxation is rounded to the nearest
integers, it is not feasible for the ILP.
Different Forms of ILP
Mixed integer linear programming (MILP)
involves problems in which only some of
the variables,
, are constrained to be
integers, while other variables are allowed
to be non-integers.
Fig 7.6 where the entries of are vectors

and is a matrix, having integer values.
Zero-one linear programming involves

problems in which the variables are
restricted to be either 0 or 1. Note that any
bounded integer variable can be expressed
as a combination of binary variables.
Example
The graph on the right shows the following
problem.
7.3.2.2 Applications
There are two main reasons for using
integer variables when modeling problems
as a linear program:
The integer variables represent quantities
that can only be integer. For example, it is
not possible to build 3.7 cars.
The integer variables represent decisions
and so should only take on the value 0 or 1
.
The feasible integer points are shown in

red, and the red dashed lines indicate their
convex hull, which is the smallest
polyhedron that contains all of these
points. The blue lines together with the
coordinate axes define the polyhedron of
the LP relaxation, which is given by the
inequalities without the integrality
constraint. The goal of the optimization is
to move the black dotted line as far
upward while still touching the
polyhedron. The optimal solutions of the
integer
problem
are
the
These considerations occur frequently in

practice and so integer linear programming
can be used in many applications areas,
some of which are briefly described below.
Production planning
Mixed integer programming has many
applications in industrial production,
including job-shop modelling. One
important example happens In agricultural
production planning involves determining
production yield for several crops that can
share resources ( Land, labor, capital,
seeds, fertilizer). A possible objective is to
maximize the total production, without
points
and
which both have an
objective value of 2. The unique optimum
of the relaxation is
with objective
375
exceeding the available resources. In some

cases, this can be expressed in terms of a
linear program, but variables must be
constrained to be integer.
7.3.2.3 Algorithms
Exact
there are a variety of algorithms that can
be used to solve integer linear programs
exactly. One class of algorithms are cutting
plane methods which work by solving the
LP relaxation and then adding linear
constraints that drive the solution towards
being integer without excluding any
integer feasible points.
Scheduling
These problems involve service and vehicle
scheduling in transportation networks. For
example, a problem may involve assigning
buses or subways to individual routes so
that a timetable can be met, and also to
equip them with drivers. Here binary
decision variables indicate whether a bus
or subway is assigned to a route and
whether a driver is assigned to a particular
train or subway.
Another class of algorithms are variants of

the branch and bound method. For
example, the branch and cut method that
combines both branch and bound and
cutting plane methods. Branch and bound
algorithms have a number of advantages
over algorithms that only use cutting
planes. One advantage is that the
algorithms can be terminated early and as
long as at least one integral solution has
been found, a feasible, although not
necessarily optimal, solution can be
returned. Further, the solutions of the LP
relaxations can be used to provide a worstcase estimate of how far from optimality
the returned solution is. Finally, branch
and bound methods can be used to return
multiple optimal solutions , this methods
are explained in further parts of this books.
Telecommunications networks
The goal of these problems is to design a
network of lines to install so that a
predefined set of communication
requirements are met and the total cost of
the network is minimal. This requires
optimizing both the topology of the
network along with the setting the
capacities of the various lines. In many
cases, the capacities are constrained to be
integer quantities. Usually there are,
depending on the technology used,
additional restrictions that can be modeled
as a linear inequalities with integer or
binary variables.
Heuristic
many problem instances are intractable
and so heuristic methods must be used
instead. For example, tabu search can be
used to search for solutions to ILPs.To use
tabu search to solve ILPs, moves can be
defined as incrementing or decrementing
an integer constrained variable of a
feasible solution, while keeping all other
integer-constrained variables constant.
The unrestricted variables are then solved
for. Short term memory can consist of
Cellular networks
The task of frequency planning in GSM
mobile networks involves distributing
available frequencies across the antennas
so that users can be served and
interference is minimized between the
antennas. This problem can be formulated
as an integer linear program in which
binary variables indicate whether a
frequency is assigned to an antenna.
376
previous tried solutions while medium

term memory can consist of values for the
integer constrained variables that have
resulted in high objective values (assuming
the ILP is a maximization problem). Finally,
long term memory can guide the search
towards integer values that have not
previously been tried.
minimizes
subject to
where xT denotes
the
vector transpose of X.
The
notation Ax b means that every entry of
the vector Ax is less than or equal to the
corresponding entry of the vector b.
Other heuristic methods that can be

applied to ILPs include
Hill climbing
Simulated annealing
Ant colony optimization
For general problems a variety of methods

are commonly used, including
neural networks
7.3.3 Quadratic programming
interior point,
is a special type of mathematical

optimization problem. It is the problem of
optimizing (minimizing or maximizing) a
quadratic function of several variables
subject to linear constraints on these
variables.
active set
gradient projection,
Example
Solve a simple quadratic programming
problem: find values of x that minimize
f(x)=(1/2)x21+x22x1x22x16x2
The quadratic programming problem

with n variables and m constraints can be
formulated
as
follows.
subject to:
x1 + x2 2
x1 + 2x2 2
2x1 + x2 3
0 x1 , 0 x2.
Given:
a
real-valued, n-dimensional
vector c,
an n n-dimensional
real Symmetric matrix Q,
an m n-dimensional
matrix A, and
In matrix notation this is

f(x)=(1/2)xTQx +fTx,
where
real
Q = [1 -1; -1 2]
f = [-2; -6]
an m-dimensional real vector b,
A = [1 1; -1 2; 2 1]
the objective of quadratic programming is

to find an n-dimensional vector x, that
b = [2; 2; 3]
377
lb = zeros(2,1)
visible for the material of this chapter

where
simplex-like
techniques
predominate.
from this point we can solve the problem

by using any optimization tool or by using
one of the previously indicated methods as
we will discuss later.
7.3.4.2 NLP Problems Understanding

A general optimization problem is to select
n decision variables x1, x2, . . . , xn from a
given feasible region in such a way as to
optimize (minimize or maximize) a given
objective function f(x1, x2, . . . , xn) of the
decision variables. The problem is called a
nonlinear programming problem (NLP) if
the objective function is nonlinear and/or
the feasible region is determined by
nonlinear
constraints.
Thus,
in
maximization form, the general nonlinear
program is stated as:
7.3.4 Non-Linear Programming

7.3.4..1 Introduction
Numerous
mathematical-programming
applications, including many introduced in
previous chapters, are cast naturally as
linear programs. Linear programming
assumptions or approximations may also
lead
to
appropriate
problem
representations over the range of decision
variables being considered. At other times,
though, nonlinearities in the form of either
nonlinear objective functions or nonlinear
constraints are crucial for representing an
application properly as a mathematical
program. This chapter provides an initial
step
toward
coping
with
such
nonlinearities, first by introducing several
characteristics of nonlinear programs and
then by treating problems that can be
solved using simplex-like pivoting
procedures. As a consequence, the
techniques to be discussed are primarily
algebra-based. The final two sections
comment on some techniques that do not
involve pivoting. As our discussion of
nonlinear programming unfolds, the
reader is urged to reflect upon the linear
programming theory that we have
developed previously, contrasting the two
theories to understand why the nonlinear
problems are intrinsically more difficult to
solve. At the same time, we should try to
understand the similarities between the
two theories, particularly since the
nonlinear results often are motivated by,
and are direct extensions of, their linear
analogs. The similarities will be particularly
Maximize : f(x1, x2, . . . , xn),

subject to:
g1(x1, x2, . . . , xn) b1,
gm(x1, x2, . . . , xn) bm,
where each of the constraint functions g1
through gm is given. A special case is the
linear program that has been treated
previously. The obvious association for this
case is
And
Note that non-negativity restrictions on

variables can be included simply by
appending the additional constraints:
gm+i(x1, x2, . . . , xn) = xi 0 (i = 1, 2, . . ,n).
378
Sometimes these constraints will be

treated explicitly, just like any other
problem constraints. At other times, it will
be convenient to consider them implicitly
in the same way that non-negativity
constraints are handled implicitly in the
simplex
method.
For
notational
convenience, we usually let x denote the
vector of n decision variables x1, x2, . . . ,
xn that is, x = (x1, x2, . . . , xn) and
write the problem more concisely as
Example
Portfolio Selection
An investor has $5000 and two potential
investments. Let xj for j = 1 and j = 2
denote his allocation to investment j in
thousands of dollars.
From historical data, investments 1 and 2
have an expected annual return of 20 and
16 percent, respectively. Also, the total risk
involved with investments 1 and 2, as
measured by the variance of total return, is
given by 2x1 2+x 2 2 +(x1+x2) 2 ,
Maximize f(x),
subject to:
so that risk increases with total

investment and with the amount of each
individual investment. The investor would
like to maximize his expected return and at
the same time minimize his risk.
gi(x) bi (i = 1, 2, . . . , m).
As in linear programming, we are not
restricted to this formulation. To minimize
f(x), we can of course maximize f(x).
Equality constraints h(x) = b can be written
as two inequality constraints h(x) b and
h(x) b. In addition, if we introduce a
slack variable, each inequality constraint is
transformed to an equality constraint.
Thus sometimes we will consider an
alternative equality form:
Clearly, both of these objectives cannot, in

general, be satisfied simultaneously. There
are several possible approaches. For
example, he can minimize risk subject to a
constraint imposing a lower bound on
expected return. Alternatively, expected
return and risk can be combined in an
objective function, to give the model:
Maximize f(x) = 20x1 + 16x2 [2x1 2 + x 2 2
+ (x1 + x2) 2 ],
Maximize f(x),
subject to:
hi(x) = bi (i = 1, 2, . . . , m) xj 0 (j = 1, 2, .
. . , n).
subject to:
g1(x) = x1 + x2 5, x1 0, x2 0,
Usually the problem context suggests

either an equality or inequality formulation
(or a formulation with both types of
constraints), and we will not wish to force
the problem into either form. The
following three simplified examples
illustrate how nonlinear programs can
arise in practice.
(that is, g2(x) = x1, g3(x) = x2). The

nonnegative constant reflects his
tradeoff between risk and return.
If = 0, the model is a linear program, and
he will invest completely in the investment
with greatest expected return. For very
large , the objective contribution due to
expected return becomes negligible and he
is essentially minimizing his risk.
379
These algorithms maintain a population of

candidate solutions, rather than a single
best solution so far. From existing
candidate solutions, they generate new
solutions
through
either
random mutation of
single
points
or crossover or recombination of two or
more existing points. The population is
then subject to selection that tends to
eliminate the worst candidate solutions
and keep the best ones. This process is
repeated, generating better and better
solutions; however, there is no way for
these methods to determine that a given
solution is truly optimal.
7.3.4.3 Important Idioms in NLP

Non-Smooth problems
The most difficult type of optimization
problem to solve is a non-smooth problem
(NSP). Such a problem normally is, or must
be assumed to be non-convex. Hence it
may not only have multiple feasible
regions and multiple locally optimal points
within each region -- because some of the
functions are non-smooth or even
discontinuous, derivative or gradient
information generally cannot be used to
determine the direction in which the
function is increasing (or decreasing). In
other words, the situation at one possible
solution gives very little information about
where to look for a better solution.
Tabu Search and Scatter Search offer

another approach to find "good" solutions
to
non-smooth
optimization
problems. These algorithms also maintain
a population of candidate solutions, rather
than a single best solution so far, and they
generate new solutions from old
ones. However, they rely less on random
selection and more on deterministic
methods. Tabu search uses memory of
past search results to guide the direction
and intensity of future searches. These
methods generate successively better
solutions, but as with genetic and
evolutionary algorithms, there is no way
for these methods to determine that a
given solution is truly optimal.
In all but the simplest problems, it is

impractical to exhaustively enumerate all
of the possible solutions and pick the best
one, even on a fast computer. Hence, most
methods rely on some sort of random
sampling of possible solutions.
Such
methods
or stochastic .
are nondeterministic
they may yield different solutions on

different runs, even when started from the
same point on the same model, depending
on which points are randomly sampled.
Genetic or Evolutionary Algorithms offer
one way to find "good" solutions to nonsmooth optimization problems. (In a
genetic algorithm the problem is encoded
in a series of bit strings that are
manipulated by the algorithm; in an
"evolutionary algorithm," the decision
variables and problem functions are used
directly. Most commercial Solver products
are based on evolutionary algorithms.)
Local VS Global Optimum

Geometrically, nonlinear programs can
behave much differently from linear
programs, even for problems with linear
constraints.
In Fig. 7.7, the portfolio-selection example
from the last section has been plotted for
several values of the tradeoff parameter .
For each fixed value of , contours of
constant objective values are concentric
380
ellipses. As Fig. 7.7 shows, the optimal

solution can occur: a) at an interior point of
the feasible region; b) on the boundary of
the feasible region, which is not an
extreme point; or c) at an extreme point of
the feasible region. As a consequence,
procedures, such as the simplex method,
that search only extreme points may not
determine an optimal solution. Figure 7.8
illustrates another feature of nonlinearprogramming problems. Suppose that we
are to minimize f(x) in this example, with 0
x 10. The point x = 7 is optimal. Note,
however, that in the indicated dashed
interval, the point x = 0 is the best feasible
point; i.e., it is an optimal feasible point in
the local vicinity of x = 0 specified by the
dashed interval. The latter example
illustrates that a solution optimal in a local
sense need not be optimal for the overall
problem. Two types of solution must be
distinguished. A global optimum is a
solution to the overall optimization
problem. Its objective value is as good as
any other point in the feasible region. A
local optimum, on the other hand, is
optimal only with respect to feasible
solutions close to that point. Points far
removed from a local optimum play no role
in its definition and may actually be
preferred to the local optimum. Stated
more formally,
whenever each variable yj is within of xj

that is, xj yj xj + and y is
feasible, then f(x) f(y).
Figure 7.7 Portfolio-selection example for various values

of . (Lines are contours of constant objective values.)
Definition. Let x = (x1, x2, . . . , xn) be a

feasible solution to a maximization
problem with objective function f(x). We
call x
1. A global maximum if f(x) f(y) for every
feasible point y = (y1, y2, . . . , yn);
2. A local maximum if f(x) f(y) for every
feasible point y = (y1, y2, . . . , yn)
sufficiently close to x. That is, if there is a
number > 0 (possibly quite small) so that,
Figure 7.8 Local and global minima
381
Global and local minima are defined

analogously. The definition of local
maximum simply says that if we place an ndimensional box (a cube in three
dimensions) about x, whose side has
length 2, then f(x) is as small as f(y) for
every feasible point y lying within the box.
(Equivalently, we can use n-dimensional
spheres in this definition.)
The dynamic programming technique,

when
applicable,
represents
or
decomposes a multistage decision
problem as a sequence of single-stage
decision problems. Thus an N-variable
problem is represented as a sequence of N
single-variable problems that are solved
successively. In most cases, these N subproblems are easier to solve than the
original problem.
For instance, if = 1 in the above example,

the one-dimensional box, or interval, is
pictured about the local minimum x = 0 in
Fig. 7.8 The concept of a local maximum is
extremely important.
The decomposition to N sub-problems is

done in such a manner that the optimal
solution of the original N-variable problem
can be obtained from the optimal solutions
of the N one-dimensional problems. It is
important to note that the particular
optimization technique used for the
optimization of the N single-variable
problems is irrelevant.
As we shall see, most general-purpose

non-linear programming procedures are
near-sighted and can do no better than
determine local maxima. We should point
out that, since every global maximum is
also a local maximum, the overall
optimization problem can be viewed as
seeking the best local maxima.
It may range from a simple enumeration

process to a differential calculus or a
nonlinear programming technique.
Multistage decision problems can also be
solved by direct application of the classical
optimization techniques. However, this
requires the number of variables to be
small, the functions involved to be
continuous
and
continuously
differentiable, and the optimum points not
to lie at the boundary points.
7.3.5 Dynamic Programming

In most practical problems, decisions have
to be made sequentially at different points
in time, at different points in space, and at
different levels, say, for a component, for a
subsystem, and/or for a system.
The problems in which the decisions are to
be made sequentially are called sequential
decision problems. Since these decisions
are to be made at a number of stages, they
are also referred to as multistage decision
problems.
Further, the problem has to be relatively

simple so that the set of resultant
equations can be solved either analytically
or
numerically.
The
nonlinear
programming techniques can be used to
solve slightly more complicated multistage
decision problems. But their application
requires the variables to be continuous
and prior knowledge about the region of
the global minimum or maximum.
Dynamic programming is a mathematical

technique well suited for the optimization
of multistage decision problems. This
technique was developed by Richard
Bellman in the early 1950s.
382
In all these cases, the introduction of

stochastic variability makes the problem
extremely complex and renders the
problem unsolvable except by using some
sort of an approximation such as chance
constrained programming.
the optimal value of temperature in the

heater, the reaction rate in the reactor,
and the number of trays in the distillation
tower such that the cost of the process is
minimum while satisfying all the
restrictions placed on the process. Figure
7.9 shows a missile resting on a launch pad
that is expected to hit a moving aircraft
(target) in a given time interval. The target
will naturally take evasive action and
attempts to avoid being hit.
Dynamic programming, on the other hand,

can deal with discrete variables, nonconvex, non-continuous, and nondifferentiable functions. In general, it can
also take into account the stochastic
variability by a simple modication of the
deterministic procedure.
The problem is to generate a set of

commands to the missile so that it can hit
the target in the specied time interval.
The Dynamic Programming technique

suffers from a major drawback, known as
the curse of dimensionality. How- ever,
despite this disadvantage, it is very suitable
for the solution of a wide range of complex
problems in several areas of decision
making.
This can be done by observing the target

and, from its actions, generate periodically
a new direction and speed for the missile.
7.3.5.2 MULTISTAGE DECISION

PROCESSES
7.3.5.2.1 Denition and Examples
As applied to dynamic programming, a
multistage decision process is one in which
a number of single-stage processes are
connected in series so that the output of
one stage is the input of the succeeding
stage.
Figure 7.9 Ground-radar-controlled missile chasing a

moving target.
Next, consider the minimum cost design of

a water tank. The system consists of a tank,
a set of columns, and a foundation. Here
the tank supports the water, the columns
support the weights of water and tank, and
the foundation supports the weights of
water, tank, and columns. The
components can be seen to be in series
and the system has to be treated as a
multistage decision problem.
Strictly speaking, this type of process

should be called a serial multistage
decision process since the individual stages
are connected head to tail with no recycle.
Serial multistage decision problems arise in
many types of practical problems.
A few examples are given below and many
others can be found in the literature.
Consider a chemical process consisting of a
heater, a reactor, and a distillation tower
connected in series. The objective is to nd
Finally, consider the problem of loading a

vessel with stocks of N items. Each unit of
383
item i has a weight wi and a monetary value

ci . The maximum permissible cargo weight
is W. It is required to determine the cargo
load that corresponds to maximum
monetary value without exceeding the
limitation of the total cargo weight.
Although the multistage nature of this
problem is not directly evident, it can be
posed as a multistage decision problem by
considering each item of the cargo as a
separate stage.
7.11. Because of some convenience, which

will be seen later, the stages n, n 1, . . . ,
i, . . . , 2, 1 are labeled in decreasing order.
For the stage, the input state vector is
denoted by si+1 and the output state vector
as s.
Since the system is a serial one, the output
from stage i + 1 must be equal to the input
to stage i.
Hence the state transformation and return
functions can be represented as
7.3.5.2.2 Representation of a Multistage

Decision Process
Si=ti(Si+1 , Xi)
A single-stage decision process (which is a

component of the multistage problem) can
be represented as a rectangular block (Fig.
7.10). A decision process can be
characterized by certain input parameters,
S (or data), certain decision variables (X),
and certain output parameters (T)
representing the outcome obtained as a
result of making the decision.
Ri=ri(si+1 , Xi)
Figure 7.10 Single-stage decision problem.
The input parameters are called input state

variables, and the output parameters
are called output state variables.
Finally, there is a return or objective
function R, which measures the
effectiveness of the decisions made and
the output that results from these
decisions. For a single-stage decision
process shown in Fig. 7.10, the output is
related to the input through a stage
transformation function denoted by
Figure 7.11 Multistage decision problem (initial value

problem).
T = t(X, S)
Since the input state of the system
inuences the decisions we make, the
return function can be represented as
7.3.5.3 Conversion of a Non-serial System

to a Serial System
According to the denition, a serial system
is one whose components (stages) are
R = r(X, S)
A serial multistage decision process can be
represented schematically as shown in Fig.
384
connected in such a way that the output of

any component is the input of the
succeeding component. As an example of a
non-serial system, consider a steam power
plant consisting of a pump, a feed water
heater, a boiler, a super heater, a steam
turbine, and an electric generator, as
shown in Fig. 7.12. If we assume that some
steam is taken from the turbine to heat the
feed water, a loop will be formed as shown
in Fig. 7.12a.
This non-serial system can be converted to

an equivalent serial system by regrouping
the components so that a loop is redened
as a single element as shown in Fig. 9.4b
and c.
Figure 7.12 Serializing a nonserial system.
3. Boundary value problem. If the values of

both the input and output variables are
specied, the problem is called a boundary
value problem. The three types of
problems are shown schematically in Fig.
7.13, where the symbol | is used to
indicate a prescribed state variable.
Thus the new serial multistage system

consists of only three components: the
pump, the boiler and turbine system, and
the electric generator. This procedure can
easily be extended to convert multistage
systems with more than one loop to
equivalent serial systems.
7.3.5.4 Types of Multistage Decision
Problems
The serial multistage decision problems
can be classied into three categories as
follows.
1. Initial value problem. If the value of the
initial state variable, sn+1, is prescribed, the
problem is called an initial value problem.
Fig 7.13
2. Final value problem. If the value of the

nal state variable, s1 is prescribed, the
problem is called a nal value problem.
Notice that a nal value problem can be
transformed into an initial value problem
by reversing the directions of
7.3.6 Stochastic Programming

Stochastic or probabilistic programming
deals with situations where some or all of
the parameters of the optimization
problem are described by stochastic (or
random or probabilistic) variables rather
than by deterministic quantities. The
si , i = 1, 2, . . . , n + 1.
385
sources of random variables may be

several, depending on the nature and the
type of problem. For instance, in the design
of concrete structures, the strength of
concrete is a random variable since the
compressive strength of concrete varies
considerably from sample to sample.
1970s several evolutionary methodologies

have been proposed, mainly genetic
algorithms, evolutionary programming,
and evolution strategies (Hammel, and
Schwefel 1997). All of these approaches
operate on a set of candidate solutions.
Using strong simplifications, this set is
subsequently modified by the two basic
principles of evolution: selection and
variation. Selection represents the
competition for resources among living
beings. Some are better than others and
more likely to survive and to reproduce
their genetic information. In evolutionary
algorithms, natural selection is simulated
by a stochastic selection process. Each
solution is given a chance to reproduce a
certain number of times, dependent on
their quality. Thereby, quality is assessed
by evaluating the individuals and assigning
them scalar fitness values. The other
principle, variation, imitates natural
capability of creating new living beings
by means of recombination and mutation.
Although the underlying principles are
simple, these algorithms have proven
themselves as a general, robust and
powerful search mechanism.
In the design of mechanical systems, the

actual dimension of any machined part is a
random variable since the dimension may
lie anywhere within a specied
(permissible) tolerance band.
Similarly, in the design of aircraft and
rockets the actual loads acting on the
vehicle depend on the atmospheric
conditions prevailing at the time of the
ight, which cannot be predicted precisely
in advance. Hence the loads are to be
treated as random variables in the design
of such ight vehicles.
Depending on the nature of equations
involved (in terms of random variables) in
the problem, a stochastic optimization
problem is called a stochastic linear,
geometric,
dynamic,
or
nonlinear
programming problem. The basic idea
used in stochastic programming is to
convert the stochastic problem into an
equivalent deterministic problem.
Basic Principles of Evolutionary Algorithms

In general, an EA is characterized by three
facts:
7.4 Modern Methods for

Solving Optimization
Problems
1. a set of solution candidates is

maintained
2. these solutions undergoes a selection
process
7.4.1 Introduction to Evolutionary

Algorithms
3. they are manipulated by genetic

operators, usually recombination and
mutation.
The term evolutionary algorithm (EA)

stands for a class of stochastic optimization
methods that simulate the process of
natural evolution. The origins of EAs can be
traced back to the late 1950s, and since the
By analogy to natural evolution, the

solution candidates are called individuals
and the set of solution candidates is called
386
the population. Each individual represents

a possible solution.
7.4.2 Simulated Annealing (SA)
A decision vector, to the problem at hand

where, however, an individual is not a
decision vector but rather encodes it based
on an appropriate structure. Without loss
of generality, this structure is assumed to
be a vector here, a bit vector or a realvalued vector, although other structures
like trees might be used as well; the set of
all possible vectors constitutes the
individual space II. In this terminology, the
population is a set of vectors i I , to be
more precise a multi-set of vectors since it
can contain several identical individuals. In
the selection process, which can be either
stochastic or completely deterministic,
low-quality individuals are removed from
the population, while high quality
individuals are reproduced.
Simulated annealing is a local search

algorithm (meta-heuristic) capable of
escaping from local optima. Its ease of
implementation, convergence properties
and its use of hill-climbing moves to escape
local optima have made it a popular
technique over the past two decades. It is
typically used to address discrete, and to a
lesser extent, continuous optimization
problems. Recent survey articles that
provide a good overview of simulated
annealings theoretical development and
domains of application include Eglese
(1990), Fleischer (1995), Koulamas et al.
(1994), and Romeo and SangiovanniVincentelli (1991). Aarts and Korst (1989)
and van Laarhoven and Aarts (1988)
devote entire books to the subject. search
algorithms for discrete optimization
problems.
The goal is to focus the search on particular

portions of the search space and to
increase the average quality within the
population. The quality of an individual
with respect to the optimization task is
represented by a scalar value, the so-called
fitness.
in 1953 [Metropolis, 1953), The algorithm

in this paper simulated the cooling of
material in a heat bath. This is a process
known as annealing.
Evolutionary Algorithms
Simulated
Anneling
Ant Colony
Neural
Network
Fuzzy
Genetic
Algorithm
387
in 1953 [Metropolis, 1953), The algorithm

in this paper simulated the cooling of
material in a heat bath. This is a process
known as annealing.
Using these mappings any combinatorial

optimisation problem can be converted
into an annealing algorithm [(Kirkpatrick,
1983), (Crny, 1985)] by sampling the
neighbourhood randomly and accepting
worse solutions with a probability.
If you heat a solid past melting point and

then cool it, the structural properties of
the solid depend on the rate of cooling. If
the liquid is cooled slowly enough, large
crystals will be formed. However, if the
liquid is cooled quickly (quenched) the
crystals will contain imperfections.
The law of thermodynamics state that at

temperature, t, the probability of an
increase in energy of magnitude, E, is
given by
P (E) = exp (-E /kt)
Metropoliss algorithm simulated the

material as a system of particles. The
algorithm simulates the cooling process by
gradually lowering the temperature of the
system until it converges to a steady,
frozen state.
(1)
Where k is a constant known as

Boltzmanns constant.
The simulation in the Metropolis algorithm
calculates the new energy of the system. If
the energy has decreased then the system
moves to this state. If the energy has
increased then the new state is accepted
using the probability returned by the
above formula.
In 1982, Kirkpatrick et al (Kirkpatrick, 1983)

took the idea of the Metropolis algorithm
and applied it to optimisation problems.
The idea is to use simulated annealing to
search for feasible solutions and converge
to an optimal solution.
A certain number of iterations are carried

out at each temperature and then the
temperature is decreased. This is repeated
until the system freezes into a steady state.
7.4.2.2 Idea of SA
In (Dowsland, 1995) table 6.1 is presented
which shows how physical annealing can
be mapped to simulated annealing. It is
repeated here
System States
Feasible Solutions
This equation is directly used in simulated

annealing, although it is usual to drop the
Boltzmann constant as this was only
introduced into the equation to cope with
different materials. Therefore, the
probability of accepting a worse state is
given by the equation
Energy
Cost
P = exp (-c/t) > r
Change of State
Neighbouring
Solutions
Where
Temperature
Control Parameter
t = the current temperature
Frozen State
Heuristic Solution
r = a random number between 0 and 1
Thermodynamic Combinatorial
Simulation
Optimisation
(2)
c = the change in the evaluation function
Table 6. 1
The probability of accepting a worse move

is a function of both the temperature of
388
the system and of the change in the cost

function. This was shown in the lectures
and a spreadsheet is available from the
web site for this course which shows the

same example that was presented in the
lectures.
Initialize the
solution(T=Th)
While T> Coolest

temp
For N (number of
iteration)
Calculate
(its the difference of
energy between
current and new state)
Accept new state

as current state
YES
If
<0
NO
Accept new state
with probability
Pr(accepted)=exp(E/KB.T
Decrease the
temperature
End
389
It can be appreciated that as the

temperature of the system decreases the
probability of accepting a worse move is
decreased. This is the same as gradually
moving to a frozen state in physical
annealing. Also note, that if the
temperature is zero then only better
moves will be accepted which effectively
makes simulated annealing act like hill
climbing, in each step of this algorithm, an
atom is given a small random displacement
and the resulting change, AE, in the energy
of the system is computed.
of parameters {x,}, it is straightforward

with the Metropolis procedure to generate
a population of configurations of a given
optimization problem at some effective
temperature.
This temperature is simply a control
parameter in the same units as the cost
function. The simulated annealing process
consists of first "melting" the system being
optimized at a high effective temperature,
then lowering the temperature by slow
stages until the system "freezes" and no
further changes occur. At each
temperature, the simulation must proceed
long enough for the system to reach a
steady
state.
The
sequence
of
temperatures and the number of
rearrangements of the {x,) attempted to
reach equilibrium at each temperature can
be considered an annealing schedule.
If E 0, the displacement is accepted,

and the configuration with the displaced
atom is used as the starting point of the
next step.
The case E > 0 is treated probabilistically:
the probability that the configuration is
accepted is P (E) = exp (- ).
Starting temperature
Random numbers uniformly distributed in

the interval (0, l) are a convenient means
of implementing the random part of the
algorithm.
The starting temperature must be hot

enough to allow a move to almost any
neighbourhood state. If this is not done,
then the ending solution will be the same
(or very close) to the starting solution.
Alternatively, we will simply implement a
hill climbing algorithm.
One such number is selected and

compared with P ().
If it is less than P (), the new
configuration is retained;
However, if the temperature starts at too

high a value then the search can move to
any neighbour and thus transform the
search (at least in the early stages) into a
random search. Effectively, the search will
be random until the temperature is cool
enough to start acting as a simulated
annealing algorithm.
If not, the original configuration is used to

start the next step.
By repeating the basic step many times,
one simulates the thermal motion of
atoms in thermal contact with a heat bath
at temperature T.
This choice of P (E) has the consequence
that the system evolves into a Boltzmann
distribution.
The problem is finding the correct starting

temperature. At present, there is no
known method for finding a suitable
starting temperature for a whole range of
Using the cost function in place of the

energy and defining configurations by a set
390
problems. Therefore, we need to consider

other ways.
Temperature Decrement
Once we have our starting and stopping
temperature we need to get from one to
the other. That is, we need to decrement
our temperature so that we eventually
arrive at the stopping criterion.
If we know the maximum distance (cost

function difference) between one
neighbour and another then we can use
this information to calculate a starting
temperature.
The way in which we decrement our

temperature is critical to the success of the
algorithm. Theory states that we should
allow enough iterations at each
temperature so that the system stabilises
at that temperature. Unfortunately, theory
also states that the number of iterations at
each temperature to achieve this might be
exponential to the problem size. As this is
impractical we need to compromise. We
can either do this by doing a large number
of iterations at a few temperatures, a small
number
of
iterations
at
many
temperatures or a balance between the
two.
Another method, suggested in (RaywardSmith, 1996), is to start with a very high

temperature and cool it rapidly until about
60% of worst solutions are being accepted.
This forms the real starting temperature
and it can now be cooled more slowly.
A similar idea, suggested in (Dowsland,
1995), is to rapidly heat the system until a
certain proportion of worse solutions are
accepted and then slow cooling can start.
This can be seen to be similar to how
physical annealing works in that the
material is heated until it is liquid and then
cooling begins (i.e. once the material is a
liquid it is pointless carrying on heating it).
One way to decrement the temperature is

a simple linear method.
Final temperature
An alternative is a geometric decrement

where
It is usual to let the temperature decrease

until it reaches zero. However, this can
make the algorithm run for a lot longer,
especially when a geometric cooling
schedule is being used.
t = t
Where < 1
Experience has shown that should be
between 0.8 and 0.99, with better results
being found in the higher end of the range.
Of course, the higher the value of , the
longer it will take to decrement the
temperature to the stopping criterion.
In practise, it is not necessary to let the

temperature reach zero because as it
approaches zero the chances of accepting
a worse move are almost the same as the
temperature being equal to zero.
Therefore, the stopping criteria can either
be a suitably low temperature or when the
system is frozen at the current
temperature (i.e. no better or worse
moves are being accepted).
Iterations at each Temperature

The final decision we have to make is how
many iterations we make at each
temperature.
A constant number of iterations at each
temperature is an obvious scheme.
391
Another method, first suggested by

(Lundy, 1986) is to only do one iteration at
each temperature, but to decrease the
temperature very slowly. The formula they
use is:
If possible, the cost function should also be

designed so that it can lead the search. One
way of achieving this is to avoid cost
functions where many states return the
same value. This can be seen as
representing a plateau in the search space
which the search has no knowledge about
which way it should proceed.
t = t / (1 + t)
Where is a suitably small value. I have
entered this formula on a spreadsheet
(available from the web site) so that you
can play around with the parameters, if
you are interested.
In the lecture an example was given that

described two possible cost functions for a
bin packing problem.
Many cost functions cater for the fact that
some solutions are illegal. This is typically
achieved using constraints. Two types of
constraints are often used.
An alternative is to dynamically change the

number of iterations as the algorithm
progresses. At lower temperatures it is
important that a large number of iterations
are done so that the local optimum can be
fully explored. At higher temperatures, the
number of iterations can be less.
Hard Constraints
:
these
constraints cannot be violated in a feasible
solution. For example, in designing the
Jubilee Campus the space allocation cost
function could define a hard constraint as
not allowing a professor to occupy an
office that is smaller than a certain size.
Cost Function
Presented with a solution to a problem,
there must be some way of measuring the
quality of the solution.
Soft Constraints
:
these
constraints should, ideally, not be violated
but, if they are, the solution is still feasible.
In defining this cost function we obviously

need to ensure that it represents the
problem we are trying to solve.
When defining constraints they are usually

weighted. Hard constraints maybe given a
large weighting so that those solutions
which violate those constraints have a high
cost function. Soft constraints are
weighted depending on their importance.
It is also important that the cost function

can be calculated as efficiently as possible,
as it will be calculated at every iteration of
the algorithm. However, the cost function
is often a bottleneck and it may sometimes
be necessary to use
Sometimes the weightings are dynamically

changed as the algorithm progresses. This
allows, for example, hard constraints to be
accepted more readily at the start of the
algorithm but later on these solutions
would be rejected.
Delta Evaluation
:
the
difference between the current solution
and the neighbourhood solution is
evaluated.
Partial Evaluation
:
a simplified
evaluation function is used that does not
give an exact result but gives a good
indication as to the quality of the solution.
Cooling
If you plot a typical cooling schedule you
are likely to find that at high temperatures
392
many solutions are accepted. If you start at

too high a temperature a random search is
emulated and until the temperature cools
sufficiently any solution can be reached
and could have been used as a starting
position.
The essential trait of ACO algorithms is the

combination of a priori information about
the structure of a promising solution with
a posteriori information about the
structure of previously obtained good
solutions.
At lower temperatures, the plot of the

cooling schedule, is likely to show that very
few worse moves are accepted; almost
making simulated annealing emulate hill
climbing.
Metaheuristic algorithms are algorithms

which, in order to escape from local
optima, drive some basic heuristic: either a
constructive heuristic starting from a null
solution and adding elements to build a
good complete one, or a local search
heuristic starting from a complete solution
and iteratively modifying some of its
elements in order to achieve a better one.
None of this should come as a surprise as

this is precisely what we would expect
from simulated annealing.
Taking this one stage further, we can say
that simulated annealing does most of its
work during the middle stages of the
cooling schedule. (Connolly, 1990) took
this argument one stage further and
suggested annealing at a constant
temperature. The problem is (again) what
the best temperature is. It must be high
enough to allow movement but not so low
that the system is frozen. But the problem
goes even further than that. The optimum
temperature will vary from one type of
problem to another and also from one
instance of a problem to another instance
of the same problem.
The metaheuristic part permits the

lowlevel heuristic to obtain solutions
better than those it could have achieved
alone, even if iterated. Usually, the
controlling mechanism is achieved either
by constraining or by randomizing the set
of local neighbor solutions to consider in
local search (as is the case of simulated
annealing or tabu search), or by combining
elements taken by different solutions (as is
the case of evolution strategies and
genetic or bionomic algorithms).
Ant Colony Optimization incrementally

generates solutions paths in the space of
such
components,
adding
new
components to a state.
7.4.3 Ant Colony Optimization (ACO)

Ant Colony Optimization (ACO) is a
paradigm for designing metaheuristic
algorithms for combinatorial optimization
problems. The first algorithm which can be
classified within this framework was
presented in 1991 and, since then, many
diverse variants of the basic principle have
been reported in the literature.
Memory is kept of all the observed

transitions between pairs of solution
components and a degree of desirability is
associated to each transition depending on
the quality of the solutions in which it
occurred so far.
While a new solution is generated, a
component y is included in a state, with a
probability that is proportional to the
393
desirability of the transition between the

last component included in the state, and
y itself .The main idea is to use the selforganizing principles to coordinate
populations of artificial agents that
collaborate to solve computational
problems. Self-organization is a set of
dynamical
mechanisms
whereby
structures appear at the global level of a
system from interactions among its lowerlevel components. The rules specifying the
interactions
among
the
systems
constituent units are executed on the basis
of purely local information, without
reference to the global pattern, which is an
emergent property of the system rather
than a property imposed upon the system
by an external ordering influence. For
example, the emerging structures in the
case of foraging in ants include
spatiotemporally organized networks of
pheromone trails. The aim of this work is
to enhance the ability of ACO by using GP
technique.
globally cooperating to find a good
7.4.3.2 Similarities and differences with
the ants current history/performance and
real ants
can be read/written by any ant accessing

the state.
solution to the task under consideration.

Although the complexity of each artificial
ant is such that it can build a feasible
solution (as a real ant can find somehow a
path between the nest and the food), high
quality solutions are the result of the
cooperation among the individuals of the
whole colony. Ants cooperate by means of
the information they concurrently
read/write on the problems states they
visit, as explained in the next item.
Pheromone trail
Artificial ants modify some aspects of their
environment as the real ants do. While real
ants deposit on the worlds state they visit
a chemical substance, the pheromone,
artificial ants change some numeric
information locally stored in the problems
state they visit. This information takes into
account
Most of the ideas of ACO stem from real

ants. In particular, the use of:
By analogy, we call this numeric

information artificial pheromone trail,
pheromone trail for short in the following.
In ACO algorithms local pheromone trails
are the only communication channels
among the ants.
(i) a colony of cooperating individuals,

(ii) an (artificial) pheromone trail for local
stigmergetic communication,
(iii) a sequence of local moves to find
shortest paths,
This stigmergetic form of communication

plays a major role in the utilization of
collective knowledge. Its main effect is to
change the way the environment (the
problem landscape) is locally perceived by
the ants as a function of all the past history
of the whole ant colony.
(iv) a stochastic decision policy using local

information and no look ahead.
Colony of cooperating individuals.
As real ant colonies, ant algorithms are
composed of a population, or colony, of
concurrent and asynchronous entities
394
Usually, in ACO algorithms an evaporation

mechanism, similar to real pheromone
evaporation,
modifies
pheromone
information over time. Pheromone
evaporation allows the ant colony to slowly
forget its past history so that it can direct
its search towards new directions without
being over-constrained by past decisions.
builds a solution, or a component of it6,

starting from an initial state selected
according to some problem dependent
criteria. While building its own solution,
each ant collects information on the
problem characteristics and on its own
performance, and uses this information to
modify the representation of the problem,
as seen by the other ants. Ants can act
concurrently and independently, showing
a cooperative behavior. They do not use
direct communication: it is the stigmergy
paradigm that governs the information
exchange among the ants.
Shortest path searching and local moves.

Artificial and real ants share a common
task: to find a shortest (minimum cost)
path joining an origin (nest) to destination
(food) sites. Real ants do not jump, they
just walk through adjacent terrains states,
An incremental constructive approach is

used by the ants to search for a feasible
solution.
and so do artificial ants, moving step-bystep through adjacent states of the

problem. Of course, exact definitions of
state and adjacency are problem-specific.
A solution is expressed as a minimum cost

(shortest) path through the states of the
Stochastic and myopic state transition

policy.
problem in accordance with the problems

constraints. The complexity of each ant is
such that even a single ant is able to find a
(probably poor quality) solution. High
quality solutions are only found as the
emergent result of the global cooperation
among all the agents of the colony
concurrently building different solutions.
Artificial ants, as real ones, build solutions

applying a probabilistic decision policy to
move through adjacent states. As for real
ants, the artificial ants policy makes use of
local information only and does not make
use of lookahead to predict future states.
Therefore, the applied policy is completely
local, in space and time. The policy is a
function of both the a priori information
represented by the problem specifications
According to the assigned notion of

neighborhood (problem-dependent), each
ant builds a solution by moving through a
(finite) sequence of neighbor states.
Moves are selected by applying a
stochastic local search policy directed (i) by
ant private information (the ant internal
state, or memory) and (ii) by publicly
available pheromone trail and a priori
problem-specific local information.
(equivalent to the terrains structure for

real ants), and of the local modifications in
the environment (pheromone trails)
induced by past ants.
7.4.3.3The ACO meta-heuristic
In ACO algorithms a finite size colony of
artificial ants with the above described
characteristics collectively searches for
good quality solutions to the optimization
problem under consideration. Each ant
The ants internal state stores information

about the ant past history. It can be used
to carry useful information to compute the
value/goodness of the generated solution
395
and/or the contribution of each executed

move. Moreover it can play a fundamental
role to manage the feasibility of the
solutions. In some problems in fact,
typically in combinatorial optimization,
some of the moves available to an ant in a
state can take the ant to an infeasible
state. This can be avoided exploiting the
strength to capture the uncertainties

associated
with
human
cognitive
processes, such as thinking and reasoning.
The
conventional
approaches
to
knowledge representation lack the means
for representing the meaning of fuzzy
concepts. As a consequence, the
approaches based on first order logic do
not provide an appropriate conceptual
framework for dealing with the
representation
of
commonsense
knowledge, since such knowledge is by its
nature both lexically imprecise and noncategorical. The developement of fuzzy
logic was motivated in large measure by
the need for a conceptual framework
which can address the issue of lexical
imprecision. Some of the essential
characteristics of fuzzy logic relate to the
following
ants memory. Ants therefore can build

feasible solutions using only knowledge
about the local state and about the effects
of actions that can be performed in the
local state.
7.4.4 Fuzzy Optimization

Many decision-making task are too
complex to be understood quantitatively,
however, humans succeed by using
knowledge that is imprecise rather than
precise. Fuzzy logic resembles human
reasoning in its use of imprecise
information to generate decisions. Unlike
classical logic which requires a deep
understanding of a system, exact
equations, and precise numeric values,
fuzzy logic incorporates an alternative way
of thinking, which allows modeling
complex systems using a higher level of
abstraction
originating
from
our
knowledge and experience. Fuzzy logic
allows expressing this knowledge with
subjective concepts such as very big and a
long time which are mapped into exact
numeric ranges. Since knowledge can be
expressed in a more natural by using fuzzy
sets, many decision (and engineering)
problems can be greatly simplified. Fuzzy
logic provides an inference morphology
that enables approximate human
reasoning capabilities to be applied to
knowledge-based systems. The theory of
fuzzy logic provides a mathematical
In fuzzy logic, exact reasoning is viewed

as a limiting case of approximate
reasoning.
In fuzzy logic, everything is a matter of
degree.
In fuzzy logic, knowledge is interpreted a
collection of elastic or, equivalently, fuzzy
constraint on a collection of variables.
Inference is viewed as a process of
propagation of elastic constraints.
Any logical system can be fuzzified. There
are two main characteristics of fuzzy
systems
that
give
them
better
performance for specific applications.
Fuzzy systems are suitable for uncertain or
approximate reasoning, especially for
systems with mathematical models that
are difficult to derive. Fuzzy logic allows
decision making with estimated values
under
incomplete
or
uncertain
information.
396
However, fuzzy set theory[3,4] which was

developed by Zadeh in 1960s and fuzzy
optimization techniques provide a useful
and efficient tool for modelling and
optimizing such systems. Modelling and
optimization under a fuzzy environment is
called fuzzy modelling and fuzzy
optimization.
high or low, fuzzy membership functions

can be used to quantify them.
Both probability distributions and fuzzy
membership functions of these uncertain
or qualitative variables can be included in
quantitative optimization models. This
chapter introduces fuzzy optimization
modelling, again for the preliminary
screening of alternative water resources
plans and management policies. 5 on the
value of x. The set A is referred to as a crisp
set.
Large, small, pure, polluted, satisfactory,

unsatisfactory, sufficient, insufficient,
excellent, good, fair, poor and so on are
words often used to describe various
attributes or performance measures of
water
resources
systems.
These
descriptors do not have crisp, welldefined boundaries that separate them
from others. A particular mix of economic
and environmental impacts may be more
acceptable to some and less acceptable to
others. Plan A is better than Plan B. The
water quality and temperature is good for
swimming. These qualitative, or fuzzy,
statements convey information despite
the imprecision of the italicized adjectives.
If one is not able to say for certain whether

or not any number x is in the set, then the
set A could be referred to as fuzzy. The
degree of truth attached to that statement
is defined by a membership function. This
function ranges from 0 (completely false)
to 1 (completely true). Consider the
statement, The water temperature should
be suitable for swimming. Just what
temperatures are suitable will depend on
the person asked.
It would be difficult for anyone to define
precisely those temperatures that are
suitable if it is understood that
temperatures outside that range are
absolutely not suitable. A membership
function defining the interval or range of
water temperatures suitable for swimming
is shown in Figure 5.1. Such functions may
be defined on the basis of the responses of
many potential swimmers.
7.4.4.2 Fuzzy Membership Functions

Consider a set A of real or integer numbers
ranging from say 18 to 25. Thus A[18, 25].
In classical (crisp) set theory, any number x
is either in or not in the set A. The
statement x belongs to A is either true or
false depending 135 Fuzzy Optimization
The precise quantification of many system
performance criteria and parameter and
decision variables is not always possible,
nor is it always necessary. When the values
of variables cannot be precisely specified,
they are said to be uncertain or fuzzy.
There is a zone of imprecision or

disagreement at both ends of the range.
The form or shape of a membership
function depends on the individual
subjective feelings of the members or
individuals who are asked their opinions.
If the values are uncertain, probability

distributions may be used to quantify
them. Alternatively, if they are best
described by qualitative adjectives, such as
dry or wet, hot or cold, clean or dirty, and
To define this particular membership

function, each individual i could be asked
397
to define his or her comfortable water

temperature interval (T1i , T2i ). The
membership value associated with any
temperature value T equals the number of
individuals who place that T within their
range (T1i , T2i ), divided by the number of
individual opinions obtained. The
assignment of membership values is based
on subjective judgments , but such
judgments seem to be sufficient for much
of human communication.
both too cold and too hot. Of course it

could be an empty set, as indeed it is in this
case, based on the two membership
functions shown in Figure 7.15. The
minimum of either function for any value
of x is 0. The complement of the
membership function for fuzzy set A is the
membership function, mA c (x), of Ac .
mA c (x) =1 mA(x)
The complement of set A (defined in Figure
7.15) would represent the range of
temperatures considered not too cold for
swimming. The complement of set B (also
defined in Figure 7.15) would represent
the range of temperatures considered not
too hot for swimming. The complement of
the union set as shown in Figure 7.16
would be the range of temperatures
considered just right. This complement set
is the same as shown in Figure 7.14.
7.4.4.3Membership Function Operations

Denote the membership function
associated with a fuzzy set A as mA(x). It
defines the degree or extent to which any
value of x belongs to the set A. Now
consider two fuzzy sets, A and B. Set A
could be the range of temperatures that
are considered too cold, and set B could be
the range of temperatures that are
considered too hot. Assume these two sets
are as shown in Figure 7.15.
The degree or extent that a value of x
belongs to either of two sets A or B is the
maximum of the two individual
membership function values. This union
membership function is defined as:
Figure 7.14. A fuzzy membership function

for suitability of water temperature for
mAB(x) = maximum(mA(x), mB(x))

This union set would represent the ranges
of temperatures that are either too cold or
too hot, as illustrated in Figure 7.16. The
degree or extent that a value of a variable
x is simultaneously in both sets A and B is
the minimum of the two individual
membership function values. This
intersection membership function is
defined as:
swimming.
Figure 7.15. Two membership functions relating to
swimming water temperature. Set A is the set defining
the fraction of all individuals who think the water
temperature is too cold, and Set B defines the fraction of
all individuals who think the water temperature is too
hot.
mAB(x) = minimum (mA(x), mB(x))

This intersection set would define the
range of temperatures that are considered
398
neurons, some having increased layers of

input neurons and output neurons. The
synapses store parameters called
"weights" that manipulate the data in the
calculations.
An ANN is typically defined by three types
of parameters:
Figure 7.16. Membership function for water temperatures that

are considered too cold or too hot.
7.4.5 Neural Optimization
1. Architecture (The interconnection

pattern between the different layers of
neurons)
In machine learning, artificial neural
networks (ANNs) are a family of models
inspired by biological neural networks and
are used to approximate functions that can
depend on a large number of inputs and
are generally unknown. ANNs are
presented as systems of interconnected
"neurons" which exchange messages
between each other. The connections have
weights that can be tuned, making neural
nets adaptive to inputs and capable of
learning. The ability of the ANNs to
accurately
approximate
unknown
functions , underlies this growing
popularity. Neural networks have been
used to solve a wide variety of tasks that
are hard to solve using ordinary rule-based
programming, including computer vision
and speech recognition, Character
Recognition, Image Compression, Stock
Market Prediction, Traveling Saleman's
Problem, and Medicine, Electronic Nose,
Security, and Loan Applications . The word
network in the term 'artificial neural
network' refers to the interconnections
between the neurons in the different
layers of each system. An example system
has three layers (see figure 7.17). The first
layer has input neurons which send data
via synapses to the second layer of neurons
(hidden layer), and then via more synapses
to the third layer of output neurons. More
complex systems will have more layers of
2. The learning process for updating the

weights of the interconnections
3. The activation function that converts a
neuron's weighted input to its output
activation.
Mathematically, a neuron's network
function F(x) is defined as a composition of
other functions gi(x) , which can further be
defined as a composition of other
functions. This can be conveniently
represented as a network structure, with
arrows depicting the dependencies
between variables. A widely used type of
composition is the nonlinear weighted
sum, where
where
K
(commonly referred to as the activation
function) is some predefined function,
such as the hyperbolic tangent. It will be
convenient for the following to refer to a
collection of functions gi as simply a vector
g=(g1 , g2 , g3..gn ).
Figure 7.17: ANN with one hidden layer
399
problems. Authors in [4] compared BP, SA,

and GA for ANN. They concluded that
although the most frequently used
algorithm for optimizing neural networks is
back propagation, it is likely to obtain local
solutions. They showed that simulated
annealing, a global search algorithm,
performs better than back propagation,
but is also uses a point to point search.
Both BP and SA have multiple user
determined parameters which may
significantly impact the solution. Since
there are no established rules for selecting
these parameters, solution outcome is
based on chance. The genetic algorithm
appears to be able to systematically obtain
superior solutions to simulated annealing
for optimizing neural networks. These
solutions provide the researcher with
superior estimates of interpolation data.
The genetic algorithms process of moving
from one population of points to another
enables it to discard potential local
solutions and also to achieve the superior
solutions in a computationally more
efficient manner.
7.4.5.2 More Understanding of ANN

Generally when we talk about ANN
optimization, the objective function is
mean square error function (loss/cost
function). We have to find optimize values
weights of neural network to minimize the
objective function. Although, gradient
based search techniques such as back
propagation are currently the most widely
used optimization techniques for training
neural networks, it has been shown that
these gradient techniques are severely
limited in their ability to find global
solutions. Global search techniques have
been identified as a potential solution to
this problem. Other techniques can be
used. Two well-known global search
techniques, Simulated Annealing and the
Genetic Algorithm can be used as well
Because of its ease of use, an
overwhelming
majority
of
these
applications have used some variation of
the gradient technique, back propagation
(BP) for optimizing the networks.
Although,
back
propagation
has
unquestionably been a major factor for the
success of past neural network
applications, it is plagued with inconsistent
and
unpredictable
performances.
Recently, demonstrated that for a variety
of complex functions the genetic algorithm
was able to achieve superior solutions for
neural network optimization than back
propagation and found that another global
search heuristic, Tabu Search (TS), is able
to systematically achieve superior
solutions for optimizing the neural
network than those achieved by back
propagation. In addition to GA and TS
another well-known global search heuristic
is simulated annealing. Simulated
annealing has been shown to perform well
for optimizing a wide variety of complex
7.4.5.2.1 Back-propagation
Back-propagation, an abbreviation for
"backward propagation of errors", is a
common method of training artificial
neural networks used in conjunction with
an optimization method such as gradient
descent. The method calculates the
gradient of a loss function with respect to
all the weights in the network. The
gradient is fed to the optimization method
which in turn uses it to update the weights,
in an attempt to minimize the loss
function. Back-propagation requires a
known, desired output for each input value
in order to calculate the loss function
gradient.
400
It is therefore usually considered to be a

supervised learning method, although it is
also used in some unsupervised networks
such asautoencoders. It is a generalization
of the delta rule to multi-layered feedforward networks, made possible by using
the chain rule to iteratively compute
gradients for each layer. Back-propagation
requires that the activation function used
by the artificial neurons (or "nodes") be
differentiable. Gradient search techniques
such as back-propagation are designed for
local search. They typically achieve the
best solution in the region of their starting
point. Obtaining a global solution is often
dependent on a fortuitous choice of
starting values. Global search techniques
are known to more consistently obtain the
optimal solution.
points generating the lowest sum of

squared errors are the most likely to be
represented in the new population. The
points comprising this new population are
then randomly paired for the crossover
operation. Each point is a vector (string) of
n parameters (weights). A position along
the vectors is randomly selected for each
pair of points and the preceding
parameters are switched between the two
points. This crossover operation results in
each new point having parameters from
both parent points. Finally, each weight
has a small probability of being replaced
with a value randomly chosen from the
parameter space.
This operation is referred to as mutation.
Mutation
enhances
the
GA
by
intermittently injecting a random point in
order to better search the entire
parameter space. This allows the GA to
possibly escape from local optima if the
new point generated is a better solution
than has previously been found, thus
providing a more robust solution. This
resulting set of points now becomes the
new population, and the process repeats
until convergence. Since this method
simultaneously
searches
in
many
directions, the probability of finding a
global optimum greatly increases. The
algorithms similarity to natural selection
inspires its name.
7.4.5.2.2 Genetic Algorithm

For difficult non-linear functions GA has
been shown to perform exceedingly well in
obtaining global solutions Basically, an
objective function, such as minimization of
the sum of squared errors or sum of
absolute errors, is chosen for optimizing
the neural network. Using the chosen
objective function, each candidate point
out of the initial population of randomly
chosen starting points is used to evaluate
the objective function. These values are
then used in assigning probabilities for
each of the points in the population. For
minimization, as in the case of sum of
squared errors, the highest probability is
assigned to the point with the lowest
objective function value. Once all points
have been assigned a probability, a new
population of points is drawn from the
original population with replacement. The
points are chosen randomly with the
probability of selection equal to its
assigned probability value. Thus, those
As the GA progresses through

generations, the parameters most
favorable for optimizing the objective
function will reproduce and thrive in future
generations, while poorly performing
parameters die out, as in survival of the
fittest.
401
Inside a factory, you could use a neural

network for quality control. Let's say
you're producing clothes washing
detergent in some giant, convoluted
chemical process. You could measure the
final detergent in various ways (its color,
acidity, thickness, or whatever), feed those
measurements into your neural network as
inputs, and then have the network decide
whether to accept or reject the batch.
7.4.5.2.3 Tabu Search

Tabu search, proposed by Glover ,is a
meta-heuristic method that has received
widespread attention recently because of
its flexible framework and several
significant successes in solving NP-hard
problems The method of neighborhood
exploration and the use of short-term and
long-term memories distinguish Tabu
Search from local search and other
heuristic search methods, and result in
lower computational cost and better space
exploration. TS involves a lot of techniques
and strategies, such as short-term
memories (tabu list), long-term memories
and other prior information about the
solution can used to improve the
intensification and diversification of the
search. It can be confirmed that the
strategy
of
intensification
and
diversification is very important at most
time, therefore, a novel adaptive search
strategy
of
intensification
and
diversification, proposed in literature by
us, was employed to improve the
efficiency of TS for neural network
optimization . proposed an adaptive TS
approach for ANNs
In security
There are lots of applications for neural
networks in security too. Suppose you're
running a bank with many thousands of
credit-card transactions passing through
your computer system every single
minute. You need a quick automated way
of identifying any transactions that might
be fraudulentand that's something for
which a neural network is perfectly suited.
Your inputs would be things like
1. Is the cardholder actually present?
2. Has a valid PIN number been used?
3. Have five or more transactions
been presented with this card in
the last 10 minutes?
7.4.5.3 Applications of ANNs
4. Is the card being used in a different

country from which it's registered?
On the basis of this example, you can

probably see lots of different applications
for neural networks that involve
recognizing patterns and making simple
decisions about them.
And so on. With enough clues, a neural

network can flag up any transactions that
look suspicious, allowing a human
operator to investigate them more closely.
In airplanes
7.4.6 Genetic Algorithm

Optimization
You might use a neural network as a basic

autopilot, with input units reading signals
from the various cockpit instruments and
output units modifying the plane's controls
appropriately to keep it safely on course.
Genetic Algorithms (GA) is a search
heuristic that
mimics
the
process
of natural selection. This heuristic is
In industry
402
routinely used to generate useful solutions

to optimization and search problems.
terminates when either a maximum

number of generations has been
produced, or a satisfactory fitness level has
been reached for the population.
Initially developed by John Holland figure

6.20, University of Michigan (1970s).
A typical genetic algorithm requires:
Hollands original GA is known as the

simple genetic algorithm (SGA).
1)a genetic representation of the solution

domain,
Provide efficient techniques for

optimization and machine learning
applications.
2)a fitness function to evaluate the

solution domain.
A standard representation of each
candidate solution is as an array of bits
Widely used in business, science

and engineering.
Arrays of other types and structures can be

used in essentially the same way. The main
property that makes these genetic
representations convenient is that their
parts are easily aligned due to their fixed
size, which facilitates simple crossover
operations.
Optimization problems
In a genetic algorithm, a population of
candidate solutions (called individuals,
creatures, or phenotypes) to an
optimization problem is evolved toward
better solutions. Each candidate solution
has a set of properties (its chromosomes or
genotype) which can be mutated and
altered; traditionally, solutions are
represented in binary as strings of 0s and
1s, but other encodings are also possible.
Variable length representations may also

be used, but crossover implementation is
more complex in this case. Tree-like
representations are explored in genetic
programming
and
graph-form
representations
are
explored
in
evolutionary programming; a mix of both
linear chromosomes and trees is explored
in gene expression programming.
The evolution usually starts from a

population of randomly generated
individuals, and is an iterative process,
with the population in each iteration called
a generation. In each generation, the
fitness of every individual in the population
is evaluated; the fitness is usually the value
of the objective function in the
optimization problem being solved.
Once the genetic representation and the

fitness function are defined, a GA proceeds
to initialize a population of solutions and
then to improve it through repetitive
application of the mutation, crossover,
inversion and selection operators.
The more fit individuals are stochastically

selected from the current population, and
each individual's genome is modified
(recombined and possibly randomly
mutated) to form a new generation.
Initialization
The population size depends on the nature
of the problem, but typically contains
several hundreds or thousands of possible
solutions. Often, the initial population is
generated randomly, allowing the entire
The new generation of candidate solutions

is then used in the next iteration of the
algorithm. Commonly, the algorithm
403
range of possible solutions (the search

space). Occasionally, the solutions may be
"seeded" in areas where optimal solutions
are likely to be found.
dynamics is used to determine the air

resistance of a vehicle whose shape is
encoded as the phenotype), or even
interactive genetic algorithms are used.
Selection
Genetic operators
During each successive generation, a

proportion of the existing population is
selected to breed a new generation.
Individual solutions are selected through a
fitness-based process, where fitter
solutions (as measured by a fitness
function) are typically more likely to be
selected. Certain selection methods rate
the fitness of each solution and
preferentially select the best solutions.
Other methods rate only a random sample
of the population, as the former process
may be very time-consuming.
Main
articles:
algorithm) and
algorithm)
Crossover
Mutation
(genetic
(genetic
The next step is to generate a second

generation population of solutions from
those selected through a combination of
genetic operators: crossover (also called
recombination), and mutation.
For each new solution to be produced, a
pair of "parent" solutions is selected for
breeding from the pool selected
previously. By producing a "child" solution
using the above methods of crossover and
mutation, a new solution is created which
typically shares many of the characteristics
of its "parents". New parents are selected
for each new child, and the process
continues until a new population of
solutions of appropriate size is generated.
Although reproduction methods that are
based on the use of two parents are more
"biology inspired", some research suggests
that more than two "parents" generate
higher quality chromosomes.
The fitness function is defined over the

genetic representation and measures the
quality of the represented solution. The
fitness function is always problem
dependent. For instance, in the knapsack
problem one wants to maximize the total
value of objects that can be put in a
knapsack of some fixed capacity. A
representation of a solution might be an
array of bits, where each bit represents a
different object, and the value of the bit (0
or 1) represents whether or not the object
is in the knapsack. Not every such
representation is valid, as the size of
objects may exceed the capacity of the
knapsack. The fitness of the solution is the
sum of values of all objects in the knapsack
if the representation is valid, or 0
otherwise.
7.4.6.2 GA definitions and terminologies

Gene:
A single encoding of part of the solution
space, i.e. either single bits or short blocks
of adjacent bits that encode an element of
the candidate solution.
Chromosome:
A string of genes that represents a
solution.
In some problems, it is hard or even

impossible to define the fitness expression;
in these cases, a simulation may be used to
determine the fitness function value of a
phenotype (e.g. computational fluid
404
Population:
The number of chromosomes available to
test.
improvements based on the variationselection-paradigm. Thereby, the variation

operators produce genetic diversity and
the selection directs the evolutionary
search.
Adaptation:
process of adjustment to given (outer)
conditions.
Evolutionary computation (EC):

Computation based on evolutionary
algorithms.
Bit mutation/ bit flipping:

Random negation of a bit position (gene)
within the genome of an individual in the
case of binary representation
Fitness:
evaluation of an individual with respect to
its reproduction capability. Selection in EA
is based on the fitness. Generally, it is
determined on the basis of the objective
value(s) of the individual in comparison
with all other individuals in the selection
pool. The fitness function may additionally
depend
on
different
side
conditions/constraints and stochastic
influences. The term ``fitness function'' is
often used as a synonym for objective
function.
Representation:
Also coding or problem representation,
the mapping of the optimization problem
to the genome level.
Convergence:
Approach to a stationary system state,
where the variation rate of the considered
state variables goes to zero. It is necessary
to
distinguish gene
convergence (or
premature convergence) from the
convergence to optimal states. The latter is
divided into local and global convergence
depending on whether a local or a global
optimum is approached.
Fitness-proportionate selection:
Stochastic selection method where the
selection probability of an individual is
proportional to its fitness. Therefore,
fitness must be non-negative (scaling
function). Only fitness maximization is a
possible evolution goal.
Crossover (CO, XO):

Special form of recombination in GA
where two parents, generally, produce two
offspring by gene exchange.
Gene pool:
Total amount of all genes in a population.
Crossover rate:
Probability of a CO execution.
Generation:
(Natural) time unit of the EA, an iteration
step of the EA, a complete cycle comprising
the procreation and evaluation (fitness
determination and selection) of one or
more individuals.
Elitism/ elitist selection:

Property of selection methods, which
guarantees the survival of the best
individual(s).
Evolutionary algorithm (EA):
Collective term for all variants of
(probabilistic)
optimization
and
approximation algorithms that are inspired
by Darwinian evolution. Optimal states are
approximated
by
successive
Genetic programming (GP):

Variant of EA, works with genomes of
variable length and is used to evolve
symbolic information. The evolution of
405
Mutation/ mutation operator:
Variation operator, which changes the
information contained in the genome of a
parent according to a given probability
distribution. In the case of bit strings, this
is realized by random negation (bit
mutation, mutation rate) of single bits. For
real-valued object parameters a random
number from a suitable distribution is
added (mutation strength).
computer code, symbolic regression, and

automatic electrical circuit design are main
application domains. A common form
of representation is
the
tree-like
arrangement of genes as a program.
Genotype:
In EA with genotype-phenotype mapping,
the genotype is the representation on
which the crossover and mutation
operators are applied to.
In GP, mutation is usually achieved by the

deletion of a sub-tree, the addition of a
randomly generated sub-tree or by the
exchange of a random block (node). If
necessary, the individual has to be
repaired after mutation in order to obtain
a consistent program.
Individual:
Carrier of the genetic information
(chromosome). It is characterized by its
state
in
the
search
space,
its fitness (objective function), and, where
applicable, a set of strategy parameters.
The individual is the unit of selection.
Mutation rate:
Mutation probability of a single
gene/object parameter of an individual.
With respect to binary representation, the
mutation rate is the probability of flipping
a single bit position. The mutation rate
does not directly relate to the mutation
strength/step size.
Intermediate recombination:
Production of one recombinant in realvalued search spaces by calculating the
centroid (center of mass) of parents, a
special case of (weighted) arithmetic
recombination.
Inversion:
Mutation operator, which extracts a
randomly
chosen
fraction
of
a chromosome and reinserts the genes in
reverse order. In ordering problems.
Objective function/ quality

function:
Also goal function, the function to be
optimized (optimization), depending on
the object parameters (also referred to as
search space parameters or phenotype
parameters).
Multi-criteria optimization:
Also multi-objective optimization, vector
optimization, or multi-criteria decision
making. Optimization with regard to
multiple objective functions aiming at a
simultaneous improvement of the
objectives.
The objective function constitutes the

implementation of the problem to be
solved. The input parameters are the
object parameters. The output is the
objective
value
representing
the
evaluation/quality
of
the
individual/phenotype. In a multi-criteria
optimization problem,
the
objective
function returns multiple objective values.
Multi-recombination:
Variation operator, which recombines
more than two parents in order to produce
one or more offspring.
406
The objective value is often referred to

as fitness.
generation to the next in search space

(e.g. progress rate) or in the space of the
objective function values (quality gain).
Object parameter:
Also search space parameter, genotype or
phenotype variables which directly
influence the fitness. The objective
function directly depends on the object
parameters. In general, EA aims at
identifying those object parameters which
maximize (or minimize) the objective
function.
Global performance measures estimate

the resources which are necessary to
locate or approximate a local/global
optimum (e.g. time complexity, expected
running time).
Permutation operator:
Mutation operator, which changes the
order of genes in a chromosome,
particularly used for ordering or scheduling
problems.
Optimization:
Main application domain of EAs.
Optimization is the process of determining
or approximating object parameter sets
that maximize resp. minimize a
given objective function(s). A distinction is
made between global and local
optimization (local search). Optimal states
are always defined with respect to a
given neighborhood in the search space.
For global optima the neighborhood
includes the whole search space.
Depending on the structure of the search
space parameter, combinatorial, or
structure optimization are distinguished.
Phenotype:
Expression of the properties coded by the
individual's genotype.
The
expression/development of the phenotype
can additionally be influenced by
(stochastic) constraints. The precise
definition is mostly problem-dependent.
For parameter optimization the phenotype
is usually identical with the object
parameters, whereas for structure
optimization (e.g. of neural networks) the
phenotype represents a specific structure.
Population:
Pool of individuals exhibiting equal or
similar genome structures, which allows
the application of genetic operators.
Parent:
Individual which has been selected for
reproduction due to its fitness.
Performance:
General term concerning the evaluation of
the optimization behavior/quality of EAs. It
is often used synonymously for monitoring
the
evolution
of fitness or objective
function values over the generations
(fitness dynamics).
Population size:
Number of individuals in a population.
Problem size:
Number
of object
parameters,
components, or degrees of freedom of an
optimization problem.
Performance measures:
Local and global performance measures
are differentiated.
Progress rate:
Theoretical
performance
measure,
expected value of the distance change to a
predefined goal (usually, the optimal
solution) in one generation
Local performance measures evaluate the

optimization power of the EA from one
407
Rank-based selection:
Also ranking selection. The individuals'
selection probabilities are assigned
according to the individuals' rank which is
based on the objective function values.
selection methods are truncation selection

, tournament
selection, fitnessproportionate
selection (realized
as
roulette selection or stochastic universal
sampling) and their transformed versions
with ranking (ranking-based selection), as
well as Boltzmann-selection. All of these
selection methods reinsert the individuals
into the current population after selection,
i.e., individuals can be selected more than
once.
Real-coded GA:
GA tailored for the optimization in realvalued search spaces. In contrast to the
canonical GA, the genome consists of (realvalued) object parameters, i.e., evolution
operates on the ``natural'' representation.
The real-coded GA employs special
recombination operators, which are hybrid
constructs
of intermediate
recombination and mutation. Real-coded
GAs can exhibit self-adaptive behavior.
Selection pool:
The part of the population containing all
individuals which constitute mutual
reproduction partners (parents).
Selection pressure:
Ratio of the best individual's selection
probability to the average selection
probability
of
all
individuals
in
the selection pool.
Recombination:
Variation operator, which merges the
genetic information of two or more parent
individuals producing one or more
descendants, the so-called recombinants.
In
canonical GAs,
normally
two
recombinants
are
produced
by crossover using two parents.
Roulette selection:
Randomized
variant
proportionate selection.
Stopping condition:
Also termination criterion or stop rule,
conditions
which
determine
the
termination of the evolutionary process
(examples: number of objective function
evaluations, maximum run time)
of fitness-
Selection/ selection operator:

Necessary operator in EA, which,
depending on the fitness or the objective
function values, respectively, decides on
the reproduction of the individuals.
Selection gives a direction to evolution, it
guides the search (e.g., in the sense of
optimization or amelioration). It conserves
successful states, but at the same time it
reduces the diversity of the population. In
this sense, selection is the antagonist
to variation, which produces diversity. If
the selection method guarantees the
conservation of the information of the best
individual, it is called ``elitist''. Popular
408
409
7.4.6.3 GA Operators
7.4.6.3.1 Selection
In selection the offspring producing
individuals are chosen. The first step is
fitness assignment. Each individual in the
selection pool receives a reproduction
probability depending on the own
objective value and the objective value of
all other individuals in the selection pool.
This fitness is used for the actual selection
step afterwards.
7.4.6.3.1.1 Selection operator types:
Figure 7.18 Roulette wheel selection
Elitism selection (ES):
Ranked based RWS (RBRWS):
In this method, each individual is assigned

a fitness value via the fitness function.
Using these scores, a percentage of the
best, most fit individuals are used as
parents.
Modified roulette wheel selection

algorithm where each individual is
assigned a fitness value equal to its rank in
the population: the highest rank has the
highest probability to be selected. The
probability is calculated as illustrated in the
Roulette wheel selection (RWS):

Figure 7.18, the conspicuous characteristic
of this selection method is the fact that it
gives to each individual i of the current
population a probability of being selected,
proportional to its fitness f (i). () =
()
= ()
following equation:
2
_( +1)
Stochastic remainder RWS (SRRWS):

In Stochastic Remainder Selection
Mechanism, relative fitness of a
chromosome is calculated as the fitness
associated with an individual divided by
average fitness, say mi = fi/favg. Integer
part of mi is used to select one parent
deterministically and then uses roulette
selection on the remaining fractional part
to fill up the free places in the mating pool.
The
drawback of this technique is the risk of

premature convergence of the GA to a
local optimum, due to the possible
presence of a dominant individual that
always wins the competition and is
selected as a parent.
Steps:
1. =
.
2. For
the
1
()
410
first
parent:
1 =
3. For
=
the
remaining
parents:
()
Stochastic universal sampling (SUS):

The SUS is a variant of RWS aimed at
reducing the risk of premature
convergence. It can be implemented
according to the following steps:
1. Calculate the mean
1
= ( ) =1 ()
Once the n probabilities are computed, the

ERS procedure can be implemented by
randomly choose 2 parents based on their
probability.
fitness:
2. Generate
[0,1]
random
Tournament selection (TS):
number:
Its principle consists in randomly selecting

a set of k individuals. These individuals are
then ranked according to their relative
fitness and the fittest individual is selected
for reproduction for the 2 parents.
3. Sum = (1) ; delta = ; j = 1 ;

4. if delta < sum
select the jth
5. individual & delta = delta+sum
Tournament selection is best explained

with a concrete example. Suppose you
want to pick 20 individuals from 100
randomly
choose
(with
uniform
probability) a small number of individuals
(typically fewer than 10) from the 100
(with replacement). Keep the fittest one.
Do this again and again until you have got
your 20.
6. else j = j+1 & sum = sum+ (i)

7. while j<n
Linear rank selection (LRS):
It is based on the rank of individuals rather
than on their fitness. The rank n is
accorded to the best individual whilst the
worst individual gets the rank 1. Thus,
based on its rank, each individual i has the
probability of being selected given by the
expression: () =
Truncation selection (TRS):

The truncation selection is a very simple
technique that orders the candidate
solutions of each population according to
their fitness. Then, only a certain portion of
the fittest individuals are selected and
()
(1)
Once all individuals of the current

population are ranked, the LRS procedure
can be implemented by randomly choose 2
parents based on their probability.
reproduced
times. It is less used in
practice than other techniques, except for

very large population.
Exponential rank selection (ERS):

The ERS is based on the same principle as
LRS, but it differs from LRS by the
probability of selecting each individual. For
ERS, this probability is given by the
expression:
Steps:
1. Order the N individuals of P (t)
according to their fitness.
With,
411
2. Set the portion p of individuals to

select (e.g. 10% < p < 50%).
3. These selected parents produce
uniform at random offspring.
Individual 1
7.4.6.3.2 Recombination / Crossover.

Recombination produces new individuals
in combining the information contained in
two or more parents. This is done by
combining the variable values of the
parents. Depending on the representation
of the variables different methods must be
used.
Individual 2
Crossover point
After crossover the new individuals are

created:
Binary/Gray crossover operators:
Child 1
This section describes recombination

methods for individuals with binary
variables. Commonly, these methods are
called 'crossover'. Thus, the notion
'crossover' will be used to name the
methods.
Child 2
Double-point crossover:
In double-point crossover two crossover
positions are selected uniformly at random
and the variables exchanged between the
individuals between these points. Then
two new offspring are produced.
During the recombination of binary

variables only parts of the individuals are
exchanged between the individuals.
Depending on the number of parts, the
individuals are divided before the
exchange of variables (the number of cross
points). The number of cross points
distinguish the methods.
Consider the following two individuals with

12 binary variables each:
Single-point crossover:
Individual 1
In single-point crossover one crossover

position k[1,2,...,Nvar-1], Nvar: number
of variables of an individual, is selected
uniformly at random and the variables
exchanged between the individuals about
this point, then two new offspring are
produced.
Individual 2
Crossover points

created:

Child 1
412
bits in the mask indicate which parent will

supply the offspring with which bits.
Child 2

Multi-point crossover:
For multi-point crossover, m crossover
positions ki[1,2,...,Nvar-1], i=1:m, Nvar:
number of variables of an individual, are
chosen at random with no duplicates and
sorted into ascending order. Then, the
variables between successive crossover
points are exchanged between the two
parents to produce two new offspring.
Individual 1
Individual 2

Sample/Mask
created:
Individual 1
Child 1
Individual Crossover points
Child 2

created:
7.4.6.3.3 Mutation
By mutation individuals are randomly
altered. These variations (mutation steps)
are mostly small. They will be applied to
the variables of the individuals with a low
probability (mutation probability or
mutation rate). Normally, offspring are
mutated after being created by recombination.
Child 1
Child 2
For the definition of the mutation steps

and the mutation rate two approaches
exist:
Uniform crossover / Masked crossover:

Uniform crossover generalizes this
scheme to make every locus a potential
crossover point. A crossover mask, the
same length as the individual structure is
created at random and the parity of the
Both parameters are constant during a

whole evolutionary run.
413
One or both parameters are adapted

according to previous mutations.
Ex:
7.4.6.3.3.1 Binary / Gray mutation
Proto-child
operators:
Single flip mutation:
Child
Select randomly one point.
7.4.6.3.3.2 Real mutation operators:
Flip the bit value :
Steps:
1. Select one gene randomly.
New =
2.
Ex:
New =
3.
Proto-child
Example:
=70%
Child
For the industries of runout
Min value = 70 %
value = 83 %
100%
Double flip mutation:

Select randomly two point.
Old
Max value =
New =
Flip the bit value:
+
10083
83+
= 91.5%.
New =
Ex:

= 83 -
8370
2
= 76.5%
Proto-child
Center inverse mutation (CIM):
Child
The chromosome is divided into two

sections. All genes in each section are
copied and then inversely placed in the
same section of a child.
Uniform randomly mutation:

Select randomly one point.
Change the bit value randomly:
414
Thrors mutation:
Produce less offspring than parents and

replace the worst parents.
Three genes are chosen randomly which

shall take the different positions not
necessarily successive i < j < k. the gene of
the position i becomes in the position j and
the one who was at this position will take
the position k and the gene that has held
this position takes the position i.
Fitness-based reinsertion:
Produce more offspring than needed for
reinsertion and reinsert only the best
offspring.
Elitist combined
reinsertion:
Select randomly i < j < k
with
fitness-based
Prevents losing of information and is the

recommended method.
7.4.6.3.4 Reinsertion
Once the offspring have been produced by
selection, recombination and mutation of
individuals from the old population, the
fitness of the offspring may be determined.
If less offspring are produced than the size
of the original population then to maintain
the size of the original population, the
offspring have to be reinserted into the old
population. Similarly, if not all offspring are
to be used at each generation or if more
offspring are generated than the size of the
old population then a reinsertion scheme
must be used to determine which
individuals are to exist in the new
population.
Reinsertion methods
Pure reinsertion:
Produce as many offspring as parents and
replace all parents by the offspring.
Uniform reinsertion:
Produce less offspring than parents and
replace parents uniformly at random.
Elitist reinsertion:
415
binary genetic algorithm in terms of

efficiency and quality of solution.
7.5 Matlab Optimization

Toolbox
7.5.1 GA toolbox:
Matlab Optimization Toolbox provides

methods that search for global solutions to
problems that contain multiple maxima or
minima. It includes global search, multistart, pattern search, genetic algorithm,
and simulated annealing solvers. You can
use these solvers to solve optimization
problems where the objective or
constraint function is continuous,
discontinuous, stochastic, does not
possess
derivatives,
or
includes
simulations or black-box functions with
undefined values for some parameter
settings.
1) Calling the genetic algorithm

function ga at the command line.
2) The graphical user interface
Genetic algorithm and pattern search

solvers support algorithmic customization.
You can create a custom genetic algorithm
variant by modifying initial population and
fitness scaling options or by defining
parent selection, crossover, and mutation
functions. You can customize pattern
search by defining polling, searching, and
other functions.
Figure 7.19 graphical user interface
Matlab optimization tool is used for the

following reasons:
Using the graphical interface is more

restricted than the command line.
Using the command line gives you more
space to write your functions and control
them.
It provides many built in auxiliary

functions
useful
for
function
optimization.
It is completely portable.
It
is efficient for
numerical
computations.
The genetic algorithm toolbox developed is
tested on a series of non-linear, multimodal, non-convex test problems
compared with results using simulated
annealing.
The genetic algorithm using a float
representation is found to be superior to
7.5.2 GA toolbox components:
416
Fitness Functions.
Variables.
Populations.
Crossover.
Mutation.
Plot.
Main GA code.
The main line of the function is

constant you shouldnt change the
sequence & it consists of three
variables (Nvar, FitnessFnc, options)
Nvar: no of variables
FitnessFnc: name of fitness
function M-file.
Options: its the options of GA.
7.5.3 How the genetic algorithm

works?
1. The algorithm begins by creating a
random initial population.
2. The algorithm then creates a sequence
of new populations, or generations. At
each step, the algorithm uses the
individuals in the current generation to
create the next generation. To create
the new generation, the algorithm
performs the following steps:
Scores each member of the current
population by computing its fitness
value.
Scales the raw fitness scores to
convert them into a more usable
range of values.
Selects parents based on their
fitness.
Produces children from the
parents. Children are produced
either by making random changes
to a single parent mutation or
by combining the vector entries of
a pair of parents crossover.
Replaces the current population
with the children to form the next
generation.
3. The algorithm stops when one of the
stopping criteria is met.
Set the total population size by two

different ways
You can set that its equal to (number).
Like: total population = 20;
You can set the population size from
the option in the main GA code.
For loop will contain the main code to
create the population
Creating population is up to you
depending on the case you are working
on.
7.5.5 Crossover
Crossover types:
Crossover heuristic.
Crossover scattered.
Crossover intermediate.
Crossover single point.
Crossover two point.
Crossover arithmetic.
You cant use the all types for your case
because every type has its method for
crossover.
7.5.4 Population
Structure:
Function pop=population(Nvar,FitnessFcn,options)
totalPopulationSize = sum(options.PopulationSize);
pop = cell(totalPopulationSize,1);
for i=1:totalPopulationSize
% main code
end
figure 7.20
417
first for loop counts the number of new

kids
r1 is the first parent
r2 is the second parent
v is the new chromosome
second for loop to define from any
parent the bit of the new chromosome
is formed.
XoverKides: is the new chromosomes
7.5.5.1 Crossover scattered

Structure:
Function xoverKids =
crossover_pop(parents,options,NVARS, ...
FitnessFcn, thisScore,thisPopulation)
nKids =length(parents)/2;
xoverKids = cell(nKids,1);
index=1;
for i=1:nKids
r1 = parents(index);
index = index + 1;
index = index + 1;
v=zeros(1,NVARS);
for j = 1:NVARS
if rand > 0.5
b= thisPopulation{r1};
v(j)= b(j);
else
v(j)= b(j);
end
end
xoverKids{i} = v;
end
Crossover scattered is a random crossover

or in other word its multi points crossover
7.5.5.2 Crossover single point
Structure
function xoverKids =
crossover_pop(parents,options,NVARS, ...
FitnessFcn,thisScore,thisPopulation)
index=1;
for i=1:nKids
r1 = thisPopulation{parents(index)};
r2 = thisPopulation{parents(index+1)};
index = index + 1;
c=NVARS;
t=randsample(c,1)
child= [ r1 (1 : t), r2 (( t + 1 ) : end) ];
xoverKids { i } = child;
end
Figure 7.21

constant, you shouldnt change the
sequence & it consists of six variables
(parents, options, NVARS, FitnessFnc,
thisScore, thisPopulation)
Parents: they are the population
which you create.
ThisScore: the score which it has
after fitness function.
ThisPopulation:the
population
during this generation, because
every generation has its own
population due to crossover and
mutation.
nKids: number of new chromosomes in
the population, it must be an integer &
even number
the population is in the form of cells
figure 7.22
single point form for crossover is the

simplest type.
The point of crossover is defined randomly
in the range of chromosome length, the
following line show that:
c=NVARS; t=randsample(c,1)
Chromosome length is equal to number of
variables.
After defining the crossover point then we
make the crossover from the following
line.
you can see it just one line
xoverKids{i} = [r1(1 : t),r2(( t + 1 ) : end)];
418
7.5.5.3 Crossover two points
7.5.6.1 Binary system
Structure
You just change zero to one or one to zero

Structure
function xoverKids =
crossover_pop (parents,options,NVARS, ...
function mutationChildren =
mutate_pop(parents ,options,NVARS, ...
FitnessFcn, state, thisScore,thisPopulation)
index=1;
for I = 1 : nKids
r1 = thisPopulation { parents (index) };
r2 = thisPopulation { parents (index+1) };
index=index+2;
t=randi (NVARS,1,2);
t=sort(t);
child =
[r1(1:t(1)), r2(t(1)+1:t(2)), r1(t(2)+1:end)];
xoverKids { i } = child;
end
mutationRate=0.1;
mutationChildren = cell (length(parents),1);
for i=1:length (parents)
child = thisPopulation {parents (i) , :} ;
mutationPoints =
find(rand(1,length(child)) < mutationRate);
for j = 1:length(mutationPoints)
b= mutationPoints(j);
if child(b)==0
child(b)=1;
else
child(b)=0;
end
end
mutationChildren{i} = child;
end
Figure 7.23
It is like single point crossover but "t" will

be matrix its dimension 1x2.
So you define the two points for crossover
random then sort them to maintain the
sequence of the integer numbers.
Figure 7.24

constant, you shouldnt change the
sequence & it consists of seven
variables (parents, options, NVARS,
FitnessFnc,
state,
thisScore,
thisPopulation)
After you sort the matrix, you used the

following line from the above code to
make the two points crossover.
The line of two point will be:
xoverKids{i} =
Mutation rate is an indication to how

much you want to mutate
If it is equal to small number, it will be
better in the performance.
As you can see that the for loop counts
all chromosomes.
Mutation is done according to the
mutation rate.
But the question here is how to change
the numbers in chromosomes?
It can be done by using the second for
loop and if condition.
[r1(1:z),r2(( z + 1 ):y), r1((y+1 : end))];
7.5.6 Mutation
Mutation types:
Mutation for binary system.
Mutation for real number.
Every type has its own way to mutate
419
7.5.6.2 Real system
7.5.6.2.2 Two variable mutation
In this type you cant change zero to

one or vice versa
Its a real number, would be integer or
not.
It has a different construction for its
oeration.
The following form is suitable for most
cases for optimization by using GA.
Structure
mutationRate=0.1;
mutationChildren = cell(length(parents),1);
for I = 1 : length (parents)
child = thisPopulation { parents(i), : };
mutationPoints =
find ( rand (1 , length(child)) < mutationRate) ;
for m=1:length(mutationPoints)
p=mutationPoints(m);
if p<=5 %% limit number for first variable
mut1=1:5;
child(p) = mut1 (randi(length(mut1),1,1));
else
mut2= zeros(1,5);
child(p) = mut2(randi(length(mut2),1,1));
end
end
end
7.5.6.2.1 One variable mutation

Structure
mutationRate=0.1;
for i=1:length(parents)
child = thisPopulation { parents(i), : };
mutationPoints =
find ( rand(1,length(child)) < mutationRate);
mut = 1.5 : 0.025 : 2.2 ;
child (mutationPoints) =
mut(randi([1 length(mut)] , ...
length(mutationPoints),1));
end
Figure 7.26
Every variable will have a known

number of bits.
So in this code to determine which
variable will mutate, we use the second
for loop with if condition.
If condition is used with "p" matrix to
determine if the position in the first or
second variable.
mut: is the matrix from which the
optimization tool chooses the values
for each variable.
Mut1: is for first variable.
Mut2: is for second variable.
If you have any numbers of variables,
you can increase the conditions in if
condition.
Figure 7.25
"mut" matrix is responsible for the

operation of the function, in this matrix
you can set the range of the variable
where the tool chooses the value of the
variable for mutation.
After you set the positions of mutation,
you change the values in this position
by values from mut matrix, thats can
be done by using the following line:
child (mutationPoints) = mut ( randi ([1
length(mut)] , length (mutationPoints),1));
After mutation the chromosome is
assigned to mutation children cell to
use them in the next generation after
selection.
You can build your function of mutation as

you want depending on your case.
The most important function in
optimization is fitness function, in this
420
function you can set what do you want to

maximize or minimize.
7.5.8 Main GA code

clc,clear
inputs=[9]; %%[comp]
FitnessFcn = @(x)cost_fitness(x,inputs);
options = gaoptimset('PopulationType', 'custom');
options =
gaoptimset(options,'CreationFcn',@population, ...
'CrossoverFcn',@crossover_pop, ...
7.5.7 Fitness function

Structure
Function
scores=cost_fitness(x,inputs)
'MutationFcn',@mutate_pop,'PlotFcn',@gaplotbestf ,
'Generations',30,'PopulationSize',20,
'SelectionFcn',@selectionroulette, ...
'StallGenLimit',6,'Vectorized','on');
numberOfVariables = 9;
[x,fval,reason,output] = ...
ga(FitnessFcn,numberOfVariables,options)
scores = zeros(size(x,1),1);
for i=1:size(x,1)
p=x{i}
% main code
scores(i)=V;
end
Figure 7.27
The function
( x , inputs)
contain
Figure 7.28
two
variables
Options of GA:
Population type is set custom because
you create the population
In the option you define every
function you have made:
Population
Crossover
Mutation
Plot: its set as default
X: is the cell that contains all population

Inputs: it is a matrix that contains inputs
which are used in the fitness function main
code.
P is a matrix that contains each
chromosome from the population, we use
P matrix in order to take each chromosome
in each loop and the for loop count is equal
to the population length.
Generation: contains the number of

generation
Population size: is the number of
chromosomes
Selection: there are many types for
selection
Selection remainder
Selection uniform
Selection stochunif
Selection roulette
Selection tournament
Stall limit: is the number of generation
at which the results still constant, after
this limit the GA stop its operation
Vectorized on: to increase the tool
speed
Main code differs from code to other

according to your case to calculate V.
V is the variable that you need to minimize
or maximize according to the case.
Each chromosome contains the values of
each variable and these variables are used
to calculate V, so V differs from
chromosome to other as they dont
contain the same values.
421
You create the options structure using the

function gaoptimset. options = gaoptimset
This returns the structure options with the
default values for its fields. options =
7.5.9 Running the Genetic Algorithm

with the Default Options
To run the genetic algorithm with the
default options, call ga with the syntax
[x fval] = ga(@fitnessfun, nvars)
PopulationType: 'doubleVector'
PopInitRange: [2x1 double]
PopulationSize: 20
EliteCount: 2
CrossoverFraction: 0.8000
MigrationDirection: 'forward'
MigrationInterval: 20
MigrationFraction: 0.2000
Generations: 100
TimeLimit: Inf
FitnessLimit: -Inf
StallLimitG: 50
StallLimitS: 20
InitialPopulation: []
InitialScores: []
PlotInterval: 1
CreationFcn: @gacreationuniform
FitnessScalingFcn: @fitscalingrank
SelectionFcn: @selectionstochunif
CrossoverFcn: @crossoverscattered
MutationFcn: @mutationgaussian
HybridFcn: []
PlotFcns: []
OutputFcns: []
Vectorized: 'off'
The input arguments to ga are:
@fitnessfun: A function handle to the

M-file that computes the fitness
function, you can use this form if the
fitness function is function in X only
without inputs.
nvars: The number of independent
variables for the fitness function.
The output arguments are:

X: The final point.
fval: The value of the fitness function at
x.
Additional Output Arguments:
To get more information about the
performance of the genetic algorithm, you
can call ga with the syntax.
[x fval reason output population scores] =
ga(@fitnessfcn, nvars)
Besides x and fval, this returns the
following additional output arguments:
Reason: Reason the algorithm
terminated.
Output:
Structure
containing
information about the performance of
the algorithm at each generation.
population: Final population.
scores: Final scores.
Figure 7.29
The function ga uses these default values

if you do not pass in options as an input
argument. The value of each option is
stored in a field of the options structure,
such as options.PopulationSize. You can
display any of these values by entering
options followed by the name of the field.
To create an options structure with a field
value that is different from the default
for example to set PopulationSize to 100
instead of its default value 20 enter:
options = gaoptimset('PopulationSize',
100)
7.5.9.1 Setting Options

You can specify any of the options that are
available in the Genetic Algorithm Tool by
passing an options structure as an input
argument to ga using the syntax.
[x fval] = ga(@fitnessfun, nvars, options)
422
This creates the options structure with all

values set to their defaults except for
PopulationSize, which is set to 100.
If you now enter:
ga (@fitnessfun, nvars, options)
ga runs the genetic algorithm with a
population size of 100.
If you subsequently decide to change
another field in the options structure, such
as setting PlotFcns to @gaplotbestf, which
plots the best fitness function value at each
generation, call gaoptimset with the syntax
options = gaoptimset (options, 'PlotFcns',
@plotbestf)
This preserves the current values of all
fields of options except for PlotFcns, which
is changed to @plotbestf. Note that if you
omit the input argument options,
gaoptimset resets PopulationSize to its
default value 20. You can also set both
PopulationSize and PlotFcns with the single
command
options = gaoptimset('PopulationSize',100,
'PlotFcns',@plotbestf)
[x, fval, reason, output, final_pop2] =

ga(@fitnessfcn, nvars);
If you want, you can then use final_pop2,
the final population from the second run,
as the initial population for a third run.
7.5.9.2 Resuming ga from the Final

Population of a Previous Run
By default, ga creates a new initial
population each time you run it. However,
you might get better results by using the
final population from a previous run as the
initial population for a new run. To do so,
you must have saved the final population
from the previous run by calling ga with the
syntax
[x, fval, reason, output, final_pop] =
ga(@fitnessfcn, nvars);
The last ouput argument, is the final
population. To run ga using final_pop as
the initial population, enter
Options=gaoptimset('InitialPop', final_pop
);
423
The Current generation field displays

the number of the current generation.
The Status and results pane displays
the message GA running.
When the algorithm terminates, the Status
and results pane displays
7.6 The graphical user

interface
7.6.1 Opening the Genetic
Algorithm Tool:
To open the tool, enter "gatool" at the

MATLAB command window.
This opens the Genetic Algorithm Tool
From Apps, optimization, solver then
choose GA
7.6.2 Defining a Problem in the

Genetic Algorithm Tool
You can define the problem you want to
solve in the following two fields:
Figure 7.32
The message: GA terminated.

The fitness function value of the
best individual in the final
generation
The
reason
the
algorithm
terminated
The coordinates of the final point
You can change many of the settings in the
Genetic Algorithm Tool while the algorithm
is running. Your changes are applied at the
next generation. Until your changes are
applied, which occurs at the start of the
next generation, the Status and Results
pane displays the message Changes
pending. At the start of the next
generation, the panel displays the message
Changes applied.
Figure 7.30
Fitness function: The function you

want to minimize. Enter a handle to an
M-file function that computes the
fitness function. Writing an M-File for
the Function You Want to Optimize
Number of variables: The number of
independent variables for the fitness
function.
7.6.3 Running the Genetic Algorithm

To run the genetic algorithm, click Start in
the Run solver pane. When you do so,
Figure 7.33
Figure 7.31
424
7.6.4 Pausing and Stopping the

Algorithm
While the genetic algorithm is running,
you can:
Click Pause to temporarily suspend the
algorithm. To resume the algorithm
using the current population at the
time you paused, click Resume.
Click Stop to stop the algorithm. The
Status and results pane displays the
Stall time limit: The algorithm runs for

the specified amount of time in
seconds with no improvement in the
fitness function.
The above options are shown in figure 7.34
fitness function value of the best point

in the current generation at the
moment you clicked Stop
Note: If you click Stop and then run the
genetic algorithm again by clicking Start,
the algorithm begins with a new random
initial population or with the population
you specify in the Initial population field. If
you want to restart the algorithm where it
left off, use the Pause and Resume
buttons.
figure 7.34
If you want the genetic algorithm to

continue running until you click Pause or
Stop, you should change the default
values of these options as follows:
7.6.4.1 Setting Stopping Criteria

The genetic algorithm uses five criteria,
listed in the Stopping criteria options, to
decide when to stop, in case you do not
stop it manually by clicking Stop. The
algorithm stops if any one of the following
conditions occur:
Generations: The algorithm reaches

the specified number of generations.
Time: The algorithm runs for the
specified amount of time in seconds.
Fitness limit: The best fitness value in
the current generation is less than or
equal to the specified value.
Stall generations: The algorithm
computes the specified number of
generations with no improvement in
the fitness function.
Set Generations to Inf.

Set Time to Inf.
Set Fitness limit to -Inf.
Set Stall generations to Inf.
Set Stall time limit to Inf.
figure 7.35
425
Note: Do not use these settings when

calling the genetic algorithm function ga
at the command line, as the function will
never terminate until you press Ctrl + C.
Instead, set Generations or Time limit to a
finite number.
point with the best fitness value in the

current generation.
7.6.6 Reproducing Your Results

To reproduce the results of the last run of
the genetic algorithm, select the Use
random states from previous run check
box. This resets the states of the random
number generators used by the algorithm
to their previous values. If you do not
change any other settings in the Genetic
Algorithm Tool, the next time you run the
genetic algorithm, it returns the same
results as the previous run.
Normally, you should leave Use random
states from previous run unselected to get
the benefit of randomness in the genetic
algorithm. Select the Use random states
from previous run check box if you want to
analyze the results of that particular run or
show the exact results to others.
7.6.5 Displaying Plots

The Plots pane enables you to display
various plots of the results of the genetic
algorithm.
figure 7.36
Select the check boxes next to the plots

you want to display. For example, if you
select Best fitness and Best individual and
run, the tool displays the plots shown in
the following figure.
7.6.7 Setting Options for the

Genetic Algorithm
You can set options for the genetic
algorithm in the Options pane
figure 7.37
The upper plot displays the best and mean

fitness values in each generation. The
lower plot displays the coordinates of the
figure 7.38
426
7.6.8 Importing and Exporting Options

and Problems
You can export options and problem
structures from the Genetic Algorithm Tool
to the MATLAB workspace, and then
import them back into the tool at a later
time. This enables you to save the current
settings for a problem and restore them
later. You can also export the options
structure and use it with the genetic
algorithm function ga at the command line.
You can import and export the following
information:
The problem definition, including
Fitness function and Number of
variables
The currently specified options
The results of the algorithm
7.6.8.1 Exporting Options and

Problems
You can export options and problems to
the MATLAB workspace so that you can
use them at a future time in the Genetic
Algorithm Tool. You can also apply the
function ga using these options or
problems at the command line.
To export options or problems, click the
Export button or select Export to
Workspace from the File menu. This opens
the dialog box shown in the following
figure
Export problem and options to a

MATLAB structure named and enter a
name for the structure. Clicking OK
saves this information to a structure in
the MATLAB workspace. If you later
import this structure into the Genetic
Algorithm Tool, the settings for Fitness
function, Number of variables, and all
options settings are restored to the
values they had when you exported the
structure.
If you want the genetic algorithm to
resume from the final population of the
last run before you exported the
problem, select Include information
needed to resume this run. Then, when
you import the problem structure and
click Start, the algorithm resumes from
the final population of the previous
run. To restore the genetic algorithms
default behavior of generating a
random initial population, delete the
population in the Initial population
field and replace it with empty
brackets, [].
To save only the options, select Export
options to a MATLAB structure named
and enter a name for the options
structure.
To save the results of the last run of the
algorithm, select Export results to a
MATLAB structure named and enter a
name for the results structure.
Note:
figure 7.39
The dialog provides the following options:

To save both the problem definition
and the current options settings, select
427
If you select Use random states from

previous run in the Run solver pane
before exporting a problem, the
Genetic Algorithm Tool also saves the
states of rand and randn at the
beginning of the last run when you
export. Then, when you import the
problem and run the genetic algorithm
with Use random states from previous
run selected, the results of the run just

before you exported the problem are
reproduced exactly.
If you select Include information
needed to resume this run, then
selecting Use random states from
previous run has no effect on the initial
population created when you import
the problem and run the genetic
algorithm on it. The latter option is only
intended to reproduce results from the
beginning of a new run, not from a
resumed run.
Number of variables
The options fields
7.6.9 Genetic Algorithm Options
Diversity
Population Options
Fitness Scaling Options
Selection Options
Reproduction Options
Mutation Options
The Crossover Fraction
Setting the Maximum Number of
Generations
7.6.8.2 Importing Options

To import an options structure from the
MATLAB workspace, select Import Options
from the File menu. This opens a dialog box
that displays a list of the genetic algorithm
options structures in the MATLAB
workspace. When you select an options
structure and click Import, the options
fields in the Genetic Algorithm Tool are
updated to display the values of the
imported options.
You can create an options structure in
either of the following ways: Calling
gaoptimset with options as the output
By saving the current options from the
Export to Workspace dialog box in the
Genetic Algorithm Tool
7.6.9.1 Diversity
7.6.8.3 Importing Problems
Population
options
control
the
characteristics of the individuals in the
population. This section describes the
following options:
One of the most important factors in

determining how well the genetic
algorithm performs is the diversity of the
population. If the average distance
between individuals is large, the diversity
is high; if the average distance is small, the
diversity is low. Diversity affects the
performance of the genetic algorithm. If
the diversity is too large or too small, the
algorithm might not perform well.
You can control the amount of diversity in
the population by various options settings,
including the Initial range and the amount
of mutation in Mutation options.
7.6.9.2 Population Options
To import a problem that you previously

exported from the Genetic Algorithm Tool,
select Import Problem from the File menu.
This opens the dialog box that displays a
list of the genetic algorithm problem
structures in the MATLAB workspace.
When you select a problem structure and
click OK.
the following fields are updated in the
Genetic Algorithm Tool:
Fitness function
7.6.9.2.1 Setting the Initial Range

By default, the Genetic Algorithm Tool
creates a random initial population using
the creation function. You can specify the
range of the vectors in the initial
population in the Initial range field in
Population options.
428
If you know approximately where the

solution to a problem lies, you should
specify the initial range so that it contains
your guess for the solution. However, the
genetic algorithm can find the solution
even if it does not lie in the initial range,
provided that the populations have
enough diversity.
reproduce too rapidly, taking over the

population gene pool too quickly, and
preventing the genetic algorithm from
searching other areas of the solution
space. On the other hand, if the scaled
values vary only a little, all individuals have
approximately the same chance of
reproduction and the search will progress
very slowly.
7.6.9.2.2 Setting the Population Size
The default fitness scaling function, Rank,

scales the raw scores based on the rank of
each individual instead of its score. The
rank of an individual is its position in the
sorted scores: the rank of the most fit
individual is 1, the next most fit is 2, and so
on. The rank scaling function assigns scaled
values so that
The Size field in Population options

determines the size of the population at
each generation. Increasing the population
size enables the genetic algorithm to
search more points and thereby obtain a
better result. However, the larger the
population size, the longer the genetic
algorithm takes to compute each
generation.
The scaled value of an individual with

rank n is proportional to (1 ).
The sum of the scaled values over the
entire population equals the number of
parents needed to create the next
generation.
Rank fitness scaling removes the effect of
the spread of the raw scores.
The following plot shows the raw scores of
a typical population of 20 individuals,
sorted in increasing order.
You can experiment with different settings

for Population size that return good results
without taking a prohibitive amount of
time to run.
Note: You should set Size to be at least the
value of Number of variables, so that the
individuals in each population span the
space being searched.
7.6.9.3 Fitness Scaling Options
Fitness scaling converts the raw fitness
scores that are returned by the fitness
function to values in a range that is suitable
for the selection function. The selection
function uses the scaled fitness values to
select the parents of the next generation.
The selection function assigns a higher
probability of selection to individuals with
higher scaled values.
The range of the scaled values affects the
performance of the genetic algorithm. If
the scaled values vary too widely, the
individuals with the highest scaled values
429
Because top scaling restricts parents to the

fittest individuals, it creates less diverse
populations than rank scaling. The
following plot compares the variances of
distances between individuals at each
generation using rank and top scaling.
Figure 7.40
Because the algorithm minimizes the

fitness function, lower raw scores have
higher scaled values. Also, because rank
scaling assigns values that depend only on
an individuals rank.
Comparing Rank and Top Scaling
To see the effect of scaling, you can
compare the results of the genetic
algorithm using rank scaling with one of
the other scaling functions, such as Top. By
default, top scaling assigns the four fittest
individuals the same scaled value, equal to
the number of parents divided by 4, and
assigns the rest the value 0. Using the
default selection function, only the four
fittest individuals can be selected as
parents.
figure 7.42
7.6.9.4 Selection Options

The selection function chooses parents for
the next generation based on their scaled
values from the fitness scaling function. An
individual can be selected more than once
as a parent, in which case it contributes its
genes to more than one child. The default
selection function, Stochastic uniform, lays
out a line in which each parent
corresponds to a section of the line of
length proportional to its scaled value. The
algorithm moves along the line in steps of
equal size. At each step, the algorithm
allocates a parent from the section it lands
on.
A more deterministic selection function is
Remainder, which performs two steps:
In the first step, the function selects
parents deterministically according to
the integer part of the scaled value for
each individual. For example, if an
individuals scaled value is 2.3, the
function selects that individual twice as
a parent.
Figure 7.41
430
with the best fitness values, the algorithm

creates.
Crossover children by selecting vector
entries, or genes, from a pair of
individuals in the current generation
and combines them to form a child.
Mutation children by applying random
changes to a single individual in the
current generation to create a child.
Both processes are essential to the genetic
algorithm.
Crossover
enables
the
algorithm to extract the best genes from
different individuals and recombine them
into potentially superior children.
Mutation adds to the diversity of a
population and thereby increases the
likelihood that the algorithm will generate
individuals with better fitness values.
Without mutation, the algorithm could
only produce individuals whose genes
were a subset of the combined genes in the
initial population.
In the second step, the selection

function selects additional parents
using the fractional parts of the scaled
values, as in stochastic uniform
selection. The function lays out a line in
sections,
whose
lengths
are
proportional to the fractional part of
the scaled value of the individuals, and
moves along the line in equal steps to
select the parents.
7.6.9.5 Reproduction Options

Reproduction options control how the
genetic algorithm creates the next
generation. The options are
Elite count: The number of individuals

with the best fitness values in the
current generation that are guaranteed
to survive to the next generation.
These individuals are called elite
children. The default value of Elite
count is 2.
When Elite count is at least 1, the best
fitness value can only decrease from
one generation to the next. This is what
you want to happen, since the genetic
a lgo r it h m min imiz e s t h e f it n e ss
function. Setting Elite count to a high
value causes the fittest individuals to
dominate the population, which can
m a k e t h e s e a r c h le s s e f f e c t i v e .
Crossover fraction: The fraction of
individuals in the next generation,
other than elite children, that are
created by crossover.
7.6.9.7 Mutation Options

The genetic algorithm applies mutations
using the function that you specify in the
Mutation function field. The default
mutation function, Gaussian, adds a
random number, or mutation, chosen from
a Gaussian distribution, to each entry of
the parent vector. Typically, the amount of
mutation, which is proportional to the
standard deviation of the distribution,
decreases as at each new generation. You
can control the average amount of
mutation that the algorithm applies to a
parent in each generation through the
Scale and Shrink options:
Scale controls the standard deviation
of the mutation at the first generation,
which is Scale multiplied by the range
of the initial population, which you
specify by the Initial range option.
7.6.9.6 Mutation and Crossover

The genetic algorithm uses the individuals
in the current generation to create the
children that make up the next generation.
Besides elite children, which correspond to
the individuals in the current generation
431
Shrink controls the rate at which the

average
amount
of
mutation
decreases. The standard deviation
decreases linearly so that its final value
equals 1-Shrink times its initial value at
the first generation. For example, if
Shrink has the default value of 1, then
the amount of mutation decreases to 0
at the final step.
Figure 7.43
7.6.9.8 The Crossover Fraction

The final point is close to the true
minimum at (1, 1). You can improve this
result by setting Hybrid function to
fminunc in Hybrid function options.
The Crossover fraction field, in the

Reproduction options, specifies the
fraction of each population, other than
elite children, that are made up of
crossover children. A crossover fraction of
1 means that all children other than elite
individuals are crossover children, while a
crossover fraction of 0 means that all
children are mutation children. The
following example show that neither of
these extremes is an effective strategy for
optimizing a function.
Figure 7.44
When the genetic algorithm terminates,

the function fminunc takes the final point
of the genetic algorithm and as its initial
point and returns a more accurate result,
as shown in the Status and results pane.
7.6.9.9 Using a Hybrid Function

A hybrid function is an optimization
function that runs after the genetic
algorithm terminates in order to improve
the value of the fitness function. The
hybrid function uses the final point from
the genetic algorithm as its initial point.
You can specify a hybrid function in Hybrid
function options.
Before adding a hybrid function, trying
running the genetic algorithm by itself, by
clicking Start. The genetic algorithm
displays the following results in the Status
and results pane.
Figure 7.45
432
7.7 Instructions to write

the code
Create the population is the first

function that you have to write
because after it has been written, you
will know the chromosome length and
you will be able to write mutation and
crossover functions and.
After you know the chromosome
length then you will be able to decide
which crossover type will be used, if
the chromosome length is small you
can use single o two points crossover,
but if the length is large then you can
use crossover scattered (multi- points).
When you write the mutation function,
you have to know the variables that
you mutate, if it is binary system or real
number to build your function.
Fitness function differs from case to
case because in each case you have a
different target and if you want to
minimize or maximize your score.
433
7.8 Optimization
Implementation
In this chapter we will illustrate the goal

that we achieved by using genetic
algorithm.
You will know the difference between

variables, constrains and target that you
reached by optimization.
Optimum solutions that have been
represented to operate the Natural Gas
National Grid.
7.8.1.3 Target (objective function)

The main targets from optimization in
network problem are:
Min cost for the operation of
compressors.
Max flowrate for the demands.
Results from the code at each case that

contains demands flow, pressures for gas
power stations and steam power stations.
7.8.1 Optimization Techniques:
If we used the compressors as a solution,

the target will be fuel or electrical cost
minimization.
The power stations pressure will be the
constraints to us as it has a required
pressure to operate at maximum efficiency
without increasing the required flow.
For another case, the operation without
compressors, the pressure of power
stations must be achieved, but there are
no compressors in the model except
Dahshour, so we have to minimize the
output flow for demands according to the
priority of each demand.
There is three main parts of optimization

techniques, you have to recognize the
difference between each part to be able to
write any code for optimization.
7.8.1.1 Variables
Variables are the parameters that you can
change in the network to achieve the
optimum solution, like:
Medium constraints:
constraints that can be accepted if you
have no choice and the hard
constraints arent broken.
Soft constraints:
constraints that if the solution broke it,
the solution will be accepted but it will
not be the best.
Valves: change the valve constant to

change the flow direction and to close
or partial close before any demands.
Compressor stations: open or close the
compressor station to achieve the
optimum pressure to demands
Regulators and part of demands flow.
So the constraints will be the power

stations pressure and the target is to
increase the demands flowrate.
7.8.1.2 Constraints
Constraints differ from case to another
depending on what you want.
You can divide it to many section just like:
Hard constraints:
constraint that cant be broken and if
the solution breaks any one of them,
the solution must be ignored.
434
7.8.2 Optimization loop between MATLAB and Synergi
Initial Population
Simulator
synergi script
Fitness Function
matlab sign
Scaling
simulation by synergi
Selection
Crossover
write output data
Mutation
synergi sign
Simulator
Fitness Function
Stopping Criteria
Yes
Optimum
Results
435
End
No
7.8.2.1 MATLAB and Synergi

First thing in this loop is begin with matlab,
Create the population randomly, each

chromosome contains number of bits
according to the variables in your case,
Population size is an important factor
as it increases, you increase the area of
search.
After you create the population, you
write them in text file to be sent to the
simulator after that a sign is sent to the
script to begin the simulation.
After simulation the output data is
written in another text file to calculate
the fitness function, then a sign is sent
to matlab.
According to the output data the
matlab compares these data with the
constraints, if it is ok, the matalab then
calculate the fitness function, if not,
the GA then ignore this solution.
After that you check on stopping
criteria, if it is true then the GA will
stop, if it is false then the GA go to
scaling.
Every chromosome in the population is
given a value in scaling according to its
score from fitness function, after
scaling the selection function chooses
chromosomes for the new generation.
Mutation and crossover is done to the
previous population to create new
chromosomes to change the solutions.
The new generation is now containing
chromosomes from the previous
population and chromosomes from
crossover
and
mutation,
the
percentage between them depends on
the selection tool.
The new generation is sent to the

simulator again and to make the
simulation and write the output data to
text file and send them to matlab.
Script is a tool to make a repeated

simulation according to the script code.
Every simulation you have a new
chromosome with new solution so you its
impossible to make each simulation
manually and take the data to the matlab
manully too.
Script code helps us to make the above
operations automatically without the need
to transport any data between matlab and
synergi manually.
436
7.9.3 Methods to solve the traveling

salesman problem
7.9 Case studies

7.9.1 Travelling salesman problem
7.9.3.1 Heuristics methods
The Traveling Salesman Problem (TSP) is

one of the most widely studied
combinatorial optimization problems. Its
statement is deceptively simple, and yet it
remains one of the most challenging
problems in Operational Research.
Hundreds of articles have been written on
the TSP.
According to Karp, we can partition the

problem to get an approximate solution
using the divide and conquer techniques.
We form groups of the cities and find
optimal tours with in these groups. Then
we combine the groups to find the optimal
tour of the original problem. Karp has given
probabilistic analyses of the performance
of the algorithms to determine the average
error and thus the average performance of
the solution compared to the optimal
solution.
The most common practical interpretation

of the TSP is that of a salesman seeking the
shortest tour through n clients or cities.
This basic problem underlies several
vehicle routing applications, but in this
case a number of side constraints usually
come into play. Several interesting
permutation problems not directly
associated with routing can also be
described as TSP.
7.9.3.2 greedy algorithms

Assume the asymmetric traveling
salesman problem. We use the symbol of
(Kn,c) where c is the weight function and n
is the number of vertices. We assume the
symmetric traveling salesman problem to
be defined in the same way but Kn symbols
a complete undirected graph. If we try to
find an approximate solution to an NP-hard
problem using heuristics, we need to
compare
the
solutions
using
computational experiments. There is a
number called domination number that
compares the performance of heuristics.
7.9.2 Example
Consider the following set of cities: [A, B, C,
D, E and F], The problem lies in finding a
minimal path passing from all vertices
once.
7.9.3.3 Using the triangle inequality

we create a minimum spanning tree the
weight of which is a lower bound on the
cost of an optimal traveling salesman tour.
Using this minimum spanning tree, we will
create a tour the cost of which is at most 2
times the weight of the spanning tree. We
present the algorithm that performs these
computations using the MST-Prim
algorithm.
For example, the path Path1 {A, B, C, D, E,

F, A}, and the path Path2 {A, B, F, C, D, E, A}
pass all the vertices.
Each path has its length, so we have to find
the minimum distance between cities.
437
After crossover it will be reversed at the

crossover point and will be
7.9.3.4 using the Gas
0 0 0 0 0 0 0 | 1 1 1 1 1 1 1
Initial Population
It just changes the sequence, so it was a

special one for this case.
Fitness Function
The reversed point is chosen randomly.

7.9.3.4.3 Mutation
In this case we use a special one, we
determine two points randomly and
change the number of cities between
them.
Selection
Crossover
If the two positions are 2 & 10 the

mutation will be like that,
Mutation
1 1 1 1 1 1 1 0 0 0 0 0 0 0
1 0 1 1 1 1 1 0 0 1 0 0 0 0
Stopping Criteria
7.9.3.4.4 Fitness function

Fitness function is main function that
control the GA, and in this case the fitness
function is depending on the total distance
between the cities.
End
7.9.3.4.1 initial population
each chromosome will have a different
sequence to pass through all cities.
GA will choose the min distance between

cities as an optimum solution.
The sequence of the cities is set randomly

by the tool.
The cities before optimization as shown in

figure 7.46
The length of the chromosome is equal to

the number of cities.
7.9.3.4.2 crossover
in this case we didnt use the ordinary
types of crossover, but we use a special
one.
We used a type to change the sequence of
the cities.
Figure 7.46
1 1 1 1 1 1 1 | 0 0 0 0 0 0 0
438
Figure 7.47 Route between cities at the beginning of optimization
7.9.3.4.5 Results from optimization
Figure 7. 48 Final route between cities
7.9.3.4.6 Output from Matlab

Optimization terminated: average change in the fitness value less than options.TolFun.
x = [1x40 double]
fval = 5.0415
reason = 1
output =
problemtype: 'unconstrained'
rngstate: [1x1 struct]
generations: 854
funccount: 51300
message: 'Optimization terminated: average change in the fitness value less than options.TolFun.
439
Total population size is a function in

options, that we set the total population
size in the options of the GA.
7.9.4 optimum pipe diameter

In this case we have three pipes, the length
of them is known and the flowrate that
passes through them is known.
Each pipe has different flowrate.
"x" is the matrix that contains the

allowable diameter range.
Flow at A > B > C
"n" is equal to variable number.
So we have to determine the optimum

diameter of each pipe to:
Pop: is the population cell that contains all

chromosomes.
In the for loop we fill the cell with the three

diameters in each chromosome randomly.
minimize the cost of the pipes

minimize the power required by the
pump which is at the beginning of the
pipeline.
7.9.4.2 Crossover
In this case we use single point crossover
due to the chromosome length.
flow direction
7.9.4.1 Population
Chromosome length will be three, contains
three diameters from the range that you
define:
Figure 7.50 crossover code
Crossover is a function in 6 variables

(parents, options, Nvar, FitnessFnc,
thisScore, ThisPopulation)
D = 0.1 : 0.025 : 2.2 "dimension in meter "

So the we will make a code that chooses
three diameters randomly from the range
which is defined.
Number of new kids is equal to half

number of parents.
Thats can be done by this code:
The crossover point is determined

randomly by using "randi" (integer
number).
In the for loop we make crossover between
every sequent chromosomes according to
their index till the final number of the kids.
Then we save the new chromosomes in
xoverKids cell.
Figure 7.49 population code
Population is a function in 3 variables

(Nvar, FitnessFnc, options)
440
7.9.4.3 Mutation
Total head loss along the pipeline
We use random points for mutation

according to mutation rate.
h =
0.81 1 2
0.82 2 2
2
0.83 3 2
3 5
= + * h
= h

1
S.P = 1000
cost = S.P *
1
m
*Price ($/Kwh) *W-hour
(hr/year)
Figure 7.51 mutation code
Initial Cost
Mutation is a function in seven variables

(parents, options, Nvar, FitnessFnc, state,
thisScore, ThisPopulation)
Mutation rate is equal to 0.1

Mutation for all parents, but the point of
mutation depending on mutation rate,
some of chromosomes doesnt mutate
according to mutation rate.
Material is Cast Iron

Density () = 7800 3
Thickness (t) = 6 mm
Cost of one Ton = 590 $
Volume of the total pipeline
V= 4 *( d 2 di 2 )* L
Total mass
If random number is higher than or equal

to mutation rate, then mutation doesnt
happen.
m=*V
Cost =
m
1000
kg
* 590 $
The random number is done by using

mutation points in the code
Constrain: Input pressure must be less
than 7 bar.
We fill the mutation children cell by using

the for loop.
Target: optimum diameters to the pipeline

sections with minimum cost.
The value of diameter that have been

changed according to the mutation, the
value that have been changed is defined
from (mut) matrix.
7.9.4.5 Results
Pipes diameters
d1 = 24 in
d2 = 21 in
d3 = 18 in
All chromosomes after mutation are saved

in mutationChilden cell.
7.9.4.4 Fitness function
Total cost = 1.3848 million dollar

Inlet pressure = 4.2 bar
Fitness function is depending on two

variables
The total head loss along the pipeline

to minimize the power of the pump.
The total cost including: the running
cost and the initial cost
441
7.9.5 National grid
7.9.5.3 solutions
7.9.5.1 Problems
Decrease the demands flowrate

according to their priority, this solution
increases the pressure in the national
grid and maximize the demand
flowrate.
Using of compressor stations to
increase the pressure of the natural gas
grid without decreasing the demand
flowrate.
Pressure of power stations is below the

allowable limit.
Allowable gas power station or
combined power station pressure
is equal to 27 bar.
Allowable steam power station
pressure is equal to 7 bar.
Required flow rate for demands is
higher than supply flow rate.
7.9.5.2 Target
Provide the required pressure for
power stations (gas and steam).
Maximize the demands flowrate.
Figure 7.52 location of compressor station
7.9.5.3.1 Compressor stations

We propose eight locations for new
compressor stations, but will we use all this
compressors?
We propose the locations according to our

experience that we have during the study
of the national grid.
Of course not, it just a proposition and the

optimization by using genetic algorithm
will choose the optimum locations of the
new compressor stations.
If you want to propose a new location, you

have to choose location in closed loop,
near to gas source that have flowrate
enough to reach the required pressure and
442
the pressure drop is large across this

pipeline.
7.9.5.3.1.1 Steady state analysis
Pressures before we use the solution of
compressor station.
Pressures after we use the solution of

compressor station.
Figure 7.54 gas ps. pressure after optimization

Figure 7.53 Gas ps. Pressures before optimization
You can see that the pressure results from

the model before we use the compressors
solution doesnt match the allowable
pressure range to operate with high
efficiency.
The target from this solution is to increase
the pressures at all power stations to the
required range with minimum power
required for all compressors.
The optimum locations of the compressors
will be related to the optimization code by
using random population each one of them
contains random locations for the
compressor and the optimum solution will
be found according to the optimization
constraints and target.
443
Figure 7.55 Steam ps. After optimization
After optimization the pressures of gas

power stations and steam power stations
are within the required range.
Four compressors are in working in the
model for optimum solution which
increase the pressure in the grid to the
required value.
Figure 7.56 Compressor stations location from

optimization
Compressors that are working in the

models are:
Dahshour (Co1)
Shabshir (Co4)
Fayoum(Co5)
UGDC(Co6)
444
7.9.5.3.1.2 Unsteady state analysis
Pressures before optimization
Figure 7.57 Gas power stations Pressures for 24 hour
The pressures of power stations are out of allowable range, so we have to choose which
compressors will make the pressures stable at the allowable range.
After we propose the eight locations, we use the GA to find the optimum solution for the
operation for 24 hr.
after optimization
Figure 7.58 Locations of compressor stations
After optimization the code choose three compressors to operate in the grid with minimum
required power for compressors operation.
By using these compressors, the pressures at all power stations are within the range.
But what about the power of compressor stations for 24hr, it will be illustrated in the figure,
every compressor has its own power according to the location of the compressor in the grid.
445
7.9.5.3.1.3 Result pressures from optimization
Figure 7.59 Gas power stations Pressures for 24 hour
Figure 7.60 steam power stations Pressures for 24 hour
pressures of compressor stations are stable for 24hr which increase the efficiency of the
operation of the national grid.
The natural gas reaches gas power stations at pressure 27 bar and steam power stations at 9
bar which Achieve our target from optimization.
446
Figure 7.61 Power of compressor stations
The power differs from compressor to other according to the pressure of the grid.
The power of compressor 4 is constant for 24hr but compressor 2 is working for few hours.
Linepack
Figure 7.62 linepack for 24hr
Linepack in the national grid for 24hr is varying between 83.5 and 97 mmsm 3, this range is
suitable for our model.
This linpack results from the variation in power stations load and load of industries.
447
7.9.5.3.2 Flow Reduction Strategy
7.9.5.3.2.1 Understanding the
notion
Here in this case study we aim to solve a
problem that already exist in the national
grid which is shortage of the amount of gas
in the grid which will deteriorate the
pressures at some power station
decreasing its efficiency and threatening
the national security .
Supplies load and pressures

No Description
Pressure Flow
by
Category
Base
Volumetric
(sm^3/hr)
1
WDGC
60.37
934500
2
LPG(badreldin) 59.14
625000
3
aboqir
51.72
259000
4
rashpetco
54.13
1085000
5
dahshour
46.73
123000
6
elmansoura
45.7
126000
7
abo madi
51.5
630000.063
8
UGDC(port
56.28
1657000
fouad)
9
north sinai
53.38
70000
10 ras shouqir
48.11
25000
11 ras bakr
48.26
25000
At this point the reaction from the

company (GASCO) which operate the grid
is reducing the flow of some industries
randomly or base upon experience and
consultants , but we now offer a better
solution which is using modern technology
, IT power and state of art science to
determine the best reduction data to all
consuming industries which will guarantee
the optimal minimum reduction of the
total flow and maintain the pressures at
acceptable operating range.
The model contain 34 power station In

total 18 of them are gas Power station and
the others are steam .
In our model the regulating pressure of the
gas power stations is 27 Bar and steam is 9
Bar.
We will use Genetic Algorithm to Achieve

this
strategy
using
the
matlab
programming and built in synergi
programming language (Script { C++}).
Gas Power Stations
Now we will implement some data about

our model which will indicate how we
achieved our idea
7.9.5.3.2.2 Model Data
After we draw the model on Arc Gis and
imported the data to synergi the model
was in balance which means that the total
load of supply field is equal to total load of
the all consumers ( industries , power
stations , and local districts )
Total Demand
(sm^3/hr)
Total Supply
(sm^3/hr)
5559499.961
5559499.961
448
Description
Pressure
krir
5
,
6
combined
elsyof gas
damanhor
combined
el
mahmoudia
combined
nubaria
combined
6 october gas0
korymat comb
banha gas
noth
cairo
combined
27
Flow
(sm^3/hr)
-90000
27
27
-50000
-46000
27
-68000
27
-380000
27
27
27
27
-150000
-123000
-95000
-168000
south
cairo
combined
wadi hoof gas
talkha combined
talkha gas
west
damietta
gas
new damietta gas
damietta
combined
elshabab gas
port said gas
el
attf
new
combined
27
-56000
27
27
27
27
-11000
-125000
-64000
-140000
27
26.96
-138000
-208000
27
27
27
-30000
-14000
-138000
Now in order to simulate our shortage

state we need to add a assumed consumer
which will cause a deterioration of the
pressures at the industrial zone , we
assume that the consumer will consume
140000 Sm^3/hr .
The result of this step is that many gas
power station decreased under minimum
operating pressures .
Steam Power Stations

Description
Pressure Flow
(sm^3/hr)
intergen 3
9
-138000
krir 1
9
-42000
abo qir new
9
-51000
abo qir steam
9
-87000
kafr el dawar 9
-74000
steam
west
cario 9
-63000
steam
shobra el khema 9
-33000
steam
teppin steam
9
-14000
talkha steam
9
-51000
west suez steam 9
-153000
mousa steam
9
156000.016
ataka steam
9
-46000.004
east port said 9
-180000
steam
al arish steam
9
-19000
abo
sultan 9
-28000.002
steam
korymat steam
9
126999.992
449
Description
krir 5,6 combined
elsyof gas
damanhor combined
el mahmoudia combined
nubaria combined
Pressure
27
27
27
27
18.44
6 october gas0
25.15
korymat comb
27
banha gas
noth cairo combined
26.22
25.43
south cairo combined

wadi hoof gas
talkha combined
26.31
26.43
26.53
talkha gas
west damietta gas
26.85
27
new damietta gas
22.96
damietta combined
22.91
elshabab gas
port said gas
el attf new combined
27
27
27
Flow
-90000
-50000
-46000
-68000
380000
150000
123000
-95000
168000
-56000
-11000
125000
-64000
140000
138000
208000
-30000
-14000
138000
7.9.5.3.2.3 Discussion of the code
Now we will run both codes
simultaneously ( matlab code and synergi)
As result of this solution the gas power

stations have been recovered as well the
entire industrial zone area with only a
reduction in industries flow with
percentage of 94%.
First matlab will produce random flow

rates for all industries but within
predefined range, this step called
population initialization.
Each one of this solution will be exported
to synergi to run and solve the network
hydraulically, then we will take the new
result pressures of power station and
export it to matlab which will guarantee
that its above 27, then matlab will
calculate a predefined objective function
trying to minimize it as possible.
This process will repeat itself till all
population solution are applied, then the
population solution will be ordered based
upon related objective function value, this
process called scaling, then best of these
solution is selected to continue to the next
generation.
Before the next generation is exported to
synergi, there is inevitable process must be
done first which will guarantee the variety
and efficiency of the solution which is
mutation and crossover.
In our case study the number of the
solutions within population is 100, the
estimated number of iterations to produce
this solution is 2300 iteration, now we will
exhibit the flow rates of the entire 87
industries before the optimization and
after it indicating the percentage of
reduction and resulted pressures of gas
power stations.
We must anotate that the objective
function is the reverse of the productions
of the percentages of the industries flow
(After Optimization/ Before Optimization)
450
Total Industries flow

Before Optimization
(sm^3/hr)
Total Industries flow

After Optimization
(sm^3/hr)
1911232
2016100
Before Optimization
No
1
2
3
NodeDescripti
on
natgas borg el
arab
natgas agamy
After Optimization
NodeFlow
(sm^3/hr)
-34620
NO Node Description Node

Flow
(sm^3/hr)
1
natgas borg el -34620
arab
2
natgas agamy
-3790
-3790
natgas
west -7300
alex
town gas alex
-15000
natgas west alex
-7300
town gas alex
-15000
natgas
kafr -8540
eldawar
town
gas -16000
portsaid
transgas delta -13000
5
6
natgas
kafr -8540
eldawar
town gas portsaid -16000
transgas delta
-40000
egypt gas delta(1) -40000
-2700
quesna(egyptgas) -2700
-40000
10
egyptgas delta(2)
-15400
11
-45000
12
egyptgas
abo -15400
zabal
egyptgas cairo 1
-45000
13
egypt
gas
delta(1)
quesna(egyptg
as)
egyptgas
delta(2)
egyptgas abo
zabal
egyptgas cairo
1
natgas cario
-15000
13
natgas cario
-15000
14
towngas cairo
-72000
14
towngas cairo
-72000
15
natgas
6 -35000
october
egyptgas cairo -45000
2
towngas
el -5000
ismailia
city gas
-124000
15
natgas 6 october
-35000
16
egyptgas cairo 2
-45000
17
towngas
ismailia
city gas
-124000
midor
medalic
star nex
19
4
5
6
7
8
9
10
11
12
16
17
18
19
20
18
and -64500
-1830
20
451
-13000
-40000
el -5000
midor
and -62249
medalic
star nex
-1825
NO
Description
Flow
NO
Description
Flow
21
misr chemical
-6100
21
misr chemical
-5542
22
alex petrolum
-19700
22
alex petrolum
-18450
23
amoic
-10500
23
amoic
-10438
24
anrpc
-6770
24
anrpc
-6534
25
polyprobelin
-21800
25
polyprobelin
-20669
26
26
27
elmogammaa
for -46000
petrol
amryia petroleum
-26600
27
elmogammaa
for -42528
petrol
amryia petroleum
-25458
28
misr petroluem
-160
28
misr petroluem
-148
29
sidpec
-7000
29
sidpec
-6986
30
petrochemical
-27000
30
petrochemical
-25516
31
carbon black factory
-7500
31
carbon black factory
-6906
32
fenosa facilities
-4000
32
fenosa facilities
-3742
33
methanics
-23000
33
methanics
-22978
34
morgan
-31000
34
morgan
-29667
35
gapco
-7000
35
gapco
-6600
36
suez petroluem
-11500
36
suez petroluem
-11380
37
elnasr petroluem
-7500
37
elnasr petroluem
-7467
38
gas derivatives
-40000
38
gas derivatives
-37901
39
abo qir fert
-129800
39
abo qir fert
-124803
40
alexandria fert
-40000
40
alexandria fert
-37301
41
uoria and natarat
-72600
41
uoria and natarat
-68085
42
helwan fert
-50000
42
helwan fert
-49450
43
egypt for fert2
-42000
43
egypt for fert2
-36919
44
suez for fert
-15000
44
suez for fert
-13403
452
NO
Description
Flow
No
Descripiton
Flow
45
amrya cement
-2720
45
amrya cement
-2420
46
dkhela steel
-80000
46
dkhela steel
-69161
47
max cement
-17550
47
max cement
-17432
48
48
49
ganob
elwadi -6000
cement
katmya cement
-7400
49
ganob
elwadi -5983
cement
katmya cement
-6427
50
helwan cement
-20000
50
helwan cement
-19401
51
national cement
-37900
51
national cement
-35627
52
kamaen el top
-75000
52
kamaen el top
-65232
53
53
54
towngas iron and -35000

steel
swedy cement
-1000
54
towngas iron and -33884

steel
swedy cement
-972
55
suez cement
-1000
55
suez cement
-955
56
arabya cement
-1000
56
arabya cement
-893
57
egypt cement
-5000
57
egypt cement
-4578
58
alex free zone
-1870
58
alex free zone
-1787
59
alex sugar
-31200
59
alex sugar
-29469
60
nobaria sugar
-18030
60
nobaria sugar
-17895
61
alex IA
-18420
61
alex IA
-17702
62
dakhlia sugar
-26000
62
dakhlia sugar
-25948
63
repco gas
-2800
63
repco gas
-2706
64
64
65
damietta pump -4400

station
gastec talkha
-800
65
damietta
pump -3964
station
gastec talkha
-645
66
gazl el mahala
66
gazl el mahala
-18551
67
cebaghet el mahla -2800
67
cebaghet el mahla
-2467
68
10 of ram
-95000
68
10 of ram
-74481
69
hawamdya sugar
-13800
69
hawamdya sugar
-13062
70
egyptgas aswan
-10000
70
egyptgas aswan
-8500
-19500
453
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
fayoum gas oftake

nile vally oftake
qift
epec
gastec suez
sinai gas( cement
and IA)
aswan
wadi elnile cement
bani swaif
upper egypt IA
bani swaif cement
elmania
elmania AI
abo qurqas
asyout
naga hamadi
qena
-22300
-14000
-17700
-45000
-400
-15000
71
72
73
74
75
76
-30000
-5000
-5000
-5000
-5000
-10000
-1000
-6000
-10000
-11500
-22800
77
78
79
80
81
82
83
84
85
86
87
454
fayoum gas oftake

nile vally oftake
qift
epec
gastec suez
sinai gas( cement and
IA)
aswan
wadi elnile cement
bani swaif
upper egypt IA
bani swaif cement
elmania
elmania AI
abo qurqas
asyout
naga hamadi
qena
-21330
-12811
-17134
-44978
-394
-11955
-27825
-4613
-4756
-4938
-4593
-8151
-952
-5316
-8296
-10966
-21786
Conclusion
Egyptian natural gas grid has problem in the power stations pressures, due to
low pressure the power stations works at low efficiency and consume more flow
than the required in normal operation.
So to solve this problem we have two solutions:
1. Converting power to gas
This solution will increase the calorific value of natural gas and increase the
amount of gas in the national grid by blending hydrogen in the grid at WDGC
with percentage equal to 3% from the total output flow from WDGC.
Increasing the calorific value of natural gas will decrease the total
consumption of power stations and that will increase the linepack and
pressure in the national grid.
2. Optimization
We have two scenarios first one is demand reduction
In this scenario we reduce the amount of the required flow for industrial
demands according to their priorities, it differs from industrial demand to
other.
The reduction of flow will increase the amount of gas flowing in the
national grid which increase the pressure in the grid and the linepack.
Our target from this scenario is to increase the pressure at all power
stations gas or steam to the required level and minimize the amount of
flow reduction.
Second scenario is to install new compressor stations
Install new compressors stations in the grid will increase the pressure in
the grid without the need to reduce the flow at any demand.
But increasing the number of new compressor stations in the grid above
the optimum limit will decrease the pressure in the grid instead of
increasing it, so to reach the optimum solution GA will search for the
optimum solution to increase the pressure at power stations to the
required level.
We installed eight compressor stations beside Dahshour compressor and
the locations are:
WDGC, Abohoumos, Shabshir, Cairo, Fayoum, UDGC, East El-Tina and
Suez.
455
Our target from this scenario is to increase the pressure at all power stations gas
or steam to the required level and minimize the total power of compressor
stations.
So after running the GA, the optimum solution was including three compressor
stations where are in Abohoumos, Shabshir and East El-Tina.
456
References
Chapter 1
Gas pipeline hydraulic

Getting the best measurements from your turbine meters.
Natural gas pipeline technology overview.
Chapter 2
Dynamic Modeling and Simulation of MSF Desalination Plants

Parabolic Trough Reference Plant for Cost Modeling with the Solar Advisor Model (SAM)
Current (2009) State-of-the-Art Hydrogen Production Cost Estimate Using Water Electrolysis
Dynamic Modeling of Multi Stage Flash (MSF) Desalination Plant
A REVIEW ON WATER ELECTROLYSIS: Centre for Renewable Energy Sources (CRES), Pikermi,
Greece
Fundamentals of Membranes for Water Treatment: Alyson Sagle and Benny Freeman
Photovoltaic Systems in Existing Residential Building in Egypt: International Journal of
Scientific & Engineering Research Volume 2, Issue 7, July-2011
RENEWABLE ENERGY TECHNOLOGIES: COST ANALYSIS SERIES: International Renewable
Energy Agency
Links:
http://www.pv-magazine.com/news/details/beitrag/power-to-gas-platform-established-ineurope_100011076/#axzz4BgHhXjsY
http://www.powermag.com/progress-for-germanys-power-to-gas-drive/
http://blogs.worldwatch.org/from-wind-and-sun-to-gas-fraunhofer%E2%80%99s%E2%80%9Crenewable-methane%E2%80%9D-energy-storage-technology/
http://www.solar-energy-at-home.com/photovoltaic-solar-power.html
http://conceptselectronics.com/diodes/structure-pn-junction-diode/
http://www.odec.ca/projects/2008/full8e2/background.html
http://www.ontariosolarfarms.com/solar-panels.html
http://www.renewablegreenenergypower.com/solar-energy-pros-and-cons-pvsystems/#.UFtfyVEfo_d
https://www.environment.gov.au/system/files/resources/ef2c1cc7-07d8-4ed8-8f79816d36fb959e/files/desalination-summary.pdf
http://www.solidcell.com/techElectrolyzer.htm
http://www.nrel.gov/docs/fy13osti/51995.pdf
http://www.gtai.de/GTAI/Navigation/EN/Invest/Industries/Smarter-business/Smartenergy/power-to-gas.html
http://www.gtai.de/GTAI/Navigation/EN/Invest/Industries/Smarter-business/Smartenergy/power-to-gas,t=model-commercialization-projects-in-germany,did=547394.html
http://www.businessdictionary.com/definition/inflation.html
http://www.investopedia.com/terms/n/npv.asp
457
Chapter 3
GAS TURBINE THERMODYNAMIC AND PERFORMANCE ANALYSIS METHODS USING

AVAILABLE CATALOG DATA, Master Thesis by K.A.B.Pathirathna.
Thermodynamic performance analysis of gas-turbine power-plant by Rahman et al.
Thermodynamics An Engineering Approach 5th Edition - Gengel, Boles.
Power-Plant-Engineering-A-K-Raja
Power Generation Handbook
Gas Turbine Handbook Second Edition
Chapter 4
GAS TURBINE THERMODYNAMIC AND PERFORMANCE ANALYSIS METHODS USING

AVAILABLE CATALOG DATA, Master Thesis by K.A.B.Pathirathna.
Thermodynamic performance analysis of gas-turbine power-plant by Rahman et al.
Thermodynamics An Engineering Approach 5th Edition - Gengel, Boles.
Power-Plant-Engineering-A-K-Raja
Power Generation Handbook
Gas Turbine Handbook Second Edition
Chapter 5
Help of ArcGIS
User Guide of ArcGIS DESK 1, 2, 3
GIS and science.com
Bgis introduction to GIS .com
Chapter 6
SynerGEE Gas 4.7.0, User Guide

software.support@dnvgl.com
Chapter 7
[Randy_L._Haupt,_Sue_Ellen_Haupt]_Pratical_Genetic
[Singiresu_S._Rao]_Engineering_Optimization_Theory
[H._Paul_Williams]_Model_Building_in_Mathematical
Genetic Algorithms and Engineering Optimization [Mitsuo Gen and Runwei Cheng]
web.mit.edu/15.053/www/AMP-Chapter-13.pdf
www.tik.ee.ethz.ch/sop/publicationListFiles/zitz1999a.pdf
www.utwente.nl/ctw/wem/education/afstuderen/Loucks_VanBeek/05_chapter05.pdf
courses.mai.liu.se/FU/MAI0083/Report_Mina_Nikanfs.pdf
Bioinformatics - Genetic Algorithm And Direct Search Toolbox - Matlab
458
Appendix
Compressors code steady state
Population
function pop=population(Nvar,FitnessFcn,options)
totalPopulationSize = sum(options.PopulationSize);
x=[0 1]; % dim. in "m"
n=Nvar;
pop = cell(totalPopulationSize,1);
for i=1:totalPopulationSize
pop{i}= randi(x,1,n);
end
Crossover
function xoverKids = crossover_pop(parents,options,NVARS, ...
index=1;
for i=1:nKids
index = index + 1;
index = index + 1;
v=zeros(1,NVARS);
for j = 1:NVARS
if(rand > 0.5)
v(j)= b(j);
else
v(j)= b(j);
end
end
xoverKids{i} = v;
end
mutation
function mutationChildren = mutate_pop(parents ,options,NVARS, ...
mutationRate=0.1;
child = thisPopulation{parents(i),:};
mutationPoints = find(rand(1,length(child)) < mutationRate);
mut=[0 1];
child(mutationPoints) = mut(randi([1 length(mut)],
length(mutationPoints),1));
end
459
fitness function
function scores=cost_fitness(x,inputs)
%%%%%%%%%%%%%%%%%%%%%% no.of components of nodes and valves
ID3=fopen('C:\Users\Ahmed Bahaa\Desktop\comp file\text
file\components.txt','wt');
fprintf(ID3,'%f\n',inputs);
fclose(ID3);
iter=0;
Syn_sign=0.5;
ddddd=55;
flow=0;
power=10000000;
prs=0;
it=1;
for i=1:size(x,1)
p=x{i};
file\syn_input.txt','wt');
fprintf(ID1,'%f\n',p);
fclose(ID1);
%%%%%%%%%%%%%% matlab sign to make synergee start simulation
file\mat_firstsign.txt','wt');
fprintf(ID2,'%f\n',it);
fclose(ID2);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%% waiting for simulation end in synergee
for wait=1:100000;
pause(0.1);
try
Syn_sign = textscanu('C:\Users\Ahmed Bahaa\Desktop\comp
file\text file\mat_sign.txt');
ddddd = cellfun(@str2double,Syn_sign );
end
if ddddd ==it+1;
break
end
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%% Reading pressure values of gas power stations
try
file\text file\syn_output.txt');
pr = cellfun(@str2double,Syn_sign );
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%pressure of steam power stations %%%%%%%%
Try
Syn_sign = textscanu('C:\Users\Ahmed Bahaa\Desktop\comp file\text
file\steam.txt');
prs = cellfun(@str2double,Syn_sign );
end
460
%%%%%%%%%%%%%%%%%%%%%%% power of compressors %%%%%%%%%%%%%%%%%%%%%

try
file\text file\comp.txt');
Power = cellfun(@str2double,Syn_sign );
end
%%%%%%%%%%%% if simulation is unsolved or not feasible
if pr==1
scores(i)=10^6;
%%%%%%%%%%%% if simulation is solved & feasible
else
pp=min(pr);
% "pp" is the matrix of pressures of power stations
ppmax=27;
% set pressure of power station
prr=pp/ppmax;
if prr>=0.95
z1=1;
elseif prr<0.95 & prr>=0.9
z1=0.5;
else
z1=.001;
end
%%%%%%%%%%%%%%%%%%%%%%%%%%
ps=min(prs);
ps_max=7;
pss=ps/ps_max;
if pss>=0.95
z2=1;
elseif pss<0.95 & prr>=0.9
z2=0.5;
else
z2=.001;
end
%%%%%%%%%%%%%%compressors%%%%%%%%%%%%%%%
power_max=120348;
w2=Power;
ww1=z1*z2;
w1=1/ww1;
scores(i)=w1*w2;
end
iter=iter+1
%%%%%% Remove the data from text file
fclose(ID1);
fclose(ID2);
file\mat_sign.txt','wt');
fclose(ID3);
it=it+1;
end
461
GA main code
clc,clear
winopen('C:\Users\Ahmed Bahaa\Desktop\comp file\network model.lnk');
inputs=[9 19 16]; %%[comp psg pss]
options = gaoptimset(options,'CreationFcn',@population, ...
'MutationFcn',@mutate_pop,'PlotFcn',@gaplotbestf , ...
'Generations',20,'PopulationSize',20,'SelectionFcn',@selectionroulet
te, ...
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%% optimum simulation data %%%%%%%%%%%%%%%%%%%%
p=x{1};
fclose(ID1);
winopen('C:\Users\Ahmed Bahaa\Desktop\comp file\optimum.lnk');
462
Compressors code unsteady state

Population, mutation and crossover have the same code of steady state
fitness function
function scores=cost_fitness(x,inputs)
%%%%%%%%%%%%%%%%%%%%%% no.of components of nodes and valves
ID3=fopen('C:\Users\Ahmed Bahaa\Desktop\comp unsteady\text
file\components.txt','wt');
fprintf(ID3,'%f\n',inputs);
fclose(ID3);
iter=0;
ddddd=55;
it=1;
for i=1:size(x,1)
p=x{i};
fclose(ID1);
%%%%%%%%%%%%%% matlab sign to make synergee start simulation
fprintf(ID2,'%f\n',it);
fclose(ID2);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%% waiting for simulation end in synergee
for wait=1:100000;
pause(0.1);
try
unsteady\text file\mat_sign.txt');
ddddd = cellfun(@str2double,Syn_sign );
end
if ddddd ==it+1;
break
end
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%% Reading pressure values of gas power stations
T1 = readtable('C:\Users\Ahmed Bahaa\Desktop\comp unsteady\press
excel\pressuresg.csv');
B1= table2array(T1);
A1=B1(:,2:20);
u1=size(A1,1);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%pressure of steam power stations %%%%%%%%%
excel\pressuress.csv');
A2=B2(:,2:17);
u2=size(A2,1);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%% power of compressors %%%%%%%%%%%%%%%%%%%%%
463

excel\power.csv');
A3=B3(:,2:10);
zz1=sum(A3);
Power=sum(zz1)
%%%%%%%%%%%% if simulation is unsolved or not feasible
u3=u1+u2
if u3~=48
scores(i)=10^20;
else
%%%%%%%%%%%% if simulation is solved & feasible
pp=min(A1);
% "A1" is the matrix of pressures of gas power
stations
ppmax=27;
% set pressure of gas power station
prr=pp/ppmax;
if prr>=0.95
z1=1;
elseif prr<0.95 & prr>=0.9
z1=0.5;
else
z1=.0000001;
end
%%%%%%%%%%%%%%%%%%%%%%%%%%
ps=min(A2); % "A2" is the matrix of pressures of steam power
stations
ps_max=9;
% set pressure of steam power station
pss=ps/ps_max;
if pss>=0.95
z2=1;
elseif pss<0.95 & prr>=0.9
z2=0.5;
else
z2=.0001;
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%compressors
w2=Power;
ww1=z1*z2
w1=1/ww1;
scores(i)=w1*w2;
end
iter=iter+1
%%%%%% Remove the data from text file
fclose(ID1);
it=it+1;
end
464
GA main code
clc,clear
winopen('C:\Users\Ahmed Bahaa\Desktop\comp unsteady\usmodel.lnk');
inputs=[9]; %%[compressors]
options = gaoptimset(options,'CreationFcn',@population, ...
'MutationFcn',@mutate_pop,'PlotFcn',@gaplotbestf , ...
'Generations',30,'PopulationSize',20,'SelectionFcn',@selectionroulet
te, ...
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%% optimum simulation data %%%%%%%%%%%%%%%%%%%
p=x{1};
fclose(ID1);
winopen('C:\Users\Ahmed Bahaa\Desktop\comp unsteady\optimum.lnk');
Functions for mutation and crossover

Mutation code for two variable
function mutationChildren = mutate_pop(parents ,options,NVARS, ...
mutationRate=0.1;
child = thisPopulation{parents(i),:};
mutationPoints = find(rand(1,length(child)) < mutationRate);
for m=1:length(mutationPoints)
p=mutationPoints(m);
if p<=5 %%%%% limit number for first variable
mut1=1:5;
else
mut2=zeros(1,5);
end
end
end
465
single point crossover

index=1;
for i=1:nKids
parent1 = thisPopulation{parents(index)};
parent2 = thisPopulation{parents(index+1)};
index=index+2;
e=randi(NVARS,1);
child = [parent1(1:e),parent2(e+1:end)];
xoverKids{i} = child;
end
Two points crossover

index=1;
for i=1:nKids
parent1 = thisPopulation{parents(index)};
parent2 = thisPopulation{parents(index+1)};
index=index+2;
b=randi(NVARS,1,2);
b=sort(b);
child =
[parent1(1:b(1)),parent2(b(1)+1:b(2)),parent1(b(2)+1:end)];
xoverKids{i} = child;
end
466
P2G alternatives codes

Simulation code for the first alternative:
clc,clear
%%%%%%%%%%%% mass balance %%%%%%%%%%%%%%%%
m1=(0.03*1083000*0.0899)/3600 % Kg/s ('the production rate of hydrogen= ');
M=2.015*(10^(-3)) ; % the molar mass of hydrogen "g/mol"
eta_w=0.74; %('the efficiency of water use= ');
M_w=18.015*(10^(-3)) ; %the molar mass of water "g/mol"
m2=(m1/(M*eta_w))*M_w
m2vol= m2/1000
m3=4*m1 %('the potable water= ');
m3vol= m3/1000
m5=(m2+m3)/0.7
m5_vol =m5/1000
m4=0.3*m5 %saults
%%%%%%%%%%%% energy balance %%%%%%%%%
F=96485.3375; %c/mol (' faraday const ');
vc=2.1 ; eta_ele=0.74; %('the electrolysis efficiency= ');
E7=(m1*vc*2*F) /((M*eta_ele))
% ( watt)
E_7energy = (E7/1000)*8760*0.9
% (KWh/year)
%%% RO %%%
rho=1000; V=m5/rho
T=25; %('thermodynamic temperature= ');
eta_pump=0.8; %('efficiency of pump= ');
Z=30*(10^5);
power=(Z*V)/eta_pump
% ( Watt)
RO_Pumpenergy = (power/1000)*8760*0.9
%%% PV %%%
P=10*(10^5); %('pressure of fluid = ');
seawaterpump = (P*V)/eta_pump
seawaterpump_energy = (seawaterpump/1000)*8760*0.9
PV_output=(((P*V)/eta_pump)+((Z*V)/eta_pump)+E7)
eta_pv=0.2; %('pv efficiency = ');
IP=PV_output/eta_pv
PV_energyout=((PV_output)/1000)*8760*0.9
%('KWh .year')
H=3200;
%('KWh / (m^2).year')
TCF = 0.8; eta_out = 0.765;
Area_PV =(PV_energyout)/(H*(eta_pv)*TCF*(eta_out))
467
Simulation code for the second alternative:

"mass balance "
"PEM"
m_1 = (0.03*1083000*0.0899)/3600
M_H2o = 18.015 * (10^(-3))
M_H2 = 2.015 * (10^(-3))
eta_elec = 0.78
(m_2)=((m_1)*(M_H2o))/((M_H2)*(eta_elec))
"MSF"
m_3 = 4 * m_1
m_3vol = (m_3)/1000
m_5 =((m_2)+(m_3))/ 0.7
m_4 = 0.3 *(m_5)
"energy balance"
"PEM"
V_c = 2
F= 96485.3375
(E_8)=((m_1)*(V_c)*2*F)/((M_H2)*(eta_elec))
E_8energy= ((E_8)/1000)*8760*0.9
"Brine Heater"
T_win = 30
T_wout = 90
C_pw = 4.2
h_s = ENTHALPY(Steam,T=100,X=1)
h_L= ENTHALPY(Steam,T=100,X=0)
(m_E)*((h_s)-(h_L))= (m_5)* (C_pw)*((T_wout)-(T_win))
"MSF"
row_w = 1000
P = 10*(10^5)
eta_pump = 0.8
eta_csp = 0.23
T_amb = 25
T_oilin = 240
T_oilout = 300
C_poil =2.5 *1000
G_B = 3200
V=(m_5)/(row_w)
(E_6)=(P*V)/(eta_pump)
E_6energy= ((E_6)/1000)*8760*0.9
(E_7)=(m_E)*(h_s) *1000
E_7energy= ((E_7)/1000)*8760*0.9
(E_out) = (E_6)+(E_7)+(E_8)
E_outenergy = ((E_out)/1000) *8760 *0.9
(E_out) = (m_oil)*(C_poil)*((T_oilout)-(T_oilin))
(E_9) = ((E_6)+(E_7)+(E_8))/(eta_csp)
E_9energy= (E_outenergy)/(eta_csp)
468
Simulation code for the third alternative:

"mass balance "
"PEM"
m_0 = (0.03*1083000*0.0899)/3600
M_H2o = 18.015 *(10^(-3))
M_H2 = 2.015 *(10^(-3))
eta_elec = 0.78
(m_2)=((m_0)*(M_H2o))/((M_H2)*(eta_elec))
"MSF"
m_3 = 4*m_0
m_3vol = (m_3)/1000
(m_5)=((m_2)+(m_3))/ 0.7
m_4 = 0.3 *(m_5)
"energy balance"
"PEM"
V_c = 2
F= 96485.3375
(E_1)=((m_0)*(V_c)*2*F)/((M_H2)*(eta_elec))
E_1energy= ((E_1)/1000)*8760*0.9
"Brine Heater"
T_win = 30
T_wout = 90
C_pw = 4.2
"MSF"
row_w = 1000
P = 10 * (10^5)
eta_pump = 0.8
eta_csp = 0.23
eta_pv=0.2
T_amb = 25
T_oilin =240
T_oilout = 300
C_poil =2.5
G_B = 3200
V=(m_5)/(row_w)
E_8energy= ((E_8)/1000)*8760*0.9
E_pvout = (E_1)+(E_8)
(E_9)=((E_1)+(E_8))/(eta_pv)
E_energyout =((E_pvout)/1000) *8760 *0.9
TCF = 0.8
eta_out = 0.765
Area_PV =(E_energyout )/((G_B)*(eta_pv)*TCF*(eta_out))
(E_6)=(m_E)*(h_s) * 1000
E_6energy = ((E_6)/1000) * 8760 * 0.9
469
Simulation code for the forth alternative:

"mass balance "
"SOEC"
m_0 =(0.03*1083000*0.0899)/3600
M_H2o = 18.015*(10^(-3))
M_H2 = 2.015*(10^(-3))
eta_elec = 0.78
(m_1)=((m_0)*(M_H2o))/((M_H2)*(eta_elec))
"MSF"
m_2 = 4*(m_0)
m_2_volume=(m_2)/1000
m_4=((m_2)+(m_1))/ 0.7
m_3 = 0.3 *(m_4)
"energy balance"
"SOEC"
V_c = 2
F= 96485.3375
E_8 =((m_0)*(V_c)*2*F)/(((M_H2)*(eta_elec)))
E_8energy= ((E_8)/1000)*8760*0.9
h_s/h = ENTHALPY(Steam,T=160,P=1)
h_sat = ENTHALPY(Steam,T=100,P=1)
delta_H = (h_s/h )- ( h_sat)
(m_1)* (C_pw)*((T_wout)-(T_win))= (m_10) * (delta_H)
eta_boiler=0.99
(E_10)=(((m_10)*(delta_H)) *1000)/((eta_boiler))
E_10energy=((E_10)/1000)*8760*0.9
"Brine Heater"
T_win = 30
T_wout = 90
C_pw = 4.2
"MSF"
row_w = 1000
P = 10*(10^5)
eta_pump = 0.8
eta_csp = 0.23
eta_pv=0.2
T_amb = 25
T_oilin = 240
T_oilout = 300
C_poil =2.5
G_B = 3200
V=(m_4)/(row_w)
E_7energy=((E_7)/1000)*8760*0.9
470
Power stations code
471
472
473
474
475
476
477
478
479

Final Book

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Final Book

Uploaded by

Copyright:

Available Formats

Alexandria University

Natural Gas Network Optimization

Under The supervision of:

This page is intentionally left blank

About the team

Natural Gas network

Eslam Mahmoud Mohamed Ibrahim

Samar Samy Anwar Zakarya Matter

Shaza Abd El-Hady Abd El-Kader

Mohamed Abd Elnaby Shams Eldyn

Mohamed Mostafa Ali Mohamed

Mohamed Moustafa Mahmod

Mohamed Mamdouh Mohamed

Mohamed Youssef Omar

Natural gas network

Natural Gas network

This page is intentionally left blank

This page is intentionally left blank

2.5.4 Polymer electrolyte membrane (PEM) electrolysis .................. 100

3.8 Cycle Improvements ............................................................... 151

4.1 Introduction ........................................................................... 182

4.7.2 WDGC compressor station .................................................. 232

6.1.6 Synergi modules .................................................................. 268

7.4.3 Ant Colony Optimization (ACO) ........................................... 393

Conclusion ................................................................................... 455

Chapter 1 | Natural Gas Network Components

flame stability, gas turbine governors, the

In chapter four, compressor stations and

In chapter one, a study of natural gas history,

Synergi Gas was used to model the network,

In chapter two a solution to reduce the

Chapter 1 | Natural Gas Network Components

So in chapter six, Synergi Gas in details is

operate the network at a specific period and

In chapter seven, Optimization of the natural

Chapter 1 | Natural Gas Network Components

Chapter 1 | Natural Gas Network Components

known natural gas well in 211 BC. In Europe,

1.1.2 Natural gas Composition:

1.1.1 Natural gas history

Heavier gaseous hydrocarbons: ethane

Chapter 1 | Natural Gas Network Components

dissolved under high pressure and at high

Mercury: very small amounts of mercury

1.1.2.2 Non-associated Gas

1.1.2.1 Gas sources

1.1.2.3 Associated Gas

Chapter 1 | Natural Gas Network Components

production of associated gas so as to retain

some of the needed processing can be

Oil and Condensate Removal

1.1.3.2 Oil and Condensate Removal

1.1.3 Processing of Natural Gas

Chapter 1 | Natural Gas Network Components

ways for further processing. The most basic

1.1.3.5 Sulfur and Carbon Dioxide Removal

1.1.3.3 Water Removal

1.1.3.4 Separation of Natural Gas Liquids

1.2.1 Transmission pipelines

Chapter 1 | Natural Gas Network Components

purposes. Mainline pipes, the principal

the installation of a typical transmission pipe

Figure 1.1 installation of a typical transmission pipe in a