Power Line
Ampacity
System
Theory, Modeling, and Applications
Anjan K. Deb, Ph.D., P.E.
Electrotech Consultant
Boca Raton London New York Washington, D.C.
CRC Press
This book contains information obtained from authentic and highly regarded sources. Reprinted material
is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable
efforts have been made to publish reliable data and information, but the author and the publisher cannot
assume responsibility for the validity of all materials or for the consequences of their use.
Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying, microﬁlming, and recording, or by any information storage or
retrieval system, without prior permission in writing from the publisher.
The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for
creating new works, or for resale. Speciﬁc permission must be obtained in writing from CRC Press LLC
for such copying.
Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431.
Trademark Notice:
Product or corporate names may be trademarks or registered trademarks, and are
used only for identiﬁcation and explanation, without intent to infringe.
© 2000 by CRC Press LLC
No claim to original U.S. Government works
International Standard Book Number 0849313066
Library of Congress Card Number 00036093
Printed in the United States of America 1 2 3 4 5 6 7 8 9 0
Printed on acidfree paper
Library of Congress CataloginginPublication Data
Deb, Anjan K.
Powerline ampacity system : theory, modeling, and applications / Anjan K. Deb.
p. cm.
Includes bibliographical references and index.
ISBN 0849313066
1. Powerline ampacity—Mathematical models. 2. Electric cables—Evaluation. 3. Electric
lines—Evaluation—Mathematical models. 4. Electric power systems—Load dispatching.
5. Electric currents—Measurement—Mathematics. 6. Amperes. I. Title.
TK3307 .D35 2000
621.319—dc21 00036093
CIP
1306/disclaimer Page 1 Tuesday, May 30, 2000 10:28 AM
Dedication
This book is dedicated to my wife Meeta and
my family and friends
1306/fm/frame Page 10 Saturday, May 27, 2000 4:36 PM
1306/fm/frame Page 4 Tuesday, May 30, 2000 12:46 PM
About the Author
Dr. Anjan K. Deb is a registered professional electrical engineer in the state of
California, and is a principal in ELECTROTECH Consultant, a transmission line
software and consulting company that he started in 1990. He has 20 years’ experience
in highvoltage power transmission lines, substation automation, and electric power
systems. He has authored or coauthored more than 20 research publications in the
area of transmission line conductor thermal ratings, and has been awarded a U.S.
patent for the invention of the LINEAMPS™ program.
Dr. Deb works as a consultant for electric power companies in all regions of the
world, offering seminars and custom software solutions for increasing transmission
line capacity by dynamic thermal ratings. As stated in this book, the LINEAMPS
software developed by the author is used in several countries.
After receiving a bachelor’s degree in electrical engineering from MACT India,
Dr. Deb began his transmission line engineering career at EMC India, where he
worked on the research and development of conductors and line hardware. He went
to Algiers to work for the National Electrical and Electronics Company, where he
designed and manufactured highvoltage substations. While working in Algeria, he
received a French government scholarship to study Electrotechnique at the Conser
vatoire National des Arts et Métiers, Paris, France, where he received the equivalent
of a master’s degree in electrical engineering. He then received training at the
Electricité de France (EDF) Research Center at Paris. EDF is the national electric
power supply company of France. At EDF, Dr. Deb performed theoretical and
experimental research on the heating of conductors and transmission line ampacity.
Dr. Deb came to the U.S. and began working for Paciﬁc Gas & Electric (PG&E),
San Francisco, where he developed and successfully implemented a realtime line
rating system for PG&E. While working at PG&E, he joined a doctoral degree
program at the Columbia Paciﬁc University, and earned a Ph.D. after completing all
courses and preparing a doctoral dissertation on the subject of transmission line
ampacity.
In addition to solving transmission line electrical and mechanical problems, Dr.
Deb is interested in adaptive forecasting, energy management and developing intel
ligent computer applications for power. He is presently working on projects related
to intelligent software development by the application of artiﬁcial intelligence, expert
systems, objectoriented modeling, fuzzy sets, and neural networks. He also main
tains the LINEAMPS website for interaction with program users, and for reporting
new developments. He can be reached by email at akdeb@aol.com, and on the Web
at http://www.lineamps.com.
1306/fm/frame Page 5 Tuesday, May 30, 2000 10:27 AM
1306/fm/frame Page 4 Tuesday, May 30, 2000 12:46 PM
Preface
It is my great pleasure to present a book on transmission line ampacity. While there
are several books devoted to transmission line voltage, there are few books that focus
on line currents, computer modeling of line ampacity with power system applica
tions, and the environmental impact of high currents. A unique contribution of this
book is the development of a complete theory for the calculation of transmission
line ampacity suitable for steady state operation and dynamic and transient condi
tions. To bring this theory into practice I have developed an objectmodel of the line
ampacity system and implemented a declarative style of programming by rules. The
end product is a stateoftheart, userfriendly windows program with a good graph
ical user interface that can be used easily in all geographic regions.
As we enter the 21
st
century we shall have to develop new methods to maximize
the capacity of existing transmission and distribution facilities. The power system
may have to be operated more closely to generation stability limits for better utili
zation of existing facilities. Adding new lines will become more difﬁcult as public
awareness of environmental protection and land use increases.
To keep pace with increasing electric energy usage in the next millennium, new
lines will be required for more efﬁcient electricity transmission and distribution.
Hopefully, with the help of material presented in this book, the transmission line
engineer will make better decisions regarding the choice of conductors, environmen
tal impact, system operation, and cost optimization.
This book is primarily for practicing electric power company engineers and
consultants who are responsible for the planning, operations, design, construction,
and costing of overhead powerlines. It is also a useful source of reference for various
government authorities, electricity regulators, and electric energy policy makers who
want to get a ﬁrm grip on technical issues concerning the movement of electric
energy from one location to another, environmental concerns, and uptodate knowl
edge of existing and future transmission line technologies.
Academicians and students will ﬁnd material covering theoretical concepts of
conductor thermal modeling, the analysis of conductor ampacity, powerline EMF
developed from Maxwell’s equations and Ampere’s law, power ﬂow with variable
line ratings, stability analysis, power electronic devices, and ﬂexible AC transmis
sion. These will complement the existing large number of excellent textbooks on
electric power systems.
This book has been developed from more than 20 years of my experience in
working with various electric power companies in Asia, Africa, Europe, and North
America. I am particularly grateful to Electricité de France, Paris, for the various
interactions with the members of the Departement Etudes et Recherche since 1978,
where I initiated research on the heating of conductors. Paciﬁc Gas & Electric
Company, San Ramon, California, offered me an excellent environment for research
1306/fm/frame Page 3 Saturday, May 27, 2000 4:36 PM
and development when I worked as a consultant on transmission line dynamic
thermal ratings.
Thanks are due to several users of the LINEAMPS program, including
TransPower, New Zealand; Hydro Quebec, Canada; and Korea Electric Power Com
pany for their valuable feedback and support which has enabled me to enhance the
computer program. The kind technical support offered by Mr. Graham of Intellicorp,
California, during the development of the LINEAMPS program is gratefully
acknowledged.
Thanks are due to Dr. Peter Pick, Dean, and Dr. John Heldt, Mentor, of Columbia
Paciﬁc University, California, for their guidance while I prepared a doctoral disser
tation, and for their continued encouragement to write this book. I thank Ms.
Genevieve Gauthier, Research Engineer, Institute de Recherche Hydro Quebec,
Canada, for going through the initial manuscript and kindly pointing out errors and
omissions. Last, but not the least, I am grateful to Professor R. Bonneﬁlle, University
of Paris VI, and Professor J. F. Rialland for their lectures and teachings on Elec
trotechnique at the Conservatoire National des Arts et Métiers, Paris, France.
As in most modern electrical engineering books, SI (System International) units
are used consistently throughout. Complex numbers are denoted by an upperscore,
for example, a complex current I
∠θ
= I·e
j
θ
is represented by , and a vector is
denoted by an upper arrow like .*
The LINEAMPS computer program described in this book is a commercial
software program available from:
ELECTROTECH Consultant
4221 Minstrell Lane
Fairfax, VA 22033, USA
(703) 3228345
For additional details of the program and to obtain new information concerning
recent developments on high currents and transmission line ampacity, readers may
visit LINEAMPS on the Web at http://www.lineamps.com.
Anjan K. Deb, Ph.D.
* I have followed the same notation used by Gayle F. Miner in
Lines and Electromagnetic Fields for
Engineers
, Oxford University Press, New York, 1996.
I
r
H
1306/fm/frame Page 4 Saturday, May 27, 2000 4:36 PM
Contents
Chapter 1
Introduction
........................................................................................ 1
1.1 Organization of Book and Chapter Description .................................. 1
1.2 Introducing the Powerline Ampacity System ...................................... 3
1.3 Electric Power System Overview......................................................... 3
1.3.1 Transmission Grid ................................................................. 3
1.3.2 Overhead Transmission Line................................................. 4
1.3.3 HighVoltage Substation........................................................ 8
1.3.4 Energy Control Center........................................................... 9
1.4 Factors Affecting Transmission Capacity and
Remedial Measures............................................................................. 11
1.5 New Developments For Transmission Capacity Enhancement ......... 12
1.6 Dynamic Line Rating Cost–Beneﬁt Analysis .................................... 12
1.7 Chapter Summary............................................................................... 12
Chapter 2
Line Rating Methods
....................................................................... 15
2.1 Historical Backround.......................................................................... 15
2.1.1 Early Works on Conductor Thermal Rating ....................... 15
2.1.2 IEEE and Cigré Standards................................................... 15
2.1.3 Utility Practice..................................................................... 15
2.2 Line Rating Methods .......................................................................... 16
2.2.1 Deﬁning the Line Ampacity Problem................................. 16
2.2.2 Static and Dynamic Line Ratings ....................................... 17
2.2.3 WeatherDependent Systems ............................................... 18
2.2.4 Online Temperature Monitoring System............................. 19
2.2.5 Online Tension Monitoring System.................................... 21
2.2.6 SagMonitoring System....................................................... 22
2.2.7 Distributed Temperature Sensor System............................. 23
2.2.8 ObjectOriented Modeling and Expert Line
Rating System...................................................................... 25
2.3 Chapter Summary............................................................................... 26
Chapter 3
Theory of Transmission Line Ampacity
........................................ 27
3.1 Introduction......................................................................................... 27
3.2 Conductor Thermal Modeling............................................................ 28
3.2.1 General Heat Equation ........................................................ 28
3.2.2 Differential Equation of Conductor Temperature ............... 29
3.2.3 SteadyState Ampacity ........................................................ 29
3.2.4 Dynamic Ampacity.............................................................. 36
3.2.5 Transient Ampacity.............................................................. 44
1306/fm/frame Page 7 Tuesday, May 30, 2000 10:27 AM
3.2.6 Radial Conductor Temperature............................................ 47
3.3 Chapter Summary............................................................................... 49
Appendix 3 AC Resistance of ACSR.......................................................... 51
Chapter 4
Experimental Veriﬁcation of Transmission Line Ampacity
......... 61
4.1 Introduction......................................................................................... 61
4.2 Wind Tunnel Experiments.................................................................. 61
4.3 Experiment in Outdoor Test Span...................................................... 63
4.4 Comparison of LINEAMPS with IEEE and Cigré............................ 66
4.4.1 SteadyState Ampacity ........................................................ 66
4.4.2 Dynamic Ampacity.............................................................. 71
4.5 Measurement of Transmission Line Conductor Temperature ........... 71
4.6 Chapter Summary............................................................................... 72
Chapter 5
Elevated Temperature Effects
......................................................... 73
5.1 Introduction......................................................................................... 73
5.1.1 Existing Programs................................................................ 74
5.2 Transmission Line Sag and Tension — A
Probabilistic Approach ....................................................................... 74
5.2.1 The Transmission Line Sag–Tension Problem................... 75
5.2.2 Methodology........................................................................ 75
5.2.3 Computer Programs............................................................. 77
5.3 Change of State Equation................................................................... 78
5.3.1 Results.................................................................................. 79
5.3.2 Conductor Stress/Strain Relationship.................................. 80
5.4 Permanent Elongation of Conductor.................................................. 80
5.4.1 Geometric Settlement .......................................................... 81
5.4.2 Metallurgical Creep ............................................................. 81
5.4.3 Recursive Estimation of Permanent Elongation ................. 82
5.5 Loss of Strength.................................................................................. 83
5.5.1 Percentile Method................................................................ 83
5.5.2 Recursive Estimation of Loss of Strength .......................... 84
5.6 Chapter Summary............................................................................... 84
Appendix 5 Sag and Tension Calculations .............................................. 87
Chapter 6
Transmission Line Electric and Magnetic Fields
......................... 93
6.1 Introduction......................................................................................... 93
6.2 Transmission Line Magnetic Field..................................................... 93
6.2.1 The Magnetic Field of a Conductor.................................... 94
6.2.2 The Magnetic Field of a ThreePhase Powerline ............... 98
6.2.3 The Magnetic Field of Different Transmission
Line Geometry................................................................... 100
6.2.4 EMF Mitigation ................................................................. 102
6.3 Transmission Line Electric Field ..................................................... 108
6.4 Chapter Summary............................................................................. 113
1306/fm/frame Page 8 Tuesday, May 30, 2000 10:27 AM
Chapter 7
Weather Modeling for Forecasting Transmission
Line Ampacity
............................................................................................. 115
7.1 Introduction....................................................................................... 115
7.2 Fourier Series Model ........................................................................ 116
7.3 RealTime Forecasting...................................................................... 123
7.4 Artiﬁcial Neural Network Model ..................................................... 127
7.5 Modeling by Fuzzy Sets................................................................... 132
7.6 Solar Radiation Model...................................................................... 137
7.7 Chapter Summary............................................................................. 139
Chapter 8
Computer Modeling
....................................................................... 143
8.1 Introduction ..................................................................................... 143
8.1.1 From Theory to Practice.................................................... 143
8.1.2 The LINEAMPS Expert System....................................... 143
8.2 Object Model of Transmission Line Ampacity System................... 144
8.2.1 LINEAMPS Object Model ................................................ 144
8.2.2 Transmission Line Object.................................................. 145
8.2.3 Weather Station Object...................................................... 148
8.2.4 Conductor Object............................................................... 150
8.2.5 Cartograph Object.............................................................. 152
8.3 Expert System Design ...................................................................... 154
8.3.1 GoalOriented Programming ............................................. 155
8.3.2 Expert System Rules ......................................................... 157
8.4 Program Description......................................................................... 159
8.4.1 LINEAMPS Windows ....................................................... 159
8.4.2 Modeling Transmission Line and Environment................ 159
8.4.3 LINEAMPS Control Panel ................................................ 159
8.5 Chapter Summary............................................................................. 162
Chapter 9
New Methods of Increasing Transmission Capacity
.................. 163
9.1 Introduction....................................................................................... 163
9.2 Advancement in Power Semiconductor Devices ............................. 163
9.3 Flexible AC Transmission ................................................................ 168
9.4 Chapter Summary............................................................................. 181
Chapter 10
Applications
................................................................................... 183
10.1 Introduction....................................................................................... 183
10.2 Economic Operation......................................................................... 183
10.2.1 Formulation of the Optimization Problem........................ 184
10.2.2 Electricity Generation Cost Saving in Interconnected
Transmission Network....................................................... 186
10.3 Stability............................................................................................. 190
10.3.1 Dynamic Stability.............................................................. 191
10.3.2 Transient Stability.............................................................. 193
10.4 Transmission Planning...................................................................... 195
10.5 LongDistance Transmission ............................................................ 198
1306/fm/frame Page 9 Tuesday, May 30, 2000 10:27 AM
10.6 Protection.......................................................................................... 201
10.7 Chapter Summary............................................................................. 204
Appendix 10.1 Transmission Line Equations .................................................... 205
Chapter 11
Summary, Future Plans, and Conclusion
.................................. 209
11.1 Summary........................................................................................... 209
11.2 Main Contributions........................................................................... 212
11.3 Suggestions for Future Work............................................................ 219
11.4 A Plan to Develop LINEAMPS for America .................................. 225
11.5 Conclusion ........................................................................................ 226
Bibliography
......................................................................................................... 229
Appendices A1–A8: Conductor Data
................................................................. 235
Appendix B: Wire Properties
............................................................................. 243
Index
...................................................................................................................... 245
1306/fm/frame Page 10 Tuesday, May 30, 2000 10:27 AM
1
1
Introduction
1.1 ORGANIZATION OF BOOK AND
CHAPTER DESCRIPTION
Chapter 1 gives a broad overview of the electric power system including transmission
lines, substations, and energy control centers. Data for electricity production in the
U.S. and the world are also given.
Chapter 2 presents the different methods of transmission line rating, including
both online and offline methods.
A complete theory of transmission line ampacity is presented in Chapter 3. A
threedimensional conductor thermal model is ﬁrst developed, and then solutions
are presented for steadystate, dynamic, and transient operating conditions.
Experimental work related to transmission line ampacity that was conducted in
different research laboratories is described in Chapter 4. The conductor thermal
models in the steadystate and dynamic and transient states are validated by com
paring results with the IEEE standard and Cigré method. Results are also compared
to laboratory experiments and measurements from actual transmission lines.
The effects of elevated temperature operation on transmission line conductors
are presented in Chapter 5. Experimentally derived models of loss of tensile strength
of conductors, as well as permanent elongation of conductors due to creep, are
presented in this chapter. The method of calculation of the loss of strength and
inelastic elongation of conductors by a recursive procedure that utilizes probability
distribution of conductor temperature in service is described. A method of generating
the probability distribution of conductor temperature in service from time series
stochastic and deterministic models is given.
The theory of transmission line electric and magnetic ﬁelds is developed from
Maxwell's equations in Chapter 6. When higher ampacity is allowed on the line, it
increases the magnetic ﬁeld radiated from the transmission line. The electric ﬁeld
from the transmission line does not change with line ampacity, but increases with
conductor temperature due to lowering of the conductor to ground clearance by sag.
Methods of reducing the level of EMF radiated from transmission lines by active
and passive shielding are presented in this chapter. This aspect of transmission line
ampacity is signiﬁcant because there is little previous work carried out in this
direction. Even though there is no evidence of environmental impact by EMF due
to increased transmission line currents, measures are suggested to lower magnetic
ﬁelds from transmission lines.
Environmental factors inﬂuence transmission capacity signiﬁcantly. For this
reason Chapter 7 is devoted entirely to weather modeling. The meteorological
variables that are most important to powerline capacity are ambient temperature,
wind speed, wind direction, and solar radiation. Statistical modeling of weather
1306/C01/frame Page 1 Tuesday, May 30, 2000 10:43 AM
2
Powerline Ampacity System: Theory, Modeling, and Applications
variables based on time series analysis, Fourier series analysis, and neural networks
are presented with examples using real data collected from the National Weather
Service. Models are developed for realtime prediction of weather variables from
measurements as well as by weather pattern recognition. Analytical expressions for
the calculation of solar radiation on a transmission line conductor are also presented
to complete the chapter on weather modeling.
Chapter 8 describes computer modeling of the LINEAMPS expert system. The
complete system of rating overhead powerlines is implemented in a computer
program called LINEAMPS. This stateoftheart software package is an expert
system for the rating of powerlines. It was developed by objectoriented modeling
and expert rules of powerline ampacity. The object of the program is to maximize
the currentcarrying capacity, or the ampacity, of existing and future overhead
powerlines as functions of present and forecast weather conditions. Methods of
objectoriented modeling of transmission lines, weather stations, and powerline
conductors are described with examples from electric power companies in the
different regions of the world. Expert system rules are developed to enable an
intelligent powerline ampacity system to check user input and explain error mes
sages like a true expert.
Chapter 9 discusses new methods of increasing line ampacity. The capacity
of electric powerlines to transport electric energy from one point to another, that
depends upon several factors, is discussed. The most important factors are trans
mission distance, voltage level, and generator stability. In many cases, adequate
stability can be maintained by electrical control of generation systems as well as
by fast control of active and reactive power supply to the system. When energy
is transported over a long distance, there may be signiﬁcant voltage drop that may
be compensated for by controlling reactive power and/or boosting voltage levels
by transformer action. Therefore, in most cases the transport capacity of overhead
powerlines is limited only by the thermal rating of the powerline conductor. An
overview of new technologies that are being developed to increase transmission
capacity up to the thermal limit by overcoming the aforementioned limitations is
presented in this chapter. These new technologies include the application of mod
ern power electronics devices that are known as FACTS (Flexible AC Transmission
System), Superconducting Magnetic Energy Storage (SMES), and distributed gen
eration systems.
Chapter 10 presents applications of the new powerline ampacity system to clearly
show its beneﬁts. In a competitive power supply business environment, it is necessary
to optimize the ampacity of overhead power transmission lines to enable the most
economic power system operation on an hourbyhour basis. Until recently, electric
power companies* have assumed that the maximum capacity of a powerline is
constant by assuming conservative weather conditions, so they followed a static line
rating system. Now certain electric power companies** are adopting a system of
line rating that is variable and dynamic depending upon actual weather conditions.
*
Regles de calcul electrique
. EDF/CERT
Directives Lignes Aeriennes
1996.
Ampacity of overhead line conductors. PG&E Engineering Standard.
** REE Spain (Soto et al., Cigre, 1998); KEPCO, South Korea (Wook et al., 1997)
1306/C01/frame Page 2 Tuesday, May 30, 2000 10:43 AM
Introduction
3
The thermal rating of a transmission line depends upon the maximum design tem
perature of the line, and the temperature of a conductor varies as a function of line
current and meteorological conditions. Therefore, for the same value of maximum
conductor temperature, higher line currents are possible if there are favorable mete
orological conditions. In this chapter, a system of equations for the economic oper
ation of diverse generation sources in an interconnected power system is developed
that utilizes a dynamic line rating system. The economic beneﬁts of a dynamic line
rating system are demonstrated by giving an example of an interconnected trans
mission network having a diverse mix of electricity generation sources. The chapter
concludes with a discussion of increased competition in the electric power supply
industry in a power pool system of operations, and the important role of the powerline
ampacity system presented in this book.
Chapter 11 gives a summary of main contributions made in this book, presents
future plans and new transmission and distribution technologies, describes the role of
Independent System Operators (ISO) and powerpool operations from the point of
view of transmission line capacity. It provides a discussion on deregulation and how
the line ampacity system facilitates greater competition in the electric supply business.
1.2 INTRODUCING THE POWERLINE
AMPACITY SYSTEM
As the demand for electricity increases, there is a need to increase electricity
generation, transmission, and distribution capacities to match demand. While the
location and construction of a generation facility is relatively easy, it is becoming
increasingly difﬁcult to construct new lines. As a result, electric power authorities
everywhere are searching for new ways to maximize the capacity of powerlines.
One of the methods used to increase line capacity is dynamic thermal rating.
The object of this book is to develop a complete system of rating overhead
powerlines by presenting theory, algorithms, and a methodology for implementation
in a computer program. The development of a computer program by objectoriented
modeling and expert system rules is also described in detail. The end product is easy
to use and suitable for application in all geographic regions. The different methods
of increasing line ampacity by FACTS are described, and the impact of higher
transmission line ampacity on electric and magnetic ﬁelds is analyzed with numerical
examples. Application of the powerline ampacity system in the economic operation
of a power system is presented, and considerable cost savings are shown by the
deferment of capital investment required for the construction of new lines, and by
enabling greater utilization of lowcost energy sources.
1.3 ELECTRIC POWER SYSTEM OVERVIEW
1.3.1 T
RANSMISSION
G
RID
The electric power system is comprised of an interconnected transmission grid that
is used to connect diverse generation sources for the distribution of electricity to
load centers in the most economical manner. Figure 1.1 shows the transmission grid
1306/C01/frame Page 3 Tuesday, May 30, 2000 10:43 AM
4
Powerline Ampacity System: Theory, Modeling, and Applications
of North America with transmission lines having voltages 230 kV and above. The
total length of transmission lines at each voltage category is shown in the Figure
1.2. The installed generation capacity is approximately 750 GW, which is expected
to grow at the rate of 2% per year. Assuming an annual load factor of 0.5, approx
imately 3200 billion
kilowatthours will be distributed through the transmission
network in the year 2000. Figures 1.3–1.5 show electricity production in the U.S.
and the world.
According to the U.S. Department of Energy,* about 10,000 circuit km of
transmission lines are planned to be added by the year 2004. The total cost of adding
new transmission lines is approximately three billion dollars. In addition to the high
cost of adding new transmission lines, environmental factors related to land use and
EMF are also required to be considered before the construction of new lines.
Dynamic rating of transmission lines offers substantial cost savings by increasing
the capacity of existing lines such that the construction of new lines may be post
poned in many cases.
1.3.2 O
VERHEAD
T
RANSMISSION
L
INE
The overhead transmission line consists of towers, conductors, insulators, and line
hardware for the jointing of conductors and for properly supporting the highvoltage
line to the transmission line tower. The most common type of transmission line
FIGURE 1.1
North American Transmission grid.
* Arthur H. Fuldner,
Upgrading Transmission Capacity for Wholesale Electric Power Trade
, U.S.
Department of Energy publication on the World Wide Web, December 30, 1998.
Pacific Ocean
Washington
Oregon
California
Alberta
British
Columbia
Manitoba
Sas
MT
ID
NV
UT
CO
AZ
NM
OK
TX
Mexico
KS
MO
NE
IA
IL
IN
SD
OH
MI
MN
ND
Hudson Bay
Ontario
WI
KY
SC
NC
MS
AR
LA
AL
GA
Florida
Atlantic Ocean
New York
NJ
DE
MD
ME
MA
Quebec
VT
NH
RA
VA
WV
TN
WY
Transmission Lines
Cluster of Power
Generation Stations
230kV/345kV/500kV
735kV/765kV
DC Transmission
1306/C01/frame Page 4 Tuesday, May 30, 2000 10:43 AM
Introduction
5
towers are selfsupporting towers, and guyed and pole towers. Some typical examples
of towers are shown in Figures 1.6–1.9.
Aluminum Conductor Steel Reinforced (ACSR) is the most widely used type of
currentcarrying conductor. All Aluminum Conductors (AAC) are used in coastal
regions for high corrosion resistance and also for applications requiring lower resis
tance, where the high strength of a steel core is not required. More recently, All
Aluminum Alloy conductors have been used for their light weight and high strength
toweight ratio, which enables longer spans with less sag. Other hybrid conductors
having various proportions of aluminum, aluminum alloy, and steel wires are also used
for special applications. The popular type of hybrid conductors are Aluminum Con
ductor Alloy Reinforced (ACAR) and Aluminum Alloy Conductor Steel Reinforced
(AACSR). Some examples of commonly used powerline conductors according to
various standards are given in Appendix A (Thrash, 1999; Koch, 1999; Hitachi, 1999).
FIGURE 1.2
Total transmission line circuit km in North America according to transmission
voltage category.
FIGURE 1.3
U.S. electric utility generation capacity.
Transmission Voltage
500kV 735kV 345kV 230kV DC
70000
60000
50000
40000
30000
20000
10000
0
K
i
l
o
m
e
t
e
r
Hydroelectric
US Electric Power Generation
(13%)
Nuclear
(14%)
Gas
(20%)
Petrolium
(10%)
Coal
(13%)
E
n
e
r
g
y
S
o
u
r
s
e
0 100 200 300 400
Generation Capacity,GW
1306/C01/frame Page 5 Tuesday, May 30, 2000 10:43 AM
6
Powerline Ampacity System: Theory, Modeling, and Applications
FIGURE 1.4
U.S. utility electric energy production.
FIGURE 1.5
World electric energy production.
FIGURE 1.6
Self supporting tower.
Hydroelectric
Nuclear
Gas
Petrolium
Coal
US Electric Energy Generation
(11%)
(20%)
(9%)
(2%)
(57%)
E
n
e
r
g
y
S
o
u
r
s
e
0 500 1000 1500 2000
Net Generation (TWh)
Hydroelectric
Nuclear
Gas
Petrolium
Coal
World Electric Energy Production
(11%)
(20%)
(9%)
(2%)
(57%)
E
n
e
r
g
y
S
o
u
r
s
e
0 2000 4000 6000 8000
Net Generation (TWh)
1306/C01/frame Page 6 Tuesday, May 30, 2000 10:43 AM
Introduction
7
FIGURE 1.7
Guyed tower.
FIGURE 1.8
Ornamental tower.
FIGURE 1.9
Tubular tower.
1306/C01/frame Page 7 Tuesday, May 30, 2000 10:43 AM
8
Powerline Ampacity System: Theory, Modeling, and Applications
Hightemperature conductors are used for bulk power transmission in heavily
loaded circuits where a high degree of reliability is required. Steel Supported Aluminum
Conductor (SSAC) allows hightemperature operation with minimum sag. In the SSAC
conductor, the currentcarrying aluminum wires are in the annealed state and do not
bear any tension. The tension is borne entirely by the highstrength steel wires. In the
newer hightemperature, highampacity conductors, aluminum zirconium alloy wires
are used to carry high current, and Invar alloy reinforced steel wires are used for the core.
Recently, compact conductor designs have been available that offer lower losses
for the same crosssectional area of the conductor. Compact design is made possible
by the trapezoidal shaping of wires instead of wires having the circular crosssections
used in conventional ACSR conductors. For better aerodynamic performance, con
ductors are also available with concentric gaps inside the conductor which offer
better damping of windinduced vibrations.
Another recent development in transmission line conductor technology is the
integration of optical ﬁber communication technology in the manufacture of pow
erline conductors. In an Optical Ground Wire (OPGW) system, a ﬁberoptic cable is
placed inside the core of the overhead ground wire. In certain transmission line
applications, the ﬁberoptic cable is placed inside the core of the power conductor.
Communication by ﬁber optics offers a noisefree system of data communication in
the electric utility environment since communication by optical ﬁber is unaffected
by electromagnetic disturbances. The different types of conductors are shown in
Figure 1.10. Important physical properties of the different types of wires used in the
manufacture of powerline conductors are given in Appendix B.
1.3.3 H
IGH
V
OLTAGE
S
UBSTATION
The electric substation is an important component of the electric power system. The
substation is a hub for receiving electricity from where electricity is distributed to
load centers, as well as to other substations. Voltage transformation is carried out
in the substation by transformers. A transmission substation is generally used for
interconnection with other substations where power can be rerouted by switching
action. In a distribution substation, electricity is received by highvoltage transmis
sion lines and transformed for distribution at lower voltages. Besides transformers,
there are other important devices in a substation, including bus bars, circuit breakers,
interrupters, isolators, wave traps, instrument transformers for the measurement of
high voltage and current, inductive and capacitive reactors for the control of reactive
power ﬂow, protective relays, metering, control and communication equipment, and
other lowvoltage equipment for station auxiliary power supply.
A typical layout of a highvoltage substation is shown in Figure 1.11. When a
dynamic line rating system is implemented in an electric utility system, it is also
important to have knowledge of the current rating of all substation equipment in
addition to powerline conductor ratings. Substation switching devices are generally
designed to withstand short circuit currents, and have sufﬁcient continuous overload
current capability. Transformer ratings, on the other hand, need to be examined more
closely. A system of dynamic rating of substation equipment may be implemented
1306/C01/frame Page 8 Tuesday, May 30, 2000 10:43 AM
Introduction
9
by realtime monitoring of equipment temperature by installing sensors, or by
inferring equipment temperature by the measurement of current ﬂowing through the
device and monitoring weather conditions at the location of the substation.
1.3.4 E
NERGY
C
ONTROL
C
ENTER
The electric power system comprising generation stations, transmission and distri
bution lines, and substations is controlled by a system of energy control centers.
Each electric power company operates its electricity supply system in a given
geographic region through one or more energy control centers, as shown in Figure
1.13. For example, Electricité de France (EDF), the national electric power supply
company of France, operates its electric power supply system through one central
control center and seven regional control centers. Control centers are responsible
for the control of power generation, load forecasting, performing load ﬂow, dynamic
and transient stability analysis, contingency analysis, and switching operations in
the substations. Control centers constantly monitor the condition of all transmission
lines and substations in their respective regions, and, in the event of a failure of a
component in the network, control actions are taken to remedy the problem.
Modern control centers are operated through a network of computers having
intelligent programs called “expert systems.” These expert systems perform a variety
of tasks from energy management to alarm processing and fault diagnosis, providing
assistance to control system operators for better decision making, which is especially
FIGURE 1.10
Transmission line conductors.
7 wire AAC 37 wire AAC 37 wire AAAC
54 EC 7 St ACSR 30 EC 7 st ACSR 19 Alloy 42 ES ACAR
42 Es 19 AS ACSR/AS 54 ES AS Compact
1306/C01/frame Page 9 Tuesday, May 30, 2000 10:43 AM
10
Powerline Ampacity System: Theory, Modeling, and Applications
useful during an emergency. The powerline ampacity system described in this book
is an expert system for the evaluation of transmission line ampacity, which is
expected to be an integral part of a modern energy control system.
FIGURE 1.11
High voltage substation.
FIGURE 1.12
Typical power system hierarchy.
T/L Earth
Switch
Main Bus
Bus Couper
Bus Sectioalizer
Transfer Bus
O
500 kV
High Voltage Transmission Line
Lokal Genereation
Distribution Feeders
33 kV
3 Winding Transformer
Symbols
Circuit Breaker
Isolator
230 kV
General Control Center
Regional Control
Center
Regional Control
Center
Substation Substation Substation Substation
1306/C01/frame Page 10 Tuesday, May 30, 2000 10:43 AM
Introduction
11
1.4 FACTORS AFFECTING TRANSMISSION CAPACITY
AND REMEDIAL MEASURES
The effects of elevated temperature operation are loss of tensile of conductors and
permanent elongation of conductors. The loss of strength model is given by Harvey
(Harvey, 1972) and (Morgan, 1978). The models for permanent elongation of con
ductor is given in a Cigré 1978 report. A recursive estimation algorithm for calcu
lating the loss of tensile strength and permanent elongation due to heating in service
from the probability distribution of conductor temperature is described by the author
(Deb et al., 1985). A study for the assessment of thermal deterioration of transmission
line conductor from conductor temperature distribution was presented recently by
Mizuno et al. (1998). The results were presented by the author (Deb, 1993) for
practical line operating conditions.
The remedial measures that are proposed to reduce the possibility of transmission
line conductor overheating comprise the use of line ampacity programs and the
monitoring of transmission line current and/or temperature.
Special conductors may be used to transfer higher currents in highly congested
transmission circuits. A recent study conducted by the author and KEPCO* (Wook,
Choi, and Deb, 1997) shows that transmission capacity may be doubled by the
application of new types of powerline conductors. The new types of conductors
are capable of operating at signiﬁcantly higher temperatures with less sag and
without any thermal deterioration. There is general agreement that transmission
line magnetic ﬁelds** have minimum impact on the environment, and there are
no harmful effects of magnetic ﬁelds on human beings. A recent research study
conducted by EPRI (Rashkes and Lordan, 1998) presents new transmission line
design considerations to lower magnetic ﬁelds. This study is important from the
FIGURE 1.13
Energy Control Center.
* Author worked as a consultant for Korea Electric Power Company (KEPCO), South Korea.
** EMF Conference. National Academy of Science, U.S.A. 1994, concluded that there are no harmful
effects due to powerline electric and magnetic ﬁelds.
1306/C01/frame Page 11 Tuesday, May 30, 2000 10:43 AM
12
Powerline Ampacity System: Theory, Modeling, and Applications
point of view of transmission line ampacity so that future transmission lines can
be constructed with higher power transfer capability and minimum magnetic ﬁeld.
1.5 NEW DEVELOPMENTS FOR TRANSMISSION
CAPACITY ENHANCEMENT
There are other electrical network constraints that must be satisﬁed before transmis
sion lines can be operated at their maximum thermal capacities. The most important
constraints are voltage levels and generator stability limits. New methods and devices
to improve transmission system voltage levels and generation stability limits include
FACTS (Flexible AC Transmission System) (Hingorani, 1995).
FACTS technology makes use of recent developments in modern power elec
tronics and superconductivity (Feak, 1997) to enhance transmission capacity. A
recent FACTS development is the Uniﬁed Power Flow Controller (UPFC) (Norozian
et al., 1997). Another important development is the invention of a new type of
generator called the “Powerformer” (
MPS Review
, 1998b) that eliminates the need
for a transformer by generating electricity at high voltage at the level of transmission
system voltage. The new type of generator produces greater reactive power to elevate
grid voltage levels, and also enhances generation stability when required. Therefore,
by connecting Powerformer directly to the transmission grid, yet higher levels of
transmission capacity may be achieved. These studies show that there is considerable
interest in maximizing the capacity of existing assets. By the introduction of these
new technologies in the electrical power system, it is now becoming possible to
operate transmission lines close to thermal ratings, when required.
1.6 DYNAMIC LINE RATING COSTBENEFIT ANALYSIS
A cost beneﬁt analysis was carried out by the cost capitalization method and the
results are presented in Table 1.1. It is assumed that line current will increase at the
rate of 2.5% per year. The results show that the capitalized cost of higher losses due
to the increase in line current by deferment of new line construction for a period of
10 years is signiﬁcantly lower than the cost of constructing a new line in the San
Francisco Bay area.
In addition to cost savings achieved by postponing the construction of new lines,
dynamic line rating systems also offer substantial operational cost savings. In Chap
ter 10, a study is presented which show 16% economy achieved by dynamic line
rating by facilitating the transfer of lowcost surplus hydroelectric energy through
overhead lines. To undertake this study, an economic load ﬂow program was devel
oped to simulate an interconnected transmission network with diverse generation
sources (Hall and Deb, 1988a; Deb, 1994; Yalcinov and Short, 1998).
1.7 CHAPTER SUMMARY
An introduction to the subject of transmission line ampacity is presented in this
chapter by giving an overview of the electric power system. The signiﬁcance of the
study and the main contributions in each chapter are summarized.
1306/C01/frame Page 12 Tuesday, May 30, 2000 10:43 AM
Introduction
13
As the demand for electricity grows, new methods and systems are required to
maximize the utilization of existing power system assets. Highvoltage transmission
lines are critical components of the electric power system. Due to environmental,
regulatory, and economical reasons it is not always possible to construct new lines,
and new methods are required to maximize their utilization. The object of this book
is to present a study of transmission line conductor thermal modeling, to develop a
methodology for the rating of transmission lines for implementation in a computer
program that is suitable for all geographic regions, and to present the applications
of line ampacity in the operation of electric power systems.
The development of a complete line ampacity system having transmission line,
weather, and conductor models that can be easily implemented in all geographic
regions was a major challenge. For this reason it was necessary to develop a computer
program that will adapt to different line operating standards followed by power
companies in the different regions of the world. This was accomplished by devel
oping an expert system and objectoriented modeling of the line ampacity system.
TABLE 1.1
Cost–Beneﬁt Analysis
Year
Line Current,
Increase @
2.5%/yr,
A
Annual Energy
Loss Increase, @
10c/kWh,
$
Present Value of
Annual Loss, @
10%/yr Interest,
$
1995 800 0 0
1996 820 2,759 2,508
1997 841 5,657 4,675
1998 862 8,702 6,538
1999 883 11,902 8,129
2000 905 15,263 9,477
2001 928 18,794 10,609
2002 951 22,505 11,548
2003 975 26,403 12,317
2004 1000 30,498 12,934
Total present value of
loss, $/mile
78,736
Cost of new line,
$/circuit/mile
200,000
Saving by LINEAMPS,
$/circuit/mile
121,264
Note
: The following assumptions are made in the above calculations:
Line load loss factor = 0.3, based on system load factor = 0.5.
Conductor is ACSR Cardinal, 1 conductor/phase, single circuit line.
Rate of interest is 10% /year.
Line load increase by 2.5% /year.
Static normal ampacity of the line is 800 A.
1306/C01/frame Page 13 Tuesday, May 30, 2000 10:43 AM
14
Powerline Ampacity System: Theory, Modeling, and Applications
The economic incentives for implementing a dynamic line rating system are
clearly established by showing the approximately 60% cost saving by the deferment
of new line construction. The factors limiting line capacity are clearly brought out,
and the means to overcome these are explained.
1306/C01/frame Page 14 Tuesday, May 30, 2000 10:43 AM
15
2
Line Rating Methods
2.1 HISTORICAL BACKROUND
2.1.1 E
ARLY
W
ORKS
ON
C
ONDUCTOR
T
HERMAL
R
ATING
Faraday was one of the early researchers who conducted theoretical and experimental
research to study the heating of wires by electric current (Faraday, 1834).* Some
early works on transmission line conductor thermal rating were conducted in France
(Legrand, 1945) that realized the importance of transmission line conductor thermal
ratings. A transmission line rating system using temperature monitoring by a thermal
image of conductors was developed in Belgium (Renchon, 1956).
A steadystate ampacity model based on the conductor heat balance equation
was presented in 1956 (House and Tuttle, 1956). For the shortterm rating of trans
mission line conductors, Davidson (1969) presented a solution to the differential
equation of conductor temperature by using the Eulers method. All of the above
research shows that there has long been considerable interest in maximizing the
transmission capacity of overhead lines. Weather modeling for transmission line
ampacity was ﬁrst presented by the author at the Cigré Symposium** on High
Currents (Deb et al., 1985).
2.1.2 IEEE
AND
C
IGRÉ
S
TANDARDS
IEEE (IEEE Standard 738, 1993) and Cigré (Cigré, 1992, 1997, 1999) offer standard
methods for the calculation of transmission line ampacity in the steady, dynamic,
and transient states. The Cigré report presents a threedimensional thermal model
of conductors for unsteadystate calculation. A similar model was presented at the
IEEE (Hall, Savoullis, and Deb, 1988) for the calculation of thermal gradient of
conductor from surface to core.
2.1.3 U
TILITY
P
RACTICE
Electric power companies*** generally assume that the ampacity of transmission
line conductors is constant. Ampacity calculations are commonly based upon the
following conservative assumptions of ambient temperature, wind speed, solar radi
ation, and maximum conductor temperature:
* Michael Faraday, Electricity, Encyclopaedia Britannica, Great Books # 42, page 686.
** Cigré Symposium: High Currents in Power Systems under Normal, Emergency and Fault Conditions,
Brussels, Belgium, 3–5 June, 1985, devoted to the subject of transmission line ampacity.
*** Electricité de France, Paris, is the national electric power company of France (Urbain, 1998); Paciﬁc
Gas & Electric Co. San Francisco, CA, (PG&E Standard 1978); Central Board of Irrigation and Power,
India (Deb et al., 1985).
1306/C02/frame Page 15 Saturday, May 27, 2000 9:53 AM
16
Powerline Ampacity System: Theory, Modeling, and Applications
• Ambient temperature = 40°C
• Wind speed = 0.61 m/s (2 ft/s)
• Solar radiation =1000 W/m2
• Maximum conductor temperature = 80°C
It is well known that weather conditions are never constant. Therefore, during favorable
weather conditions when ambient temperature is lower than the assumed maximum
or wind speed is higher than the assumed minimum, or during cloudy sky conditions,
higher ampacity is possible without exceeding the allowable maximum temperature
of the powerline conductor. For the above reasons, many utilities have started adapting
line ratings to actual weather conditions to increase line capacity. A dynamic line rating
system can provide further increase in line ampacity for short durations by taking into
consideration the heatstorage capacity of conductors.
2.2 LINE RATING METHODS
2.2.1 D
EFINING
THE
L
INE
A
MPACITY
P
ROBLEM
The problem of determining the thermal rating of an overhead powerline can be
stated as follows: based on existing and forecast weather conditions at several
locations along the transmission line route, determine the maximum current that can
be passed through the line at a given time (t) such that the conductor temperature
(Tc) at any section of the line does not exceed the design maximum temperature
(Tmax) of the line.
Stated formally,
∀
l
I
l,t
= min(I
l,j,t
) (2.1)
I
l,j,t
= f(Ws
k,j,t
, Wd
k,j,t
, Ta
k,j,t
, Sr
k,j,t
, Tc
l,j
, C
l,j
, D
l,j
) (2.2)
Tc
l,j
≤
Tmax
l,j
(2.3)
Where,
I = Ampacity (Ampere)
Ws = Wind speed
Wd = Wind direction
Ta = Ambient temperature
Sr = Solar radiation
Tc = Conductor temperature
C = Conductor
D = Direction of line
l = 1,2,3… L transmission lines
j = 1,2,3… J line sections
t = 1,2,3… T time (T = 168 h in LINEAMPS)
k = 1,2,3… K weather stations
1306/C02/frame Page 16 Saturday, May 27, 2000 9:53 AM
Line Rating Methods
17
The LINEAMPS computer program described in this book ﬁnds a solution to the
above line ampacity problem.
2.2.2 S
TATIC
AND
D
YNAMIC
L
INE
R
ATINGS
Transmission line rating methods are broadly classiﬁed into two categories: static
and dynamic line rating. The static line rating system is widely used because of its
simplicity, as it does not require monitoring weather conditions or installation of
sensors on the transmission line conductor. The static rating of transmission lines
in a region is generally determined by analysis of historical weather data of that
region for the different types of conductors used in the transmission lines. Generally,
static line ratings are ﬁxed for a particular season of the year, and many electric
power utilities have different line ratings for summer and winter. For example, the
static line rating of some typical conductor sizes used by PG&E in the region of the
San Francisco Bay area is given in Table 2.1.
Dynamic line ratings are obtained by online or ofﬂine methods. Online line rating
methods include monitoring conductor temperature or tension, and weather condi
tions all along the transmission line route. Conductor temperature is monitored by
installing conductor temperature sensors at certain sections of the transmission line.
Conductor tension is overseen by tension monitors that are attached to insulators on
TABLE 2.1
Ampacity of ACSR Conductors
Summer Coastal
Winter Coastal
Conductor Size, mm
2
Normal Emergency Normal Emergency
210 382 482 550 616
264 442 558 640 716
375 550 697 801 898
624 752 959 1108 1243
749 919 1133 1218 1393
874 1060 1312 1448 1625
1454 1319 1642 1814 2040
Basis for Table 2.1
Summer ambient temperature = 37°C with sun
Winter ambient temperature = 16°C without sun
Wind velocity = 0.6 m/s perpendicular to conductor axis
Conductor temperature, normal condition = 80°C
Conductor temperature, emergency condition A = 90°C (100 hr total)
Conductor temperature, emergency condition B = 100°C (100 hr total)
(Emergency B ratings are shown in the table)
Emissivity = 0.5
Conductivity of aluminum = 61% IACS
1306/C02/frame Page 17 Saturday, May 27, 2000 9:53 AM
18
Powerline Ampacity System: Theory, Modeling, and Applications
tension towers. Unlike temperature monitoring systems, tension monitors are
required to be located only at anchor towers. In both monitoring systems, sensor
data is communicated to a base station computer by a radio communication device
installed on the sensor, and the ampacity of the line is calculated at the base station
computer from this data.
In the ofﬂine system, line ratings are obtained uniquely by monitoring weather
conditions along the transmission line route. An ofﬂine system may also include
monitoring conductor sag by pointing a laser beam at the lowest point of the
conductor in a span. The ampacity of a line is calculated from conductor sag and
weather data by taking a series of measurements of conductor sag at different
transmission line spans along the length of the line.
2.2.3 W
EATHER
D
EPENDENT
S
YSTEMS
Weatherdependent line rating systems were proposed by several researchers
(Cibulka et al., 1992; Douglass, 1986; Hall and Deb, 1988b; Mauldin et al., 1988;
Steeley et al., 1991). These methods require weather data on a continuous basis.
Diurnal weather patterns of the region are considered for the prediction of line
ampacity several hours in advance. The existence of daily and seasonal cyclical
weather patterns are well known, and their usefulness to forecast powerline ampacity
was recognized by many researchers (Foss and Maraio, 1989), (Hall and Deb,
1988b).
A weatherdependent line rating system developed in the UK is described in a
Cigré article (Jackson and Price, 1985). Similarly, a weatherdependent realtime
line rating system has been developed for the Spanish 400 kV transmission network
(Soto, et al., 1998). In the Spanish system, realtime measurements of wind speed,
wind direction, ambient temperature, and solar radiation from several weather sta
tions are entered into a computer where a line ampacity program calculates steady
state and dynamic ampacity. Foss and Maraio (1989) described a line ampacity
system for the power system operating environment. They were also interested in
forecasting transmission line ampacity. In their method, line ampacity is adjusted
based on previous 24hour weather data. Because of these assumptions, the accuracy
of the system in forecasting transmission line ampacity several hours ahead is
somewhat limited.
In the LINEAMPS program (Deb, 1995a, 1995b), the periodic cyclical pattern
of wind speed and ambient temperature are considered in a unique manner to forecast
powerline ampacity. Weather patterns of a region are stored in Fourier series in each
weather station object. A method in each weather station object generates hourly
values of meteorological data from this series. The powerline objects have a plurality
of virtual weather sites that receive their data from a plurality of weather station
objects, and a method in each powerline object determines the minimum hourly
values of line ampacity up to seven days in advance. The number of virtual weather
stations that can be accommodated in a powerline is limited only by the computer
processing speed and memory, whereas installing an unlimited number of temper
ature sensors on a transmission line is not economical. Due to these reasons, a
1306/C02/frame Page 18 Saturday, May 27, 2000 9:53 AM
Line Rating Methods
19
weatherdependent line rating system is expected to be more reliable and more
accurate than systems utilizing realtime measurements from a limited number of
locations.
2.2.4 O
NLINE
T
EMPERATURE
M
ONITORING
S
YSTEM
U.S. Patent 5140257 (system for rating electric power transmission lines and equip
ment, 1992) was given for a transmission line rating system that calculates the current
carrying capacity of one or more powerlines by the measurement of conductor
temperature and meteorological conditions on the line. In this method, line ampacity
is calculated by the measurement of conductor temperature and by the solution of
the conductor heat balance equation as follows:
(2.4)
where,
P
r
= Heat lost by radiation, W/m
P
c
= Heat lost by convection due to the cooling effect of wind, W/m
P
s
= Heat gained by solar radiation, W/m
R
ac
= AC resistance of conductor, ohm/m
In the above equation, P
r
, P
c
, P
s
, and R
ac
are functions of conductor temperature. An
online temperature monitoring system using Power Donut ™ temperature sensors is
shown in Figure 2.1. The Power Donut ™ temperature sensor is shown in Figure 2.2.
Davis’s system (Davis, 1977) required the installation of conductor temperature
sensors as well as meteorological sensors at several locations along powerlines. Real
time conductor temperature, meteorological data, and line current are continuous
input to a computer system where line ampacity is calculated. The computer system
requires special hardware and software for data acquisition from remote sensor
locations via special telecommunication networks. The online monitoring systems
described in the IEEE and Cigré papers (Davis, 1977; Howington and Ramon, 1984;
Renchon and Daumerie, 1956) are not widely used because of transmission distance,
communication requirements, and maintenance problems. The new line ampacity
system does not require realtime continuous input of meteorological data, line
current, or conductor temperature measurements from the powerline. Powerline and
conductor ampacity is estimated by the program from user input and by synthetic
generation of weather data from selfgenerating weather station objects. General
purpose weather forecast data available from the internet are used in the LINEAMPS
program.
A realtime dynamic linerating model was proposed (Black and Byrd, 1983).
In this method, line ampacity is predicted accurately by realtime numerical solution
of the following conductor temperature differential equation at a location:
I
P P P
R
r c s
ac
=
+ –
1306/C02/frame Page 19 Saturday, May 27, 2000 9:53 AM
20
Powerline Ampacity System: Theory, Modeling, and Applications
(2.5)
Where,
M =
γ
· A, conductor mass, kg/m
A = conductor area, m
2
T
av
= average conductor temperature, °C
γ
= conduct or density, kg/m
3
(2.6)
T
c
= Conductor core temperature, °C
T
s
= Conductor surface temperature, °C measured by line temperature sensor
P
r
= Heat lost by radiation, W/m
P
c
= Heat lost by convection due to the cooling effect of wind, W/m
P
s
= Heat gained by solar radiation, W/m
P
m
= Magnetic heating, W/m
P
j
= Joule heating, W/m
The following expression for the calculation of realtime dynamic ampacity was
obtained by the author:
FIGURE 2.1
Online temperature monitoring system is comprised of Power Donut™
temperature sensors, and weather station and ground station RTU. (© Courtesy Nitech, Inc.)
SUBSTATION
CONTROL HOUSE
POWER DONUT
SENSORS
GROUND STATION
RTU
TM
M c
dT
dt
P P P P P
p
av
j s m r c
⋅ = + + – –
T
T T
av
c s
=
+
2
1306/C02/frame Page 20 Saturday, May 27, 2000 9:53 AM
Line Rating Methods
21
(2.7)
T
max
= Max conductor temperature
T
initial
= Initial temperature T
initial
, and time
∆
t
C
1
, C
2
= Constants
The different terms in the above equation are described in Chapter 3.
The calculation of dynamic ampacity by the above equation requires realtime
conductor temperature and meteorological data on a continuous basis.
2.2.5 O
NLINE
T
ENSION
M
ONITORING
S
YSTEM
The online tension monitoring system is used to predict transmission line ampacity
by measurement of conductor tension at tension towers along the transmission line
(Seppa, et al., 1998). Since conductor tension is a function of conductor temperature,
the ampacity of the transmission can be obtained by realtime monitoring of con
ductor tension as follows.
FIGURE 2.2
Power Donut™ temperature sensor. (© Courtesy Nitech, Inc.)
I
T T t
C t
C
initial
=
( ) { }
( ) { }
max
– exp –
– exp –
–
∆
∆
τ
τ
1
2
1
1306/C02/frame Page 21 Saturday, May 27, 2000 9:53 AM
22
Powerline Ampacity System: Theory, Modeling, and Applications
(2.8)
σ
1
,
σ
2
= stress at state1 and state2, respectively, kg/mm
2
Tc
1
, Tc
2
= conductor temperature at state1 and state2, °C
E = Young’s modulus of elasticity, kg/mm
2
ϖ
= speciﬁc weight of conductor, kg/m/mm
2
L = span length, m
∆
Ec = inelastic elongation (creep) mm/mm
α
= coefﬁcient of linear expansion of conductor, °C
–1
Therefore, by measurement of conductor tension and by knowledge of initial
conditions, the temperature of the conductor is obtained by the solution of the
above equation. Transmission line ampacity is then calculated by the solution of
conductor heat balance, Equation 2.5. This method of monitoring a transmission
line has the added advantage of monitoring iceloads as well. The major disad
vantages of this method are that it requires taking the transmission line out of
service for installation and maintenance. It may be feasible to install such devices
on certain heavily loaded lines, but is impractical and expensive to install tension
monitors on all transmission and distribution lines for line ampacity predictions
of all overhead lines in a system.
2.2.6 S
AG
M
ONITORING
S
YSTEM
This is an ofﬂine method of realtime line rating by monitoring conductor sag. It is
an ofﬂine method because it does not require the installation of any device on the
transmission line conductor. Therefore, this system does not require taking the line
out of service during installation or maintenance of the sagmonitoring device. In
this method, conductor sag is measured by pointing a laser beam at the lowest point
of the conductor in a span. Ampacity is calculated from conductor sag and by
measurement of weather conditions. The ampacity of the transmission line is then
obtained by taking a series of measurements at different transmission line spans
along the length of the line.
FIGURE 2.3
Sagmonitoring line ampacity system.
σ ϖ
σ
α
σ ϖ
σ
2
2
2
2 2 1
1
2
1
2
24 24 E
L
Tc Tc Ec
E
L
– – –
⋅ ( )
+
( )
+ =
⋅ ( )
∆
Transmission line spans
1 3 2 4 5
Laser beam
Sag measuring instrument
1306/C02/frame Page 22 Saturday, May 27, 2000 9:53 AM
Line Rating Methods
23
Conductor sag is calculated approximately by the wellknown parabola equation:
(2.9)
W = conductor weight, kg/m
T = conductor tension, kg
Having calculated conductor tension T from (2.9), the temperature of the conductor
is then obtained from (2.8), and then ampacity is calculated from (2.5).
2.2.7 DISTRIBUTED TEMPERATURE SENSOR SYSTEM
With the development of the powerline communication system by a ﬁberoptic cable
integrated with a powerline conductor, it is now possible to have a distributed system
of ﬁberoptic conductor temperature sensors that will span the entire length of the
transmission line. A distributed temperature sensor system will result in a more
accurate realtime line rating system, since a ﬁberoptic cable will be used for data
transmission and will eliminate the need for a separate communication system for
transmission of conductor temperature data from a transmission line to utility power
control center. At the present time, the ﬁberoptic cable is embedded inside the core
of powerline ground wire. This kind of conductor, with a ﬁberoptic cable in the
core, is called an OPGW conductor. A ﬁberoptic cable also may be placed within
the core of a phase conductor in highvoltage lines. In lowvoltage distribution lines,
the ﬁberoptic cable may be wrapped over the conductor. An example of an OPGW
conductor on overhead line is shown in Figure 2.4.
Probabilistic ratings are used by several power companies (Deb et al., 1985;
Giacomo, Nicolini, and Paoli, 1979; Koval and Billinton, 1970; Urbain, 1998).
Probabilistic ratings are determined by Monte Carlo simulation of meteorological
variables, and by the solution of the conductor heat balance equation. From the
resultant probability distribution of conductor ampacity, line ratings are determined.
An attractive feature of this technique is that continuous input of realtime weather
data is not required. A limitation is that line ampacity is not adaptive to real weather
conditions. However, it must be mentioned that probability modeling of conductor
temperature is useful for the prediction of conductor performance in service.
Probability modeling of conductor temperature is used to predict the loss of
tensile strength and permanent elongation of conductor during the lifetime of the
transmission line conductor (Deb, 1985, 1993; Hall and Deb, 1988b; Mizuno et al.,
1998). The author has obtained the probability distribution of conductor temperature
by Monte Carlo simulation of time series stochastic models of the meteorological
variables and transmission line current. By using time series stochastic models, it
is possible to consider the correlation between the different variables (Douglass,
1986). For example, it is well known that electricity demand depends upon weather
conditions.
Sag
WL
T
=
2
8
1306/C02/frame Page 23 Saturday, May 27, 2000 9:53 AM
24 Powerline Ampacity System: Theory, Modeling, and Applications
A time series stochastic and deterministic model was used to predict realtime
probabilistic ratings of transmission line ampacity up to 24hours in advance based
on ambient temperature measurements only, and by assuming constant wind speed
(Steeley et al., 1991). A stochastic model was also used to forecast solar radiation
(Mauldin et al., 1991) and wind speed (Hall and Deb, 1988b). The general form of
the stochastic model is given below:
Ta(t) = A + A
2
·Sin(wt) + A
3
·Sin(2wt)
+ A
4
·Cos(wt) + A
5
·Cos(2wt)
+ A
6
·Z(t – 1) + A
7
·Z(t – 2) (2.10)
Z(t – 1), Z(t – 2) = difference of measured and predicted temperature at time
(t – 1) and (t – 2) respectively
A
1
, A
2
, A
3
, A
4
, A
5
, A
6
, A
7
are the coefﬁcients of the model
ω = 2π/T = fundamental frequency
T = 24 hour = period
In the LINEAMPS program, Fourier series models of ambient temperature and
wind speed are used to generate weather data. Because weather patterns are stored
FIGURE 2.4 Power line conductors with ﬁberoptic cable.
1306/C02/frame Page 24 Saturday, May 27, 2000 9:53 AM
Line Rating Methods 25
in weather objects, it eliminates the need for realtime measurements on a con
tinuous basis. Weather data is required only when weather conditions change. One
of the limitations of the stochastic model is that it is unsuitable for the predictions
of hourly values of ambient temperature for more than 24 hours in advance. This
is to be expected, because timeseries models are statistical models that do not
consider a physical model of the atmosphere. In this regard, National Weather
Service forecasts are generally more accurate for longterm weather predictions
because they are derived from atmospheric models. Therefore, the parameters of
the weather models in the LINEAMPS program are adjusted to National Weather
Service forecasts.
2.2.8 OBJECTORIENTED MODELING AND EXPERT LINE
RATING SYSTEM
The estimation of powerline ampacity by the application of objectoriented modeling
and expert system rules was ﬁrst presented by the author (Deb, 1995). It was shown
for the ﬁrst time how objectoriented modeling of transmission line ampacity enabled
program users to easily create new lines and new conductors, and for weather stations
to estimate line ampacity up to seven days in advance. It was shown that powerline
objects not only have methods to predict line ampacity, but are also convenient
repositories for the storage of line data and ampacity that are easily retrieved and
displayed on a computer screen.
Important electric power companies are embracing the objectmodel approach
to meet their information requirements for the year 2000 and beyond (MPS Review
article, 1998). For example, EDF has selected objectmodel technology for the
management and operation of the transmission grid in France. Expert systems
(Kennedy, 2000) are developed in the electric power system for power quality
machine diagnosis, alarmprocessing (Taylor et al., 1998) and for power system fault
analysis (Negnevitsky, M., 1998). LINEAMPS is the ﬁrst expert system for the
estimation of transmission line ampacity (Deb, Anjan K., 1995).
Integrated Line Ampacity System
LINEAMPS (Deb, 1995a, 1995b), (Wook et al., 1997) is an integrated line ampacity
system having a transmission line model, conductor model, and weather model to
forecast line ampacity up to seven days in advance. It has provision for steadystate
rating, dynamic line rating, and transient rating. The concept of steady, dynamic,
and transient line ratings are used in this program for the ﬁrst time. A direct solution
of the conductor temperature differential equation is used, and an analytical expres
sion for the direct solution of dynamic line ampacity is presented for the ﬁrst time
(Deb,1998a). A threedimensional conductor thermal model is used to calculate
conductor thermal gradient (Deb, 1998a, 1998b).
An important contribution of the new line ampacity system is the ability to self
generate hourly values of weather data from statistical and analytical models, elim
inating the need for realtime weather data on a continuous basis. The author
previously developed an algorithm (Deb, 1993) for synthetic generation of weather
1306/C02/frame Page 25 Saturday, May 27, 2000 9:53 AM
26 Powerline Ampacity System: Theory, Modeling, and Applications
data by timeseries analysis and recursive estimation. The idea of selfgeneration,
synthetic generation, or artiﬁcial generation of meteorological data by a Fourier
series weather model of ambient temperature and wind speed of a region evolved
from these developments. Synthetic generation of weather data from a model is also
useful to evaluate the probability distribution of transmission line conductor tem
perature in service (Hall, Deb, 1988b). The probability distribution of transmission
line conductor temperature is required to calculate the thermal deterioration of
transmission line conductor (Mizuno et al., 1998, 2000).
2.3 CHAPTER SUMMARY
Transmission line rating methods are introduced by presenting a critical review of
literature, from the early works on conductor thermal rating to modern applications
of objectoriented modeling, expert systems, recursive estimation, and realtime
transmission line ratings. The line ampacity problem is clearly deﬁned, and the
various methods of rating power transmission lines are critically examined to show
the advantages and deﬁciencies of each method. The various methods of rating
transmission lines includes static and dynamic thermal ratings, weatherdependent
system, temperature monitoring system, tension monitoring system, sag monitoring,
distributed ﬁberoptic sensors, and probabilistic rating methods.
The development of an integrated powerline ampacity system having transmis
sion line, weather, and conductor thermal models that can be easily implemented in
all geographic regions was a major challenge. For this reason, it was necessary
to develop a computer program that will adapt to the different line operating stan
dards followed by power companies in the different regions of the world. This was
accomplished by objectoriented modeling and by developing an expert system
computer program called LINEAMPS
1306/C02/frame Page 26 Saturday, May 27, 2000 9:53 AM
27
3
Theory of Transmission
Line Ampacity
3.1 INTRODUCTION
As mentioned in the previous chapters, the currentcarrying capacity of a transmis
sion line conductor is not constant but varies with weather conditions, conductor
temperature, and operating conditions. Line ampacity is generally based on a max
imum value of conductor temperature determined by the type of conductor, and
depends upon the following operating conditions:
• Steadystate
• Dynamic state
• Transient state
The conductor is assumed to be in the steadystate during normal operating condi
tions when the heat gained due to line current and solar radiation equals the heat
lost by cooling due to wind and radiation. During steadystate conditions, the
transmission line current is considered constant, weather conditions are assumed
stable, and the temperature of the conductor is fairly uniform.
Dynamic conditions arise when there is a step change in line current. Line
energization or sudden changes in line current due to a failure on one circuit are
examples of dynamic operating conditions. A typical example of dynamic loading
is when the load from the faulted circuit in a double circuit line is transferred to the
healthy circuit. Due to the thermal inertia of the conductor, shortterm overloads
may be supplied through the line without overheating the conductor before steady
state conditions are reached.
Transient conditions arise due to shortcircuit or lightning current. During tran
sient conditions there is no heat exchange with the exterior, and adiabatic conditions
are assumed.
In the following section, the equations for the calculation of conductor temper
ature and ampacity are derived from the general heat equation for steadystate,
dynamic state, and transient conditions. Then, equations for the radial conductor
temperature differential from surface to core of a conductor are developed from the
same equations.
This chapter prepares the framework for computer modeling of the line ampacity
system described in Chapter 8.
1306/C03/frame Page 27 Saturday, May 27, 2000 10:15 AM
28
Powerline Ampacity System: Theory, Modeling, and Applications
3.2 CONDUCTOR THERMAL MODELING
3.2.1 G
ENERAL
H
EAT
E
QUATION
Starting with the general heat equation of the transmission line conductor, the
solution of the transmission line ampacity is found for each of the above operating
conditions as follows:
The general heat equation* for a transmission line conductor is given by,
(3.1)
Where,
In cylindrical coordinates,
(3.2)
T = conductor temperature
r = radial length
ϕ
= azimuth angle
z = axial length
q = power per unit volume
k = thermal conductivity of conductor
α
= thermal diffusivity given by,
α
=
c
p
= speciﬁc heat capacity
γ
= mass density
From (3.1) and (3.2) we obtain,
(3.3)
λ
= Thermal conductivity
* J.F. Hall, A.K. Deb, J. Savoullis, Wind Tunnel Studies of Transmission Line Conductor, IEEE,
Trans
actions on Power Delivery
, Volume 3, Number 2, April 1988.
∇ +
∂
∂
2
1
T
q
k
T
t α
∇
∇
∂
∂
+
∂
∂
+
∂
∂
2
2
2
2
2
2
2
Laplacian operator,
2
x y z
∇
∂
∂
+
∂
∂
+
∂
∂
+
∂
∂
2
2
2 2
2
2
2
2
1 1
T
T
r r
T
r r
T T
z ϕ
k
c
p
γ
∂
∂
⋅
∂
∂
+
∂
∂

(
'
`
J
J
+
T
t c
T
r r
T
r
q
p
λ
γ
2
2
1
1306/C03/frame Page 28 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity
29
Equation 3.3 may be solved numerically with appropriate initial and boundary
condition,
1,2,3
or solved analytically by making certain simplifying assumptions that
are presented in the following section.
3.2.2 D
IFFERENTIAL
E
QUATION
OF
C
ONDUCTOR
T
EMPERATURE
For practical consideration of transmission line conductor heating, it is possible to
make the following assumption with sufﬁcient accuracy:*
(3.4)
Where,
T
av
= average conductor temperature
T
c
= conductor core temperature
T
s
= conductor surface temperature
With the above assumption, we can calculate the average conductor temperature by
the solution of the following differential equation obtained from (3.3):
(3.5)
Where,
M =
γ⋅
A, conductor mass, kg/m
A = conductor area, m
2
P
j
= joule heating, W/m
P
s
= solar heating, W/m
P
m
= magnetic heating, W/m
P
r
= radiation heat loss, W/m
P
c
= convection heat loss, W/m
3.2.3 S
TEADY
S
TATE
A
MPACITY
The calculation of transmission line ampacity may be simpliﬁed if steadystate con
ditions are assumed. The following assumptions are made in steadystate analysis:
* V.T. Morgan, The radial temperature distribution and effective radial thermal conductivity in bare
solid and stranded conductors,
IEEE Transactions on Power Delivery
, Volume 5, pp. 1443–1452, July
1990
The thermal behaviour of overhead conductors. Section 3: Mathematical model for evaluation of con
ductor temperature in the unsteady state, Cigré Working Group WG 22.12 report,
Électra
No. 174,
October 1997.
T
T T
av
c s
+
2
M c
dT
dt
P P P P P
p
av
j s m r c
⋅ + + – –
1306/C03/frame Page 29 Saturday, May 27, 2000 10:15 AM
30
Powerline Ampacity System: Theory, Modeling, and Applications
• Conductor temperature remains constant for one hour.
• Conductor current remains constant for one hour.
• Ambient temperature, solar radiation, wind speed, and wind direction are
constant for one hour.
The steadystate solution is obtained by setting in (3.5), resulting in the
conductor heat balance equation:
P
j
+ P
s
+ P
m
– P
r
– P
c
= 0 (3.6)
By substitution,
P
j
+ P
m
= I
2
R
ac
(Tc) (3.7)
The following steadystate solution of conductor ampacity (I) is obtained:
(3.8)
Since the AC resistance of ACSR conductor varies as a function of conductor current
(Appendix 3), the ampacity of ACSR conductor is calculated by iteration as shown
in Example 1.
Description of symbols:
I = ampacity, A
R
ac
= R
dc
k
ac
[1 +
α
0
(T
c
– T
0
] (3.9)
R
ac
= AC resistance of conductor, ohm/m (R
ac
may be obtained directly from con
ductor manufacturer’s data sheet or calculated as shown in the Appendix).
R
dc
= DC resistance of conductor at the reference temperature T
o
, ohm/m
k
ac
=
α
0
= temperature coefﬁcient of resistance, /°C
T
c
= conductor temperature, °C
T
o
= reference conductor temperature, °C
P
s
= heat gains by solar radiation, W/m
P
s
=
α
s
D(S
b
+ S
d
) (3.10)
α
s
= coefﬁcient of solar absorption
D = conductor diameter, m
dT
dt
av
0
I
P P P
R Tc
r c s
ac
+
( )
–
R
R
ac
dc
1306/C03/frame Page 30 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity
31
S
b
= beamed solar radiation, W/m
2
S
d
= diffused solar radiation, W/m
2
S
b
= S
ext
τ
b
cos(z) (3.11)
S
d
= S
ext
τ
d
cos(
θ
) (3.12)
S
ext
= 1353 W/m
2
, normal component of the extra terrestrial solar radiation
measured outside the earth’s atmosphere
τ
b
= atmospheric transmittance of beamed radiation
τ
d
= atmospheric transmittance of diffused radiation
z =
zenith angle, degree
θ
= angle of beamed radiation with respect to conductor axis, degree
P
r
= heat loss by radiation, W/m
P
r
=
σεπ
D{(T
c
+ 273)
4
– (T
a
+ 273)
4
} (3.13)
T
a
= ambient temperature, °C
σ
= 5.67
⋅
10
–8
, Stephan Boltzman constant, (W/m
2
K
4
)
ε
= Emissivity of conductor
P
c
= heat loss by convection, W/m
P
c
= h ·
π
· D(T
c
– T
a
) (3.14)
h = coefﬁcient of heat transfer from conductor surface to ambient air,
W/(m
2
· °C)
h =
λ
· Nu · K
wd
/D (3.15)
λ
= thermal conductivity of ambient air, W/(m
· °C)
Nu = Nusselt number, dimensionless
Nu = 0.64 Re
0.2
+ 0.2 Re
0.61
(3.16)
Re = Reynolds number, dimensionless
Re = D · w
s
/ν
f
(3.17)
w
s
= wind speed, m/s
v
f
= kinematic viscosity of air, m
2
/s
k
wd
= wind direction correction factor given by (Davis. 1977)
K
wd
= 1.194 – sin(ω) – 0.194 cos(2ω) + 0.364 sin(2ω) (3.18)
ω = wind direction with respect to conductor normal, degree
1306/C03/frame Page 31 Saturday, May 27, 2000 10:15 AM
32 Powerline Ampacity System: Theory, Modeling, and Applications
The AC resistance, R
ac
, of a conductor is generally available from the manufacturer’s
catalog for standard conductor sizes at a certain speciﬁed conductor temperature.
The value of the AC resistance, R
ac
, of the conductor of any size for any temperature
may be calculated by the procedure given in Appendix 3. As shown in the Appendix,
the AC resistance of conductor is calculated from the current distribution inside the
conductor, conductor construction, and the magnetic properties of the steel core in
ACSR.
The AC resistance of conductors having a steel core can be 5 to 15% or more
higher than its DC resistance due to the current induced in the steel core as shown
in Appendix 3. For stranded conductors without a magnetic core, the AC resistance
may be 2 to 5% higher than the DC resistance due to skin effect.
A ﬂow chart of the steadystate current method is shown in Figure 3.1, and a
numerical application is given in Example 1. A ﬂow chart of the steadystate temper
ature calculation method is given in Figure 3.2, and a numerical application is given
in Example 2. These methods are used in the LINEAMPS program.
Example 1
Calculate the steadystate ampacity of an ACSR Cardinal conductor. Conductor
temperature is 80°C. The meteorological conditions and conductor surface charac
teristics are as follows:
Ambient temperature = 20°C
Wind speed = 1 m/s
Wind direction = 90° (perpendicular to conductor axis)
Solar radiation = 1000 W/m
2
Emissivity = 0.5
Absorptivity = 0.5
FIGURE 3.1 Flow chart of steadystate ampacity method.
Receive input from steady state
session window
Heat gain by radiation
Ps
Radiation Heat Loss
Pr
Convection Heat Loss
Pc
I =
P
R
P P
s r
ac
c
1306/C03/frame Page 32 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity 33
Solution
Calculate joule heat gain P
j
P
j
= I
2
⋅
k
ac
⋅ Rdc
20
{1 + α
0
(T
c
– T
0
)}
P
j
= I
2
⋅
k
ac
⋅ 0.5973⋅ 10
–3
{1 + 0.004(80 – 20)}
P
j
= I
2
⋅
k
ac
⋅ 0.0741⋅ 10
–3
W/m
Calculate solar heat gain P
s
P
s
= α
s
· D · F
s
P
s
= 0.5 · 30.39 · 10
–3
· 1000
P
s
= 15.19 W/m
Calculate convection heat loss by wind P
c
P
c
= k
f
· Nu · π(T
c
– T
a
)
k
f
= 2.42 · 10
–2
+ T
f
· 7.2 · 10
T
f
= 0.5(80 + 20) = 50°C
k
f
= 2.42 · 10
–2
+ 50 · 7.2 · 10
–1
= 0.0278 W/(m·°K)
FIGURE 3.2 Flow chart of steady state conductor temperature method.
Steady:Conductor Temperature
Tc
Abs(Error) > 0.01W/m
Error = Ps + Pj + Pm Pr  Pc
Convention beat loss
Pc
Radiation heat loss
Pr
Joule + Mag heat gain
Pj +Pm
Assume initial conductor temperature
equal to ambient temperature
Heat gain by solar radiation
Ps
Receive input from steady
state session window
New estimate of
conductor temperature
1306/C03/frame Page 33 Saturday, May 27, 2000 10:15 AM
34 Powerline Ampacity System: Theory, Modeling, and Applications
Nu = 0.64 · Re
0.2
+ 0.2 · Re
0.61
Re =
M
air
= air density = 1.103 kg/m
3
v = 1 m/s
vf
o
= 1.32 · 10
–5
+ T
f
· 9.5 · 10
–8
vf
o
= 1.32 · 10
–5
+ 50 · 9.5 · 10
–8
= 1.795 · 10
–5
m
2
/s
H = Altitude, m
Altitude at sea level, H = 0
vf
h
= vf
o
= 1.795 · 10
–5
m
2
/s
Nu = 0.64(1867)
0.2
+ 0.2(1867)
0.061
= 23
P
c
= 0.0278 ⋅ 30.39 ⋅ 10
–3
⋅ π ⋅ (T
c
–T
a
) = 0.0278 ⋅ 23 ⋅ π ⋅ (80–20)
P
c
= 120 W/m
Heat lost by radiation P
r
P
r
= s · ε · π · D[(T
c
+ 273)
4
– (T
a
+ 273)
4
]
P
r
= 5.67 · 10
–8
· 0.5 · π · 30.39 · 10
–3
[(T
c
+ 273)
4
– (T
a
+ 273)
4
]
Pr = 22.08 W/m
By substitution in the steadystate heat balance equation,
P
j
+ P
s
= P
c
+ P
r
We obtain,
M v D
vf
air
h
⋅ ⋅
vf vf H
h o
⋅
( )
]
]
]
]
1
6 5 10
288 16
3
5 2561
–
.
.
–
– .
Re
. .
.
–
–
⋅ ⋅
⋅
1 103 30 39 10
1 795 10
1867
3
5
I k
I
k
ac
ac
2
3
0 0741 15 19 120 2 08
120 22 08 15 19
0 0741 10
⋅ ⋅ + +
+
⋅ ⋅
. . .
. – .
.
–
1306/C03/frame Page 34 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity 35
assume k
ac
= 1,
I = 1308 A
For this current k
ac
= 1.16
The revised value I from (3.8 ) is,
I = 1214 A
For this current k
ac
= 1.14
This process is repeated until convergence, and the ﬁnal value of current is found
to be,
I = 1220 A
The calculation of AC resistance of ACSR Cardinal conductor is shown in the
Appendix 1.
Example 2
Calculate the temperature of an ACSR Cardinal conductor. Conductor current is
1220 A, and all other conditions are the same as in Example 1.
Solution
From (3.8) and (3.9) we have,
P
j
= I
2
· k
ac
· Rdc
20
{1 + α
0
(T
c
– T
0
)}
For I = 1220 A we obtain k
ac
= 1.14, and the joule heat gain P
j
is then calculated as,
P
j
= 1220
2
· 1.14 · 0.05973 · 10
–3
{1 + 0.004(T
c
– 20)}
From Example 1 we obtain the value of solar heat gain P
s
,
P
s
= 15.19 W/m
Using the value of k
f
and Nu calculated in Example 1, the convection heat loss P
c
is obtained from (3.14):
P
c
= 0.0278 ⋅ 23 ⋅ π ⋅ (T
c
– 20)
The heat loss by radiation P
r
is given by (3.13):
P
r
= 5.67 · 10
–8
· 0.5 · π · 30.39 · 10
–3
[(T
c
+ 273)
4
– (T
a
+ 273)
4
]
1306/C03/frame Page 35 Saturday, May 27, 2000 10:15 AM
36 Powerline Ampacity System: Theory, Modeling, and Applications
The conductor temperature T
c
is calculated from the steadystate heat balance
equation:
P
j
+ P
s
= P
c
+ P
r
By substitution in the above equation we obtain:
1220
2
· 1.14 · 0.05973 · 10
–3
{1 + 0.004(T
c
– 20)} + 15.19
= 0.0278 · 23 · π(T
c
– 20) + 5.67 · 10
–8
· 0.5 · π · 30.39 · 10
–3
[(T
c
+ 273)
4
– (20 + 273)
4
]
The above equation is solved for T
c
by iteration by giving an initial value T
c
= T
a
.
The converged value of T
c
is found to be 80°C. A direct solution of steadystate
conductor temperature T
c
is also obtained from the following quartic equation (Davis,
1977):
(3.19)
where,
and obtain
T
c
= 80°C
The result of conductor temperature T
c
obtained by the direct solution of the quartic
equation (3.19) is also found to be 80°C.
3.2.4 DYNAMIC AMPACITY
The transmission line conductor is assumed to be in the dynamic state when there
is a shortterm overload on the line due to line energization or a step change in load.
The duration of such overload condition is generally less than 30 minutes. In the
dynamic state, the heat storage capacity of the conductor is considered, which allows
a T a T a T a T k
c c c c 1
4
2
3
3
2
4 5
0 ⋅ + ⋅ + ⋅ + ⋅ +
a D
a a
a a
a a Nu
a P P a T a T a T a T
j s a a a a
1
2 1
3 1
2
4 1
3
5 1
4
2
3
3
2
4
4 273
6 273
4 273
⋅ ⋅ ⋅
⋅ ⋅
⋅ ⋅
⋅ ⋅ + ⋅ ⋅
+ + ⋅ + ⋅ + ⋅ + ⋅
( )
π ε σ
π λ
–
1306/C03/frame Page 36 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity 37
higher than normal static line loading to be allowed on the line for a short duration.
The temperature of the conductor in the dynamic state is obtained by the solution
of the following differential equation (3.5):
(3.20)
The nonlinear differential equation can be solved numerically by Euler’s method as
follows (Davidson, 1969):
(3.21)
where,
T
i
= initial temperature
By selecting a suitable time interval ∆t = dt, we can replace the above integral by
a summation such that,
(3.22)
The above equation is solved by the algorithm shown in the ﬂow chart of Figure
3.3a, which is suitable for realtime calculations. A solution of the above nonlinear
differential equation by the Runge Kutte method is given by Black (Black et al.,
1983).
Direct Solution of Dynamic Conductor Temperature
A direct solution of the nonlinear differential equation is possible by making some
simplifying assumptions to linearize the equation. We deﬁne an overall heat transfer
coefﬁcient h
o
(Dalle et al., 1979) such that,
P
c
+ P
r
= π · D · h
o
· (T
c
– T
a
) (3.23)
Combining joule and magnetic heating,
P
j
+ P
m
= I
2
· R
ac
= I
2
· k · Rdc
20
{1 + a
0
(T
av
– T
0
)} (3.24)
we obtain,
(3.25)
M c
dT
dt
P P P P P
p
av
j s m r c
⋅ + + – –
T
P P P P P dt
M c
T
av
t
j s m r c
p
i
+ +
( )
⋅
+
∫
0
– –
T
P P P P P t
M c
T
av
j s m r c
p
t
i
+ +
( )
⋅
+
∑
– – ∆
0
M c
dT
dt
I k Rdc T T D F D h T T
p
av
av s s o aw s
⋅ ⋅ ⋅ +
( ) { ¦
+ ⋅ ⋅ ⋅ ⋅ ⋅
( )
2
20 0 0
1 α α π – – –
1306/C03/frame Page 37 Saturday, May 27, 2000 10:15 AM
38 Powerline Ampacity System: Theory, Modeling, and Applications
Conductor Temperature
The solution of the linear differential equation (3.25) (Dalle et al., 1979),
T
ch(i)
= θ
1
– (θ
1
– T
ch(i–1)
· exp(–∆t/τ
h
) (3.26)
T
cc(i)
= θ
2
– (θ
2
– T
cc(i–1)
· exp(–∆t/τ
h
) (3.27)
Where,
T
ch(i)
= conductor temperature during heating
T
cc(i)
= conductor temperature during cooling
(3.28)
(3.29)
The heating time constant is given by,
(3.30)
FIGURE 3.3a Real time calculation of dynamic conductor temperature.
T = T
i
i c
c
T = P t
c t
.
T = T + T
c
t j m
c
s c r
c i
T = T + T
c c i
T = Initial temperature
i
START T = T
Input Ws, Wd, Ta, Sr
At 1 min intervals
P = P + P + P  P  P
t = 60s
T = Conductor temperature
Figure 3.3.a Realtime calculation dynamic conductor temperature
θ
α π
π α
1
1
2
0
0 1
2
1
⋅ ⋅
( )
+ ⋅ + ⋅ ⋅
( )
⋅ ⋅ ⋅ ⋅
R I T D P h T
D h R I
ac ref s o a
o ac
–
–
θ
α π
π α
2
2
2
0
0 2
2
1
⋅
( )
+ ⋅ + ⋅ ⋅
( )
⋅ ⋅ ⋅ ⋅
R I T D P h T
D h R I
ac ref s o a
o ac
–
–
τ
π α
h
p
ac
M c
D h R I
⋅
⋅ ⋅ ⋅ ⋅
0 0 1
2
–
1306/C03/frame Page 38 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity 39
The cooling time constant is given by,
(3.31)
The coefﬁcient of heat transfer during heating is,
(3.32)
The coefﬁcient of heat transfer during cooling is,
(3.33)
(3.34)
(3.35)
I
1
= overload current
I
2
= post overload current
∆T
cl
= T
cl
– T
a
(3.36)
∆T
c2
= T
c2
– T
a
(3.37)
T
c1
= steadystate preload conductor temperature
T
c2
= steadystate overload conductor temperature
∆t = time step, t
i
– t
i–1
A = sectional area of conductor, m
2
D = diameter of conductor, m
α
0
= temperature coefﬁcient of resistance, /°C
R
0
= DC resistance of conductor at reference temperature T
ref
, ohm/m
T
ref
= reference temperature, generally 20°C or 25°C
T
a
= ambient temperature, °C
R
ac
= AC resistance of conductor, ohm/m
c
p
= speciﬁc heat capacity, (J/kg °K) at 20°C. For elevated temperature operation
the speciﬁc heat may be calculated by, c
p
(Tc) = c
p
(T
20
){1 + β(T
c
– T
20
)}
β = temperature coefﬁcient of speciﬁc heat capacity, /°C ( Table 1).
A Flow Chart of Dynamic Temperature method is given in Figure 3.3b and a
numerical application is presented in Example 3.
τ
π α
c
p
ac
M c
D h R I
⋅
⋅ ⋅ ⋅ ⋅
0 0 2
2
–
h
P P
D T
o
j s
c
+
⋅ ⋅
1
1
π ∆
h
P P
D T
c
j s
c
+
⋅ ⋅
2
2
π ∆
P I R
j ac 1 1
2
⋅
P I R
j ac 2 2
2
⋅
1306/C03/frame Page 39 Saturday, May 27, 2000 10:15 AM
40 Powerline Ampacity System: Theory, Modeling, and Applications
Direct Solution of Dynamic Ampacity
From Equations (3.27) and (3.28) we may obtain the maximum value of dynamic
ampacity approximately as follows:
(3.38)
(3.39)
(3.40)
I = dynamic ampacity, A
T
max
= maximum conductor temperature, °C
T
initial
= intial conductor temperature, °C
FIGURE 3.3b Flow chart of dynamic temperature method used in the LINEAMPS program.
Calculate Xh
Receives message
from Rule check
dynamic data
Calculateτ
Calculate θ
Calculate θ
Calculate Tcc(i)
Calculate Tch(i)
Calculate Xc
Incerment time
i = i +1
Tch(i) >= Tc(max)?
Time >=120 min?
c
1
2
Calculateτ
h
Yes
Yes
No
No
Result: Dynamic
Temperature
Incerment time
i = i +1
I
T T t
C t
C
initial
( ) { ¦
− ( ) { ¦
max
– exp –
exp –
–
τ
τ
1
1
2
C
R T
D I R
ac ref
ac
1
0
2
0
1
( )
⋅ ⋅ ⋅ ⋅
–
–
α
π γ α
C
D P T
R T
s s a
ac ref
2
0
1
⋅ + ⋅ ⋅
( )
⋅ ⋅
( )
α π γ
α –
1306/C03/frame Page 40 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity 41
τ = conductor heating time constant, s
t = duration of overload current, s
A numerical application of dynamic ampacity method is presented in Example 3.4.
Formulas for calculation of constants of conductors composed of different material
are given in Table 3.1.
Result of Conductor Temperature in Dynamic State
The result of conductor temperature vs. time in the dynamic state is obtained from
(3.26) and presented in Figure 3.4. In this example the analysis is carried out by
selecting a typical transmission line Zebra conductor by using the Line Ampacity
System (LINEAMPS) software developed by the author.*
TABLE 3.1
(Cigré 1999)
Constant
1
Formula
2
Resistivity, ρ, Ω · m
Temperature coefﬁcient of resistance, α /° K
Speciﬁc heat, c, J/(kg · °K)
Temperature coefﬁcient of speciﬁc heat, β /° K
Mass, kg/m
1
Constants are calculated at 20°C.
2
Subscripts a, s are for aluminum and steel
m
a
, m
s
are mass density of aluminum and steel respectively, kg/m
3
A = area, m
2
* Anjan K. Deb. Objectoriented expert system estimates transmission line ampacity, IEEE Computer
Application in Power, Volume 8, Number 3, July 1995.
ρ
ρ ρ
ρ ρ
+
+
( )
a s a s
a s sd a
A A
A A
α
α α
ρ ρ
α
ρ
α
ρ
ρ ρ
α
ρ
α
ρ
+ + +
+ + +

(
'
`
J
J

(
'
`
J
J

(
'
`
J
J

(
'
`
J
J

(
'
`
J
J
a s
a
s
s
a
a
s
s
s
a
a
a
a
s
s
a
a
a
s
s
s
A A A A
A A A A
c
c m A c m A
m A m A
a a a s s s
a a s s
+
+
β
β β
β β
+
+
c m c m
m m
a a a s s m
a a s s
M
A m A m
A A
a a s s
a s
+
+
1306/C03/frame Page 41 Saturday, May 27, 2000 10:15 AM
42 Powerline Ampacity System: Theory, Modeling, and Applications
Example 3.3
Calculate the temperature of ACSR Cardinal conductor when 1475 A overload
current is passed through the conductor for 20 minutes. Normal load current is 1260
A. All other conditions are as follows:
Ambient temperature = 20°C
Wind speed = 1 m/s
Wind direction = 90°
Sun = 1000 W/m
2
Emissivity = 0.5
Solar absorption = 0.5
Maximum average conductor temperature = 100°C
Steadystate normal conductor temperature at 1260 A = 80°C
Solution
The overall heat transfer coefﬁcient h
o
remains fairly constant for a given set of
meteorological conditions within a range of conductor temperatures and evaluated
by (3.33),
substituting for P
j
and P
s
we have,
FIGURE 3.4 Dynamic conductor temperature as a function of time for ACSR Zebra con
ductor due to a step change in current.
Dynamic Conductor Temperature: ACSR Zebra
Dynamic Temperature
ACSR Zebra
T
e
m
p
e
r
a
t
u
r
e
,
C
Time, min
0
0
20
20
40
40
60
60
80
80
100
100
120
120
0
h
P P
D T T
o
j s
c a
⋅ ⋅
( )
π –
h
I R D F
D T T
h
h W m C
o
ac s s
c a
o
o
⋅ + ⋅ ⋅
⋅ ⋅
( )
⋅ ⋅ ⋅ ⋅ + ( ) { ¦
+ ⋅ ⋅ ⋅
⋅ ⋅ ⋅ ( )
⋅ °
( )
2
2 3 3
3
2
1260 0 05973 10 1 1 1 0 004 80 20 0 5 30 39 10 1000
30 37 10 80 20
25 3
α
π
π
–
. . . – . .
. –
.
– –
–
1306/C03/frame Page 42 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity 43
Substituting values in (3.26),
where,
we obtain,
An average temperature of conductor after 20 minute is obtained from (3.26),
Example 3.4
Calculate the dynamic ampacity of ACSR Cardinal conductor for 20 minutes. All
other conditions are the same as in Example 3.1.
Solution
The dynamic ampacity is calculated directly from (3.38),
T T
t
ch i ch i
h
( ) − ( )
− −
( )

(
'
`
J
J

¦








θ θ
τ
1 1 1
exp
–
θ
α π
π α
1
2
0
0
2
1
⋅ ⋅
( )
+ ⋅ + ⋅ ⋅
( )
⋅ ⋅ ⋅ ⋅
R I T D P h T
D h R I
ac ref s o a
o ac
–
–
θ
π
π
θ
1
3 2 3
3 3 2
1
0 05973 10 1 1 1475 1 0 004 20 30 39 10 0 5 1000 25 3 20
30 39 10 25 3 004 0 05973 10 1 1 1475
106
⋅ ⋅ ⋅ ⋅ + ⋅ ⋅ ⋅ + ⋅ ⋅
⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅
°
( ) ( ) . . – . . . .
. . – . . .
– –
– –
C
τ
π α
τ
π
τ
h
p
o
h
h
m c
D h R I
s
⋅
⋅ ⋅ ⋅ ⋅
⋅
⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅
–
.
. . – . . .
– –
0 0
2
3 3 2
1 828 826
30 39 10 25 3 004 0 05973 10 1 1 1475
822
T
T C
av
av
( ) ⋅

(
`
J
°
106 106 80
1200
822
100
– – exp
–
I
T T t
C t
C
C
R T
D I R
initial
ac ref
ac
( ) { ¦
( ) { ¦
( )
⋅ ⋅ ⋅ ⋅
max
– exp –
– exp –
–
–
–
τ
τ
α
π γ α
1
1
0
2
0
1
2
1
1306/C03/frame Page 43 Saturday, May 27, 2000 10:15 AM
44 Powerline Ampacity System: Theory, Modeling, and Applications
Since current I is not known, we assume a steadystate current I = 1440 A at 100°C
to calculate C1,
Substituting in (3.38) we obtain the value of dynamic ampacity I,
For greater accuracy we may recalculate C1 with the new value of current I. The
ﬁnal value of dynamic ampacity is found to be 1475 A, which is the same as
Example 3.
3.2.5 TRANSIENT AMPACITY
Transient conditions arise when there is shortcircuit or lightning current. The dura
tion of transient current is generally in the range of milliseconds as most power
system faults are cleared within few cycles of the 60 Hz frequency. During this time,
adiabatic condition is assumed (Cigré, 1999) when there is no heat exchange with
the exterior.
Algorithm for the Calculation of Transient
Conductor Temperature
Transient conductor temperature response due to shortcircuit current is obtained
from the solution of the following differential equation:
(3.41)
C
C
1
3
3 2 3
5
1 1 0 0593 10 1 0 004 20
25 2 30 39 10 1440 1 1 0 0593 10 0 004
1 3 24310 10
⋅ ⋅ ⋅ ( )
⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅
⋅
. . – .
. . – . . .
.
–
– –
–
π
C
D P T
R T
C
C
s s a
ac ref
2
1
2
30 39 10 0 5 1000 25 2 20
1 1 0 0593 10 1 0 004 20
2 1 04810 10
0
3
3
5
⋅ ⋅ + ⋅ ⋅
( )
⋅ ⋅
( )
⋅ ⋅ ⋅ + ⋅ ⋅ ( )
⋅ ⋅ ⋅ ⋅ ( )
⋅
α π λ
α
π
–
. . .
. . – .
.
–
–
I
I A
⋅ ( ) { ¦
⋅ ( ) { ¦
⋅
100 80 1200 822
3 23 10 1 1200 822
1 08 10
1484
5
6
– exp –
. – exp –
– .
–
M c
dT
dt
P P
p
av
j m
⋅ +
1306/C03/frame Page 44 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity 45
where,
During adiabatic condition there is no heat exchange with the exterior therefore,
P
s
= 0
P
r
= 0
P
c
= 0
The solution of the differential equation is given by,
(3.42)
Where,
T
i
= initial conductor temperature, °C
t = time, s
T
o
= reference temperature, °C
α
o
= temperature coefﬁcient of DC resistance of conductor, /°C
R
ac
= AC resistance of conductor at reference temperature T
o
, ohm/m
I
sc
= short circuit current, A
M = conductor mass, kg/m
c
p
= speciﬁc heat of conductor, J/Kg · °K
Equation (3.42) provides the temperature of the conductor during heating by a short
circuit current. The temperature during cooling of the conductor is obtained from
the dynamic equation.
A ﬂow chart of the transient ampacity method is shown in the Figure 3.5 and a
numerical application is shown in Example 5.
Example 5
Calculate the temperature of ACSR Cardinal conductor after a shortcircuit current
of 50 kA is applied through the conductor for 1 second. The conductor was carrying
1260 A steadystate current when the shortcircuit current was applied. All other
conditions are the same as in Example 4.
Solution
From Example 4 we obtain the initial temperature of the conductor to be equal to
80°C when the shortcircuit current is applied. The following additional data were
calculated in Example 4:
P P I R T T
j m sc ac c
+ +
( )
[ ]
2
0 0
1 α –
T T T
R I t
M c
c i o
ac sc
p
+

(
'
`
J
J
⋅

(
'
`
J
J
]
]
]
]
– – exp
1
1
0
0
2
α
α
1306/C03/frame Page 45 Saturday, May 27, 2000 10:15 AM
46 Powerline Ampacity System: Theory, Modeling, and Applications
cp = 826 J/(kg ⋅ °C)
R
ac20
= 6.57 ⋅ 10
–5
From (3.42 ) we obtain the temperature of ACSR Cardinal conductor for the fol
lowing condition:
Short circuitcurrent, I
sc
= 50 kA
Duration of shortcircuit current, t = 1
s
By substitution of values in the above equation, the temperature of the conductor
Tc is calculated as,
FIGURE 3.5 Flow chart of transient ampacity method.
Input from transient
session window
Input data
error
Yes
Yes
Yes Yes
No
No No
Calculate prefault
steady state conductor
temperature
Calculate Pj
Calculate Tc(i)
Incremente time
i = i + 1
Time > duration
of short circuit
Update lineplot
conductor heating
Update lineplot
conductor cooling
Increment time
i = i + 1
Time > 120 min
Calculate Tcc(i)
Calculate AA
Calculate τ
c
Calculate Xc
Post error message
T T T
R I t
M c
c i o
ac sc
p
+

(
'
`
J
J
⋅

(
'
`
J
J
]
]
]
]
– – exp
1
1
0
0
2
α
α
T
T C
c
c
+

(
`
J
⋅ ⋅ ⋅ ⋅
( )
⋅
⋅

(
'
'
`
J
J
J
]
]
]
]
°
80 20
1
004
1
0 004 6 57 10 50 10 1
1 828 826
205
5 3
2
–
.
– exp
. .
.
–
1306/C03/frame Page 46 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity 47
The above example shows the importance of clearing faults by a highspeed fault
protection system using modern circuit breakers and protective relaying that can
detect and clear faults within a few cycles.
Result of Conductor Temperature in Transient State
Calculated by Program
Results obtained by the application of the transient ampacity algorithm by using
the LINEAMPS program are presented in Figure 3.6. Conductor temperature as a
function of time is shown by a line graph when a shortcircuit current equal to 50
kA is applied for 0.5s. The conductor is ACSR Zebra.
3.2.6 RADIAL CONDUCTOR TEMPERATURE
In the previous section, dynamic ampacity calculations were carried out by assuming
average conductor temperature. Often, the surface temperature of a conductor is
available by measurement, and core temperature is required to calculate sag. The
calculation of the radial temperature differential in the conductor is also required
for dynamic ampacity calculation. Radial temperature gradient is particularly impor
tant for highampacity transmission line conductors since they are capable of oper
ating at high temperatures. For high value of ampacity, substantial radial temperature
differences from 1 – 5°C were measured in the wind tunnel. Based on the radial
temperature differential an average value of conductor temperature can be estimated.
In this section the radial temperature of the conductor is derived from the general
heat equation. From the general heat equation (3.1) we obtain
(3.43)
FIGURE 3.6 Transient temperature as a function of time for ACSR Zebra conductor due
to a shortcircuit current.
Transient Temperature
ACSR Zebra
T
e
m
p
e
r
a
t
u
r
e
,
C
Time, millisec
0
0
20
20
40
40
60
60
80
80
100
100
120
∂
+
∂
∂
+ ⋅
∂
∂
+
∂
∂
+
( )
⋅
∂
∂
2
2 2
2
2
2
2
1 1 1 T
r r
T
r r
T T
z
q r
k
T
t
r
φ α
1306/C03/frame Page 47 Saturday, May 27, 2000 10:15 AM
48 Powerline Ampacity System: Theory, Modeling, and Applications
For a 1mlong cylindrical conductor we may assume that,
In the steady state,
Assuming constant heat generation per unit volume, q(r) = q = constant.
By the application of above conditions we obtain,
(3.44)
where,
r = radial distance from conductor axis, m
k
r
= radial thermal conductivity, W/(m ⋅ °K)
q is the internal heat generation by unit volume obtained by,
(3.45)
For homogeneous conductors the following boundary conditions are applied,
T(r) = T
s
at r = r
s
r
s
= conductor radius, m
The solution to (3.44) is then,
(3.46)
∂
∂
∂
∂
T
T
z
φ
0
0
∂
∂
T
t
0
∂
∂
+
∂
∂
+
( )
2
2
1
0
T
r r
T
r
q r
k
r
q
I R
A
ac
al
2
∂ ( )
∂
T r
r
0 at r = 0
T r T r I R
r
r
s s ac
s
( )

(
'
`
J
J

¦








– –
2 2
2
1
1306/C03/frame Page 48 Saturday, May 27, 2000 10:15 AM
Theory of Transmission Line Ampacity 49
Substituting,
Pj + P
m
= I
2
R
ac
The radial temperature difference from conductor core to surface ∆T in homogeneous
conductor is obtained as,
(3.47)
For bimetallic conductor (ACSR), the boundary conditions are,
By the application of the above boundary conditions for ACSR conductor, the radial
conductor differential is obtained by,
(3.48)
where,
T
c
= conductor core temperature, °C
T
s
= conductor surface temperature, °C
3.3 CHAPTER SUMMARY
Starting with a threedimensional transmission line conductor thermal model, a
differential equation of conductor temperature with respect to time is developed in
this chapter. Steadystate solutions of the differential equation are given for the
calculation of conductor ampacity and conductor temperature. Differential equations
are developed for dynamic and transient conditions, and their closed form solutions
are given. The radial temperature differential in the conductor due to the difference
in the surface and core temperature is also derived. Algorithms for the calculation
of transmission line conductor ampacity and temperature are presented with worked
out practical examples. The AC resistance of ACSR conductors increases with
A r
A
al s
al
π
2
Aluminum area, m
2
T o T T AAC
P P
k
s
j m
r
( ) ( )
+
– ∆
4π
T r T
T r
r
s
( )
∂ ( )
∂
at r = r
at r = r
s
c
0
T T
r I R
A k
r
r
r
r
r
r
c s
s ac
al r
c
s
c
s
c
s
– – ln

(
'
`
J
J
+

(
'
`
J
J

(
'
`
J
J

¦








2 2
2
4
1 2
1306/C03/frame Page 49 Saturday, May 27, 2000 10:15 AM
50 Powerline Ampacity System: Theory, Modeling, and Applications
conductor temperature as well as conductor current. The calculation of AC resistance
of ACSR conductor, including magnetic heating and current redistribution in the
different layers of the conductor, is presented in Appendix 3 at the end of this chapter
with a numerical application.
1306/C03/frame Page 50 Saturday, May 27, 2000 10:15 AM
51
Appendix 1
AC Resistance of ACSR
The AC resistance of conductors having magnetic cores is greater than their DC
resistance because of the transformer action created by the spiraling effect of current
in the different layers of aluminum wires. The increase in the AC resistance of ACSR
conductors are mainly due to current redistribution in the aluminum wire layers, and
the magnetic power loss in the steel core due to eddy current and hysteresis loss.
Therefore, the AC resistance of ACSR conductors may be considered to be composed
of the following:
1. DC resistance
2. Increment in resistance due to current redistribution
3. Increment in resistance due to magnetic power losses in the steel core
The resistance and inductance model of a threelayer ACSR conductor is shown in
Figure A1.1 (Vincent, M., 1991), (Barrett et al., 1986). As shown in the ﬁgure, the
reactance of each layer of aluminum wire is due to the selfinductance, L
nn
; mutual
inductance; L
m,n
, due to the longitudinal ﬂux; and the circular inductance, L
c
, due
to the circular ﬂux. The circular inductance model assumes that there is 21%
contribution due to inner ﬂux, and 79% contribution by the outer ﬂux of each wire
in a layer (Vincent, M., 1991), (Barrett et al., 1986). The longitudinal inductances
lead to the longitudinal self reactances X
mm
, and mutual reactances, X
m,n
. Similarly,
the inner and outer circular inductances lead to the inner and outer circular reac
tances, Xc
n,i
and Xc
n,o
, respectively.
The current redistribution in the different layers of the aluminum wires are due
to longitudinal and circular ﬂux, which are calculated as follows (Vincent, M., 1991),
(Barrett et al., 1986).
Longitudinal Flux
The magnetic ﬁeld intensity, H (A/m), of a wire carrying current, I, is given by
Ampere’s current law,
(A 1.1)
N = number of turns of aluminum wires over the steel core given by,
H dl N I
c
⋅ ⋅
∫
1306/appendix 1.1/frame Page 51 Saturday, May 27, 2000 9:00 AM
52
Powerline Ampacity System: Theory, Modeling, and Applications
(A 1.2)
s
i
= lay length of layer i, m
The magnetic ﬂux
φ
due to the magnetic ﬁeld B (T) is obtained as,
(A 1.3)
Therefore, for layer n, the magnetic ﬂux,
φ
n
, is,
φ
n
= B
n
· A
n
(A 1.4)
B
n
= magnetic ﬁeld, Tesla
Applying,
B =
µ
H
µ
=
µ
0
µ
r
FIGURE A1.1
Electric circuit model of ACSR.
d
S
2
3
1
D
D
D
D
I I
j(X I X I +X I ) R I
R I
R I
R I
Xc
3o
3
2
1
s
3 31
21
11
32
22
12
33
23
23
1
1
1
2
2
2
3
3
3
2
1
s
3i
2o
2i
1o
n
n
n
Xc
Xc
Xc
Xc
Xc
j(X I +X I X I )
j(X I X I X I )
Resistance of
layer
Longitudinal
Inductance of
layer
Circular Inductance
of layer
Aluminum wire
Steel wire
N
s
i
=
1
φ = ⋅
∫
B ds
s
1306/appendix 1.1/frame Page 52 Tuesday, May 30, 2000 12:26 PM
Appendix 1 AC Resistance of ACSR
53
The ﬂux in layer n is obtained by,
φ
n
=
µ
0
µ
r
H
n
A
n
(A 1.5)
(A 1.6)
The selfinductance of layer n, L
nn
(
µ
r(al)
= 1) is given by,
The selfreactance of layer n, X
nn
is,
(A 1.7)
The mutual inductance of layer m, n is,
(A 1.8)
And the mutual reactance of layer m, n is,
(A 1.9)
Circular Flux
By assuming that the layers currents are concentrated at the center of each layer,
the outer circular ﬂux due to layer, n, of length, l, is obtained by,
(A 1.10)
Which has for solution,
(A 1.11)
Similarly the inner circular ﬂux due to layer (n + 1) is obtained as,
φ µ π µ
n n c r c
n
n
r A A
I
s
( )
+
[ ] 0
2
–
X
f r A A
s
nn
n c r c
n
( )
+
[ ]
2
0
2
2
π µ π µ –
M M
r A A
s s
m n n m
n c r c
n q
, ,
–
( )
+
[ ]
⋅
µ π µ
0
2
X X
f r A A
s s
m n n m
n c r c
n q
, ,
–
( )
+
[ ]
⋅
2
0
2
π µ π µ
φ
π
µ µ
n outer
D
D
n
n
N
r
n
n
I
r
dr dz
,
–
.
∫ ∫
∑
1
0
1
0
0
2
φ
π
µ µ
n outer
n
n
N
r
n
n
I
D
D d
,
ln
–
∑

(
'
`
J
J
0
0
2
1306/appendix 1.1/frame Page 53 Saturday, May 27, 2000 9:00 AM
54
Powerline Ampacity System: Theory, Modeling, and Applications
(A 1.12)
It was previously shown (Vincent, M., 1991), (Barrett et al., 1986) that the layer
current contributes 21% to the inner ﬂux, and 79% to the outer ﬂux because of
current distribution in a wire as shown in Figure A1.1.
The voltage drop per meter along each layer is given by,
The voltage drop V
1
in layer 1 is,
(A 1.13)
The voltage drop V
2
in layer 2 is,
(A1.14)
The voltage drop in layer 3 is,
(A 1.15)
φ
π
µ µ
n inner
n
n
N
r
n
n
I
D d
D
+
+
∑

(
'
`
J
J 1
0
0
1
2
,
ln
–
V I R jI X jI X jI X j f I I
D
D d
j f I I I
D d
D
j f I I I
D
D d
j
s
s s
1 1 1 1 11 2 12 3 13 0 1
1
1
0 1 2
2
1
0 1 2
2
2
0 79
0 21 0 79
+ + + +
( )

(
'
`
J
J
+ + +
( )

(
'
`
J
J
+ + +
( )

(
'
`
J
J
+
– . ln
–
. ln
–
. ln
–
µ
µ µ
µµ µ
0 1 2 3
3
2
0 1 2 3
3
3
0 21 0 79 f I I I I
D d
D
j f I I I I
D
D d
s s
+ + +
( )

(
'
`
J
J
+ + + +
( )

(
'
`
J
J
. ln
–
. ln
–
V I R jI X jI X jI X j f I I I
D
D d
j f I I I I
D d
D
j f I I I
D
D d
s
s s
2 2 2 1 21 2 22 3 23 0 1 2
2
2
0 1 2 3
3
1
0 1 2
2
2
0 79
0 21 0 79
+ + +
( )

(
'
`
J
J
+ + + +
( )

(
'
`
J
J
+ + +
( )

(
– – – . ln
–
. ln
–
. ln
–
µ
µ µ
''
`
J
J
+ + + +
( )

(
'
`
J
J
+ + + +
( )

(
'
`
J
J
j f I I I I
D d
D
j f I I I I
D
D d
s s
µ µ
0 1 2 3
3
2
0 1 2 3
3
3
0 21 0 79 . ln
–
. ln
–
V
3
V I R jI X jI X jI X j f I I I
D
D d
s 3 3 3 1 31 2 32 3 23 0 1 3
3
3
0 79 + + + + +
( )

(
'
`
J
J
– . ln
–
µ
1306/appendix 1.1/frame Page 54 Saturday, May 27, 2000 9:00 AM
Appendix 1 AC Resistance of ACSR
55
The voltage drop in the steel core is,
(A 1.16)
Equations (A1.13)–(A 1.16) may be set up as a set of four simultaneous equations
with four unknown currents, , which satisfy the following conditions:
The sum of layer currents must equal total current ,
(A 1.17)
The voltage drop of each layer are equal,
(A 1.18)
From the calculated layer currents we obtain the voltage drop, V. The AC resistance
of the conductor is then found by,
The calculation of the AC resistance of a conductor is carried out iteratively because
the complex relative permeability,
µ
r
, of steel core is a nonlinear function of the
magnetic ﬁeld intensity, H. The magnetic ﬁeld intensity, H, is a function conductor
current. The following relation may be used to calculate complex relative perme
ability,
µ
r
, for H
≤
1000 A/m.
µ
r
= [40 – 0.0243
Έ
H
Έ
+ 0.000137
Έ
H
Έ
2
] – j[5 + 1.03 · 10
–10
Έ
H
Έ
2
]
Example 6
Calculate the AC resistance of a 54/7 ACSR Cardinal conductor for the following
operating conditions:
Conductor current = 1000 A
Average conductor temperature = 80°C
V
s
V I R j f I I
D d
D
j f I I
D
D d
j f I I I
D d
D
j f I I I
D
D
s s s s
c
s
s s
+ +
( )
−

(
'
`
J
J
+ +
( )

(
'
`
J
J
+ + +
( )

(
'
`
J
J
+ + +
( )
µ µ
µ µ
0 1
1
0 1
2
1
0 1 2
2
1
0 1 2
2
2
0 21 0 79
0 21 0 79
. ln . ln
–
. ln
–
. ln
––
. ln
–
. ln
–
d
j f I I I I
D d
D
j f I I I I
D
D d
s s

(
'
`
J
J
+ + + +
( )

(
'
`
J
J
+ + + +
( )

(
'
`
J
J
µ µ
0 1 2 3
3
2
0 1 2 3
3
3
0 21 0 79
I I I I
1 2 3 2
, , ,
I
I I I I I
s 1 2 3
+ + +
V V V V
s 1 2 3
R
V
I
ac

(
'
`
J
J
Re
1306/appendix 1.1/frame Page 55 Saturday, May 27, 2000 9:00 AM
56
Powerline Ampacity System: Theory, Modeling, and Applications
Solution
Cardinal conductor data:
Number of steel wires, n
s
= 7
Number of aluminum wires, n
al
= 54
Number of aluminum wires in layer 1 = 12
Number of aluminum wires in layer 2 = 18
Number of aluminum wires in layer 3 = 24
Wire diameter, d = 3.376 mm
Conductor diameter, D = 30.38 mm
Aluminum resistivity,
ρ
a
= 0.028126
Ω ⋅
mm
2
/m
Steel resistivity,
ρ
s
= 0.1775
Ω ⋅
mm
2
/m
The following layer lengths are assumed:
λ
s
= 0.253 m
λ
1
= 0.219 m
λ
2
= 0.236 m
λ
3
= 0.456 m
The dc resistance of Layer i is given by,
Initial layer current,
Rdc
D d
s
A n
i
i
i
i
i
i i
+
( ) { ¦
ρ
π
1
10
3
–
–
1, 2, 3 layers
I
Rdc
Rc Rdc Rdc
I
I Rdc
Rdc
I
I Rdc
Rdc
I
I Rdc
R
s
s
1
1
2 3
2
1 1
2
3
1 1
3
1 1
1
1
1 1 1
+ + +

¦




]
]
]
]
⋅
⋅
⋅
1306/appendix 1.1/frame Page 56 Saturday, May 27, 2000 9:00 AM
Appendix 1 AC Resistance of ACSR
57
We obtain initial layer currents,
I
1
= 327 A
I
2
= 489 A
I
3
= 652 A
I
c
= 31 A
The voltage drop in Layer 1 is,
The voltage drop, V
2
, in Layer 2 is,
The voltage drop, , in Layer 3 is,
The voltage drop, , in the steel core is,
V I R jI X jI X jI X j f I I
I I I I I I
I I I I I
s
s s
s s
1 1 1 1 11 2 12 3 13 0 1
1 2 1 2
1 2 3
0 79
5
4
0 21
6
5
0 79
7
6
0 21
8
7
+ + + +
( )
[

(
`
J
+ + +
( )

(
`
J
+ + +
( )

(
`
J
+ + + +
( )

(
`
J
+
– . ln
. ln . ln
. ln
µ
++ + +
( )

(
`
J
]
]
]
I I I
1 2 3
0 79
9
8
. ln
V I R jI X jI X jI X j f I I I
I I I I I I I
I I I I
s
s s
s
2 2 2 1 21 2 22 3 23 0 1 2
1 2 3 1 2
1 2 3
0 79
7
6
0 21
8
5
0 79
7
6
0 21
8
7
+ + +
( )
[

(
`
J
+ + + +
( )

(
`
J
+ + +
( )

(
`
J
+ + + +
( )

– – – . ln
. ln . ln
. ln
µ
((
`
J
+ + + +
( )

(
`
J
]
]
]
I I I I
s 1 2 3
0 79
9
8
. ln
V
3
V I R jI X jI X jI X j f I I I I
s 3 3 3 1 31 2 32 3 23 0 1 2 3
0 79
9
8
+ + + + + +
( )

(
`
J
– . ln µ
V
s
V I R j f I I I I
I I I I I I
I I I I I
s s s s s
s s
s s
+ +
( )
[

(
`
J
+ +
( )

(
`
J
+ + +
( )

(
`
J
+ + +
( )

(
`
J
+ + + +
( )

(
`
J
+
µ
0 1 1
1 2 1 2
1 2 3
0 21
4
3
0 79
5
4
0 21
6
5
0 79
7
6
0 21
8
7
. ln . ln
. ln . ln
. ln ++ + +
( )

(
`
J
]
]
]
I I I
1 2 3
0 79
9
8
. ln
1306/appendix 1.1/frame Page 57 Saturday, May 27, 2000 9:00 AM
58
Powerline Ampacity System: Theory, Modeling, and Applications
Where,
The magnetic ﬁeld, H (A/m), is obtained by,
The complex relative permeability, , of the steel core is given by,
The sum of all layer currents is equal to total current,
Voltage drop in each layer is equal,
The above problem was solved by Mathcad®* Solver giving initial values and
the following results were obtained,
* Mathcad 8® is registered trademark of Mathsoft, Inc., http://www.mathsoft.com/
X
A
s
X
A
s s
X
A
s s
X
A
s
X
A
s s
X
A
s
c
c
c
c
c
c
11 0
1
2
12 0
1 2
13 0
1 3
22 0
2
2
23 0
2 3
33 0
3
2
2 60
2 60
2 60
2 60
2 60
2 60
⋅ ⋅
⋅ ⋅
⋅
⋅ ⋅
⋅
⋅ ⋅
⋅ ⋅
⋅
⋅ ⋅
π µ
π µ
π µ
π µ
π µ
π µ
H
I
s
I
s
I
s
+
1
1
2
2
3
3
–
µ
r
µ
r
H H j H +
[ ]
+ ⋅
[ ]
40 0 0243 0 000137 5 1 03 10
2
10
4
– . . – .
–
I I I I I
s 1 2 3
+ + +
V V V V
s 1 2 3
V V V V j
s 1 2 3
0 109 0 024 + . .
1306/appendix 1.1/frame Page 58 Saturday, May 27, 2000 9:00 AM
Appendix 1 AC Resistance of ACSR
59
Current density:
Layer 1 = 2.72 A/m
2
Layer 2 = 3.34 A/m
2
Layer 3 = 3.08 A/m
2
The ac resistance of the conductor is given by,
The ac/dc ratio, k, is composed of a factor k
1
due to current redistribution in the
layers, and a factor k
2
due to magnetic power loss in a ferromagnetic core, and is
given by,
The current redistribution factor, k
1
, is obtained by,
The magnetic power loss factor, k
2
, is obtained by,
I j
I j
I j
j
r
1
2
3
273 104
534 62
659 52
131 76
+
+
–
– µ
R
V
I
R
R
ac
ac
dc

(
'
`
J
J
⋅
⋅
⋅
Re .
.
.
.
–
–
–
7 432 10
7 432 10
6 432 10
1 127
5
5
5
R
R
k k k
ac
dc
⋅
1 2
k
I R I R I R I R
I R
c c
dc
1
2
1
2
1 2
2
2 3
2
3
2
⋅ + ⋅ + ⋅ + ⋅
⋅
k
k
k
2
1
1306/appendix 1.1/frame Page 59 Saturday, May 27, 2000 9:00 AM
61
4
Experimental Veriﬁcation
of Transmission
Line Ampacity
I have always endeavoured to make experiment the test and controller of
theory and opinion.
Michael Faraday on Electricity
4.1 INTRODUCTION
The object of this chapter is to present experimental data on transmission line
ampacity for the development and validation of theory, hypotheses, and assump
tions. The data presented in this chapter is compiled from the different tests that
I have either conducted myself, or were conducted by other people in different
research laboratories. As much as possible, the data presented here are from
published literature. I have selected Michael Faraday’s (1834) quotation for this
discussion not only for its general applicability to all experimental research, but
also for his particular interest in the subject of electricity and the heating of wires
by electric current.
4.2 WIND TUNNEL EXPERIMENTS*
Experiments were carried out at in a wind tunnel to verify conductor thermal
modeling for static and dynamic thermal ratings, and to determine the radial thermal
conductivity of conductors. Atmospheric conditions of wind speed and ambient
temperature were simulated in a wind tunnel that was specially built for these studies.
A transmission line conductor was installed in the wind tunnel, and current was
passed through it to study the effects of environmental variables on conductor
heating.
In Table 4.1, wind tunnel data is compared to the values calculated by the
program, showing excellent agreement between measured and calculated values.
The measured value of steadystate ampacity is 1213 Amperes in the wind tunnel
with 2.4 m/s wind, 90° wind direction, and 43°C ambient temperature. The same
* “Wind Tunnel Studies of Transmission Line Conductor Temperatures,” by J.F. Hall, Paciﬁc Gas &
Electric Co., and Anjan. K. Deb, Consultant, Innova Corporation, presented and published in
IEEE
Transactions in Power Delivery
, Vol. 3, No. 2, April 1988, pages 801–812.
1306/C04/frame Page 61 Saturday, May 27, 2000 10:23 AM
62
Powerline Ampacity System: Theory, Modeling, and Applications
value is calculated by the LINEAMPS program. When wind speed is zero, the
measured value of ampacity is 687 Amperes compared to 718 Amperes calculated
by the program. Due to the inherent uncertainties in the measurement of atmospheric
variables, we may easily expect 5 to 10% measurement error. The difference between
measurement and calculations are within this range in all of the data presented in
Table 4.1.
In addition to the comparison of wind tunnel data with the program, data from
several other sources are presented to show their excellent agreement with results
obtained by the LINEAMPS program. The data presented in Table 4.1 represents a
diverse sampling of line ampacity results obtained in the different regions of the
world, including the northern and southern hemispheres of the globe, which have
different national standards. For example, Southwire is a wellknown conductor
manufacturing company in the U.S. PG&E is the largest investorowned electric
utility in the U.S., EDF is the national electric supply company of France, and Dr.
Vincent Morgan is a leading authority on conductor thermal rating in Australia
(Morgan, 1991). In all of the above examples, the results obtained from the LIN
EAMPS program compared well with the data presented in Table 4.1.
A sketch of the wind tunnel is given in Figure 4.1, showing the placement of
the conductor inside the wind tunnel to achieve different wind angles. A 25hp motor
was used to power four 36inch fans at 0 to 960 rpm. Wind speed in the range of 0
to 20 mph was generated inside the wind tunnel and measured by propellertype
anemometers manufactured by R. M. Young Co. Two sizes of fourblade polystyrene
propellers were used for low and high windspeed measurements. The smaller
propellers were 18 cm in diameter and 30 cm in pitch, and the larger propellers
were 23 cm in diameter and 30 cm in pitch.
From wind tunnel experimental data of conductor temperature at various wind
speeds, the following empirical relationship between the Nusselt number (Nu) and
the Reynolds number (Re) is determined for the calculation of forced convection
cooling in conductor:
Nu = exp{3.96 – 0.819 ·
ᐉ
n Re + 0.091 · (
ᐉ
n Re)
2
} (4.1)
1000
≤
Re
≤
15000
The results from the above equation are compared to data presented by other
researchers in Figure 4.2 with excellent agreement.
When higherthannormal transmission line ampacity is allowed through a line,
it is also necessary to evaluate the probability distribution of conductor temperature.
The loss of conductor tensile strength (Mizuno et al., 1998), the permanent elonga
tion of the conductor due to creep (Cigré, 1978), the safety factor of the line, and
transmission line sag as a function of the life of the line (Hall and Deb, 1988b) are
calculated from the probability distribution of conductor temperature and discussed
further in Chapter 5.
1306/C04/frame Page 62 Saturday, May 27, 2000 10:23 AM
Experimental Veriﬁcation of Transmission Line Ampacity
63
4.3 EXPERIMENT IN OUTDOOR TEST SPAN
An outdoor test span is useful for the veriﬁcation of transmission line sag and tension
calculated by the LINEAMPS program. The computer program for the calculation
of sag and tension uses the following transmission line conductor change of state
equation:
(4.2)
TABLE 4.1
Ampacity Test Results
Source Conductor Sun Ta °C
Ws
m/s Wd° Tc °C Rating Type
Source
Amp
LINEAMP
S Amp
Southwire Drake N 40
0 50
Summer Steady
320 319
Southwire Drake Y 40
1.2
90
75
Summer Steady
880 880
Southwire Drake Y 40
0.61
90
100
Summer Emergency
15 min
1160 1160
PG&E Cardinal Y 43
0.61
90
80
Summer Steady
838 830
Wind Tunnel Cardinal N 43
0 78.1
Measured Steady
687 718
Wind Tunnel Cardinal N 43
2.4
90
75.8
Measured Steady
1213 1213
EDF Aster 570 Y 30
1
90
60
Summer Steady
830 826
EDF Aster 851 Y 15
1
90
60
Winter Steady
1350 1366
EDF Aster 570 Y 15
1
90
75
Winter Dynamic
20 min
1393 1390
EDF Aster 570 Y 15
1
90
150
Transient 1 sec
44.5 kA 44.5 kA
Morgan V. T. Curlew N 0
0.6
90
80
Winter Steady Night
1338 1324
Morgan V. T. Curlew Y 5
0.6
90
80
Winter Steady Noon
1182 1233
Notes:
Ta = Ambient temperature, Degree Celsius
Ws = Wind speed, meter per second
Wd = Wind direction, Degree
Tc = Conductor surface temperature, Degree Celsius
Y = Yes, N = No
Source = Name of company or research publication from where data was obtained for this test.
• Southwire is a trademark of Southwire Company, Carrolton, GA.
• PG&E is the Paciﬁc Gas & Electric Company, San Francisco.
• EDF is Electricité de France, Paris.
• Vincent T. Morgan is author of
Thermal Behavior of Electrical Conductors
, published by Wiley, Inc., New York, 1991.
• LINEAMPS is
Line Ampacity System
, an objectoriented expert line ampacity system. U.S. Patent 5,933,355 issued
August 1999 to Anjan K. Deb.
• Wind Tunnel data from: “Wind Tunnel Studies of Transmission Line Conductor Temperatures,”
IEEE Transactions
on Power Delivery
, Vol. 3, No. 2, April 1988, Authors: J. F. Hall, Anjan K. Deb, J. Savoullis.
σ ϖ
σ
α
σ ϖ
σ
2
2
2
2 2 1
1
2
1
2
24 24 E
L
Tc Tc Ec
E
L
– – –
⋅ ( )
+
( )
+ =
⋅ ( )
∆
1306/C04/frame Page 63 Saturday, May 27, 2000 10:23 AM
64
Powerline Ampacity System: Theory, Modeling, and Applications
where,
σ
1
,
σ
2
= stress at state1 and state2 respectively, kg/mm
2
Tc
1
, Tc
2
= conductor temperature at state1 and state 2, °C
E = Young’s modulus of elasticity, kg/mm
2
FIGURE 4.1
Wind tunnel.
FIGURE 4.2
Forced convection Nusselt number vs Reynolds number relationship obtained
from wind tunnel experiments and comparison with results from other researchers.
Plan View
23
,
A
25 HP
Motor
Wind Flow
90
o
Anemomet
Pivot point for conductor angle
Wind Flow
Wind Straightner
Conductor
A
Four 36 Fans
0 to 960 RPM
,,
SECTION A  A
15 6
, ,,
8
,
3
PVC
pipe 6
long
,
4
,,
Venturi
WIND TUNNEL
Reynolds Number, Re
HallDeb JohannetDalle Morgan
0
N
u
s
s
e
l
t
s
N
u
m
b
e
r
,
N
FORSED CONVECTION HEAT TRANSFER
Nusselt Number vs Reynolds Number
500 1000 5000 2000 10000 15000 30000
20
40
60
80
100
120
140
160
180
200
5
1306/C04/frame Page 64 Saturday, May 27, 2000 10:23 AM
Experimental Veriﬁcation of Transmission Line Ampacity
65
ϖ
= speciﬁc weight of conductor, kg/m/mm
2
L = span length, m
∆
Ec = inelastic elongation (creep) mm/mm
α
= coefﬁcient of linear expansion of conductor, °C
–1
Results obtained by the application of above equation are presented in Table 4.2.
The sag and tension program is further veriﬁed by comparison with ﬁeld data
from various electric power companies* with excellent agreement (Wook, Choi, and
Deb, 1997).
As shown in Figure 4.3, the transmission line conductor was energized by a 100
kVA transformer, and the temperature of the conductor was controlled by varying
the current passing through it. Conductor sag at midspan was measured by a scale
which compared well with the sag calculated by the program.
The loss of strength of aluminum alloy wires was determined experimentally
by heating individual wires at elevated temperatures. The results of this experiment
are presented in Table 4.3. These experiments were conducted at the EDF laboratory
at Paris, France (Deb, 1978).
* Thanks are due to Mr. Wally Sun, Transmission Line Engineer, PG&E, San Francisco, CA, for providing
transmission line sag and tension data . A report was submitted to Mr. Sun which shows the result of
this comparison, June 1990.
TABLE 4.2
Veriﬁcation of Transmission Line Security
(ACSR Cardinal Conductor)
Tc °C
Wind,
Pa
LOS
%RTS
Creep
µ
strain
Tension
kN
Safety
Factor
Sag
m
Life
Year
15 1480 4 1200 73.03 2.00 2.85 50
80 – 4 1200 20.56 7.00 10.12 50
100 – 4 1200 19.40 7.47 10.72 50
1 Pascal (Pa) = 0.02 lbf/ft2
1 kN (Kilo Newton) = 224.8 lbf
mstrain = micro strain = mm/km
RTS = Rated Tensile Strength
Tc = Average conductor temperature, °C
Wind = Wind pressure on projected area of conductor, Pa
LOS = Loss of Strength
SF = Safety Factor of conductor
Initial sag after stringing = 9.1m @100 °C
1306/C04/frame Page 65 Saturday, May 27, 2000 10:23 AM
66
Powerline Ampacity System: Theory, Modeling, and Applications
4.4 COMPARISON OF LINEAMPS WITH IEEE AND CIGRÉ
4.4.1 S
TEADY
S
TATE
A
MPACITY
The IEEE* recommends a standard method for the calculation of currentcarrying
capacity of overhead line conductors based on theoretical and experimental research
FIGURE 4.3
Test setup of high temperature conductor sag measurement in outdoor
test span.
TABLE 4.3
Loss of Strength of Aluminum Alloy Wires
at Elevated Temperatures
Conductor
Temperature, °C
Duration,
hr
Loss of Strength,
%
150 100 32.35
150 20 24.71
150 5 7.65
150 4 4.80
150 2 2.40
130 10 2.06
130 10 1.47
Note:
Aluminum alloy wire size is 3.45 mm in diameter.
* IEEE Standard 7381993.
IEEE Standard for calculating the currenttemperature relationship of bare
overhead conductors.
+ +
+
~ 220v/50Hz/20 Deg
220/100
Variable Transformer
100kVA
100m
4 m
1.5 m
Insulator
Conductor
1306/C04/frame Page 66 Saturday, May 27, 2000 10:23 AM
Experimental Veriﬁcation of Transmission Line Ampacity
67
carried out by several researchers. Similarly, Cigré* (Conférence International de la
Grande Réseaux Electrique), the international conference on large electrical networks,
has proposed a method for calculating the thermal rating of overhead conductors. The
two methods of ampacity calculation were compared by PG&E engineer N.P.
Schmidt,** and the results of this comparison were presented in a 1997 IEEE paper
(Schmidt. 1997). The comparisons are based on steadystate conditions only. The study
shows that there may be up to 10% variation in the two methods of ampacity calculation.
In this section, LINEAMPS results are compared to the results given by
Schmidt (1997) in Figures 4.4–4.9. These results show that the values calculated
by LINEAMPS are within 10% of those of the IEEE (IEEE Std. 738, 1993) and
Cigré (1997, 1992). The assumptions regarding the transmission line and the
various meteorological conditions are presented in Table 4.4 from the IEEE
paper.
It is appropriate to mention here that these comparisons were made on the
assumptions that the meteorological conditions comprised of wind speed, wind
direction, sky condition, and ambient temperature are the same all along the trans
mission line route. It is important to note that IEEE and Cigré provide methods to
calculate line ampacity when the ambient conditions are given. They do not include
* The thermal behaviour of overhead conductors. Section 1, 2, and 3. Report prepared by Cigré Working
Group 22.12. Section 1 and 2,
Electra
, October 1992, Section 3,
Electra
, October 1997.
** N. P. Schmidt.
Comparison between IEEE and Cigré ampacity standards
. IEEE Power Engineering
Society conference paper # PE749PWRD0061997. Anjan K. Deb, Discussion contribution to this
paper, October 1997.
FIGURE 4.4
Ampacity calculated by LINEAMPS program is compared to the IEEE Stan
dard and Cigré method of calculating conductor thermal rating in the steady state. Figure
shows the variation of conductor ampacity as a function of ambient temperature. All other
assumptions are speciﬁed in Table 4.4.
40 30 50
10 20
Ambient Temperature, C
LINEAMPS
IEEE
CIGRE
1300
1250
1200
1150
1100
1050
1000
950
900
850
800
A
m
p
a
c
i
t
y
,
A
COMPARISON OF LINEARAMPS, IEEE, CIGRE
ACSR DRAKE AMPASITY (Ambient Temperature Effects)
Comparision of Ambient Temperature Effects
1306/C04/frame Page 67 Saturday, May 27, 2000 10:23 AM
68
Powerline Ampacity System: Theory, Modeling, and Applications
methods for modeling variations in meteorological conditions along the transmission
line route. The different meteorological conditions along the transmission line route
are considered in a unique manner by the LINEAMPS program, as stated in a
discussion contribution recently prepared by this author (Deb, 1998). Faraday also
FIGURE 4.5
Ampacity calculated by LINEAMPS program is compared to the IEEE Stan
dard and Cigré method of calculating conductor thermal rating in the steady state. Figure
shows the variation of conductor ampacity as a function of wind speed. All other assumptions
are speciﬁed in Table 4.4.
FIGURE 4.6
Ampacity calculated by LINEAMPS program is compared to the IEEE Stan
dard and Cigré method of calculating conductor thermal rating in the steady state. Figure
shows the variation of conductor ampacity as a function of wind direction. All other assump
tions are speciﬁed in Table 4.4.
1600
1400
1200
1000
800
600
400
200
0
0 1 2 3 4 5 6 7 8
LINEAMPS
IEEE
CIGRE
COMPARISON OF LIMEAMPS, IEEE, CIGRE
ACSR DRAKE AMPACITI (Wind Speed Effectc)
A
m
p
a
c
i
t
y
,
A
Wind speed, ft/s
30 50 10 2
Wind Direction, Degree
1100
950
900
850
800
700
A
m
p
a
c
i
t
y
,
A
COMPARISON OF LINEAMPS, IEEE, CIGRE
ACSR DRAKE (Wind Direction Effects)
0 60 70 80 90
IEEE
CIGRE
LINEAMPS
40
650
600
1050
1000
750
1306/C04/frame Page 68 Saturday, May 27, 2000 10:23 AM
Experimental Veriﬁcation of Transmission Line Ampacity
69
realized the problem of changing cooling effects on wire when different parts of the
wire are exposed to different cooling conditions when he stated:*
FIGURE 4.7
Ampacity calculated by LINEAMPS program is compared to the IEEE Stan
dard and Cigré method of calculating conductor thermal rating in the steady state. The effect
of solar radiation on conductor ampacity is shown as a function of time of day. All other
assumptions are speciﬁed in Table 4. LINEAMPS considers both direct beam and the diffused
solar radiation hence the predicted ampacity is slightly lower than IEEE and Cigré. Diffused
radiation was neglected in the comparison made in the IEEE paper. However, the effect of
solar radiation on line ampacity is comparatively small when compared to the effects of
ambient temperature and wind.
FIGURE 4.8
Ampacity calculated by LINEAMPS program in dynamic state is compared
to Cigré method of calculation of dynamic ampacity when wind speed is 0.5 m/s.
* Michael Faraday on Electricity.
“
On the absolute quantity of electricity associated with the particles
or atoms of matter.”
Encyclopedia Britannica
. Great Books # 42, page 295. January 1834.
11 12 10
Time of Day, hr
1100
1080
1060
1040
1020
1000
980
960
940
920
900
A
m
p
a
c
i
t
y
,
A
COMPARISON OF LINEAMPS, IEEE, CIGRE
ACSR DRAKE AMPACITY (Solar Effects)
13 14
IEEE
CIGRE
LINEAMPS
Comparison of Solar Effects
400
600
700
1000
1100
1200
1300
1400
Comparison of Dynamic Ampacity (Wind Speed Effect: 0.5 m/s)
COMPARISON OF DYNAMIC AMPACITY
Wind Speed = 0.5 m/s
Interval of Overload, min
D
y
n
a
m
i
c
A
m
p
a
c
i
t
y
,
A
3 10 30
CIGRE
LINEAMP
500
800
900
1306/C04/frame Page 69 Saturday, May 27, 2000 10:23 AM
70
Powerline Ampacity System: Theory, Modeling, and Applications
The same quantity of electricity which, passed in a given time, can heat an inch of
platina wire of a certain diameter redhot can also heat a hundred, a thousand, or any
length of the same wire to the same degree, provided the cooling circumstances are
the same for every part in all cases.
FIGURE 4.9
Ampacity calculated by LINEAMPS program in dynamic state is compared
to Cigré method for the calculation of dynamic ampacity when wind speed is 2 m/s.
TABLE 4.4
Data for Line Ampacity Calculations Presented in Figures 4.4–4.9
Transmission Line Conductor 795 kcmil 26/7 ACSR Drake
Wind Speed 2 ft/s
Wind Direction Perpendicular to Line
Latitude 30 °
Azimuth of Conductor 90°
Atmosphere Clear
Solar Heating On
Diffuse Solar Radiation 0 (ignored in IEEE & Cigré), considered in LINEAMPS
Emissivity 0.5
Absorptivity 0.5
Elevation above Sea Level 0 m
Ground Surface Type Urban
Time of Day 11:00 am
Time of Year June 10
Maximum Conductor Temperature 100°C
Source: N.P. Schmidt, Comparison between IEEE and Cigré ampacity standards, IEEE Power
Engineering Society conference paper # PE749PWRD0061997. Anjan K. Deb, Discussion
contribution, October 1997.
400
600
800
1200
1400
1600
1800
2000
CIGRE
Comparison of Dynamic Ampacity (Wind Speed Effect: 2m/s)
COMPARISON OF DYNAMIC AMPACITY
Wind Speed = 2m/s
Interval of Overload, min
D
y
n
a
m
i
c
A
m
p
a
c
i
t
y
,
A
3 10 30
LINEAMPS
1000
1306/C04/frame Page 70 Saturday, May 27, 2000 10:23 AM
Experimental Veriﬁcation of Transmission Line Ampacity
71
In an overhead power transmission line the cooling effects are generally not the
same at all sections of the line because of its length. A transmission line may be 10
or 100 miles long (or greater), and the meteorological conditions cannot be expected
to remain the same everywhere. LINEAMPS takes into consideration the different
cooling effects on the transmission line conductor, due to varying meteorological
conditions in space and in time, by objectoriented modeling of transmission lines
and weather stations, by introducing the concept of virtual weather sites,* and by
expert rules described in Chapter 8.
4.4.2 D
YNAMIC
A
MPACITY
In the dynamic state, shortterm overload currents greater than steadystate ampacity
are allowed on a transmission line by taking into consideration the energy stored in
a transmission line conductor. The energy stored in a transmission line conductor is
shown by the differential equation (3.5) in Chapter 3. In addition, Section 3 of a
recent Cigré report** presents data on dynamic ampacity. The data was compared
to the values calculated by a LINEAMPS Dynamic model with excellent agreement,
as shown in Figures 4.8 and 4.9.
4.5 MEASUREMENT OF TRANSMISSION LINE
CONDUCTOR TEMPERATURE
345 kv Transmission line
The results obtained from the LINEAMPS program were also veriﬁed by comparison
with the ampacity of a real transmission line by measurement. Measurements were
made by temperature sensors installed on various locations of a 345 kV overhead
transmission line operated by the Commonwealth Edison Company (ComEd) in the
region of Chicago, IL. The results of this comparison are presented in Figure 4.10,
showing that the ampacity of the transmission line calculated by the LINEAMPS
program never exceeded the measured values at all locations during daytime for the
period considered in the study. These results clearly indicate that the program safely
and reliably offers substantial increase in line ampacity over the present method of
static line rating.
As seen in Figure 4.10, LINEAMPS ratings never exceeded measured (ComEd)
ampacity at different hours of the day. It also accurately predicted the lowest value
of line ampacity at noontime. The ampacity predicted by LINEAMPS offers sub
stantially higher line capacity than the present method of static line rating. The static
line ampacity is 1000 A, as shown in Figure 4.10.
* LINEAMPS User Manual, 1998.
** The thermal behaviour of overhead conductors. Section 3: Mathematical model for evaluation of
conductor temperature in the unsteady state. Cigré Working Group 22.12 Report,
Electra
, October 1997.
1306/C04/frame Page 71 Saturday, May 27, 2000 10:23 AM
72
Powerline Ampacity System: Theory, Modeling, and Applications
4.6 CHAPTER SUMMARY
In this chapter the results calculated by the LINEAMPS program are compared to
experimental data from different power company data, as well as transmission line
conductor manufacturers’ catalog data. The results of these comparisons show that
line ampacity calculated by program is in good agreement with actual data from the
ﬁeld. The calculations are also compared to data presented at the IEEE and Cigré
conferences by several researchers. In all of these comparisons, there is excellent
agreement with the results obtained by program. Therefore, according to Faraday,
the proposed theory of transmission line ampacity, conductor thermal models,
hypotheses, and the correctness of various assumptions are validated by verifying
results obtained by program with experimental data.
FIGURE 4.10
Transmission line ampacity calculated by LINEAMPS program is compared
to the ampacity measured on a real 345 kV overhead transmission line operated by Common
wealth Edison Company in the region of Chicago, IL, USA.
11:00
Time of Day, hr
ComED
LINEAMPS
STATIC
0
A
m
p
a
c
i
t
y
,
A
COMPARISON OF LINEAAMPS WITH MEASURED
TRANSMISSION LINE AMPASITY
9:00 10:00 12:00 13:00 14:00 15:00 16:00
500
1000
1500
2000
2500
3000
3500
8:00
1306/C04/frame Page 72 Saturday, May 27, 2000 10:23 AM
73
5
Elevated Temperature
Effects
5.1 INTRODUCTION
The advantages of higher transmission line ampacity discussed in Chapter 1 include
the deferment of the capital investment required for the construction of new lines
and economic energy transfer. As a result of achieving higher line ampacity, elec
tricity costs are reduced and there is less environmental impact. While there are
signiﬁcant beneﬁts to increasing transmission line ampacity, its effects must be
clearly understood and evaluated accurately. In this chapter the effects of higher
transmission line ampacity are evaluated from the point of view of elevated temper
ature operation of conductors. The problem of electric and magnetic ﬁelds due to
higher ampacity are presented in Chapter 6. This chapter includes a study of trans
mission line conductor sag and tension, permanent elongation, and the loss of tensile
strength of the powerline conductor due to elevated temperature operation.
As stated in the previous chapters, the main objective of the powerline ampacity
system is to accurately predict transmission line ampacity based upon actual and
forecast weather conditions. The line ampacity system will ensure that the allowable
normal and emergency operating temperatures of the conductor are not exceeded.
The line ampacity system program should also verify that the loss of tensile strength
of a conductor is within acceptable limits, and that any additional conductor sag
caused by permanent elongation will not exceed design sag and tension during the
lifetime of the transmission line conductor. When line ampacity is increased con
ductor temperature increases, consequently, there may be greater loss of tensile
strength of conductor and higher sag, which must be evaluated properly.
The loss of tensile strength and permanent elongation of a conductor is calculated
recursively from a speciﬁed conductor temperature distribution by using the empirical
equations found in the literature (Harvey, 1972; Morgan, 1978; Cigré, 1978; Deb et
al., 1985; Mizuno, 1998). The Cigré report did not describe how the temperature
distribution was obtained, and did not include the design sag and tension of conductor.
An elegant method to calculate sag and tension by a strain summation procedure is
described in a report prepared by Ontario Hydro.* The unique contribution made in
this chapter is the development of an uniﬁed approach to determine sag and tension
during the lifetime of a transmission line conductor by consideration of the probability
distribution of transmission line conductor temperature.
* Development of an accurate model of ACSR conductors at high temperatures. Canadian Electricity
Association Research Report.
1306/C05/frame Page 73 Saturday, May 27, 2000 11:17 AM
74
Powerline Ampacity System: Theory, Modeling, and Applications
5.1.1 E
XISTING
P
ROGRAMS
Most sagtension computer programs presently used are based on the assumption
that conductor temperature will remain constant for the entire life of the line. In
reality, as we all know, conductor temperature is never constant. For example,
conductor sag at 100 °C cannot be expected to remain the same if it has been operated
at that temperature for 100 hours or 10,000 hours. Therefore, sag based on the
probability distribution of conductor temperature is required. Conductor sag and
tension are important transmission line design parameters upon which depend the
security of the line. A line security analysis was carried out (Hall, Deb, 1988) based
upon different line operating conditions. This study showed how conductor sag and
tension varies with conductor temperature frequency distributions.
5.2 TRANSMISSION LINE SAG AND TENSION —
A PROBABILISTIC APPROACH
A method of calculation of conductor sag and tension is presented in this section
by consideration of the probability distribution of transmission line conductor tem
perature in service. The probability distribution of conductor temperatures is
obtained by the synthetic generation of meteorological data from timeseries sto
chastic and deterministic models. This method of generating probability distribution
of conductor temperatures takes into account the correlation between the meteoro
logical variables and the transmission line current.*
The effects of elevated temperature operation of conductors comprising inelastic
elongation and the loss of tensile strength of conductor are considered by the
recursive formulation of inelastic elongation and annealing models found in the
literature. The equations and algorithm that were used to calculate conductor sag
and tension from the probability distribution of conductor temperatures are presented
and implemented in a computer program. Results are presented that show good
agreement with data from other computer programs.
There is considerable interest in the industry in the probabilistic design of
overhead lines.** Ghanoum (1983) described a method for the structural design of
transmission lines based upon probabilistic concepts of limit loads and return period
of wind. Probabilitybased transmission line rating methods are described by several
authors (Koval and Billinton, 1970; Deb et al., 1985, 1993; Morgan, 1991; Redding,
1993; Urbain, 1998). Redding*** 1993 presented probability models of ambient
temperature and wind speed, but the resulting conductor temperature distribution
was not given. Not much attention has been given to probabilistic design of sag and
tension of overhead line conductors, which depend upon conductor temperature
probability distributions. The probability distribution of conductor temperature is a
* See discussion contribution by J.F. Hall and Anjan K. Deb on the IEEE paper (Douglass, 1986).
** Ghanoum, E., “Probabilistic Design of Transmission Lines,” Part I, II.
IEEE Transactions on Power
Apparatus and Systems
, Vol. PAS102, No. 9, 1983.
*** J.L. Redding, “A Method for Determining Probability Based Allowable Current Ratings for BPA’s
Transmission Lines, “IEEE/PES 1993 Winter Meeting Conference Paper # 93WM 0778PWRD, Colum
bus, Ohio, January 31–February 5, 1993.
1306/C05/frame Page 74 Saturday, May 27, 2000 11:17 AM
Elevated Temperature Effects
75
function of conductor current and the meteorological conditions on the line (Hall
and Deb, 1988; Morgan, 1991; Mizuno et al., 1998, 2000). Mizuno et al. considered
the loss of strength of conductor as the index of thermal deterioration. In this chapter
I have considered transmission line sag and tension as the determining factor. This
includes both the loss of strength and the permanent elongation of the conductor.
5.2.1 T
HE
T
RANSMISSION
L
INE
S
AG
T
ENSION
P
ROBLEM
Given the maximum mechanical loading of a conductor due to wind and ice, and a
probable distribution of conductor temperature with time, it is required to calculate
the sag and tension of the conductor at different temperatures. The probability
distribution of conductor temperature is obtained from line ampacity simulations.
Transmission line sag and tension are critical line design parameters that are required
to verify conductortoground clearance and the safety factor of the conductor at the
maximum working tension.
5.2.2 M
ETHODOLOGY
In order to predict conductor sag at the highest conductor temperature, the permanent
elongation of the conductor due to metallurgical creep is required to be estimated.
This requires an estimate of the conductor temperature distribution during the
expected life of the line. Future conductor temperature distributions require knowl
edge of the line current as well as the meteorological conditions. The conductor
temperature distribution is also required to calculate the loss of strength of the
conductor.
Future meteorological conditions may be estimated by the random generation
of weather data from their speciﬁed probability distributions, or by taking a typical
set of weather data and assuming it to repeat itself every year (Giacomo et al., 1979).
In the method proposed by the author (Deb, 1993), meteorological data is generated
by Monte Carlo simulation of the following time series stochastic and deterministic
model:
Y(t) = X(t)
T
· A(t) +
η
(t) (5.1)
Y(t)
∈
(Ta, Nu, Sr) input variables*
X(t)
T
= {1, Sin(
ω
t), Sin(2
ω
t), Cos(
ω
t), Cos(2
ω
t), z(t1), z(t2)}
A(t) = model coefﬁcients
ω
= fundamental frequency
z(t1), z(t2) are the stochastic variables at lag 1 and lag 2 respectively
η
(t) = uncorrelated white noise
By this method it is possible to take into consideration timeofday effects of weather
and line current in the analysis. Using real weather data in chronological order allows
* Ta = ambient temperature, Nu = Nusselt number (coefﬁcient of heat transfer), Sr = solar radiation (Sr
is also calculated analytically by the method given in Chapter 7).
1306/C05/frame Page 75 Saturday, May 27, 2000 11:17 AM
76
Powerline Ampacity System: Theory, Modeling, and Applications
this also. Timeofday effects are lost when weather data is generated from known
probability distributions (Mizuno et al., 1998), (Deb et al., 1985).
AAC Bluebell conductor temperature distribution is shown in Figure 5.5 which
was obtained by the synthetic generation of California meteorological data from
time series stochastic models. Examples of synthetic generations of meteorological
data from time series stochastic models are shown in Figures 5.1–5.4.
The conductor temperature distribution of Figure 5.5 assumes constant line
current equal to the static line rating and is used to calculate transmission line
conductor sag and tension.
FIGURE 5.1
Hourly values of ambient temperature generated by Monte Carlo simulation
for the San Francisco Bay area.
FIGURE 5.2
Hourly values of conductor heat transfer coefﬁcient of AAC Bluebell trans
mission line conductor generated by Monte Carlo simulation for the San Francisco Bay area.
MonteCarlo Simulation of
Ambient Temperature
T
e
m
p
e
r
a
t
u
r
e
,
D
e
g
r
e
e
C
45
40
35
30
25
20
15
10
5
0
0 24 48 72 96 120 144 168
Hour
MonteCarlo Simulation of Heat Transfer Coefficient
N
u
s
s
e
l
t
N
u
m
b
e
r
Hour
0 24 48 72 96 120 144 168
120
100
80
60
40
20
0
1306/C05/frame Page 76 Saturday, May 27, 2000 11:17 AM
Elevated Temperature Effects
77
5.2.3 C
OMPUTER
P
ROGRAMS
The following computer programs are required for the prediction of sag and tension
at high temperature based upon the probabilistic distribution of conductor temperature:
1. Conductor temperature predictor
2. Probability distribution generator of ambient temperature, line current,
solar radiation, and conductor heat transfer coefﬁcient from timeseries
stochastic models
3. Probability distribution generator of conductor temperatures
4. Inelastic elongation (creep) predictor
5. Loss of strength predictor
6. Sag and Tension Calculator
FIGURE 5.3
Hourly values of solar radiation on conductor surface of AAC Bluebell trans
mission line conductor in the San Francisco Bay area simulated by program.
FIGURE 5.4
Hourly values of conductor temperature of AAC Bluebell transmission line
conductor in the San Francisco Bay area using simulated weather data.
Solar Radiation Simulation
Hour
S
o
l
a
r
R
a
d
i
a
t
i
o
n
,
W
/
m
2
0 24 48 72 96 120 144 168
1400
1200
1000
800
600
400
200
0
Conductor Temperature Simulation
T
e
m
p
e
r
a
t
u
r
e
,
D
e
g
r
e
e
C
Hour
80
70
60
50
40
30
20
10
0
0 24 48 72 96 120 144 168
1306/C05/frame Page 77 Saturday, May 27, 2000 11:17 AM
78
Powerline Ampacity System: Theory, Modeling, and Applications
The above modules constitute the Sag and Tension Program developed by the author.
A ﬂow chart for the calculation of transmission line sag and tension from the
probability distribution of conductor temperature is given in Figure 5.7. The equa
tions and algorithms are developed in the following sections.
5.3 CHANGE OF STATE EQUATION*
The change of state equation given below is used for the calculation of transmission
line sag and tension. If the conductor is at State 1 given by conductor stress
σ
1
and
temperature Tc
1
and goes to State 2 given by stress
σ
2
and temperature Tc
2
,
then
the new sag and tension of the conductor at State 2 is calculated from the following
change of state equation:
FIGURE 5.5
Frequency distribution of AAC Bluebell transmission line conductor temper
ature in the San Francisco Bay area simulated by program.
FIGURE 5.7
Flow chart for the calculation of transmission line conductor sag and tension
from probability distribution of conductor temperature.
* P. Hautefeuille, Y. Porcheron, Lignes Aeriennes,
Techniques de l’Ingenieur
, Paris.
J.P. Bonicel, O. Tatat, Aerial optical cables along electrical power lines,
REE
No. 3, March 1998, SEE
France.
20
15
10
5
0
10 15 20 25 30 35 40 45 50 55 60 65 70 75
Temperature, Degree C
F
r
e
q
u
e
n
c
y
,
%
Conductor Temperature
Frequency Distribution
MonteCarlo Simulation of
Ambient Temperature
Tem
perature, D
egree C
45
40
35
30
25
20
15
10
5
0
0 24 48 72 96 120 144 168
Hour
Heat Transfer Simulation
N
u
sse
lt N
u
m
b
e
r
Hour
0 24 48 72 96 120 144 168
120
100
80
60
40
20
0
Solar Radiation Simulation
Hour
S
o
la
r R
a
d
ia
tio
n
, W
/m
2
0 24 48 72 96 120 144 168
1400
1200
1000
800
600
400
200
0
Conductor Temperature Simulation
Tem
perature, D
egree C
Hour
80
70
60
50
40
30
20
10
0
0 24 48 72 96 120 144 168
20
15
10
5
0
10 15 20 25 30 35 40 45 50 55 60 65 70 75
Temperature, Degree C
F
re
q
u
e
n
cy, %
Conductor Temperature
Frequency Distribution
Transmission
Line Sag &
Tension
1306/C05/frame Page 78 Saturday, May 27, 2000 11:17 AM
Elevated Temperature Effects
79
(5.2)
σ
1
,
σ
2
= stress at State1 and State2 respectively, kg/mm
2
Tc
1
, Tc
2
= conductor temperature at State1 and State2, °C
E = Young’s modulus of elasticity, kg/mm
2
ϖ
= speciﬁc weight of conductor, kg/m/mm
2
L = span length, m
∆
Ec = inelastic elongation (creep) mm/mm
α
= coefﬁcient of linear expansion of conductor, °C
–1
Conductor sag is calculated approximately by the following wellknown parabola
equation:
(5.3)
where,
Sag is in meters, m
W = conductor weight, kg/m
T = conductor tension, kg
The above equation is used to calculate transmission line conductor sag and tension.
It requires a knowledge of Young’s modulus of elasticity, the coefﬁcient of linear
expansion of the conductor and the permanent elongation of the conductor due to
elevated temperature creep. Young’s modulus is obtained from the stress/strain curve
shown in Figure 5.6. The coefﬁcient of linear expansion is a property of the con
ductor. The elevated temperature creep Ec (metallurgical creep) is estimated sepa
rately by the creep predictor program.
5.3.1 R
ESULTS
The probability distribution of conductor temperature is shown in Table A5.1 in
Appendix 5 at the end of this chapter, the loss of strength is given in Table A5.2,
and the permanent elongation of conductor during the lifetime of a transmission line
FIGURE 5.6
Stress/strain relation of AAC conductor
σ ϖ
σ
α
σ ϖ
σ
2
2
2
2 2 1
1
2
1
2
24 24 E
L
Tc Tc Ec
E
L
– – –
⋅ ( )
+
( )
+ =
⋅ ( )
∆
Sag
WL
T
=
2
8
Strain
O
E1 E2
Stress
A
B
1306/C05/frame Page 79 Saturday, May 27, 2000 11:17 AM
80
Powerline Ampacity System: Theory, Modeling, and Applications
conductor is presented in Table A5.3. Calculations are based upon the conductor
temperature distributions generated from Figure 5.5. The sag and tension of trans
mission line conductors is presented in Table A5.6 of Appendix 5. Comparison of
sag and tension results with other programs are given in the Tables A5.8 through
A5.10. Further results of transmission line sag and tension of special conductors are
given in Table A5.11 (Choi et al., 1997). The study was prepared by the author with
KEPCO* for line uprating with highampacity conductors. The results obtained by
program show excellent agreement with ﬁeld measurements on an actual transmis
sion line in Korea.
5.3.2 C
ONDUCTOR
S
TRESS
/S
TRAIN
R
ELATIONSHIP
The stres/strain relationship of an all aluminum conductor is shown in the Figure
5.6 for the purpose of illustration of some of the concepts discussed in this section.
The stress/strain relationship of an ACSR conductor is somewhat complicated,
though the general concepts remain the same for any type of conductor.
When tension is applied to an unstretched conductor, the ratio stress/strain of
the conductor follows the curve OA. When the tension is lowered at Point A, this
ratio becomes linear and follows the trace AE1. If the tension is increased again it
follows the linear path E1A until it reaches Point A. If the tension is further increased
at Point A it becomes nonlinear again, as shown by the curve AB. When tension is
lowered again at Point B, it follows the linear path BE2. In Figure 5.6, the section
of the stress/strain curve OA and AB represents the initial stress/strain curve, which
is nonlinear. Consequently, the Young’s modulus in this region becomes nonlinear
and is generally approximated by ﬁtting a polynomial function of degree N to the
data. The ﬁnal modulus of elasticity given by the slope of the linear portion of the
curves AE1 and BE2 is constant.
The sections OE1 and E1E2 are the permanent elongation due to creep (geo
metrical settlement). The advantage of pretensioning the conductor becomes obvious
from Figure 5.6. The permanent stretch OE1 and E1E2 can be removed if the
conductor is pretensioned by a load to reach Point B close to the allowed maximum
working tension of the line. The stress/strain curve then becomes linear.
5.4 PERMANENT ELONGATION OF CONDUCTOR
Permanent or irreversible elongation of the conductor is known to occur due to
elevated temperature operation of the conductor. It results in increase of conductor
sag and reduces the midspan clearance to ground. Elevated temperature creep is a
function of the conductor temperature, its duration, and the conductor tension. Two
factors cause permanent elongation of the conductor (Cigré, 1978):
1. Geometric settlement
2. Metallurgical creep
* Korea Electric Power Company.
1306/C05/frame Page 80 Saturday, May 27, 2000 11:17 AM
Elevated Temperature Effects
81
5.4.1 G
EOMETRIC
S
ETTLEMENT
Geometric settlement depends upon conductor stringing tension and occurs very
rapidly as it only involves the settling down of strands. Generally, the process starts
with conductor stringing and is completed within 24 hours. Higher than normal
stringing tensions, “pretension,” is sometimes applied to a conductor to accelerate
the process of geometric settlement. The geometric settlement, Es, is calculated by
the following formula (Cigré, 1978):
Es = 750(d – 1) (1 – exp(–m/10)) (MWT/RTS)
2.33
(5.4)
d = wire diameter, mm
m = aluminum/steel sectional area, ratio
MWT = Maximum Working Tension, kg
RTS = Rated Tensile Strength, kg
5.4.2 M
ETALLURGICAL
C
REEP
Metallurgical creep is a function of conductor temperature, tension, and time. There
fore, elevated temperature operation of a line for short duration is not as much of a
concern as continuous operation at high temperatures. Metallurgical creep is estimated
by using the following empirical formula determined experimentally (Cigré, 1978):
(5.5)
E
c
= elongation, mm/km
K,
φ
,
α
,
µ
,
δ
are constants (Table 5.1 gives values for typical ACSR conductor
sizes)
Tc = average conductor temperature, °C
σ
= average conductor stress, kg/mm
2
t = time, hr
The factor
β
takes into account the effect of conductor type, stranding and material
and is calculated as follows,
(5.6)
N = number of aluminum strands
n
i
= number of wires in layer i
β
i
= angle of the tangent in a point of a wire in layer i with conductor axis
E K T t
c c
= ⋅
( )
+
1
2
cos
exp
α
α µ σ
β
φ σ
δ
β
β
=
=
=
∑
∑
n
n
i i
i
N
i
i
N
1
1
1306/C05/frame Page 81 Saturday, May 27, 2000 11:17 AM
82
Powerline Ampacity System: Theory, Modeling, and Applications
5.4.3 R
ECURSIVE
E
STIMATION
OF
P
ERMANENT
E
LONGATION
The inelastic elongation of a conductor due to metallurgical creep, E
c
, is calculated
in small steps,
∆
E
c
, by the following recursive equations:
Ec
i,j
= Ec
i,j1
+
∆
Ec (5.7)
(5.8)
i = 1,2… n line loading intervals obtained from the frequency distribution of
Table A5.1
j = 1,2… f sub intervals
tq
i,j
= equivalent time at present temperature Tc
i
for the past creep Ec
i,j1
(5.9)
Total creep is estimated by summation,
E
total
= Ec
n,f
+ Es
f
(5.10)
Ec
n,f
and Es
f
are the ﬁnal inelastic elongation due to metallurgical creep and geo
metrical settlement of the transmission line conductor.
Results are presented in Table A5.3 of Appendix 5 for the frequency distribution
of Table A5.1. When Table A5.3 values are used to calculate sag, the results are
given in Table A5.6. When ﬁnal sags are compared to the initial conditions speciﬁed
in Table A5.5, it is seen that the sag of the AAC Bluebell conductor increases by
5.9 ft (14.4% increase) over initial conditions after 30 years. When maximum
conductor temperature is 75°C, the increase in sag is only 10.25% after 30 years.
Transmission line design conditions will determine whether the increase in sag or
loss of strength is the limiting factor for a particular transmission line.
TABLE 5.1
Value of coefﬁcients in equation (5.5)
(Cigré, 1978)
ACSR Al/St K f a m d
54/7 1.1 0.0175 2.155 0.342 0.2127
30/7 2.2 0.0107 1.375 0.183 0.0365
tq Ec K Tc
i j i j i j i . , – ,
–
–
exp – = ′
( )
{ }
1
σ φ
α
µ σ
δ
′ =
+
K
Cos
K
2 α
β
1306/C05/frame Page 82 Saturday, May 27, 2000 11:17 AM
Elevated Temperature Effects 83
5.5 LOSS OF STRENGTH
The loss of tensile strength of a conductor results in lowering the design safety
factor of the conductor. Generally, T/L conductor tension is designed with a safety
factor of two at the worst loading condition.* The worst condition results in the
conductor being subjected to the maximum tension. Such conditions arise when
the conductor is exposed to high winds and/or ice covering. To give an example,
an ACSR Cardinal conductor under extreme wind (~ 100 mph) and ice loading
may result in the tension of the Cardinal conductor to reach 17,000 lbf which is
approximately 50% of the rated tensile strength of the conductor. Therefore, a
10% reduction in the tensile strength of the conductor would also lower the safety
factor of the conductor by 10%. Generally, a loss of strength up to 10% is
acceptable (Mizuno et al., 1998).
5.5.1 PERCENTILE METHOD
A recent study (Mizuno et al., 1998) describes the calculation of thermal deteriora
tion of a transmission line conductor by a probability method. The reduction in
tensile strength of the conductor was used as the index of thermal deterioration. The
loss of tensile strength is calculated as a function of conductor temperature, Tc, and
the time duration, t, at which the temperature, Tc, is sustained (Morgan, 1978);
(Harvey, 1972).
W = exp{C(ln t – A – BT)} (5.11)
W is loss of strength (%) and A,B,C are constants that are characteristics of the
conductor. This is an empirical equation based upon laboratory tests on individual
wire strands. The total loss of strength is then obtained by,
(5.12)
where t
i
is time duration when the conductor temperature is T
i
and t
i
is given by,
(5.13)
5.5.2 RECURSIVE ESTIMATION OF LOSS OF STRENGTH
The author has developed a recursive method for the calculation of loss of strength
of conductors as follows:
W = W
a
[1 – exp{–exp(A
3
+ B
3
T
c
+ n
1
ln t + Kln (R/80)}] (5.14)
Wi = Wa[1 – exp{–exp(A3 + B3Tci + n
1
ln(ti + tqWi1) + Kln (R/80)}] (5.15)
* National Electrical Safety Code C21997.
W t t t t t t t t t t t
n n n n n
c
∑
=
( )
+
( )
+
( )
)
+
)
[ ] − −
K L
1 2 1 2 3 2 1 1
ln ln t A BT
i
= +
1306/C05/frame Page 83 Saturday, May 27, 2000 11:17 AM
84 Powerline Ampacity System: Theory, Modeling, and Applications
tqW
i1
= exp[ln ln{1/(1 – W
i1
/W
a
)} + A
3
+ B
3
T
ci
+ n
1
ln(t
i
+ tqW
i1
) + Kln (R/80)}]/n
i
(5.16)
A
3
, B
3
, n
1
, K = constants given in Table 5.2 (Morgan. 1978).
R = Reduction of wire by drawing from rod to wire (Morgan, 1978).
(5.17)
D, D
o
are the diameters of wire and rod, respectively
i = 1, 2, 3….n intervals of time
tqW
i–1
= equivalent time for loss W
i–1
at temperature Tc
i
Results are presented in Appendix 5 in Table A5.2 for the frequency distribution of
Table A5.1. These results show that the loss of tensile strength for the AAC Bluebell
conductor is greater than 10% when the maximum temperature of the conductor is
95°C and the life is 25 years. For this reason, All Aluminum Conductors (AAC) are
generally operated below 90°C under normal conditions.*
5.6 CHAPTER SUMMARY
When transmission line ampacity is increased it is necessary to properly evaluate
the thermal effects of the powerline conductors, which includes loss of tensile
strength of the conductor, permanent elongation, and conductor sag. In this chapter
a uniﬁed approach to modeling and evaluation of the elevated temperature effects
of transmission line conductors is presented. Conductor loss of strength and perma
nent elongation are evaluated recursively from the probability distribution of con
ductor temperature. The probability distribution of conductor temperature is gener
ated by Monte Carlo simulation of weather data from timeseries stochastic models
and transmission line current. Therefore, a new method is developed to determine
the sag and tension of overhead line conductors with elevated temperature effects.
TABLE 5.2
Value of Coefﬁcients in Equations
(5.15), (5.16) (Morgan 1978)
Wire A
3
B
3
n
1
K
Aluminum –8.3 0.035 0.285 9
Aluminum Alloy –14.5 0.060 0.79 18
Copper –7.4 0.0255 0.40 11
* PG&E Line Rating Standard.
R
D
D
o
= [ −
100 1
2
1306/C05/frame Page 84 Saturday, May 27, 2000 11:17 AM
Elevated Temperature Effects 85
A study of the AAC Bluebell conductor is presented to show the longterm
effects of elevated temperature. Results are presented which clearly show that the
loss of strength of the conductor is less than 10% when the maximum temperature
of the conductor is 90°. The increase in sag is less than 15% for the same maximum
operating temperature.
The accuracy of the sagtension program was tested with ALCOA’s Sag and
Tension program, Ontario Hydro’s STESS program, and KEPCO’s transmission line
ﬁeld data. Results are presented that compare well with all of the above data.
1306/C05/frame Page 85 Saturday, May 27, 2000 11:17 AM
87
Appendix 5
Sag and Tension Calculations
FREQUENCY DISTRIBUTION OF
CONDUCTOR TEMPERATURE
A conductor temperature distribution is shown in Table A5.1. This temperature
distribution was generated by assuming constant load current equal to the static line
rating, and by the artiﬁcial generation of meteorological data. From the basic tem
perature distribution assumption shown in Column 1 of Table A5.1, the conductor
temperatures were increased in steps of 5°C up to the maximum temperature of
100°C. The corresponding frequency distributions of conductor temperature are also
shown in this table.
TABLE A5.1
Frequency Distribution of Conductor Temperature
Maximum Conductor Temperature, °C
75 80 85 90 95 100 Frequency,%
10 15 20 25 30 35 2.0
15 20 25 30 35 40 11.0
20 25 30 35 40 45 13.0
25 30 35 40 45 50 14.0
30 35 40 45 50 55 16.0
35 40 45 50 55 60 15.0
40 45 50 55 60 65 12.5
45 50 55 60 65 70 8.0
50 55 60 65 70 75 4.0
55 60 65 70 75 80 2.0
60 65 70 75 80 85 1.5
65 70 75 80 85 90 0.5
70 75 80 85 90 95 0.3
75 80 85 90 95 100 0.2
Σ
100%
Table A5.1 provides the starting point in the calculation of loss of strength
and permanent elongation of conductor. The sag and tension of the conductor
are then calculated from this data. The factor of safety of the conductor and
the line to ground clearances can now be veriﬁed.
1306/appendix 5.1/frame Page 87 Saturday, May 27, 2000 9:17 AM
88
Powerline Ampacity System: Theory, Modeling, and Applications
L
OSS
OF
T
ENSILE
S
TRENGTH
OF
C
ONDUCTOR
Based upon the distributions shown in Table A5.1, the loss of strength of the
conductor was calculated at different conductor maximum temperatures for the
conductor life, from 10 to 30 years. The results are shown in Table A5.2 for the
AAC Bluebell conductor.
P
ERMANENT
E
LONGATION
OF
C
ONDUCTOR
The permanent elongation of the AAC Bluebell conductor due to metallurgical creep
calculated by the program is shown in Table A5.3 and is based on the same assump
tions of conductor temperature distribution as shown in Table A5.1.
The loss of strength and permanent elongation of the conductor shown in the
Tables A5.2 and A5.3, respectively are used as input data to the sag and tension
program. The sag and tension of the 1034 Kcmil AAC Bluebell is given in Table A5.4.
TABLE A5.2
Loss of Strength of AAC Bluebell Calculated by Program
Conductor Life, Year
Maximum Conductor
Temperature °C 10 5 20 25 30
Loss of Strength, %
75 4.3 4.8 5.2 5.5 5.8
80 5.1 5.6 6.1 6.5 6.8
85 6.0 6.7 7.2 7.6 8.0
90 7.0 7.8 8.5 9.0 9.4
95 8.3 9.2 9.9 10.5 11.0
100 9.7 10.8 11.6 12.3 12.9
TABLE A5.3
Permanent Elongation of AAC Bluebell Calculated by Program
Life of Conductor, Year
Maximum Conductor
Temperature °C 10 15 20 2 30
Permanent Elongation, Micro Strain
75 809 859 899 939 969
80 889 959 999 1049 1079
85 989 1069 1119 1169 1209
90 1109 1189 1249 1289 1329
95 1239 1309 1369 1419 1459
100 1359 1449 1529 1579 1619
1306/appendix 5.1/frame Page 88 Saturday, May 27, 2000 9:17 AM
Appendix 5 Sag and Tension Calculations
89
S
AG
AND
T
ENSION
C
ALCULATION
BY
P
ROGRAM
TABLE A5.4
Input Data
1034 Kcmil AAC Bluebell
Conductor diameter, in 1.17
Mass, lb/ft 0.97
Rated Tensile Strength, lbf 18500
Modulus of elasticity, psi 8.5·10
6
Coefﬁcient of linear expansion, /°C 23·10
–6
Sectional area, in
2
0.81
Final unloaded tension, lbf 3700
Final unloaded temperature, °F 50
TABLE A5.5
Sag and Tension Initial Condition after SaggingIn
Span
ft
Wind
lb/in2
Ice
in
Tc
°F
Tc
°C
Tension
lb
SF
#
Sag
ft
1000 8.0 0 25 –3.9 4972 3.7 31.9
1000 0 0 60 15.6 3641 4.9 33.4
1000 0 0 167 75.0 3045 5.8 40.0
1000 0 0 176 80.0 3006 5.9 40.5
1000 0 0 185 85.0 2969 5.9 41.0
1000 0 0 194 90.0 2934 5.9 41.5
1000 0 0 203 95.0 2899 5.9 42.0
1000 0 0 212 100.0 2866 5.9 42.5
TABLE A5.6
Final Sag
Life, year
Temperature °C 10 15 20 25 30
Sag, ft
75 43.4 43.6 43.8 44.0 44.1
80 44.2 44.5 44.7 44.9 45.0
85 45.1 45.4 45.6 45.8 46.0
90 46.0 46.4 46.6 46.7 46.9
95 47.0 47.3 47.5 47.7 47.8
100 47.9 48.2 48.5 48.7 48.9
1306/appendix 5.1/frame Page 89 Saturday, May 27, 2000 9:17 AM
90
Powerline Ampacity System: Theory, Modeling, and Applications
S
AG
AND
T
ENSION
C
OMPARISON
WITH
STESS P
ROGRAM
OF
O
NTARIO
H
YDRO
To verify the accuracy of the results obtained from the sag and tension program, it
was compared to the STESS program. The results are provided below.
TABLE A5.7
Input
ACSR Drake
Conductor diameter, mm 28.13
Mass, kg/m 1.628
Rated Tensile Strength, kN 140.1
Modulus of elasticity, kg/mm 8360
Coefﬁcient of linear expansion /°C 19·10
6
Sectional area, mm
2
468.7
Final unloaded tension, kN 27.80
Final unloaded temperature, °C 20
TABLE A5.8
Sag and Tension Comparison with STESS Program of
Ontario Hydro
Span
ft
Wind
Pa
Tc
°C
Tension
kN
Safety
Factor, #
Sag, m
STESS
Sag, m
Program*
300 0 20 28.0 5.0 6.47 6.4
300 0 30 26.4 5.3 6.84 6.8
300 0 40 25.1 5.6 7.20 7.2
300 0 50 23.8 5.9 7.56 7.5
300 0 60 22.8 6.1 7.92 7.9
300 0 70 21.8 6.4 8.27 8.2
300 0 80 21.0 6.7 8.62 8.6
300 0 90 20.2 6.9 8.96 8.9
300 0 100 19.5 7.2 9.27 9.3
300 0 110 18.8 7.4 9.45 9.5
300 0 120 18.3 7.7 9.64 9.6
300 0 130 17.7 7.9 9.82 9.8
300 0 140 17.2 8.1 10.01 9.9
300 0 150 16.8 8.4 10.20 10.1
Note
:
1 Pa = 0.02 lb/ft
2
1 kN = 225.8 lbf
Wind = Wind Pressure on Projected Area of Conductor, Pa
Tc = Average Conductor Temperature, °C
*Calculated by the sag and tension program described in this chapter.
1306/appendix 5.1/frame Page 90 Saturday, May 27, 2000 9:17 AM
Appendix 5 Sag and Tension Calculations
91
S
AG
AND
T
ENSION
C
OMPARISON
WITH
ALCOA P
ROGRAM
*
Results obtained by new program are compared to the ALCOA Sag and Tension
Program (see footnote).
* Craig B. Lankford, ALCOA’s Sag and Tension Program Enhanced for PC Use,
Transmission and
Distribution Journal
, Vol. 41, No. 11, November 1989.
TABLE A5.9
Input (ALCOA Program Data)
795 AS33 ACSR DRAKE (26 al + 7 st)
Conductor diameter, in 1.108
Mass, lb/ft 1.0940
Rated Tensile Strength, lb 31500
Modulus of elasticity, psi 11.89·10
6
Coefﬁcient of linear expansion, /°C 19.5·10
–6
Sectional area, in
2
0.7264
Final unloaded tension, lbf 4403
Final unloaded temperature, °F 60
TABLE A5.10
Sag and Tension Comparison with ALCOA Program
Span
ft
Wind
lb/ft2
T
°Fc
Tension
lbf
Safety
Factor, SF
Sag, ft
ALCOA
Sag, ft
Program*
750 33.0 60 11496 2.7 21.8 21.9
750 0 70 4293 7.1 18.0 18.0
750 0 80 4171 7.3 18.5 18.5
750 0 90 4057 7.5 19.0 19.0
750 0 100 3951 7.7 19.5 19.5
750 0 110 3852 7.9 20.0 20.0
750 0 120 3784 8.1 20.5 20.3
750 0 170 3578 9.0 21.5 21.5
750 0 205 3450 9.7 22.3 22.3
Wind = Wind Pressure on Projected Area of Conductor, lb/ft
2
Tc = Average Conductor Temperature, °F
* Calculated by the sag and tension program described in this chapter.
1306/appendix 5.1/frame Page 91 Saturday, May 27, 2000 9:17 AM
92
Powerline Ampacity System: Theory, Modeling, and Applications
TABLE A5.11
Sag and Tension Comparison with KEPCO Line Data
154 kV Double Circuit Line
INPUT ACSR STACIR
Conductor dia, mm 25.30 25.30
Mass, kg/m 1.30 1.30
Rated Tensile Strength, kN 98.00 98.00
Modulus of Elasticity, kg/mm
2
mm 8360.00 16500.00
Coeff. of lin. expan. /deg C 19E·06 3GE·07
Sectional Area, mm
2
mm 379.60 379.60
Final unloaded tension, kN 24.50 24.50
Final unloaded temperature, /deg C 10.00 10.00
OUTPUT
ACSR
STACIR
Span,
m
Wind
Pa
Tc
Deg C
Tension
kN
Sag
m
Tension
kN
Sag
m
300 400 10 29.2 6.3 29.2 6.3
300 0 10 24.5 5.9 24.5 5.9
300 0 90 16.4 8.8 18.0 8.2
300 0 150 NA NA 17.1 8.4
300 0 200 NA NA 16.4 8.7
300 0 210 NA NA 16.3 8.8
300 0 240 NA NA 16.0 9.0
NA = Not Applicable
Table adapted from KEPCO highampacity transmission line (Choi et al., 1997).
1306/appendix 5.1/frame Page 92 Saturday, May 27, 2000 9:17 AM
93
6
Transmission Line Electric
and Magnetic Fields
6.1 INTRODUCTION
The magnetic ﬁeld of a transmission line increases with line ampacity, and increases
at ground level with conductor sag. Typical powerline conﬁgurations are evaluated
to show their magnetic ﬁelds. It is shown that the magnetic ﬁeld of overhead
transmission lines is within acceptable limits. Line designs and EMF mitigation
methods are developed to lower transmission line magnetic ﬁelds in sensitive areas.
Electric ﬁeld limit at ground level is not exceeded by higher transmission line
ampacity if the maximum design temperature of the conductor is not exceeded and
minimum conductortoground distance is maintained.
Even though transmission line voltage remains unchanged with higher ampacity,
electric ﬁelds are also evaluated since the level of an electric ﬁeld at ground is
affected by conductor sag. It must be mentioned that the lowering of transmission
line conductor to ground distance due to higher sag will raise the level of electric
and magnetic ﬁelds at ground level. Therefore, it is important to calculate transmis
sion line sag accurately for the calculation of electric and magnetic ﬁelds at ground
level.
The study of transmission line magnetic ﬁelds is also important from the point
of view of transmission line ampacity. In the previous chapters it was shown that
transmission line capacity may be increased by dynamic line ratings, which will
result in the lowering of the cost of electricity. On the other hand, increasing line
ampacity also increases the level of magnetic ﬁeld. With public concern for electric
and magnetic ﬁelds, transmission line engineers are required to accurately evaluate
the impact of increased line capacity on the environment due to higher electric and
magnetic ﬁelds.
6.2 TRANSMISSION LINE MAGNETIC FIELD
The magnetic ﬁeld of a currentcarrying transmission line conductor is calculated
by the application of Maxwell’s equation. The Electric Power Research Institute
(EPRI) conducted a study* in which they proposed methods for the reduction of
transmission line magnetic ﬁelds. The study presents data to quantify more accu
rately the magnetic ﬁelds of different transmission line conﬁgurations.
* V.S. Rashkes, R. Lordan, “Magnetic Field Reduction Methods: Efﬁciency and Costs,”
IEEE Transactions
on Power Delivery
, Vol. 13, No. 2, April 1998.
1306/C06/frame Page 93 Saturday, May 27, 2000 11:26 AM
94
Powerline Ampacity System: Theory, Modeling, and Applications
In this section, a general method of calculation of the magnetic ﬁeld of overhead
transmission line conductors is presented. This method is suitable for any transmis
sion line conﬁguration. The magnetic ﬁelds of typical transmission line conﬁgura
tions are also presented to show that magnetic ﬁelds are within acceptable limits. It
is shown that there is minimum environmental impact due to higher transmission
line ampacity. EMF mitigation methods are also given.
6.2.1 T
HE
M
AGNETIC
F
IELD
OF
A
C
ONDUCTOR
In the following section we shall derive expressions for the calculation of the
magnetic ﬁeld of a currentcarrying conductor by the application of Maxwell’s
equations.
(6.1)
(6.2)
(6.3)
(6.4)
Hr, H
ϕ
, Hz are the components of the magnetic ﬁeld, H, along r,
ϕ
, and z axes, and
is current density. From equations (6.1) through (6.4) the following solutions are
obtained for the calculation of the magnetic ﬁeld of a currentcarrying conductor.
Considering a simple case of an inﬁnitely long cylindrical conductor carrying a
direct current density, (A/m
2
), as shown in Figure 6.1, we have,
The current I through the conductor is,
I = J
z
π
R
2
(6.5)
where R is conductor radius.
∇ × ( ) ⋅
∂
∂
∂
∂
H
r
Hz H
z
J
r r
1
ϕ
ϕ
–
∇ × ( ) ⋅
∂
∂
∂
∂
H
r
Hz
z
Hz
r
J
ϕ φ
1
–
∇ × ( ) +
∂
∂
∂
∂
H
r
H
H
r
Hr
J
z z
1
ϕ
ϕ
ϕ
–
∇⋅ +
∂
∂
+ ⋅
∂
∂
+
∂
∂
H
r
Hr
Hr
r r
H Hz
z
1 1
0
φ
φ
r
J
r
J
∂
∂
( )
∂
∂
( )
ϕ
0
0
due to circular symmetry
due to infinitely long conductor
z
1306/C06/frame Page 94 Saturday, May 27, 2000 11:26 AM
Transmission Line Electric and Magnetic Fields
95
Field Outside of Conductor from (6.1)–(6.4)
(6.6)
∇
×
H = 0 (6.7)
(6.8)
(6.9)
(6.10)
Since
from (6.6) and (6.8) we obtain,
Since
from (6.10) we have,
(6.11)
FIGURE 6.1
Magnetic ﬁeld of a conductor carrying dc current j.
j
R
r
ϕ
z
r
j
o
0
∇ × ( )
∂
∂
∂
∂
H
r
Hz H
z
r
1
0
ϕ
ϕ
–
∇ × ( )
∂
∂
∂
∂
H
r
Hr
z
Hz
r
ϕ
1
0 –
∇ × ( ) +
∂
∂
∂
∂
H
r
H
H
r
Hr
z
1
0 ϕ
ϕ
ϕ
–
∂
∂
z
0
∂
∂
Hz
r
or Hz 0 constant = 0
∂
∂
ϕ
0
1
0
r
H
H
r
ϕ
ϕ
+
∂
∂
1306/C06/frame Page 95 Saturday, May 27, 2000 11:26 AM
96
Powerline Ampacity System: Theory, Modeling, and Applications
and obtain,
(6.12)
where A is a constant.
From (6.4) we have,
(6.13)
since,
(6.14)
we have,
(6.15)
and obtain the following solution,
(6.16)
where B is a constant.
Field Inside the Conductor
The current, , inside the conductor has the following components,
J
r
= 0
J
ϕ
= 0
J
z
≠
0
Since , from (6.10) we have,
(6.17)
H
A
r
ϕ
∇⋅
∇⋅ +
∂
∂
+
∂
∂
+
∂
∂
H
H
r
Hr
Hr
r r
H Hz
z
0
1 1
0
ϕ
ϕ
∂
∂
∂
∂
ϕ
0 0 ,
z
1
0
r
Hr
Hr
r
+
∂
∂
Hr
B
r
r
J
∂
∂
ϕ
0
1
r
H
H
r
j
z
ϕ
ϕ
+
∂
∂
1306/C06/frame Page 96 Saturday, May 27, 2000 11:26 AM
Transmission Line Electric and Magnetic Fields
97
which has for solution,
(6.18)
where K and L are two constants which are determined from the following boundary
conditions,
as r
→
0, H
ϕ
→
∞
or K= 0
substituting, we have
(6.19)
and obtain,
(6.20)
Summary of Equations
The magnetic intensity, (A/m), inside and outside a conductor with current, I, is
found as follows:
Inside the conductor: 0
≤
r
≤
R
H
r
= 0 (6.21)
(6.22)
H
z
= 0 (6.23)
Outside the conductor: r > R
H
r
= 0 (6.24)
(6.25)
H
z
= 0 (6.26)
H
K
r
Lr ϕ +
∂
∂
+
+
H
r
K
r
L
Lr
r
L j
z
ϕ
–
2
H
I r
R
ϕ
π
⋅
⋅ ⋅ 2
2
r
H
H
Ir
R
φ
π
⋅ 2
2
H
I
r
φ
π
⋅ 2
1306/C06/frame Page 97 Saturday, May 27, 2000 11:26 AM
98
Powerline Ampacity System: Theory, Modeling, and Applications
For environmental impact studies, we are interested in the ﬁeld outside the conductor
in free space. Equation (6.25) gives the magnetic ﬁeld, H
φ
, outside the conductor.
The radial, H
r
, and horizontal component, H
z
, of the magnetic ﬁeld outside the
conductor is zero. It is seen that the magnetic ﬁeld at ground level increases with
transmission line ampacity (I) and by lowering of the distance (r) from conductor
to ground. The distance (r) is a function of conductor sag.
6.2.2 T
HE
M
AGNETIC
F
IELD
OF
A
T
HREE
P
HASE
P
OWERLINE
The magnetic ﬁeld of a polyphase transmission line at a point in space can be
calculated from (6.25) by vector addition of the magnetic ﬁeld of each conductor
as follows:
(6.27)
Therefore, for a threephase transmission line having one conductor per phase, the
magnetic ﬁeld is,
(6.28)
is the resultant magnetic ﬁeld (A/m) of the transmission line at a point in space
and are the individual contributions of the magnetic ﬁeld of each phase
conductor at the same point in space.
Example 6.1
Calculate the magnetic ﬁeld of a threephase single circuit 750 kV transmission line
at a point, M, 1 m above ground, and at a distance 30 m from the line shown in
Figure 6.2. The line is loaded to its summer maximum thermal rating of 1500 A per
phase.
FIGURE 6.2
Phase conﬁguration of 750 kV line (Example 6.1).
r r
H H
r i
i
n
∑
1
r r r r
H H H H
r
+
1 2 3
r
H
r r r r
H H H
1 2 3
, ,
1 m
M
30 m
15 m 15 m
1 2 3
20 m
0
y
x
1306/C06/frame Page 98 Saturday, May 27, 2000 11:26 AM
Transmission Line Electric and Magnetic Fields
99
Solution
Transmission Line Conﬁguration
Phase distance
,
x
1
= –15 m
x
2
= 0 m
x
3
= 15 m
y
1
= 20 m
y
2
= 20 m
y
3
= 20 m
Phase current,
= 1500
∠
0
= 1500
∠
–120
= 1500
∠
120
The magnetic ﬁeld at point M is calculated from (6.25),
I = phase current, A
i =1,2,3 phase
r
im
, = distance from phase conductor i to point m, m
I
1
I
2
I
3
H
I
r
i m
i
i m
,
,
2π
H
u u
u u u u
m
x y
x y x y
1
2 2
1500 0
2 19 45
1500 0
2 48 8
20 1
48 8
15 30
48 8
1 9 4 5
,
.
–
. .
. . ,
∠
+
∠ ( )
+
+ ( )
]
]
]
+
( )
π
π
r r
r r r r
are unit vectors in x and y direction
H
u u
j ux j uy
m
x y
2
2 2
1500 120
2 19 30
1500 120
2 35 5
19
35 5
30
35 5
1 79 3 09 2 84 4 88
,
–
. . .
. . . .
∠ −
+
∠
+
]
]
]
+ ( ) + + ( )
π
π
r r
r r
1306/C06/frame Page 99 Saturday, May 27, 2000 11:26 AM
100 Powerline Ampacity System: Theory, Modeling, and Applications
The magnetic ﬁeld at point M is obtained by,
The magnetic ﬁeld of a three phase powerline may be obtained approximately* by,
(6.29)
B = µ
0
H is the magnetic ﬁeld strength, Tesla
µ
0
= permeability of free space
H = magnetic ﬁeld intensity, A/m
I = positive sequence current, A
P = phase to phase distance, m
r = distance from the axis of the line to the point of measurement, m
A = numerical coefﬁcient depending upon line design (geometry)
n = number of subphases for a split phase line
The above equation is consistent with Equation (6.25) after taking into consideration
the effect of multiple conductors, threephase AC, and the transmission line geometry.
6.2.3 THE MAGNETIC FIELD OF DIFFERENT TRANSMISSION
LINE GEOMETRY
In this section results are presented to show the magnetic ﬁelds of highvoltage
transmission lines having different geometry. The calculations are based upon very
optimistic line ampacities that would only be possible by adopting a dynamic line
rating system. Conventional transmission line conductor conﬁgurations are shown
in Figures 6.6, 6.7, and 6.8, and a new conductor conﬁguration is shown in Figure
6.9. The geometry of Figure 6.9 provides a compact transmission line design with
reduced magnetic ﬁeld. The magnetic ﬁeld of each conﬁguration is shown in Figure
6.10, and the magnetic ﬁeld at 30 m distance from the axis of the transmission line
for each line conﬁguration is shown in Table 6.1.
* V.S. Rashkes, R. Lordan, “Magnetic Field Reduction Methods: Efﬁciency and Costs,” IEEE Transactions
on Power Delivery, Vol. 13, No. 2, April 1998.
H
ux uy
j ux j uy
m 3
2 2
1500 120
2 19 15
1500 120
2 35 5
19
24 2
15
24 2
3 87 6 65 3 05 5 25
,
. . .
. – . . .
∠
+
∠
+
]
]
]
( ) + + ( )
π
π
r r
r r
r
r
H H H H
H A m
M m m m
M
+ +
∠
1 2 3
5 38 15 5
, , ,
. .
B A
P
r
I
n
n
⋅

(
'
`
J
J
⋅
+1
1306/C06/frame Page 100 Saturday, May 27, 2000 11:26 AM
Transmission Line Electric and Magnetic Fields 101
The above values are well within the acceptable limit* of 1330 microTeslas for
continuous exposure. A Cigré 1998 paper (Bohme et al. 1998) indicates 100
microTesla as the upper limit. It must also be mentioned that transmission line
magnetic ﬁelds shown in Figure 6.10 are based upon high transmission line
FIGURE 6.6 Single circuit horizontal conﬁguration
FIGURE 6.7 Single circuit delta conﬁguration
FIGURE 6.8 Double circuit vertical conﬁguration
FIGURE 6.9 Compact line with phase splitting
* Restriction on human exposures to static and timevarying EM ﬁelds and radiation. Documents of the
NRPB 4(5): 1–69, 1993. Exposure limits for powerfrequency ﬁelds, as well as static ﬁelds and MW/RF
frequencies; the standards apply to both residential and occupational exposure. For 60Hz the limits
recommended are 10 kV/m for the electric ﬁeld and 1,330 micro T for the magnetic ﬁeld. Copyright©,
1993–1998, by John E. Moulder, Ph.D. and the Medical College of Wisconsin.
1306/C06/frame Page 101 Saturday, May 27, 2000 11:26 AM
102 Powerline Ampacity System: Theory, Modeling, and Applications
conductor ampacity, 1000 A for an ACSR Zebra conductor. During normal operating
conditions the line current will be less than 1000 A, and, therefore, the magnetic
ﬁeld will be lower.
6.2.4 EMF MITIGATION
Even though overhead transmission lines are designed with low levels of EMF, even
tighter control over the level of EMF radiated from a line can be achieved by EMF
mitigation measures. Compact line designs having low levels of EMF are now used
extensively.* Active and passive shielding of lines by the addition of auxiliary
conductors on certain sections of the line are also used to lower the magnetic ﬁeld
at critical locations. These approaches have resulted in lowering the magnetic ﬁeld
to about 0.2 µT at the edge of transmission line rightofway, and are recommended
for areas such as schools, hospitals, and other areas where the public may be exposed
to EMF continuously (Bohme et al., 1998).
Passive Shielding
Passive shielding of overhead lines is accomplished by the addition of auxiliary
shield wires connected in a loop at certain critical sections of the line where
FIGURE 6.10 Transmission line magnetic ﬁeld, ACSR Zebra 1000 A
TABLE 6.1
Conﬁguration
Magnetic Field
at 30 m, µT
Horizontal, single circuit, Figure 6.6 2.92
Delta, single circuit, Figure 6.7 2.06
Vertical, double circuit, Figure 6.8 0.94
Compact Star, phase splitting, Figure 6.9 0.01
* “Compacting Overhead Transmission Lines,” Cigré Symposium, Leningrad, USSR, 3–5 June, 1991.
30
25
20
15
10
5
0
M
i
c
r
o
T
e
s
l
a
Figure 6.6
Figure 6.7
Figure 6.8
Figure 6.9
10 20 30 40 50
Distance, m
Transmission Line Magnetic Field
1306/C06/frame Page 102 Saturday, May 27, 2000 11:26 AM
Transmission Line Electric and Magnetic Fields 103
controlling line EMF is important. In this method, current ﬂows through the
auxiliary conductor by induction from the powerline conductor. The magnitude
and phase of the current in the auxiliary conductor is managed by controlling loop
impedance. In this manner the auxiliary conductor generates a ﬁeld that opposes
the ﬁeld produced by the power conductor, thereby lowering the total ﬁeld from
the line.
Active Shielding
Active shielding is similar to passive shielding, but instead of induced current in the
auxiliary conductors, an external power supply is used to circulate a current through
the shield wires. In this manner even greater control over the ﬁeld generated by a
line is possible by controlling the amplitude and phase of the current through the
shield wires. The following example will help in the analysis of magnetic ﬁeld
shielding of overhead powerlines.
Example 6.2
Calculate the magnetic ﬁeld of a threephase 750 kV single circuit transmission line
at a point M, 1 m above ground and 30 m from the line shown in Figure 6.2. The
line is loaded to its summer maximum thermal rating of 1500 A per phase. Two
auxiliary conductors, M
1
and M
2
, shown in Figure 6.3, are used for magnetic ﬁeld
shielding by forming a 1 km current loop.
FIGURE 6.3 750 kV line with auxiliary shield wires (Example 6.2)
FIGURE 6.4 750 kV line showing shielding angles (Example 6.2)
20 m
M
1
30m
M
2
1 2 3
15m 15m
1m
M
0
y
13m
x
7m
30m
15m 15m
1 2 3
30m
M
1
M
2
Ë
1
Ë
2
Ë
Ò
1306/C06/frame Page 103 Saturday, May 27, 2000 11:26 AM
104 Powerline Ampacity System: Theory, Modeling, and Applications
Solution
The current I
l
ﬂowing through the loop is calculated from,
where,
= voltage induced in the loop
= loop impedance
The induced voltage is calculated from
= jωφᐉ
where,
ω = angular frequency, radian/s
φ = total ﬂux penetrating the loop, wb/m
ᐉ = length of the loop
The total ﬂux φ is calculated by
where
S = area of the loop, m
2
B = total ﬂux density, Tesla
FIGURE 6.5 Magnetic ﬁeld of 750 kV 3 phase transmission line with auxiliary shield wires.
6 3 0 3 6
2
1
0
1
2
Magnetic Field 750 kV Line
A
/
m
Hy
i
A/m
Hx
i
I
V
Z
1
1
1
V
1
Z
1
V
1
φ
∫
BdS
s
.
1306/C06/frame Page 104 Saturday, May 27, 2000 11:26 AM
Transmission Line Electric and Magnetic Fields 105
Since only the y component of the ﬂux density vector contributes to the above
surface integral,
are the ﬂux density components due to current in the phase
conductors respectively and produce the ﬂux , respectively.
Calculating from the integral of equation,
Substituting
B = µ
o
H
H from (6.25) we have,
from,
and,
we ﬁnd the ﬂux due to current ,
From ﬁgure we have,
and obtain the solution to the integral,
B
r
r r r
B B B B
ly y y
+ +
2 3
r r r
B B B
ly y y
, ,
2 3
I I I
1 2 3
, ,
φ φ φ
1 2 3
, ,
φ
1
φ
1
⋅
∫
r
B dn dz
ly
s
r
B
I
ly
µ θ
πρ
0 1
2
sin
θ
θ
( )
( dx h
cos
cot
dx h d
dz
( )
cot θ θ
1
φ
1
I
1
φ
µ
π
θ θ θ θ
θ
θ
1
0 1
2
1
2
( ) ( ) ( )
∫
I
d sin cos cot
θ
θ
1
2
0
0
76 8
.
φ
1
4
2 35 10 0 ⋅ ∠ .
–
wb
1306/C06/frame Page 105 Saturday, May 27, 2000 11:26 AM
106 Powerline Ampacity System: Theory, Modeling, and Applications
Similarly, the ﬂux due to current is calculated
Flux φ
3
due to current I
3
,
The total ﬂux is,
The induced loop voltage is,
ᐉ = 1000 m (1 km loop)
ω = 2⋅π⋅60 rad/s
= j(2.2 · 10
–4
∠ – 120) · 377 · 10
3
= 83.2 ∠ – 30
φ
2
I
2
I
2
1
0
2
0
1500 120
64 6
64 6
∠
–
.
.
θ
θ
φ
µ
π
θ θ θ θ
φ
θ
θ
2
0 2
64 6
64 6
2
4
2
4 55 10 120
1
2
( ) ( ) ( )
⋅ ∠
∫
I
d
wb
sin cos cot
. –
.
.
–
I
3
1
0
2
0
1500 120
76 6
0
∠
θ
θ
– .
φ
µ
π
θ θ θ θ
φ
θ
θ
3
0 3
76 6
0
3
4
2
2 35 10 120
1
2
( ) ( ) ( )
⋅ ∠
∫
I
d
wb
sin cos cot
.
.
–
φ φ φ φ
φ
+ +
⋅ ∠
1 2 3
4
2 2 10 120 . –
–
wb
V
l
V j
l
l ⋅ ⋅ ⋅ φ ω
V
l
V
l
1306/C06/frame Page 106 Saturday, May 27, 2000 11:26 AM
Transmission Line Electric and Magnetic Fields 107
The auxiliary conductor loop impedance is selected as,
The loop current is,
Magnetic Field with Shielding
The magnetic ﬁeld, Ha, at point M, due to the two auxiliary conductors, is calculated,
as in Example 6.1, from,
Ia = current in auxiliary conductors, A
i =1,2
r
i,m
= distance from auxiliary conductors, conductor i to point M
Za
Za ∠ 0 3 30 . – , Ω
I
I A
l
l
∠
∠
83 2 30
0 3 30
277 3
. –
. –
.
Ha
Ia
r
i m
i
i m
,
,
2π
Ha
u u
u u
m
x y
x y
1
2 2
277 3
2 12 45
277 3
2 46 57
13 1
46 57
15 30
46 57
0 24 0 92
,
.
.
.
–
. .
– . – .
+
( )
+
+ ( )
]
]
]
π
π
r r
r r
H
ux uy
ux uy
m 2
2 2
277 3
2 12 45
277 3
2 19 2
12
19 2
15
19 2
1 43 1 79
,
.
.
. . .
– . – .
+
+
]
]
]
π
π
r r
r r
r
r r
r
r r
H ux ux
H uy uy
a x
a y
,
,
– . . – .
– . . – .
+ ( )
+ ( )
0 24 1 43 1 67
0 92 1 79 2 71
1306/C06/frame Page 107 Saturday, May 27, 2000 11:26 AM
108 Powerline Ampacity System: Theory, Modeling, and Applications
The magnetic ﬁeld is an ellipse
6.3 TRANSMISSION LINE ELECTRIC FIELD
The electric ﬁeld, E, of the transmission line at any point in space is a function of
line voltage and the distance of the point from the transmission line conductor.
Therefore, the electric ﬁeld at ground level is affected by conductor sag since an
increase in conductor temperature will increase sag and result in lowering the
distance of the powerline conductor to the ground. The effect of conductor sag due
to higher transmission line ampacity was studied in Chapter 5. In this section we
shall study the method of calculation of the electric ﬁeld of a transmission line and
determine the effect of conductor sag on the electric ﬁeld at ground level.
The electric ﬁeld strength E may be deﬁned as gradient of the potential V given
as,
(6.30)
There exists an electric ﬁeld if there is a potential difference between two points
having potential V
1
and V
2
separated by a distance, r, such that,
(6.31)
In a perfect conductor the potential difference ( ) is zero, the gradient of the
voltage is effectively zero, and, hence, the electric ﬁeld inside a perfect conductor
is also zero.
Electric ﬁeld calculation
The various methods of calculation of the electric ﬁeld by numerical and analytical
methods are given in a Cigré report (Cigré, 1980). In this section we shall apply the
r
r r
r
r r
H H H j ux j ux
H H H j uy j uy
x M x a x
y M y a y
+ ( ) { ¦
( )
+ ( ) { ¦
( )
, ,
, ,
. – . – . . – .
. – . – . . – .
3 77 3 56 1 67 2 08 3 56
1 93 0 37 2 71 1 31 0 37
r
H
r r r
r
H H H
H
H
H
x y
y
x
+

(
'
'
`
J
J
J
∠
2 2
0
4 3
18 3
4 3 18 3
.
.
. .
Angle = tan
–1
r r
E grad V V m
( )
–
grad V
V V
r
r
r r
( )
1 2
–
r r
V V
1 2
–
1306/C06/frame Page 108 Saturday, May 27, 2000 11:26 AM
Transmission Line Electric and Magnetic Fields 109
analytical method of equivalent charges to calculate the electric ﬁeld of threephase
transmission lines.
The electric ﬁeld, E, at a distance, r, from a charge, q, is calculated by Gauss’s
law,
(6.32)
The q charges carried by transmission line conductors is calculated by,
[q] = [C] · [V] (6.33)
where [q] is a column vector of charges, [C] is the capacitance matrix of the multi
conductor circuit, and [V] a column vector of phase voltages.
The capacitance matrix [C] is calculated from Maxwell’s potential coefﬁcients
[λ] deﬁned as the ratio of the voltage to charge.
The elements, λ
ii
, of the matrix of potential coefﬁcients are calculated by,
(6.34)
where,
h
i
= height of the conductor i above ground
r
i
= radius of conductor i
For bundle conductor system an equivalent radius is calculated as,
(6.35)
r = subconductor radius
n = number of subconductors in bundle
R = geometric radius of the bundle
The elements λ
ij
of the matrix of potential coefﬁcients are calculated by,
(6.36)
where D
i,j
is the distance between conductors i and j, and D′
i,j
is the distance between
image conductors i’ and j’.
E
q
r
2
0
πε
λ
πε
ii
i
i
n
h
r
1
2
2
0
l
r R n
nr
R
eq
⋅
λ
πε
ij
ij
ij
n
D
D
′
1
2
2
0
l
1306/C06/frame Page 109 Saturday, May 27, 2000 11:26 AM
110 Powerline Ampacity System: Theory, Modeling, and Applications
The matrix [C] is calculated by inversion,
[C] = [λ]
–1
(6.37)
Knowing [C] and [V], we calculate [q] from,
[q] = [C] · [V] (6.38)
The electric ﬁeld, E, is then calculated by the application of Gauss’s law by vector
summation of the individual ﬁelds due to the charge on each conductor
(6.39)
where,
(6.40)
The following example illustrates the important concepts presented in this section
by showing the calculation of the electric ﬁeld of the transmission line in Example 1.
Example 6.3
Calculate the electric ﬁeld of a three phase single circuit 750 kV transmission line
at a point, M, 1 m above ground, and at a distance 30 m away from the line as
shown in Figure 6.1. The line is loaded to its summer maximum thermal rating of
1500 A per phase.
Solution
The height of conductor above ground is given as,
h
1
= h
2
= h
3
= 20m
The equivalent radius of four conductor bundle is,
FIGURE 6.6 Threephase conﬁguration of a 750 kV transmission line
r r
E E
i
i
i n
∑
1
r
E
q
r
i
i
2
0
πε
1306/C06/frame Page 110 Saturday, May 27, 2000 11:26 AM
Transmission Line Electric and Magnetic Fields 111
Maxwell’s potential coefﬁcients are easily calculated in Mathcad® as,
For a threephase transmission line, n = m = 3, resulting in the following matrix of
potential coefﬁcients,
And we obtain the capacitance matrix [C] from,
The charge, q, is obtained by substitution of [C] and [V] ,
r R n
nr
R
r
m
eq
eq
⋅
⋅ ⋅
⋅ ⋅
⋅
150 10 4
4 30 10
150 10
0 54
3
3
3
–
–
–
.
λ
π ε
π ε
n m
h
r
if i j
D
D
i
i
i j
i j
, :
ln
ln
,
,
( ) ∈
∈
←
⋅

(
'
`
J
J
⋅ ⋅
←
′
( )
]
]
]
]
⋅ ⋅
for i l..n
for j l..m
A
A otherwise
A
i, j
i, j
2
2 0
2 0
λ 3 3
7 76 10 1 884 10 9 195 10
1 884 10 7 76 10 1 884 10
9 195 10 1 884 10 7 76 10
10 10 9
10 10 10
9 10 10
,
. . .
. . .
. . .
( )
⋅ ⋅ ⋅
⋅ ⋅ ⋅
⋅ ⋅ ⋅
]
]
]
]
]
C
C
[ ]
[ ]
⋅ ⋅ ⋅
⋅ ⋅ ⋅
⋅ − ⋅ ⋅
]
]
]
]
]
λ
–
– – –
– – –
– – –
. – . – .
– . . – .
– . . .
1
11 12 13
12 11 12
13 12 11
1 375 10 3 126 10 8 7 10
3 126 10 1 44 10 3 126 10
8 7 10 3 126 10 1 375 10
1306/C06/frame Page 111 Saturday, May 27, 2000 11:26 AM
112 Powerline Ampacity System: Theory, Modeling, and Applications
giving,
The electric ﬁeld, E, is obtained by the application of Gauss’s law,
The resultant E ﬁeld at M is calculated by vector summation by adding X and Y
components of the individual elements of [E]. The X and Y components are obtained
as,
q C V
q
[ ]
[ ][ ]
[ ]
⋅ ⋅ ⋅
⋅ ⋅ ⋅
⋅ − ⋅ ⋅
]
]
]
]
]
⋅ ∠
⋅ ∠
⋅ ∠
1 375 10 3 126 10 8 7 10
3 126 10 1 44 10 3 126 10
8 7 10 3 126 10 1 375 10
750 10 0
750 10 120
750 10 240
11 12 13
12 11 12
13 12 11
3
3
3
. – . – .
– . . – .
– . . .
– – –
– – –
– – –
]
]
]
]
]
q
j
j
j
[ ]
⋅ ⋅
⋅ + ⋅
⋅ ⋅
]
]
]
]
]
1 18 10 1 45 10
6 574 10 1 131 10
4 636 10 1 088 10
5 6
6 5
6 5
. – .
– . .
– . – .
– –
– –
– –
E
q
q
q
[ ]
]
]
]
]
]
]
]
]
]
]
]
1
2
0
1
1
2
2
3
3
πε
ρ
ρ
ρ
Ey E
y h
x d y h
y h
x d y h
Ex E
x d
x d y h
x d
x d
i i i
i i
i i i i
i i
i i i i
i i i
i i
i i i i
i i
i i
:
–
– –
–
–
:
–
– –
–
–
⋅ ⋅
( )
( )
+
( )
+
( )
( )
+ +
( )
]
]
]
]
⋅ ⋅
( )
( )
+
( )
+
( )
( )
ρ
ρ
2 2 2 2
2 2 2
++ +
( )
]
]
]
]
y h
i i
2
1306/C06/frame Page 112 Saturday, May 27, 2000 11:26 AM
Transmission Line Electric and Magnetic Fields 113
The resultant electric ﬁeld is an ellipse as seen in the Figure 6.7.
6.4 CHAPTER SUMMARY
The magnetic ﬁeld of a transmission line at ground level is a function of line current
and the distance of phase conductors from ground. The magnetic ﬁeld inside and
outside of a currentcarrying conductor has been developed from Maxwell’s equation
and Ampere’s law. As shown by the equations developed in this chapter, an increase
in line current increases the magnetic ﬁeld at ground level. The magnetic ﬁeld at
ground level also increases with higher sag. If conductor temperature is higher than
normal due to higher current through the line, then the magnetic ﬁeld at ground level
will become more signiﬁcant due to the combined effect of high current and reduced
distance of conductor to ground. A numerical example is provided in this chapter for
the calculation of the magnetic ﬁeld of threephase transmission line. Methods of
reducing magnetic ﬁelds by active and passive shielding are also presented in this
chapter.
The electric ﬁeld from a transmission line at ground level is indirectly affected
by line ampacity only if an increase in line ampacity raises the maximum design
temperature of the transmission line conductor. If conductor temperature is higher
than the maximum allowed for the line, then sag will increase. Consequently the
distance from conductor to ground will become less than normal which will raise
the electric ﬁeld at ground level. A method of calculation of electric ﬁeld at ground
level due to higher transmission line ampacity is given in this chapter with a
numerical example.
Since both electric and magnetic ﬁelds of a transmission line depend upon
conductor temperature, it is very important that increasing line ampacity does not
exceed the maximum design temperature of the conductor. Therefore, for EMF
considerations also, it is important to follow a dynamic line rating system that will
maintain normal conductor temperature within a speciﬁed limit.
FIGURE 6.7 The electric ﬁeld of 750 kV threephase transmission line.
Ey i
Ex i
i
i
i
i
∑
∑
⋅ + ⋅
+
1
3
3 3
1
3
3
9 475 10 7 602 10
127 271 1 181 10
. .
– . .
1306/C06/frame Page 113 Saturday, May 27, 2000 11:26 AM
115
7
Weather Modeling for
Forecasting Transmission
Line Ampacity
7.1 INTRODUCTION
Since weather is an important parameter in the determination of transmission line
ampacity, the development of weather models of ambient temperature, wind speed,
wind direction, and solar radiation are presented in this chapter. These are statistical
weather models based upon timeseries analysis and National Weather Service fore
casts. Hourly values of future meteorological conditions from 1 to 24 hours ahead,
or even up to 1 week in advance, are now becoming possible due to developments
in weather forecasting.
The solution of the differential equations for the heating of a conductor by
current in the steady, dynamic, and transient states requires the knowledge of the
following meteorological variables:
• Ambient Temperature
• Wind Speed
• Wind Direction
• Solar Radiation
When transmission line ampacity is required for the present time, the above meteoro
logical data can be obtained by measurement from weather stations. For the prediction
of line ampacity several hours in advance, a weather model is required. In this chapter,
stochastic and deterministic models of ambient temperature, wind speed, and wind
direction, and an analytical model of solar radiation shall be developed from time
series data. Neural network models are also presented for forecasting hourly values
of meteorological conditions as well as for weather pattern recognition.
The prediction of transmission line ampacity several hours in advance has
become more important today due to competition in the electric power supply
industry, and greater need for the advance planning of electricity generation and
transmission capacity* (Deb, 1998, 1997, 1995; Cibulka, Williams, and Deb, 1991;
Hall and Deb, 1988c; Douglass, 1986; Foss and Maraio, 1989). Numerical examples
* M. Aganagic, K. H. AbdulRahman, J.G. Waight.
Spot pricing of capacities for generation and
transmission of reserve in an extended Poolco model.
IEEE Transactions on Power Systems, Vol. 13,
No. 3, August 1998
1306/C07/frame Page 115 Saturday, May 27, 2000 11:35 AM
116
Powerline Ampacity System: Theory, Modeling, and Applications
of weather forecasting are presented in this chapter, followed by an example of line
ampacity forecast generated by a program.
Having precise knowledge of future transmission line capacity will greatly
facilitate the purchase of competitively priced electricity from remote locations. In
the future we expect an increase in the number of power producers requiring access
to utility transmission lines for the supply of electricity. For these reasons, a trans
mission line ampacity program with forecast capability is essential.
The LINEAMPS program uses a weather model based on historical weather
data as well as weather forecast data prepared by the National Weather Service. Two
alternative approaches to weather modeling are developed. In the ﬁrst approach,
hourly values of historical weather data for different seasons of the year are ﬁtted
by a Fourier series. In the second approach, weather patterns are recognized by
training an unsupervised neural network using Kohonen’s learning algorithm
(Haykin, 1999; Eberhart and Dobbins, 1990). These patterns are then adjusted to
forecast weather data available from the National Weather Service or other weather
service companies. Hourly values of future ambient temperature and wind speed
data are then generated from these patterns as described in the following section.
When continuous input of realtime meteorological data is available, a Kalman
ﬁltertype algorithm is developed for the recursive estimation of weather variables
for realtime prediction of transmission line ampacity.
7.2 FOURIER SERIES MODEL
Hourly values of ambient temperature (Ta) and wind speed (W
s
) at time (t) are
generated by AmbientGen and WindGen methods in the weather station object of
the program by ﬁtting Fourier series to historical weather data of the region. The
Fourier series model is given by,
(7.1)
Description of symbols
Y(t)
∈
{Ta(t), W
s
(t)}
A
0
, C
i
, B
i
for i = 1...n are coefﬁcients of the model
ω
= 2
π
/24 = fundamental frequency
k = factor used to adjust Fourier series to National Weather Service forecast.
It is calculated by,
(7.2)
Y t A k C Sin t B Cos t
i
i
n
i i
i
n
i
( ) +
( )
+
( )
∑ ∑
0
1 1
ω ω
k
Y t Y t
F t F t
f f
( ) ( )
( ) ( )
max min
max min
–
–
1306/C07/frame Page 116 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity
117
Y
f
(t
max
) = daily maximum value of ambient temperature or wind speed forecast
by the National Weather Service.
Y
f
(t
min
)= daily minimum value of ambient temperature or wind speed forecast
by the National Weather Service.
F( t
max
) and F( t
min
) are found from the Fourier series (7.1),
when t = t
max
and t = t
min
respectively.
The development of a weather model (Figure 7.1) requires determining the
parameters of the model from historical weather data of each of the meteorological
variables. The unknown parameters of the model [A
0
, K, A
i
, B
i
, n,
ω
] are determined
by least square regression analysis. The fundamental frequency
ω
and the coefﬁcients
r
N
(
ω
) of the frequency spectra are also determined by spectral analysis (Priestley
1981) as shown in the Figures 7.2, 7.4, and 7.6 where,
(7.3)
e
–j
ω
t
= A(
ω
) +jB(
ω
) (7.4)
From Figure 7.2 we see that the dominant frequency is equal to 0.042, which is also
the fundamental frequency. Since the period T = 1/f , the fundamental period is
found to be approximately equal to 24 hours, as we should expect for the region of
San Francisco. The same phenomenon is observed in all of the meteorological
variables comprising ambient temperature, wind speed, wind direction, and solar
radiation in this region as seen in Figures 7.2, 7.4, and 7.6.
Examples of Fourier series patterns of hourly ambient temperature and wind
speed that were developed for the San Francisco Bay area during summer time are
shown in Figure 7.14. It is appropriate to mention here that these patterns are
applicable to the region of the San Francisco Bay area only. A similar analysis is
required for transmission lines in other regions.
FIGURE 7.1
Hourly averaged ambient temperature data during summer time in San Fran
cisco.
T
e
m
p
e
r
a
t
u
r
e
D
e
g
.
C
Ta
j
Signal: Ambient Temperature
j
Hour
40
30
20
10
0 500 1000 1500 2000 2500
r
N
X e
N i
t
N
j t
ω
ω
( ) ⋅
∑
2
1
–
1306/C07/frame Page 117 Saturday, May 27, 2000 11:35 AM
118
Powerline Ampacity System: Theory, Modeling, and Applications
The daily cyclical behavior of the meteorological variables is further supported
by the autocorrelations that were calculated from the hourly averaged values of each
time series as shown in the Figures 7.157.17.
The autocorrelation r
k
is calculated by,
FIGURE 7.2
Ambient temperature spectrum.
FIGURE 7.3
Hourly averaged wind speed data during summer time in San Francisco.
FIGURE 7.4
Wind speed spectrum.
Ambient Temperature Spectrum
r
j
C
o
e
f
f
i
c
i
e
n
t
j
2048
Frequency
100
80
60
40
20
0
0 0.1 0.2 0.3 0.4 0.5
Signal: Wind Speed
Ws
j
W
i
n
d
S
p
e
e
d
,
m
/
s
j
Hour
2000 1500 1000 500
0
0
5
10
15
20
Wind Speed Spectrum
40
30
20
10
r
j
C
o
e
f
f
i
c
i
e
n
t
0
0 0.1 0.2
j
2048
Freqency
0.3 0.4 0.5
1306/C07/frame Page 118 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity
119
(7.5)
FIGURE 7.5
Hourly averaged wind direction data during summer time in San Francisco.
FIGURE 7.6
Wind direction spectrum.
FIGURE 7.7
Ambient temperature pattern, June.
Signal: Wind Direction
2000 1500 1000 500
j
Hour
0
0
100
Wd
j
200
300
W
i
n
d
D
i
r
e
c
t
i
o
n
D
e
g
r
e
e
400
Wind Direction Spectrum
600
400
200
r
j
C
o
e
f
f
i
c
i
e
n
t
0
0 0.1 0.2 0.3 0.4
j
2048
Frequency
0.5
Ambient Temperature Pattern
June
Hour
35
30
25
20
15
10
5
0
0
4 8 12 16 20
A
m
b
i
e
n
t
T
e
m
p
e
r
a
t
u
r
e
,
C
o
r
c
c
k
k
0
1306/C07/frame Page 119 Saturday, May 27, 2000 11:35 AM
120
Powerline Ampacity System: Theory, Modeling, and Applications
The autocovariance c
k
is given by,
(7.6)
z
t
= average hourly value of the meteorological variable at time t
FIGURE 7.8
Ambient temperature pattern, July.
FIGURE 7.9
Ambient temperature pattern, August.
FIGURE 7.10
Ambient temperature pattern, September.
Ambient Temperature Pattern
Hour
35
30
25
20
15
10
5
0
0
4 8 12 16 20
A
m
b
i
e
n
t
T
e
m
p
e
r
a
t
u
r
e
,
C
July
Ambient Temperature Pattern
August
Hour
35
30
25
20
15
10
5
0
0
4 8 12 16 20
A
m
b
i
e
n
t
T
e
m
p
e
r
a
t
u
r
e
,
C
Ambient Temperature Pattern
August
Hour
35
30
25
20
15
10
5
0
0
4 8 12 16 20
A
m
b
i
e
n
t
T
e
m
p
e
r
a
t
u
r
e
,
C
c
N
z z z z k
k t
t
N k
t k
( )( )
−
+
∑
1
1
– – 0, 1, 2 n lags K
1306/C07/frame Page 120 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity
121
The process mean is,
(7.7)
N = number of observations in the time series
FIGURE 7.11
Wind speed pattern, June.
FIGURE 7.12
Wind speed pattern, July.
FIGURE 7.13
Wind speed pattern, August.
Wind Speed Pattern
June
Hour
10
8
6
4
2
0
0 4 8 12 16 20
W
i
n
d
S
p
e
e
d
,
m
/
s
Wind Speed Pattern
July
Hour
10
8
6
4
2
0
0 4 8 12 16 20
W
i
n
d
S
p
e
e
d
,
m
/
s
Wind Speed Pattern
August
Hour
10
8
6
4
2
0
0 4 8 12 16 20
W
i
n
d
S
p
e
e
d
,
m
/
s
z
z
N
t
t
N
∑
1
1306/C07/frame Page 121 Saturday, May 27, 2000 11:35 AM
122
Powerline Ampacity System: Theory, Modeling, and Applications
One of the important advantages of a Fourier series model (Figure 7.1) for the
prediction of hourly values of ambient temperature and wind speed is that it does
not require continuous input of weather data. Hourly values of future weather data
are generated by the model by adjusting the coefﬁcients with general purpose weather
forecast data. The daily maximum and minimum values forecast by the weather
FIGURE 7.14
Wind speed pattern, September.
FIGURE 7.15
Ambient temperature autocorrelations.
FIGURE 7.16
Wind speed autocorrelations.
Wind Speed Pattern
September
Hour
10
8
6
4
2
0
0 4 8 12 16 20
W
i
n
d
S
p
e
e
d
,
m
/
s
A
u
t
o
c
o
r
r
e
l
a
t
i
o
n
Ambient Temperature
Autocorrelations
Lag, hr
1.2
1
0.8
0.6
0.4
0.2
0
0 12 24 36 48
A
u
t
o
c
o
r
r
e
l
a
t
i
o
n
Wind Speed Autocorrelations
Lag, hr
1.2
1
0.8
0.6
0.4
0.2
0
0 12 24 36 48
1306/C07/frame Page 122 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity
123
service are used to adjust model coefﬁcients using Figure 7.2. Figure 7.19 shows
hourly values of ambient temperature generated by the Fourier series model in
comparison with measured data for one week during summer in the San Francisco
Bay area. The model is also useful for simulation purpose as shown in Figure 7.20,
and as discussed in Chapter 5.
7.3 REALTIME FORECASTING
As stated earlier, forecasting of transmission line ampacity several hours in advance
is beneﬁcial for the advance planning of generation and transmission resources. Due
to deregulation in the electric utility industry there is even greater competition for
the supply of electric energy. Utilities and power producers are making advance
arrangements for the purchase and sale of electricity, which requires ensuring
adequate transmission capacity. Transmission line capacities are predicted in
advance by the LINEAMPS program by taking into account weather forecast data
and the weather models developed in the previous section.
FIGURE 7.17
Ambient temperature and wind speed crosscorrelations.
FIGURE 7.18
Selection of number of harmonics in the Fourier series model of
ambient temperature.
Ambient Temperature and Wind
Speed CrossCorrelations
C
r
o
s
s

C
o
r
r
e
l
a
t
i
o
n
s
Lag, hr
0.4
0.3
0.2
0.1
0
0.1
0.2
0.3
0.4
0 24 48
Selection of Numbers of Harmonics
Ambient Temperature Model
S
u
m
o
f
S
q
u
a
r
e
s
20000
15000
10000
5000
0
0 1 2 3 4 5
Numbers of Harmonics
1306/C07/frame Page 123 Saturday, May 27, 2000 11:35 AM
124
Powerline Ampacity System: Theory, Modeling, and Applications
7.3.1 F
ORECASTING
A
MPACITY
FROM
W
EATHER
P
ATTERNS
The algorithm for forecasting line ampacity by Fourier series weather patterns is
given in the ﬂow chart of Figure 7.31. An example of forecasting hourly values of
line ampacity up to seven days in advance by the LINEAMPS program is given in
the Figure 7.32.
7.3.2 R
EAL
T
IME
F
ORECASTING
OF
T
RANSMISSION
L
INE
A
MPACITY
When realtime weather data is available continuously, it is possible to forecast
hourly values of weather data on an hourbyhour basis by the application of the
Kalman ﬁlter algorithm, which is suitable for realtime predictions. It is a recursive
algorithm that calculates future values of the meteorological variables consisting of
ambient temperature, wind speed, and wind direction based on previous measure
ments of these variables. The predicted values of meteorological variables are then
entered into the transmission line heat balance equation to calculate line ampacity.
The following recursive algorithm is developed for the prediction of hourly values
of ambient temperature, wind speed, and wind direction.
FIGURE 7.19
Forecasting hourly values of ambient temperature.
FIGURE 7.20
Simulation of hourly values of ambient temperature by Fourier series model
and a second order autoregressive stochastic model.
40
35
30
25
20
15
10
5
0
5
10
0 24 48 72 96 120 144 168
Measured
Predicted
Error
Hour
T
e
m
p
e
r
a
t
u
r
e
,
C
Forecasting Ambient Temperature from Daily
Max/Min Forecast Weather Data
Ambient Temperature Simulations
T
e
m
p
e
r
a
t
u
r
e
,
D
e
g
.
C
Hour
50
40
30
20
10
0
0 24 48 72 96 120 144 168
1306/C07/frame Page 124 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity
125
The measurements, y(t), comprising ambient temperature, wind speed, and wind
direction are considered to be composed of a periodic component, p(t), and a
stochastic component, z(t),
y(t) = p(t) + z(t) (7.8)
where y(t) represents the hourly values of measurement of ambient temperature,
wind speed, and wind direction, or a coefﬁcient of heat transfer, Hc.
The Auto Regressive Moving Average process with exogenous Variables
(ARMAV) was selected to model z(t) as given below,
A(q
–1
,t)z(t) = B(q
–1
,t)U(td) + C(q
–1
,t)S(t) (7.9)
where , A(q
–1
,t), B(q
–1
,t), C(q
–1
,t) are timevariable polynomials in the backward
shift operator q
–1
:
A(q
–1
,t) = 1 + a
1
(t) q
–1
+ a
2
(t) q
–1
+ … + a
na
(t) q
–1
B(q
–1
,t) = 1 + b
1
(t) q
–1
+ b
2
(t) q
–1
+ … + b
nb
(t) q
–1
C(q
–1
,t) = 1 + b
1
(t) q
–1
+ b
2
(t) q
–1
+ … + c
nb
(t) q
–1
In the above equations, z(t), u(t), and s(t) represent the output, input, and white noise
sequence, respectively.
The periodic term, p(t), is represented by a Fourier series given by,
p(t) = m(t) + f
i
(t) sin(iωt) + g
i
(t) cos(iωt) (7.10)
where,
m(t) = process mean
f
i
, g
i
i = 1,2 … nh are the coefﬁcients of the model
nh = number of harmonics
ω = 2π/24 = fundamental period
Writing the parameter vector compactly,
x
T
(t) = {a
1
(t) … a
na
(t), b
1
(t) … b
nb
(t), c
1
(t) … c
np
(t),
m(t), f
i
(t) … f
nh
(t), g
i
(t) … g
nh
(t)} (7.11)
the problem now becomes that of estimating x
T
(t) at each instant (t) based on the
measurement y(t). This is carried out recursively by the Kalman ﬁlter algorithm.
1306/C07/frame Page 125 Saturday, May 27, 2000 11:35 AM
126 Powerline Ampacity System: Theory, Modeling, and Applications
State Equation
The true value of the parameter vector is assumed to vary according to,
x(t+1) = x(t) + v(t) (7.12)
where v(t) is a sequence of independent gaussien random vector.
Measurement Equation
From equations ( 7.7) – (7.11) we may write the measurement equation as,
y(t) = H
T
(t) x(t) + e(t) (7.13)
where the matrix H (actually a row vector) is given by,
H
T
= { y(t1), y(t2), … y(tnp),
1, u(td1), u(td2), … u(tdnw),
n(t1), n(t2), … .n(tnf),
1, sin(ωt), sin(2ωt), … sin(nhωt),
cos(ωt), cos(2ωt), … cos(nhωt)} (7.14)
Equations (69), (70) constitute the state and measurement equation, respectively,
and, therefore, the problem of parameter estimation is reduced to the problem of
state estimation. The Kalman ﬁlter algorithm can now be applied to estimate the
state vector x(t).
7.3.3 KALMAN FILTER ALGORITHM
State Update Equation
x(t) = x(t1) + n(t)k(t) (7.15)
Innovations
n(t) = y(t)  H
T
x(t1) (7.16)
Kalman Gain
k(t) = p(t1)H(t)[R
2
(t) + H
T
(t)p(t1)H(t)]
1
(7.17)
1306/C07/frame Page 126 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity 127
where the error covariance matrix p(t) is given by,
(7.18)
The results obtained by the application of the above algorithm* to predict hourly
values of ambient temperature are shown in the Figure 7.21.
7.4 ARTIFICIAL NEURAL NETWORK MODEL
Neural network is an important subject of research in artiﬁcial intelligence where
computations are based on mimicking the functions of a human brain. Neural
networks consist of many simple elements called neurons that are linked by con
nections of varying strengths as shown in Figure 7.22. The neural networks used in
numerical analysis today are gross abstractions of the human brain. The brain
consists of very large numbers of far more complex neurons that are interconnected
with far more complex and structured couplings (Haykin. 1999).
A supervised neural network using the back propagation algorithm, and an
unsupervised neural network using Kohonen’s learning algorithm, are the two types
of neural networks that were used in weather modeling for line ampacity predictions.
A neural network is trained to forecast hourly values of ambient temperature and
wind speed by using the back propagation algorithm. An unsupervised neural net
work is also developed for weather pattern recognition by using Kohonen’s learning
algorithm. Results obtained by the application of the above neural networks are
presented in Figures 7.23 and 7.24.
* There is recent interest in the application of the Kalman ﬁlter algorithm for the efﬁcient solution of
nonlinear recurrent neural networks for realtime prediction. For example, a recent book by Simon Haykin,
Neural Networks: A Comprehensive Foundation, published by PrenticeHall in 1999 recommends the
Kalman ﬁlter for realtime recurrent learning.
FIGURE 7.21 Recursive estimation of San Francisco Bay area ambient temperature during
summer time by the application of Kalman ﬁlter algorithm.
p t p t
R t p t H t p t
R t H t p t H t
T
T
( ) ( ) +
( ) ( ) ( ) ( )
[ ]
( ) + ( ) ( ) ( )
[ ]
–
– – –
–
1
1 1
1
1
2
RECURSIVEESTIMATIONOFAMBIENT
TEMPERATURE
Measured
Predicted
1 21 41 61 81 101
Hour
35
30
25
20
15
10
5
0
T
e
m
p
e
r
a
t
u
r
e
,
˚
C
1306/C07/frame Page 127 Saturday, May 27, 2000 11:35 AM
128 Powerline Ampacity System: Theory, Modeling, and Applications
According to Haykin,* a neural network is a massively parallel distributed
processor that has a natural propensity for storing experiential knowledge and mak
ing it available for use. It resembles the brain in two respects:
1. Knowledge is acquired by the network through a learning process.
2. Interneuron connection strengths known as synaptic weights are used to
store the knowledge.
FIGURE 7.22 A threelayer artiﬁcial neural network with three inputs and three outputs
and a hidden layer.
FIGURE 7.23 Example of neural network application to forecast next hour ambient tem
perature in the San Francisco Bay area. The network trains by supervised learning using the
back propagation algorithm. Neural network results are compared to a statistical forecasting
model and actual data.
* Haykin, S. (1994), Neural Networks: A Comprehensive Foundation, Macmillan, NY, p. 2.
Input Layer Hidden Layer Output Layer
Input 1
Input 2
Input 3
Output 1
Output 2
Output 3
Figure 7. 22 A three layer Artificial Neural Network with three inputs
and three outputs and a hidden layer.
TEMPERATURE FORECASTING BYARTIFICIAL
NEURAL NETWORK (ANN)
Measured
ANN
S/D Model
95 48
Hours
1
0
5
10
15
20
25
30
35
40
T
e
m
p
e
r
a
t
u
r
e
,
1306/C07/frame Page 128 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity 129
7.4.1 TRAINING OF THE NEURAL NETWORK
A neural network is ﬁrst trained by feeding it with data, from which it learns the
inputoutput relationship of a system. Once trained, the network provides the correct
output from a set of input data. If the training set is sufﬁciently large, then the neural
net will provide the correct output from a set of input data that is different from the
training set. A neural network is different from a lookup table. Unlike a lookup
table, the dynamics of the system are represented by a trained neural network. It is
therefore clear that a neural network is particularly useful to predict the outcome of
a phenomenon that cannot be formulated otherwise.
7.4.2 SUPERVISED AND UNSUPERVISED LEARNING
Learning is the key to AI—the artiﬁcial neural network learns from data and dem
onstrates intelligent capability. Learning in an artiﬁcial neural network is either
supervised or unsupervised.
Supervised Learning
In a supervised neural network, learning or training is carried out by a back propa
gation algorithm, where network output is compared to a target and the difference
is used to adjust the strength of the connections. Training is completed when the
sum of squares of errors is minimized. The back propagation algorithm is presented
below, and an application of this algorithm to forecast hourly ambient air temperature
in the San Francisco Bay area is shown in Figure 7.23.
Unsupervised Learning
In unsupervised learning there is no teacher, in other words, there is no target
response with which to compare output. The network organizes by itself (self
organizing neural network) and learns to recognize patterns within data. Unsuper
vised learning is carried out by Kohonen’s learning algorithm, which is used for
FIGURE 7.24 Application of unsupervised neural network for ambient temperature pattern
recognition in the San Francisco Bay area.
Pattern Recognition By Neural Network
30
25
20
15
10
5
T
e
m
p
e
r
a
t
u
r
e
,
C
o
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour of Day
1306/C07/frame Page 129 Saturday, May 27, 2000 11:35 AM
130 Powerline Ampacity System: Theory, Modeling, and Applications
pattern recognition and data classiﬁcation. In Kohonen’s learning algorithm the
strength of interconnections is adjusted by minimizing the Euclidean distance of
output neuron. The output neuron having the minimum Euclidean distance is
declared the winner and set to 1; all others are set to 0.
For the above reasons, a selforganizing neural network is also called a “winner
takeall” algorithm, because when the network is trained only a certain output will
go high, depending upon the characteristics of the input vector. During training, the
weights of the connections are adjusted until subsequent iterations do not change
weights. A winnertakeall selforganizing neural network due to Kohonen’s learning
algorithm is presented below, and its application to pattern recognition of daily
ambient temperature is shown in Figure 7.24. The above types of neural networks
are examples of nonrecurrent networks.
Another important type of network is the recurrent network due to Hopﬁeld,
where there is continuous feedback from output to input. Recurrent networks ﬁnd
applications in nonlinear optimization problems whose solutions are difﬁcult by
conventional means.
7.4.3 BACK PROPAGATION ALGORITHM*
The back propagation algorithm is composed of the following steps:
1. Apply an input vector x
2. Calculate the error e between the output vector y and a known target
vector z
3. Minimize errors
4. Calculate error signal of output layer δ(l)
δ(l) =[z(l) – y(l)] · y(l) · [l – y(l)]
5. Calculate error signal of input layer δ(j)
* Rumelhart, David E., McClelland, James, L., Parallel Distributed Processing, Volume 1, M.I.T. Press,
Cambridge, MA, 1988.
e z y
l
n
( ) ( )
[ ]
∑
1 1
2
1
– n = length of training vector
∂
∂ ( )
∂
∂
⋅
∂
∂ ( )
( )
e
w j i
e
y
z
w l j
l
, ,
δ
δ δ j y j y j w l j l
l
l nl
( ) ( ) ⋅ ( )
[ ]
( ) ( )
∑
1
0
– ,
1306/C07/frame Page 130 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity 131
6. Update weights by the learning rule
w(j, i) new = w(j, i) old + δ(j) · y(i) + α[∆w(j, i)old]
w(l, i) new = w(l, i) old + δ(l) · y(l) + α[∆w(l, i)old]
Steps 1 through 6 are repeated until the sum of squares of errors is minimized.
A neural network for the prediction of hourly values of ambient temperature is
developed by using the back propagation algorithm and the results are presented in
Figure 7.23.
7.4.4 UNSUPERVISED NEURAL NETWORK TRAINING ALGORITHM*
The unsupervised neural network training algorithm is due to Kohonen and is
composed of the following steps:
1. Apply an input vector x
2. Calculate the Euclidean space d(j) between x and the weight vector w of
each neuron as follows:
n = number of training vector x
w(i, j) = weight from input i to neuron j
3. The neuron that has the weight vector closest to x is declared the winner.
This weight vector, called w
c
, becomes the center of a group of weight
vectors that lie within a distance d from w
c
.
4. Update nearby weight vectors as follows:
w(i, j) new = w(i, j) old + α[x – w(i, j) old]
where, α is a timevarying learning coefﬁcient normally in the range 0.1 < α < 1.
It starts with a low value of 0.1 and gradually increases to 1 as learning takes place.
Steps 1 through 4 are repeated until weight change between subsequent iterations
becomes negligible.
A selforganizing neural network is also developed for ambient temperature
pattern recognition by using Kohonen’s learning algorithm, and the results are
presented in Figure 7.25.
* Wasserman, Philip D. 1989 Neural Computing, Van Nostrand, New York.
d j x i w i j
i l
n
( ) ( ) ( ) ( )
[ ]
∑
– ,
2
1306/C07/frame Page 131 Saturday, May 27, 2000 11:35 AM
132 Powerline Ampacity System: Theory, Modeling, and Applications
7.5 MODELING BY FUZZY SETS
Fuzzy Set Theory was introduced by Professor Lotﬁ Zadeh* of the University of
California, Berkeley during the 1960s. Fuzzy set theory accepts many valued logic
and departs from the classical logic of Aristotle which allows a proposition to be
either true or false. The idea of manyvalued logic was developed by Jan
Lukasiewicz, a Polish logician in the 1920s and applied by Max Black in 1937.
Zadeh formally developed multivalued set theory in 1965 and called it fuzzy set
theory.
The main idea of fuzzy sets is that they allow partial membership of an element
in a set, as shown in Figure 7.25. A fuzzy set F in a universe of discourse U is
deﬁned to be a set of ordered pairs,
F = {(u, m
F
(u))Έu ∈ U} (7.19)
where m
F
(u) is called the membership function of u in U. When U is continuous, F
can be written as,
(7.20)
and when U is discrete, F is represented as,
(7.21)
where n is the number of elements in the fuzzy set F.
* Zadeh, L. A., “Fuzzy sets as a basis for a theory of possibility.” Fuzzy Sets and Systems, 10, (3),
243–260, 1978.
FIGURE 7.25 A fuzzy set of wind speed is represented by four fuzzy levels: “very low,”
“low,” “medium,” “high,” “very high.” By allowing varying degrees of membership, fuzzy
sets enables the process of decision making better under uncertainty.
µ (wind speed)
Very Low Low Medium High Very High
1.0
0.5
0 1 2 3 4 5 6 7
wind speed, m/s
F u u
F
u
( )
∫
µµ
F u
F
i l
n
∑
µµ
1306/C07/frame Page 132 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity 133
7.5.1 LINGUISTIC VARIABLES
Fuzzy set theory enables modeling a system in the natural language by making use
of linguistic variables. A linguistic variable is characterized by a quintuple,
(x, T(x), U, G, M)
where,
x = name of variable [Example: wind speed]
T(x) = Term set of x [Example: T(wind speed) = {very low, low, medium,
high, very high}]
U = Universe of discourse [Example: U(wind speed) = (0, 7) m/s]
G = Syntactic rule for generating the name of values of x
M = Semantic rule for associating a meaning with each value
The terms Very Low, Low, Medium, High, and Very High wind speeds represent
fuzzy sets whose membership functions are shown in of Figure 7.25.
If A and B are fuzzy sets with membership function µ
A
(u) and µ
B
(u), respectively,
then the membership function of the union, intersection, and complement, and the
fuzzy relation involving the two sets, are as follows:
Union (AND)
Example: IF A AND B THEN C
µ
C
(u) = µ
A∪B
= max(µ
A
(u), µ
B
(u)) u∈U
Intersection (OR)
Example: IF A OR B THEN C
µ
C
(u) = µ
A∩B
= min(µ
A
(u), µ
B
(u)) u∈U
Complement (NOT)
Example: NOT C
FUZZY RELATION
Two or more fuzzy IF/THEN rules of the form,
Y is B
i
IF X is A
i
, i = 1,2,… n
can be connected by the fuzzy relation R,
µ µ c u c u ( ) ( ) 1 –
1306/C07/frame Page 133 Saturday, May 27, 2000 11:35 AM
134 Powerline Ampacity System: Theory, Modeling, and Applications
(7.22)
where the membership function of the cartesian product (A
i
x B
i
) is given by,
(7.23)
One of the most successful applications of fuzzy sets is in the design of Fuzzy Logic
Controller (FLC). A fuzzy logic controller may be developed to control tranmission
line ampacity. The steps in FLC design follow:
• Deﬁne input and control variables. For example, in ampacity calculation,
the input variables are wind speed and direction, solar radiation, and air
temperature. The control variable is ampacity.
• Describe the input and control variables as fuzzy sets (fuzziﬁcation).
• Design the rule base (fuzzy control rules).
• Develop the fuzzy computational algorithm and fuzzy output.
• Transform fuzzy outputs to crisp control actions (defuzziﬁcation).
An important step in the design of FLC is the selection of membership function and
fuzzy IF/THEN rules. Generally, they are obtained by experimentation with data.
More recently, neural networks have been used to learn membership function and
the rules. Figure 7.26 is a schematic representation of a fuzzy logic controller with
learning by neural networks.
An example of a fuzzy logic system for the calculation of transmission line ampacity
is given in Figure 7.27. Only two rules are shown in the ﬁgure for illustration of
main concepts, and a calculation to obtain a crisp value of line ampacity from fuzzy
sets is given below. This is an excellent example of fuzzy logic because the mete
orological variables comprising wind speed and ambient temperature, are best
described by fuzzy sets. In the example, wind speed, ambient temperature and
ampacity are represented by fuzzy sets having four fuzzy levels (very low, low,
medium, high, very high). In this manner, we can represent weather parameters by
linguistic variables and consider the uncertainty in weather data comprising wind
speed and ambient temperature accurately. Furthermore, weather forecast data from
FIGURE 7.26 Schematic representation of fuzzy logic controller that learns from a
Neural network.
R A xB
i i
i
( )
∑
µ µ µ
A A n A l A
x l n
u u u
KK
KK KK
1
1 1
, min , ,
( )
( )
{
FLC Membership Function
and Rules
Transmission
line
Output
Neural Network
1306/C07/frame Page 134 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity 135
the National Weather Service or other sources are normally presented in a similar
manner, which can be directly utilized to calculate transmission line ampacity as
shown in Figure 7.27.
Example of Ampacity Calculation by Fuzzy Sets
Typical input data for ampacity calculation may be as follows:
Sunny, wind northwest 8–12 km/h, temperature low 2–15°C.
We are required to calculate ampacity from the above data.
The following two rules are activated:
Rule 1: If Ambient Temperature (T) is very low and Wind Speed (W) is
medium, then Ampacity (A) is high.
Rule 2: If Ambient Temperature is low and Wind Speed is high, then
Ampacity is very high.
The degree of membership of fuzzy variables T and W in the four fuzzy levels is
given in Table 1.
From Rule 1, the membership of input T denoted as m
VL
(T), and the membership
of input W denoted as m
M
(W), is obtained from Figure 7.27,
FIGURE 7.27 Fuzzy logic system of calculation of transmission line ampacity.
Temperature (T)
Very Low (T) Input (T)
Input (T)
Wind Speed (W)
Input (W)
Input (W)
Medium (W)
Ampacity (A)
High (A)
Rule 1
Rule 2
Low (T)
High (W)
Very High (A)
Final value of Ampacity (A)
Fuzzy centroid (A)
Rule 1: If Ambient Temperature (T) is Very Low and Wind
Speed (W) is Medium then Ampacity (A) is High
Rule 2: If Ambient Temperature is Low and Wind Speed (W) is
High then Ampacity (A) is Very High
1306/C07/frame Page 135 Saturday, May 27, 2000 11:35 AM
136 Powerline Ampacity System: Theory, Modeling, and Applications
m
VL
(T) = 0.75
m
M
(W) = 0.55
Therefore, Rule 1 activates the consequent fuzzy set H of A to degree
m
H
(A) = min(0.75, 0.55) = 0.55
Similarly, from Rule 2, the membership of input T (m
L
(T)), and the membership of
input W (m
H
(W)), are obtained from Figure 7.27,
m
L
(T) = 0.6
m
M
(W) = 0.75
Therefore, Rule 2 activates the consequent fuzzy set VH of A to degree
m
VH
(A) = min(0.6, 0.75) = 0.6
Therefore, the combined output fuzzy set of A, m
o
(y
j
) = (0,0,0,0.55,0.6).
A crisp value of ampacity is obtained by centroid defuzziﬁcation of the combined
output fuzzy set,
y = (1, 2.5, 3.5, 4.5, 5.5)
Where the elements y
j
of vector y are the mean value of each fuzzy level.
TABLE 1
Var/Level VL L M H VH
T (2–15) °C 0.75 0.6 0 0 0
W (2.5–3.2) m/s 0 0 0.55 0.75 0
A S
y m y
m y
f
j
j l
p
o j
o
j l
p
j
⋅
( )
( )
∑
∑
1306/C07/frame Page 136 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity 137
Ampacity = 2008 A
Since the membership functions given in Figure 7.26 are based on scaled values the
actual value of ampacity for a transmission line having an ACSR Cardinal conductor
is obtained by multiplying with a scaling factor, S
f
= 400.
7.6 SOLAR RADIATION MODEL
During daytime, the transmission line conductor is heated by the energy received
from the sun. Depending upon the condition of the sky and the position of the sun
with respect to the conductor, the temperature of the conductor may increase by
1–10°C by solar heating alone. To calculate conductor heating by solar radiation,
the energy received on the surface of the conductor from the sun (E
s
) is obtained as,
E
s
= α
s
(S
b
+ S
d
) (7.24)
Where,
α
s
= solar absorption coefﬁcient, 0 < α
s
< 1
S
b
= Solar energy received by conductor due to beamed radiation
S
d
= Solar energy received by conductor due to diffused radiation
The beamed radiation S
b
is calculated as,
S
b
= S
n
· cos(θ) (7.25)
The diffused radiation S
d
is calculated as,
S
d
= S
n
· cos(θ
z
) (7.26)
Where,
S
n
= the component of solar radiation that is normal to the surface of the earth
θ = Angle of deviation from the normal
θ
z
= Zenith angle, given by,
cos(θ
z
) = sin(φ) sin(δ) + cos(φ) cos(δ) cos(ω) (7.27)
The normal component of the beamed solar radiation inside the earth’s atmosphere
is obtained as,
A S
A S x
f
f
× + × + × + × + ×
+
0 5 0 1 5 0 2 5 0 4 5 0 55 5 5 0 6
0 55 0 6
5 02
. . . . . . .
. .
.
1306/C07/frame Page 137 Saturday, May 27, 2000 11:35 AM
138 Powerline Ampacity System: Theory, Modeling, and Applications
S
nb
= S
n(o)
· τ
b
(7.28)
The normal component of the diffused solar radiation inside the earth’s atmosphere
is obtained as,
S
nd
= S
n(o)
· τ
d
(7.29)
S
n(o)
= Normal component of the solar radiation outside the earth’s atmosphere which
is obtained as,
(7.30)
J
d
= 1,2..365 day number
S
c
= Solar constant = 1353 W/m
2
(measured value outside the earth’s atmo
sphere)
τ
b
= atmospheric transmittance of beamed radiation. It takes into account
attenuation by the earth’s atmosphere of the extraterrestrial radiation and
is given as,*
(7.31)
τ
d
= atmospheric transmittance of diffused radiation given by,
τ
d
= 0.2710 – 0.2939τ
b
(7.32)
a
0
= 0.4237 – 0.00821(6Alt)
2
a
1
= 0.5055 – 0.00595(6.5Alt)
2
k = 0.2711 – 0.01858(2.5Alt)
2
Alt = Altitude of the transmission line above MSL, km
The angle of deviation θ of the beamed radiation with respect to the normal to
surface of the conductor is given by the following formula,
cos(θ) = sin(δ)sin(φ)cos(β)sin(δ)cos(φ)sin(β)cos(γ)
– sin(δ)cos(φ)sin(β)cos(γ)
+ cos(δ)cos(φ)cos(β)cos(ω) + cos(δ)sin(φ)sin(β)cos(γ)cos(ω)
+ cos(δ)sin(φ)sin(β)cos(γ)cos(ω) + cos(δ)sin(β)sin(γ)sin(ω) (7.33)
* Dufﬁe, John A. and Beckman, William A. 1980 Solar Engineering of Thermal Processes, John Wiley
& Sons, New York.
S S Cos
J
n o c
d
( )
+ ⋅
⋅

(
'
`
J
J

¦




1 0 0033
360
365
.
τ
θ
b
z
a a
k
Cos
+
( )

(
'
`
J
J 0 1
exp
–
1306/C07/frame Page 138 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity 139
φ = Latitude. Angle between north or south of the equator, north +ve:
–90° =< φ =< 90°
δ = Declination. Angular position of the sun at solar noon with respect to the
plane of the equator, north +ve: –23.5° =< δ =< 23.5°. The declination
angle is calculated as,
(7.34)
β = Slope. Angle between the conductor axis and the horizontal: 0 =< β =<
180°
γ = Line orientation angle (azimuth). South zero, East ve, West +ve: –180°
=< γ=< 180°
ω = Hour angle. Angular displacement of the sun east or west of the local
meridian due to rotation of the earth on its axis at the rate of 15°/hr, morning
–ve, afternoon +ve. The hour angle is given by,
ω = (12 – SolarTime) ⋅ 15 (7.35)
SolarTime = StandardTime + 4(L
std
–L
loc
) + Eqt (7.36)
L
std
= Longitude of Standard Meridian (Example: L
std
San Francisco = 120°)
L
loc
= Longitude of location
Eqt = Equation of time = 9.87sin(2B) –7.53cos(B) –1.5sin(B) (7.37)
(7.38)
θ = Angle of incidence. The angle between the beam radiation on a surface
and the normal to the surface. These angles are shown in Figures 7.28 and
7.29. The result of solar radiation calculation by the program is for one day
during the month of July in the region of San Francisco and is shown in
Figure 7.30.
Figure 7.31 is a ﬂow chart showing the line ampacity forecasting procedure from
weather models AmbientGen, WindGen, and SolarGen.
7.7 CHAPTER SUMMARY
In this chapter, weather modeling for the prediction of transmission line ampacity
is presented ﬁrstly by Fourier analysis of weather data. Ambient temperature and
δ
+
( )

¦








23 45 360
284
365
.
J
d
B
J
d
( )
360 81
364
–
1306/C07/frame Page 139 Saturday, May 27, 2000 11:35 AM
140 Powerline Ampacity System: Theory, Modeling, and Applications
wind speed models were developed by ﬁtting Fourier series to hourly weather data
available from the National Weather Service. A Kalman ﬁltertype algorithm is then
used to model the uncertainty in the Fourier series, and a realtime forecasting
algorithm is presented that uses a recursive estimation procedure for the prediction
of ambient temperature and wind speed. The forecasts are adapted to the daily high
and low values of ambient temperature that are forecast daily by the National Weather
Service.
FIGURE 7.28 Transmission line solar angles.
FIGURE 7.29 Slope angle between conductor and horizontal.
FIGURE 7.30 Global Solar Radiation (Direct Beam + Diffused Radiation) on a transmission
line conductor surface in San Francisco, calculated by program for one day during the month
of July. Transmission Line EW direction and the slope angle is 5 °.
Atmosphere
Beamed
Radiation
Earth
Normal
Conductor
Tower
Equator
Sun
90
0
0
Time of Day, hr
Global Solar Radiation
San Francisco, CA, July
W
/
m
2
0 4 8 12 16 20 24
0
200
400
600
800
1000
1200
1400
1306/C07/frame Page 140 Saturday, May 27, 2000 11:35 AM
Weather Modeling for Forecasting Transmission Line Ampacity 141
Neural network models for prediction and weather pattern recognition are devel
oped by using the back propagation algorithm, and a selforganizing neural network
is developed using Kohonen’s learning algorithm. Neural networks offer an alterna
tive method of forecasting weather variables and pattern recognition.
Basic fuzzy logic concepts are presented and a system for developing a fuzzy
logic controller of transmission line ampacity is proposed for further research.
Analytical expressions for the calculation of hourly values of solar radiation are also
developed which take into account transmission line location and line geometry.
FIGURE 7.31 Flow chart for forecasting transmission line ampacity from weather models.
FIGURE 7.32 Forecasting hourly ampacity values of a 220 kV transmission line seven days
in advance in the region of North Island, New Zealand.
Input forecast weather data:
Ta. Ws. Sr
Method AmbientGen
generates hourly vcalues T a(t)
Method WinGen
generates hourly values Ws(t)
Method SolarGen
generates hourly values Sr(t)
Method Steady Static Ampacity
generates hourly values of
Transmission Line Ampacity at each
line segment (I )
Calculate the minimum value of
ampacity of all line segments
Min (I )
j,t
j,t
1306/C07/frame Page 141 Saturday, May 27, 2000 11:35 AM
142 Powerline Ampacity System: Theory, Modeling, and Applications
For realtime prediction of transmission line ampacity, a recursive estimation
algorithm for weather forecasting is developed based on Kalman ﬁlter equations.
Examples of weather models developed by the program are shown for the region of
San Francisco.
1306/C07/frame Page 142 Saturday, May 27, 2000 11:35 AM
143
8
Computer Modeling
8.1 INTRODUCTION
For the proper evaluation of transmission line capacity, it is necessary to support
the theory developed in Chapter 4 by a practical knowledge of the transmission
system and its environment. Line ampacity is calculated from steady, dynamic, and
transient thermal models by developing an object model and expert rules of the
transmission line system. This is the object of computer modeling of line ampacity
system as described in this chapter.
In this chapter the LINEAMPS (Line Ampacity System) transmission line expert
system computer program is developed by the integration of theory and practical
knowledge of the transmission line system. Examples of objectoriented modeling
and expert system rules are presented here to demonstrate how practical knowledge
is embedded in program, which will enable a transmission line engineer to easily
evaluate power line capacity in any geographic region.
8.1.1 F
ROM
T
HEORY
TO
P
RACTICE
A transmission line is composed of the conductors that carry current, structures that
support conductors in air, insulators to safely protect transmission tower structures
from the high voltage conductor, connectors for the splicing of conductors, and other
hardware necessary for the attachment of conductors to the tower. The transmission
line environment is vast as they traverse different kinds of terrain—plains, forests,
mountains, deserts, and water. They are also exposed to varying atmospheric con
ditions—sun, wind, temperature, and rain. Some sections of a line may be exposed
to industrial pollution as well. All of these environmental factors affect line capacity
to a certain degree. Modeling such a system is not easy. A systematic approach using
practical knowledge and simplifying assumptions is required to achieve a realistic
line ampacity system with sufﬁcient accuracy. This is the object of the line ampacity
expert system program.
8.1.2 T
HE
LINEAMPS E
XPERT
S
YSTEM
LINEAMPS*
is an expert system computer program developed by the author based
on a systematic approach of objectoriented modeling and expert rules. Objects are
used to model the transmission line system and its environment. Expert system rules
are used to incorporate practical knowledge. The end product is an intelligent line
ampacity system resembling a human expert. It is also a humble contribution and a
* Anjan K. Deb, Object oriented expert system estimates transmission line ampacity
, IEEE Computer
Application in Power
, Volume 8, Number 3, July 1995.
1306/C08/frame Page 143 Saturday, May 27, 2000 12:19 PM
144
Powerline Ampacity System: Theory, Modeling, and Applications
practical demonstration of research in the ﬁeld of artiﬁcial intelligence,* where a
computer system demonstrates intelligent behavior. In the following sections the
objectmodel and expert system design of the line ampacity system are presented
in greater detail.
8.2 OBJECT MODEL OF TRANSMISSION LINE
AMPACITY SYSTEM
System modeling by object orientation is a new way of data representation and
programming.** Its attributes and behavior deﬁne an object. Objects can store data,
whether it is temporary or permanent. Methods or stored procedures in the objects
give them behavior, which enables them to perform a required action. Methods have
the ability to use data contained in their own objects as well as other objects. For
example, data stored in weather objects of the line ampacity system program are
environmental data comprising weather conditions, terrain, latitudes, longitude, and
elevation. Data relating to the electrical and mechanical properties of the transmis
sion line are contained in the transmission line object.
Once an object is created it is easier to create newer instances of the same
object by inheritance. Objects inherit data as well as methods—this is an important
property of all objects. For example, once a transmission line object is created,
several lines may be produced by inheritance. Similarly, weather station objects
are created. These objects have methods to receive weather forecast data from
external sources, for example, the National Weather Service, the Internet, or even
the daily newspaper. By using forecast weather data, as well as the stored proce
dures and weather patterns of the region, hourly values of transmission line
ampacity are forecast several hours in advance. The object model of the line
ampacity system is given in Figure 8.1, and the hierarchical structure of transmis
sion line, weather station, and conductor object of the transmission line ampacity
system is shown in the Figures 8.2,8.3, and 8.7.
8.2.1 LINEAMPS O
BJECT
M
ODEL
The object model of the LINEAMPS system shown in Figure 8.1 is comprised of
transmission line object, weather station object, conductor object, and cartograph
object. In addition, there is a system of objects for the presentation of data, and a
userfriendly graphical user interface. The KappaPC*** object oriented modeling
tool is used to implement the object model. The following sections describe the
object model and expert rules of the line ampacity system.
* Lawrence Stevens, Artiﬁcial Intelligence. The Search for the Perfect Machine, Hayden Book Company,
1993.
** G. Booch, ObjectOriented Design with Applications, Benjamin Cummings Publishing Co. 1991.
*** KappaPC® objectoriented development software v 2.4, 1997, Intellicorp, Mountainview, CA.
1306/C08/frame Page 144 Saturday, May 27, 2000 12:19 PM
Computer Modeling
145
8.2.2 T
RANSMISSION
L
INE
O
BJECT
The Line object class is shown in the Figure 8.2. Powerline objects are classiﬁed
by categories of transmission and distribution lines. In each class there are subclasses
of lines by voltage levels. In each subclass there are several instances of powerlines,
for example, the transmission line object class is comprised of 500kV, 345kV, and
230kV lines. The distribution line class comprises 15kV, 480 V, and other line
instances created by the user. Each of these transmission line objects derives its
attributes and behavior from the general class of lines.
The subclasses of line voltages in the line object classes are deﬁned by the user.
The line object has all the data and methods pertaining to the overhead line that are
necessary for the evaluation of powerline ampacity. Data are stored in slots, and
methods perform the action of evaluating ampacity. Table 8.1 shows an example of
the data in one instance of a transmission line object. Only a partial list of attributes
and methods are shown for the purpose of illustration.
Following is a complete list of attributes of the line object:
Line object attributes:
LineName
LineVoltage
LineLength
ConductorCodeName
ConductorType
ConductorDiameter
FIGURE 8.1
Line ampacity system objectmodel.
1306/C08/frame Page 145 Saturday, May 27, 2000 12:19 PM
146
Powerline Ampacity System: Theory, Modeling, and Applications
ConductorArea
ConductorAlpha
ResistanceAtTemperature
Conductor DC Resistance
ConductorEmissivity
ConductorAbsorbtivity
ConductorSpeciﬁcHeat
ConductorMass
ConductorAluminumMass
ConductorSteelMass
NumberOfSites
SiteList (List)
AssociatedWeatherSites(List)
Site#x(List), where x =1,2..NumberOfSites
AmbientTemperatureSite#x(List)
WindSpeedSite#x(List)
WindDirectionSite#x(List)
SkyConditionSite#x(List)
NormalAmpacityOneDay(List)
NormalAmpacitySevenDays(List)
EmergencyAmpacityOneDay(List)
EmergencyAmpacitySevenDays(List)
TimeOfDayEnergyPrice(List)
Each Site#x comprises a list having the following values: elevation, slope, latitude,
longitude, standard longitude, line orientation at the location, and the type of terrain.
Line object comprise the methods shown in Table 8.2.
A graphical representation of transmission line by voltage category and by the
type of line, transmission or distribution, greatly facilitates the task of a transmission
line engineer to view and modify line data by simply clicking on a transmission line
object shown in Figure 8.2.
The following is an example of creating a transmission line object, assigning
attributes and giving them behavior.
TABLE 8.1
Transmission Line Object: 350kV_Line10
Attributes Values Method
Name San Francisco, Berkeley Ampacity
Distance 50 km Steady_State_Ampacity
Conductor acsr cardinal Dynamic_Ampacity
Duration 30 min Transient_Ampacity
Region Coastal Draw Line
1306/C08/frame Page 146 Saturday, May 27, 2000 12:19 PM
Computer Modeling
147
TABLE 8.2
Method: Function:
Amp7Days Calculates hourly values of line ampacity up to seven days in advance.
SteadyStateCurrent Calculates steady state current.
SteadyStateTemperature Calculates steady state temperature.
DynamicAmpacity Calculates conductor temperature response due to step change in line
current.
TransientAmpacity Calculates conductor temperature response due to short circuit and
lightning current.
DrawLine Draws the line in the cartogram window.
CheckInput Veriﬁes the correctness of input data.
WeatherData Obtains data from the associated weather stations.
AdjustWeather Weather data is adjusted in AmbientTemperatureSite#x slot and
WindSpeedSite#x slot based on terrain in Site#x slot.
MakeNewLine Makes an instance of a new line.
UpdateLineList Updates the list of lines when a new line is created.
MakeSiteData Makes virtual weather sites along the route of the line.
EnergyDeliveryCost Calculates hourly values of energy delivery cost based on time of day
energy price and forecast ampacity.
FIGURE 8.2
Classiﬁcation of transmission line objects.
1306/C08/frame Page 147 Saturday, May 27, 2000 12:19 PM
148
Powerline Ampacity System: Theory, Modeling, and Applications
Creating the object
MakeClass(Transmission, Lines);
MakeInstance(350kV_Line10, Transmission);
Attributes
MakeSlot(350kV_Line10, Name, SanFrancisco_Danville, 50);
MakeSlot(350kV_Line10, Distance, 50);
MakeSlot(350kV_Line10, Conductor, Cardinal);
Behavior
MakeMethod(Transmission, Ampacity, [time, ambient, sun, wind, interval]);
Action
SendMessage(350kV_Line10, Ampacity, [12:00, 20, C, 2, 60]);
Result:
1000 A
8.2.3 W
EATHER
S
TATION
O
BJECT
Since LINEAMPS calculates transmission ampacity from weather data, modeling
of weather by developing an object model is an important aspect of this program.
The purpose of the weather object is to reproduce, as closely as possible by software,
the behavior of an actual weather station. This is the main objective of the LIN
EAMPS weather station object. The behavior of a weather station object is obtained
by modeling weather patterns of the region by Fourier analysis from historical
weather data and regional weather forecasts prepared daily by the National Weather
Service.
To enable modeling of a weather station object, it is divided into subclasses of
regions and region types so that weather station instances inherit class attributes.
Station objects have all of the meteorological data and geographic information
required in the calculation of transmission line ampacity. The weather station object
hierarchy is shown in Figure 8.3 and is comprised of:
1. Subclass of regions. Example: Region1, Region2… Region#x.
2. Subclass of region types. Example: Coastal, Interior, Mountain, Desert.
3. Instances of weather stations. Example: San Francisco, Oakland, Liver
more.
Weather station objects have the following attributes.
Attributes of weather station object
• StationName
• AmbientMax(List)
1306/C08/frame Page 148 Saturday, May 27, 2000 12:19 PM
Computer Modeling
149
• AmbientMin(List)
• HourAmbientMax
• HourAmbientMin
• AmbientPattern#x(List), where x =1,2...12 months
• WindSpeedMax(List)
• WindSpeedMin(List)
• HourWindSpeedMax
• HourWindSpeedMin
• WindSpeedPattern#x(List), where x =1,2...12 months
• WindDirection(List)
• SkyCondition(List)
• ForecastTemperature(List)
• ForecastWind(List)
• ForecastSolarRadiation(List)
• Latitude
• Longitude
A “List” inside parentheses is used to indicate that the attribute has a list of
values. Weather station objects comprise the following methods:
N
EW
Z
EALAND
E
XAMPLE
*
To ﬁx ideas, an example of
New Zealand weather station object is presented in
Figure 8.3. The object has North and South Island subclasses. North Island is Region
1, and South Island Region 2. Each region is further divided into subclasses of
Coastal, Interior, Mountain, and Desert. In each subclass there are instances of
weather stations. These instances derive their attributes and behaviors from the
general class of weather stations, and their characteristics are reﬁned by the prop
erties of each region
Table 8.4 shows data in one instance of a weather station object.
TABLE 8.3
Method Function
AmbientGen Generates hourly values of ambient temperature, (Figure 8.4).
WindGen Generates hourly values of wind speed, Figure 8.5.
SolarGen Generates hourly values of solar radiation, Figure 7.3
SelectPattern Selects ambient temperature and wind speed pattern of the month
DisplayAmbient Display ambient temperature in a line plot and a transcript image
DisplayWind Display wind speed in a line plot and a transcript image
OnLineData Reads weather data downloaded from AmericaOnLine.
MakeNewStation Makes an instance of a new weather station.
* Deb, Anjan K.,
LINEAMPS for New Zealand, A Software User’s Guide
, 1996.
1306/C08/frame Page 149 Saturday, May 27, 2000 12:19 PM
150
Powerline Ampacity System: Theory, Modeling, and Applications
Examples of Ambient_Gen and Wind_Gen methods used to generate hourly
values of ambient temperature and wind speed are shown in Figures 8.4 and 8.5.
Seven days’ forecast weather data from the National Weather Service are shown in
Figure 8.6.
8.2.4 C
ONDUCTOR
O
BJECT
A conductor object class is shown in Figure 8.7. It is comprised of the following
subclasses of conductor types: AAAC, AAC, ACAR, SSAC, ACSR_AW, ACSR,
ACSR_TW, COPPER, COPPERWELD and ACSR_INVAR. Each conductor type
has plurality of conductor instances. The user may also create other subclasses of
conductor types and new instances of conductor objects.
FIGURE 8.3
Classiﬁcation of weather station objects.
TABLE 8.4
Weather Station Object:
Wellington Example
Attribute Value Method
Name Wellington Ambient_Gen
Latitude 41° 18 ‘ S Wind_Gen
Longitude 174° 47’ E Solar_Gen
Elevation 100 Draw_on_Map
Region Coastal Show_Value
1306/C08/frame Page 150 Saturday, May 27, 2000 12:19 PM
Computer Modeling
151
FIGURE 8.4
Temperature modeling.
FIGURE 8.5
Wind speed modeling.
FIGURE 8.6
National Weather Service seven day weather forecast.
1306/C08/frame Page 151 Saturday, May 27, 2000 12:19 PM
152
Powerline Ampacity System: Theory, Modeling, and Applications
Conductor object has following attributes:
Attributes of Conductor object:
Conductor code name
Conductor type
Conductor diameter
Conductor area
DC resistance of conductor
Emissivity of conductor
Absorptivity of conductor
Speciﬁc heat of conductor
Conductor mass
Aluminum mass
Steel mass
Conductor object has following methods:
Data pertaining to one instance of a transmission line conductor is shown in
Figure 8.8.
8.2.5 C
ARTOGRAPH
O
BJECT
A cartograph window is used to show the location of weather stations and the
transmission line route in a geographic map of the region, as seen in Figure 8.9.
FIGURE 8.7
Classiﬁcation of transmission line conductor objects.
1306/C08/frame Page 152 Saturday, May 27, 2000 12:19 PM
Computer Modeling
153
The maximum and minimum ambient temperatures of the day are also displayed at
the location of each weather station. By displaying the transmission line in a map,
one obtains a better picture of the transmission line route and its environment.
A unique feature of the program realized by objectoriented modeling is the
ability to create new lines and weather stations by inheritance. A transmission line
object is created by entering the latitudes and longitudes of the line at discrete
intervals, and by specifying the type of terrain through which the line passes.
Similarly, weather station objects are also created. A transmission line appears on
the map when the line is selected from the database. It is generated automatically
by the program with a DrawLine method using data stored in the transmission line
object shown in Figure 8.9.
TABLE 8.5
Methods Function
SpeciﬁcHeat Calculates the speciﬁc heat of conductor
SteadyStateCurrent Calculates steady state current
SteadyStateTemperature Calculates steady state temperature
DynamicAmpacity Calculates conductor ampacity n the dynamic state.
DynamicTemperature Calculates conductor temperature versus time in the dynamic state.
TransientAmpacity Calculates conductor temperature versus time in the transient state.
MakeNewConductor Makes an instance of a new conductor
FIGURE 8.8
Data in a transmission line conductor object.
1306/C08/frame Page 153 Saturday, May 27, 2000 12:19 PM
154
Powerline Ampacity System: Theory, Modeling, and Applications
8.3 EXPERT SYSTEM DESIGN
The line ampacity expert system is accomplished by a system of rules and goals to
be achieved by the program. Expert systems are capable of ﬁnding solutions to a
problem by a description of the problem only. The rules and the data in the objects
describe the problem. This declarative style of rule based programming offers an
alternative to the traditional procedural programming method of solving problems.
For example, to solve a problem by rules, we specify what rules to apply and a goal.
Reasoning is then carried out automatically by an inference engine, which ﬁnds a
solution by using a backward or forward chaining mechanism.*
An expert system is generally composed of the following:
• Goals, facts, database
• Rules or knowledge base
• Inference engine (reasoning capability)
• Explanation facility
• Man machine interface
• Learning capability
FIGURE 8.9
A cartograph is shown in the right window of the LINEAMPS program and
the corresponding weather station objects are shown in the left window. The cartograph shows
the geographic boundary of the region that was created by program. The location of weather
stations, daily maximum and minimum values of ambient temperature (max/min) and the
trace of a selected 230 kV transmission line from Sacramento to Livermore, California USA
is shown in the cartograph window.
* Waterman, Donald A.,
A Guide to Expert Systems
. AddisonWesley, Reading, MA, 1986.
1306/C08/frame Page 154 Saturday, May 27, 2000 12:19 PM
Computer Modeling
155
In the following section, the transmission line expert system is described by
presenting an example of goaloriented programming, rules, inference engine, and
explanation facility. These features were used in the program to check user input
data and explain error messages. An example of a man machine interface was given
in a previous IEEE publication.* Learning by artiﬁcial neural network is described
later in this chapter.
8.3.1 G
OAL
O
RIENTED
P
ROGRAMMING
Goaloriented programming by rules greatly facilitates the task of computer pro
gramming as the programmer is not required to code a detailed logic to solve a
problem. In traditional programming by procedures, a programmer must precisely
code the logic, of a mathematical equation for example, to solve a problem. Rules
not only facilitate a declarative style of programming, but also provide a practical
method of incorporating practical and imprecise knowledge such as “rules of thumb”
that are not easily amenable to formal mathematical treatment. In addition, rules are
easily understood and maintained. Following is a simple example of programming
by rules and a goal in the line ampacity system.
Goal: SteadyStateGoal
Action:
{ SetExplainMode( ON );
ForwardChain( [ ASSERT ], SteadyStateGoal, Global:SteadyStateRules );
If ( Transmission:Problem # = N )
Then CalcSteadyTemperature( );
};
The object of the above action statement is to satisfy a SteadyState Goal by verifying
all of the steadystate rules stored in the Object:Slot pair Global:SteadyStateRules. In
the KappaPC objectoriented development environment, slots are provided to store
data of an object. The program proceeds with the calculation of transmission line
conductor temperature only if there are no problems detected in the data entered by the
user. Setting the explain mode to ON enables the user to receive explanations of expert
system generated error messages. When new facts are generated by the ﬁring of rules,
[ASSERT] ensures that the new facts are automatically inserted into a fact database.
Result:
In the above example, the user input data were checked by the expert system rules.
The SteadyStateGoal was satisﬁed and the program correctly evaluated steadystate
conductor temperature to be equal to 60°C.
In the following example, the user entered a value of conductor temperature less
than ambient temperature. The expert system correctly detected the problem and
* Deb, Anjan, K., Object oriented expert system estimates transmission line ampacity,
IEEE Computer
Application in Power
, Volume 8, Number 3, July 1995.
1306/C08/frame Page 155 Saturday, May 27, 2000 12:19 PM
156
Powerline Ampacity System: Theory, Modeling, and Applications
generated an error message, as shown in Figure 8.10. By clicking on the Explain
button, the expert system generated the required explanations, and the story of the
transmission line ampacity problem started to unfold (Figure 8.11).*
FIGURE 8.10
Example of error message given by program when user entered incorrect
value of conductor temperature.
FIGURE 8.11
Explanation of error message given by program when user clicks on the
explain button.
*
Towards the Learning Machine
, Richard Forsyth.
1306/C08/frame Page 156 Saturday, May 27, 2000 12:19 PM
Computer Modeling
157
8.3.2 E
XPERT
S
YSTEM
R
ULES
The expert system knowledge base is comprised of the abovementioned system of
objects and rules. In LINEAMPS, rules are used to offer expert advice to users in
the event of erroneous input or conﬂicting data, or to caution the user during speciﬁc
operating conditions. Some examples of rules used in the program are:
Rule 1.
If ambient temperature is greater than conductor temperature, then advise user.
Rule 2.
If the temperature of the selected conductor is greater than the allowable
maximum for the conductor type, then advise user.
Rule 3.
If user input, is low 2 or 4 ft/s wind speed, and the National Weather Service
forecast is high wind speed, then advise user.
Rule 4.
In the dynamic state, if preload current results in a higher than maximum
allowable conductor temperature, then advise user.
Rule 5.
In the dynamic state, if post overload current results in a higher than maximum
allowable conductor temperature, then advise user.
Rule 6.
In the dynamic state, if the user speciﬁed preload current results in a conductor
temperature that is higher than the allowable maximum, then advise user.
Rule 7.
In the transient state, if the duration of transient current is greater than the
speciﬁed maximum, then advise user.
Rule 8.
If the age of the conductor is old and the conductor temperature is high, then
advise user.
Rule 9.
If the age of the conductor is old or the line passes through areas of industrial
pollution, and the coefﬁcient of solar absorbtivity is low and/or the emissivity
of the conductor is high, then advise user.
Rule 10.
In the transient state, if the transient current is high and the line is old, then
advise user.
Rule 11.
If the line passes through urban areas with highrise buildings or where wind
is restricted by tall structures or trees and line ampacity is high, then advise
user.
1306/C08/frame Page 157 Saturday, May 27, 2000 12:19 PM
158
Powerline Ampacity System: Theory, Modeling, and Applications
Rule 12.
If the value of conductor emissivity is less than or equal to 0 or greater than
1, then advise user.
Rule 13.
If the value of conductor absorbtivity is less than or equal to 0 or greater than
1 then advise user.
The following example shows a listing of two rules used by the program.
/***************************************************
**** RULE 1: Conductor Rule
**** If conductor temperature is greater than the allowed
**** maximum for the conductor type, then reduce current
***********************************************/
MakeRule( ConductorRule, [],
Not( Member?( ConductorTypes:AllowableValues, Steady:ConductorType ) )
Or Steady:Temperature > Steady:ConductorType:MaxTemperature,
{
Transmission_Line_Ampacity:Problem = “High conductor temperature”;
PostMessage( "Please check valid conductor type and conductor temper
ature" );
} );
SetRuleComment( ConductorRule, “If conductor temperature is greater
than the
allowed maximum for the conductor type, then reduce current” );
/***************************************************
**** RULE 2: SteadyState Temperature
**** Conductor temperature must be greater than
**** ambient temperature.
***************************************************/
MakeRule( SteadyStateTemperature, [],
Steady:Calculate #= Current And
Steady:ShowTemperature <= Steady:AmbientTemperature,
{
Line_Ampacity:Problem = “Conductor temperature”;
PostMessage( “Check ambient temperature” );
} );
SetRuleComment( SteadyStateTemperature,
“Conductor temperature must be greater than ambient temperature”);
In the above steadystate analysis window of the LINEAMPS program, the value
of conductor temperature entered by the user is 35°C, which is inconsistent with
the value of the ambient temperature, 40°C. One of the rules in the expert system
detects this problem and displays the error message shown in Figure 8.10.
After receiving an explanation, users also have the ability to request other
relevant information regarding various input data required in this session window.
In Figure 8.11 explanations of temperature, emissivity, and solar radiation are given.
1306/C08/frame Page 158 Saturday, May 27, 2000 12:19 PM
Computer Modeling
159
8.4 PROGRAM DESCRIPTION
8.4.1 LINEAMPS W
INDOWS
In this section, the main features and functions of the LINEAMPS program are
brieﬂy described. LINEAMPS is designed as a system of windows where users
conduct various sessions with the program on different aspects of power line ampac
ity. Following are the main session windows that are presently available in the
program for the analysis and planning of transmission line ampacity.
• LINEAMPS Control Panel
• SteadyState Analysis
• Dynamic Analysis
• Transient Analysis
• Forecast Ampacity
• Power Lines
• Conductors
• Weather
8.4.2 M
ODELING
T
RANSMISSION
L
INE
AND
E
NVIRONMENT
The following additional session windows are available for the modeling of a
transmission line and its environment, conductor modeling, weather modeling, and
cartograph. These windows may be opened by clicking icons, or by selection from
the LINEAMPS Control Panel Window.
• Power Lines
• Conductor
• Static Rating
• Cartograph
• Ambient Temperature
• Wind Speed
• Daily Weather Forecast
• Extended Weather Forecast
• Object Browser
The functions and operations of the LINEAMPS program in each window are
described in detail in the user manual.*
8.4.3 LINEAMPS C
ONTROL
P
ANEL
The program is operated by a system of icons, each representing a unique function.
While full details of the program is presented in the user manual, a brief description
of each icon shown in the Control Panel window (Figure 8.12) is presented here.
The control panel window has the following icons:
*
LINEAMPS User Manual
, 1998.
1306/C08/frame Page 159 Saturday, May 27, 2000 12:19 PM
160
Powerline Ampacity System: Theory, Modeling, and Applications
SteadyState Analysis
Clicking this button opens the SteadyState Analysis window. It is used
for the analysis and display of transmission line ampacity and conductor
temperature in the steady state.
Dynamic Analysis
Clicking this button opens the Dynamic Analysis window. It is used for
the analysis and display of transmission line ampacity and conductor
temperature in the dynamic state.
Transient Analysis
Clicking this button opens the Transient Analysis window. It is used for
the analysis and display of transmission line ampacity and conductor
temperature in the transient state.
Forecast Ampacity
Clicking this button opens the Forecast Ampacity window. It is used for
the calculation and display of hourly values of transmission line ampacity
up to seven days in advance.
Extended Weather Forecast
The Extended Weather Forecast window is opened by clicking this button.
It is used to view the forecast weather data of the region, as well as for
input of forecast weather data.
Ambient Temperature Model
Clicking this button opens the Ambient Temperature Model window. It is
used for selecting weather patterns and for generating hourly values of
ambient temperature data from forecast weather data.
Wind Speed Model
Clicking this button opens the Wind Speed Model window. It is used for
selecting weather patterns and for generating hourly values of wind speed
data from forecast weather data.
FIGURE 8.12 LINEAMPS Control Panel window.
LINEAMPS: Control Panel
D W A B H atabase eather nalysis rowser elp
$
?
Exit
EE
EE
1306/C08/frame Page 160 Saturday, May 27, 2000 12:19 PM
Computer Modeling 161
Cartograph
Clicking this button opens the Cartograph window. It is used to view a
geographic map of the region, the location of weather stations, and the
transmission line route.
Powerlines
Clicking this button opens the Powerlines window. This window is used
for creating new lines, and viewing and updating data in existing line
objects.
Conductor
Clicking this button opens the Conductor window. It is used for conductor
selection, creating new transmission line conductors, and viewing data in
conductor objects.
Transmission Cost
Clicking this button opens the hourly Transmission Cost window. This
window is used to generate hourly values of transmission costs for seven
days in advance.
Welcome
The Welcome window is opened by clicking this button. It is used for
navigating the LINEAMPS system of windows.
Help
Clicking this button opens the Help window. This window is used for
generating help text on userrequested topics.
Print
Clicking this button opens the Print window. It is used to print LIN
EAMPSgenerated reports on steadystate analysis, dynamic analysis, and
transient analysis, and forecast hourly values of ampacity for seven days
in advance.
Exit
Exit the program by clicking this button.
$
?
Exit
1306/C08/frame Page 161 Saturday, May 27, 2000 12:19 PM
162 Powerline Ampacity System: Theory, Modeling, and Applications
8.5 CHAPTER SUMMARY
This chapter describes computer modeling of the transmission line ampacity system
by objects and rules. The theory developed in the previous chapter and practical
knowledge of the transmission line system are implemented in the LINEAMPS
program by objectoriented modeling and expert rules.
The object model of the complete line ampacity system is presented, followed
by the component object models consisting of transmission line object, weather
station object, conductor object, and a cartograph object. The creation of objects,
and methods and the sending of messages between objects are presented to show
how an elegant system of objects having messaging capability is realized in the
LINEAMPS program.
A combination of procedures, goals, and rules are used to ﬁnd a solution to the
powerline ampacity problem. Procedures or methods are used in objects when a
mathematical model is available. Rules are used to incorporate practical knowledge.
Decisions based on rules are generated automatically by an inference engine. Exam
ples are presented to show how the system calculates powerline ampacity by objects
and rules, checks user input, and explains error messages to the user, thus demon
strating the intelligent behavior like a true expert.
A brief description of the various functions and features of the program and the
graphical user interface are presented to demonstrate the realization of a complete
line ampacity system that is suitable for all geographic regions.
1306/C08/frame Page 162 Saturday, May 27, 2000 12:19 PM
163
9
New Methods of
Increasing Transmission
Capacity
9.1 INTRODUCTION
AC transmission circuits are mostly composed of passive elements having very little
controllability. Therefore, when the load increases, existing network control methods
are not sufﬁcient to properly accommodate increased power ﬂows. As a result, certain
lines are more heavily loaded and stability margins are reduced. Due to the difﬁculty
of controlling power ﬂows by existing methods, new types of power electronic
devices called FACTS (Flexible AC Transmission System) are used to control exist
ing T&D networks. FACTS devices are installed at all voltage levels up to 800 kV.
FACTS devices are also used in lowvoltage distribution networks. Due to greater
utilization of electrotechnologies in the industry, and increasing use of high tech
nology electronic equipment like computers, TV, and other similar devices in homes
and businesses, consumers are paying greater attention to power quality in distribu
tion circuits. FACTS devices are, therefore, used in power distribution networks to
maintain power quality, and voltage regulation, and to lower harmonics and minimize
voltage ﬂicker.
This chapter presents an overview of FACTS and the various power semicon
ductor devices it uses. When transmission capacity increases are planned by dynamic
thermal ratings and reconductoring of existing circuits with special or highersize
conductors, it will be necessary to properly evaluate their impact on power system
performance and load ﬂows. Installation of FACTS devices may be required for
better control of the current and voltage in a transmission line. It is hoped that by
gaining knowledge of the various FACTS technologies presented in this chapter, a
power transmission and distribution line engineer will make better decisions regard
ing the selection of FACTS technology best suited for his or her requirements.
9.2 ADVANCEMENT IN POWER SEMICONDUCTOR
DEVICES
Recent advances in power electronics semiconductor devices have made it possible
to develop equipment with sufﬁcient current and voltage ratings to enable their
utilization in electric power circuits for better control of voltage and current. An
important application of power electronic semiconductor devices in electric networks
1306/C09/frame Page 163 Saturday, May 27, 2000 4:08 PM
164
Powerline Ampacity System: Theory, Modeling, and Applications
is to control the amount of current that can pass through the device in response to
a control action. The most common semiconductor device used for the control of
current ﬂowing through a circuit is a siliconcontrolled rectiﬁer device called a
“thyristor,” shown in Figure 9.1.
T
HYRISTOR
A thyristor is a threeterminal device with an anode, cathode, and gate, as shown in
Figures 9.29.5. It is a special type of diode. Like a diode, a thyristor requires a
certain positive anodetocathode voltage, but unlike a diode it also requires a pulse
having a certain voltage and current to be applied to the gate to turn on the thyristor
for conduction to begin from anode to cathode. Similar to a diode, conduction takes
place during the positive half of the AC cycle, when the anodetocathode voltage
is positive. It is blocking during the negative half of the AC cycle when the anode
tocathode voltage becomes negative.
The current through a thyristor is regulated by controlling the gatetriggering
time. This is accomplished by supplying a current pulse to the gate at a desired
triggering time. The ability to control current in a circuit by controlling the ﬁring
angle,
θ,
of a thyristor is illustrated by a simple circuit in Figure 9.6.
Considering a resistive load, R, the instantaneous load current, i
r
, is
FIGURE 9.1
Thyristor
FIGURE 9.2
Thyristor Construction
Cathode Gate Cathode
Silicon Base Plate
Anode
N
P
N
P
1
1
2
2
1306/C09/frame Page 164 Saturday, May 27, 2000 4:08 PM
New Methods of Increasing Transmission Capacity
165
The average halfwave rectiﬁed DC load current, I
r
, is
FIGURE 9.3
Schematic
FIGURE 9.4
Two transistor model of Thyristor
FIGURE 9.5
Thyristor Symbol
FIGURE 9.6
Current control by Thristor in a 5 kV line to ground distribution circuit.
N
1
P
1
N
2
P
2
Cathode
Gate
Anode
Gate
Anode
N2
N1
T1
P1
P2
T2
Cathode
Gate
Anode
Cathode
V
5kV/60Hz/0Deg
R
40
I
r
i
V t
R
t
r
=
( )
≤ ≤
sin ω
ω π 0
1306/C09/frame Page 165 Saturday, May 27, 2000 4:08 PM
166
Powerline Ampacity System: Theory, Modeling, and Applications
where
θ
is the angle at which a pulse is applied to the thyristor gate.
The variation of load current as a function of the ﬁring angle,
θ
, in the circuit
of Figure 9.6 is shown in Figure 9.7.
The main disadvantage of a thyristor is that it cannot be turned off easily by
applying a control signal at the gate. This limitation is overcome by the newer type
of semiconductor switching devices like the IGBT discussed later. Other than the
thyristor, there are several types of semiconductor devices that are suitable for various
power system applications. These are primarily electronic switching devices derived
from transistor technology and adapted for highcurrent and highvoltage applica
tions. Their important characteristics are listed in the Table 9.1 and are brieﬂy
discussed here.
MOSFET
The MOSFET is a voltagedriven source used mainly in lowvoltage applications.
It can be turned on or off rapidly and is suitable for high switchingfrequency
operations. Being a low voltage device, it is not suitable for direct connection to a
highvoltage network, and is therefore used in lowvoltage FACTS control circuits,
for example, to control a GTO device.
GTO
The Gate Turn Off Thyristor (GTO) is turned on by a short pulse of gate current,
and is turned off by applying a reverse gate signal. It has a short turnoff time in
the order of tens of nanoseconds, which is much faster than a thyristor. The main
disadvantage of a GTO is the high current requirement to turn off current. It has a
FIGURE 9.7
Control of circuit current as a function of thyristor ﬁring angle.
I
V Sin t
R
d t
I
V Cos
R
r
r
=
⋅ ( )
=
⋅ + ( ) ( )
∫
1
2
1
2
π
ω
ω
θ
π
θ
π
Thyristor Current Control
Firing angle
C
u
r
r
e
n
t
,
A
50
40
30
20
10
0
0
2
0
4
0
6
0
8
0
1
0
0
1
2
0
1
4
0
1
6
0
1
8
0
1306/C09/frame Page 166 Saturday, May 27, 2000 4:08 PM
New Methods of Increasing Transmission Capacity
167
low ratio of commutation current to turn off current, generally in the range of 3 to
5. This reduces the power that can be commuted by a GTO in a FACTS application.
IGBT
The IGBT takes advantage of the high commutation speed of a power MOSFET
and the low resistance offered by a bipolar transistor. The term “insulated gate”
refers to the metal oxide insulated gate which requires very little control power. This
feature is particularly useful in highvoltage FACTS applications because the control
signal is transmitted from ground potential to the gate of the IGBT device at very
high voltage. The switch can be turned on or off quite rapidly by the application of
a control signal at the gate. It is therefore suitable for highvoltage applications,
where several of these devices can be connected in series.
GBTR
The Giant Bipolar Transistor (GBTR) is older than the IGBT, but is available with
sufﬁcient power capability that is comparable to a thyristor. Even though the cost
of the GBTR is somewhat less than IGBT and power MOSFET, it is not widely
used due to the complex electronics required to control it.
MCT
The MOSControlled Thyristor is considered the power semiconductor switching
device of the future. The MCT device under development is expected to offer high
commutation power similar to a GTO, with control capability similar to an IGBT.
FACTS Semiconductor Valve Assembly
Since the present state of the art in semiconductor assembly allows a thyristor or an
IGBT device to be built with voltage rating up to 10 kV, several of these devices
are required to be connected in series to withstand high commutation voltage for
highvoltage applications. Therefore a thyristor valve assembly is made modular in
structure for ease of installation and maintenance. At the individual thyristor level
it is comprised of a control unit for controlling the ﬁring angle of the thyristor, a
unit for cooling the thyristor, and an electrical ﬁlter for the elimination of noise
generated by switching action. A module may consist of four to six thyristors
connected in series with a reactor. Several of these modules are then connected in
series to develop the full transmissionlinetoground voltage.
A schematic of a thyristor valve assembly is shown in Figure 9.8. A thyristor
valve assembly is comprised of a system for the communication and control of the
various thyristor units, and a system for the distribution of cooling ﬂuid to the various
units and support insulators. Highvoltage DC convertors and SVC stations are
generally located indoors, where a stack of thyristor modules forming a highvoltage
valve is either ﬂoor standing or suspended from a ceiling.
A number of such devices have been developed which are widely known as FACTS
(Flexible AC Transmission System) devices. By connecting these active devices at
1306/C09/frame Page 167 Saturday, May 27, 2000 4:08 PM
168
Powerline Ampacity System: Theory, Modeling, and Applications
suitable locations in the T&D network, it is now possible to implement a wide range
of control over an AC network in a manner that was not possible before.
By having better control over the ﬂow of electricity in a T&D network by FACTS
devices, it is expected to lower electricity costs by efﬁcient utilization of existing
assets, better outage management, greater operational ﬂexibility, and faster recovery
from system disturbances. These devices will also enable greater utilization of
economic energy sources (including distributed energy resources) and facilitate
competition in the electric supply business.
9.3 FLEXIBLE AC TRANSMISSION
HVDC
The development of FACTS technology has evolved from the early days of High
Voltage DC Transmission (HVDC) where converter stations are used to rectify AC
to DC, and an inverter station is used to convert DC back to AC. A converter station
is generally composed of 12 or 24 pulse thyristorcontrolled electronic bridge cir
cuits. Unlike an AC transmission circuit, a wide variety of control is possible in an
HVDC transmission system by controlling the ﬁring angle of thyristors.
FIGURE 9.8
Construction of a thyristor valve module.
TABLE 9.1
Semiconductor Device Properties
Semiconductor
Device
Rated
Voltage
Rated
Current Speed
Switching
Frequency
MOSFET 200 V 100 A 200 ns 100 kHz
IGBT 1200 V 300 A 1
µ
s 10 kHz
GBTR 1200 V 300 A 5
µ
s 3 kHz
GTO 4000 V 3000 A 40
µ
s 1 kHz
Thyristor 5000 V 4000 A – 300 Hz
MCT 3000 V 30 A 40
µ
s 1 kHz
Adapted from “Technologie des FACTS,” Ph. Lataire, SEE Confer
ence 1994.
1
2
4
1
2
4
1
2
4
1
2
4
3
4
1306/C09/frame Page 168 Saturday, May 27, 2000 4:08 PM
New Methods of Increasing Transmission Capacity
169
There are several HVDC transmission systems operational in the USA. The
Paciﬁc Intertie is one example of an HVDC transmission system that is used to bring
lowcost hydroelectricity from the Paciﬁc Northwest to California by means of a
bipolar ± 500kV HVDC transmission line. HVDC offers several advantages over
AC transmission, including lower transmission cost, asynchronous operation of two
electrical systems, and support to AC systems when required.
As consumers require better power quality, greater utilization of FACTS devices
are also expected in the distribution systems for better voltage regulation, lower
harmonics, minimum voltage ﬂicker, and greater reliability of service. The use of
FACTS devices in distribution systems will also lead to greater integration of smaller
generation systems with lower environmental impact. HVDC LIGHT is another new
development for DC power distribution by underground cables to remote locations
where there is no local generation available, or as a backup resource for local
generation.
S
TATIC
VAR C
OMPENSATOR
(SVC)
Static VAR compensators are reactive power devices that generate or absorb reactive
power as required by the electric power system. Passive shunt capacitors and reactors
have been used for a long time for reactive power supply, power factor correction,
and voltage support, but were limited by their ability to provide continuously variable
output to match system requirements. SVC systems are active devices whose output
can be controlled continuously to match system requirements very precisely. These
devices are used to maintain constant voltage levels, enhance power ﬂow, improve
stability, and provide various other improvements listed at the end of this section.
SVC is widely used in the electric power system to increase the transmission
capacity of existing lines. The performance of the SVC device to augment power
ﬂow in a transmission line is seen in Figure 9.9. In an uncompensated transmission
line, power ﬂow is limited by voltage drop on the line, such that any further increase
in power transfer results in voltage collapse. If an SVC is connected to the trans
mission line, the power transmission capacity of the line may be increased up to the
full thermal rating of the transmission line conductor as shown in Figure 9.9. This
example is given for a 132 kV transmission line having an ACSR Cardinal conductor.
Without an SVC device connected on this line, the transmission capacity is limited
to 750 MW. When an SVC device is connected, it is possible to utilize the full
thermal capacity of the line, up to 1000 MW or more.
The most common types of SVC systems used at present consist of a thyristor
controlled reactor (TCR) and a thyristorswitched capacitor (TSC). The basic ele
ments of a TCR are shown in Figure 9.10, and a TSC is shown in Figure 9.11.
A TCR consists of a reactor in series with a bidirectional thyristor pair. A
thyristor is a fastacting electronic switch, basically a fourlayer pnpn device,
consisting of anode, cathode, and gate. The thyristor is turned on by applying a short
pulse across the gate and cathode, and turned off by applying a reverse voltage
across the anode and cathode. The reactive power absorbed by the TCR device is
controlled by regulating the current ﬂow through the reactor by directing the ﬁring
angle of the thyristor between 90° and 180°. There is full conduction at 90° ﬁring
1306/C09/frame Page 169 Saturday, May 27, 2000 4:08 PM
170
Powerline Ampacity System: Theory, Modeling, and Applications
angle. Increasing the ﬁring angle decreases current, and conduction is blocked when
the angle is 180°. A continuously variable lagging reactor power, Q
L
, is made possible
by controlling the ﬁring angle of the thyristor given by,
Q
ᐉ
= I
ᐉ
(
α
)
2
X
ᐉ
where,
Q
ᐉ
= reactive power (lagging)
I
ᐉ
= reactor current
α
= ﬁring angle of thyristor
X
ᐉ
= inductive reactance
FIGURE 9.9
Increasing power transmission line capacity by FACTS using a Static Var
Compensator (SVC).
FIGURE 9.10
Thyristor Controlled Reactor (TCR)
FIGURE 9.11
Thyristor Switched Capacitors (TSC)
Power Transmission Capacity
Voltage,kV
120.4
97.3
74.1
51.0
27.9
4.7
With SVCand
LINEAMPS
WithoutSVC
Power,MW
0 500 1000
V
I (a)
L
L
V
V
c
L
c
V
L
I
1306/C09/frame Page 170 Saturday, May 27, 2000 4:08 PM
New Methods of Increasing Transmission Capacity
171
Thyristors having voltage and current ratings up to 5 kV and 4000 A are presently
available for switching frequency up to 300 Hz. A basic model of a thyristorswitched
capacitor (TSC) system is shown in the Figure 9.11. A TSC is used to deliver leading
reactive power, Q
c
, by switching on thyristors and is given by,
Where,
Q
c
= leading reactive power delivered to the system
V
c
= capacitor voltage
X
c
= capacitive reactance
A combination of TCR and TCS is often connected in parallel to offer a continuously
variable SVC system providing leading or lagging reactive power, Q
r
, obtained as
follows: Q
r
= Q
l
– Q
c
. An example of an SVC system having a combination of TCR
and TCS in an actual 735 kV substation is shown in Figure 9.12, and the single
line diagram is shown in Figure 9.13.
There are more than 100 SVC installations worldwide operating at voltages up
to 500kV with capacity in the range of
±
50 MVA to
±
400 MVA.
STATCOM
The development of Gate TurnOff (GTO) Thyristor technology has recently led to
the making of an all solidstate VAR generator called STATCOM (Static Synchro
nous Compensator). Unlike thyristors, GTOs are capable of turning off by changing
their state from conduction to nonconduction by pulsing the gate. At present, GTO
FIGURE 9.12
A 735 kV SVC substation, Hydro Quebec Canada.
Q v X
c c c
=
2
1306/C09/frame Page 171 Saturday, May 27, 2000 4:08 PM
172
Powerline Ampacity System: Theory, Modeling, and Applications
voltage and current rating is 4 kV and 3000 A, respectively, 1 kHz commutation
frequency, and 40
µ
s speed. This makes the GTO more suitable for faster turning
off, requiring less power than a conventional thyristor. Unlike the TCS and TCR
devices described previously, STATCOM offers continuously variable reactive power
without the use of capacitors or reactors. As a result, the size and cost of static VAR
generating equipment is reduced considerably.
The basic operating feature of a STATCOM is similar to the rotating synchronous
condenser, but has the advantage of solidstate technology, having no moving parts
and with a high degree of reliability. The STATCOM consists of a DC to AC converter
circuit, which provides a threephase output voltage in phase with the AC system
voltage as shown in Figure 9.14.
FIGURE 9.13
Singleline diagram of the 735 kV SVC system with TCR and TSC
FIGURE 9.14
STATCOM, 6 pulse GTO converter and equivalent circuit
110
MVAR
TCR
110
MVAR
TSC
110
MVAR
TSC
110
MVAR
TSC
330 MVA 3 phase
Transformer
110
MVAR
TSC
110
MVAR
TSC
110
MVAR
TSC
110
MVAR
TCR
330 MVA 3 phase
Transformer
735 kV line to line voltage
Va
Ia
Ia + Ib
Vb Ib
GTO
Diode
V
Transformer
V
O
dc
L
V
1306/C09/frame Page 172 Saturday, May 27, 2000 4:08 PM
New Methods of Increasing Transmission Capacity
173
By controlling the magnitude of the converter output voltages, Vo, using GTO,
the reactive power delivered by STATCOM can be controlled from full leading to
full lagging. Increasing the magnitude of converter output voltages above AC system
voltage, V
L
, delivers leading VARS to the AC system. A reduction in converter output
voltage delivers lagging VARS.
S
ERIES
C
OMPENSATION
Series compensation has been used for a long time to increase the power transmission
capacity of long transmission lines by connecting capacitors in series at suitable
locations along the line. Early series compensation devices used mechanical switches
to connect capacitor banks in series with a transmission line to reduce line impedance
and increase transmission capacity.
Today, controllable series capacitors with GTO thyristors are used to control the
degree of reactive power compensation more precisely as a function of network
operating conditions. Thus, with a thyristorcontrolled series capacitors, (TCSC) it
is now possible to control the current ﬂowing through a speciﬁc highvoltage line,
as well to maximize the ampacity of the line. Some examples of possible series
capacitor locations in long transmission lines are shown in Figures 9.15 and 9.16,
and a thyristorcontrolled series capacitor with a protection device is shown in
Figure 9.17.
FIGURE 9.15
Series compensated line of Example 1.
FIGURE 9.16
Location of series capacitors on long transmission lines.
FIGURE 9.17
Controllable series capacitor
Vs
j250
Vr
1000 km
(A)
(B)
(C)
4
3
2
5
6
1
1 = Capacitor
2 = Zno varistor
3 = GTO thyristors
4 = Bypass breaker
5 = Damping reactor
6 = Transmission line
1306/C09/frame Page 173 Saturday, May 27, 2000 4:08 PM
174
Powerline Ampacity System: Theory, Modeling, and Applications
Static Synchronous Series Compensators (SSSC) are also developed that offer
the faster response time required for the damping of power system oscillations caused
by faults and other system disturbances. A GTO circuit is used for SSSC, which is
similar to a STATCOM, as shown in Figure 9.18. A complex voltage Vseries is
generated by connecting the GTO converter to a transformer connected in series
with the transmission line. This complex voltage V series is then added to the AC
voltage of the transmission line to control transmission line voltage.
Example 1
A 1000 km 765 kV transmission line delivers natural load. It is proposed to increase
the transmission capacity by two times the natural load. A series capacitor having
a reactance of j250 ohm is installed at the middle of the line, as shown in Figure
9.18. Determine the voltage regulation of the line.
Transmission line data
Receiving end voltage = 765 kV
Rdc = 0.041 ohm/km
L = 8.35 x 10
–4
H/km
C = 12.78 x 10
–9
F/km
G = 0
FIGURE 9.18
Static Synchronous Series Compensator (SSSC) and equivalent circuit
Vseries
Va
Va + Vseries
GTO
V
ds
Diode
Transformer
Vseries V
L
I
1306/C09/frame Page 174 Saturday, May 27, 2000 4:08 PM
New Methods of Increasing Transmission Capacity
175
Solution
For a 1000 km line, the following longline equations, derived in Appendix 10.1 at
the end of Chapter 10, will be used.
(9.1)
(9.2)
x = distance from receiving end, km
= Voltage at a point, x, in the line
= Voltage at the receiving end of the line
= Current at the receiving end of the line
= Current at a point, x, in the line
= propagation constant
Zc = characteristic impedance of the line
The above equation is derived in the appendix at the end of this chapter. The values
can be represented by the following matrix equation utilizing the wellknown A, B,
C, D constants of the line:
The natural load of the line is given by the Surge Impedance Loading (SIL), (See
Example 4, Section 10.5).
SIL = 2289 MVA
Therefore, the load current, Ir, is,
The propagation constant and the characteristic impedance are obtained from (Exam
ple 4, Section 10.5).
V x V Zc x I
x r r
= ( ) + ⋅ ( ) cosh sinh γ γ
I
x V
Zc
x I
x
r
r
=
( )
+ ( )
sinh
cosh
γ
γ
V
x
V
r
I
r
I
x
γ
Vs
Is
A B
C D
Vr
Ir
=
Ir
SIL
Ir A
=
⋅ ⋅
=
3 765 10
1728
3
1306/C09/frame Page 175 Saturday, May 27, 2000 4:08 PM
176
Powerline Ampacity System: Theory, Modeling, and Applications
The sending end voltage Vs is obtained from Section 10.5,
Sending end voltage when load is 2 times SIL,
The sending ending voltage is too high and the regulation is unacceptable.
A series capacitor bank having a total reactance of j250 ohm is now added at
the middle of the line. The ABCD constants of a 500 km section of line are:
γ = ⋅ + ⋅
= −
2 10 1 2 10
255 6 4 15
5 3 – –
.
. .
j
Zo j
V j
j j
V j
V
V V
V
s
s
s
s r
r
= ⋅ + ⋅
( )
⋅ ⋅ ⋅
{ }
+ ( ) ⋅ ⋅ + ⋅
( )
⋅
{ }
⋅
= + ( ) ⋅
= ⋅
=
( )
⋅
=
cosh .
. – . sinh .
. .
.
–
. %
– –
– –
2 10 1 23 10 10 441 10
255 6 4 16 2 10 1 23 10 10 1728
156 7 424 9 10
452 8 10
100
2 52
5 5 3 3
5 3 3
3
3
Regulation
V j
V
s
s
= + ( ) ⋅
= ⋅
= ⋅ ⋅
( )
⋅
⋅
=
166 4 841 4 10
857 7 10
857 7 10 441 7 10
100
441 7 10
94 1
3
3
3 3
3
. .
.
. – .
.
. %
Regulation
A
A D j
j
= ⋅ ( )
= = ⋅ + ⋅
( )
⋅
= + ⋅
cosh
cosh .
. .
– –
–
γ l
2 10 1 23 10 500
0 816 5 79 10
5 3
3
1306/C09/frame Page 176 Saturday, May 27, 2000 4:08 PM
New Methods of Increasing Transmission Capacity 177
The ABCD constants of the series capacitor are:
The modiﬁed ABCD constants of the series compensated transmission line are:
B Zo
B j j
B j
= ⋅ ⋅ ( )
= ( ) ⋅ ⋅ + ⋅
( ) { }
⋅
= +
sinh
. – . sinh .
. .
– –
γ l
255 6 4 15 2 10 1 23 10 500
4 49 147 6
5 3
C
Zo
C
j
j
C j
=
⋅ ( )
−
⋅ + ⋅
( ) { }
⋅
( )
⋅
= − ⋅ + ⋅
sinh
sinh .
. – .
. .
– –
– –
γ l
2 10 1 23 10 500
255 6 4 15
4 77 10 2 26 10
5 3
6 3
′ =
′ =
′ =
′ =
A
B j
C
C
1
250
0
1
′′ ′′
′′ ′′
=
′ ′
′ ′
=
+ ( ) + ( )
⋅ ⋅
( )
+ ( )
A B
C D
A B
C D
A B
C D
A B
C D
j j
j j
j 0 816 5 79 4 49 147 6
4 77 10 2 26 10 0 816 5 79
1 250
0
6 3
, . . .
– . – . . .
–
– –
11
0 816 5 79 4 49 147 6
4 77 10 2 26 10 0 816 5 79
0 794 0 023 7 993 74 515
3 935 10 4 966 10 0 794 0 023
6 3
5 3
+ ( ) + ( )
⋅ ⋅
( )
+ ( )
=
+ ( ) + ( )
⋅ ⋅
( )
+ ( )
. . . .
– . – . . .
. . . .
– . – . . .
– –
–
j j
j j
j j
j j
1306/C09/frame Page 177 Saturday, May 27, 2000 4:08 PM
178 Powerline Ampacity System: Theory, Modeling, and Applications
The above sending end voltage and regulation is acceptable.
The above example shows that by adding a series capacitor at the middle of a
line, the maximum power delivered by the line is increased from 2289 MVA natural
load to 4578 MVA, which is twice the natural load of the line. Having a GTO
thyristorcontrolled series capacitor as shown in Figure 9.17 allows control of the
current passing through the capacitor and, hence, the reactance, by modifying the
A,B,C,D constants of the line according to network operating conditions.
UNIFIED POWER FLOW CONTROLLER (UPFC)
The UPFC is a new development that offers multiple compensation functions by
providing independent control of the following transmission line parameters:
• Bus voltage
• Active power
• Reactive power
The UPFC is made of two GTObased converters connected by a common DC link.
They can also operate independently with Converter 1 acting as a STATCOM and
Converter 2 like an SSSC. A block diagram of the UPFC converter, control domain,
and equivalent circuit is shown in Figure 9.19. Recent UPFC installations worldwide
are in the range ± 100 to ± 200 MVA at transmission voltage levels. The AEP*
installation (1997–1998) is rated at ± 160 MVA and is located on a 138 kV trans
mission line.
CUSTOM POWER
Custom Power concept covers a number of power electronics devices suitable for
connection at the distribution system level. These devices typically have a range
from 1 to 10 MVA and may be connected at customer point of connection to provide
better voltage regulation, and prevent plant shutdown during adverse voltage con
* World’s ﬁrst Uniﬁed Power Flow Controller on the AEP system, B.A. Rena, et al. Cigré 1998.
V A V B I
V j j
V j
V
s r r
s
s
s
= ′′ ⋅ + ′′ ⋅
= + ( ) ⋅ ⋅ + + ( ) ⋅
= ⋅ + ⋅
= ⋅
. . . . .
. .
.
794 0 023 441 7 10 7 993 74 515 3455
3 782 10 2 677 10
463 4 10
3
5 5
3
Regulation = 463 4 10 441 7 10
100
441 7 10
4 91
3 3
3
. – .
.
. %
⋅ ⋅
( )
⋅
⋅
=
1306/C09/frame Page 178 Saturday, May 27, 2000 4:08 PM
New Methods of Increasing Transmission Capacity 179
ditions such as voltage dips and power surges. A distribution system STATCOM is
used at the distribution level for power factor correction as more current is passed
through the distribution circuit, and for protecting the distribution system from the
effects of nonlinear loads and ﬂicker reduction applications.
Power electronics devices also enable greater integration of smaller generating
systems located closer to the loads. Small generation technologies include micro
turbines, fuel cells, wind turbines, and photovoltaic and other renewable energy
sources. In this scenario, the installation of power electronic devices offers faster
connection to the main supply in the event of a failure, damping of oscillations, and
smoothing out the voltage delivered by the small generation systems.
SMES
The Superconducting Magnetic Energy Storage (SMES) device is another FACTS
device (Feak, 1997), (Borgard, L., 1999), (Buckles et al. 2000) that is used for
voltage support, damping of power system oscillations, and improvement of power
system stability.
FIGURE 9.19a Uniﬁed Power Flow Controller (UPFC), Converter schematic
FIGURE 9.19b Uniﬁed Power Flow Controller (UPFC), Control domain
FIGURE 9.19c Uniﬁed Power Flow Controller (UPFC), Equivalent circuit1
Qin
Pin
Va
Qout
Pout
Vb
Vb
Vseries
Qout
Pout
Qin
Pin
Va
Control
Converter 2 Converter 1
Vdc
Re
Vseries
Vb
Va
Im
Ia Vseries Ib
Vb Ic
Va
+
+
1306/C09/frame Page 179 Saturday, May 27, 2000 4:08 PM
180 Powerline Ampacity System: Theory, Modeling, and Applications
During normal operating conditions, energy is stored in a superconducting coil,
6, comprise the SMES device. The stored energy released to the network when there
is a disturbance is used to provide voltage support, as well as for damping power
system oscillations. A simpliﬁed SMES device is shown in Figure 9.20. The two
GTO thyristor converters, 4 and 5, are required for conversion of DC power supplied
by the superconducting coil to 60 Hz ac. The AC output voltage of the converters
is stepped up to the transmission line voltage by a threewinding transformer, 3.
LIST OF FACTS APPLICATIONS
A number of FACTS device systems are in use worldwide to solve various power
system problems:
• Voltage control
• Load balancing
• Increase active power transmission capacity of existing and future lines
• Increase transient stability margin
• Increase damping of power oscillations caused by a disturbance in the
power system
• Facilitate greater use of dispersed, small, and distributed energy sources
• Reduce temporary overvoltages
• Damp subsynchronous resonance
• Provide reactive power to ACDC converters
• FACTS devices for distribution system and custom power
MANUFACTURERS
There are several manufacturers of FACTS devices in the US and worldwide. Fol
lowing is a list of important manufacturers.
ABB
GE Power Electronics Division
GEC Alsthom T&D Power Electronics System
Hitachi
SIEMENS
Westinghouse Electric Company
FIGURE 9.20 SMES device connected to a high voltage transmission line
1
2
3
4
5
6
1 = Transmission line
2 = Harmonic filters
3 = Three winding transformer
4 = GTO thyristor converter 1
5 = GTO thyristor converter 2
6 = SMES
1306/C09/frame Page 180 Saturday, May 27, 2000 4:08 PM
New Methods of Increasing Transmission Capacity 181
FUTURE RESEARCH AND DEVELOPMENT
The FACTS devices used at present are mostly independent units and are not yet
widely integrated with other FACTS devices in the network. Further R&D is required
to develop integrated control of several FACTS devices by communication between
devices in order to exercise much wider and distributed control of the power system.
For HV and EHV network applications, there is a need to increase the current
and voltage rating of power semiconductor devices as well as to increase the switch
ing frequency for higher commutation speed and fewer losses.
This chapter mainly focused on FACTS technology for transmission systems,
with a brief discussion on distribution system applications. As consumers require
better power quality, greater utilization of FACTS devices is expected for better
voltage regulation, lower harmonics, minimum voltage ﬂicker, and greater reliability
of service. The use of FACTS devices in distribution systems will also lead to greater
integration of smaller generation systems with lower environmental impact.
Another interesting application of FACTS for distribution systems is HVDC
LIGHT. This technology is providing competition for distributed generation systems.
HVDC LIGHT applies FACTS technology for low voltage DC transmission by
underground cables. Lowvoltage DC converters are used to tap off connections
from HV and EHV AC lines and convert AC to lowvoltage DC. The lowvoltage
DC power supply is then distributed by DC cables to remote areas. Since DC voltage
is used for distribution it does not have the problem of the high capacitive reactance
of AC underground cables. Therefore, it appears that lowvoltage underground DC
transmission can supply power for longer distances to remote areas. Underground
DC is also more acceptable for environmental reasons.
9.4 CHAPTER SUMMARY
Power electronics semiconductor device technology for FACTS has evolved consid
erably during the last decade, and several applications of the different technologies
are presented in this chapter. For power semiconductor devices, the main require
ments are fast turnon and turnoff times with high current and voltage ratings. At
the present time these devices are developed with voltage ratings up to 10 kV, and
maximum current ratings in the range of 1000 A to 3000 A. Superconducting
Magnetic Energy Storage (SMES) systems are developed for application in all
voltage levels for faster response time. SMES provides continuous VAR support and
maintains the stability of the interconnected transmission grid.
The development and application of FACTS devices are largely dependent upon
semiconductor switching devices. The IGBT device appears to be promising for
highvoltage applications requiring fast turnon and turnoff times and having high
switching frequency. GTO thyristors are also available with sufﬁcient voltage and
current rating and fast switching capability. Further development of semiconductor
power electronic devices is required to obtain greater current and voltage ratings
with faster response time and higher switching frequency.
A wide variety of FACTS devices are presented in this chapter, which includes
both static series and shuntcompensated devices, and HVDC LIGHT for lowvoltage
1306/C09/frame Page 181 Saturday, May 27, 2000 4:08 PM
182 Powerline Ampacity System: Theory, Modeling, and Applications
distribution to remote locations. Thyristor controlled series capacitors are particu
larly useful for increasing the power transmission capacity of long lines without
causing synchronous resonance problems associated with generators. The develop
ment and application of FACTS technology is essential for the control of current
and voltage in transmission and distribution circuits. This will lead to increased
reliability of service, the lowering of electricity cost, and better power quality. When
lines are operated close to thermal ratings, these new FACTS devices will respond
quickly in the event of a system disturbance by providing voltage support, and will
maintain generator stability. It is expected that these developments will enable
maximum utilization of existing transmission and distribution system resources by
making use of the line ampacity system.
1306/C09/frame Page 182 Saturday, May 27, 2000 4:08 PM
183
10
Applications
10.1 INTRODUCTION
The application of the powerline ampacity system to the economic operation of
transmission network, power system stability, and transmission planning is presented
in this chapter. The economic load dispatch problem and the optimal power ﬂow
problem are presented by consideration of variable transmission line ratings. Results
are presented by example of a hypothetical utility generation and transmission
network, which clearly demonstrates the economic beneﬁt of operating a power
system by utilizing a dynamic powerline ampacity system.
For transmission planning purposes, it is necessary to clearly understand the
factors affecting the choice of conductors for overhead transmission line applications.
For this purpose, a formulation of the economic sizing of conductors is presented,
followed by a study of economic conductor temperatures. Based on the above
considerations, the cost of adding a new line in the region of the San Francisco Bay
area, for example, is compared to the increased cost of losses due to higher trans
mission line current. It is shown that the construction of a new line can be deferred
by at least ten years by adopting a dynamic line rating system in this region.
For future transmission system planning, due consideration must be given to
utilizing renewable energy sources that are generally located far away from major
metropolitan areas and industrial load centers. Extra High Voltage (EHV) transmis
sion lines are required to transport electric energy economically over long distances.
For this reason, some important factors affecting longdistance transmission, mainly
from the point of view of transmission line ampacity, are discussed.
The maximum power transmission capacity of certain overhead powerlines is
sometimes limited by the stability of generators connected to transmission lines.
Therefore, the effect of increasing line ampacity on generator stability is presented.
The steady, dynamic, and transient stabilities of generators are then related to the
steady, dynamic, and transient line ratings.
10.2 ECONOMIC OPERATION
In an earlier study, the author performed economic evaluation of an interconnected
electricity generation and transmission system by a dynamic line rating system (Hall,
Deb, 1988a). In this study, actual PG&E transmission and generation system costs
were presented to show cost savings by dynamic line ratings. We presented a case
study without performing a formal network optimization analysis. Therefore, the
present work on generation and transmission system cost optimization by dynamic
line rating is an extension of previous work on this subject.
1306/C10/frame Page 183 Saturday, May 27, 2000 4:17 PM
184
Powerline Ampacity System: Theory, Modeling, and Applications
10.2.1 F
ORMULATION
OF
THE
O
PTIMIZATION
P
ROBLEM
It is required to minimize the total cost of electricity production, including generation
and transmission cost for a given load demand and operating system constraints. In
an interconnected electric utility system comprising dispersed generation sources,
the problem is how to allocate the required load demand among the available
generation units on an hourbyhour basis. This is carried out by minimizing an
objective function, F, that represents the total cost of producing electricity.
The objective function is given as,
(10.1)
where,
a
i
, b
i
, c
i
, = cost coefﬁcients of the ith generator
n = number of generator units
P
i
= power output of the ith generator
The objective function (10.1) is subject to the following constraints:
Power Balance
This requires that the sum of all power generated be equal to the sum of demand
and transmission loss.
(10.2)
P
D
= demand
P
L
= transmission loss given by,
(10.3)
B
i
is the transmission loss coefﬁcient.
Nodal Balance of Active and Reactive Powers
(LoadFlow Equations)
Active Power
(10.4)
F a b P c P
i i i i i
i
n
+ ⋅ + ⋅
( )
∑
2
1
P P P
i
D L
i l
n
– –
∑
0
P B P
L
i
i
n
i
∑
1
2
P V V g Cos b Sin i n nodes
i i
k
n
k i k i k i k i k
( )
+
( )
[ ]
∑
1
1
, ,
– – , θ θ θ θ K
1306/C10/frame Page 184 Saturday, May 27, 2000 4:17 PM
Applications
185
Reactive Power
(10.5)
Transmission Line Capacity Limits (Line Ratings)
Line current between nodes i and k:
(10.6)
Generator Capacity Limits
The active (P) and reactive (Q) power output of the ith generator should be between
minimum and maximum generation limits:
P
min,i
≤
P
i
< P
max,i
i = 1,….Ng (10.7)
Q
min,i
< Q
i
≤
Q
max,i
i = 1,….Ng (10.8)
Ng = Number of generator nodes
Bus Voltage Limits
V
min,i
≤
V
i
≤
V
max,i
i= 1,…N nodes (10.9)
Transformer Tap Limits
VT
min,i
≤
VT
i
≤
VT
max,i
i= 1,…Ntap (10.10)
The above equations describe the optimal powerﬂow problem. Additional con
straints may be added to consider emission levels in fossil fuelbased generation
units. A solution to the above nonlinear optimization problem is available in a modern
power system analysis textbook (Bergen, A., 1986). More recently, artiﬁcial neural
Q V V g Cos b Sin i n nodes
V
V
i i
k
n
k i k i k i k i k
i
( )
+
( )
[ ]
≅ ∠
∑
1
1
, ,
– – , θ θ θ θ
θ
K
node voltage
I
V V
z
I Max
z
g jb
I Max
i k
i i k k
i k
i k
i k
i k i k
i k
,
,
,
,
, ,
,
–
∠ ∠
≤ ( )
+
( )
θ θ
1
Dynamic line rating
1306/C10/frame Page 185 Saturday, May 27, 2000 4:17 PM
186
Powerline Ampacity System: Theory, Modeling, and Applications
network (Lee, K.Y. et al., 1998), (Yalcinov, Short, 1998) and genetic algorithm GA
(Wong, Yuryevich, 1998) solutions have also been proposed. A simpliﬁed solution
of the above problem is presented in Table 2.3 and Figures 10.3 and 10.4 for a static
and dynamic line rating system. The object of this study is to show the signiﬁcance
of dynamic transmission line ratings in the economic operation of an interconnected
power system having diverse generation sources.
10.2.2 E
LECTRICITY
G
ENERATION
C
OST
S
AVING
IN
I
NTERCONNECTED
T
RANSMISSION
N
ETWORK
To illustrate the main concepts presented in the previous section, a hypothetical electric
utility generation and transmission system was created as shown in Figures 10.3 and
10.4. The transmission network receives power from ﬁve generation sources (G1G5).
The generation sources G1 and G3 are hydroelectric, and all other generation sources
are thermal. The transmission network is comprised of ﬁve 230 kV double circuit lines
and the network data is shown in the Table 10.1. An ACSR Cardinal conductor is used
in each phase of the line. The dimensions of the 230kV double circuit line are given
in Figure 10.1. The transmission line electrical
π
equivalent model shown in Figure
10.2 was used to obtain network loadﬂow solutions.
A comparison of electricity generation costs is made in the following sections using
static and dynamic line rating systems. It is shown that substantial economy in
electricity generation cost can be achieved by adopting a dynamic line rating system.
Electricity Generation Cost Using Static Line Rating System
Electric power companies generally follow a static line rating system by assuming
constant line capacity based on conservative assumptions of the meteorological
parameters that affect line capacity. Many electric power companies have seasonal
line ratings for summer and winter. There are several reasons for operating lines
based on static line ratings. For example, transmission lines are operated more easily
TABLE 10.1
230 kV Transmission network data with ACSR Cardinal
Conductor
From Bus To Bus Line #
Impedance
Z
1
= R
1
+ jX
1
HalfLine Charging
Capacitance
1 2 1 0.02 + j0.13 j0.03
1 5 2 0.05 + j0.32 j0.025
2 3 3 0.016 + j0.10 j0.02
3 4 4 0.02 + j0.13 j0.02
4 5 5 0.016 + j0.10 j0.015
Ys
2
j C
2
ωω
1306/C10/frame Page 186 Saturday, May 27, 2000 4:17 PM
Applications
187
by following a static line rating system. The transmission line protection scheme
does not have to be changed because line current limits are held constant. Also,
close monitoring of weather conditions all along the transmission line route, as well
as monitoring transmission line conductor temperatures, conductor tension, and other
parameters are not required.
In the example of Figure 10.3, the static line rating of each transmission line
phase conductor is assumed to be 849 A for normal operating conditions. An ACSR
FIGURE 10.1
230 kV Double Circuit Transmission Line Tower.
FIGURE 10.2
Transmission Line Electrical PI model.
FIGURE 10.3
Electricity generation and transmission line current using static line rating.
4.5 m
10.5 m
5.9 m
7.6 m
7.6 m
30.5 m
R jXl
Ys/2 Ys/2
1022MW 400MW 500MW
500MW
600MW
1697A 1109A
1244A
829A
320A
1000MW
800MW
600MW 400MW
G1 G2
G3
G4 G5
1306/C10/frame Page 187 Saturday, May 27, 2000 4:17 PM
188
Powerline Ampacity System: Theory, Modeling, and Applications
Cardinal conductor is used in all lines. Based on the above assumptions, the optimum
powerﬂow solution of the network is shown in Figure 10.3 and Table 10.2. In Figure
10.3 it is seen that the ampacity of Line 1 has reached its maximum static rating (2
×
848 = 1996 A), and the total cost of electricity generation for meeting the demand
for one hour is $ 48,877.32.
Electricity Generation Cost Using LINEAMPS Rating
Due to environmental and economic considerations, many electric power companies
have now started to use a system of rating transmission lines that is variable, depending
upon actual transmission line operating conditions (Soto et al., 1998), (Wook et al.,
1997), (Steeley et al., 1991). As stated in the previous chapters, the variable system
of line ratings is commonly known as “dynamic line rating system,” where transmission
line ampacity is adapted to actual and forecast weather conditions. In this section the
economic beneﬁts of a dynamic line rating system are demonstrated by an example
of the transmission network considered in the previous section.
In the optimum powerﬂow solution of Figure 10.3, the capacity of Line 1
reached its maximum static ampacity, which limited further addition of cheaper
hydroelectricity available from the generator (G1). It is shown that by utilizing a
dynamic line rating system the network is able to utilize a greater percentage of
hydroelectric generation from the generation source (G1).
The optimum powerﬂow solution achieved by dynamic line rating is presented
in Figure 10.4. As seen in this ﬁgure the network was able to receive 1549 MW of
cheaper hydroelectric energy from generator source (G1) due to the higher trans
mission line ampacity (2 x 1063.5 = 2127 A) offered by LINEAMPS. A similar
result was obtained in a previous study (Hall, Deb, 1988a).*
Cost Savings by LINEAMPS
The power ﬂow solution presented in Table 10.3 utilizes dynamic line rating, which
has resulted in an electricity cost saving of $ 4020.87 for one hour. Such favorable
conditions exist often during the life of a transmission line when it is required to
FIGURE 10.4
Electricity generation and line current using dynamic line rating.
* Assuming typical electricity generation cost at PG&E.
1549 MW 400 MW
600 MW 500 MW
400 MW
G2
G4 G5
G3 G1
2127 A 1544 A
1136 A
800 MW 1000 MW
200 MW 400 MW
1731 A
378 A
1306/C10/frame Page 188 Saturday, May 27, 2000 4:17 PM
Applications
189
transfer power greater than static line rating. For a more realistic example, the
network of Figure 10.3 may be considered a simpliﬁed transmission network of the
San Francisco region. The generation source (G1) may represent the cheaper source
of hydroelectric energy from the Paciﬁc Northwest supplying power to San Francisco
Bay area. Similarly, the generation source (G3) could well represent hydroelectric
energy supplied from the Sierras. Sometimes there is surplus electric energy available
from these sources due to greater than normal rainfall or snow in these areas. Using
LINEAMPS ratings, these cheaper sources of electricity can be utilized in the Bay
Area with substantial cost savings as shown in Figure 10.4 and Table 10.3.
In addition to operational cost savings by facilitating economy energy transfer
as mentioned above, it is also possible to save the capital investment required for
the construction of new lines or the reconductoring of existing lines. By following
the economic analysis presented in Section 10.4, it is shown in Figure 10.5 that
capital investment for new line construction can be deferred for at least ten years
by increasing the capacity of existing lines for different overload conditions.
TABLE 10.2
Electricity Production Cost: Static Line Rating
Bus
Number
Generation
MW
Load
MW
Generation
Cost $/MWh
Generation
Cost $/hr
1 1022.07 0 10.30 10527.32
2 400 600 24.20 9680.00
3 500 500 12.50 6250.00
4 400 800 25.30 10120.00
5 600 1000 20.50 12300.00
Total Generation
Cost, $/hr
48877.32
TABLE 10.3
Electricity Production Cost: Dynamic Line Rating
Bus
Number
Generation
MW
Load
MW
Generation
Cost $/MWh
Generation
Cost $/hr
1 1549.17 0 10.30 15956.45
2 400 600 24.20 9680.00
3 400 500 12.50 5000.00
4 400 800 25.30 10120.00
5 200 1000 20.50 4100.00
Total Generation
Cost, $/hr
44856.45
1306/C10/frame Page 189 Saturday, May 27, 2000 4:17 PM
190
Powerline Ampacity System: Theory, Modeling, and Applications
10.3 S
TABILITY
The maximum power transmission capacity of certain overhead powerlines are
limited by voltage drop and the stability of generators connected to transmission
lines. In Chapter 9 the various methods of increasing transmission capacity by the
installation of modern power electronics devices were presented. In this section the
method of calculation of generation stability when supplying power over a trans
mission line is presented brieﬂy. The following types of stability (Nasar, Trutt, 1998)
problems are discussed:
• Steadystate stability
• Dynamic stability
• Transient stability
We are concerned with the steadystate stability of a generator when the power
transferred over a line is increased slowly. The steadystate stability limit of a single
generator supplying a load through a transmission line is given by,
(10.11)
where,
Ps = steady state power transfer
Vs = sending end voltage
Vr = receiving end voltage
FIGURE 10.5
Cost of adding a new line is compared to the capitalized cost of increased
transmission line losses for a range of overload conditions. The analysis is carried out for a
period of ten years. For example, if we consider 1000 A overload for 2 hr/Day we can see
from this graph that it is more economical to allow the overload current than to build a new
line having cost greater than $75,000 /km. The above example is for a 230 kV line with ACSR
Cardinal conductor having a static line rating of 849 A.
0 1 2 3 4 5 6 7 8 9 10 11 12
Overload hr/Day
350
300
250
200
150
100
50
0
Cost Saving by LINEAMPS
Deferment of Investment for New Line
C
o
s
t
o
f
N
e
w
L
i
n
e
,
$
x
1
0
0
0
1000A
1200A
1400A
Ps
Vs Vr Sin
X
( ) δ
1306/C10/frame Page 190 Saturday, May 27, 2000 4:17 PM
Applications
191
X = line reactance
δ
= generator rotor angle
Maximum steadystate power, Pmax, is obtained when the rotor angle
δ
= 90°.
Increasing
δ
> 90° will result in lower power transfer and, ultimately, loss of
steadystate stability as Ps approaches zero. From Equation 10.11, one possible
means of increasing steadystate stability is to add capacitors in series with the
line to lower the reactance, X, of the line. This was discussed in Chapter 9 in
Series Compensation.
Example 10.1
A generator is supplying a load through a 50 km 230 kV double circuit line. Calculate
the maximum power transfer capability of the line. The transmission line data is
given below.
Conductor = ACSR Cardinal
Reactance of line = 0.5 ohm/km
Transmission line sending and receiving end voltage magnitude = 230kV
Solution
Selecting generator base MVA = 1000 MVA, 3 phase
Selecting transmission line base kV = 230 kV
Transmission line base impedance = = 52.9 ohm
Line impedance = = 0.47 pu
Pmax sin 90
Pmax = 2.12 pu
Pmax = 2.12x1000 = 2120 MVA
10.3.1 D
YNAMIC
S
TABILITY
Dynamic stability is concerned with generator oscillations due to step changes in
load or other small disturbances. Small changes in generator output due to load
variations result in generator rotor oscillations. If oscillations increase, in time the
system becomes unstable. The system is dynamically stable if the oscillations dimin
ish with time, and the generators return to a stable state. Generally, the dynamic
condition oscillations remain for several seconds until steadystate conditions are
reached.
The differential equation governing rotor motion,
∆δ
, with respect to time, t,
due to a small increase in power,
∆
P, is obtained by (Saadat et al., 1998):
230
1000
2
50 0 5
52 9
⋅ .
.
1
0 47 .
1306/C10/frame Page 191 Saturday, May 27, 2000 4:17 PM
192
Powerline Ampacity System: Theory, Modeling, and Applications
(10.12)
(10.13)
where,
∆δ
= small deviation in power angle from initial operating point
δ
0
D = damping constant
H = Inertia constant
f
0
= frequency
ω
n
= natural frequency of oscillation
ζ
= damping ratio given by,
(10.14)
The above differential equation is obtained by linearization of the swing equation
(Bergen 1986) and is applicable for small disturbances only.
The solution of the above differential equation is ( Saadat et al 1998):
(10.15)
Example 10.2
The transmission line of Example 10.1 delivers a load of 1000 MVA under steady
state conditions. Show that the system will remain stable if the load is suddenly
increased to 1200 MVA. Assume generator frequency is 60 Hz during normal
operating conditions, and the inertia constant is H = 6 pu. The damping constant is
D = 0.138 pu.
Solution
Initial operating angle
Ps = P max · cos(δ
0
) = 2.04 pu
H
f
d
dt
D
d
dt
P P
o
s
π
δ δ
δ
2
2
∆ ∆
∆ ∆ + +
d
dt
d
dt
f P
H
n n
2
2
2 0
2
∆ ∆
∆
∆ δ
ςω
δ
ω δ
π
+ +
ς
π
⋅
⋅
D f
H P
s
2
0
∆
∆
δ
π
ω
ς
ω θ
ςω
⋅ ⋅
⋅
⋅ +
( )
]
]
]
]
f P
H
e t
n
t
d
n
0
2
2
1
1
1
–
–
sin
–
δ
0
1

(
`
J
sin
max
–
Pm
P
δ
0
1
0 6
2 128
16 38

(
`
J
° sin
.
.
.
–
1306/C10/frame Page 192 Saturday, May 27, 2000 4:17 PM
Applications 193
θ = cos
–1
(ζ)
The solution of the above equation for a period of 3 s is shown in Figure 10.6.
As seen in the ﬁgure, the system returns to a stable state after a sudden increase in
transmission line load from 1000 MVA to 1200 MVA.
10.3.2 TRANSIENT STABILITY
Transient stability is concerned with generator oscillations due to sudden changes
in power transfer levels caused by large disturbances which are due to shortcircuit,
large scale loadshedding, or generator or transmission line outage. Since we are
dealing with large disturbances, linearization of the swing equation is not possible.
Numerical solution of the nonlinear differential equation is obtained by Euler’s or
the Runge Kutte method. A simpliﬁed swing equation neglecting damping for
transient stability studies is given by,
FIGURE 10.6 Generator rotor angle oscillations due to sudden increase in transmission line
load current from 1000 MVA to 1200 MVA.
ω π
n
H Ps
⋅
60
ζ
π
⋅
⋅
⋅
D
H Ps
60
2
0 271 .
ω ω ζ
d n
⋅ 1 7 709
2
– .
δ π t e
t
t
( ) + ⋅ ⋅
⋅
⋅
⋅ + ( )
]
]
]
⋅ ⋅ ( )
0 286 60
0 2
6 8
1
7 707 1 29
1 0 271
2
0 2718
2
.
.
–
sin . .
– .
– .
38
36
34
32
30
28
26
24
22
20
0 1 2 3 0.5 1.5 2.5
Time, s
Dynamic Stability High Ampacity Line
G
e
n
e
r
a
t
o
r
A
n
g
l
e
,
D
e
g
1306/C10/frame Page 193 Saturday, May 27, 2000 4:17 PM
194 Powerline Ampacity System: Theory, Modeling, and Applications
(10.16)
Following Euler’s method, we obtain the change in angle, ∆δ
n
, during a small
interval, ∆t:
Example 10.3
The transmission line of Example 10.2 is delivering 1200 MVA through both circuits
when a shortcircuit occurs on one transmission line that lowers the power to 400
MVA. The fault is cleared in 0.125s. The power delivered by the line is 1200 MVA
after the fault. Determine if the system will remain stable and attain a steadystate
operating condition.
Solution
Initial operating angle,
The accelerating power is ∆P = 0 before the fault.
Acceleration power, ∆P, after the fault is:
∆P = P
0
– 0.4 sin δ
0
= 0.96
Starting with t = 0 and δ = 36° and time interval ∆t = 0.05 we ﬁnd,
d
dt
f P
H
2
2
0
0
δ π
–
⋅ ⋅
∆
∆
∆ ∆ ∆
∆ ∆
∆
∆ ∆
∆ ∆
δ δ δ
δ δ ω ω
π
δ δ
π
n n n
n n n n
n
n n
n
t
t t
f P
H
f P t
H
⋅
( )
⋅
⋅ ⋅
+
⋅ ⋅ ⋅ ( )
− − −
−
−
−
–
– –
–1
1 1 2 3 2
0 1
1
0 1
2
δ
0
1 1
1 2
2 04
36

(
'
`
J
J

(
`
J
°
sin
max
sin
.
.
– –
P
P
o
∆ ∆
∆ ∆
δ δ
n n
n
t P
H P
+
⋅ ⋅ ( ) ⋅
⋅
+
−
1
2
1
0
180 60
1306/C10/frame Page 194 Saturday, May 27, 2000 4:17 PM
Applications 195
During fault we obtain the change in rotor ∆δ angle during a small interval, ∆t, by,
and after the fault,
The above equation is solved for a period of 2 s. The swing curve is stable as shown
in Figure 10 7.
10.4 TRANSMISSION PLANNING
In the previous section, the application of the powerline ampacity system to power
system operations was presented. In this section, applications to transmission system
planning and design are discussed. This study includes transmission system cost
analysis, optimum sizing of transmission line conductors, and the evaluation of
optimum conductor temperatures. The following factors are considered:
• Capital cost of line
• Cost of capital (interest rate)
• Cost of energy, $/MWH
• Load factor
• Conductivity of conductor material
FIGURE 10.7 Swing curve for a transmission line fault cleared in 0.125 s.
∆ ∆
∆
δ δ
δ
n n
n
t P
H P
+
⋅ ⋅ ( ) ⋅
( )
[ ]
⋅
−
−
1
2
0 1
0
180 60 0 4 – . sin
∆ ∆
∆
δ δ
δ
n n
n
t P
H P
+
⋅ ⋅ ( ) ⋅
( )
[ ]
⋅
−
−
1
2
0 1
0
180 60 1 2 – . sin
100
90
80
70
60
50
40
30
20
10
0
0.00 0.50 1.00 1.50 2.00
G
e
n
e
r
a
t
o
r
A
n
g
l
e
,
D
e
g
Time, s
Transient Stability High Ampacity Line
1306/C10/frame Page 195 Saturday, May 27, 2000 4:17 PM
196 Powerline Ampacity System: Theory, Modeling, and Applications
Speciﬁc Transmission Cost
The economic evaluation of transmission lines is carried out by comparing the
speciﬁc transmission cost (Vp) of different alternatives. The speciﬁc transmission
cost is deﬁned as the cost per MVA/Km of power delivered by the line given as
follows (Hall, Deb, 1988a):
(10.17)
Wp = Present worth of line = Cp + Co
Cp = Capital cost of line
Co = Capitalized cost of line operation
Capital Cost of Line
For estimation purposes the capital cost of line, Cp, may be obtained by,
Cp = a
1
+ a
2
S + a
3
V (10.18)
Where, a
1
, a
2
, a
3
are the coefﬁcients of the line cost model obtained by statistical
ﬁtting of historical data of line costs at different transmission voltage (V) and
conductor size (S).
Capitalized Cost of Line Operation
The capitalized cost of line operations (Co) includes the cost of losses (Cl) and the
cost of line maintenance (Cm). It is calculated as follows,
Co = k(Cl
j
+ Cm
j
)
j = 1,2…n years (10.19)
k = capitalization factor =
n = life of the line, years
i = interest rate
The annual cost of line losses C
l
is obtained by,
(10.20)
I = conductor current, A
R
ac
= ac resistance of conductor, ohm/km
L
s
= load loss factor
d = cost of energy, $/MWH
Vp
Wp
MVA
1
1
+ ( )
∑
i
j
j
n
–
C
I R L d
S
ac s
1
2
3 8760
⋅ ⋅ ⋅ ⋅ ⋅
1306/C10/frame Page 196 Saturday, May 27, 2000 4:17 PM
Applications 197
The load loss factor L
s
is related to the load factor L
f
by,
(10.21)
Where, k
1
, k
2
are constants (Hall, Deb, 1988a).
The load factor L
f
is deﬁned as,
(10.22)
Optimum Size of Conductor
When planning a new transmission line, it is required to select the optimum size of
conductor for a given maximum power transfer. The optimum size of the conductor
is obtained by minimizing the present worth (Wp) of the total transmission cost as
follows:
(10.23)
From equations 10.17–10.23 we obtain,
(10.24)
With the help of Equation 10.24, we can perform transmission cost evaluation studies
with alternative conductor designs for transmission planning purposes as shown in
the Table 10.4 adapted from (Anand et al. 1985).
r = resistivity of conductor, ohm⋅mm
2
⋅ m
1
l = length of line = 1 km
S = sectional area of conductor, mm
2
Differentiating Wp with respect to S and setting , we obtain the optimum
size of conductor,
(10.25)
L k L k L
s f f
⋅ + ⋅
1 2
2
L
f
×
Energy supplied by the line in a year
Maximum demand 8760
Min Wp
dWp
dS
( ) 0
Wp k
I r l L d
S
Cm a a S a V
s
⋅ ⋅ ⋅ ⋅ ⋅ ⋅
+
]
]
]
+ + +
( )
3 8760
2
1 2 3
R
r l
S
ac
.
dWp
dS
0
S I
k L d r
a
optimum
s
⋅ ⋅ ⋅ ⋅ ⋅ 3 8760
2
1306/C10/frame Page 197 Saturday, May 27, 2000 4:17 PM
198 Powerline Ampacity System: Theory, Modeling, and Applications
and the optimum current density,
(10.26)
Equation 10.26 shows that optimum current density depends upon the factor a
2
,
which represents that portion of the capital cost of line that depends upon conductor
size, S; interest rate, (k ∝ i) ; Load factor, (L
f
∝ L
s
); energy cost, d; and the resistivity
of conductor material, r. The value of r is important — as we approach supercon
ductivity, the optimum current density, J, will become very high.
10.5 LONGDISTANCE TRANSMISSION
Extra High Voltage (EHV) transmission lines are used for the transportation of
electric energy over long distances economically. At the present time the highest
transmission line voltage in North America is 765 kV; there are some experimental
lines capable of reaching voltages up to 1100 kV but they are not in operation.
Transmission line voltage up to 800 kV is operational in many countries, and a
1150 kV EHV AC line is operating in Russia (Alexandrov et al. 1998).
For longdistance transmission line analysis, a lumped parameter equivalent of
a line is no longer accurate, and a distributed parameter representation of the line
is used for transmission line analysis. The following distributed parameter equations
of the transmission line are used for the analysis of transmission line voltage and
current along the length of the line:
(10.27)
TABLE 10.4
Line Data
ACSR
54/7
AAAC(1)
61
AAAC(2)
61
ACAR
54/7
ACSR/AS20
54/7
Compact
54/7
MVA capacity 780 790 810 815 805 800
Economic span, m 425 475 425 400 425 425
Tension, kN 43 48 38 36 43 43
Loss
1
1.0 0.99 0.97 1.06 0.98 0.98
Line cost
1
1.0 0.95 0.96 0.97 0.97 0.97
Transmission cost
1
1.0 0.97 0.96 1.01 0.97 0.97
1
Cost with respect to ACSR
Conductor diameter = 31 mm
AAAC (1) = 53% IACS
AAAC (2) = 56% IACS
J
a
k L d r
optimum
s
⋅ ⋅ ⋅ ⋅ ⋅
2
3 8760
V x V Zc x I
x r r
( ) + ⋅ ( ) cosh sinh γ γ
1306/C10/frame Page 198 Saturday, May 27, 2000 4:17 PM
Applications 199
(10.28)
x = distance from receiving end, km
= Voltage at a point x in line
= Voltage at the receiving end of line
= Current at the receiving end of line
= Current at a point x in line
= propagation constant
Zc = characteristic impedance of the line
The above equation is derived from Appendix 10 at the end of this chapter. It can
be represented by the following matrix equation utilizing the well known A, B, C,
D constants of the line.
(10.29)
The following example will illustrate some interesting features of longdistance
transmission.
Example 10.4
It is proposed to supply large amounts of cheap hydroelectricity by Extra High
Voltage (EHV) transmission line from a location 2500 km away from the load center.
The transmission line voltage is 765 kV AC. Find the following:
1. Surge Impedance Loading for this line
2. Line ampacity and maximum power transmission capacity
3. Transmission line current as a function of line distance for power trans
mission equal to 0.5 SIL, 1 SIL, and 2 SIL.
4. Transmission line voltage as a function of line distance for power trans
mission equal to 0.5 SIL, 1 SIL, and 2 SIL.
The following line constants are assumed for the 765kV line:
R = 0.01 ohm/km
L = 8.35 × 10
–4
H/km
G = 0
C = 12.78 × 10
–9
F/km
Conductor Type: ACSR
Diameter: 35.1 mm
Rdc @ 20°C: 0.04 ohm/km
I
x V
Zc
x I
x
r
r
( )
+ ( )
sinh
cosh
γ
γ
V
x
V
r
I
r
I
x
γ
Vs
Is
A B
C D
Vr
Ir
]
]
]
]
]
]
]
]
]
1306/C10/frame Page 199 Saturday, May 27, 2000 4:17 PM
200 Powerline Ampacity System: Theory, Modeling, and Applications
Number of subconductors / phase: 4
Number of circuit: 1
Frequency: 60 Hz
Latitude: 54°N
Longitude: 77°W
Time of day: 2 pm
Day: Dec 12
Meteorological Conditions:
Ambient temperature: 0°C
Wind speed: 1 m/s
Wind direction: 90° with respect to conductor
Sky condition: Clear sky
Solution
1. Surge Impedance Loading (SIL)
Propagation constant is calculated,
The characteristic impedance is calculated,
2. Line ampacity is calculated by the program by following the procedure
described in Chapter 3 for the speciﬁed transmission line conductor,
meteorological conditions, and by consideration of four subconductors
per transmission line phase.
Ampacity/subconductor = 1920 A
Line Ampacity = 4 × 1920 = 7680 A
γ
γ ω ω
γ
⋅
+ ⋅ ( ) + ⋅ ( )
⋅ + ⋅
Z Y
R jL G jC
j 2 10 1 2 10
5 3 – –
.
Zc
R jL
G jC
Zc j
+ ⋅ ( )
+ ⋅ ( )
−
ω
ω
255 6 4 15 . .
SIL
SIL MVA
⋅
( )
⋅
765 10
255 6
10
2289
3
2
6
.
–
1306/C10/frame Page 200 Saturday, May 27, 2000 4:17 PM
Applications 201
Maximum power transmission capacity of the line =
· 765 · 7680 ·
10
–3
= 5875.2 MVA
= 5875.2 MVA
To give an idea, this power is sufﬁcient for many metropolitan cities.
3. Transmission line current as a function of line distance
The receiving end currentm I
r
, is,
Line to ground voltage V
r
,
The transmission line current as a function of distance is obtained from
Equation 10.28,
The value of line current as a function of the distance from receiving end
is shown in Figure 10.12. It is interesting to observe the variation of line
current as a function of the distance. From this ﬁgure we can see that the
thermal limit of the line is not exceeded even at two times the surge
impedance loading (2 SIL) of the line.
4. Transmission line voltage as a function of line distance is calculated
similarly from Equation 10.27 and is shown in Figure 10.13. It is inter
esting to observe the variation of line voltage as a function of line distance.
From this ﬁgure we can see that there is substantial increase in line voltage
at midpoint when power transfer is increased beyond the surge impedance
load of the line.
10.6 PROTECTION
The effect of variable transmission line ratings on system protection requires careful
evaluation to ensure proper functioning of the protective relaying system for both
transmission and distribution lines. Transmission and distribution systems are generally
provided with overcurrent and earth fault relays, impedance relays, differential relays,
and voltage and underfrequency relays. These protective devices are designed to offer
3
I
Ir
r
⋅ ⋅
2289
3 765 10
3455
3
Vr
Vr kV
⋅
765 10
3
441
3
I
j x
j
j x
x
⋅ + ⋅
( )
⋅
{ ¦
⋅ ⋅
+ + ⋅
( )
⋅
{ ¦
⋅
sinh . .
. – .
cosh . .
– –
– –
2 10 1 2 10 441 10
255 6 4 15
2 10 1 2 10 3455
5 3 3
5 3
1306/C10/frame Page 201 Saturday, May 27, 2000 4:17 PM
202 Powerline Ampacity System: Theory, Modeling, and Applications
FIGURE 10.8 Optimum conductor temperature is 80°C as seen in the above ﬁgure.
FIGURE 10.9 Optimum value of transmission line ampacity is 848 A.
FIGURE 10.10 Optimum conductor temperature as a function of energy cost. As seen in
this ﬁgure the optimum value of transmission line conductor temperature increases with lower
electric energy cost.
2500
2000
1500
1000
500
0
50 70 90 110 130 150 170
o
190 210
Conductor Temperature, C
Optimum Conductor Temperature
230 kV Line ACSR Cardinal
T
r
a
n
s
m
i
s
s
i
o
n
C
o
s
t
,
$
/
M
V
A
/
K
m
Optimum Conductor Ampacity
230 kV Line ACSR Cardinal
2500
2000
1500
1000
500
0
T
r
a
n
s
m
i
s
s
i
o
n
C
o
s
t
,
Ampacity, A
$
/
M
V
A
/
K
m
5
0
0
8
4
0
1
0
5
7
1
2
2
4
1
3
6
1
1
4
7
8
1
5
8
3
1
6
7
8
1
8
0
0
Energy Cost, $/MWh
T
e
m
p
e
r
a
t
u
r
e
,
C
o
Optimum Conductor Temperature vs.
Energy Cost
100
90
80
70
60
50
40
30
50
20
10
20 30 40 50 60 70 80 90 100
1306/C10/frame Page 202 Saturday, May 27, 2000 4:17 PM
Applications 203
FIGURE 10.11 The optimum value of transmission line ampacity increases with lower
electric energy cost.
FIGURE 10.12 Variation of transmission line current as a function of distance for different
power transmission levels.
FIGURE 10.13 Variation of transmission line voltage as a function of distance for different
power transmission levels.
Energy Cost, $/MWh
A
m
p
a
c
i
t
y
,
A
OptimumTransmission Line Ampacity
vs. Energy Cost
1200
1000
800
600
400
200
0
20 30 40 50 60 70 80 90 100
8000
7000
6000
5000
4000
3000
2000
1000
0
0 500
Distance, km
1000 1500 2000 2500
Transmission Line Current vs Distance
L
i
n
e
C
u
r
r
e
n
t
,
A
SIL
0.5 SIL
2 SIL
0 500 1000 1500 2000 2500
2 SIL
0.5 SIL
SIL
Transmission Distance, km
S
e
n
d
i
n
g
a
n
d
R
e
c
e
i
v
i
n
g
V
o
l
t
a
g
e
R
a
t
i
o
2.5
2
1.5
1
0.5
0
Transmission Line Voltage vs Distance
1306/C10/frame Page 203 Saturday, May 27, 2000 4:17 PM
204 Powerline Ampacity System: Theory, Modeling, and Applications
protection to transmission lines, distribution feeders, transformers, substation busbars,
and generators, as well as the loads they serve. For satisfactory functioning of the
protection system, the system must be able to distinguish between permissible overload
current and a fault current to avoid faulty tripping during an acceptable overload
condition.
To ensure fault discrimination, protection relays are timecoordinated so that the
circuit breaker closest to the fault opens ﬁrst. Backup protection is provided so that
if the breaker closest to the fault fails to operate, the next breaker will open.
Automatic reclosures are also provided on most circuits for automatic recovery from
temporary faults. Reclosurers are circuit breakers that close automatically at prede
termined intervals after opening a circuit to eliminate faults that are temporary in
nature. For all of the above protection schemes, the magnitude of the overcurrent
and relay operation times are determined from network shortcircuit studies with
proper prefault and postfault line operating conditions.
Traditionally, the relay pickup current setting for overload protection was deter
mined by static line ratings. In a static system, the ratings of transmission lines,
transformers, and other substation currentcarrying equipment are usually considered
to be constant during a season, resulting in winter and summer ratings. For networks
having dynamic line ratings, the protective relaying settings will have to be updated
continuously, preferably on a realtime basis. This is the subject of adaptive relaying
and beyond the scope of this book. The interested reader is referred to an excellent
book on the subject of adaptive computer relaying (Phadke et al.1988) and other
excellent technical papers in IEEE, Cigré, and similar conferences.
10.7 CHAPTER SUMMARY
Applications of powerline ampacity system to power system economic operation, load
ﬂow, generator stability, transmission line planning, and design considerations of over
head powerlines in view of powerline ampacity were presented in this chapter. A
formulation of the optimal power ﬂow problem was given to show the signiﬁcance of
transmission line dynamic thermal ratings in the economic operation of an intercon
nected electric power system having diverse generation sources. Electricity production
costs were evaluated with static and dynamic line rating in a transmission network
having diverse generation sources. Results were presented to show the savings in
electricity production cost achieved by the dynamic rating of transmission lines using
LINEAMPS.
Examples were provided in the chapter to show the impact of high transmission
line ampacity on steadystate generator stability, and dynamic and transient stability.
It was shown that power system stability limits are enhanced by dynamic line ratings.
The application of a powerline ampacity system in the planning and design of
new overhead lines includes the selection of optimum conductor size, optimum
current density, and the evaluation of alternative conductor designs. For this purpose,
a complete formulation of transmission line economics was presented. Economic
analysis of existing lines show that the construction of new lines, or the reconduc
toring of existing wires, may be postponed in many cases by dynamic line ratings,
with substantial cost savings.
1306/C10/frame Page 204 Saturday, May 27, 2000 4:17 PM
205
Appendix 10
Transmission Line Equations
Applying Kirchoff’s law to a small section dx (dx <<
λ
) of the line we have,
(A10.1)
(A10.2)
where the product of differential quantities are neglected. From above, the following
secondorder linear differential equations are obtained,
Voltage,
(A10.3)
(A10.4)
Current,
(A10.5)
FIGURE A10.1
Long transmission line model.
V V
1
1 x
x x
x
x
x
x 2
x
V V = V – dV
I = I – dI
Y
d
d
I I
Z
x
0
dV V V dZ I dI dZ I Z I dx
x x x x x x x
= ′ =
( )
≈ ⋅ = ⋅ – –
dI I I dY V Y V dx
x x x x x
= ′ = ⋅ = ⋅ –
d V
dx
Z Y V V
x
x x
2
2
= ⋅ ⋅ = γ
d V
dx
V
x
x
2
2
0 − = γ
d I
dx
Z Y I I
x
x x
2
2
= ⋅ ⋅ = γ
1306/appendix 10/frame Page 205 Saturday, May 27, 2000 9:08 AM
206
Powerline Ampacity System: Theory, Modeling, and Applications
(A10.6)
where the propagation constant is given by,
(A10.7)
The solutions of differential equations (10.4), (10.6) are,
(A10.8)
(A10.9)
at x = 0 we have,
(A10.10)
(A10.11)
(A10.12)
(A10.13)
(A10.14)
(A10.15)
(A10.16)
(A10.17)
substituting x =
ᐉ
, we obtain the sending end voltage current as follows,
d I
dx
I
x
x
2
2
0 − = γ
γ
γ = ⋅ Z Y
V k e k e
x i
x
r
x
= +
γ γ –
I
k
Zc
e
k
Zc
e
x
i x r x
= −
γ γ –
V k k
i r 2
= +
Z I k k
c i r 2
= −
k
V Z I
i
c
=
+
2 2
2
k
V Z I
r
c
=
−
2 2
2
V
V e e
Z I e e
x
x x
c
x x
=
+
( )
+ −
( )
2
2
2
γ γ
γ γ
–
–
I
V e e
Zc
I
e e
x
x x
x x
=
+
( )
+ −
( )
2
2
2 2
γ γ
γ γ
–
–
V x V Zc x I
x
= ( ) + ⋅ ( ) cosh sinh γ γ
2 2
I
x V
Zc
x I
x
=
( )
+ ( )
sinh
cosh
γ
γ
2
2
1306/appendix 10/frame Page 206 Saturday, May 27, 2000 9:08 AM
Appendix 10 Transmission Line Equations
207
(A10.18)
(A10.19)
Equations (A10.18) and (A10.19) in matrix form relating sending end voltage, Vs,
and current, Is, to receiving end voltage, Vr, and current, Ir, is,
(A10.20)
where,
V V Zc I
s
= ⋅ ( ) + ⋅ ⋅ ( ) cosh sinh γ γ l l
2 2
I
V
Zc
I
s
=
⋅ ( )
+ ⋅ ( )
sinh
cosh
γ
γ
l
l
2
2
Vs
Is
A B
C D
Vr
Ir
=
A = ⋅ ( ) cosh γ l
B I
r
= ⋅ ( ) sinh γ l
C
Zc
=
⋅ ( ) sinh γ l
D = ⋅ ( ) cosh γ l
1306/appendix 10/frame Page 207 Saturday, May 27, 2000 9:08 AM
209
11
Summary, Future Plans
and Conclusion
11.1 SUMMARY
As the demand for electricity grows in all regions of the world, there is greater need
to develop better ways of operating existing Transmission and Distribution (T&D)
networks for more efﬁcient utilization of existing facilities. Historically, when
demand increased, electric utilities added new T&D capacity by the construction of
new lines and substations. Due to public concern about protecting the environment,
and to population growth, public authorities are paying greater attention to properly
locating electric power facilities and proper land use. As a result, there is greater
R&D effort by power companies and concerned government agencies to minimize
environmental impact, increase energy efﬁciency, and improve land use.
In this book, a complete system of rating overhead powerlines is developed
giving theory, algorithms, and a methodology suitable for implementation in a
computer program. The program is not only an operational tool enabling better
utilization of existing transmission and distribution facilities, but is also a valuable
planning tool for line maintenance, and a design tool for the construction of future
T&D facilities. The line ampacity system described in this book offers an integrated
line ampacity system comprising a transmission line model, a conductor model, and
a weather model for the ﬁrst time. By developing a system of rating overhead lines
as a function of forecasting weather conditions by an objectmodel and expert
system, a very userfriendly program is realized that is easily implemented in all
geographic regions.
Chapter 1 introduced the subject of transmission line ampacity and presented
the line ampacity problem. It is stated that voltage and stability limits can be
improved by control of reactive power and/or boosting voltage levels by transformer
action. Therefore, in many cases the transport capacity of overhead powerlines is
limited only by the thermal rating of powerline conductors.
Chapter 2 described line rating methods from the early works of Ampere,
Faraday, and several researchers from different countries who were concerned with
the problem of transmission line ampacity, and who offered solutions to increase
line capacity. Utility line rating practices, and online and ofﬂine methods are dis
cussed, including realtime ratings, forecast ratings, probabilistic ratings, and static
line ratings.
Chapter 3 provided a complete theory of conductor thermal ratings. First, a
threedimensional conductor thermal model is developed, and then steadystate,
dynamic, and transient thermal rating models are developed from it. The concept of
1306/C11/frame Page 209 Saturday, May 27, 2000 4:27 PM
210
Powerline Ampacity System: Theory, Modeling, and Applications
steady, dynamic, and transient ratings are introduced for the ﬁrst time in this chapter
and are related to machine stability in Chapter 10.
Chapter 4 described experimental research in the laboratory, on outdoor trans
mission line test span, as well as ﬁeld measurements on real transmission line
circuits. This chapter presents uptodate knowledge in the ﬁeld of line ampacity
calculations by comparing different line ampacity calculation methods. Theoretical
results that were obtained by calculation from the transmission line conductor ther
mal models developed in Chapter 3 are compared to IEEE and Cigré standards with
excellent agreement.
Chapter 5 presented elevated temperature effects of different types of transmis
sion line conductors. Experimentallydetermined empirical models of loss of tensile
strength of conductors and permanent elongation due to elevated temperature oper
ations were presented with results from each model. A probability method of cal
culation of transmission line sag and tension was presented for the ﬁrst time. The
probability distribution of conductor temperature in service was obtained by a
MonteCarlo simulation of time series stochastic models. Computer algorithms were
presented for the recursive estimation of loss of strength and permanent elongation.
Results were compared with data from industry standards and transmission line ﬁeld
data with excellent agreement.
Chapter 6 presented a study of the electric and magnetic ﬁelds from highvoltage
power transmission lines. This aspect of transmission line ampacity is signiﬁcant
because there is little previous work carried out in this direction. A complete theory
with examples of the method of calculation of the electric and magnetic ﬁelds from
a line were presented to form a clear understanding of the subject. Even though
there is no evidence of any signiﬁcant environmental impact by EMF due to increased
transmission line current, measures are suggested for the reduction of electric and
magnetic ﬁelds from transmission lines by new conductor conﬁgurations, and by
active and passive shielding.
Chapter 7 described various approaches to weather modeling for the prediction
of line ampacity. Since weather is an important parameter in transmission line
ampacity calculations, the development of weather models of ambient temperature,
wind speed, wind direction, and solar radiation were presented. First, statistical
weather models based upon timeseries analysis of National Weather Service fore
casts were developed. Then, neural network modeling was presented for forecasting
and pattern recognition. Examples of hourly values of future meteorological condi
tions generated from the models are given. Application of fuzzy set modeling of
transmission line ampacity is also given.
Chapter 8 described the computer modeling method of the line ampacity system
and its implementation in a computer program called LINEAMPS. A signiﬁcant
contribution made by the program is that it not only predicts line ampacity, but also
offers a complete environment for the management, planning, and operation of
overhead powerlines. The program offers a userconﬁgurable transmission line data
base for the ﬁrst time, thanks to recent developments in object oriented modeling
technology (Cox 2000). The database is comprised of powerlines, different types of
powerline conductors, and weather stations for any geographic region.
1306/C11/frame Page 210 Saturday, May 27, 2000 4:27 PM
Summary, Future Plans and Conclusion
211
LINEAMPS provides the ability to create new line, new conductors and new
weather stations. It is a userfriendly program created by the application of artiﬁcial
intelligence using objectmodel and expert system rules. Users of the program have
less chance of making errors in data input because expert rules check user input and
explain error messages. It is also a contribution to the ﬁeld of cognitive science applied
to electric power transmission where a computer program resembles a human expert.
Real examples* of computer modeling of the line ampacity system in two
different geographic regions of the world are presented in Chapter 8. It shows the
suitability of the program in all geographic regions of the world. As of this writing,
LINEAMPS is being used or evaluated in the following power companies in the
different regions of the world.
• TransPower, New Zealand (Figure 11.5)
• Korea Electric Power Company, South Korea (Figure 11.4)
• Electricité de France (EDF), Paris, France (Figure 11.3)
• Hydro Quebec, Canada (Figure 11.2)
Chapter 9 discussed the state of the art in the development of power semiconductor
devices for FACTS applications. The various types of FACTS devices are described
in this chapter, including Static Var Compensating (SVC) devices, Thyristor Con
trolled Series Compensation (TCSC), STATCOM, UPFC, and Superconducting
Magnetic Energy Storage (SMES). An example of an actual FACTS installation in
a highvoltage substation is also provided.
Chapter 10 presented applications of the line ampacity system to transmission
system planning, economic power system operation, and loadﬂow and generator
stability issues. Cost saving by the implementation of a variable line rating system
in an interconnected electric network was presented by example. An important
feature of the new line ampacity system in a competitive electricity supply market
is the ability to forecast hourly values of line ampacity up to seven days in advance,
enabling advance purchase and sale of electricity.
A formulation of the optimal power ﬂow problem was presented in Chapter 10
to show the signiﬁcance of transmission line dynamic thermal ratings to the eco
nomic operation of an electric power system with diverse generation sources. The
application of the powerline ampacity system in the planning and design of new
overhead lines includes the selection of optimum conductor size, optimum current
density, and the evaluation of alternative conductor designs. For this purpose, a
complete formulation of transmission line economics was presented. Economic
analysis of existing lines shows that the construction of new lines or the reconduc
toring of existing wires can be postponed in many cases by dynamic line ratings,
with substantial cost savings. The cost of electricity may be reduced by greater
utilization of economy energy sources and, last but not least, there is a lowering of
environmental impact, both visual and ecological.
*
LINEAMPS for New Zealand User Manual
, 1995, describes modeling transmission lines and weather
stations for the region of New Zealand, North and South Island.
LINEAMPS for Korea User Manual
, 1996, describes modeling transmission lines and weather stations
for the region of South Korea.
1306/C11/frame Page 211 Saturday, May 27, 2000 4:27 PM
212
Powerline Ampacity System: Theory, Modeling, and Applications
11.2 MAIN CONTRIBUTIONS
The Concept of Steady, Dynamic and Transient Line Rating
Most of the literature on transmission line conductor thermal rating considers the
shortterm thermal rating of a conductor as its transient thermal rating. In effect, it
would be more appropriate to call shortterm rating “dynamic line rating,” because
the time span is of the order of several seconds, which corresponds well with the
time span considered in dynamic stability studies on generators.
The transient thermal rating of the conductor is considered when there is a short
circuit or lightning currents. Therefore, the time span of transients is much shorter,
generally in the order of milliseconds, and corresponds well with the time span
considered in transient stability studies on generators. Thus, the concept of steady
state rating, dynamic rating,* and transient line rating were introduced for the ﬁrst
time in relation to generation stability as presented in Chapter 10.
The above terminology of the different transmission line ratings is consistent
with other areas of the electric power system relating to the steadystate stability,
dynamic stability, and transient stability of electrical generators. The steadystate
transmission line conductor thermal rating is applicable to the condition of steady
state operation of a power system; dynamic thermal rating is applicable during power
system dynamic operations; and transient thermal rating is applicable during power
system transient operating conditions.
ThreeDimensional Conductor Thermal Model
A threedimensional conductor thermal model was used in a previous wind tunnel
study conducted by the author at PG&E to determine radial temperature distribution
inside a transmission line conductor (Hall, Savoullis, Deb, 1988). A threedimen
sional conductor thermal model was developed in a recent report (Cigré, 1997) for
the calculation of transmission line ampacity.
In Chapter 4, the differential equation of conductor temperature was developed
from a threedimensional conductor thermal model. The main reason for selecting
this model is that it enables the calculation of radial temperature distribution and
the average conductor temperature within the conductor. As we may recall from
Chapter 5, the sag and tension of a transmission line conductor is calculated from
the average conductor temperature.
Transmission Line Risk Evaluation
By evaluating weather conditions and transmission line temperatures in the region
of Detroit (Davis, 1977), it was shown that transmission line capacity was greater
than static line capacity for a large percentage of the time in that region. Davis also
showed that static line rating is not risk free, and there exists a small percentage of
time when maximum conductor temperature is exceeded even by static line rating.
* For a discussion on dynamic rating terminology, please see the discussion contribution in the references
(Hall, Deb, 1988a).
1306/C11/frame Page 212 Saturday, May 27, 2000 4:27 PM
Summary, Future Plans and Conclusion
213
Similarly, the research carried out at PG&E in California also showed that actual
transmission line ampacity is signiﬁcantly greater than static line ratings (Hall and
Deb, 1988b) for a large percentage of time. A transmission line risk analysis study
was carried out at PG&E which showed there is minimum risk to a dynamic line
rating system (PG&E Report. 1989).* The results presented in this book show that
there is minimum risk as conductor sag; loss of strength and EMF were calculated
and found to be within acceptable limits.
RealTime Rating
Realtime line rating systems (Davis, 1977), (Seppa, T.O., 1998), (Soto et al., 1998),
(Deb, 1998) calculate line ampacity by the measurement of weather variables,
conductor temperature, and/or conductor tension. In Davis’s system, realtime line
ratings were calculated by monitoring conductor temperature with sensors installed
at different sections of the line. In the PG&E system (Mauldin et al., 1988), (Steeley
et al., 1991), (Cibulka et al., 1992), line ampacity is calculated by monitoring ambient
temperature only. The PG&E line rating system also offered one to 24 hours ahead
forecast rating capability. For this reason, realtime stochastic and deterministic
models were developed by the author at PG&E to forecast transmission line ampacity
up to 24 hours in advance (Hall, Deb, 1988b), (Steeley et al., 1991).
Another realtime line rating system calculates ampacity by the measurement
of conductor tension (Seppa, T.O., 1998). The relationship between transmission
line conductor temperature and tension is presented in the conductor change of state
equation in Chapter 5. By the application of the change of state equation we can
easily calculate conductor temperature if conductor tension is measured. Ampacity
is then calculated from conductor temperature by the application of the conductor
thermal models presented in the Chapter 3. It is difﬁcult to obtain forecast ratings
by monitoring conductor tension or conductor temperature alone. It is easier to
forecast line ratings by forecasting weather conditions.
Installation of temperature sensors or tension monitors requires taking the line
out of service for periodic maintenance. It also requires communication with sensors
installed on the line to the utility power control center computer where line ampacity
is calculated by a program. The prediction of line ampacity by monitoring weather
conditions does not require any new hardware to be installed on a transmission line.
Weather conditions are generally monitored in electric utility power systems for
other reasons, such as load forecasting, etc. Therefore, realtime weather data is
available in most electric power company control centers at no additional cost. It is
also expensive to install temperature or tension monitoring devices on all transmis
sion lines. Thus, at the present time, these devices are installed on a limited number
of heavilyloaded critical transmission circuits.**
For the above reasons, a line monitoring system is expected to complement the
more generalpurpose line rating system developed in this book. The line rating
* The report presents a quantitative analysis of the risk of an ambient adjusted line rating system developed
at PG&E.
** (Waldorf, Stephen, P., Engelhardt, John S., 1998). This paper describes realtime ratings of critical
transmission circuits by monitoring temperature of overhead lines, switchgear, and power transformers.
1306/C11/frame Page 213 Saturday, May 27, 2000 4:27 PM
214
Powerline Ampacity System: Theory, Modeling, and Applications
methodology presented in this book offers a more economical system that can be
easily implemented for the rating of all lines in all geographic regions.* (See Figures
11.2–11.5). The Spanish system (Soto et al., 1998) is already beginning to follow
this approach. However, their system lacks forecast rating capability, weather mod
eling, and the ability to make new conductors, new weather stations, and new
transmission lines, which are the objects of the author’s patent (Deb. 1999).**
*
LINEAMPS for HydroQuebec, Canada, Software Users Guide
, 1998, describes the line campacity
system developed for HydroQuebec in Canada.
LINEAMPS for EDF, France, Software Users Guide
, 1998, describes the line ampacity system developed
for EDF, France.
LINEAMPS for TransPower, New Zealand, Software Users Guide
, 1996, describes the line ampacity
system developed for TransPower, NZ.
LINEAMPS for KEPCO, S. Korea, Software Users Guide
, 1996, describes the line ampacity system
developed for TransPower, NZ.
FIGURE 11.2
LINEAMPS for Hydro Quebec. Map shows the geographic region of Quebec,
Canada and weather stations created by program. A 765 kV transmission line from La Grande
hydroelectric stations to Montreal, Quebec is shown in the ﬁgure.
** Anjan K. Deb, 1999. “Objectoriented line ampacity expert system.” U.S. Patent.
1306/C11/frame Page 214 Saturday, May 27, 2000 4:27 PM
Summary, Future Plans and Conclusion
215
Forecast Rating
In the line ampacity system developed in this book, ampacity is forecast up to seven
days in advance by adjusting the LINEAMPS weather model to National Weather
Service forecasts. Hourly values of ambient temperature, wind speed, and solar
radiation are generated by AmbientGen, WindGen, and SolarGen methods in each
weather station object as described in Chapter 7. At the present time, line ampacity
forecasts are limited to seven days in advance because weather forecasts are less
accurate beyond that period.*
Weather and Line Current Modeling
A timeseries stochastic model of ambient temperature for the prediction of trans
mission line conductor temperatures was ﬁrst presented by the author (Deb, 1985)
at the (Cigré Symposium 1985)** on High Currents and (Hall, Deb, 1988b). The
BoxJenkins forecasting model is quite accurate, but the model coefﬁcients could not
be easily adapted in realtime. Therefore, a recursive least square estimation model
was developed at PG&E suitable for realtime calculation (Steeley, Norris, and Deb,
1991), (Cibulka, Steeley, Deb, 1992). Statistical models based on hourly differences
of ambient temperature were also proposed by others (Douglass, 1986), (Foss, Maraio,
FIGURE 11.3
LINEAMPS for EDF, France. EDF is the national electric company of France.
Weather stations are created by the LINEAMPS program. A 400 kV transmission line from
Paris to Bordeaux created by program is also shown in the ﬁgure.
* Weather section,
USA Today
, November 23, 1998.
** The Cigré Symposium was exclusively devoted to the subject of transmission line ampacity.
1306/C11/frame Page 215 Saturday, May 27, 2000 4:27 PM
216
Powerline Ampacity System: Theory, Modeling, and Applications
1989) but it was shown in a discussion contribution* that realtime recursive estimation
was more accurate. The following forecasting models were evaluated:
• Fourier series model
• Recursive least square estimation algorithm
• Kalman Filter
• Neural network
The results presented in the Chapter 3 and Chapter 4 show the accuracy of each
model. A transmission line current model is useful for the prediction of conductor
temperature, and may be developed by using the forecasting models mentioned
above (Deb et al., 1985).
The Fourier series model of ambient temperature and wind speed is used in the
LINEAMPS program. This model is recommended for its simplicity, and is suitable
when a generalpurpose weather forecast of a region is available.** A neural network
is useful for weather pattern recognition, as shown in Section 4.3. A recursive
FIGURE 11.4
LINEAMPS for KEPCO, South Korea. KEPCO is the national electric
company of South Korea. Weather stations are created by the LINEAMPS program. A 345
kV transmission line from Seoul to Wonju, also created by program, is shown in the ﬁgure.
* Discussion contribution by Anjan K. Deb and J.F. Hall to the IEEE paper (Douglass, D.A., 1988).
** Generalpurpose weather forecast issued by the National Weather Service includes daily maximum
and minimum values of ambient temperature for three to seven days in advance. The data is generally
published in daily newspapers as well as in several Internet web sites: www.intellicast.com,
www.nws.noaa.gov.
1306/C11/frame Page 216 Saturday, May 27, 2000 4:27 PM
Summary, Future Plans and Conclusion
217
estimation algorithm and Kalman ﬁlter are suitable for realtime forecasting of
transmission line ampacity on an hourly basis.
Transmission Line ObjectModel
Objectoriented modeling is a new way of developing computer programs that makes
use of inheritance, polymorphism, communication by message, and delegation of
messages (Booch, G., 1993), (Cox, E., 2000), (KappaPC).* The objectmodel is
particularly suitable for electric power system applications for the modeling of
transmission lines, generators, and other substation equipment, and is seriously
considered by electric power companies (
MPS Review
Article, 1998). The object
model approach is used to model transmission lines, weather stations and transmis
sion line conductors in the LINEAMPS program (Deb, 1995, 1997, 1998). By using
the objectmodel approach, it is shown in Chapter 5 how transmission line objects
are easily created with attributes and behavior by class inheritance.
Examples of the objectmodel that were created for the regions of New Zealand
and South Korea are presented in Chapter 5. It is seen in this chapter how a
systematic method of transmission line ampacity system data classiﬁcation is
developed. Transmission lines are classiﬁed by voltage levels, weather station data
FIGURE 11.5
LINEAMPS for TransPower, New Zealand. TransPower operates the trans
mission grid in New Zealand. The weather stations shown in the above map are created by
the LINEAMPS program. A 375 kV transmission line from Auckland to Wellington in the
North Island of New Zealand is also shown in the ﬁgure.
* KappaPC ver 2.4 provides connection to industry standard databases by ODBC, which emables the
LINEAMPS program to access transmission line and weather data.
1306/C11/frame Page 217 Saturday, May 27, 2000 4:27 PM
218
Powerline Ampacity System: Theory, Modeling, and Applications
is classiﬁed by geographic regions, and conductor data is classiﬁed by conductor
types. These are implemented in the LINEAMPS program to enable users to easily
create their own objects.
Line Ampacity ExpertSystem
The declarative style of programming by rules was used to develop the LINEAMPS
program. There are several advantages of programming by rules instead of purely
by procedures (Kronfeld, Kevin M., Tribble, Alan C., 1998). For example, expert
systems use an inference engine to make decisions, while in a procedural language
the programmer writes the code for decision making. The LINEAMPS program uses
an expert system to check user input data. Similarly, there are other expert systems
developed for the power industry for fault diagnosis (Taylor et al., 1998),* intelligent
tutoring systems (Negnevitsky, 1998), and power quality (Kennedy, B. 2000).
LINEAMPS is the ﬁrst transmission line expert system using objectoriented
modeling and rules (Deb, 1995). As shown in Chapter 8, rules of thumb and practical
knowledge are easily implemented by an expert system. Examples are presented to
show how the program calculates powerline ampacity by objects and rules, checks
user input data, and explains error messages like a true expert.
Experimental Veriﬁcation of Transmission Line Ampacity
Experimental work carried out at the PG&E wind tunnel to verify the conductor
thermal model (Hall, Savoullis, Deb, 1988) is described in Chapter 6. A new equation
was developed from wind tunnel data to determine the Nusselt number as a function
of the Reynolds number (Ozisik, M.N., 1985) for the calculation of convection
cooling by wind. Results of steady state ampacity and dynamic and transient ampac
ity obtained from the LINEAMPS program are presented in a table and compared
to wind tunnel data and other data compiled from various power companies (Urbain
J.P., 1998), (PG&E Standard, 1978)** and a conductor manufacturer’s catalog
(Southwire, 1994)*** with excellent agreement. The values of steadystate ampacity
and dynamic ampacity calculated by the LINEAMPS program also compared well
with the IEEE standard and a recent Cigré report.
Hourly values of an actual 345kV transmission line ampacity was available from
Commonwealth Edison Company**** (ComEd), Chicago. The line ampacity data
was obtained from a realtime line rating system operated by ComEd. A comparison
of this data from 8:00 a.m. to 4:00 p.m. showed that LINEAMPS ratings never
exceeded realtime ratings at any time. LINEAMPS also accurately predicted the
lowest rating at 12:00 noon.
* (Taylor et al., 1998) provides a list of expert systems developed for the power industry worldwide.
** (PG&E Standard, 1978) gives the static line rating for summer and winter of all conductors used by
PG&E for normal and emergency conditions, and the corresponding assumptions of meteorological
conditions.
*** (Southwire, 1994) provides the ampacity of all commonly used conductors in the U.S., and the
corresponding assumptions of meteorological conditions.
**** Thanks are due to Mr. S. Nandi, Technical Expert, Transmission Line Thermal Rating Studies,
Commonwealth Edison, Chicago, IL.
1306/C11/frame Page 218 Saturday, May 27, 2000 4:27 PM
Summary, Future Plans and Conclusion
219
The above example showed that LINEAMPS ratings have minimum risk of
exceeding allowable maximum conductor temperatures.
Effects of Higher Transmission Line Ampacity
When transmission line ampacity is increased, it is necessary to properly evaluate
the thermal effects of the powerline conductor as well as the electric and magnetic
ﬁelds of the transmission line. A uniﬁed approach to the modeling and evaluation
of the effects of higher transmission line ampacity is presented in Chapter 5. The
thermal effects include conductor loss of strength and permanent elongation of the
transmission line conductor. These effects are evaluated recursively from the prob
ability distribution of conductor temperature by using Morgan’s equation (Morgan,
1978).
Morgan calculated the loss of tensile strength of conductor by a percentile
method. The same method was recently used to calculate the thermal deterioration
of powerline conductors in service in Japan (Mizuno et al., 1998, 2000). In their
study, the reduction in tensile strength of the conductor was used as the index of
thermal deterioration of the conductor. They showed by simulation that conductor
loss of strength is less than 1% by static line rating. They also showed that the
frequency distribution of a conductor should consider the relationship between
weather and line current. In Chapter 5, the probability distribution of conductor
temperature is generated by a MonteCarlo simulation of weather data from time
series stochastic models and transmission line currents which consider the correlation
between weather and load current.
In addition to the loss of tensile strength, conductor sag and tension are also
affected by elevated temperature operation due to high currents (Cigré, 1978). A
new method is developed to determine the sag and tension of overhead line conduc
tors with elevated temperature effects. The accuracy of sagtension program was
tested with ALCOA program (Lankford, 1989), STESS program (CEA Report,
1980), and KEPCO’s transmission line ﬁeld data (Wook, Choi, Deb, 1997). Results
are presented which compare well with all of the above data.
The magnetic ﬁeld at ground level of a transmission line increases with line
ampacity and conductor sag (Rashkes, Lordan, 1998). Most studies on magnetic
ﬁelds from powerlines are conducted with typical transmission line currents equal
to or less than static line ratings. The magnetic ﬁelds of conductors with dynamic
ampacity are presented in Chapter 6.
Typical powerline conﬁgurations are evaluated to show the magnetic ﬁelds with
high currents. It is shown that the magnetic ﬁelds of overhead transmission lines
with dynamic line ampacity are within acceptable limits. Line design and EMF
mitigation methods are suggested to lower transmission line magnetic ﬁelds in
sensitive areas (Böhme et al., 1998). The electric ﬁeld of a highvoltage line at
ground level does not depend upon transmission line ampacity if the maximum
design temperature of the conductor is not exceeded and minimum conductorto
ground distance is maintained.
1306/C11/frame Page 219 Saturday, May 27, 2000 4:27 PM
220
Powerline Ampacity System: Theory, Modeling, and Applications
Line Ampacity Applications
The application of a dynamic line rating system in the economic operation of an
electric power system was ﬁrst presented by (Hall, J. F., Deb, Anjan K., 1988a), and
later on by (Deb, 1994). Deb developed the economic analysis method of dynamic
line rating system with adaptive forecasting to demonstrate power system operational
cost savings by dynamic thermal rating. Substantial capital cost savings by the
deferment of capital investment required for the construction of new lines and
environmental beneﬁts were shown, as fewer lines are required.
A recent study (Yalcinov, T. and Short, M. J., 1998) on generation system cost
optimization by neural network now conﬁrms that substantial cost saving is possible
by increasing transmission line capacity. The theory and mathematical model of a
hypothetical utility system are presented in this book to demonstrate electric power
system economy achieved by the new line ampacity system. In addition to the solution
of the optimal power ﬂow problem by a classical solution of the nonlinear optimization
problem (Bergen, A., 1986), artiﬁcial neural network (Lee, K.Y. et al., 1998), and
genetic algorithm GA (Wong, Yuryevich, 1998) were developed in the industry to
obtain faster and more efﬁcient solutions to the optimal power ﬂow problem.
11.3 SUGGESTIONS FOR FUTURE WORK
Overhead powerlines constitute the single most important component of the electrical
power system. Therefore, the contributions made in this book to increase line capac
ity of existing overhead power transmission and distribution lines will have a sig
niﬁcant impact in the improvement of the overall performance of the electrical power
system. The power system will continue to evolve as our demand for electricity
increases in a fair and competitive environment based on free market principles.
Power plants and transmission lines require substantial investments, which must
be carefully evaluated before new facilities are added. Following is a list of areas
that require further research for optimum utilization of existing assets, and the
development of new transmission line technologies with greater emphasis on renew
able energy sources and minimum environmental impact. A plan to develop the line
ampacity system further for application in a deregulated electricity production,
supply, and distribution environment is also proposed. It is hoped that these studies
will lead to the optimum utilization of all resources.
Overcoming Limitations of Existing AC Networks by the
Improvement of Power System Stability, Economy, Energy
Transfer, and Reliability
In the past, the ampacity of existing overhead powerlines could not be fully utilized
because existing Alternating Current (AC) circuits are mostly composed of passive
elements having very little controllability. Therefore, when demand increases, exist
ing network control methods are not sufﬁcient to properly accommodate increased
power ﬂows. As a result, certain lines are more heavily loaded and stability margins
are reduced. Due to the difﬁculty of controlling power ﬂow by existing methods,
1306/C11/frame Page 220 Saturday, May 27, 2000 4:27 PM
Summary, Future Plans and Conclusion
221
FACTS devices are being developed and are used to control existing T&D networks.
These devices are presently installed at a number of locations worldwide, at all
voltage levels up to 800 kV, with varying capacities.
FACTS Technology
Due to the development of FACTS technology it is now possible to fully exploit
existing transmission line capacities. By locating FACTS devices at suitable locations
on existing networks, it is now possible to exercise a range of control over the AC
network in a manner that was not possible before. Thyristor controlled shunt and
series capacitors are now widely used in the electric power system to increase the
transmission capacity of existing lines. Similarly, SMES devices are installed for
faster response and voltage support, and to maintain stability.
Increasing transmission line capacity by dynamic thermal rating adjusted to
actual weather conditions is not only useful for economic energy transfer, it is also
beneﬁcial for the enhancement of power system reliability, including dynamic and
transient stability. In addition to the SVC system, there are other new developments
such as UPFC, PowerFormer, and Superconducting Magnetic Energy Storage that
will enable even greater control over power ﬂow through transmission lines.
The above examples demonstrate the importance of developing a dynamic trans
mission line thermal rating system with FACTS for the improvement of power system
stability, and increasing the power transfer capability of existing lines to achieve
greater economy.
Greater Utilization of Renewable Energy Sources
It has been said that the electric power system is the most complex system ever
created by human beings. Since transmission lines constitute the most important
component of a power system, the powerline ampacity system presented in this book
will continue to develop further and is expected to play a vital role in the planning
and operation of existing and future powerlines in all regions.
With public concern about global warming, there will be even tighter control
on the emission levels of electricity generated by burning fossil fuels. There is now
greater emphasis on producing electricity from renewable energy sources including
solar, wind, and hydroelectricity. Therefore, an interconnected transmission network
is essential to connect all of these sources of energy.
The world has an abundance of renewable energy sources. For example, hydro
electric reserves in Alaska and Canada in North America, and in the Amazon in
South America, are yet to be developed. Similarly, there exists a large potential for
solar energy development in the Sahara desert. When renewable energy sources are
fully developed, it is expected to meet all of our energy requirements for this planet.
As these sources of energy are generally at remote locations, Extra High Voltage
(EHV) transmission lines will be required to bring electricity from remote locations
to major metropolitan areas and industrial centers.
In the past, electricity could be transmitted efﬁciently to a distance of about
2500 km. Research from the International Conference on Large High Voltage
1306/C11/frame Page 221 Saturday, May 27, 2000 4:27 PM
222
Powerline Ampacity System: Theory, Modeling, and Applications
Networks (Cigré) now shows that it is technically and economically feasible to
transport electricity to about 7000 km by UltraHigh Voltage (UHV) lines (
IEEE
Power Engineering Review
, 1998).* This modern trend in electricity transmission is
already beginning to happen. EHV lines are presently transporting low cost elec
tricity to New York generated from renewable energy sources in Canada.** Similarly,
EHV and HVDC lines are planned to bring lowcost hydroelectricity from Siberia
to Japan.
Interconnection of a highvoltage power transmission network will also enable
the ﬂow of electricity across time zones, thereby taking advantage of the difference
in timeofday demand for electricity. As transmission lines extend beyond political
boundaries, it is hoped this will lead to greater cooperation between nations, bringing
greater peace and prosperity. With further development of the highvoltage trans
mission network to accommodate the diverse sources of energy, and interconnection
with neighboring countries, the powerline ampacity system will become more useful
for the precise control of the current ﬂowing through transmission line wires.
Development of New Transmission Line Technology
New transmission line technology is required to achieve greater energy efﬁciency
and reliability of supply by interconnection, with minimum environmental impact.
Advanced systems of communication are being developed using powerline commu
nication by ﬁber optics (J.P. Bonicel, O. Tatat, 1998), which is free from electro
magnetic disturbances. The ﬁber optic core can be easily used as a continuous wire
temperature sensor, thus eliminating the uncertainty of presentday line ratings.
Furthermore, the Internet will be used extensively for utility data communication,
and a utility transmission line network could become a part of the Internet for the
reliable transmission of data.
Modern highvoltage transmission networks will be used to carry electric energy
and data for the precise control of power ﬂow through the wires, in addition to other
business and commercial uses. Here, again, there is greater opportunity for the line
ampacity system to develop further. It will send signals through the powerline
communication network to remote power electronics (FACTS) devices for the precise
control of power ﬂow through the lines.
In the ﬁeld of overhead transmission line conductor technology, more research
is required to develop high temperature conductors using advanced aluminum alloys
(Wook, Choi, Deb, 1997) for higher transmission capacity, and compact conductors
to lower transmission losses (JeanLuc Bousquet et al., 1997). Conductors with ﬁber
optic communication technology will be required for longdistance communication
with minimum number of booster stations.
*
IEEE Power Engineering Review
, 1998, presents articles from several authors in different countries for
the development of renewable energy sources and UHV transmission to meet the energy requirements
of this planet in the 21st century.
** North American Electric Reliability (NERC) Transmission Map. NERC, 101 College Rd, Princeton,
NJ 08540.
1306/C11/frame Page 222 Saturday, May 27, 2000 4:27 PM
Summary, Future Plans and Conclusion
223
Environmental Impact
The world is paying greater attention to the environment (Alexandrov, G.N. et al.,
Russia, 1998), (Awad et al., Egypt, 1998)* by greater consideration of factors such
as EMF from powerlines, land use and ecological effects, and the emission level of
fossil fuel generators. In Chapter 6, the electric and magnetic ﬁelds of typical
transmission line conﬁgurations is calculated by the application of Maxwell’s equa
tion. It is shown that the magnetic ﬁeld of a line is minimum at ground level, and
well below acceptable limits. EMF mitigation measures are also suggested by com
pact line design, modiﬁed phase conﬁguration, and active shielding to minimize
EMF near sensitive areas.
Underground Transmission
Underground lines are recommended for aesthetic reasons in cities and scenic areas.
Since they are underground, they have no visual impact on the surrounding envi
ronment. Underground transmission lines have no electric ﬁelds on ground surfaces
since the outer layer of the cable is generally connected to ground and remains at
ground potential. Unless properly shielded, there is a magnetic ﬁeld from the cable
at ground surface.
There are certain disadvantages of underground cables.** Underground cables
take up a much greater area of land than that needed for overhead lines of the same
capacity. To place a 400,000volt line underground would typically involve digging
a trench the width of a threelane road and 1
1
/
2
meters in depth to accommodate up
to 12 separate cables. The space is needed because the highvoltage cable generates
a great deal of heat — equivalent to a onebar electric ﬁre every two meters. Whereas
the surrounding air directly cools overhead lines, underground cables have to be
well spaced to allow for natural cooling and to avoid overheating. Additional land
is needed at cable terminals where underground cables are joined to overhead lines.
The space required for cable terminations may be around 2,000 m
2
in order to place
a terminal tower (pylon) somewhat heavier in appearance than a normal suspension
tower, a small building, and other transmission equipment.
AC underground lines by cable also have limitations on transmission distance
due to the high capacitive reactance of cables. This problem is overcome by HVDC
transmission. HVDC Light is a new underground power transmission and distribution
system technology that is being developed for electricity distribution to remote areas.
In the future, the powerline ampacity system methodology developed in this book
will be extended to include underground cables and other currentcarrying transmis
sion and distribution line equipment.
* There were other papers on this subject at Cigre 1998.
** This discussion is based on a report on the World Wide Web by the National Grid Company (NGC),
UK, 1998.
1306/C11/frame Page 223 Saturday, May 27, 2000 4:27 PM
224
Powerline Ampacity System: Theory, Modeling, and Applications
Dynamic Rating of Substation Equipment
The powerline rating system methodology may be easily adapted to determine the
ampacity of other transmission and distribution line equipment. The calculation of
equipment ampacity is much easier because, unlike a transmission line, it is situated
at one location in a substation. For the calculation of equipment rating, a thermal
model of the device is required along with a knowledge of the weather conditions
at that location. Weather conditions are generally monitored in utility substations,
enabling equipment ratings to be easily determined from this data. A simple equip
ment rating model (Douglass and Edris, 1996), (Soto et al., 1998) for implementation
in the LINEAMPS program is given below:
I = Equipment ampacity at ambient temperature T
a
I
r
= Nameplate rating of equipment
Tmax = Maximum equipment operating temperature
Tmax(r) = Rated maximum temperature of equipment speciﬁed for nameplate
rating at rated ambient temperature Ta(r).
τ
= Thermal time constant of substation equipment
t = time (for steady state rating t =
∞
)
Ta = Ambient temperature (actual or forecast)
a = experimental constant
If device ratings limit line capacity it is relatively less expensive to replace a device
with a higher rating device.
Line Maintenance
A line ampacity system having forecast capability is required for the planning of
transmission line maintenance. When taking a line out of service to examine joints,
inspect conductors, clean insulators, etc., it is necessary to ensure that adjacent
transmission circuits will have sufﬁcient line ampacity to safely carry the additional
load of the line taken out of service.
Korea Electric Power Company (KEPCO) engineers used the LINEAMPS pro
gram (Choi et al., 1997) to safely reconductor existing transmission circuits by
replacing ACSR conductors with highampacity Zirconium aluminum alloy conduc
tor and Invar core. Conductor replacement was carried out one circuit at a time on
a doublecircuit line by leaving the other circuit energized. For the planning of
conductor replacement work they used LINEAMPS ratings to ensure that the ener
gized circuit had sufﬁcient capacity to carry the load of both circuits.
I I
T T
T T
t
t
r
a
r a r
= ⋅
( )
− ( )
( ) ( )
max
max
–
–
– exp –
exp –
τ
τ 1
1306/C11/frame Page 224 Saturday, May 27, 2000 4:27 PM
Summary, Future Plans and Conclusion
225
As transmission line ampacity is increased by dynamic line ratings, better tech
niques will be required to ensure that lines operate safely. Line maintenance using
infrared imaging from a helicopter is becoming routine for many utilities (Burchette,
Sam N., 1989). Similarly, devices for the robotic maintenance of conductors and
line hardware are being developed so that long transmission circuits can be inspected
rapidly without human intervention in areas of high electric and magnetic ﬁelds.
Deregulation, Independent System Operator and
Power Pool Operations
The deregulation of the electric supply business is under active consideration world
wide (Cigré Panel Session, 1998), (Cigré Regional Meeting, 1996) and is already
implemented in certain states (Barkovitch, B., Hawk, D.) of the U.S.A. and Canada
(
Electricity Today
, 1998). Figure 11.1 is a simpliﬁed diagram of an unbundled system
to illustrate the operation of a deregulated electric supply business. Independent
generation companies operate power generation (G), distribution of electricity to
consumers is done by independent distribution companies (D), and the transmission
network is operated by a transmission company (T). An independent system operator
(ISO) would ensure the reliability of the system as well as act as a clearinghouse
for the purchase and sale of electricity through a competitive bidding process. In
this scenario, a line ampacity system will provide assistance to the ISO in determin
ing transmission line capacity and pricing for the transmission of power from (G)
to (D).
According to (Cigré Panel Session, 1998), the concept of electricity deregulation
is accepted almost universally, but the rules governing a competitive electric energy
marketplace, ISO, and power pool operations are not yet well established. There is
general consensus, however, that the rules should result in optimal system operations
and encourage optimum system expansion plans. This is not an easy task and much
remains to be done in this area. The application of the LINEAMPS program in the
ideal operation of a power system as described in Chapter 8 is a contribution in this
direction.
FIGURE 11.1
Operation of an Open Electricity Market.
G
G1 G2 G3
T
D
PD1 PD2 PD3
1306/C11/frame Page 225 Saturday, May 27, 2000 4:27 PM
226
Powerline Ampacity System: Theory, Modeling, and Applications
11.4 A PLAN TO DEVELOP LINEAMPS FOR AMERICA
Discussions with various electric power companies in the U.S. have revealed that
many companies are still using a static line rating. The LINEAMPS program can
be easily adapted for implementation in these utilities with minimum cost. Near
realtime weather data is available from the National Weather Service (NWS). The
NWS has an extensive network of weather stations connecting most regions of the
U.S., and near realtime data is available on the Internet.* In addition, there are
several private weather service companies (Figure 11.6) that deliver custom weather
data. These data are adequate for LINEAMPS to determine powerline ampacity in
all regions of the U.S. Similarly, Canadian and Mexican weather data is easily
available from various agencies.
As deregulation of the electricity supply business is embraced by all states in
the U.S., Canada, and even Mexico, it will enable the development of a more
competitive business environment for the purchase and sale of electricity. In this
scenario, determining transmission capacity will become more important. Advance
knowledge of transmission capacity (hourly, daily, weekly, or longer) will enable
* US WEATHER SERVICE NOAA Internet Web Site: http://www.nws.noaa.gov/
FIGURE 11.6
A proposal for LINEAMPS for North America (including USA, Canada and
Mexico).
LINEAMPS
Weather
Data
LINEAMPS for America
Thursday December 3, 1998
Boston
Fairbanks
Honolulu
New Orleans
St Louis
Chicago
Mpls./St. Paul
Wichita
Denver
Great Falls
Phoenix
Los Angeles
San Francisco
Boise
Seattle
Dallas
Miami
Atlanta
Washington
NewYork
1
81
82
69
63
57
74
76
67
66
70
40
69
65
56
40
41
65
57
73
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
o
D
G
T
PD1
G1 G2 G3
PD2 PD3
1306/C11/frame Page 226 Saturday, May 27, 2000 4:27 PM
Summary, Future Plans and Conclusion
227
more efﬁcient electric energy futures trading by competitive bidding and spot pricing
of electricity (Schweppe, 1987). All players in the electricity market can easily use
the LINEAMPS program because it does not require any hardware connection with
the transmission line or any other power system equipment.
A plan to develop LINEAMPS for America is shown in Figure 11.6. As shown in
the ﬁgure, weather service data from different weather stations are input to LINEAMPS.
The user of the program will simply select a transmission line of interest from the stored
database of transmission lines to determine the ampacity of that line. In addition to
forecast ratings, the program will also offer realtime dynamic and transient line ratings
when realtime line current data is automatically entered to the program.
A realtime interface has already been developed by the author by establishing
a DDE* connection with a Microsoft Excel spreadsheet. As seen in Figure 11.7, the
program can be used by generation companies to determine the maximum generation
capacity they can deliver through the transmission lines; by the transmission com
panies to determine transmission line capacity and determine transmission conges
tion and transmission pricing; and by distribution companies to determine the max
imum generation capacity they can buy from lowcost generation supply sources.
As of this writing, the author has contacted the South West Power Pool; OG&E,
Oklahoma City; Houston Light & Power; Orange & Rockland Utilities, New York;
PEPCO, Washington DC; PJM; and Virginia Power, Richmond, VA, for support in
the implementation of this program. The implementation of this proposal will result
in the development of a more efﬁcient transmission network in America, enabling
greater competition for the generation, transmission, and distribution of electricity
with the objective of minimizing costs and lowering the price of electricity. In fact
there is no reason why an electricity user should not be able to shop around in the
electric energy market for the best rates, just like any other commodity.
11.5 CONCLUSION
In this chapter, the main contributions in the book are summarized and a plan is laid
out for further research and development to maximize transmission capacity.
Electricity is the prime mover of modern society and its efﬁciency depends upon
research and development of the electric power system. The electric power trans
mission network is one of the most important components of the electric power
system that is responsible for the reliable operation of the interconnected system
that comprises diverse generation sources, all connected together, to serve the load
in the most economical manner.
As the demand for electricity grows, the need for higher transmission capacity
will increase, requiring better planning, operation, and maintenance of lines. In the
past, new transmission lines were constructed to meet the needs of the electricity
supply industry. With public concern about environmental protection, the cost of
land, and other economic and demographic factors, it is becoming difﬁcult to build
new lines, and new methods are required to maximize the utilization of existing
lines. This is the object of the new Line Ampacity System developed in the book.
* LINEAMPS Dynamic Data Exchange (DDE) for Real Time Ampacity Calculations. Unpublished report.
Anjan K. Deb 1998.
1306/C11/frame Page 227 Saturday, May 27, 2000 4:27 PM
229
Bibliography
Ampère, AndréMarie, 1827. Memoir on the Mathematical Theory of Electrodynamic Phe
nomena, Uniquely Deduced from Experience. Encyclopedia Britannica.
Alexandrov et al., 1998. Overhead line designing in view of environmental constraints.
Compact overhead lines. Cigré, 1998, Paris.
Anand, V.P., Deb, Anjan K., and Mukherjee, P.K. 1984.Choice of conductors for 400kV
transmission lines. Central Board of Irrigation and Power, New Delhi, India.
Barkovitch, B.D. Hawk. 1996. Starting a New Course in California.
IEEE Spectrum
, Vol. 33,
July 1996, p26.
Bergen, Arthur R., 1986.
Power System Analysis,
Prentice Hall.
Black, W.Z. et al., 1987. Critical span analysis of overhead conductors. IEEE/PES Conference
Paper Number 87 SM 5606, San Francisco.
Black, W.Z. and W.R. Byrd, 1983. Realtime ampacity model for overhead lines. IEEE
Transactions on Power Apparatus and Systems, Vol. PAS102, No. 7.
Böhme et al., 1998. Overhead transmission lines: design aimed to reduce the permitting time.
Cigre, 1998 paper Number 22/33/3606.
Bonicel, J.P., Tatat, O., 1998. Aerial optical cables along electrical power lines.
REE
, No. 3,
March 1998.
Booch, G., 1993. Object Oriented Design with Applications, The Benjamin/Cummings Pub
lishing Co.
Borgard, L., 1999. Grid voltage support at your ﬁnger tips. Portable SMES units provide
Wisconsin Public Service Corporation voltage and VAR support.
Transmission & Dis
tribution
, Vol. 51, No. 9, October 1999.
Bousquet, JeanLuc, Loreau, B., Bechet, D., and Delomel, J.C., 1997. Un nouveau conducteur
compact homogène en alliage d’aluminium pour les lignes 400 kV.
REE
No.10, Novem
ber 1997.
Buckles, W., Hassenzahl, W.V., 2000. Superconducting Magnetic Energy Storage. IEEE Power
Engineering Review, Volume 20, Number 5, p. 16–20, May 2000.
Burchette, Sam N., 1989. Helicopter maintenance on energized EHV transmission lines.
Transmission & Distribution
, Vol. 41, No. 11, November 1989.
CEA Report, 1980. Development for an accurate model of ACSR conductor for calculating
sags at high temperaturesPart II: SagTension ProgramSTESS.
Cibulka, L., Steeley, W.J., and Deb, A.K., 1992. Le système ATLAS de PG&E d’évaluation
dynamique de la capacité thermique d’une ligne de transport. Conférence Internationale
des Grands Réseaux Électriques, Paris.
Cigré, 1999. The use of weather predictions for transmission line thermal ratings. Electra No.
186, October 1999.
Cigré, 1999. The thermal behaviour of overhead conductors. Section 4. Mathematical model
for evaluation of conductor temperature in the adiabatic state. Working Group 22.12.
_____1997. The thermal behaviour of overhead conductors. Section 3. Report prepared by
Cigré Working Group 22.12.
_____1992. Section 1 and 2, Electra # 144.
1306/bibliography/frame Page 229 Saturday, May 27, 2000 9:36 AM
230
Powerline Ampacity System: Theory, Modeling, and Applications
_____1978. Permanent elongation of conductors predictor equations and evaluation methods.
Cigré report # WG052278.
Cigré Panel Session, 1998. Impact of Deregulation on Operation and Planning of Large Power
Systems.
Cigré Regional Meeting, 1996. Power Pool Arrangements and Economic Load Dispatch. New
Delhi, India, 13–14 October 1996.
Cigré Symposium, 1991. Compacting Overhead Transmission Lines, Leningrad, USSR, 3–5
June 1991.
_____1985. High Currents in Power Systems under Normal, Emergency and Fault Conditions.
Cigré Symposium, Brussels, Belgium, 3–5 June 1985.
Cox, Earl, 2000. What’s the object? Bringing object oriented machine intelligence to web
hosted applications. PC AI, March/April 2000.
Davidson, Glen, 1969. Shortterm thermal rating for bare overhead conductors. IEEE Vol.
PAS88, Number 3.
Davis, M.W. 1977. A new thermal rating approach: the realtime thermal rating system for
strategic overhead conductor transmission lines. Part I: General description and justiﬁ
cation of the realtime thermal rating system. IEEE Transaction PAS, Vol. 96.
Deb, Anjan Kumar, 1999. Objectoriented expert powerline ampacity system. US Patent
Number 5,933,355 issued on August 3, 1999.
_____. 1998a. LINEAMPS: Line Ampacity System Comparison of Result with IEEE and
Cigré. Contribution to Cigré 98, Paris, France.
_____. 1998b. A method for the calculation of transmission line conductor temperature and
current from general heat equation. A Syntopicon Discussion on Descartes. Project IV,
Columbia Paciﬁc University, San Rafael, CA.
_____. 1998c. Application of neural network and fuzzy set theory. Response 6 CPU Course
# EG 628: Ecology. January 23, 1998.
_____. 1995a. Object oriented expert system estimates power line ampacity. IEEE Computer
Application in Power, Vol. 8, No. 3.
_____. 1995b. Objectoriented expert line ampacity system. Cigré Regional Meeting on Power
Pool Arrangements and Economical Load Despatch. New Delhi, India.
_____. 1994. Thermal rating of overhead line conductor, SIEMENS, Electrizitwirtschaft,
Germany
_____. 1993. Probabilistic design of transmission line sag and tension. International Confer
ence on Electricity Distribution System, Birmingham, UK.
_____. 1982. Etude de la capacite thermique des conducteurs de lignes aeriennes. Memoire
EDF/CNAM.
Deb, Anjan K., Singh, S.N., and Ghoshal,T.K. 1985. Higher service current in overhead lines.
Cigré Brussels Symposium on High Currents in Electric Network under Normal and
Emergency Conditions, Brussels.
Waterman, Donald A., 1986. A Guide to Expert Systems, Addison Wesley, Reading, MA.
Dorf, Richard C., 1993.
The Electrical Engineering Handbook
, CRC Press, Boca Raton.
Douglass, Dale A., 1986. Weatherdependent versus static thermal line ratings. IEEE Power
Engineering Society, Transmission and Distribution Meeting, Anaheim, California, Sep
tember 1419, 1986, Paper 86 T&D 5037. Discussion contribution of above paper by
Anjan K. Deb.
Dufﬁe, John A., Beckman, William A., 1980.
Solar Engineering of Thermal Processes
, John
Wiley & Sons, New York, 1980.
Eberhart, R.C., Dobbins, R.W., 1990.
Neural Networks PC Tools – A Practical Guide
,
Academic Press, New York.
1306/bibliography/frame Page 230 Saturday, May 27, 2000 9:36 AM
Bibliography
231
Electricity Today, 1998. An Industrial Customer’s Guide to an Open Electricity Market, Vol.
10, No. 8, September 1998.
Faraday, Michael, 1834. Experimental Researches in Electricity. Encyclopaedia Britannica.
Feak, S.D., 1997. Superconducting magnetic energy storage (SMES) utility application stud
ies. IEEE PES SM 4721 TPWRS.
Foss, Stephen D., Maraio, Robert, 1989. Dynamic line rating in the operating environment.
IEEE Transmission and Distribution Conference paper # 89 TD 4318 PWRD.
Ghanoum, E., 1983. “Probabilistic Design of Transmission Lines”, Part I, II
.
IEEE Transac
tions on Power Apparatus and Systems, Vol. PAS102, No. 9, 1983.
Giacomo, D., Nicolini, G.P., and Paoli, P., 1979. Criteria for the statistical evaluation of the
temperature in conductors, thermal rating problems. Report prepared by CIGRE working
group 22. Cigré Sienna Colloquium. Italy.
Hall, J. F., Savoullis, J., and Deb, Anjan K., 1988. Wind tunnel studies of transmission line
conductor temperature. IEEE Transactions on Power Delivery, Vol. 3, No. 4, pp. 801–812.
Hall, J.F. and Deb, Anjan K., 1988a. Economic evaluation of dynamic thermal rating by
adaptive forecasting. IEEE Transactions on Power Delivery, Vol. 3, No. 4, pp. 2048–2055.
_____. 1988b. Prediction of overhead transmission line ampacity by stochastic and determin
istic models. IEEE Transactions on Power Delivery, Vol. 3, No. 2.
Harvey, J. R., 1972. Effect of elevated temperature operation on the strength of aluminum
conductors. IEEE Transactions, Power Engineering Society.
Haykin, S., 1999.
Neural Networks  A Comprehensive Foundation
. 2nd Edition, Prentice
Hall, Englewood Cliffs, NJ.
Hingorani, N.G., Gyugyi, L., 2000. Understanding FACTSConcepts and Technology of
Flexible AC Transmission System. IEEE Press, 2000.
Hitachi, 1999. Wire catalog.
Howington, B.S. and Ramon, G.J., 1984. Dynamic thermal line rating, summary and status
of the stateoftheart technology. IEEE/PES 1984 Summer Meeting, Seattle.
House, P.D. and Tuttle, H., 1957. Current carrying capacity of ACSR. AIEE Transactions.
IEEE Power Engineering Review. 1998. Supergrids. A WinWin Solution for Sustainable
Development. Volume 18, Number 8, August 1998.
IEEE Std 738. 1993. IEEE Standard for calculating the currenttemperature relationship of
bare overhead conductors.
Jackson, R.L. and Price, C.F., 1985. Examination of continuous and thermal capacity of
overhead lines. Cigré Brussels Symposium on High Currents in Electric Network under
Normal and Emergency Conditions, Brussels.
Johannet, P., Dalle, B., 1979. Calcul des chutes de tension, des échauffements et des efforts
électrodynamiques en cas de courtcircuit. Lignes Aériennes. Techniques de l’ingénieur
121979.
KappaP.C., 1998. ObjectOriented Application Development Software, Version 2.4. Intelli
corp, Mountain View California, 1998.
Kennedy, B. 2000. Artiﬁcial intelligence improves power quality. Electrical World Vol. 214,
No. 3, May/June 2000.
Koch, B., 1999. Conductor developments help extend T&D systems.
Electrical World
, Vol.
213, No. 3, May/June 1999.
Koval, D.O., Billinton, R., 1970. Determination of transmission line ampacities by probability
and numerical methods. IEEE Transactions on Power Apparatus and Systems, Vol. 89,
No. 7.
Kronfeld, Kevin M., Tribble, Alan C., 1998. Expert System vs. Procedural Language Devel
opment. PC AI, July/August 1998.
1306/bibliography/frame Page 231 Saturday, May 27, 2000 9:36 AM
232
Powerline Ampacity System: Theory, Modeling, and Applications
Lankford, Craig B., 1989. ALCOA’s Sag and Tension Program Enhanced for PC Use. Trans
mission and Distribution Journal, Vol 41, No 11, November 1989.
Stevens, Lawrence, 1993.
Artiﬁcial Intelligence. The Search for the Perfect Machine
. Hayden
Book Company.
Lee, K.Y. et al., 1998. Adaptive Hopﬁeld Neural Networks for Economical Load Dispatch.
IEEE Transactions on Power Apparatus and Systems, Vol. 13, No. 2, May 1998.
Legrand, M.J.C., 1945. Les limites de puissancet des lignes électriques aériennes du fait de
l’échauffement. Cigré.
Mauldin, T. Paul, Steeley, William J., Deb, A.K. 1988. Dynamic thermal rating of transmission
lines independent of critical span analysis. International Conference on High Technology
in the Power Industry, Phoenix. March 1–4.
Miner, Gayle F., 1996.
Lines and Electromagnetic Fields for Engineers
. Oxford University
Press, New York.
Machowski, Jan, Bialek, Janusz W., Bumby, James R., 1998.
Power System Dynamics and
Stability
, John Wiley & Sons.
Mizuno, Yukio, Nakamura, Hisahide, Adomah, Kwabena, Naito, Katsuhiko, 1998. Assessment
of thermal deterioration of transmission line conductor by probabilistic method.
IEEE
Transactions on Power Delivery
, Vol. 13, No. 1.
Mizuno, Yukio, Adomah, Kwabena, Naito, Katsuhiko, 2000. Probabilistic Assessment of the
Reduction in Tensile Strength of an Overhead Line’s Conductor with Reference to
Climate Data. IEEE paper number PE015PRD (22000).
Morgan, Vincent T., 1991. Thermal behaviour of electrical conductors. Research Studies Press
Ltd., Somerset, UK.
Morgan, V.T., 1978. The loss of tensile strength of harddrawn conductors by annealing in
service.
IEEE Transactions, Power Engineering Society.
MPS Review Article. 1998a. New system improves IT functionality.
Modern Power System
,
UK.
_____. 1998b.Generation without transformers: Introducing Powerformer.
Modern Power
Systems
,UK.
Muhamed Aganagic, K.H. AbdulRahman, Waight, J.G., 1998. Spot pricing of capacities for
generation and transmission of reserve in an extended Poolco model
.
IEEE Transactions
on Power Systems
, Vol. 13, No. 3, August 1998.
Nasar, S.A., Trutt, F.C., 1999.
Electric Power Systems
, CRC Press, Boca Raton.
National Electric Safety Code. 1997. C21997.
Negnevitsky, Michael, 1998. A knowledgebased tutoring system for teaching fault analysis.
IEEE Transactions on Power Systems, Vol. 13, No. 1, February.
Noroozian, M., Ängquist, L., Ghandhari, M., Andersen, G., 1997. Use of UPFC for optimal
power ﬂow control. IEEE Transactions on Power Delivery,Vol. 12, No. 4.
Ozisik, M. Necati, 1985. Heat Transfer. McGraw Hill, New York.
Pfaffenberger, R.C., Patterson, J.H.,
Statistical Methods for Business and Economics
, Richard
D. Irwin, Inc.
PG&E Standard. 1978. Ampacity of Overhead Conductors.
PG&E Report. 1989. Risk Analysis of ATLAS at three PG&E locations. PG&E report 1989.
Prepared by Anjan K. Deb with contributions from William J. Steeley (PG&E) and
Benjamin L. Norris (PG&E).
Porcheron, Y., Hautefeuille, P., 1982. Lignes Aériennes.
Techniques de l’Ingénieur
, Paris.
Priestley, M.B., 1981.
Spectral Analysis and Time Series
, Volume 1, II. Academic Press, New
York.
Rashkes, V.S., Lordan, R., 1998. Magnetic Field Reduction Methods: Efﬁciency and Costs
.
IEEE Transactions on Power Delivery, Vol. 13, No. 2.
1306/bibliography/frame Page 232 Saturday, May 27, 2000 9:36 AM
Bibliography
233
Redding, J.L. 1993. A Method for Determining Probability Based Allowable Current Ratings
For BPA’s Transmission Lines. IEEE/PES 1993 Winter Meeting Conference Paper #
93WM 0778PWRD, Columbus, Ohio, January 31–February 5, 1993.
Renchon, R., Daumerie, G., 1956. Image thermique de ligne aérienne. Cigré, (30 May–9 June).
Saadat, H., 1999.
Power System Analysis
, McGraw Hill, New York.
Schmidt, N.P., 1997. Comparison between IEEE and Cigré Ampacity Standards. IEEE PE
749PWRD0061997. Discussion contribution by Anjan K. Deb.
Schweppe, Fred C. et al.,1988. Spot Pricing of Electricity. Kluwer Academic Publishers
Seppa, T.O., et al., 1998. Use of online tension monitoring for realtime thermal ratings, ice
loads, and other environmental effects. Cigré 98 paper # 22–102.
Soto, F. et al., 1998. Increasing the capacity of overhead lines in the 400kV Spanish trans
mission network: realtime thermal ratings. Cigré ‘98 paper # 22211.
Southwire, 1994. Overhead Conductor Manual. Southwire Company, Carrolton, GA.
Steeley, W.J., Norris, B.L., and Deb, A.K., 1991. Ambient temperature corrected dynamic
transmission line ratings at two PG&E locations. IEEE Transactions on Power Delivery,
Vol. 6, No. 3.
Subrahmanyam, V., 1996.
Power Electronics
, John Wiley & Sons, New York.
System for rating electric power transmission lines and equipment.1992. U.S. Patent
5,140,257.
Taylor, Odin, Smith, Peter, MacIntyre, John, Tait, John., 1998. Expert Systems for Power
Industry. Handbook of Expert Systems. Editor: Jay Liebowitz. CRC Press, Boca Raton.
Thrash, R.F., 1999. Transmission conductors – A review of the design and selection criteria.
Southwire Company.
Urbain, J.P., 1998. Intensité admissibles dans les conducteurs de lignes aériennes HTB lors
des regimes temporaires de secours. EDF CNIR 98.
Waldorf, Stephen P., Engelhardt, John S., 1998. The ﬁrst ten years of realtime ratings on
underground transmission circuits, overhead lines, switchgear and power transformers.
Electricity Today. Vol. 10, No. 2, February 1998.
Waterman, Donald A., 1987.
A Guide to Expert Systems
, Addison Wesley, Reading, MA.
Wong. K.P., Yuryevich., 1998. EvolutionaryProgrammingBased Algorithm for Environmen
tallyConstrained Economic Dispatch.
IEEE Transactions on Power Delivery
, Vol. 13,
No. 2, May 1998, p. 301–306.
Wood, Allen J., Wollenberg, Bruce
F., 1996.
Power Generation Operation and Control,
John
Wiley & Sons, New York.
Wook, M. Byong, Choi, Michael, Deb, Anjan K., 1997. Line rating system boosts economical
energy transfer. IEEE Computer Application in Power, Vol. 8, No. 3.
Yalcinov, T., Short, M.J., 1998. Neural Networks for Solving Economic Dispatch Problem
with Transmission Capacity Constraints. IEEE Transactions on Power Systems, Vol. 13,
No. 2.
1306/bibliography/frame Page 233 Saturday, May 27, 2000 9:36 AM
235
Appendix A
APPENDIX A.1
1
US CONDUCTOR SIZES: ACSR/AS ( ASTM B54983)
Construction
Rdc
Conductor
Aluminun
Steel
Area mm
2
Diameter mm
Weight kg/km RTS
20°C
Code No
Wire
dia No
Wire
dia Al St ACSR St ACSR St Al kN ohm/km
Kiwi 72 4.41 7 2.94 1097.71 47.46 44.07 8.82 3364 311 3053 215 0.0261
Bluebird 84 4.07 19 2.44 1091.75 88.87 44.76 12.20 3620 584 3036 264 0.026
Chukar 84 3.70 19 2.22 901.74 73.51 40.68 11.10 2991 483 2508 218 0.0314
Falcon 54 4.36 19 2.62 805.45 102.07 39.23 13.08 2910 671 2240 241 0.0347
Lapwing 45 4.78 7 3.18 805.43 55.67 38.20 9.55 2594 365 2229 186 0.0352
Bobolink 45 4.53 7 3.02 724.58 50.12 36.23 9.06 2334 329 2005 167 0.0391
Martin 54 4.02 19 2.41 684.36 86.63 36.16 12.05 2472 569 1903 204 0.0408
Pheasant 54 3.90 19 2.34 644.42 81.60 35.09 11.69 2328 536 1792 192 0.0433
Bittern 45 4.27 7 2.85 644.08 44.54 34.16 8.54 2074 292 1782 149 0.044
Skylark 36 4.78 1 4.78 644.35 17.90 33.42 4.78 1891 117 1775 115 0.0443
Grackle 54 3.77 19 2.27 603.76 76.58 33.97 11.33 2182 503 1679 181 0.0462
Bunting 45 4.14 7 2.76 603.99 41.74 33.08 8.27 1945 274 1672 139 0.0469
Finch 54 3.65 19 2.19 563.81 71.47 32.83 10.94 2037 469 1568 169 0.0495
Bluejay 45 3.99 7 2.66 563.79 39.00 31.96 7.99 1816 256 1560 130 0.0502
Curlew 54 3.51 7 3.51 523.14 67.81 31.62 10.54 1892 445 1448 161 0.0531
Ortolan 45 3.85 7 2.57 523.06 36.15 30.78 7.70 1685 237 1448 121 0.0541
Tanager 36 4.30 1 4.30 523.26 14.53 30.12 4.30 1536 95 1441 93 0.0546
Cardinal 54 3.38 7 3.38 483.13 62.63 30.38 10.13 1748 411 1337 149 0.0575
Rail 45 3.70 7 2.47 483.08 33.42 29.59 7.40 1556 219 1337 111 0.0586
Canary 54 3.28 7 3.28 455.77 59.08 29.51 9.84 1649 387 1261 140 0.0609
Mallard 30 4.14 19 2.48 402.66 91.88 28.95 12.41 1721 604 1117 172 0.067
Condor 54 3.08 7 3.08 402.39 52.16 27.73 9.24 1456 342 1114 124 0.069
Tern 45 3.38 7 2.25 402.61 27.82 27.01 6.75 1297 182 1114 93 0.0703
Coot 36 3.77 1 3.77 402.51 11.18 26.42 3.77 1182 73 1109 72 0.071
Drake 26 4.44 7 3.45 402.72 65.56 28.13 10.36 1544 430 1115 142 0.0682
Cuckoo 24 4.62 7 3.08 402.65 52.16 27.74 9.24 1456 342 1114 125 0.069
Redwing 30 3.92 19 2.35 362.25 82.51 27.45 11.76 1547 542 1005 154 0.0745
Starling 26 4.21 7 3.28 362.44 59.01 26.69 9.83 1390 387 1003 128 0.0758
Gannet 26 4.07 7 3.16 337.59 54.94 25.75 9.49 1294 360 934 119 0.0814
Flamingo 24 4.23 7 2.82 337.74 43.76 25.40 8.47 1222 287 935 105 0.0822
Swift 36 3.38 1 3.38 322.09 8.95 23.63 3.38 945 58 887 57 0.0887
Egret 30 3.70 19 2.22 322.05 73.51 25.89 11.10 1376 483 893 138 0.0838
Scoter 30 3.70 7 3.70 322.05 75.15 25.89 11.09 1386 493 893 143 0.0836
Grosbeak 26 3.97 7 3.09 322.17 52.43 25.16 9.27 1235 344 892 114 0.0853
Rook 24 4.14 7 2.76 322.13 41.74 24.81 8.27 1165 274 892 100 0.0862
Kingbird 16 4.78 1 4.78 286.38 17.90 24.81 8.27 910 117 793 65 0.0991
Teal 30 3.61 19 2.16 306.40 69.85 25.25 10.82 1309 459 850 131 0.0881
Wood Duck 30 3.61 7 3.61 306.40 71.49 25.25 10.82 1319 469 850 136 0.0879
Peacock 24 4.03 7 2.69 306.59 39.76 24.21 8.07 1109 261 848 95 0.0906
Eagle 30 3.46 7 3.46 281.77 65.75 24.21 10.38 1213 431 782 125 0.0956
Dove 26 3.72 7 2.89 281.83 45.93 23.54 8.67 1081 301 780 100 0.0975
1
Conductor data is calculated by program based on individual wire properties given in Appendix B. For exact values refer to the standard.
1306/appendix A/frame Page 235 Saturday, May 27, 2000 9:24 AM
236
Powerline Ampacity System: Theory, Modeling, and Applications
APPENDIX A.2
2
US CONDUCTOR SIZES: ACSR ( ASTM B23281)
Construction
Rdc
Conductor
Aluminun
Steel
Area mm
2
Diameter mm
Weight kg/km
RTS 20°C
Code No
Wire
dia No
Wire
dia Al St ACSR St ACSR St Al kN ohm/km
Kiwi 72 4.41 7 2.94 1097.71 47.46 44.07 8.82 3423 371 3053 212 0.0263
Bluebird 84 4.07 19 2.44 1091.75 88.87 44.76 12.20 3732 696 3036 257 0.0263
Chukar 84 3.70 19 2.22 901.74 73.51 40.68 11.10 3083 575 2508 212 0.0319
Falcon 54 4.36 19 2.62 805.45 102.07 39.23 13.08 3039 799 2240 233 0.0355
Lapwing 45 4.78 7 3.18 805.43 55.67 38.20 9.55 2664 435 2229 181 0.0356
Bobolink 45 4.53 7 3.02 724.58 50.12 36.23 9.06 2397 391 2005 163 0.0396
Martin 54 4.02 19 2.41 684.36 86.63 36.16 12.05 2581 678 1903 198 0.0417
Pheasant 54 3.90 19 2.34 644.42 81.60 35.09 11.69 2431 639 1792 186 0.0443
Bittern 45 4.27 7 2.85 644.08 44.54 34.16 8.54 2130 348 1782 145 0.0445
Skylark 36 4.78 1 4.78 644.35 17.90 33.42 4.78 1914 139 1775 113 0.0446
Grackle 54 3.77 19 2.27 603.76 76.58 33.97 11.33 2278 599 1679 175 0.0473
Bunting 45 4.14 7 2.76 603.99 41.74 33.08 8.27 1998 326 1672 136 0.0475
Finch 54 3.65 19 2.19 563.81 71.47 32.83 10.94 2127 559 1568 163 0.0507
Bluejay 45 3.99 7 2.66 563.79 39.00 31.96 7.99 1865 305 1560 127 0.0509
Curlew 54 3.51 7 3.51 523.14 67.81 31.62 10.54 1978 530 1448 156 0.0543
Ortolan 45 3.85 7 2.57 523.06 36.15 30.78 7.70 1730 282 1448 118 0.0548
Tanager 36 4.30 1 4.30 523.26 14.53 30.12 4.30 1554 113 1441 92 0.0549
Cardinal 54 3.38 7 3.38 483.13 62.63 30.38 10.13 1826 489 1337 144 0.0588
Rail 45 3.70 7 2.47 483.08 33.42 29.59 7.40 1598 261 1337 109 0.0594
Catbird 36 4.14 1 4.14 483.20 13.42 28.95 4.14 1435 104 1331 85 0.0594
Canary 54 3.28 7 3.28 455.77 59.08 29.51 9.84 1723 461 1261 136 0.0623
Mallard 30 4.14 19 2.48 402.66 91.88 28.95 12.41 1836 719 1117 165 0.0697
Condor 54 3.08 7 3.08 402.39 52.16 27.73 9.24 1521 407 1114 120 0.0706
Tern 45 3.38 7 2.25 402.61 27.82 27.01 6.75 1332 217 1114 91 0.0712
Drake 26 4.44 7 3.45 402.72 65.56 28.13 10.36 1627 512 1115 137 0.0702
Cuckoo 24 4.62 7 3.08 402.65 52.16 27.74 9.24 1522 407 1114 121 0.0706
Redwing 30 3.92 19 2.35 362.25 82.51 27.45 11.76 1651 646 1005 148 0.0775
Starling 26 4.21 7 3.28 362.44 59.01 26.69 9.83 1464 461 1003 123 0.0780
Gannet 26 4.07 7 3.16 337.59 54.94 25.75 9.49 1363 429 934 115 0.0838
Flamingo 24 4.23 7 2.82 337.74 43.76 25.40 8.47 1277 342 935 102 0.0841
Swift 36 3.38 1 3.38 322.09 8.95 23.63 3.38 957 70 887 57 0.0891
Egret 30 3.70 19 2.22 322.05 73.51 25.89 11.10 1469 575 893 132 0.0872
Secret 30 3.70 7 3.70 322.05 75.15 25.89 11.09 1480 587 893 137 0.0871
Grosbeak 26 3.97 7 3.09 322.17 52.43 25.16 9.27 1301 410 892 110 0.0878
Kingbird 18 4.78 1 4.78 322.17 17.90 23.88 4.78 1027 139 887 68 0.0887
Teal 30 3.61 19 2.16 306.40 69.85 25.25 10.82 1397 547 850 125 0.0916
Wood Duck 30 3.61 7 3.61 306.40 71.49 25.25 10.82 1408 558 850 130 0.0916
Peacock 24 4.03 7 2.69 306.59 39.76 24.21 8.07 1159 311 848 92 0.0927
Eagle 30 3.46 7 3.46 281.77 65.75 24.21 10.38 1295 514 782 120 0.0996
Dove 26 3.72 7 2.89 281.83 45.93 23.54 8.67 1139 359 780 96 0.1003
2
Conductor data is calculated by program based on individual wire properties given in Appendix B. For exact values refer to the standard.
1306/appendix A/frame Page 236 Saturday, May 27, 2000 9:24 AM
Appendix A
237
APPENDIX A.3
3
US CONDUCTOR SIZES: ACSR ( ASTM B23281)
Construction
Rdc
Conductor
Aluminun
Steel
Area mm
2
Diameter mm
Weight kg/km
RTS 20°C
Code No
Wire
dia No
Wire
dia Al St ACSR St ACSR St Al kN ohm/km
Osprey 18 4.47 1 4.47 281.70 15.65 22.33 4.47 898 122 776 60 0.1015
Hen 30 3.20 7 3.20 241.60 56.37 22.42 9.61 1111 440 670 103 0.1161
Hawk 26 3.44 7 2.67 241.38 39.32 21.78 8.02 975 307 668 82 0.1171
Flicker 24 3.58 7 2.39 241.60 31.34 21.49 7.16 913 245 669 73 0.1176
Pelican 18 4.14 1 4.14 241.60 13.42 20.68 4.14 770 104 665 51 0.1183
Lark 30 2.92 7 2.92 201.35 46.98 20.47 8.77 926 367 559 85 0.1393
Ibis 26 3.14 7 2.44 201.11 32.74 19.88 7.32 812 256 557 68 0.1406
Brant 24 3.27 7 2.18 201.33 26.09 19.61 6.54 761 204 557 61 0.1411
Chickadee 18 3.77 1 3.77 201.25 11.18 18.87 3.77 641 87 554 43 0.1421
Oriole 30 2.69 7 2.69 170.41 39.76 18.83 8.07 783 311 473 72 0.1646
Linnet 26 2.89 7 2.24 170.23 27.69 18.29 6.73 687 216 471 58 0.1661
Merlin 18 3.47 1 3.47 170.33 9.46 17.36 3.47 543 74 469 36 0.1678
Ostrich 26 2.73 7 2.12 151.89 24.72 17.27 6.36 613 193 420 52 0.1862
Partridge 26 2.57 7 2.00 135.12 22.02 16.30 6.01 546 172 374 46 0.2093
Waxwing 18 3.09 1 3.09 135.00 7.50 15.45 3.09 430 58 372 29 0.2118
Penguin 6 4.77 1 4.77 107.17 17.86 14.31 4.77 433 139 294 37 0.2612
Pigeon 6 4.25 1 4.25 84.95 14.16 12.74 4.25 343 110 233 30 0.3295
Quail 6 3.78 1 3.78 67.37 11.23 11.35 3.78 272 87 185 24 0.4155
Raven 6 3.37 1 3.37 53.52 8.92 10.11 3.37 216 69 147 19 0.5229
Robin 6 3.00 1 3.00 42.39 7.07 9.00 3.00 171 55 116 15 0.6603
Sparate 7 2.47 1 3.30 33.63 8.54 8.25 3.30 159 66 92 15 0.8217
Sparrow 6 2.67 1 2.67 33.63 5.60 8.02 2.67 136 44 92 12 0.8323
Swanate 7 1.96 1 2.61 21.13 5.36 6.54 2.61 100 42 58 10 1.3079
Swan 6 2.12 1 2.12 21.13 3.52 6.35 2.12 85 27 58 7 1.3247
Turkey 6 1.68 1 1.68 13.28 2.21 5.04 1.68 54 17 36 5 2.1080
Cochin 12 3.37 7 3.37 107.05 62.44 16.85 10.11 784 488 296 91 0.2488
Brahma 16 2.86 19 2.48 102.95 91.88 18.14 12.41 1004 719 285 122 0.2481
Dorking 12 3.20 7 3.20 96.64 56.37 16.01 9.61 708 440 267 82 0.2756
Dotterel 12 3.08 7 3.08 89.59 52.26 15.42 9.25 656 408 248 76 0.2972
Guinea 12 2.92 7 2.92 80.54 46.98 14.62 8.77 590 367 223 68 0.3307
Leghorn 12 2.69 7 2.69 68.16 39.76 13.45 8.07 499 311 189 58 0.3907
Minorca 12 2.44 7 2.44 56.13 32.74 12.20 7.32 411 256 155 48 0.4745
Petrel 12 2.34 7 2.34 51.54 30.06 11.69 7.02 377 235 143 44 0.5167
Grouse 8 2.54 1 4.24 40.52 14.13 9.32 4.24 221 110 112 23 0.6761
3
Conductor data is calculated by program based on individual wire properties given in Appendix B. For exact values refer to the standard.
1306/appendix A/frame Page 237 Saturday, May 27, 2000 9:24 AM
238
Powerline Ampacity System: Theory, Modeling, and Applications
APPENDIX A.4
4
BRITISH CONDUCTOR SIZES: ACSR (BS 215: Part 2: 1970)
Construction
Rdc
Conductor
Aluminun
Steel
Area mm
2
Diameter mm
Weight kg/km
RTS 20°C
Code No
Wire
dia No
Wire
dia Al St ACSR St ACSR St Al kN ohm/km
Gopher 6 2.36 1 2.36 26.23 4.37 7.08 2.36 106 34 72 9.17 1.067
Weasel 6 2.59 1 2.59 31.60 5.27 7.77 2.59 128 41 87 11.05 0.8859
Ferret 6 3.00 1 3.00 42.39 7.07 9.00 3.00 171 55 116 14.82 0.6603
Rabbit 6 3.35 1 3.35 52.86 8.81 10.05 3.35 213 69 145 18.48 0.5295
Horse 12 2.79 7 2.79 73.33 42.77 13.95 8.37 537 334 203 62.11 0.3632
Dog 6 4.72 7 1.57 104.93 13.54 13.95 8.37 396 106 290 31.97 0.2708
Wolf 30 2.59 7 2.59 157.98 36.86 18.13 7.77 726 288 438 67.05 0.1776
Dingo 18 3.35 1 3.35 158.57 8.81 16.75 3.35 505 69 437 33.66 0.1803
Lynx 30 2.79 7 2.79 183.32 42.77 19.53 8.37 843 334 509 77.81 0.153
Caracal 18 3.61 1 3.61 184.14 10.23 18.05 3.61 587 80 507 39.08 0.1553
Panther 30 3.00 7 3.00 211.95 49.46 21.00 9.00 974 386 588 89.96 0.1324
Jaguar 18 3.86 1 3.86 210.53 11.70 19.30 3.86 671 91 580 44.68 0.1358
Zebra 54 3.18 7 3.18 428.66 55.57 28.62 9.54 1620 434 1186 127.54 0.0663
Fox 6 2.79 1 2.79 36.66 6.11 8.37 2.79 148 48 100 12.82 0.7634
Mink 6 3.66 1 3.66 63.09 10.52 10.98 3.66 255 82 173 22.06 0.4436
Skunk 12 2.59 7 2.59 63.19 36.86 12.95 7.77 463 288 175 53.52 0.4214
Beaver 6 3.99 1 3.99 74.98 12.50 11.97 3.99 303 97 205 26.21 0.3733
Raccoon 6 4.09 1 4.09 78.79 13.13 12.27 4.09 318 102 216 27.54 0.3552
Otter 6 4.22 1 4.22 83.88 13.98 12.66 4.22 339 109 230 29.32 0.3337
Cat 6 4.50 1 4.50 95.38 15.90 13.50 4.50 385 124 261 33.34 0.2935
Hare 6 4.72 1 4.72 104.93 17.49 14.16 4.72 424 136 288 36.68 0.2667
Hyena 7 4.39 7 1.93 105.90 20.47 14.16 4.72 449 159 290 40.39 0.2633
Leopard 6 5.28 7 1.75 131.31 16.83 14.16 4.72 491 131 360 39.87 0.2144
Tiger 30 2.36 7 2.36 131.16 30.60 16.52 7.08 603 239 364 55.67 0.2139
Coyote 26 2.54 7 1.91 131.68 20.05 15.89 5.73 521 157 364 43.14 0.2151
Lion 30 3.18 7 3.18 238.15 55.57 22.26 9.54 1095 434 661 101.08 0.1178
Bear 30 3.30 5 7.00 256.46 192.33 22.26 9.54 2214 1502 711 267.01 0.1017
Batang 18 4.78 7 1.68 322.85 15.51 22.26 9.54 1017 121 896 65.62 0.0893
Goat 30 3.71 7 3.71 324.14 75.63 25.97 11.13 1490 591 899 137.59 0.0865
Antelope 54 2.97 7 2.97 373.92 48.47 26.73 8.91 1413 379 1035 111.25 0.076
Sheep 30 3.99 7 3.99 374.92 87.48 27.93 11.97 1723 683 1040 159.14 0.0748
Bison 54 3.00 7 3.00 381.51 49.46 27.00 9.00 1442 386 1056 113.51 0.0745
Deer 30 4.27 7 4.27 429.38 100.19 29.89 12.81 1974 783 1191 182.26 0.0653
Camel 54 3.30 5 7.00 461.63 192.33 29.89 12.81 2783 1502 1281 296.95 0.0592
Elk 30 4.50 7 4.50 476.89 111.27 31.50 13.50 2192 869 1323 202.42 0.0588
Moose 54 3.53 7 3.50 528.22 67.31 31.68 10.50 1988 526 1462 155.78 0.0538
4
Conductor data is calculated by program based on individual wire properties given in Appendix B. For exact values refer to the standard.
1306/appendix A/frame Page 238 Saturday, May 27, 2000 9:24 AM
Appendix A
239
APPENDIX A.5
5
GERMAN CONDUCTOR SIZES: ACSR (DIN 482041974)
Construction
Rdc
Conductor
Aluminun
Steel
Area mm
2
Diameter mm
Weight kg/km
RTS 20°C
Size No
Wire
dia
N
o
Wire
dia Al St ACSR St ACSR St Al kN ohm/km
16/2.5 6 1.80 1 1.80 15.26 2.54 5.40 1.80 62 20 42 5.3 1.8341
25/4 6 2.25 1 2.25 23.84 3.97 6.75 2.25 96 31 65 8.3 1.1738
35/6 6 2.70 1 2.70 34.34 5.72 8.10 2.70 139 45 94 12.0 0.8152
44/32 14 2.00 7 2.40 43.96 31.65 8.10 2.70 367 246 120 44.4 0.5891
50/8 6 3.20 1 3.20 48.23 8.04 9.60 3.20 195 63 132 16.9 0.5803
50/30 12 2.33 7 2.33 51.14 29.83 11.65 6.99 375 233 142 43.3 0.5207
70/12 26 1.85 7 1.44 69.85 11.39 11.72 4.32 282 89 193 23.8 0.4048
95/15 26 2.15 7 1.67 94.35 15.33 13.61 5.01 381 120 261 32.1 0.2997
95/55 12 3.20 7 3.20 96.46 56.27 16.00 9.60 706 440 267 81.7 0.2761
105/75 14 3.10 19 2.25 105.61 75.51 16.00 9.60 882 590 292 106.0 0.2476
120/20 26 2.44 7 1.90 121.51 19.84 15.46 5.70 491 155 336 41.4 0.2327
120/70 12 3.60 7 3.60 122.08 71.22 18.00 10.80 894 556 338 103.4 0.2181
125/30 30 2.33 7 2.33 127.85 29.83 16.31 6.99 588 233 355 54.3 0.2194
150/25 26 2.70 7 2.10 148.79 24.23 17.10 6.30 601 189 412 50.6 0.1900
170/40 30 2.70 7 2.70 171.68 40.06 18.90 8.10 789 313 476 72.9 0.1634
185/30 26 3.00 7 2.33 183.69 29.83 18.99 6.99 741 233 508 62.4 0.1539
210/35 26 3.20 7 2.49 209.00 34.07 20.27 7.47 845 266 578 71.2 0.1353
210/50 30 3.00 7 3.00 211.95 49.46 21.00 9.00 974 386 588 90.0 0.1324
230/30 24 3.50 7 2.33 230.79 29.83 20.99 6.99 872 233 639 69.3 0.1231
240/40 26 3.45 7 2.68 242.93 39.47 21.84 8.04 981 308 672 82.6 0.1164
265/35 24 3.74 7 2.49 263.53 34.07 22.43 7.47 995 266 729 79.1 0.1078
300/50 26 3.86 7 3.00 304.10 49.46 24.44 9.00 1228 386 842 103.4 0.0930
305/40 54 2.68 7 2.68 304.46 39.47 24.12 8.04 1151 308 843 90.6 0.0933
340/30 48 3.00 7 2.33 339.12 29.83 24.99 6.99 1172 233 939 84.0 0.0843
380/50 54 3.00 7 3.20 381.51 56.27 27.60 9.60 1495 440 1056 121.6 0.0743
385/35 48 3.20 7 2.49 385.84 34.07 26.67 7.47 1334 266 1068 95.8 0.0741
435/55 54 3.20 7 3.20 434.07 56.27 28.80 9.60 1641 440 1201 129.2 0.0655
450/40 48 3.45 7 2.68 448.49 39.47 28.74 8.04 1549 308 1241 111.2 0.0638
490/65 54 3.40 7 3.40 490.03 63.52 30.60 10.20 1852 496 1356 145.8 0.0580
495/35 45 3.74 7 2.49 494.11 34.07 29.91 7.47 1634 266 1367 111.2 0.0580
510/45 48 3.68 7 2.87 510.28 45.26 30.69 8.61 1766 354 1412 126.9 0.0560
550/70 54 3.60 7 3.60 549.37 71.22 32.40 10.80 2077 556 1520 163.5 0.0517
560/50 48 3.86 7 3.00 561.42 49.46 32.16 9.00 1940 386 1554 139.2 0.0509
570/40 45 4.02 7 2.68 570.87 39.47 32.16 8.04 1888 308 1580 128.6 0.0502
650/45 45 4.30 7 2.87 653.16 45.26 34.41 8.61 2161 354 1808 147.3 0.0439
680/85 54 4.00 19 2.40 678.24 85.91 36.00 12.00 2559 672 1886 196.2 0.0421
1,045/45 72 4.30 7 2.87 1045.05 45.26 43.01 8.61 3260 354 2906 201.6 0.0277
5
Conductor data is calculated by program based on individual wire properties given in Appendix B. For exact values refer to the standard.
1306/appendix A/frame Page 239 Saturday, May 27, 2000 9:24 AM
240
Powerline Ampacity System: Theory, Modeling, and Applications
APPENDIX A.6
6
FRENCH STANDARD SIZES: AAAC (NF C 34120 1976
Conductor
Construction AAAC
Area Diameter Weight RTS Rdc 20°C
Size, mm
2
Wires Wire Dia, mm mm
2
mm kg.km kN ohm/km
22 7 2.00 21.99 6 59.4 7 1.4793
34 7 2.50 34.36 8 92.9 11 0.9469
55 7 3.15 54.55 9 147.5 17 0.5966
76 19 2.25 75.55 11 204.4 23 0.4311
117 19 2.80 116.99 14 316.7 36 0.2785
148 19 3.15 148.07 16 401.0 46 0.2201
182 37 2.50 181.62 18 492.2 56 0.1796
228 37 2.80 227.83 20 617.7 70 0.1432
288 37 3.15 288.35 22 782.2 89 0.1132
366 37 3.55 366.23 25 994.1 113 0.0892
570 61 3.45 570.24 31 1550.4 176 0.0574
851 91 3.45 850.69 38 2317.8 262 0.0386
1144 91 4.00 1143.54 44 3120.1 347 0.0287
1596 127 4.00 1595.93 52 4363.9 484 0.0206
6
Conductor data is calculated by program based on individual wire properties given in Appendix B. For
exact values refer to the standard.
APPENDIX A.7
7
FRENCH STANDARD SIZES: ACSR (NF C 34120 1976)
Construction
Rdc
Conductor
Aluminun
Steel
Area mm
2
Diameter mm
Weight kg/km
RTS 20°C
Size No
Wire
dia No
Wire
dia Al St ACSR St ACSR St Al kN ohm/km
38 9 2.00 3 2.00 28.26 9.42 6.30 2.30 151 73 77 15.5 0.9669
60 12 2.00 7 2.00 37.68 21.98 10.00 6.00 276 172 104 31.91 0.7068
76 12 2.25 7 2.25 47.69 27.82 11.25 6.75 349 217 132 40.39 0.5584
116 30 2.00 7 2.00 94.20 21.98 14.00 6.00 433 172 261 39.98 0.2978
116 30 2.00 7 2.00 94.20 21.98 14.00 6.00 433 172 261 39.98 0.2978
147 30 2.25 7 2.25 119.22 27.82 15.75 6.75 548 217 331 50.61 0.2353
147 30 2.25 7 2.25 119.22 27.82 15.75 6.75 548 217 331 50.61 0.2353
182 30 2.50 7 2.50 147.19 34.34 17.50 7.50 677 268 408 62.48 0.1906
182 30 2.50 7 2.50 147.19 34.34 17.50 7.50 677 268 408 62.48 0.1906
228 30 2.80 7 2.80 184.63 43.08 19.60 8.40 849 337 512 78.37 0.1519
228 30 2.80 7 2.80 184.63 43.08 19.60 8.40 849 337 512 78.37 0.1519
288 30 3.15 7 3.15 233.67 54.52 22.05 9.45 1074 426 648 99.19 0.1201
288 30 3.15 7 3.15 233.67 54.52 22.05 9.45 1074 426 648 99.19 0.1201
297 36 2.80 19 2.25 221.56 75.51 22.45 11.25 1206 591 615 119.65 0.1247
412 32 3.60 19 2.40 325.56 85.91 26.40 12.00 1576 672 903 146.86 0.0858
612 42 2.61 19 2.65 224.59 104.74 23.69 13.25 1443 820 623 153.9 0.1208
865 66 3.72 19 3.15 716.97 147.99 30.63 15.75 3147 1158 1989 275.77 0.0393
1185 66 3.47 37 2.80 623.84 227.71 33.48 19.60 3513 1782 1731 354.37 0.0441
7
Conductor data is calculated by program based on individual wire properties given in Appendix B. For exact values refer to the standard.
1306/appendix A/frame Page 240 Saturday, May 27, 2000 9:24 AM
Appendix A
241
HIGH AMPACITY CONDUCTOR COMPARED TO ACSR (A1.8)
ACSR
High Ampacity Conductor
Conductor 240mm
2
330mm
2
410mm
2
480mm
2
240mm
2
330mm
2
410mm
2
480mm
2
Construction
Al. alloy
Invar steel
30/3.2
7/3.2
26/4.0
7/3.1
26/4.5
7/3.5
45/3.7
7/2.47
30/3.2
7/3.2
26/4.0
7/3.1
26/4.5
7/3.5
45/3.7
7/2.47
Rated tensile strength,
kgf
10,210 10,930 13,890 11,800 9,170 10,000 12,720 10,500
Weight, kg/km 1100 1320 1,673 1,699 1,122 1,330 1,687 1,611
dc resistance, ohm/km 0.1200 0.0888 0.0702 0.06994 0.1220 0.0904 0.0714 0.0607
Modulus of elasticity,
kg/mm
2
9,081 8,346 8,368 7,253 8,231
1
16,500
2
7,720
1
16,500
2
7,730
1
16,500
2
6,960
1
16,500
2
expansion coefﬁcient,
107/°C
17.97 18.97 18.95 20.84 15.3
1
3.6
2
17.0
1
3.6
2
17.0
1
3.6
2
19.8
1
3.6
2
Ampacity, A 595 720 850 917 1,173 1,425 1,675 1,810
Courtesy KEPCO
1
Normal temperature,
2
High temperature
1306/appendix A/frame Page 241 Saturday, May 27, 2000 9:24 AM
243
Appendix B
WIRE PROPERTIES (B.1)
Wire Type Copper HC Al HD Al EC
AA
6201
AA
Almelec
AA
Zr
AS
20
AS
40
HDG
Steel
Conductivity % 100 62 61 52.5 51 60 20 40 9
Tensile Strength
kg/mm
2
200 16 16 33 31.9 16 135 70 126.8
Max Temp
Steady °C
80 80 80 80 80 200 90 90 300
Max Temp
Dynamic °C
100 100 100 100 100 230 120 120 300
Max Temp
Transient °C
180 180 180 180 180 300 200 200 300
Coeff. of linear
expn. /°C
×
10
–3
16.9 23 23 23 23 23 12.6 15.5 11
Density, kg/m
3
8890 2700 2700 2700 2700 2700 6530 4800 7780
Modulus of
elasticity, kg/mm
2
11900 11900 6300 6700 6700 6700 15800 11100 20000
Speciﬁc heat,
J/(kg K)
383 897 897 909 909 909 518 630 481
Temp. coeff of
speciﬁc heat /°C
0.000335 0.00038 0.00038 0.00045 0.00045 0.00045 0.00014 0.0001
Temp. coeff of
Rdc /°C
0.00393 0.004 0.004 0.0036 0.0036 0.0036 0.0036 0.0036 0.00327
Conductivity is based on % IACS (International Annealed Copper Standard)
Cu (100% IACS) = 0.017241 ohm.mm
2
/m at 20°C
HC AL = High conductivity aluminum
HD AL = Hard drawn High aluminum (EC grade)
AA 6201 = Aluminum alloy (Al, Mg, Si) 6201 ASTM Standard
AA Almelec = Aluminum alloy (Al, Mg, Si) French Standard
AA Zr = Aluminum Zirconium alloy (Thermo resistant wire)
AS 20 = Aluminum clad steel wire (20% conductivity)
AS 40 = Aluminum clad steel wire (40% conductivity)
HDG Steel = Galvanized stee
l
1306/appendix B/frame Page 243 Saturday, May 27, 2000 9:30 AM
245
Index
A
AAC (All Aluminum Conductors), 5
AACSR (Aluminum Alloy Conductor Steel
Reinforced), 5
AC resistance of ACSR conductor, 30, 32, 51–59
ACAR (Aluminum Conductor Alloy Reinforced),
7
ACSR (Aluminum Conductor Steel Reinforced),
6, 30, 32, 51–59
ACSR conductor ampacity calculation, 30
Active shielding of transmission lines, 103
All Aluminum Alloy conductors, 5–8
Aluminum conductors, 5–8
Ambient temperature in ampacity calculations, 16
AmbientGen, 116, 139, 150, 215
Ampacity
calculations, 15–16
deﬁning problem, 16–17
effect of elevated temperature, 73
substation equipment, 224
transmission line, 219
Anode, 164
Applications, 183
economic operation, 183
electricity generation cost, 186–190
formulation of optimization problem,
184–186
long distance transmission, 198–201
protection, 201–204
stability of generators, 190
dynamic stability, 191–193
transient stability, 193–195
transmission system planning, 195–198
ARMAV (Auto Regressive Moving Average)
process, 125
Artiﬁcial neural network model, 127–131, 216
Average conductor temperature, 29
B
Back propagation algorithm, 130
BoxJenkins forecasting model, 215
Bus voltage limits, 185
C
Capacity limits, 185
Cartograph object in computer model, 152–153
Cathode, 164
Cigré
LINEAMPS comparison with standard, 66–71
long distance electricity transportation,
221–222
standard, 15
Circular ﬂux AC resistance of ACSR conductor,
53–54
Coefﬁcient of heat transfer, 37–39, 42
Coefﬁcient of linear expansion, 79
Communication by ﬁber optics, 8, 23, 222
Complex voltage series, 174
Computer modeling, 143
expert system design, 154
expert system rules, 157–158
goaloriented programming, 155
LINEAMPS expert system, 143
object model, 144
cartograph object, 152–153
conductor object, 150–152
LINEAMPS object model, 144
transmission line object, 145–148
weather station object, 148–150
program description, 159
LINEAMPS control panel, 159–161
LINEAMPS windows, 159
modeling transmission line and
environment, 159
Computer programs
LINEAMPS (see LINEAMPS program)
required for prediction of sag and tension,
77–78
Conductivity, thermal, of transmission line, 28
Conductor conﬁgurations, 100–101
Conductor elongation.
see
Elongation of
conductors
Conductor heat balance equation, 19
Conductor object in computer model, 150–152
Conductor sag.
see
Sag
Conductor temperature
differential equation, 19–21, 25, 29
frequency distribution, 87, 213
measurement, 71
probability distribution, 62, 73–75
1306/Index/frame Page 245 Saturday, May 27, 2000 8:15 PM
246
Powerline Ampacity System: Theory, Modeling, and Applications
use in ampacity calculations, 16
Conductor tension.
see
Tension
Conductor thermal modeling
differential equation of conductor
temperature, 19–21, 25, 29
dynamic ampacity, 36–44
general heat equation, 28–29
radial conductor temperature, 47–49, 61, 212
steadystate ampacity, 29–36
transient ampacity, 44–47
wind tunnel experiments, 61, 212
Conductor thermal ratings comparisons, 67–71
Conductors, 5–8, 11–12
Control center, 9
Control panel in LINEAMPS, 159–161
Cooling of transmission line, 62, 69–71
Cost savings, 186–190
Costbeneﬁt analysis, dynamic line rating, 12
Costs of transmission lines, 195–198
Creep, 81–82, 88
Current
calculation of, 104, 205
line current modeling, 215–217
Custom power, 178–179
D
DC convertors, 167
DC resistance of ACSR conductor, 32
Deregulation of electric supply business, 225
Design of expert system, 154–158
Deterministic model, time series, 24
Differential equation of conductor temperature,
19–21, 25, 29
Differential relays, 201
Diode, 164
Distributed temperature sensor system, 23–25
Distribution substation, 8–9, 223
Dynamic ampacity, 20–21, 36–44, 71
Dynamic conductor temperature, 36–44
Dynamic line rating systems
electricity generation costs, 188
operational cost savings, 12
shortterm rating, 212
Dynamic line ratings, 18, 19–21
Dynamic stability of generators, 191–193
Dynamic state operating conditions, 27
E
Earth fault relays, 201
Economic operation, 183
electricity generation cost, 186–190
formulation of optimization problem,
184–186
EHV (extra high voltage) network applications,
181
EHV transmission lines, 183, 198, 221–222
Electric ﬁeld of transmission line, 93, 108–113,
219
Electric power system overview, 3–9
Electricity generation costs, 186
cost savings by LINEAMPS, 188–190
using LINEAMPS rating, 188
using static line rating system, 186–188
Elevated temperature effects, 73
change of state equation, 78–80
elongation of conductor, permanent, 81–83
loss of strength, 83–84
transmission line sag and tension, 74–79
Elongation of conductors
caused by elevated temperature, 11, 77,
79–80, 218
geometric settlement, 81
metallurgical creep, 81, 88
predicted by probability modeling, 23
recursive estimation of, 82
EMF mitigation measures, 102–103, 223
Energy control center, 9
Energy sources, renewable, 221–222
Environmental impact, 11, 220, 223
Equations
allocation of load demand, 184
ARMAV process in realtime forecasting, 125
back propagation algorithm, 130–131
capital cost of line, 196–197
change of state equation, 63–65, 78–79, 213
circular ﬂux, 53–54
conductor heat balance equation, 19
conductor sag, 22, 63–66, 78–79
conductor temperature differential equation,
20–21, 25, 29
conductor tension, 63–66, 213, 219
current through conductor, 94
dynamic conductor temperature equation, 37
electric ﬁeld, 109–110
equipment rating model, 224
forced convection cooling in conductor, 62
Fourier series weather model, 116–117
fuzzy set theory for weather model, 132
general heat equation, 28
geometric settlement, 81
inelastic elongation, 82
Kalman ﬁlter algorithm, 126–127
Kohonen's learning algorithm, 131
line ampacity problem deﬁnition, 16–17
load current on thyristor, 165–166
longdistance transmission, 198–199
longitudinal ﬂux, 51–53
1306/Index/frame Page 246 Saturday, May 27, 2000 8:15 PM
Index
247
loss of tensile strength, 83–84
magnetic ﬁeld inside conductor, 96–97
magnetic ﬁeld of threephase powerline, 98
magnetic ﬁeld outside conductor, 95–96
Maxwell's, 94, 223
metallurgical creep, 81–82
optimum size of conductor, 197
reactive power, 170–171
realtime dynamic ampacity, 20–21
solar radiation, 137–139
speciﬁc transmission cost, 196
spectral analysis weather model, 117
stochastic model, 24
temperature of conductor, 19–20
time series stochastic and deterministic model,
75
transient conductor temperature, 44–45
transmission line, 205–207
voltage drop per meter, 54–55
Equipment ampacity, 224
Examples
ACSR conductor
dynamic ampacity, 43–44
resistance, 55–59
steadystate ampacity, 32–35
temperature, 35–36, 41–43, 45–46
electric ﬁeld, 110–113
electricity generation costs, 186
fuzzy set calculation of ampacity, 135–137
generator stability, 191, 192–193, 194
longdistance transmission, 199–201
magnetic ﬁeld, 98–100
magnetic ﬁeld with shielding, 103–108
New Zealand weather station object, 149–150
object creation in LINEAMPS, 145–148
rules used by LINEAMPS, 157–158
series compensation, 174–178
Experimental veriﬁcation of ampacity, 61
conductor temperature measurement, 71
IEEE and Cigré compared to LINEAMPS,
66–71
outdoor test span experiment, 63–66
wind tunnel experiments, 61–63
Expert line rating system, 25–26
Expert system design, 153–158
Expert systems, 9, 218
F
FACTS
applications list, 180
future R&D, 181, 220
increasing transmission capacity (see Increasing
transmission capacity, FACTS)
manufacturers, 180
recent developments, 12
semiconductor valve assembly, 167–168
Fault discrimination, 201–202
Faults, importance of clearing fast, 47
Fiberoptic cable, 8, 23, 222
Fiberoptic conductor temperature sensors, 23
Firing angle of thyristor, 170
Flexible AC Transmission System.
see
FACTS
Forecasting models, 216
Fossil fuels, 221, 222
Fourier series weather model, 116–123, 216
Fuzzy set weather model, 132–137
G
Gate, 164
GBTR (giant bipolar transistor), 167
General heat equation, 28–29
Generator capacity limits, 185
Generator rotor oscillations, 191–192, 193
Generator, Powerformer, 12
Geometric settlement, 81
Geometries of transmission lines, 100–101
Global warming, 221
Goaloriented programming, 155
GTO (gate turnoff) thyristor, 166
H
Heat effects on conductors, 11
Heat exchange during adiabatic condition, 45
Heat transfer coefﬁcient, 37–39, 42
Helicopter, 224
Highvoltage substation, 8–9
HV network applications, 180
HVDC (highvoltage DC transmission), 168–169,
223
HVDC LIGHT, 181, 223
Hydroelectric reserves, 221
I
Ice loads, 22
IEEE standard, 15, 66–71
IGBT (insulated gate bipolar transistor), 166–167
Impedance relays, 201
Increasing transmission capacity, 163
FACTS, 168
custom power, 178–179
future R&D, 180–181
HVDC, 168–181
list of applications, 180
manufacturers, 180
1306/Index/frame Page 247 Saturday, May 27, 2000 8:15 PM
248
Powerline Ampacity System: Theory, Modeling, and Applications
series compensation, 173–178
SMES, 179–180
STATCOM, 171–173
static VAR compensator, 169–171
UPFC, 178
New power semiconductor devices, 163
FACTS semiconductor valve, 167–168
GBTR, 167
GTO, 166
IGBT, 167
MCT, 167
MOSFET, 166
Thyristor, 16, 164–166
Independent system operators, 225
Infrared imaging, 225
Inheritance of objects, 144, 153
Integrated line ampacity system, 25–26
Invar alloy reinforced steel wires, 8, 224
K
Kalman ﬁlter algorithm, 126–127, 216
KappaPC, 144, 155
Kohonen's learning algorithm, 131
L
Learning in neural network, 129–130
Line ampacity problem deﬁnition, 16–17
Line maintenance, 224–225
Line rating methods, 15
distributed temperature sensor system, 23–25
dynamic line ratings, 17–18, 20–21
expert line rating system, 25
historical background, 15–16
integrated line ampacity system, 25–26
line ampacity problem, 16–17
objectoriented modeling, 25
online temperature monitoring system, 19–21
online tension monitoring system, 21–22
rating systems (see Rating transmission lines)
sag monitoring system, 22–23
static line ratings, 17
weatherdependent systems, 18–19
Line security analysis, 74
LINEAMPS program
computer modeling (see Computer modeling)
conductor temperature
dynamic state, 41
steadystate, 32
transient state, 47
deﬁnition, 143–144
evaluated by power companies, 211
forecasting powerline ampacity, 18–19, 124
generating weather data, 24
IEEE and Cigré comparison, 66–71
outdoor test span comparison, 63–65
overview, 25
plan to develop in America, 226–227
sag calculation, 63–66
summary, 209–211
temperature measurement of transmission
line, 71
tension calculation, 63–66
weather modeling, 116
wind tunnel data comparison, 62
LINEAMPS ratings, 188–190
Long distance transmission, 198–201
Longitudinal ﬂux, AC resistance of ACSR
conductor, 51–53
Loss of strength
calculations for AAC Bluebell, 88
caused by elevated temperature, 11, 75, 77,
79, 219
determined experimentally, 65
percentile method to calculate, 83
predicted by probability modeling, 23
recursive estimation of, 83–84
M
Magnetic ﬁeld impact on environment, 11, 219,
223
Magnetic ﬁeld of transmission line, 93
conductor magnetic ﬁeld, 94–95
ﬁeld inside conductor, 96–97
ﬁeld outside conductor, 95–96, 97
different geometry, 100–102
EMF mitigation, 102–108
threephase powerline magnetic ﬁeld, 98–100
Manufacturers of FACTS devices, 180
Mathcad Solver, 58, 111
MCT (MOScontrolled thyristor), 167
Metallurgical creep, 81–82, 88
Meteorological conditions.
see
Weather
conditions
Methods in computer program, 144
MOSFET, 166
N
National Weather Service
data supplied, 116
forecasts used in LINEAMPS program,
24–25, 148, 149–150, 226
Neural network weather model, 127–131, 216
1306/Index/frame Page 248 Saturday, May 27, 2000 8:15 PM
Index
249
O
Objectoriented modeling, 25–26, 217
Objects in computer program, 144
Ofﬂine line ratings, 18, 22
Online line ratings, 18
Online temperature monitoring system, 19–21
Online tension monitoring system, 21–22
Operating condition of transmission line, 27
OPGW (Optical Ground Wire) conductor, 8, 23
Oscillations of generator rotor, 191–192, 193
Outdoor test span, 63–66
Overcurrent relays, 201
Overhead conductors, thermal ratings, 67–71
Overhead transmission lines, 4–8, 74
Overload protection, 201–204
P
Passive shielding of transmission lines, 102
Passive shunt capacitors and reactors, 169
Power balance equation, 184
Power companies using LINEAMPS, 211
Power Donut temperature sensors, 19
Power pool operations, 225
Powerformer generator, new technology, 12, 221
Powerline communication system, 8, 23
Pretensioning, 81
Probabilistic design of overhead lines, 74
Probabilistic rating methods, 23–25
Probability distribution of conductor temperature,
62, 73–75
Probability modeling of conductor temperature,
23
Program description of LINEAMPS, 159–161
Protection, 201–204
R
Radial conductor temperature, 47–49, 61, 212
Rating transmission lines
dynamic, 12, 18, 188–189, 212
LINEAMPS, 188–190
static, 17, 186–188
transient, 212
Realtime weather forecasting, 124–127, 213
Recursive estimation, 25, 216
Renewable energy sources, 221–222
Robotic maintenance, 225
Rotor angle, 191
Rotor oscillations, 191–192, 193
Rules used in LINEAMPS, 157–158
S
Sag
calculation comparisons, 89–92
calculation of, 23, 63–66, 78–79, 219
caused by elevated temperature, 74–78, 83,
219
effect on electric and magnetic ﬁelds, 93, 98,
108
monitoring of, 18, 22
problem, 75
Security of transmission line, 74
Semiconductors, 163–168
Series compensation, 173–178
Shielding of transmission lines, 102–103
Shunt and series capacitors, 221
SMES (Superconducting Magnetic Energy
Storage), 179–180, 221
Solar energy development, 221
Solar radiation, 16, 137–139
SolarGen, 139, 149, 215
SSAC (Steel Supported Aluminum Conductor), 8
SSSC (Static Synchronous Series Compensator),
174
Stability of generators, 190
dynamic stability, 191–193
transient stability, 193–195
STATCOM (Static Synchronous Compensator),
171–173
Static line ratings, 17, 186–188
Steadystate ampacity, 29–36, 66–71
Steadystate line rating, 212
Steadystate operating conditions, 27
Steadystate stability of generators, 190–191
Stochastic model, time series, 23–25
Strength, loss of.
see
Loss of strength
Stress/strain relationship, 80
Substation, electric, 8–9, 224
Supervised learning in neural network, 129
SVC (static VAR compensator), 169–171, 221
SVC stations, 167
T
TCR (thyristorcontrolled reactor), 169–171
TCSC (thyristorcontrolled series capacitor), 173
Temperature of conductor.
see
Conductor
temperature; Conductor thermal modeling
Temperature of transmission line
effects of elevated (see Elevated temperature
effects)
measured by temperature sensors, 71
Temperature sensors
on conductors, 18, 19
1306/Index/frame Page 249 Saturday, May 27, 2000 8:15 PM
250
Powerline Ampacity System: Theory, Modeling, and Applications
on transmission lines, 71
Tensile strength loss.
see
Loss of strength
Tension
and elevated temperature, 74–79, 219
calculation comparisons, 89–92
calculation of, 63–66
in stress/strain relationship, 80
problem, 75
realtime rating, 213
Tension monitors on conductors, 18, 21–22
Theory of transmission line ampacity, 27
differential equation of conductor
temperature, 29
dynamic ampacity, 36–44
general heat equation, 28–29
operating conditions, 27
radial conductor temperature, 47–49
steadystate ampacity, 29–36
transient ampacity, 44–47
Thermal conductivity of transmission line
conductor, 28
Thermal modeling.
see
Conductor thermal
modeling
Thermal ratings of overhead conductors, 67–71
Threedimensional conductor thermal model, 25
Threephase powerline magnetic ﬁeld, 98–100
Thyristor, 164–166, 167
Time series analysis, 25
Timeseries stochastic and deterministic models,
74–76
Towers, 6–7
Transient ampacity, 44–47
Transient line rating, 212
Transient stability of generators, 193–195
Transient state operating conditions, 27
Transmission capacity
enhancement, 12
factors, 11
new methods of increasing (see Increasing
transmission capacity)
Transmission grid of North America, 3–4
Transmission line
ampacity theory (see Theory of transmission
line ampacity)
capacity limits (line ratings), 185
components, 143
conductor temperature measurement, 71
electric ﬁeld (see Electric ﬁeld of transmission
line)
magnetic ﬁeld (see Magnetic ﬁeld of
transmission line)
new technology, 222
object in computer model, 145–148
overhead components, 4–8
rating systems (see Rating transmission lines)
sag (see Sag)
security, 74
tension (see Tension)
Transmission substation, 8–9, 224
Transmission system planning, 195–198
Transmission tap limits, 185–186
TSC (thyristorswitched capacitor), 169–171
U
UHV (ultrahigh voltage) lines, 222
Underfrequency relays, 201
Underground transmission, 223
Unsupervised learning in neural network,
129–130
UPFC (Uniﬁed Power Flow Controller), 12, 178,
221
V
Voltage
and electric ﬁeld, 108
calculation of, 104, 205
listed in computer model, 145
longdistance transmission, 201
substation, highvoltage, 8–9
Voltage drop, AC resistance of ACSR conductor,
54–55
Voltage relays, 201
W
Weather conditions
affect probability distribution of conductor
temperature, 74–75
affecting ampacity, 16, 115
affecting overhead transmission lines, 67–68
and dynamic line rating system, 188
and static line rating system, 186–187
realtime forecasting, 124–126, 213
required for calculation of equipment rating,
224
Weather dependent line rating systems, 18–19
Weather modeling, 115, 215–217
Fourier series model, 116–123, 216
fuzzy set model, 132–137
neural network model, 127–131, 216
realtime forecasting, 124–126
solar radiation model, 137–139
Weather station object in computer model,
148–150
Wind speed, 16
Wind tunnel experiments, 61–63, 212
1306/Index/frame Page 250 Saturday, May 27, 2000 8:15 PM
Index
251
WindGen, 116, 141, 149, 215
Wire conductors, 4–8
Y
Young's modulus of elasticity, 64, 79, 80
Z
Zirconium aluminum alloy conductors, 7–8, 224
1306/Index/frame Page 251 Saturday, May 27, 2000 8:15 PM
Power Line Ampacity System
Theory, Modeling, and Applications
Anjan K. Deb, Ph.D., P.E.
Electrotech Consultant
CRC Press Boca Raton London New York Washington, D.C.
Library of Congress CataloginginPublication Data
Deb, Anjan K. Powerline ampacity system : theory, modeling, and applications / Anjan K. Deb. p. cm. Includes bibliographical references and index. ISBN 0849313066 1. Powerline ampacity—Mathematical models. 2. Electric cables—Evaluation. 3. Electric lines—Evaluation—Mathematical models. 4. Electric power systems—Load dispatching. 5. Electric currents—Measurement—Mathematics. 6. Amperes. I. Title. TK3307 .D35 2000 621.319—dc21 00036093 CIP
This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microﬁlming, and recording, or by any information storage or retrieval system, without prior permission in writing from the publisher. The consent of CRC Press LLC does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Speciﬁc permission must be obtained in writing from CRC Press LLC for such copying. Direct all inquiries to CRC Press LLC, 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identiﬁcation and explanation, without intent to infringe.
© 2000 by CRC Press LLC No claim to original U.S. Government works International Standard Book Number 0849313066 Library of Congress Card Number 00036093 Printed in the United States of America 1 2 3 4 5 6 7 8 9 0 Printed on acidfree paper
Dedication
This book is dedicated to my wife Meeta and my family and friends
While working in Algeria. objectoriented modeling. Dr. and is a principal in ELECTROTECH Consultant. He has authored or coauthored more than 20 research publications in the area of transmission line conductor thermal ratings. where he worked on the research and development of conductors and line hardware. EDF is the national electric power supply company of France. Dr. and neural networks. Paris. after completing all courses and preparing a doctoral dissertation on the subject of transmission line ampacity. he joined a doctoral degree program at the Columbia Paciﬁc University.com. He can be reached by email at akdeb@aol. He went to Algiers to work for the National Electrical and Electronics Company. He also maintains the LINEAMPS website for interaction with program users. the LINEAMPS software developed by the author is used in several countries. patent for the invention of the LINEAMPS™ program. and began working for Paciﬁc Gas & Electric (PG&E). Dr.D. He is presently working on projects related to intelligent software development by the application of artiﬁcial intelligence. expert systems. Dr. and on the Web at http://www.About the Author Dr. where he received the equivalent of a master’s degree in electrical engineering. and electric power systems. San Francisco. While working at PG&E. substation automation. and earned a Ph.lineamps. Deb performed theoretical and experimental research on the heating of conductors and transmission line ampacity. In addition to solving transmission line electrical and mechanical problems.S. energy management and developing intelligent computer applications for power. France. As stated in this book. At EDF.com. where he developed and successfully implemented a realtime linerating system for PG&E. Dr. offering seminars and custom software solutions for increasing transmission line capacity by dynamic thermal ratings. He has 20 years’ experience in highvoltage power transmission lines. where he designed and manufactured highvoltage substations.S. Deb came to the U. a transmission line software and consulting company that he started in 1990. Deb works as a consultant for electric power companies in all regions of the world. Deb is interested in adaptive forecasting. and for reporting new developments. He then received training at the Electricité de France (EDF) Research Center at Paris. he received a French government scholarship to study Electrotechnique at the Conservatoire National des Arts et Métiers. Anjan K. fuzzy sets. Deb is a registered professional electrical engineer in the state of California. . After receiving a bachelor’s degree in electrical engineering from MACT India. and has been awarded a U. Deb began his transmission line engineering career at EMC India.
.
electricity regulators. operations. and electric energy policy makers who want to get a ﬁrm grip on technical issues concerning the movement of electric energy from one location to another. environmental concerns. and uptodate knowledge of existing and future transmission line technologies. new lines will be required for more efﬁcient electricity transmission and distribution. powerline EMF developed from Maxwell’s equations and Ampere’s law. To bring this theory into practice I have developed an objectmodel of the line ampacity system and implemented a declarative style of programming by rules. where I initiated research on the heating of conductors. This book has been developed from more than 20 years of my experience in working with various electric power companies in Asia. Africa. These will complement the existing large number of excellent textbooks on electric power systems. San Ramon. offered me an excellent environment for research . with the help of material presented in this book. stability analysis. I am particularly grateful to Electricité de France. Paris. While there are several books devoted to transmission line voltage. environmental impact. Hopefully. userfriendly windows program with a good graphical user interface that can be used easily in all geographic regions. power ﬂow with variable line ratings. Adding new lines will become more difﬁcult as public awareness of environmental protection and land use increases. California. A unique contribution of this book is the development of a complete theory for the calculation of transmission line ampacity suitable for steady state operation and dynamic and transient conditions. the transmission line engineer will make better decisions regarding the choice of conductors. there are few books that focus on line currents. and cost optimization.Preface It is my great pleasure to present a book on transmission line ampacity. and North America. The power system may have to be operated more closely to generation stability limits for better utilization of existing facilities. construction. Academicians and students will ﬁnd material covering theoretical concepts of conductor thermal modeling. The end product is a stateoftheart. To keep pace with increasing electric energy usage in the next millennium. This book is primarily for practicing electric power company engineers and consultants who are responsible for the planning. and ﬂexible AC transmission. It is also a useful source of reference for various government authorities. Paciﬁc Gas & Electric Company. the analysis of conductor ampacity. As we enter the 21st century we shall have to develop new methods to maximize the capacity of existing transmission and distribution facilities. and the environmental impact of high currents. computer modeling of line ampacity with power system applications. and costing of overhead powerlines. power electronic devices. for the various interactions with the members of the Departement Etudes et Recherche since 1978. Europe. system operation. design.
Institute de Recherche Hydro Quebec. for going through the initial manuscript and kindly pointing out errors and omissions. University of Paris VI. and a vector is denoted by an upper arrow like H . for example. Dean. California. Anjan K. readers may visit LINEAMPS on the Web at http://www. I thank Ms.* The LINEAMPS computer program described in this book is a commercial software program available from: ELECTROTECH Consultant 4221 Minstrell Lane Fairfax. but not the least. and Korea Electric Power Company for their valuable feedback and support which has enabled me to enhance the computer program. Rialland for their lectures and teachings on Electrotechnique at the Conservatoire National des Arts et Métiers. Mentor. As in most modern electrical engineering books. Bonneﬁlle. including TransPower. * I have followed the same notation used by Gayle F. France. John Heldt. Last. New Zealand. . and Professor J. Canada. Complex numbers are denoted by an upperscore.com. F. and Dr. 1996. during the development of the LINEAMPS program is gratefully acknowledged. USA (703) 3228345 For additional details of the program and to obtain new information concerning recent developments on high currents and transmission line ampacity. Hydro Quebec. Oxford University Press. Thanks are due to Dr. Deb.and development when I worked as a consultant on transmission line dynamic thermal ratings. New York. I am grateful to Professor R. for their guidance while I prepared a doctoral dissertation. Miner in Lines and Electromagnetic Fields for Engineers.D. Peter Pick. Thanks are due to several users of the LINEAMPS program. Paris. California. and for their continued encouragement to write this book. of Columbia Paciﬁc University. Genevieve Gauthier. VA 22033. The kind technical support offered by Mr.lineamps. SI (System International) units are used consistently throughout. Graham of Intellicorp. Ph. a complex currentrI∠θ = I·ejθ is represented by I . Canada. Research Engineer.
...............2 Introducing the Powerline Ampacity System ....................2.............................................................2..5 Transient Ampacity...................19 2.....3.........................18 2..25 2.....................1 1.................................................4 1..............................................2 Line Rating Methods .................1 1...7 Chapter Summary ...............8 ObjectOriented Modeling and Expert Line Rating System........27 3.........28 3.......3 1..................4 Dynamic Ampacity .....6 Dynamic Line Rating Cost–Beneﬁt Analysis ..2..........................................1 Deﬁning the Line Ampacity Problem ......15 2.............4 Factors Affecting Transmission Capacity and Remedial Measures......................................12 1......2 IEEE and Cigré Standards............1 Introduction.................15 2...........3 1.................................................2............5 Online Tension Monitoring System .....2.............1 General Heat Equation ...............3 HighVoltage Substation........3 Utility Practice ..................................................22 2..............4 Energy Control Center.....................................44 ..2..........................................1...................2................................3 WeatherDependent Systems ..........................................................................1..16 2.......................................................................16 2..........1 Organization of Book and Chapter Description ...........12 1..................2............5 New Developments For Transmission Capacity Enhancement .......................................2 Overhead Transmission Line.........................2......................................15 2.2.........................................................................................1 Historical Backround ...3 Chapter Summary ..............................................7 Distributed Temperature Sensor System ..................11 1.......................................2 Static and Dynamic Line Ratings .......................................2.................................................................26 Chapter 3 Theory of Transmission Line Ampacity ...23 2........9 1.................36 3...................15 2............29 3...........2 Differential Equation of Conductor Temperature ..................27 3.................................3 1..................................................8 1.......1.15 2......2 Conductor Thermal Modeling .1 Transmission Grid ..........................3 Electric Power System Overview...................................................6 SagMonitoring System................1 Early Works on Conductor Thermal Rating .....28 3....2.......3...3.................2...................................21 2........................................17 2........12 Chapter 2 Line Rating Methods ..................3.........................4 Online Temperature Monitoring System...Contents Chapter 1 Introduction .................................................29 3..3 SteadyState Ampacity ................................
......................2 Transmission Line Sag and Tension — A Probabilistic Approach ...................................................73 5...........................84 5...........................3 Experiment in Outdoor Test Span...............73 5.....................................................108 6........................................................1...............2 Conductor Stress/Strain Relationship...1 Geometric Settlement ..4.......................51 Chapter 4 Experimental Veriﬁcation of Transmission Line Ampacity................................................3 Change of State Equation..2 Metallurgical Creep ...................81 5......................................................75 5...2........................................................2...47 3........94 6..71 4....4............74 5...2 Recursive Estimation of Loss of Strength ......61 4.....................................................................3....79 5.....................4 EMF Mitigation ....................................................................2.....77 5.............4.....................66 4.........................5 Loss of Strength................................3 The Magnetic Field of Different Transmission Line Geometry ....78 5............74 5.......................1 SteadyState Ampacity ........3 Recursive Estimation of Permanent Elongation ...........2 Wind Tunnel Experiments .............................3 Chapter Summary .............1 The Magnetic Field of a Conductor......................72 Chapter 5 Elevated Temperature Effects.........................61 4............................1 Introduction...................4 Permanent Elongation of Conductor............2..........................93 6.......................93 6......................................82 5.......................................................49 Appendix 3 AC Resistance of ACSR ........81 5......100 6.........................2 The Magnetic Field of a ThreePhase Powerline ....................................................102 6......................................................6 Chapter Summary ....................80 5.....................2........4.........4 Chapter Summary ......................80 5............2..98 6.................................................................6 Chapter Summary ............2 Transmission Line Magnetic Field............2 Dynamic Ampacity .........................75 5......63 4.........................................................................................1 Percentile Method.............5 Measurement of Transmission Line Conductor Temperature .......1 Introduction................5.......93 6.........................................71 4.........2..................................4............1 Existing Programs............1 Results...........................................83 5....3 Transmission Line Electric Field ...........................................................................61 4.......................................................5......87 Chapter 6 Transmission Line Electric and Magnetic Fields ..............1 Introduction..............................................3...........6 Radial Conductor Temperature........................4 Comparison of LINEAMPS with IEEE and Cigré...........................................................................................................2 Methodology .......................113 ...................................................1 The Transmission Line Sag–Tension Problem .........2.....83 5.......................3..66 4.84 Appendix 5 Sag and Tension Calculations ...............3 Computer Programs...................................
..............................................159 8.......3 LINEAMPS Control Panel ..............................................................................................163 9............................2 Electricity Generation Cost Saving in Interconnected Transmission Network...........4 Conductor Object..................................6 Solar Radiation Model..............................1 Introduction ....4 Program Description....................143 8....................................115 7.........................................................159 8................3......144 8............168 9............................................198 ......................1 From Theory to Practice..................................154 8.......1 Introduction..145 8..............................................2 Advancement in Power Semiconductor Devices ............3 Stability .....139 Chapter 8 Computer Modeling ...............................................................................1 Introduction......1....1 Introduction...............143 8................................137 7...........................................186 10...............................................116 7...............................................1 LINEAMPS Windows ...........2 Modeling Transmission Line and Environment........................................2 The LINEAMPS Expert System ..................4 Chapter Summary ..................4 Artiﬁcial Neural Network Model ...................................................................190 10..........................................................152 8...........2 Expert System Rules ...............................132 7....................................2..........................2...................183 10.....................2 Transmission Line Object..........................................................................115 7......195 10....................................................................4........................................4.127 7........................................7 Chapter Summary .............2....................2 Object Model of Transmission Line Ampacity System.......3 Expert System Design ....144 8.....................2 Economic Operation ..184 10............148 8............................4 Transmission Planning...........................2.............143 8................1 Formulation of the Optimization Problem...3............159 8............5 Chapter Summary ..........163 9..................................5 Cartograph Object..........183 10...................157 8.2....................................................................................................193 10..........................................................191 10....1 Dynamic Stability .............................2 Fourier Series Model ........................................................................................................................................159 8........................3 Weather Station Object..............162 Chapter 9 New Methods of Increasing Transmission Capacity ................5 LongDistance Transmission .........................................2.........................................................Chapter 7 Weather Modeling for Forecasting Transmission Line Ampacity ..............................................3...1..........2...........1 GoalOriented Programming ..................183 10......................................................................155 8....181 Chapter 10 Applications..........143 8..............123 7.................................................................150 8......3 RealTime Forecasting....1 LINEAMPS Object Model.............................................3 Flexible AC Transmission ......163 9.....................4......2 Transient Stability.......3................5 Modeling by Fuzzy Sets...................................
....................................10.................................................219 A Plan to Develop LINEAMPS for America .............................3 11............................................................................235 Appendix B: Wire Properties ..........212 Suggestions for Future Work.......................5 Summary.......................................................6 Protection .......................................................205 Chapter 11 11................................................................4 11.................245 ............204 Appendix 10.......225 Conclusion ..................................................................................201 10........................................ and Conclusion...........................................................................................................................................................................................2 11............ Future Plans.......226 Bibliography ..................................................209 Main Contributions............243 Index........1 11....229 Appendices A1–A8: Conductor Data........................................................1 Transmission Line Equations ..............................................209 Summary .................7 Chapter Summary .......................
The effects of elevated temperature operation on transmission line conductors are presented in Chapter 5. The theory of transmission line electric and magnetic ﬁelds is developed from Maxwell's equations in Chapter 6. The method of calculation of the loss of strength and inelastic elongation of conductors by a recursive procedure that utilizes probability distribution of conductor temperature in service is described. wind direction. For this reason Chapter 7 is devoted entirely to weather modeling. and the world are also given. The conductor thermal models in the steadystate and dynamic and transient states are validated by comparing results with the IEEE standard and Cigré method. are presented in this chapter. and then solutions are presented for steadystate. A complete theory of transmission line ampacity is presented in Chapter 3. and energy control centers. but increases with conductor temperature due to lowering of the conductor to ground clearance by sag. This aspect of transmission line ampacity is signiﬁcant because there is little previous work carried out in this direction. wind speed. When higher ampacity is allowed on the line. Experimentally derived models of loss of tensile strength of conductors. and transient operating conditions. A threedimensional conductor thermal model is ﬁrst developed. substations. Even though there is no evidence of environmental impact by EMF due to increased transmission line currents. Methods of reducing the level of EMF radiated from transmission lines by active and passive shielding are presented in this chapter. A method of generating the probability distribution of conductor temperature in service from time series stochastic and deterministic models is given.1 Introduction 1. Chapter 2 presents the different methods of transmission line rating. Experimental work related to transmission line ampacity that was conducted in different research laboratories is described in Chapter 4. including both online and offline methods. it increases the magnetic ﬁeld radiated from the transmission line. Environmental factors inﬂuence transmission capacity signiﬁcantly. and solar radiation.S. The electric ﬁeld from the transmission line does not change with line ampacity. Results are also compared to laboratory experiments and measurements from actual transmission lines. Statistical modeling of weather 1 .1 ORGANIZATION OF BOOK AND CHAPTER DESCRIPTION Chapter 1 gives a broad overview of the electric power system including transmission lines. Data for electricity production in the U. as well as permanent elongation of conductors due to creep. measures are suggested to lower magnetic ﬁelds from transmission lines. The meteorological variables that are most important to powerline capacity are ambient temperature. dynamic.
An overview of new technologies that are being developed to increase transmission capacity up to the thermal limit by overcoming the aforementioned limitations is presented in this chapter. Superconducting Magnetic Energy Storage (SMES). EDF/CERT Directives Lignes Aeriennes 1996. adequate stability can be maintained by electrical control of generation systems as well as by fast control of active and reactive power supply to the system. The most important factors are transmission distance. It was developed by objectoriented modeling and expert rules of powerline ampacity. The capacity of electric powerlines to transport electric energy from one point to another. Fourier series analysis. Ampacity of overhead line conductors. Chapter 9 discusses new methods of increasing line ampacity. weather stations. 1998). Until recently. and generator stability. Chapter 10 presents applications of the new powerline ampacity system to clearly show its beneﬁts. South Korea (Wook et al. Analytical expressions for the calculation of solar radiation on a transmission line conductor are also presented to complete the chapter on weather modeling. Modeling. Models are developed for realtime prediction of weather variables from measurements as well as by weather pattern recognition. 1997) . Expert system rules are developed to enable an intelligent powerline ampacity system to check user input and explain error messages like a true expert. and Applications variables based on time series analysis. Methods of objectoriented modeling of transmission lines. and powerline conductors are described with examples from electric power companies in the different regions of the world. ** REE Spain (Soto et al. KEPCO. there may be signiﬁcant voltage drop that may be compensated for by controlling reactive power and/or boosting voltage levels by transformer action. electric power companies* have assumed that the maximum capacity of a powerline is constant by assuming conservative weather conditions. so they followed a static line rating system. voltage level. In a competitive power supply business environment.. Now certain electric power companies** are adopting a system of line rating that is variable and dynamic depending upon actual weather conditions. PG&E Engineering Standard. or the ampacity. and distributed generation systems. This stateoftheart software package is an expert system for the rating of powerlines. In many cases. Chapter 8 describes computer modeling of the LINEAMPS expert system.2 Powerline Ampacity System: Theory. * Regles de calcul electrique.. Cigre. When energy is transported over a long distance. is discussed. in most cases the transport capacity of overhead powerlines is limited only by the thermal rating of the powerline conductor. The complete system of rating overhead powerlines is implemented in a computer program called LINEAMPS. and neural networks are presented with examples using real data collected from the National Weather Service. of existing and future overhead powerlines as functions of present and forecast weather conditions. Therefore. The object of the program is to maximize the currentcarrying capacity. These new technologies include the application of modern power electronics devices that are known as FACTS (Flexible AC Transmission System). it is necessary to optimize the ampacity of overhead power transmission lines to enable the most economic power system operation on an hourbyhour basis. that depends upon several factors.
One of the methods used to increase line capacity is dynamic thermal rating. The chapter concludes with a discussion of increased competition in the electric power supply industry in a power pool system of operations. and considerable cost savings are shown by the deferment of capital investment required for the construction of new lines. The different methods of increasing line ampacity by FACTS are described. transmission. and the impact of higher transmission line ampacity on electric and magnetic ﬁelds is analyzed with numerical examples.1 TRANSMISSION GRID The electric power system is comprised of an interconnected transmission grid that is used to connect diverse generation sources for the distribution of electricity to load centers in the most economical manner. The economic beneﬁts of a dynamic line rating system are demonstrated by giving an example of an interconnected transmission network having a diverse mix of electricity generation sources. Chapter 11 gives a summary of main contributions made in this book. and the important role of the powerline ampacity system presented in this book. While the location and construction of a generation facility is relatively easy. In this chapter. and by enabling greater utilization of lowcost energy sources. higher line currents are possible if there are favorable meteorological conditions. It provides a discussion on deregulation and how the line ampacity system facilitates greater competition in the electric supply business. it is becoming increasingly difﬁcult to construct new lines. and the temperature of a conductor varies as a function of line current and meteorological conditions. for the same value of maximum conductor temperature. As a result. and a methodology for implementation in a computer program. The object of this book is to develop a complete system of rating overhead powerlines by presenting theory.1 shows the transmission grid . and distribution capacities to match demand. The development of a computer program by objectoriented modeling and expert system rules is also described in detail.3.Introduction 3 The thermal rating of a transmission line depends upon the maximum design temperature of the line. algorithms. presents future plans and new transmission and distribution technologies.3 ELECTRIC POWER SYSTEM OVERVIEW 1. a system of equations for the economic operation of diverse generation sources in an interconnected power system is developed that utilizes a dynamic line rating system. 1. Figure 1. Application of the powerline ampacity system in the economic operation of a power system is presented. Therefore. there is a need to increase electricity generation. describes the role of Independent System Operators (ISO) and powerpool operations from the point of view of transmission line capacity. The end product is easy to use and suitable for application in all geographic regions. electric power authorities everywhere are searching for new ways to maximize the capacity of powerlines. 1.2 INTRODUCING THE POWERLINE AMPACITY SYSTEM As the demand for electricity increases.
The total cost of adding new transmission lines is approximately three billion dollars. The most common type of transmission line * Arthur H. According to the U. Fuldner. and the world.2 OVERHEAD TRANSMISSION LINE The overhead transmission line consists of towers.3–1.1 North American Transmission grid.5 show electricity production in the U. Modeling. Figures 1. and Applications Alberta British Columbia Sas MT ID Washington WY Oregon California NV UT ND SD Manitoba Hudson Bay Quebec Ontario MN WI VT NH RA OH WV VA IN MI MA ME Pacific Ocean Transmission Lines 230kV/345kV/500kV DC Transmission 735kV/765kV Cluster of Power Generation Stations IA NE IL CO KS MO KY NC SC TN MS AZ NM OK AR LA AL GA TX MD DE New York NJ Atlantic Ocean Florida Mexico FIGURE 1. conductors.4 Powerline Ampacity System: Theory.S. Upgrading Transmission Capacity for Wholesale Electric Power Trade. Dynamic rating of transmission lines offers substantial cost savings by increasing the capacity of existing lines such that the construction of new lines may be postponed in many cases.2. 1. and line hardware for the jointing of conductors and for properly supporting the highvoltage line to the transmission line tower.000 circuit km of transmission lines are planned to be added by the year 2004. approximately 3200 billion kilowatthours will be distributed through the transmission network in the year 2000. environmental factors related to land use and EMF are also required to be considered before the construction of new lines. 1998. In addition to the high cost of adding new transmission lines.3.S. U.S. Department of Energy. The total length of transmission lines at each voltage category is shown in the Figure 1.* about 10.5. . December 30. which is expected to grow at the rate of 2% per year. Assuming an annual load factor of 0. of North America with transmission lines having voltages 230 kV and above. The installed generation capacity is approximately 750 GW. Department of Energy publication on the World Wide Web. insulators.
All Aluminum Alloy conductors have been used for their light weight and high strengthtoweight ratio. and guyed and pole towers.GW FIGURE 1. The popular type of hybrid conductors are Aluminum Conductor Alloy Reinforced (ACAR) and Aluminum Alloy Conductor Steel Reinforced (AACSR).9. More recently. US Electric Power Generation Hydroelectric Energy Sourse Nuclear Gas Petrolium Coal (10%) (13%) (13%) (14%) (20%) 0 100 200 300 400 Generation Capacity.3 U. Some typical examples of towers are shown in Figures 1. where the high strength of a steel core is not required.S. which enables longer spans with less sag. Aluminum Conductor Steel Reinforced (ACSR) is the most widely used type of currentcarrying conductor. Some examples of commonly used powerline conductors according to various standards are given in Appendix A (Thrash.6–1. . and steel wires are also used for special applications.2 Total transmission line circuit km in North America according to transmission voltage category. electric utility generation capacity. towers are selfsupporting towers. 1999. aluminum alloy. Other hybrid conductors having various proportions of aluminum. Koch. All Aluminum Conductors (AAC) are used in coastal regions for high corrosion resistance and also for applications requiring lower resistance. Hitachi. 1999. 1999).Introduction 5 70000 60000 50000 Kilometer 40000 30000 20000 10000 0 230kV 345kV 500kV 735kV Transmission Voltage DC FIGURE 1.
and Applications US Electric Energy Generation Hydroelectric Energy Sourse Nuclear Gas Petrolium Coal 0 500 1000 1500 2000 (2%) (57%) (9%) (11%) (20%) Net Generation (TWh) FIGURE 1.4 U.6 Self supporting tower. . Modeling. World Electric Energy Production Hydroelectric Energy Sourse Nuclear Gas Petrolium Coal 0 2000 4000 6000 (20%) (11%) (9%) (2%) (57%) 8000 Net Generation (TWh) FIGURE 1.6 Powerline Ampacity System: Theory.S. utility electric energy production. FIGURE 1.5 World electric energy production.
.Introduction 7 FIGURE 1. FIGURE 1.9 Tubular tower. FIGURE 1.8 Ornamental tower.7 Guyed tower.
In the newer hightemperature. When a dynamic line rating system is implemented in an electric utility system. In the SSAC conductor. electricity is received by highvoltage transmission lines and transformed for distribution at lower voltages. 1.3 HIGHVOLTAGE SUBSTATION The electric substation is an important component of the electric power system.3. Modeling. highampacity conductors. including bus bars. the currentcarrying aluminum wires are in the annealed state and do not bear any tension. Voltage transformation is carried out in the substation by transformers. and Applications Hightemperature conductors are used for bulk power transmission in heavily loaded circuits where a high degree of reliability is required.10. Communication by ﬁber optics offers a noisefree system of data communication in the electric utility environment since communication by optical ﬁber is unaffected by electromagnetic disturbances. and Invar alloy reinforced steel wires are used for the core. For better aerodynamic performance. Compact design is made possible by the trapezoidal shaping of wires instead of wires having the circular crosssections used in conventional ACSR conductors. circuit breakers.8 Powerline Ampacity System: Theory. it is also important to have knowledge of the current rating of all substation equipment in addition to powerline conductor ratings. A system of dynamic rating of substation equipment may be implemented . In a distribution substation. Transformer ratings. control and communication equipment. A transmission substation is generally used for interconnection with other substations where power can be rerouted by switching action. Besides transformers. isolators. aluminum zirconium alloy wires are used to carry high current. as well as to other substations. The tension is borne entirely by the highstrength steel wires. Steel Supported Aluminum Conductor (SSAC) allows hightemperature operation with minimum sag. The different types of conductors are shown in Figure 1. interrupters. Substation switching devices are generally designed to withstand short circuit currents. In an Optical Ground Wire (OPGW) system. protective relays. Important physical properties of the different types of wires used in the manufacture of powerline conductors are given in Appendix B. the ﬁberoptic cable is placed inside the core of the power conductor. The substation is a hub for receiving electricity from where electricity is distributed to load centers. on the other hand. inductive and capacitive reactors for the control of reactive power ﬂow. need to be examined more closely. A typical layout of a highvoltage substation is shown in Figure 1. Recently. conductors are also available with concentric gaps inside the conductor which offer better damping of windinduced vibrations. Another recent development in transmission line conductor technology is the integration of optical ﬁber communication technology in the manufacture of powerline conductors. and have sufﬁcient continuous overload current capability. a ﬁberoptic cable is placed inside the core of the overhead ground wire. wave traps. In certain transmission line applications. metering. there are other important devices in a substation.11. and other lowvoltage equipment for station auxiliary power supply. compact conductor designs have been available that offer lower losses for the same crosssectional area of the conductor. instrument transformers for the measurement of high voltage and current.
and. the national electric power supply company of France.Introduction 9 7 wire AAC 37 wire AAC 37 wire AAAC 19 Alloy 42 ES ACAR 30 EC 7 st ACSR 54 EC 7 St ACSR 42 Es 19 AS ACSR/AS 54 ES AS Compact FIGURE 1.” These expert systems perform a variety of tasks from energy management to alarm processing and fault diagnosis. transmission and distribution lines. operates its electric power supply system through one central control center and seven regional control centers. in the event of a failure of a component in the network.4 ENERGY CONTROL CENTER The electric power system comprising generation stations. 1.13. or by inferring equipment temperature by the measurement of current ﬂowing through the device and monitoring weather conditions at the location of the substation. load forecasting. providing assistance to control system operators for better decision making. performing load ﬂow. Electricité de France (EDF).3. as shown in Figure 1. and switching operations in the substations. which is especially .10 Transmission line conductors. Control centers are responsible for the control of power generation. Control centers constantly monitor the condition of all transmission lines and substations in their respective regions. control actions are taken to remedy the problem. by realtime monitoring of equipment temperature by installing sensors. Modern control centers are operated through a network of computers having intelligent programs called “expert systems. contingency analysis. dynamic and transient stability analysis. and substations is controlled by a system of energy control centers. Each electric power company operates its electricity supply system in a given geographic region through one or more energy control centers. For example.
and Applications High Voltage Transmission Line T/L Earth Switch 500 kV Main Bus O Transfer Bus Bus Sectioalizer Bus Couper 230 kV 33 kV Symbols Distribution Feeders 3 Winding Transformer Circuit Breaker Isolator Lokal Genereation FIGURE 1. General Control Center Regional Control Center Regional Control Center Substation Substation Substation Substation FIGURE 1.11 High voltage substation. Modeling. The powerline ampacity system described in this book is an expert system for the evaluation of transmission line ampacity. which is expected to be an integral part of a modern energy control system. useful during an emergency.10 Powerline Ampacity System: Theory.12 Typical power system hierarchy. .
The models for permanent elongation of conductor is given in a Cigré 1978 report. 1994. 1985).A. The results were presented by the author (Deb. The remedial measures that are proposed to reduce the possibility of transmission line conductor overheating comprise the use of line ampacity programs and the monitoring of transmission line current and/or temperature. ** EMF Conference.. and Deb. Special conductors may be used to transfer higher currents in highly congested transmission circuits. The new types of conductors are capable of operating at signiﬁcantly higher temperatures with less sag and without any thermal deterioration.Introduction 11 FIGURE 1. 1. A recursive estimation algorithm for calculating the loss of tensile strength and permanent elongation due to heating in service from the probability distribution of conductor temperature is described by the author (Deb et al. 1993) for practical line operating conditions. concluded that there are no harmful effects due to powerline electric and magnetic ﬁelds. A study for the assessment of thermal deterioration of transmission line conductor from conductor temperature distribution was presented recently by Mizuno et al. and there are no harmful effects of magnetic ﬁelds on human beings.13 Energy Control Center.S. 1998) presents new transmission line design considerations to lower magnetic ﬁelds.4 FACTORS AFFECTING TRANSMISSION CAPACITY AND REMEDIAL MEASURES The effects of elevated temperature operation are loss of tensile of conductors and permanent elongation of conductors. 1997) shows that transmission capacity may be doubled by the application of new types of powerline conductors. South Korea. A recent research study conducted by EPRI (Rashkes and Lordan. 1978). The loss of strength model is given by Harvey (Harvey. (1998). Choi. A recent study conducted by the author and KEPCO* (Wook. There is general agreement that transmission line magnetic ﬁelds** have minimum impact on the environment. 1972) and (Morgan. This study is important from the * Author worked as a consultant for Korea Electric Power Company (KEPCO). National Academy of Science. U. .
and also enhances generation stability when required. 1. In Chapter 10.6 DYNAMIC LINE RATING COSTBENEFIT ANALYSIS A cost beneﬁt analysis was carried out by the cost capitalization method and the results are presented in Table 1.5 NEW DEVELOPMENTS FOR TRANSMISSION CAPACITY ENHANCEMENT There are other electrical network constraints that must be satisﬁed before transmission lines can be operated at their maximum thermal capacities. 1998b) that eliminates the need for a transformer by generating electricity at high voltage at the level of transmission system voltage. . Deb. Another important development is the invention of a new type of generator called the “Powerformer” (MPS Review. To undertake this study. The signiﬁcance of the study and the main contributions in each chapter are summarized. The most important constraints are voltage levels and generator stability limits. an economic load ﬂow program was developed to simulate an interconnected transmission network with diverse generation sources (Hall and Deb. New methods and devices to improve transmission system voltage levels and generation stability limits include FACTS (Flexible AC Transmission System) (Hingorani. it is now becoming possible to operate transmission lines close to thermal ratings. Yalcinov and Short. FACTS technology makes use of recent developments in modern power electronics and superconductivity (Feak.12 Powerline Ampacity System: Theory. 1997). The new type of generator produces greater reactive power to elevate grid voltage levels.. Modeling.1. In addition to cost savings achieved by postponing the construction of new lines. It is assumed that line current will increase at the rate of 2.5% per year. A recent FACTS development is the Uniﬁed Power Flow Controller (UPFC) (Norozian et al. 1995). The results show that the capitalized cost of higher losses due to the increase in line current by deferment of new line construction for a period of 10 years is signiﬁcantly lower than the cost of constructing a new line in the San Francisco Bay area. 1988a. These studies show that there is considerable interest in maximizing the capacity of existing assets. 1. and Applications point of view of transmission line ampacity so that future transmission lines can be constructed with higher power transfer capability and minimum magnetic ﬁeld. yet higher levels of transmission capacity may be achieved.7 CHAPTER SUMMARY An introduction to the subject of transmission line ampacity is presented in this chapter by giving an overview of the electric power system. by connecting Powerformer directly to the transmission grid. Therefore. 1997) to enhance transmission capacity. dynamic line rating systems also offer substantial operational cost savings. 1998). By the introduction of these new technologies in the electrical power system. 1994. when required. a study is presented which show 16% economy achieved by dynamic line rating by facilitating the transfer of lowcost surplus hydroelectric energy through overhead lines. 1.
new methods and systems are required to maximize the utilization of existing power system assets.403 30.264 Year 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Note: The following assumptions are made in the above calculations: Line load loss factor = 0. Static normal ampacity of the line is 800 A.548 12.675 6.3. @ 10%/yr Interest.263 18.317 12.934 78. $ 0 2. regulatory.498 Total present value of loss. Line load increase by 2.5% /year. and new methods are required to maximize their utilization. Conductor is ACSR Cardinal. $/mile Cost of new line. As the demand for electricity grows.902 15. The development of a complete line ampacity system having transmission line.794 22.5%/yr.000 121.759 5. Highvoltage transmission lines are critical components of the electric power system.736 200. @ 10c/kWh. $/circuit/mile Present Value of Annual Loss.129 9. 1 conductor/phase. based on system load factor = 0.477 10. The object of this book is to present a study of transmission line conductor thermal modeling. A 800 820 841 862 883 905 928 951 975 1000 Annual Energy Loss Increase. $/circuit/mile Saving by LINEAMPS. and to present the applications of line ampacity in the operation of electric power systems. Increase @ 2. $ 0 2. and economical reasons it is not always possible to construct new lines. .1 Cost–Beneﬁt Analysis Line Current. Due to environmental.Introduction 13 TABLE 1. For this reason it was necessary to develop a computer program that will adapt to different line operating standards followed by power companies in the different regions of the world.5.538 8.508 4.702 11. and conductor models that can be easily implemented in all geographic regions was a major challenge.609 11. to develop a methodology for the rating of transmission lines for implementation in a computer program that is suitable for all geographic regions. This was accomplished by developing an expert system and objectoriented modeling of the line ampacity system.657 8.505 26. weather. single circuit line. Rate of interest is 10% /year.
The factors limiting line capacity are clearly brought out. . and Applications The economic incentives for implementing a dynamic line rating system are clearly established by showing the approximately 60% cost saving by the deferment of new line construction. and the means to overcome these are explained. Modeling.14 Powerline Ampacity System: Theory.
Brussels. is the national electric power company of France (Urbain.1 HISTORICAL BACKROUND EARLY WORKS ON CONDUCTOR THERMAL RATING Faraday was one of the early researchers who conducted theoretical and experimental research to study the heating of wires by electric current (Faraday. *** Electricité de France. A transmission line rating system using temperature monitoring by a thermal image of conductors was developed in Belgium (Renchon. 15 . 1992. 3–5 June. 1998). 1993) and Cigré (Cigré. dynamic. Savoullis. CA. A steadystate ampacity model based on the conductor heat balance equation was presented in 1956 (House and Tuttle..1. ** Cigré Symposium: High Currents in Power Systems under Normal. 1956). 1997.3 UTILITY PRACTICE Electric power companies*** generally assume that the ampacity of transmission line conductors is constant. A similar model was presented at the IEEE (Hall.2 2. Electricity. Ampacity calculations are commonly based upon the following conservative assumptions of ambient temperature. 1985. Paciﬁc Gas & Electric Co. Belgium. wind speed. For the shortterm rating of transmission line conductors. 1988) for the calculation of thermal gradient of conductor from surface to core. and transient states.* Some early works on transmission line conductor thermal rating were conducted in France (Legrand. Emergency and Fault Conditions. The Cigré report presents a threedimensional thermal model of conductors for unsteadystate calculation.1. Paris. page 686. Weather modeling for transmission line ampacity was ﬁrst presented by the author at the Cigré Symposium** on High Currents (Deb et al. India (Deb et al. devoted to the subject of transmission line ampacity. 1956). Great Books # 42.2 IEEE AND CIGRÉ STANDARDS IEEE (IEEE Standard 738. and Deb.1. and maximum conductor temperature: * Michael Faraday. 2. 2. San Francisco. 1985). 1999) offer standard methods for the calculation of transmission line ampacity in the steady. 1945) that realized the importance of transmission line conductor thermal ratings. All of the above research shows that there has long been considerable interest in maximizing the transmission capacity of overhead lines. Encyclopaedia Britannica. (PG&E Standard 1978).1 Line Rating Methods 2. Central Board of Irrigation and Power. 1834). 1985). solar radiation. Davidson (1969) presented a solution to the differential equation of conductor temperature by using the Eulers method..
A dynamic line rating system can provide further increase in line ampacity for short durations by taking into consideration the heatstorage capacity of conductors. Therefore.t = min(Il.j. determine the maximum current that can be passed through the line at a given time (t) such that the conductor temperature (Tc) at any section of the line does not exceed the design maximum temperature (Tmax) of the line. Dl.j. many utilities have started adapting line ratings to actual weather conditions to increase line capacity.2 LINE RATING METHODS 2. and Applications • • • • Ambient temperature = 40°C Wind speed = 0.2.16 Powerline Ampacity System: Theory. 2. ∀lIl.2.t.t = f(Wsk. Cl. Modeling.j.1 DEFINING THE LINE AMPACITY PROBLEM The problem of determining the thermal rating of an overhead powerline can be stated as follows: based on existing and forecast weather conditions at several locations along the transmission line route. Wdk. Tcl. Stated formally.t.3… K weather stations (2. Tak.j Where.3… J line sections t = 1.j.61 m/s (2 ft/s) Solar radiation =1000 W/m2 Maximum conductor temperature = 80°C It is well known that weather conditions are never constant. during favorable weather conditions when ambient temperature is lower than the assumed maximum or wind speed is higher than the assumed minimum.1) (2.2.3… L transmission lines j = 1.3… T time (T = 168 h in LINEAMPS) k = 1.j) Tcl. Srk.2.j.2) (2.t) Il. For the above reasons. I = Ampacity (Ampere) Ws = Wind speed Wd = Wind direction Ta = Ambient temperature Sr = Solar radiation Tc = Conductor temperature C = Conductor D = Direction of line l = 1.t.j.3) .j. higher ampacity is possible without exceeding the allowable maximum temperature of the powerline conductor.t.j ≤ Tmaxl. or during cloudy sky conditions.2.j.
as it does not require monitoring weather conditions or installation of sensors on the transmission line conductor. static line ratings are ﬁxed for a particular season of the year. The static rating of transmission lines in a region is generally determined by analysis of historical weather data of that region for the different types of conductors used in the transmission lines. For example. TABLE 2.2 STATIC AND DYNAMIC LINE RATINGS Transmission line rating methods are broadly classiﬁed into two categories: static and dynamic line rating. and many electric power utilities have different line ratings for summer and winter. and weather conditions all along the transmission line route. emergency condition A = 90°C (100 hr total) Conductor temperature. Conductor tension is overseen by tension monitors that are attached to insulators on . Conductor temperature is monitored by installing conductor temperature sensors at certain sections of the transmission line. Online line rating methods include monitoring conductor temperature or tension.1 Ampacity of ACSR Conductors Summer Coastal Conductor Size.1 Summer ambient temperature = 37°C with sun Winter ambient temperature = 16°C without sun Wind velocity = 0. emergency condition B = 100°C (100 hr total) (Emergency B ratings are shown in the table) Emissivity = 0.1. the static line rating of some typical conductor sizes used by PG&E in the region of the San Francisco Bay area is given in Table 2. The static line rating system is widely used because of its simplicity. mm2 210 264 375 624 749 874 1454 Normal 382 442 550 752 919 1060 1319 Emergency 482 558 697 959 1133 1312 1642 Winter Coastal Normal 550 640 801 1108 1218 1448 1814 Emergency 616 716 898 1243 1393 1625 2040 Basis for Table 2. normal condition = 80°C Conductor temperature.Line Rating Methods 17 The LINEAMPS computer program described in this book ﬁnds a solution to the above line ampacity problem.2. Generally.6 m/s perpendicular to conductor axis Conductor temperature. 2.5 Conductivity of aluminum = 61% IACS Dynamic line ratings are obtained by online or ofﬂine methods.
1985). Due to these reasons. and solar radiation from several weather stations are entered into a computer where a line ampacity program calculates steadystate and dynamic ampacity. tension monitors are required to be located only at anchor towers.. Similarly. 1988b. The ampacity of a line is calculated from conductor sag and weather data by taking a series of measurements of conductor sag at different transmission line spans along the length of the line. Unlike temperature monitoring systems. 1995b). 1989). the accuracy of the system in forecasting transmission line ampacity several hours ahead is somewhat limited. Hall and Deb. Diurnal weather patterns of the region are considered for the prediction of line ampacity several hours in advance. 1998). realtime measurements of wind speed. In the ofﬂine system. line ratings are obtained uniquely by monitoring weather conditions along the transmission line route. Mauldin et al. 1991).. In both monitoring systems. A method in each weather station object generates hourly values of meteorological data from this series. Steeley et al. and their usefulness to forecast powerline ampacity was recognized by many researchers (Foss and Maraio. Because of these assumptions. These methods require weather data on a continuous basis.18 Powerline Ampacity System: Theory. a weatherdependent realtime line rating system has been developed for the Spanish 400 kV transmission network (Soto. whereas installing an unlimited number of temperature sensors on a transmission line is not economical. A weatherdependent line rating system developed in the UK is described in a Cigré article (Jackson and Price.. line ampacity is adjusted based on previous 24hour weather data. The powerline objects have a plurality of virtual weather sites that receive their data from a plurality of weather station objects. and Applications tension towers. wind direction. and the ampacity of the line is calculated at the base station computer from this data.2. 1988. sensor data is communicated to a base station computer by a radio communication device installed on the sensor. The existence of daily and seasonal cyclical weather patterns are well known.. et al. the periodic cyclical pattern of wind speed and ambient temperature are considered in a unique manner to forecast powerline ampacity. In the Spanish system.3 WEATHERDEPENDENT SYSTEMS Weatherdependent line rating systems were proposed by several researchers (Cibulka et al. 2. The number of virtual weather stations that can be accommodated in a powerline is limited only by the computer processing speed and memory. 1986. 1988b). ambient temperature. In the LINEAMPS program (Deb. and a method in each powerline object determines the minimum hourly values of line ampacity up to seven days in advance. 1995a. In their method. a . 1992. An ofﬂine system may also include monitoring conductor sag by pointing a laser beam at the lowest point of the conductor in a span. Modeling. Weather patterns of a region are stored in Fourier series in each weather station object. Foss and Maraio (1989) described a line ampacity system for the power system operating environment. (Hall and Deb. They were also interested in forecasting transmission line ampacity. Douglass.
Line Rating Methods 19 weatherdependent line rating system is expected to be more reliable and more accurate than systems utilizing realtime measurements from a limited number of locations. meteorological data. Pr.4 ONLINE TEMPERATURE MONITORING SYSTEM U. W/m Pc = Heat lost by convection due to the cooling effect of wind. Pr = Heat lost by radiation. Davis’s system (Davis. Ps. An online temperature monitoring system using Power Donut ™ temperature sensors is shown in Figure 2. Powerline and conductor ampacity is estimated by the program from user input and by synthetic generation of weather data from selfgenerating weather station objects. General purpose weather forecast data available from the internet are used in the LINEAMPS program. 1977) required the installation of conductor temperature sensors as well as meteorological sensors at several locations along powerlines. 1992) was given for a transmission line rating system that calculates the current carrying capacity of one or more powerlines by the measurement of conductor temperature and meteorological conditions on the line. The online monitoring systems described in the IEEE and Cigré papers (Davis. In this method.S. or conductor temperature measurements from the powerline.2. ohm/m In the above equation. and maintenance problems. line ampacity is predicted accurately by realtime numerical solution of the following conductor temperature differential equation at a location: Pr + Pc – Ps R ac (2. A realtime dynamic linerating model was proposed (Black and Byrd. The Power Donut ™ temperature sensor is shown in Figure 2. 2.4) . and line current are continuous input to a computer system where line ampacity is calculated. 1956) are not widely used because of transmission distance. and Rac are functions of conductor temperature. The computer system requires special hardware and software for data acquisition from remote sensor locations via special telecommunication networks.1. Pc. communication requirements. Realtime conductor temperature. Renchon and Daumerie. The new line ampacity system does not require realtime continuous input of meteorological data. 1984. W/m Rac = AC resistance of conductor. 1983). Patent 5140257 (system for rating electric power transmission lines and equipment. Howington and Ramon. 1977.2. line ampacity is calculated by the measurement of conductor temperature and by the solution of the conductor heat balance equation as follows: I= where. line current. In this method. W/m Ps = Heat gained by solar radiation.
W/m Pm = Magnetic heating. W/m Pc = Heat lost by convection due to the cooling effect of wind. conductor mass.20 Powerline Ampacity System: Theory. kg/m A = conductor area. and weather station and ground station RTU. °C measured by line temperature sensor Pr = Heat lost by radiation. W/m Ps = Heat gained by solar radiation.5) M = γ · A. Inc. m2 Tav = average conductor temperature. W/m Pj = Joule heating. W/m The following expression for the calculation of realtime dynamic ampacity was obtained by the author: . kg/m3 Tav = Tc + Ts 2 (2.) M ⋅ cp Where. and Applications POWER DONUT SENSORS TM GROUND STATION SUBSTATION CONTROL HOUSE RTU FIGURE 2.6) Tc = Conductor core temperature. Modeling. °C γ = conduct or density. (© Courtesy Nitech. °C Ts = Conductor surface temperature.1 Online temperature monitoring system is comprised of Power Donut™ temperature sensors. dTav = Pj + Ps + Pm – Pr – Pc dt (2.
and time ∆t C1. C2 = Constants The different terms in the above equation are described in Chapter 3.5 ONLINE TENSION MONITORING SYSTEM The online tension monitoring system is used to predict transmission line ampacity by measurement of conductor tension at tension towers along the transmission line (Seppa. (© Courtesy Nitech. et al. . The calculation of dynamic ampacity by the above equation requires realtime conductor temperature and meteorological data on a continuous basis.Line Rating Methods 21 FIGURE 2.2 Power Donut™ temperature sensor. Inc.2. 1998).. 2. Since conductor tension is a function of conductor temperature.7) Tmax = Max conductor temperature Tinitial = Initial temperature Tinitial.) I= {T max C1 {1 – exp( – ∆t τ)} – Tinitial exp( – ∆t τ)} – C2 (2. the ampacity of the transmission can be obtained by realtime monitoring of conductor tension as follows.
kg/mm2 Tc1. It may be feasible to install such devices on certain heavily loaded lines. by measurement of conductor tension and by knowledge of initial conditions. Transmission line spans 1 2 3 4 5 Laser beam Sag measuring instrument FIGURE 2. °C–1 (2. m ∆Ec = inelastic elongation (creep) mm/mm α = coefﬁcient of linear expansion of conductor. respectively. 2. Transmission line ampacity is then calculated by the solution of conductor heat balance. The ampacity of the transmission line is then obtained by taking a series of measurements at different transmission line spans along the length of the line. °C E = Young’s modulus of elasticity.5. The major disadvantages of this method are that it requires taking the transmission line out of service for installation and maintenance. conductor sag is measured by pointing a laser beam at the lowest point of the conductor in a span. Therefore. . In this method. the temperature of the conductor is obtained by the solution of the above equation. σ2 = stress at state1 and state2.6 SAGMONITORING SYSTEM This is an ofﬂine method of realtime line rating by monitoring conductor sag.3 Sagmonitoring line ampacity system. kg/m/mm2 L = span length. but is impractical and expensive to install tension monitors on all transmission and distribution lines for line ampacity predictions of all overhead lines in a system.2.22 Powerline Ampacity System: Theory. Ampacity is calculated from conductor sag and by measurement of weather conditions. Equation 2. and Applications σ 2 (ϖ ⋅ L)2 σ (ϖ ⋅ L)2 + α(Tc 2 – Tc1 ) + ∆Ec = 1 – – 2 E 24σ 2 E 24σ1 2 σ1. kg/mm2 ϖ = speciﬁc weight of conductor. Modeling.8) Therefore. this system does not require taking the line out of service during installation or maintenance of the sagmonitoring device. This method of monitoring a transmission line has the added advantage of monitoring iceloads as well. Tc2 = conductor temperature at state1 and state2. It is an ofﬂine method because it does not require the installation of any device on the transmission line conductor.
line ratings are determined. and Paoli. However.4.. Koval and Billinton. Mizuno et al. and then ampacity is calculated from (2. A limitation is that line ampacity is not adaptive to real weather conditions. In lowvoltage distribution lines. Nicolini. An example of an OPGW conductor on overhead line is shown in Figure 2. 1985. it must be mentioned that probability modeling of conductor temperature is useful for the prediction of conductor performance in service. the temperature of the conductor is then obtained from (2. Hall and Deb. An attractive feature of this technique is that continuous input of realtime weather data is not required. and by the solution of the conductor heat balance equation.9). kg Having calculated conductor tension T from (2.2. 1998). 1985. it is well known that electricity demand depends upon weather conditions. 1993. This kind of conductor. A ﬁberoptic cable also may be placed within the core of a phase conductor in highvoltage lines. it is now possible to have a distributed system of ﬁberoptic conductor temperature sensors that will span the entire length of the transmission line. it is possible to consider the correlation between the different variables (Douglass. the ﬁberoptic cable may be wrapped over the conductor. with a ﬁberoptic cable in the core.9) 2. 1970. Urbain. 1986). . Giacomo. the ﬁberoptic cable is embedded inside the core of powerline ground wire. 1988b. At the present time.5). Probability modeling of conductor temperature is used to predict the loss of tensile strength and permanent elongation of conductor during the lifetime of the transmission line conductor (Deb.Line Rating Methods 23 Conductor sag is calculated approximately by the wellknown parabola equation: Sag = W = conductor weight. WL2 8T (2. since a ﬁberoptic cable will be used for data transmission and will eliminate the need for a separate communication system for transmission of conductor temperature data from a transmission line to utility power control center. 1998). A distributed temperature sensor system will result in a more accurate realtime line rating system.8). kg/m T = conductor tension.7 DISTRIBUTED TEMPERATURE SENSOR SYSTEM With the development of the powerline communication system by a ﬁberoptic cable integrated with a powerline conductor. For example. is called an OPGW conductor.. Probabilistic ratings are determined by Monte Carlo simulation of meteorological variables. 1979. The author has obtained the probability distribution of conductor temperature by Monte Carlo simulation of time series stochastic models of the meteorological variables and transmission line current. From the resultant probability distribution of conductor ampacity. Probabilistic ratings are used by several power companies (Deb et al. By using time series stochastic models.
and Applications FIGURE 2.4 Power line conductors with ﬁberoptic cable. 1991) and wind speed (Hall and Deb. A2. A3. Because weather patterns are stored . A4. A stochastic model was also used to forecast solar radiation (Mauldin et al. A5. Z(t – 2) = difference of measured and predicted temperature at time (t – 1) and (t – 2) respectively A1. A6.10) Z(t – 1).24 Powerline Ampacity System: Theory. The general form of the stochastic model is given below: Ta(t) = A + A2·Sin(wt) + A3·Sin(2wt) + A4·Cos(wt) + A5·Cos(2wt) + A6·Z(t – 1) + A7·Z(t – 2) (2. 1988b). Modeling.. A time series stochastic and deterministic model was used to predict realtime probabilistic ratings of transmission line ampacity up to 24hours in advance based on ambient temperature measurements only. Fourier series models of ambient temperature and wind speed are used to generate weather data. and by assuming constant wind speed (Steeley et al. A7 are the coefﬁcients of the model ω = 2π/T = fundamental frequency T = 24 hour = period In the LINEAMPS program. 1991)..
1995a. 1998). alarmprocessing (Taylor et al.. A direct solution of the conductor temperature differential equation is used. (Wook et al. and weather model to forecast line ampacity up to seven days in advance. but are also convenient repositories for the storage of line data and ampacity that are easily retrieved and displayed on a computer screen. An important contribution of the new line ampacity system is the ability to selfgenerate hourly values of weather data from statistical and analytical models. 1993) for synthetic generation of weather . It has provision for steadystate rating..Line Rating Methods 25 in weather objects. Anjan K.2. It was shown that powerline objects not only have methods to predict line ampacity. This is to be expected. Expert systems (Kennedy. and for weather stations to estimate line ampacity up to seven days in advance.8 OBJECTORIENTED MODELING RATING SYSTEM AND EXPERT LINE The estimation of powerline ampacity by the application of objectoriented modeling and expert system rules was ﬁrst presented by the author (Deb. 1998) and for power system fault analysis (Negnevitsky.. and an analytical expression for the direct solution of dynamic line ampacity is presented for the ﬁrst time (Deb. dynamic line rating. 1995b). 1998a. Important electric power companies are embracing the objectmodel approach to meet their information requirements for the year 2000 and beyond (MPS Review article. The concept of steady. 1995). Integrated Line Ampacity System LINEAMPS (Deb. 1995). The author previously developed an algorithm (Deb. conductor model. 1998b).. dynamic. M. 1997) is an integrated line ampacity system having a transmission line model. LINEAMPS is the ﬁrst expert system for the estimation of transmission line ampacity (Deb. it eliminates the need for realtime measurements on a continuous basis. A threedimensional conductor thermal model is used to calculate conductor thermal gradient (Deb. the parameters of the weather models in the LINEAMPS program are adjusted to National Weather Service forecasts. For example.1998a). eliminating the need for realtime weather data on a continuous basis. It was shown for the ﬁrst time how objectoriented modeling of transmission line ampacity enabled program users to easily create new lines and new conductors. 1998). 2. 2000) are developed in the electric power system for power quality machine diagnosis. because timeseries models are statistical models that do not consider a physical model of the atmosphere. and transient rating. EDF has selected objectmodel technology for the management and operation of the transmission grid in France. Therefore. In this regard. One of the limitations of the stochastic model is that it is unsuitable for the predictions of hourly values of ambient temperature for more than 24 hours in advance. and transient line ratings are used in this program for the ﬁrst time. National Weather Service forecasts are generally more accurate for longterm weather predictions because they are derived from atmospheric models. Weather data is required only when weather conditions change.
and conductor thermal models that can be easily implemented in all geographic regions was a major challenge. or artiﬁcial generation of meteorological data by a Fourier series weather model of ambient temperature and wind speed of a region evolved from these developments. 1988b). sag monitoring. recursive estimation. tension monitoring system. and the various methods of rating power transmission lines are critically examined to show the advantages and deﬁciencies of each method. The probability distribution of transmission line conductor temperature is required to calculate the thermal deterioration of transmission line conductor (Mizuno et al. from the early works on conductor thermal rating to modern applications of objectoriented modeling. and probabilistic rating methods. For this reason. This was accomplished by objectoriented modeling and by developing an expert system computer program called LINEAMPS . The development of an integrated powerline ampacity system having transmission line.. The line ampacity problem is clearly deﬁned. Synthetic generation of weather data from a model is also useful to evaluate the probability distribution of transmission line conductor temperature in service (Hall. temperature monitoring system. The idea of selfgeneration. and realtime transmission line ratings. Deb. synthetic generation. Modeling.26 Powerline Ampacity System: Theory. expert systems. 2000). it was necessary to develop a computer program that will adapt to the different line operating standards followed by power companies in the different regions of the world. weatherdependent system. The various methods of rating transmission lines includes static and dynamic thermal ratings. distributed ﬁberoptic sensors. 1998. 2. and Applications data by timeseries analysis and recursive estimation.3 CHAPTER SUMMARY Transmission line rating methods are introduced by presenting a critical review of literature. weather.
Transient conditions arise due to shortcircuit or lightning current. In the following section. and operating conditions. equations for the radial conductor temperature differential from surface to core of a conductor are developed from the same equations. Due to the thermal inertia of the conductor. During steadystate conditions. During transient conditions there is no heat exchange with the exterior. the equations for the calculation of conductor temperature and ampacity are derived from the general heat equation for steadystate. and transient conditions. This chapter prepares the framework for computer modeling of the line ampacity system described in Chapter 8. Then.1 INTRODUCTION As mentioned in the previous chapters. and depends upon the following operating conditions: • Steadystate • Dynamic state • Transient state The conductor is assumed to be in the steadystate during normal operating conditions when the heat gained due to line current and solar radiation equals the heat lost by cooling due to wind and radiation. Line ampacity is generally based on a maximum value of conductor temperature determined by the type of conductor. conductor temperature. and the temperature of the conductor is fairly uniform. the currentcarrying capacity of a transmission line conductor is not constant but varies with weather conditions. dynamic state. shortterm overloads may be supplied through the line without overheating the conductor before steadystate conditions are reached. 27 . and adiabatic conditions are assumed. weather conditions are assumed stable. the transmission line current is considered constant. Dynamic conditions arise when there is a step change in line current.3 Theory of Transmission Line Ampacity 3. Line energization or sudden changes in line current due to a failure on one circuit are examples of dynamic operating conditions. A typical example of dynamic loading is when the load from the faulted circuit in a double circuit line is transferred to the healthy circuit.
Number 2.2) ∂2 ∂2 ∂2 + 2 + 2 ∂x 2 ∂y ∂z q 1 ∂T = k α ∂t (3. A. Savoullis. ∇2 T = ∂ 2 T 1 ∂T 1 ∂ 2 T ∂ 2 T + + + ∂r 2 r ∂r r 2 ∂ϕ 2 ∂z 2 (3.1) and (3. α= k γc p cp = speciﬁc heat capacity γ = mass density From (3.2. ∂T λ ∂ 2 T 1 ∂T = + +q ∂t γ ⋅ c p ∂r 2 r ∂r λ = Thermal conductivity * J.3) . Modeling.2) we obtain. ∇2 T + Where. ∇ 2 = Laplacian operator. Transactions on Power Delivery. Volume 3. April 1988. Deb. and Applications 3. Hall. ∇2 = In cylindrical coordinates. IEEE.1) T = conductor temperature r = radial length ϕ = azimuth angle z = axial length q = power per unit volume k = thermal conductivity of conductor α = thermal diffusivity given by.28 Powerline Ampacity System: Theory. the solution of the transmission line ampacity is found for each of the above operating conditions as follows: The general heat equation* for a transmission line conductor is given by. (3. Wind Tunnel Studies of Transmission Line Conductor. J.1 GENERAL HEAT EQUATION Starting with the general heat equation of the transmission line conductor.2 CONDUCTOR THERMAL MODELING 3.F.K.
3): M ⋅ cp Where. M = γ⋅A.3 may be solved numerically with appropriate initial and boundary condition. 174.Theory of Transmission Line Ampacity 29 Equation 3.2 DIFFERENTIAL EQUATION OF CONDUCTOR TEMPERATURE For practical consideration of transmission line conductor heating. W/m Pr = radiation heat loss. W/m Pm = magnetic heating.3 STEADYSTATE AMPACITY The calculation of transmission line ampacity may be simpliﬁed if steadystate conditions are assumed.T. m2 Pj = joule heating. 3. The following assumptions are made in steadystate analysis: * V. .2. 1443–1452. Morgan.1. October 1997.2. W/m Pc = convection heat loss. Cigré Working Group WG 22. pp. conductor mass.12 report. The radial temperature distribution and effective radial thermal conductivity in bare solid and stranded conductors. W/m dTav = Pj + Ps + Pm – Pr – Pc dt (3. W/m Ps = solar heating.3 or solved analytically by making certain simplifying assumptions that are presented in the following section. IEEE Transactions on Power Delivery. we can calculate the average conductor temperature by the solution of the following differential equation obtained from (3. Volume 5. Tav = average conductor temperature Tc = conductor core temperature Ts = conductor surface temperature With the above assumption.2. it is possible to make the following assumption with sufﬁcient accuracy:* Tav = Where.4) 3. Section 3: Mathematical model for evaluation of conductor temperature in the unsteady state. kg/m A = conductor area. Électra No.5) Tc + Ts 2 (3. July 1990 The thermal behaviour of overhead conductors.
the ampacity of ACSR conductor is calculated by iteration as shown in Example 1. Modeling.6) (3. m (3. A Rac = Rdckac[1 + α0(Tc – T0] (3.9) Rac = AC resistance of conductor. °C Ps = heat gains by solar radiation. and Applications • Conductor temperature remains constant for one hour. °C To = reference conductor temperature. Description of symbols: I = ampacity.30 Powerline Ampacity System: Theory. resulting in the dt conductor heat balance equation: Pj + Ps + Pm – Pr – Pc = 0 By substitution. solar radiation. /°C Tc = conductor temperature. W/m Ps = αsD(Sb + Sd) αs = coefﬁcient of solar absorption D = conductor diameter. Pj + Pm = I2Rac(Tc) The following steadystate solution of conductor ampacity (I) is obtained: I= Pr + Pc – Ps R ac (Tc) (3. Rdc = DC resistance of conductor at the reference temperature To. • Ambient temperature.10) . ohm/m R kac = ac R dc α0 = temperature coefﬁcient of resistance.8) Since the AC resistance of ACSR conductor varies as a function of conductor current (Appendix 3). ohm/m (Rac may be obtained directly from conductor manufacturer’s data sheet or calculated as shown in the Appendix). wind speed. dT The steadystate solution is obtained by setting av = 0 in (3.7) (3. • Conductor current remains constant for one hour. and wind direction are constant for one hour.5).
13) h = coefﬁcient of heat transfer from conductor surface to ambient air.16) (3. m2/s kwd = wind direction correction factor given by (Davis.2 Re0. degree θ = angle of beamed radiation with respect to conductor axis.14) (3.12) Sext = 1353 W/m2. m/s vf = kinematic viscosity of air. Stephan Boltzman constant. dimensionless Nu = 0.17) (3. W/m2 Sb = Sext τb cos(z) Sd = Sext τd cos(θ) (3.194 cos(2ω) + 0.67⋅10–8. normal component of the extra terrestrial solar radiation measured outside the earth’s atmosphere τb = atmospheric transmittance of beamed radiation τd = atmospheric transmittance of diffused radiation z = zenith angle.194 – sin(ω) – 0. degree (3. W/m Pc = h · π · D(Tc – Ta) (3.364 sin(2ω) ω = wind direction with respect to conductor normal.15) .64 Re0. W/m Pr = σεπD{(Tc + 273)4 – (Ta + 273)4} Ta = ambient temperature. W/(m2 · °C) h = λ · Nu · Kwd/D λ = thermal conductivity of ambient air.2 + 0. W/(m · °C) Nu = Nusselt number.18) (3. W/m2 Sd = diffused solar radiation.61 Re = Reynolds number. (W/m2 K4) ε = Emissivity of conductor Pc = heat loss by convection. °C σ = 5.Theory of Transmission Line Ampacity 31 Sb = beamed solar radiation. dimensionless Re = D · ws/νf ws = wind speed.11) (3. degree Pr = heat loss by radiation. 1977) Kwd = 1.
of the conductor of any size for any temperature may be calculated by the procedure given in Appendix 3. conductor construction. Receive input from steady state session window Heat gain by radiation Ps Radiation Heat Loss Pr Convection Heat Loss Pc I= Ps Pr R ac Pc FIGURE 3. For stranded conductors without a magnetic core.1. Conductor temperature is 80°C. Rac.32 Powerline Ampacity System: Theory. A ﬂow chart of the steadystate temperature calculation method is given in Figure 3. The AC resistance of conductors having a steel core can be 5 to 15% or more higher than its DC resistance due to the current induced in the steel core as shown in Appendix 3. The value of the AC resistance.5 . As shown in the Appendix. The meteorological conditions and conductor surface characteristics are as follows: Ambient temperature = 20°C Wind speed = 1 m/s Wind direction = 90° (perpendicular to conductor axis) Solar radiation = 1000 W/m2 Emissivity = 0.5 Absorptivity = 0. Rac. Example 1 Calculate the steadystate ampacity of an ACSR Cardinal conductor. the AC resistance of conductor is calculated from the current distribution inside the conductor. and the magnetic properties of the steel core in ACSR. the AC resistance may be 2 to 5% higher than the DC resistance due to skin effect. A ﬂow chart of the steadystate current method is shown in Figure 3. of a conductor is generally available from the manufacturer’s catalog for standard conductor sizes at a certain speciﬁed conductor temperature. Modeling. and a numerical application is given in Example 1. and a numerical application is given in Example 2. These methods are used in the LINEAMPS program.1 Flow chart of steadystate ampacity method. and Applications The AC resistance.2.
004(80 – 20)} Pj = I2 ⋅ kac ⋅ 0.0741⋅ 10–3 W/m Calculate solar heat gain Ps Ps = αs · D · Fs Ps = 0.2 · 10 Tf = 0. Solution Calculate joule heat gain Pj Pj = I2 ⋅ kac ⋅ Rdc20{1 + α0(Tc – T0)} Pj = I2 ⋅ kac ⋅ 0.Theory of Transmission Line Ampacity 33 Receive input from steady state session window Heat gain by solar radiation Ps Assume initial conductor temperature equal to ambient temperature Joule + Mag heat gain Pj + Pm Radiation heat loss Pr New estimate of conductor temperature Convention beat loss Pc Error = Ps + Pj + Pm .2 · 10–1 = 0.Pr .39 · 10–3 · 1000 Ps = 15.42 · 10–2 + Tf · 7.5 · 30.42 · 10–2 + 50 · 7.19 W/m Calculate convection heat loss by wind Pc Pc = kf · Nu · π(Tc – Ta) kf = 2.5973⋅ 10–3 {1 + 0.2 Flow chart of steady state conductor temperature method.5(80 + 20) = 50°C kf = 2.01W/m Steady:Conductor Temperature Tc FIGURE 3.Pc Abs(Error) > 0.0278 W/(m·°K) .
795 ⋅ 10 –5 Nu = 0.795 · 10–5 m2/s Re = 1.19 = 120 + 2.5 · 10–8 = 1.2 + 0.795 · 10–5 m2/s H = Altitude.5 · π · 30.103 kg/m3 v = 1 m/s 6. and Applications Nu = 0.39 ⋅ 10 –3 = 1867 1.0278 ⋅ 30.19 k ac ⋅ 0.64(1867)0.08 – 15.34 Powerline Ampacity System: Theory.2 · Re0.5 · 10–8 vfo = 1. I 2 ⋅ k ac ⋅ 0.2(1867)0.61 Re = M air ⋅ v ⋅ D vfh Mair = air density = 1.0278 ⋅ 23 ⋅ π ⋅ (80–20) Pc = 120 W/m Heat lost by radiation Pr Pr = s · ε · π · D[(Tc + 273)4 – (Ta + 273)4] Pr = 5.08 I= 120 + 22.16 ( ) –5.0741 ⋅ 10 –3 . H = 0 vfh = vfo = 1.67 · 10–8 · 0.2561 vfo = 1.64 · Re0.39 · 10–3[(Tc + 273)4 – (Ta + 273)4] Pr = 22.32 · 10–5 + Tf · 9.32 · 10–5 + 50 · 9.0741 + 15. Modeling. m Altitude at sea level.39 ⋅ 10–3 ⋅ π ⋅ (Tc –Ta) = 0.08 W/m By substitution in the steadystate heat balance equation.103 ⋅ 30.5 ⋅ 10 –3 vfh = vfo 1 – H 288. Pj + Ps = Pc + Pr We obtain.061 = 23 Pc = 0.2 + 0.
Theory of Transmission Line Ampacity 35 assume kac = 1.39 · 10–3 [(Tc + 273)4 – (Ta + 273)4] . Example 2 Calculate the temperature of an ACSR Cardinal conductor. and all other conditions are the same as in Example 1.67 · 10–8 · 0.13): Pr = 5. and the ﬁnal value of current is found to be.16 The revised value I from (3.14 · 0. Ps = 15. I = 1308 A For this current kac = 1.9) we have. Pj = 12202 · 1. the convection heat loss Pc is obtained from (3.05973 · 10–3 {1 + 0.14. and the joule heat gain Pj is then calculated as. Solution From (3.14): Pc = 0.5 · π · 30. I = 1214 A For this current kac = 1.14 This process is repeated until convergence. Conductor current is 1220 A.004(Tc – 20)} From Example 1 we obtain the value of solar heat gain Ps. I = 1220 A The calculation of AC resistance of ACSR Cardinal conductor is shown in the Appendix 1.8) and (3.19 W/m Using the value of kf and Nu calculated in Example 1.8 ) is. Pj = I2 · kac · Rdc20 {1 + α0(Tc – T0)} For I = 1220 A we obtain kac = 1.0278 ⋅ 23 ⋅ π ⋅ (Tc – 20) The heat loss by radiation Pr is given by (3.
19 = 0. the heat storage capacity of the conductor is considered. Modeling. A direct solution of steadystate conductor temperature Tc is also obtained from the following quartic equation (Davis.4 DYNAMIC AMPACITY The transmission line conductor is assumed to be in the dynamic state when there is a shortterm overload on the line due to line energization or a step change in load.14 · 0.19) is also found to be 80°C.004(Tc – 20)} + 15.19) ( ) 3.5 · π · 30. a1 = π ⋅ D ⋅ ε ⋅ σ a 2 = a 1 ⋅ 4 ⋅ 273 a 3 = a 1 ⋅ 6 ⋅ 2732 a 4 = a 1 ⋅ 4 ⋅ 2733 + π ⋅ λ ⋅ Nu a 5 = – Pj + Ps + a 1 ⋅ Ta4 + a 2 ⋅ Ta3 + a 3 ⋅ Ta2 + a 4 ⋅ Ta and obtain Tc = 80°C The result of conductor temperature Tc obtained by the direct solution of the quartic equation (3.39 · 10–3 [(Tc + 273)4 – (20 + 273)4] The above equation is solved for Tc by iteration by giving an initial value Tc = Ta.36 Powerline Ampacity System: Theory. In the dynamic state.05973 · 10–3 {1 + 0. 1977): a 1 ⋅ Tc4 + a 2 ⋅ Tc3 + a 3 ⋅ Tc2 + a 4 ⋅ Tc + k 5 = 0 where. (3. The duration of such overload condition is generally less than 30 minutes.67 · 10–8 · 0.0278 · 23 · π(Tc – 20) + 5. and Applications The conductor temperature Tc is calculated from the steadystate heat balance equation: Pj + Ps = Pc + Pr By substitution in the above equation we obtain: 12202 · 1. The converged value of Tc is found to be 80°C. which allows .2.
The temperature of the conductor in the dynamic state is obtained by the solution of the following differential equation (3.22) The above equation is solved by the algorithm shown in the ﬂow chart of Figure 3. 1983).21) ∑ 0 t (P + P + P j s m – Pr – Pc ∆t M ⋅ cp ) + Ti (3.Theory of Transmission Line Ampacity 37 higher than normal static line loading to be allowed on the line for a short duration. Pc + Pr = π · D · ho · (Tc – Ta) Combining joule and magnetic heating.3a. A solution of the above nonlinear differential equation by the Runge Kutte method is given by Black (Black et al. Ti = initial temperature By selecting a suitable time interval ∆t = dt.24) (3. Tav = ∫ 0 t (P + P + P j s m – Pr – Pc dt M ⋅ cp ) + Ti (3.23) { } .. M ⋅ cp dTav = I 2 ⋅ k ⋅ Rdc 20 1 + α 0 (Tav – T0 ) + α s ⋅ D ⋅ Fs – π ⋅ D ⋅ h o ⋅ (Taw – Ts ) dt (3. We deﬁne an overall heat transfer coefﬁcient ho (Dalle et al.20) The nonlinear differential equation can be solved numerically by Euler’s method as follows (Davidson. which is suitable for realtime calculations. Pj + Pm = I2 · Rac = I2 · k · Rdc20 {1 + a0 (Tav – T0)} we obtain. 1979) such that. we can replace the above integral by a summation such that.25) (3.5): M ⋅ cp dTav = Pj + Ps + Pm – Pr – Pc dt (3.. Direct Solution of Dynamic Conductor Temperature A direct solution of the nonlinear differential equation is possible by making some simplifying assumptions to linearize the equation. 1969): Tav = where.
Ta. and Applications START Tc= T i Ti = Tc Input Ws. 1979).3.P c .3a Real time calculation of dynamic conductor temperature. Sr At 1 min intervals T c = Pt. Tch(i) = conductor temperature during heating Tcc(i) = conductor temperature during cooling θ1 = 2 R ac ⋅ I1 (1 – α 0 ⋅ Tref ) + D ⋅ ( Ps + π ⋅ h o ⋅ Ta ) 2 π ⋅ D ⋅ h o – α 0 ⋅ R ac ⋅ I1 (3.38 Powerline Ampacity System: Theory.25) (Dalle et al.P r t = 60s Tc = Conductor temperature Tc = T c+ T i T i = Initial temperature Figure 3..29) The heating time constant is given by.27) (3.a Realtime calculation dynamic conductor temperature FIGURE 3.26) (3.30) . Wd. t Tc = Tc+ T i Pt = Pj + Pm + P s . Conductor Temperature The solution of the linear differential equation (3.28) θ2 = R ac ⋅ I 2 (1 – α 0 Tref ) + D ⋅ ( Ps + π ⋅ h o ⋅ Ta ) 2 2 π ⋅ D ⋅ h o – α 0 ⋅ R ac ⋅ I 2 (3. Tch(i) = θ1 – (θ1 – Tch(i–1) · exp(–∆t/τh) Tcc(i) = θ2 – (θ2 – Tcc(i–1) · exp(–∆t/τh) Where. τh = M ⋅ cp 2 π ⋅ D ⋅ h 0 – α 0 ⋅ R ac ⋅ I1 (3. Modeling.
For elevated temperature operation the speciﬁc heat may be calculated by.34) (3. τc = M ⋅ cp 2 π ⋅ D ⋅ h 0 – α 0 ⋅ R ac ⋅ I 2 (3. hc = Pj2 + Ps π ⋅ D ⋅ ∆Tc 2 (3. generally 20°C or 25°C Ta = ambient temperature.36) (3.33) 2 Pj1 = I1 ⋅ R ac (3. ohm/m Tref = reference temperature. ho = Pj1 + Ps π ⋅ D ⋅ ∆Tc1 (3. cp(Tc) = cp(T20){1 + β(Tc – T20)} β = temperature coefﬁcient of speciﬁc heat capacity. ohm/m cp = speciﬁc heat capacity. m2 D = diameter of conductor. A Flow Chart of Dynamic Temperature method is given in Figure 3. ti – ti–1 A = sectional area of conductor.35) Pj2 = I 2 ⋅ R ac 2 I1 = overload current I2 = post overload current ∆Tcl = Tcl – Ta ∆Tc2 = Tc2 – Ta (3.31) The coefﬁcient of heat transfer during heating is. m α0 = temperature coefﬁcient of resistance. /°C ( Table 1).Theory of Transmission Line Ampacity 39 The cooling time constant is given by.3b and a numerical application is presented in Example 3. . /°C R0 = DC resistance of conductor at reference temperature Tref. °C Rac = AC resistance of conductor.37) Tc1 = steadystate preload conductor temperature Tc2 = steadystate overload conductor temperature ∆t = time step. (J/kg °K) at 20°C.32) The coefﬁcient of heat transfer during cooling is.
Direct Solution of Dynamic Ampacity From Equations (3.28) we may obtain the maximum value of dynamic ampacity approximately as follows: I= {T max C1 {1 − exp( – t τ)} R ac (1 – α 0 Tref ) – Tinitial exp( – t τ)} – C2 (3.38) C1 = π ⋅ γ ⋅ D – I 2 ⋅ R ac ⋅ α 0 D(α s ⋅ Ps + π ⋅ γ ⋅ Ta ) R ac ⋅ (1 – α 0 ⋅ Tref ) (3. °C .40 Powerline Ampacity System: Theory. and Applications Receives message from Rule check dynamic data Calculate Xh Calculateτh Calculate θ1 Calculate Tch(i) Incerment time i = i +1 No Tch(i) >= Tc(max)? Yes Calculate Xc Calculateτc Calculate θ2 Calculate Tcc(i) Incerment time i = i +1 No Time >=120 min? Yes Result: Dynamic Temperature FIGURE 3.27) and (3.3b Flow chart of dynamic temperature method used in the LINEAMPS program.40) I = dynamic ampacity. °C Tinitial = intial conductor temperature. Modeling.39) C2 = (3. A Tmax = maximum conductor temperature.
s t = duration of overload current. kg/m3 A = area. TABLE 3. Formulas for calculation of constants of conductors composed of different material are given in Table 3. kg/m 1 2 M= Constants are calculated at 20°C. July 1995. Deb.Theory of Transmission Line Ampacity 41 τ = conductor heating time constant.1 (Cigré 1999) Constant1 Resistivity. Objectoriented expert system estimates transmission line ampacity. Volume 8. ms are mass density of aluminum and steel respectively. J/(kg · °K) Temperature coefﬁcient of speciﬁc heat. time in the dynamic state is obtained from (3.4. ρ. s A numerical application of dynamic ampacity method is presented in Example 3. Number 3.26) and presented in Figure 3. In this example the analysis is carried out by selecting a typical transmission line Zebra conductor by using the Line Ampacity System (LINEAMPS) software developed by the author. m2 Result of Conductor Temperature in Dynamic State The result of conductor temperature vs. Ω · m Formula2 ρa ρs A a + A s ρ a A s + ρ sd A a ρ= ( ) αaαs Temperature coefﬁcient of resistance. c. . β /° K β= Mass. IEEE Computer Application in Power.4. α /° K α= ρa ρs ρa ρs + + αa + αs Aa As As Aa ρa ρs ρs ρa + + αa + αs Aa As As Aa c= ca m a A a + cs ms As ma Aa + msAs c a m a β a + c s m sβ m m a β a + m sβ s Aa ma + Asms Aa + As Speciﬁc heat.1. s are for aluminum and steel ma.* * Anjan K. Subscripts a.
5 Maximum average conductor temperature = 100°C Steadystate normal conductor temperature at 1260 A = 80°C Solution The overall heat transfer coefﬁcient ho remains fairly constant for a given set of meteorological conditions within a range of conductor temperatures and evaluated by (3.5 Solar absorption = 0.05973 ⋅ 10 –3 ⋅ 1.42 Powerline Ampacity System: Theory. and Applications Dynamic Conductor Temperature: ACSR Zebra Dynamic Temperature ACSR Zebra 120 Temperature. ho = substituting for Pj and Ps we have.37 ⋅ 10 –3 ⋅ (80 – 20) π ⋅ D ⋅ (Tc – Ta ) Pj Ps ho = h o = 25. Example 3.5 ⋅ 30. All other conditions are as follows: Ambient temperature = 20°C Wind speed = 1 m/s Wind direction = 90° Sun = 1000 W/m2 Emissivity = 0.39 ⋅ 10 –3 ⋅ 1000 π ⋅ 30.4 Dynamic conductor temperature as a function of time for ACSR Zebra conductor due to a step change in current.3 Calculate the temperature of ACSR Cardinal conductor when 1475 A overload current is passed through the conductor for 20 minutes.1 ⋅ {1 + 0. 0C 100 80 60 40 20 0 0 20 40 60 80 100 120 Time. ho = I 2 ⋅ R ac + α s ⋅ D ⋅ Fs π ⋅ D ⋅ (Tc – Ta ) 1260 2 ⋅ 0.3 W m 2 ⋅ °C ( ) . Normal load current is 1260 A. min FIGURE 3.33). Modeling.004(80 – 20)} + 0.
–1200 Tav = 106 – (106 – 80) ⋅ exp 822 Tav = 100°C Example 3.3 – . θ1 = we obtain.1.1 ⋅ 14752 –3 τ h = 822 s An average temperature of conductor after 20 minute is obtained from (3.05973 ⋅ 10 –3 ⋅ 1.5 ⋅ 1000 + π ⋅ 25. I= {T max C1 {1 – exp( – t τ)} R ac (1 – α 0 Tref ) – Tinitial exp( – t τ)} – C2 C1 = π ⋅ γ ⋅ D – I 2 ⋅ R ac ⋅ α 0 .39 ⋅ 10 ⋅ 25.3 – .05973 ⋅ 10 –3 ⋅ 1.05973 ⋅ 10 –3 ⋅ 1. Solution The dynamic ampacity is calculated directly from (3. –t Tch ( i ) = θ1 − θ1 − Tch ( i −1) exp τh ( ) where.1 ⋅ 1475 2 R ac ⋅ I 2 (1 – α 0 ⋅ Tref ) + D ⋅ ( Ps + π ⋅ h o ⋅ Ta ) π ⋅ D ⋅ h o – α 0 ⋅ R ac ⋅ I 2 θ1 = 106°C τh = τh = m ⋅ cp π ⋅ D ⋅ ho – α0 ⋅ R0 ⋅ I2 1.004 ⋅ 0. All other conditions are the same as in Example 3.1 ⋅ 1475 2 (1 – 0.004 ⋅ 20) + 30.38).Theory of Transmission Line Ampacity 43 Substituting values in (3.39 ⋅ 10 –3 ⋅ 25.828 ⋅ 826 π ⋅ 30.3 ⋅ 20) π ⋅ 30.4 Calculate the dynamic ampacity of ACSR Cardinal conductor for 20 minutes.26).39 ⋅ 10 –3 ⋅ (0. θ1 = 0.26).004 ⋅ 0.
1 ⋅ 0. Algorithm for the Calculation of Transient Conductor Temperature Transient conductor temperature response due to shortcircuit current is obtained from the solution of the following differential equation: M ⋅ cp dTav = Pj + Pm dt (3. The ﬁnal value of dynamic ampacity is found to be 1475 A.41) .2 ⋅ 20) 1. 3.5 ⋅ 1000 + π ⋅ 25. and Applications Since current I is not known. I= {100 – 80 ⋅ exp( – 1200 822)} – 1.0593 ⋅ 10 –3 (1 – 0.004 ⋅ 20) π ⋅ 25.5 TRANSIENT AMPACITY Transient conditions arise when there is shortcircuit or lightning current.004 C1 = 3.2. Modeling.08 ⋅ 10 3.39 ⋅ 10 –3 ⋅ (0.44 Powerline Ampacity System: Theory. The duration of transient current is generally in the range of milliseconds as most power system faults are cleared within few cycles of the 60 Hz frequency. C1 = 1. During this time. we assume a steadystate current I = 1440 A at 100°C to calculate C1.0593 ⋅ 10 –3 ⋅ (1 – 0.0593 ⋅ 10 –3 ⋅ 0. which is the same as Example 3.2 ⋅ 30.04810 ⋅ 10 5 Substituting in (3. adiabatic condition is assumed (Cigré.1 ⋅ 0.39 ⋅ 10 –3 – 1440 2 ⋅ 1. 1999) when there is no heat exchange with the exterior.38) we obtain the value of dynamic ampacity I.24310 ⋅ 10 –5 C2 = C2 = D ⋅ (α s ⋅ Ps + π ⋅ λ ⋅ Ta ) R ac ⋅ (1 – α 0 ⋅ Tref ) 30.1 ⋅ 0.004 ⋅ 20) C2 = 1.23 ⋅ 10 {1 – exp( – 1200 822)} –5 6 I = 1484 A For greater accuracy we may recalculate C1 with the new value of current I.
Solution From Example 4 we obtain the initial temperature of the conductor to be equal to 80°C when the shortcircuit current is applied.Theory of Transmission Line Ampacity 45 where. 2 Pj + Pm = I sc R ac 1 + α 0 (Tc – T0 ) [ ] During adiabatic condition there is no heat exchange with the exterior therefore. °C αo = temperature coefﬁcient of DC resistance of conductor.5 and a numerical application is shown in Example 5. /°C Rac = AC resistance of conductor at reference temperature To. The following additional data were calculated in Example 4: (3. ohm/m Isc = short circuit current. The conductor was carrying 1260 A steadystate current when the shortcircuit current was applied. α R I 2t 1 1 – exp 0 ac sc Tc = Ti + To – α0 M ⋅ cp Where. J/Kg · °K Equation (3. All other conditions are the same as in Example 4.42) provides the temperature of the conductor during heating by a short circuit current. s To = reference temperature. Example 5 Calculate the temperature of ACSR Cardinal conductor after a shortcircuit current of 50 kA is applied through the conductor for 1 second. °C t = time. Ti = initial conductor temperature. The temperature during cooling of the conductor is obtained from the dynamic equation.42) . A ﬂow chart of the transient ampacity method is shown in the Figure 3. A M = conductor mass. kg/m cp = speciﬁc heat of conductor. Ps = 0 Pr = 0 Pc = 0 The solution of the differential equation is given by.
the temperature of the conductor Tc is calculated as.828 ⋅ 826 ( ) 2 ⋅ 1 Tc = 205°C . –5 3 20 – 1 1 – exp 0. cp = 826 J/(kg ⋅ °C) Rac20 = 6.46 Powerline Ampacity System: Theory. Isc = 50 kA Duration of shortcircuit current.57 ⋅ 10–5 From (3.57 ⋅ 10 ⋅ 50 ⋅ 10 Tc = 80 + .5 Flow chart of transient ampacity method. and Applications Input from transient session window Yes Input data error Post error message No Calculate prefault steady state conductor temperature Calculate Xc Calculate τc Calculate Pj Calculate AA Calculate Tc(i) Calculate Tcc(i) Incremente time i=i+1 No Time > duration of short circuit Yes Update lineplot conductor heating Yes Time > 120 min No Increment time i=i+1 Yes Update lineplot conductor cooling FIGURE 3. Modeling.42 ) we obtain the temperature of ACSR Cardinal conductor for the following condition: Short circuitcurrent.004 ⋅ 6.004 1. t = 1s α R I 2t 1 1 – exp 0 ac sc Tc = Ti + To – α0 M ⋅ cp By substitution of values in the above equation.
6 Transient temperature as a function of time for ACSR Zebra conductor due to a shortcircuit current.C 80 60 40 20 0 0 20 40 60 80 100 Time. Radial temperature gradient is particularly important for highampacity transmission line conductors since they are capable of operating at high temperatures. In this section the radial temperature of the conductor is derived from the general heat equation. Often. dynamic ampacity calculations were carried out by assuming average conductor temperature. For high value of ampacity. The calculation of the radial temperature differential in the conductor is also required for dynamic ampacity calculation.1) we obtain ∂ 2 T 1 ∂T 1 ∂ 2 T ∂ 2 T q( r ) 1 ∂T + = ⋅ + ⋅ + + α ∂t r2 r ∂r r 2 ∂φ 2 ∂z 2 kr (3. millisec FIGURE 3. Result of Conductor Temperature in Transient State Calculated by Program Results obtained by the application of the transient ampacity algorithm by using the LINEAMPS program are presented in Figure 3. substantial radial temperature differences from 1 – 5°C were measured in the wind tunnel.2. 3. the surface temperature of a conductor is available by measurement. From the general heat equation (3. Transient Temperature ACSR Zebra 120 100 Temperature.6. Based on the radial temperature differential an average value of conductor temperature can be estimated. The conductor is ACSR Zebra.5s. Conductor temperature as a function of time is shown by a line graph when a shortcircuit current equal to 50 kA is applied for 0.43) .6 RADIAL CONDUCTOR TEMPERATURE In the previous section.Theory of Transmission Line Ampacity 47 The above example shows the importance of clearing faults by a highspeed fault protection system using modern circuit breakers and protective relaying that can detect and clear faults within a few cycles. and core temperature is required to calculate sag.
∂T =0 ∂φ ∂T =0 ∂z In the steady state. and Applications For a 1mlong cylindrical conductor we may assume that.45) For homogeneous conductors the following boundary conditions are applied.44) (3. ∂ 2 T 1 ∂T q( r ) + + =0 ∂r 2 r ∂r kr where. ∂T =0 ∂t Assuming constant heat generation per unit volume. Modeling. q(r) = q = constant. T(r) = Ts at r = rs rs = conductor radius. m ∂T( r ) = 0 at r = 0 ∂r The solution to (3.44) is then.46) . q= I 2 R ac A al (3. By the application of above conditions we obtain. W/(m ⋅ °K) q is the internal heat generation by unit volume obtained by. r = radial distance from conductor axis. r 2 T( r ) – Ts = rs 2 I 2 R ac 1 – rs (3. m kr = radial thermal conductivity.48 Powerline Ampacity System: Theory.
r 2 I 2 R ac Tc – Ts = s A al 4 k r where.47) For bimetallic conductor (ACSR). T(o) – Ts = ∆T(AAC) = Pj + Pm 4πk r (3. Steadystate solutions of the differential equation are given for the calculation of conductor ampacity and conductor temperature. The AC resistance of ACSR conductors increases with . a differential equation of conductor temperature with respect to time is developed in this chapter. A al = πrs2 A al = Aluminum area. °C r 2 rc rc c 1 – + 2 ln rs rs rs (3. Differential equations are developed for dynamic and transient conditions. The radial temperature differential in the conductor due to the difference in the surface and core temperature is also derived. m 2 Pj + Pm = I2Rac The radial temperature difference from conductor core to surface ∆T in homogeneous conductor is obtained as. °C Ts = conductor surface temperature.3 CHAPTER SUMMARY Starting with a threedimensional transmission line conductor thermal model. the radial conductor differential is obtained by. Algorithms for the calculation of transmission line conductor ampacity and temperature are presented with workedout practical examples.48) 3.Theory of Transmission Line Ampacity 49 Substituting. the boundary conditions are. and their closed form solutions are given. Tc = conductor core temperature. T( r ) = Ts at r = rs ∂T( r ) = 0 at r = rc ∂r By the application of the above boundary conditions for ACSR conductor.
is presented in Appendix 3 at the end of this chapter with a numerical application. and Applications conductor temperature as well as conductor current. Modeling. including magnetic heating and current redistribution in the different layers of the conductor. .50 Powerline Ampacity System: Theory. The calculation of AC resistance of ACSR conductor.
The increase in the AC resistance of ACSR conductors are mainly due to current redistribution in the aluminum wire layers.n. and mutual reactances. 51 . the reactance of each layer of aluminum wire is due to the selfinductance. The longitudinal inductances lead to the longitudinal self reactances Xmm. and the circular inductance. Xcn. of a wire carrying current.. which are calculated as follows (Vincent.. 1986). Lm.1 (Vincent. 1986). the inner and outer circular inductances lead to the inner and outer circular reactances. The current redistribution in the different layers of the aluminum wires are due to longitudinal and circular ﬂux.. Similarly. 1986).i and Xcn.. The circular inductance model assumes that there is 21% contribution due to inner ﬂux. Therefore. H (A/m). M. mutual inductance. due to the longitudinal ﬂux. (Barrett et al.. respectively. Increment in resistance due to magnetic power losses in the steel core The resistance and inductance model of a threelayer ACSR conductor is shown in Figure A1. I. M.1) N = number of turns of aluminum wires over the steel core given by. Lnn. Xm. (Barrett et al. Increment in resistance due to current redistribution 3. and the magnetic power loss in the steel core due to eddy current and hysteresis loss. 1991).. As shown in the ﬁgure.o. Longitudinal Flux The magnetic ﬁeld intensity. DC resistance 2. M. due to the circular ﬂux. the AC resistance of ACSR conductors may be considered to be composed of the following: 1. (Barrett et al. Lc.n. 1991). 1991). ∫ H ⋅ dl = N ⋅ I c (A 1. and 79% contribution by the outer ﬂux of each wire in a layer (Vincent. is given by Ampere’s current law.Appendix 1 AC Resistance of ACSR The AC resistance of conductors having magnetic cores is greater than their DC resistance because of the transformer action created by the spiraling effect of current in the different layers of aluminum wires.
m 1 si (A 1. φn = Bn · An Bn = magnetic ﬁeld. φ= ∫ B ⋅ ds s (A 1.3) Therefore. N= si = lay length of layer i.2) The magnetic ﬂux φ due to the magnetic ﬁeld B (T) is obtained as. Tesla Applying.52 Powerline Ampacity System: Theory. for layer n.4) . the magnetic ﬂux. φn. is. and Applications I I j(X31 I1 X 32 I 2 +X33 I3) R3 I 3 j(X21 I1 +X22 I2 X23 I3) R2 I 2 j(X11 I1 X12I 2 X 23 I3) R1 I 1 Rs Is Xc 3o Xc 3i Xc 2o Xc 2i Xc 1o Xc Resistance of layer n Longitudinal Inductance of layer n Circular Inductance of layer n d DS D1 D2 D3 Steel wire Aluminum wire FIGURE A1.1 Electric circuit model of ACSR. Modeling. B = µH µ = µ0 µr (A 1.
dz (A 1.5) [( ) ] sI n n (A 1.9) By assuming that the layers currents are concentrated at the center of each layer. φ n .n = X n .10) Which has for solution. 2 πfµ 0 πrn2 – A c + µ r A c s 2 n X nn = [( ) ] ] (A 1. n is.6) The selfinductance of layer n. l.n = M n . .m = Circular Flux [( ) ] (A 1.m = [( ) (A 1.outer = ∫∫ N 1 D n –1 ∑I n=0 N n 0 Dn 2 πr µ r µ 0 dr.7) The mutual inductance of layer m.outer ∑I n=0 Dn µ r µ 0 ln 2π Dn – d n (A 1.11) Similarly the inner circular ﬂux due to layer (n + 1) is obtained as. Lnn (µr(al) = 1) is given by. of length. Xnn is. φ n . φn = µ0µrHnAn φ n = µ 0 πrn2 – A c + µ r A c (A 1. is obtained by. µ 0 πrn2 – A c + µ r A c sn ⋅ sq M m .8) And the mutual reactance of layer m. The selfreactance of layer n. n.Appendix 1 AC Resistance of ACSR 53 The ﬂux in layer n is obtained by. the outer circular ﬂux due to layer. 2 πfµ 0 πrn2 – A c + µ r A c sn ⋅ sq X m . n is.
15) D3 – d ( ) .79 I 2 ln D2 – d ( ) D – d D2 + jµ 0 f I s + I1 + I 2 + 0.inner ∑I n=0 N D – d µ r µ 0 ln n +1 2π Dn n (A 1.79 I 2 ln D – d D1 2 ( ) ( ) D3 D – d + jµ 0 f I s + I1 + I 2 + 0. The voltage drop per meter along each layer is given by.21I 3 ln 3 + jµ 0 f I s + I1 + 0. D3 V3 = I 3 R 3 + jI1X 31 – jI 2 X 32 + jI 3 X 23 + jµ 0 f I s + I1 + 0. 1986) that the layer current contributes 21% to the inner ﬂux. 1991).. Modeling.54 Powerline Ampacity System: Theory.13) ( ) ( ) The voltage drop V2 in layer 2 is. and 79% to the outer ﬂux because of current distribution in a wire as shown in Figure A1.12) It was previously shown (Vincent.79 I 3 ln D – d D2 3 (A 1.21I 3 ln 3 + jµ 0 f I s + I1 + I 2 + 0. (Barrett et al.79 I1 ln D1 – d ( ) D – d D2 + jµ 0 f I s + I1 + 0.79 I 3 ln D – d D2 3 (A1.21I 2 ln 2 + jµ 0 f I s + I1 + 0..1. and Applications φ n +1. M. D1 V1 = I1R1 + jI1X11 – jI 2 X12 + jI 3 X13 + jµ 0 f I s + 0.21I 3 ln 3 + jµ 0 f I s + I1 + I 2 + 0.14) ) The voltage drop V3 in layer 3 is. D2 V2 = I 2 R 2 – jI1X 21 – jI 2 X 22 – jI 3 X 23 + jµ 0 f I s + I1 + 0.79 I 3 ln (A 1. The voltage drop V1 in layer 1 is.79 I 2 ln D – d D1 2 ( ( ) ) ( ( ) D3 D – d + jµ 0 f I s + I1 + I 2 + 0.
V R ac = Re I The calculation of the AC resistance of a conductor is carried out iteratively because the complex relative permeability.79 I 3 ln D – d D2 3 (A 1. I1 . µr = [40 – 0. The following relation may be used to calculate complex relative permeability.79 I1 ln D – d 1 Dc ( ) ( ) D – d D2 + jµ 0 f I s + I1 + 0. µr. D − d D2 Vs = I s R s + jµ 0 f I s + 0. V1 = V2 = V3 = Vs (A 1.Appendix 1 AC Resistance of ACSR 55 The voltage drop Vs in the steel core is. is a function conductor current. I 2 . H.17) From the calculated layer currents we obtain the voltage drop. of steel core is a nonlinear function of the magnetic ﬁeld intensity.21I 3 ln 3 + jµ 0 f I s + I1 + I 2 + 0. The AC resistance of the conductor is then found by. V.16) ( ) ( ) Equations (A1. µr. The magnetic ﬁeld intensity.21I1 ln 1 + jµ 0 f I s + 0.18) (A 1.79 I 2 ln D – d D1 2 ( ) ( ) D3 D – d + jµ 0 f I s + I1 + I 2 + 0. which satisfy the following conditions: The sum of layer currents must equal total current I . I 3 .03 · 10–10 H 2] Example 6 Calculate the AC resistance of a 54/7 ACSR Cardinal conductor for the following operating conditions: Conductor current = 1000 A Average conductor temperature = 80°C .0243 H + 0.21I 2 ln 2 + jµ 0 f I s + I1 + 0.16) may be set up as a set of four simultaneous equations with four unknown currents. for H ≤ 1000 A/m. I 2 .000137 H 2] – j[5 + 1. I1 + I 2 + I 3 + I s = I The voltage drop of each layer are equal.13)–(A 1. H.
236 m λ3 = 0.56 Powerline Ampacity System: Theory. D = 30.028126 Ω ⋅ mm2/m Steel resistivity. ρa = 0. 3 layers Aini 1 1 1 1 + + 1 + Rdc1 Rc Rdc 2 Rdc 3 I1 ⋅ Rdc1 Rdc 2 I1 ⋅ Rdc1 Rdc 3 I1 ⋅ Rdc1 Rs I2 = I3 = Is = . d = 3. I1 = {π(D i – d )10 –3 si } i = 1. 2. Modeling. ρi 1 + Rdc i = Initial layer current.219 m λ2 = 0.376 mm Conductor diameter.253 m λ1 = 0. ns = 7 Number of aluminum wires.456 m The dc resistance of Layer i is given by.1775 Ω ⋅ mm2/m The following layer lengths are assumed: λs = 0. and Applications Solution Cardinal conductor data: Number of steel wires.38 mm Aluminum resistivity. nal = 54 Number of aluminum wires in layer 1 = 12 Number of aluminum wires in layer 2 = 18 Number of aluminum wires in layer 3 = 24 Wire diameter. ρs = 0.
V2.21I 3 ln + I s + I1 + 0.21I 3 ln + I s + I1 + I 2 + 0. in Layer 2 is.79 I 2 ln 6 [( ) 8 7 + I s + I1 + I 2 + 0.79 I 2 ln 5 6 ( ( ) ) ( ) 8 9 + I s + I1 + I 2 + 0.21I 3 ln + I s + I1 + I 2 + 0.79 I 2 ln 5 6 ( ) ( ) 9 8 + I s + I1 + I 2 + 0. in Layer 3 is.79 I 3 ln 7 8 ( ) The voltage drop.79 I 3 ln 7 8 ( ) ( ) .79 I 3 ln 8 ( ) The voltage drop. I1 = 327 A I2 = 489 A I3 = 652 A Ic = 31 A The voltage drop in Layer 1 is. 5 V1 = I1R1 + jI1X11 – jI 2 X12 + jI 3 X13 + jµ 0 f I s + 0.79 I1 ln 4 [( ) ) 6 7 + I s + I1 + 0. 7 V2 = I 2 R 2 – jI1X 21 – jI 2 X 22 – jI 3 X 23 + jµ 0 f I s + I1 + 0.21I 2 ln + I s + I1 + 0.21I 3 ln + I s + I1 + I 2 + 0.79 I 2 ln 5 6 ( ( ) ( 8 9 + I s + I1 + I 2 + 0.79 I1 ln 3 4 [( ) ( ) 6 7 + I s + I1 + 0. V3 . 4 5 Vs = I s R s + jµ 0 f I s + 0. 9 V3 = I 3 R 3 + jI1X 31 – jI 2 X 32 + jI 3 X 23 + jµ 0 f I s + I1 + I 2 + 0.79 I 3 ln 7 8 ) ( ) The voltage drop.21I1 ln + I s + 0. in the steel core is.Appendix 1 AC Resistance of ACSR 57 We obtain initial layer currents.21I 2 ln + I s + I1 + 0. Vs .
H= I1 s1 I2 s2 + I3 s3 Ac 2 s1 Ac s1 ⋅ s 2 Ac s1 ⋅ s 3 Ac s2 2 Ac s2 ⋅ s3 Ac 2 s3 – The complex relative permeability. of the steel core is given by.03 ⋅ 10 –10 H 4 ] The sum of all layer currents is equal to total current. µ r = 40 – 0.com/ . http://www.000137 H [ 2 ] – j[5 + 1. X11 = 2 π ⋅ 60 ⋅ µ 0 X12 = 2 π ⋅ 60 ⋅ µ 0 X13 = 2 π ⋅ 60 ⋅ µ 0 X 22 = 2 π ⋅ 60 ⋅ µ 0 X 23 = 2 π ⋅ 60 ⋅ µ 0 X 33 = 2 π ⋅ 60 ⋅ µ 0 The magnetic ﬁeld. and Applications Where.. V1 = V2 = V3 = Vs = 0.58 Powerline Ampacity System: Theory. Inc. µ r .024 * Mathcad 8® is registered trademark of Mathsoft. Modeling. V1 = V2 = V3 = Vs The above problem was solved by Mathcad®* Solver giving initial values and the following results were obtained.109 + j0.mathsoft. is obtained by. H (A/m).0243 H + 0. I1 + I 2 + I 3 + I s = I Voltage drop in each layer is equal.
432 ⋅ 10 –5 The ac/dc ratio.34 A/m2 Layer 3 = 3.127 R dc 6.432 ⋅ 10 –5 = = 1.Appendix 1 AC Resistance of ACSR 59 I1 = 273 – j104 I 2 = 534 + j62 I 3 = 659 + j52 µ r = 131 – j76 Current density: Layer 1 = 2. k2. k1 = I c ⋅ R c + I1 ⋅ R1 + I 2 ⋅ R 2 + I 3 ⋅ R 3 I 2 ⋅ R dc 2 2 2 2 The magnetic power loss factor. and is given by. k1. is obtained by. is obtained by. and a factor k2 due to magnetic power loss in a ferromagnetic core.432 ⋅ 10 –5 I R ac 7. is composed of a factor k1 due to current redistribution in the layers.72 A/m2 Layer 2 = 3. k. V R ac = Re = 7. R ac = k = k1 ⋅ k 2 R dc The current redistribution factor. k2 = k k1 .08 A/m2 The ac resistance of the conductor is given by.
Deb. As much as possible. and Anjan. Vol. Atmospheric conditions of wind speed and ambient temperature were simulated in a wind tunnel that was specially built for these studies. Hall. Consultant.2 WIND TUNNEL EXPERIMENTS* Experiments were carried out at in a wind tunnel to verify conductor thermal modeling for static and dynamic thermal ratings.1 INTRODUCTION The object of this chapter is to present experimental data on transmission line ampacity for the development and validation of theory. Michael Faraday on Electricity 4.” by J. The same * “Wind Tunnel Studies of Transmission Line Conductor Temperatures. I have selected Michael Faraday’s (1834) quotation for this discussion not only for its general applicability to all experimental research.4 Experimental Veriﬁcation of Transmission Line Ampacity I have always endeavoured to make experiment the test and controller of theory and opinion. 61 . K. In Table 4. but also for his particular interest in the subject of electricity and the heating of wires by electric current. showing excellent agreement between measured and calculated values. wind tunnel data is compared to the values calculated by the program. The measured value of steadystate ampacity is 1213 Amperes in the wind tunnel with 2. 90° wind direction. A transmission line conductor was installed in the wind tunnel.1. and assumptions. presented and published in IEEE Transactions in Power Delivery.4 m/s wind. 4. hypotheses. Innova Corporation. and to determine the radial thermal conductivity of conductors. April 1988. the data presented here are from published literature.F. 2. and current was passed through it to study the effects of environmental variables on conductor heating. pages 801–812. and 43°C ambient temperature. The data presented in this chapter is compiled from the different tests that I have either conducted myself.. 3. No. or were conducted by other people in different research laboratories. Paciﬁc Gas & Electric Co.
Southwire is a wellknown conductor manufacturing company in the U. The difference between measurement and calculations are within this range in all of the data presented in Table 4. 1978). and Applications value is calculated by the LINEAMPS program. When higherthannormal transmission line ampacity is allowed through a line. the following empirical relationship between the Nusselt number (Nu) and the Reynolds number (Re) is determined for the calculation of forced convection cooling in conductor: Nu = exp{3. M. showing the placement of the conductor inside the wind tunnel to achieve different wind angles.1.. and the larger propellers were 23 cm in diameter and 30 cm in pitch. Young Co. Wind speed in the range of 0 to 20 mph was generated inside the wind tunnel and measured by propellertype anemometers manufactured by R. PG&E is the largest investorowned electric utility in the U. 1988b) are calculated from the probability distribution of conductor temperature and discussed further in Chapter 5. Modeling.62 Powerline Ampacity System: Theory. A 25hp motor was used to power four 36inch fans at 0 to 960 rpm.091 · ( n Re)2} 1000 ≤ Re ≤ 15000 The results from the above equation are compared to data presented by other researchers in Figure 4. A sketch of the wind tunnel is given in Figure 4. The smaller propellers were 18 cm in diameter and 30 cm in pitch. Vincent Morgan is a leading authority on conductor thermal rating in Australia (Morgan. including the northern and southern hemispheres of the globe. Two sizes of fourblade polystyrene propellers were used for low and high windspeed measurements. we may easily expect 5 to 10% measurement error. In addition to the comparison of wind tunnel data with the program.819 · n Re + 0. When wind speed is zero. EDF is the national electric supply company of France.1.S. the safety factor of the line.2 with excellent agreement.1. 1998). the permanent elongation of the conductor due to creep (Cigré.1 represents a diverse sampling of line ampacity results obtained in the different regions of the world. it is also necessary to evaluate the probability distribution of conductor temperature.96 – 0. (4. For example. data from several other sources are presented to show their excellent agreement with results obtained by the LINEAMPS program. which have different national standards.. 1991).1) . In all of the above examples. From wind tunnel experimental data of conductor temperature at various wind speeds. the measured value of ampacity is 687 Amperes compared to 718 Amperes calculated by the program. The loss of conductor tensile strength (Mizuno et al. and Dr. Due to the inherent uncertainties in the measurement of atmospheric variables. the results obtained from the LINEAMPS program compared well with the data presented in Table 4.S. The data presented in Table 4. and transmission line sag as a function of the life of the line (Hall and Deb.
5 kA 1338 1182 319 880 1160 830 718 1213 826 1366 1390 44. Morgan is author of Thermal Behavior of Electrical Conductors.2) . Carrolton. Conductor Drake Drake Drake Cardinal Cardinal Cardinal Aster 570 Aster 851 Aster 570 Aster 570 Curlew Curlew Sun N Y Y Y N N Y Y Y Y N Y Ta °C 40 40 40 43 43 43 30 15 15 15 0 5 Ws m/s Wd° Tc °C Rating Type Summer Steady Summer Steady Summer Emergency 15 min Summer Steady Measured Steady Measured Steady Summer Steady Winter Steady Winter Dynamic 20 min Transient 1 sec Winter Steady Night Winter Steady Noon Source Amp LINEAMP S Amp 0 1.355 issued August 1999 to Anjan K. an objectoriented expert line ampacity system. Savoullis. April 1988. Patent 5. • Southwire is a trademark of Southwire Company. T. T. J. Authors: J.6 0. • PG&E is the Paciﬁc Gas & Electric Company.3 EXPERIMENT IN OUTDOOR TEST SPAN An outdoor test span is useful for the veriﬁcation of transmission line sag and tension calculated by the LINEAMPS program. • LINEAMPS is Line Ampacity System. • Wind Tunnel data from: “Wind Tunnel Studies of Transmission Line Conductor Temperatures. F. • EDF is Electricité de France. 3. Deb. Degree Celsius Y = Yes. • Vincent T. 4.Experimental Veriﬁcation of Transmission Line Ampacity 63 TABLE 4. published by Wiley.61 0. 1991. Paris. San Francisco.2 0. Degree Celsius Ws = Wind speed. Deb. 2..4 1 1 1 1 0.61 0 2. Anjan K. N = No Source = Name of company or research publication from where data was obtained for this test.1 Ampacity Test Results Source Southwire Southwire Southwire PG&E Wind Tunnel Wind Tunnel EDF EDF EDF EDF Morgan V.5 kA 1324 1233 Notes: Ta = Ambient temperature.933.S. meter per second Wd = Wind direction. Degree Tc = Conductor surface temperature.6 90 90 90 90 90 90 90 90 90 90 50 75 100 80 78. The computer program for the calculation of sag and tension uses the following transmission line conductor change of state equation: σ 2 (ϖ ⋅ L)2 σ (ϖ ⋅ L)2 + α(Tc 2 – Tc1 ) + ∆Ec = 1 – – 2 2 E 24σ 2 E 24σ1 (4. Morgan V. U. Hall. Inc. GA. Vol.” IEEE Transactions on Power Delivery.1 75.8 60 60 75 150 80 80 320 880 1160 838 687 1213 830 1350 1393 44. No. New York.
Modeling. 25 HP Motor Four 36 Fans 0 to 960 RPM Wind Flow Pivot point for conductor angle Anemomet .64 Powerline Ampacity System: Theory. pipe 6 long 8 . °C E = Young’s modulus of elasticity.A . σ1. kg/mm2 Nusselts Number. Plan View FIGURE 4.1 Wind tunnel. σ2 = stress at state1 and state2 respectively. . where. Tc2 = conductor temperature at state1 and state 2.. kg/mm2 Tc1. SECTION A .. and Applications WIND TUNNEL 3 . Re HallDeb JohannetDalle Morgan 5 FIGURE 4.2 Forced convection Nusselt number vs Reynolds number relationship obtained from wind tunnel experiments and comparison with results from other researchers. 4 Venturi PVC . N A Conductor . 15 6 90o Wind Flow A Wind Straightner 23 .. FORSED CONVECTION HEAT TRANSFER Nusselt Number vs Reynolds Number 200 180 160 140 120 100 80 60 40 20 0 500 1000 2000 5000 10000 15000 30000 Reynolds Number.
* Thanks are due to Mr.40 Safety Factor 2. Sun which shows the result of this comparison. France (Deb.00 7. June 1990. The results of this experiment are presented in Table 4. m ∆Ec = inelastic elongation (creep) mm/mm α = coefﬁcient of linear expansion of conductor. the transmission line conductor was energized by a 100 kVA transformer.3. Pa 1480 – – LOS %RTS 4 4 4 Creep µstrain 1200 1200 1200 Tension kN 73.2.1m @100 °C As shown in Figure 4. °C–1 Results obtained by the application of above equation are presented in Table 4.00 7.56 19. . and the temperature of the conductor was controlled by varying the current passing through it. 1997).12 10.47 Sag m 2. A report was submitted to Mr. CA. These experiments were conducted at the EDF laboratory at Paris. for providing transmission line sag and tension data .72 Life Year 50 50 50 1 Pascal (Pa) = 0. °C Wind = Wind pressure on projected area of conductor.2 Veriﬁcation of Transmission Line Security (ACSR Cardinal Conductor) Tc °C 15 80 100 Wind. kg/m/mm2 L = span length.85 10. The loss of strength of aluminum alloy wires was determined experimentally by heating individual wires at elevated temperatures. TABLE 4.03 20. Wally Sun. Transmission Line Engineer.02 lbf/ft2 1 kN (Kilo Newton) = 224.8 lbf mstrain = micro strain = mm/km RTS = Rated Tensile Strength Tc = Average conductor temperature. 1978). The sag and tension program is further veriﬁed by comparison with ﬁeld data from various electric power companies* with excellent agreement (Wook.Experimental Veriﬁcation of Transmission Line Ampacity 65 ϖ = speciﬁc weight of conductor. Pa LOS = Loss of Strength SF = Safety Factor of conductor Initial sag after stringing = 9.3. PG&E. San Francisco. and Deb. Choi. Conductor sag at midspan was measured by a scale which compared well with the sag calculated by the program.
65 4.3 test span. °C 150 150 150 150 150 130 130 Duration.47 Note: Aluminum alloy wire size is 3. Modeling.3 Loss of Strength of Aluminum Alloy Wires at Elevated Temperatures Conductor Temperature.35 24.06 1. IEEE Standard for calculating the currenttemperature relationship of bare overhead conductors.45 mm in diameter.5 m 4m 100m 220/100 + ~ + + 220v/50Hz/20 Deg Variable Transformer 100kVA FIGURE 4. .66 Powerline Ampacity System: Theory. % 32.1 STEADYSTATE AMPACITY The IEEE* recommends a standard method for the calculation of currentcarrying capacity of overhead line conductors based on theoretical and experimental research * IEEE Standard 7381993.40 2.80 2. and Applications Conductor Insulator 1. 4. hr 100 20 5 4 2 10 10 Loss of Strength. Test setup of high temperature conductor sag measurement in outdoor TABLE 4.4.71 7.4 COMPARISON OF LINEAMPS WITH IEEE AND CIGRÉ 4.
Similarly. Comparision of Ambient Temperature Effects COMPARISON OF LINEARAMPS. The two methods of ampacity calculation were compared by PG&E engineer N. and ambient temperature are the same all along the transmission line route. 1997). Schmidt. wind direction. 2. has proposed a method for calculating the thermal rating of overhead conductors. The assumptions regarding the transmission line and the various meteorological conditions are presented in Table 4. 738. Deb. sky condition.Experimental Veriﬁcation of Transmission Line Ampacity 67 carried out by several researchers. Section 1. Comparison between IEEE and Cigré ampacity standards. They do not include * The thermal behaviour of overhead conductors.9. 1992). CIGRE ACSR DRAKE AMPASITY (Ambient Temperature Effects) 1300 1250 1200 1150 Ampacity. ** N. 1993) and Cigré (1997. The comparisons are based on steadystate conditions only.4 from the IEEE paper. October 1997. It is important to note that IEEE and Cigré provide methods to calculate line ampacity when the ambient conditions are given.4. and 3. Section 3. In this section. .12. Anjan K. LINEAMPS results are compared to the results given by Schmidt (1997) in Figures 4. Figure shows the variation of conductor ampacity as a function of ambient temperature. IEEE Power Engineering Society conference paper # PE749PWRD0061997.** and the results of this comparison were presented in a 1997 IEEE paper (Schmidt.P. the international conference on large electrical networks. October 1997. October 1992. Report prepared by Cigré Working Group 22. P. A 1100 1050 1000 950 900 850 800 10 20 30 40 Ambient Temperature. Discussion contribution to this paper.4–4. It is appropriate to mention here that these comparisons were made on the assumptions that the meteorological conditions comprised of wind speed. The study shows that there may be up to 10% variation in the two methods of ampacity calculation. Cigré* (Conférence International de la Grande Réseaux Electrique). C LINEAMPS IEEE CIGRE 50 FIGURE 4. Schmidt. IEEE. Electra.4 Ampacity calculated by LINEAMPS program is compared to the IEEE Standard and Cigré method of calculating conductor thermal rating in the steady state. All other assumptions are speciﬁed in Table 4. These results show that the values calculated by LINEAMPS are within 10% of those of the IEEE (IEEE Std. Electra. Section 1 and 2.
6 Ampacity calculated by LINEAMPS program is compared to the IEEE Standard and Cigré method of calculating conductor thermal rating in the steady state. IEEE. and Applications COMPARISON OF LIMEAMPS. methods for modeling variations in meteorological conditions along the transmission line route.4. Figure shows the variation of conductor ampacity as a function of wind direction. All other assumptions are speciﬁed in Table 4. Degree 70 LINEAMPS IEEE CIGRE 80 90 FIGURE 4. Modeling. ft/s FIGURE 4. COMPARISON OF LINEAMPS. 1998). Figure shows the variation of conductor ampacity as a function of wind speed.68 Powerline Ampacity System: Theory.4. All other assumptions are speciﬁed in Table 4. as stated in a discussion contribution recently prepared by this author (Deb. A 900 850 800 750 700 650 600 0 10 2 30 50 60 40 Wind Direction. IEEE.5 Ampacity calculated by LINEAMPS program is compared to the IEEE Standard and Cigré method of calculating conductor thermal rating in the steady state. The different meteorological conditions along the transmission line route are considered in a unique manner by the LINEAMPS program. Faraday also . CIGRE ACSR DRAKE AMPACITI (Wind Speed Effectc) 1600 1400 1200 Ampacity. CIGRE ACSR DRAKE (Wind Direction Effects) 1100 1050 1000 950 Ampacity. A 1000 800 600 400 LINEAMPS 200 0 0 1 2 3 4 5 6 IEEE CIGRE 7 8 Wind speed.
“On the absolute quantity of electricity associated with the particles or atoms of matter. However. The effect of solar radiation on conductor ampacity is shown as a function of time of day. min 30 LINEAMP CIGRE FIGURE 4. IEEE.5 m/s 1400 1300 Dynamic Ampacity. Diffused radiation was neglected in the comparison made in the IEEE paper. All other assumptions are speciﬁed in Table 4. page 295. CIGRE ACSR DRAKE AMPACITY (Solar Effects) 1100 1080 1060 1040 Ampacity. A 1200 1100 1000 900 800 700 600 500 400 3 10 Interval of Overload. LINEAMPS considers both direct beam and the diffused solar radiation hence the predicted ampacity is slightly lower than IEEE and Cigré. Great Books # 42. A 1020 1000 980 960 940 920 900 10 11 12 13 14 LINEAMPS IEEE CIGRE Time of Day.5 m/s. realized the problem of changing cooling effects on wire when different parts of the wire are exposed to different cooling conditions when he stated:* * Michael Faraday on Electricity. January 1834. Comparison of Dynamic Ampacity (Wind Speed Effect: 0. hr FIGURE 4. the effect of solar radiation on line ampacity is comparatively small when compared to the effects of ambient temperature and wind.Experimental Veriﬁcation of Transmission Line Ampacity 69 Comparison of Solar Effects COMPARISON OF LINEAMPS.7 Ampacity calculated by LINEAMPS program is compared to the IEEE Standard and Cigré method of calculating conductor thermal rating in the steady state.8 Ampacity calculated by LINEAMPS program in dynamic state is compared to Cigré method of calculation of dynamic ampacity when wind speed is 0.5 m/s) COMPARISON OF DYNAMIC AMPACITY Wind Speed = 0. .” Encyclopedia Britannica.
70 Powerline Ampacity System: Theory. Modeling.9 Ampacity calculated by LINEAMPS program in dynamic state is compared to Cigré method for the calculation of dynamic ampacity when wind speed is 2 m/s. . Comparison between IEEE and Cigré ampacity standards. provided the cooling circumstances are the same for every part in all cases. October 1997. or any length of the same wire to the same degree. a thousand. min 30 FIGURE 4. and Applications Comparison of Dynamic Ampacity (Wind Speed Effect: 2m/s) COMPARISON OF DYNAMIC AMPACITY Wind Speed = 2m/s 2000 1800 Dynamic Ampacity. TABLE 4.P. Deb.4–4. Anjan K. A 1600 1400 1200 1000 800 LINEAMPS 600 CIGRE 400 3 10 Interval of Overload. Discussion contribution. can heat an inch of platina wire of a certain diameter redhot can also heat a hundred. IEEE Power Engineering Society conference paper # PE749PWRD0061997. Schmidt.4 Data for Line Ampacity Calculations Presented in Figures 4.9 Transmission Line Conductor Wind Speed Wind Direction Latitude Azimuth of Conductor Atmosphere Solar Heating Diffuse Solar Radiation Emissivity Absorptivity Elevation above Sea Level Ground Surface Type Time of Day Time of Year Maximum Conductor Temperature 795 kcmil 26/7 ACSR Drake 2 ft/s Perpendicular to Line 30 ° 90° Clear On 0 (ignored in IEEE & Cigré).5 0.5 0m Urban 11:00 am June 10 100°C Source: N. The same quantity of electricity which. considered in LINEAMPS 0. passed in a given time.
and the meteorological conditions cannot be expected to remain the same everywhere. The static line ampacity is 1000 A. by introducing the concept of virtual weather sites.10. * LINEAMPS User Manual.5 MEASUREMENT OF TRANSMISSION LINE CONDUCTOR TEMPERATURE 345 kv Transmission line The results obtained from the LINEAMPS program were also veriﬁed by comparison with the ampacity of a real transmission line by measurement. ** The thermal behaviour of overhead conductors. These results clearly indicate that the program safely and reliably offers substantial increase in line ampacity over the present method of static line rating. In addition. Electra. The ampacity predicted by LINEAMPS offers substantially higher line capacity than the present method of static line rating.9. Cigré Working Group 22.2 DYNAMIC AMPACITY In the dynamic state. It also accurately predicted the lowest value of line ampacity at noontime.Experimental Veriﬁcation of Transmission Line Ampacity 71 In an overhead power transmission line the cooling effects are generally not the same at all sections of the line because of its length. due to varying meteorological conditions in space and in time.10. 4. The results of this comparison are presented in Figure 4. As seen in Figure 4. LINEAMPS ratings never exceeded measured (ComEd) ampacity at different hours of the day.10. shortterm overload currents greater than steadystate ampacity are allowed on a transmission line by taking into consideration the energy stored in a transmission line conductor. as shown in Figures 4. by objectoriented modeling of transmission lines and weather stations. The data was compared to the values calculated by a LINEAMPS Dynamic model with excellent agreement. 1998. The energy stored in a transmission line conductor is shown by the differential equation (3. Section 3: Mathematical model for evaluation of conductor temperature in the unsteady state. LINEAMPS takes into consideration the different cooling effects on the transmission line conductor. October 1997. as shown in Figure 4.8 and 4. showing that the ampacity of the transmission line calculated by the LINEAMPS program never exceeded the measured values at all locations during daytime for the period considered in the study.12 Report.5) in Chapter 3. . Section 3 of a recent Cigré report** presents data on dynamic ampacity.* and by expert rules described in Chapter 8. Measurements were made by temperature sensors installed on various locations of a 345 kV overhead transmission line operated by the Commonwealth Edison Company (ComEd) in the region of Chicago.4. A transmission line may be 10 or 100 miles long (or greater). IL. 4.
. hr FIGURE 4. and the correctness of various assumptions are validated by verifying results obtained by program with experimental data. according to Faraday. USA. A 2000 1500 1000 500 0 8:00 9:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 ComED LINEAMPS STATIC Time of Day. The results of these comparisons show that line ampacity calculated by program is in good agreement with actual data from the ﬁeld. and Applications COMPARISON OF LINEAAMPS WITH MEASURED TRANSMISSION LINE AMPASITY 3500 3000 2500 Ampacity. Therefore. Modeling.10 Transmission line ampacity calculated by LINEAMPS program is compared to the ampacity measured on a real 345 kV overhead transmission line operated by Commonwealth Edison Company in the region of Chicago.72 Powerline Ampacity System: Theory. the proposed theory of transmission line ampacity. as well as transmission line conductor manufacturers’ catalog data. IL. there is excellent agreement with the results obtained by program. conductor thermal models. 4.6 CHAPTER SUMMARY In this chapter the results calculated by the LINEAMPS program are compared to experimental data from different power company data. The calculations are also compared to data presented at the IEEE and Cigré conferences by several researchers. hypotheses. In all of these comparisons.
The line ampacity system will ensure that the allowable normal and emergency operating temperatures of the conductor are not exceeded. 1998). This chapter includes a study of transmission line conductor sag and tension. consequently. Canadian Electricity Association Research Report. Morgan. 1985. While there are signiﬁcant beneﬁts to increasing transmission line ampacity.5 Elevated Temperature Effects 5. 1972. As stated in the previous chapters. 1978. Deb et al. and did not include the design sag and tension of conductor.1 INTRODUCTION The advantages of higher transmission line ampacity discussed in Chapter 1 include the deferment of the capital investment required for the construction of new lines and economic energy transfer. An elegant method to calculate sag and tension by a strain summation procedure is described in a report prepared by Ontario Hydro. and the loss of tensile strength of the powerline conductor due to elevated temperature operation. * Development of an accurate model of ACSR conductors at high temperatures. Cigré. 73 . In this chapter the effects of higher transmission line ampacity are evaluated from the point of view of elevated temperature operation of conductors. Mizuno. permanent elongation. and that any additional conductor sag caused by permanent elongation will not exceed design sag and tension during the lifetime of the transmission line conductor.* The unique contribution made in this chapter is the development of an uniﬁed approach to determine sag and tension during the lifetime of a transmission line conductor by consideration of the probability distribution of transmission line conductor temperature. there may be greater loss of tensile strength of conductor and higher sag. The problem of electric and magnetic ﬁelds due to higher ampacity are presented in Chapter 6.. the main objective of the powerline ampacity system is to accurately predict transmission line ampacity based upon actual and forecast weather conditions. its effects must be clearly understood and evaluated accurately. As a result of achieving higher line ampacity. The Cigré report did not describe how the temperature distribution was obtained. 1978. which must be evaluated properly. The loss of tensile strength and permanent elongation of a conductor is calculated recursively from a speciﬁed conductor temperature distribution by using the empirical equations found in the literature (Harvey. electricity costs are reduced and there is less environmental impact. The line ampacity system program should also verify that the loss of tensile strength of a conductor is within acceptable limits. When line ampacity is increased conductor temperature increases.
74 Powerline Ampacity System: Theory. For example.* The effects of elevated temperature operation of conductors comprising inelastic elongation and the loss of tensile strength of conductor are considered by the recursive formulation of inelastic elongation and annealing models found in the literature. but the resulting conductor temperature distribution was not given. 1985.” Part I. 9. Columbus. A line security analysis was carried out (Hall. Not much attention has been given to probabilistic design of sag and tension of overhead line conductors. The probability distribution of conductor temperatures is obtained by the synthetic generation of meteorological data from timeseries stochastic and deterministic models. Probabilitybased transmission line rating methods are described by several authors (Koval and Billinton. There is considerable interest in the industry in the probabilistic design of overhead lines. and Applications 5.. Modeling. conductor temperature is never constant. IEEE Transactions on Power Apparatus and Systems.1.F. Urbain. No.. Redding*** 1993 presented probability models of ambient temperature and wind speed. “Probabilistic Design of Transmission Lines. sag based on the probability distribution of conductor temperature is required. E. This method of generating probability distribution of conductor temperatures takes into account the correlation between the meteorological variables and the transmission line current. Ohio. “IEEE/PES 1993 Winter Meeting Conference Paper # 93WM 0778PWRD. 1988) based upon different line operating conditions. 1986). January 31–February 5. The probability distribution of conductor temperature is a * See discussion contribution by J. In reality. Results are presented that show good agreement with data from other computer programs. This study showed how conductor sag and tension varies with conductor temperature frequency distributions. Hall and Anjan K. 1998).000 hours. .** Ghanoum (1983) described a method for the structural design of transmission lines based upon probabilistic concepts of limit loads and return period of wind. 1993. Redding. conductor sag at 100 °C cannot be expected to remain the same if it has been operated at that temperature for 100 hours or 10. Deb et al. Morgan. as we all know. ** Ghanoum. 1970. Deb. PAS102. 5. II. 1993. Vol. 1993. Deb on the IEEE paper (Douglass. 1983.2 TRANSMISSION LINE SAG AND TENSION — A PROBABILISTIC APPROACH A method of calculation of conductor sag and tension is presented in this section by consideration of the probability distribution of transmission line conductor temperature in service. Redding. which depend upon conductor temperature probability distributions. The equations and algorithm that were used to calculate conductor sag and tension from the probability distribution of conductor temperatures are presented and implemented in a computer program. Therefore. 1991. “A Method for Determining Probability Based Allowable Current Ratings for BPA’s Transmission Lines.L.1 EXISTING PROGRAMS Most sagtension computer programs presently used are based on the assumption that conductor temperature will remain constant for the entire life of the line. *** J. Conductor sag and tension are important transmission line design parameters upon which depend the security of the line.
z(t2) are the stochastic variables at lag 1 and lag 2 respectively η(t) = uncorrelated white noise By this method it is possible to take into consideration timeofday effects of weather and line current in the analysis.2. The conductor temperature distribution is also required to calculate the loss of strength of the conductor. In this chapter I have considered transmission line sag and tension as the determining factor. 1979). Sin(ωt). 1988. This requires an estimate of the conductor temperature distribution during the expected life of the line. This includes both the loss of strength and the permanent elongation of the conductor. 5. or by taking a typical set of weather data and assuming it to repeat itself every year (Giacomo et al. Sr = solar radiation (Sr is also calculated analytically by the method given in Chapter 7). In the method proposed by the author (Deb.. meteorological data is generated by Monte Carlo simulation of the following time series stochastic and deterministic model: Y(t) = X(t)T · A(t) + η(t) Y(t) ∈ (Ta. Sr) input variables* X(t)T = {1. 5. 1998. Morgan. it is required to calculate the sag and tension of the conductor at different temperatures. Mizuno et al.1) * Ta = ambient temperature. Using real weather data in chronological order allows (5. 1991.Elevated Temperature Effects 75 function of conductor current and the meteorological conditions on the line (Hall and Deb. Mizuno et al. 2000). z(t2)} A(t) = model coefﬁcients ω = fundamental frequency z(t1).1 THE TRANSMISSION LINE SAGTENSION PROBLEM Given the maximum mechanical loading of a conductor due to wind and ice.2 METHODOLOGY In order to predict conductor sag at the highest conductor temperature. Cos(2ωt). The probability distribution of conductor temperature is obtained from line ampacity simulations. z(t1). Sin(2ωt). Future conductor temperature distributions require knowledge of the line current as well as the meteorological conditions. Transmission line sag and tension are critical line design parameters that are required to verify conductortoground clearance and the safety factor of the conductor at the maximum working tension.. 1993). and a probable distribution of conductor temperature with time. Nu. considered the loss of strength of conductor as the index of thermal deterioration. . Future meteorological conditions may be estimated by the random generation of weather data from their speciﬁed probability distributions. the permanent elongation of the conductor due to metallurgical creep is required to be estimated. Nu = Nusselt number (coefﬁcient of heat transfer).2. Cos(ωt).
MonteCarlo Simulation of Ambient Temperature 45 Temperature. Modeling. Degree C 40 35 30 25 20 15 10 5 0 0 24 48 72 Hour 96 120 144 168 FIGURE 5.. (Deb et al. The conductor temperature distribution of Figure 5. Examples of synthetic generations of meteorological data from time series stochastic models are shown in Figures 5. and Applications this also. MonteCarlo Simulation of Heat Transfer Coefficient 120 100 Nusselt Number 80 60 40 20 0 0 24 48 72 Hour 96 120 144 168 FIGURE 5. 1985).76 Powerline Ampacity System: Theory..1 Hourly values of ambient temperature generated by Monte Carlo simulation for the San Francisco Bay area. Timeofday effects are lost when weather data is generated from known probability distributions (Mizuno et al. . AAC Bluebell conductor temperature distribution is shown in Figure 5.2 Hourly values of conductor heat transfer coefﬁcient of AAC Bluebell transmission line conductor generated by Monte Carlo simulation for the San Francisco Bay area.5 which was obtained by the synthetic generation of California meteorological data from time series stochastic models. 1998).5 assumes constant line current equal to the static line rating and is used to calculate transmission line conductor sag and tension.4.1–5.
3 COMPUTER PROGRAMS The following computer programs are required for the prediction of sag and tension at high temperature based upon the probabilistic distribution of conductor temperature: 1. Conductor temperature predictor 2. Inelastic elongation (creep) predictor 5. Degree C 70 60 50 40 30 20 10 0 0 24 48 72 96 Hour 120 144 168 FIGURE 5.3 Hourly values of solar radiation on conductor surface of AAC Bluebell transmission line conductor in the San Francisco Bay area simulated by program. 5. W/m2 1200 1000 800 600 400 200 0 0 24 48 72 96 Hour 120 144 168 FIGURE 5. Probability distribution generator of conductor temperatures 4.Elevated Temperature Effects 77 Solar Radiation Simulation 1400 Solar Radiation. solar radiation. Probability distribution generator of ambient temperature. and conductor heat transfer coefﬁcient from timeseries stochastic models 3. line current.4 Hourly values of conductor temperature of AAC Bluebell transmission line conductor in the San Francisco Bay area using simulated weather data.2. Loss of strength predictor 6. Conductor Temperature Simulation 80 Temperature. Sag and Tension Calculator .
5. Bonicel. Paris. The equations and algorithms are developed in the following sections. O. W/m 2 120 Nusselt Number 100 80 60 40 20 0 Solar Radiation Simulation 45 40 35 30 25 20 15 10 5 0 0 24 48 72 96 120 144 168 Hour Temperature. Y. % 15 10 5 0 10 15 20 25 30 35 40 45 50 55 60 65 70 75 Temperature. March 1998.78 Powerline Ampacity System: Theory. The above modules constitute the Sag and Tension Program developed by the author. REE No. Porcheron. Lignes Aeriennes. If the conductor is at State 1 given by conductor stress σ1 and temperature Tc1 and goes to State 2 given by stress σ2 and temperature Tc2.5 Frequency distribution of AAC Bluebell transmission line conductor temperature in the San Francisco Bay area simulated by program. Degree C Transmission Line Sag & Tension FIGURE 5. . MonteCarlo Simulation of Ambient Temperature Heat Transfer Simulation Solar Radiation. % 15 10 5 0 10 15 20 25 30 35 40 45 50 55 60 65 70 75 Temperature.7. Techniques de l’Ingenieur. then the new sag and tension of the conductor at State 2 is calculated from the following change of state equation: * P. A ﬂow chart for the calculation of transmission line sag and tension from the probability distribution of conductor temperature is given in Figure 5.3 CHANGE OF STATE EQUATION* The change of state equation given below is used for the calculation of transmission line sag and tension. 3. Tatat. J. Aerial optical cables along electrical power lines. Degree C FIGURE 5.7 Flow chart for the calculation of transmission line conductor sag and tension from probability distribution of conductor temperature. Modeling. Degree C 1400 1200 1000 800 600 400 200 0 0 24 48 72 96 Hour 120 144 168 0 24 48 72 96 Hour 120 144 168 Conductor Temperature Simulation 80 Temperature. SEE France. and Applications Conductor Temperature Frequency Distribution 20 Frequency.P. Hautefeuille. Degree C 70 60 50 40 30 20 10 0 0 24 48 72 Hour 96 120 144 168 Conductor Temperature Frequency Distribution 20 Frequency.
Elevated Temperature Effects
79
σ 2 (ϖ ⋅ L)2 σ (ϖ ⋅ L)2 + α(Tc 2 – Tc1 ) + ∆Ec = 1 – – 2 E 24σ 2 E 24σ1 2 σ1, σ2 = stress at State1 and State2 respectively, kg/mm2 Tc1, Tc2 = conductor temperature at State1 and State2, °C E = Young’s modulus of elasticity, kg/mm2 ϖ = speciﬁc weight of conductor, kg/m/mm2 L = span length, m ∆Ec = inelastic elongation (creep) mm/mm α = coefﬁcient of linear expansion of conductor, °C–1
(5.2)
Conductor sag is calculated approximately by the following wellknown parabola equation: Sag = where, Sag is in meters, m W = conductor weight, kg/m T = conductor tension, kg The above equation is used to calculate transmission line conductor sag and tension. It requires a knowledge of Young’s modulus of elasticity, the coefﬁcient of linear expansion of the conductor and the permanent elongation of the conductor due to elevated temperature creep. Young’s modulus is obtained from the stress/strain curve shown in Figure 5.6. The coefﬁcient of linear expansion is a property of the conductor. The elevated temperature creep Ec (metallurgical creep) is estimated separately by the creep predictor program.
B A Stress
WL2 8T
(5.3)
O
E1
E2
Strain
FIGURE 5.6 Stress/strain relation of AAC conductor
5.3.1
RESULTS
The probability distribution of conductor temperature is shown in Table A5.1 in Appendix 5 at the end of this chapter, the loss of strength is given in Table A5.2, and the permanent elongation of conductor during the lifetime of a transmission line
80
Powerline Ampacity System: Theory, Modeling, and Applications
conductor is presented in Table A5.3. Calculations are based upon the conductor temperature distributions generated from Figure 5.5. The sag and tension of transmission line conductors is presented in Table A5.6 of Appendix 5. Comparison of sag and tension results with other programs are given in the Tables A5.8 through A5.10. Further results of transmission line sag and tension of special conductors are given in Table A5.11 (Choi et al., 1997). The study was prepared by the author with KEPCO* for line uprating with highampacity conductors. The results obtained by program show excellent agreement with ﬁeld measurements on an actual transmission line in Korea.
5.3.2
CONDUCTOR STRESS/STRAIN RELATIONSHIP
The stres/strain relationship of an all aluminum conductor is shown in the Figure 5.6 for the purpose of illustration of some of the concepts discussed in this section. The stress/strain relationship of an ACSR conductor is somewhat complicated, though the general concepts remain the same for any type of conductor. When tension is applied to an unstretched conductor, the ratio stress/strain of the conductor follows the curve OA. When the tension is lowered at Point A, this ratio becomes linear and follows the trace AE1. If the tension is increased again it follows the linear path E1A until it reaches Point A. If the tension is further increased at Point A it becomes nonlinear again, as shown by the curve AB. When tension is lowered again at Point B, it follows the linear path BE2. In Figure 5.6, the section of the stress/strain curve OA and AB represents the initial stress/strain curve, which is nonlinear. Consequently, the Young’s modulus in this region becomes nonlinear and is generally approximated by ﬁtting a polynomial function of degree N to the data. The ﬁnal modulus of elasticity given by the slope of the linear portion of the curves AE1 and BE2 is constant. The sections OE1 and E1E2 are the permanent elongation due to creep (geometrical settlement). The advantage of pretensioning the conductor becomes obvious from Figure 5.6. The permanent stretch OE1 and E1E2 can be removed if the conductor is pretensioned by a load to reach Point B close to the allowed maximum working tension of the line. The stress/strain curve then becomes linear.
5.4 PERMANENT ELONGATION OF CONDUCTOR
Permanent or irreversible elongation of the conductor is known to occur due to elevated temperature operation of the conductor. It results in increase of conductor sag and reduces the midspan clearance to ground. Elevated temperature creep is a function of the conductor temperature, its duration, and the conductor tension. Two factors cause permanent elongation of the conductor (Cigré, 1978): 1. Geometric settlement 2. Metallurgical creep
* Korea Electric Power Company.
Elevated Temperature Effects
81
5.4.1
GEOMETRIC SETTLEMENT
Geometric settlement depends upon conductor stringing tension and occurs very rapidly as it only involves the settling down of strands. Generally, the process starts with conductor stringing and is completed within 24 hours. Higher than normal stringing tensions, “pretension,” is sometimes applied to a conductor to accelerate the process of geometric settlement. The geometric settlement, Es, is calculated by the following formula (Cigré, 1978): Es = 750(d – 1) (1 – exp(–m/10)) (MWT/RTS)2.33 d = wire diameter, mm m = aluminum/steel sectional area, ratio MWT = Maximum Working Tension, kg RTS = Rated Tensile Strength, kg (5.4)
5.4.2
METALLURGICAL CREEP
Metallurgical creep is a function of conductor temperature, tension, and time. Therefore, elevated temperature operation of a line for short duration is not as much of a concern as continuous operation at high temperatures. Metallurgical creep is estimated by using the following empirical formula determined experimentally (Cigré, 1978): Ec = 1 cos
2 +α
β
K ⋅ exp(φTc )σ α t µ
σδ
(5.5)
Ec = elongation, mm/km K,φ,α,µ, δ are constants (Table 5.1 gives values for typical ACSR conductor sizes) Tc = average conductor temperature, °C σ = average conductor stress, kg/mm2 t = time, hr The factor β takes into account the effect of conductor type, stranding and material and is calculated as follows,
β=
∑n β
i i =1 N
N
i
∑n
i =1
(5.6)
i
N = number of aluminum strands ni = number of wires in layer i βi = angle of the tangent in a point of a wire in layer i with conductor axis
82
Powerline Ampacity System: Theory, Modeling, and Applications
5.4.3
RECURSIVE ESTIMATION
OF
PERMANENT ELONGATION
The inelastic elongation of a conductor due to metallurgical creep, Ec, is calculated in small steps, ∆Ec, by the following recursive equations: Eci,j = Eci,j1 + ∆Ec tq i. j = Ec i, j–1K ′σ i–,α exp( – φTc i ) j (5.7)
{
}
– µ σδ
(5.8)
i = 1,2… n line loading intervals obtained from the frequency distribution of Table A5.1
TABLE 5.1 Value of coefﬁcients in equation (5.5) (Cigré, 1978)
ACSR Al/St 54/7 30/7 K 1.1 2.2 f 0.0175 0.0107 a 2.155 1.375 m 0.342 0.183 d 0.2127 0.0365
j = 1,2… f sub intervals tqi,j = equivalent time at present temperature Tci for the past creep Eci,j1 K′ = Total creep is estimated by summation, Etotal = Ecn,f + Esf (5.10) Cos 2 +α β K (5.9)
Ecn,f and Esf are the ﬁnal inelastic elongation due to metallurgical creep and geometrical settlement of the transmission line conductor. Results are presented in Table A5.3 of Appendix 5 for the frequency distribution of Table A5.1. When Table A5.3 values are used to calculate sag, the results are given in Table A5.6. When ﬁnal sags are compared to the initial conditions speciﬁed in Table A5.5, it is seen that the sag of the AAC Bluebell conductor increases by 5.9 ft (14.4% increase) over initial conditions after 30 years. When maximum conductor temperature is 75°C, the increase in sag is only 10.25% after 30 years. Transmission line design conditions will determine whether the increase in sag or loss of strength is the limiting factor for a particular transmission line.
Elevated Temperature Effects
83
5.5 LOSS OF STRENGTH
The loss of tensile strength of a conductor results in lowering the design safety factor of the conductor. Generally, T/L conductor tension is designed with a safety factor of two at the worst loading condition.* The worst condition results in the conductor being subjected to the maximum tension. Such conditions arise when the conductor is exposed to high winds and/or ice covering. To give an example, an ACSR Cardinal conductor under extreme wind (~ 100 mph) and ice loading may result in the tension of the Cardinal conductor to reach 17,000 lbf which is approximately 50% of the rated tensile strength of the conductor. Therefore, a 10% reduction in the tensile strength of the conductor would also lower the safety factor of the conductor by 10%. Generally, a loss of strength up to 10% is acceptable (Mizuno et al., 1998).
5.5.1
PERCENTILE METHOD
A recent study (Mizuno et al., 1998) describes the calculation of thermal deterioration of a transmission line conductor by a probability method. The reduction in tensile strength of the conductor was used as the index of thermal deterioration. The loss of tensile strength is calculated as a function of conductor temperature, Tc, and the time duration, t, at which the temperature, Tc, is sustained (Morgan, 1978); (Harvey, 1972). W = exp{C(ln t – A – BT)} (5.11)
W is loss of strength (%) and A,B,C are constants that are characteristics of the conductor. This is an empirical equation based upon laboratory tests on individual wire strands. The total loss of strength is then obtained by,
∑ W = [(K((t t
1 2
t1 ) + t 2 t 3 t 2 + L t n −1 t n t n −1 ) + t n )t n
)
)
]
c
(5.12)
where ti is time duration when the conductor temperature is Ti and ti is given by, ln t i = A + BT ln (5.13)
5.5.2
RECURSIVE ESTIMATION
OF
LOSS
OF
STRENGTH
The author has developed a recursive method for the calculation of loss of strength of conductors as follows: W = Wa[1 – exp{–exp(A3 + B3Tc + n1 ln t + Kln (R/80)}] (5.14)
Wi = Wa[1 – exp{–exp(A3 + B3Tci + n1 ln(ti + tqWi1) + Kln (R/80)}] (5.15)
* National Electrical Safety Code C21997.
2 Value of Coefﬁcients in Equations (5. and Applications tqWi1 = exp[ln ln{1/(1 – Wi1/Wa)} + A3 + B3Tci + n1 ln(ti + tqWi1) + Kln (R/80)}]/ni A3.2 for the frequency distribution of Table A5.3 –14. which includes loss of tensile strength of the conductor. The probability distribution of conductor temperature is generated by Monte Carlo simulation of weather data from timeseries stochastic models and transmission line current. Therefore.* 5.035 0. and conductor sag.1. TABLE 5.5 –7.2 (Morgan.4 B3 0.0255 n1 0. K = constants given in Table 5.16) R = Reduction of wire by drawing from rod to wire (Morgan.40 K 9 18 11 (5. a new method is developed to determine the sag and tension of overhead line conductors with elevated temperature effects. In this chapter a uniﬁed approach to modeling and evaluation of the elevated temperature effects of transmission line conductors is presented. 1978).16) (Morgan 1978) Wire Aluminum Aluminum Alloy Copper A3 –8. 2.15).n intervals of time tqWi–1 = equivalent time for loss Wi–1 at temperature Tci Results are presented in Appendix 5 in Table A5. Modeling. (5. * PG&E Line Rating Standard. All Aluminum Conductors (AAC) are generally operated below 90°C under normal conditions.6 CHAPTER SUMMARY When transmission line ampacity is increased it is necessary to properly evaluate the thermal effects of the powerline conductors.060 0. 1978).84 Powerline Ampacity System: Theory. These results show that the loss of tensile strength for the AAC Bluebell conductor is greater than 10% when the maximum temperature of the conductor is 95°C and the life is 25 years. B3. permanent elongation. n1.79 0.285 0. 3…. . Conductor loss of strength and permanent elongation are evaluated recursively from the probability distribution of conductor temperature.17) D. Do are the diameters of wire and rod. D R = 100 [1 − Do 2 (5. For this reason. respectively i = 1.
Results are presented that compare well with all of the above data. The increase in sag is less than 15% for the same maximum operating temperature. Results are presented which clearly show that the loss of strength of the conductor is less than 10% when the maximum temperature of the conductor is 90°. . and KEPCO’s transmission line ﬁeld data. Ontario Hydro’s STESS program. The accuracy of the sagtension program was tested with ALCOA’s Sag and Tension program.Elevated Temperature Effects 85 A study of the AAC Bluebell conductor is presented to show the longterm effects of elevated temperature.
5 8. the conductor temperatures were increased in steps of 5°C up to the maximum temperature of 100°C.1 Frequency Distribution of Conductor Temperature Maximum Conductor Temperature.0 14.0 16.2 Σ 100% Table A5.Appendix 5 Sag and Tension Calculations FREQUENCY DISTRIBUTION OF CONDUCTOR TEMPERATURE A conductor temperature distribution is shown in Table A5. °C 75 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Frequency. 87 . TABLE A5. The corresponding frequency distributions of conductor temperature are also shown in this table. From the basic temperature distribution assumption shown in Column 1 of Table A5.% 2.5 0.0 2.0 11. The factor of safety of the conductor and the line to ground clearances can now be veriﬁed. and by the artiﬁcial generation of meteorological data. The sag and tension of the conductor are then calculated from this data.1.1.3 0.1 provides the starting point in the calculation of loss of strength and permanent elongation of conductor.5 0.0 13. This temperature distribution was generated by assuming constant load current equal to the static line rating.0 1.0 15.0 12.0 4.
0 12.5 11.5 7.1 6. The sag and tension of the 1034 Kcmil AAC Bluebell is given in Table A5. from 10 to 30 years.4.2 and A5.8 6.8 8.5 9.3 9.9 10.1 6. . TABLE A5.8 9.2 10. TABLE A5.5 6.8 5.4 11.3 5. respectively are used as input data to the sag and tension program.7 7.6 8.0 8.6 6. and Applications OF LOSS TENSILE STRENGTH OF CONDUCTOR Based upon the distributions shown in Table A5.2 7. Micro Strain 859 899 939 959 999 1049 1069 1119 1169 1189 1249 1289 1309 1369 1419 1449 1529 1579 969 1079 1209 1329 1459 1619 The loss of strength and permanent elongation of the conductor shown in the Tables A5.1.1. % 5.2 for the AAC Bluebell conductor.88 Powerline Ampacity System: Theory.7 4. Year Maximum Conductor Temperature °C 10 15 20 2 30 75 80 85 90 95 100 809 889 989 1109 1239 1359 Permanent Elongation.3 and is based on the same assumptions of conductor temperature distribution as shown in Table A5.3 5.2 5. Year Maximum Conductor Temperature °C 10 5 20 25 30 75 80 85 90 95 100 4. The results are shown in Table A5.0 9. Modeling.9 PERMANENT ELONGATION OF CONDUCTOR The permanent elongation of the AAC Bluebell conductor due to metallurgical creep calculated by the program is shown in Table A5.8 Loss of Strength.0 7.2 Loss of Strength of AAC Bluebell Calculated by Program Conductor Life. the loss of strength of the conductor was calculated at different conductor maximum temperatures for the conductor life.6 12.3.3 Permanent Elongation of AAC Bluebell Calculated by Program Life of Conductor.0 9.
4 Input Data 1034 Kcmil AAC Bluebell Conductor diameter.9 43.1 45.5 45.97 18500 8.6 46.0 41.2 45.9 5.8 5. psi Coefﬁcient of linear expansion.5 TABLE A5.9 33.0 46.8 46.8 48.3 48.0 0 0 0 0 0 0 0 Ice in 0 0 0 0 0 0 0 0 Tc °F 25 60 167 176 185 194 203 212 Tc °C –3. lbf Modulus of elasticity.9 5.4 47.4 46.4 40. °F 1.9 5.7 45.8 44.Appendix 5 Sag and Tension Calculations 89 SAG AND TENSION CALCULATION BY PROGRAM TABLE A5.5 Sag and Tension Initial Condition after SaggingIn Span ft 1000 1000 1000 1000 1000 1000 1000 1000 Wind lb/in2 8.6 47.9 47.1 46.9 Sag ft 31.7 47.0 95.6 75.17 0. lb/ft Rated Tensile Strength. /°C Sectional area.5 48.5 25 30 75 80 85 90 95 100 43. in2 Final unloaded tension.0 85.5 42.0 47. lbf Final unloaded temperature.0 46.4 44.9 45.7 4.0 90.0 40.9 5.7 48.9 .6 Final Sag Life.0 44.0 47.6 44.9 15.81 3700 50 TABLE A5.9 5.5 41.0 Tension lb 4972 3641 3045 3006 2969 2934 2899 2866 SF # 3. in Mass.2 44.0 100.0 80.7 44.5·106 23·10–6 0.0 42. ft 43. year Temperature °C 10 15 20 Sag.
8 lbf Wind = Wind Pressure on Projected Area of Conductor.8 6.0 6.13 1.0 6.2 8. and Applications WITH SAG AND TENSION COMPARISON ONTARIO HYDRO STESS PROGRAM OF To verify the accuracy of the results obtained from the sag and tension program. # Sag.80 20 TABLE A5.4 300 0 30 26.1 8360 19·106 468.84 6.45 9.96 8.64 9.01 9.8 Sag and Tension Comparison with STESS Program of Ontario Hydro Span ft Wind Pa Tc °C Tension kN Safety Factor.56 7. m STESS Sag. it was compared to the STESS program.9 300 0 100 19.90 Powerline Ampacity System: Theory. kg/mm Coefﬁcient of linear expansion /°C Sectional area.6 7. Pa Tc = Average Conductor Temperature. kN Modulus of elasticity.27 9. Modeling. .9 300 0 150 16.7 9.6 300 0 90 20.20 7.0 5.02 lb/ft2 1 kN = 225.3 300 0 110 18.8 7.3 7.1 10.4 10. mm2 Final unloaded tension.7 Input ACSR Drake Conductor diameter.5 300 0 60 22.9 300 0 70 21.4 8.8 8. °C *Calculated by the sag and tension program described in this chapter. TABLE A5.1 5.5 300 0 120 18.5 7.47 6.82 9.1 7. °C 28.3 6.8 300 0 40 25.2 300 0 50 23.4 5.8 300 0 140 17.7 7. m Program* 300 0 20 28.1 Note: 1 Pa = 0.2 6.9 8.8 6.9 7.628 140. mm Mass. The results are provided below.92 7.7 27. kN Final unloaded temperature.2 9.8 5.2 300 0 80 21.27 8.20 10.62 8.6 300 0 130 17.9 9. kg/m Rated Tensile Strength.7 8.4 9.
5·10–6 0.10 Sag and Tension Comparison with ALCOA Program Span ft 750 750 750 750 750 750 750 750 750 Wind lb/ft2 33.5 20.0 0 0 0 0 0 0 0 0 T °Fc 60 70 80 90 100 110 120 170 205 Tension lbf 11496 4293 4171 4057 3951 3852 3784 3578 3450 Safety Factor.7 7.5 22.1 7. SF 2. psi Coefﬁcient of linear expansion. 11. °F * Calculated by the sag and tension program described in this chapter.7264 4403 60 TABLE A5. TABLE A5.1 9.9 Input (ALCOA Program Data) 795 AS33 ACSR DRAKE (26 al + 7 st) Conductor diameter.3 21.0 19. . lb Modulus of elasticity.7 7. lb/ft2 Tc = Average Conductor Temperature. 41. November 1989.0 19. in Mass.0 18.3 Wind = Wind Pressure on Projected Area of Conductor.0940 31500 11.3 7. /°C Sectional area.8 18.0 9. No.89·106 19. Lankford.5 21.7 Sag. in2 Final unloaded tension. Transmission and Distribution Journal.5 22. °F 1. ft ALCOA 21.9 8. * Craig B.3 Sag.5 20.0 20. lbf Final unloaded temperature.0 20. ft Program* 21. lb/ft Rated Tensile Strength.5 19.9 18.0 18. ALCOA’s Sag and Tension Program Enhanced for PC Use.108 1.5 19.Appendix 5 Sag and Tension Calculations 91 SAG AND TENSION COMPARISON WITH ALCOA PROGRAM* Results obtained by new program are compared to the ALCOA Sag and Tension Program (see footnote).5 7. Vol.
1 16.30 98. /deg C OUTPUT Span.92 Powerline Ampacity System: Theory.00 STACIR Tension kN 29.30 1.2 24.00 19E·06 379.8 NA NA NA NA STACIR 25.30 1. mm2mm Final unloaded tension. expan.0 17. kg/m Rated Tensile Strength.0 NA = Not Applicable Table adapted from KEPCO highampacity transmission line (Choi et al.5 16.5 18.3 5. .3 5.3 16.60 24. and Applications TABLE A5.00 ACSR Wind Pa 400 0 0 0 0 0 0 Tc Deg C 10 10 90 150 200 210 240 Tension kN 29. Modeling.60 24.2 24. /deg C Sectional Area. kg/mm2mm Coeff.00 8360. mm Mass.4 NA NA NA NA Sag m 6.00 16500.11 Sag and Tension Comparison with KEPCO Line Data 154 kV Double Circuit Line INPUT Conductor dia. 1997).50 10. kN Modulus of Elasticity.9 8. of lin.50 10.9 8.4 16.8 9.0 Sag m 6.4 8.7 8. m 300 300 300 300 300 300 300 ACSR 25. kN Final unloaded temperature.30 98.00 3GE·07 379.2 8..
It must be mentioned that the lowering of transmission line conductor to ground distance due to higher sag will raise the level of electric and magnetic ﬁelds at ground level.2 TRANSMISSION LINE MAGNETIC FIELD The magnetic ﬁeld of a currentcarrying transmission line conductor is calculated by the application of Maxwell’s equation.1 INTRODUCTION The magnetic ﬁeld of a transmission line increases with line ampacity. 2. The study of transmission line magnetic ﬁelds is also important from the point of view of transmission line ampacity. Rashkes. Line designs and EMF mitigation methods are developed to lower transmission line magnetic ﬁelds in sensitive areas. electric ﬁelds are also evaluated since the level of an electric ﬁeld at ground is affected by conductor sag. No. Typical powerline conﬁgurations are evaluated to show their magnetic ﬁelds. On the other hand. it is important to calculate transmission line sag accurately for the calculation of electric and magnetic ﬁelds at ground level. It is shown that the magnetic ﬁeld of overhead transmission lines is within acceptable limits. In the previous chapters it was shown that transmission line capacity may be increased by dynamic line ratings. Lordan. The study presents data to quantify more accurately the magnetic ﬁelds of different transmission line conﬁgurations. Therefore. and increases at ground level with conductor sag. Electric ﬁeld limit at ground level is not exceeded by higher transmission line ampacity if the maximum design temperature of the conductor is not exceeded and minimum conductortoground distance is maintained.S. Even though transmission line voltage remains unchanged with higher ampacity. which will result in the lowering of the cost of electricity. R.” IEEE Transactions on Power Delivery. 6. The Electric Power Research Institute (EPRI) conducted a study* in which they proposed methods for the reduction of transmission line magnetic ﬁelds. With public concern for electric and magnetic ﬁelds. * V.6 Transmission Line Electric and Magnetic Fields 6. 93 . April 1998. transmission line engineers are required to accurately evaluate the impact of increased line capacity on the environment due to higher electric and magnetic ﬁelds. 13. Vol. “Magnetic Field Reduction Methods: Efﬁciency and Costs. increasing line ampacity also increases the level of magnetic ﬁeld.
Hz are the components of the magnetic ﬁeld. Hϕ. J (A/m2).4) the following solutions are obtained for the calculation of the magnetic ﬁeld of a currentcarrying conductor. ϕ. The magnetic ﬁelds of typical transmission line conﬁgurations are also presented to show that magnetic ﬁelds are within acceptable limits.3) ∂Hr 1 ∂Hφ ∂Hz 1 ∇ ⋅ H = Hr + + ⋅ + =0 ∂r r ∂z r ∂φ (6. From equations (6. and z axes. as shown in Figure 6. and r J is current density. ∂ = 0 (due to circular symmetry) ∂ϕ ∂ = 0 (due to infinitely long conductor) ∂z The current I through the conductor is.4) Hr. H. (∇ × H ) r 1 ∂Hz ∂Hϕ – = ⋅ = Jr r ∂ϕ ∂z 1 ∂Hz ∂Hz – ⋅ = Jφ r ∂z ∂r 1 ∂Hϕ ∂Hr – Hϕ + = Jz r ∂r ∂ϕ (6.94 Powerline Ampacity System: Theory.2) (6. and Applications In this section. along r. Considering a simple case of an inﬁnitely long cylindrical conductor carrying a r direct current density.1. This method is suitable for any transmission line conﬁguration.1) (∇ × H ) ϕ = (∇ × H ) z = (6. Modeling. EMF mitigation methods are also given. we have. a general method of calculation of the magnetic ﬁeld of overhead transmission line conductors is presented.2. It is shown that there is minimum environmental impact due to higher transmission line ampacity. (6. I = JzπR2 where R is conductor radius.1) through (6.1 THE MAGNETIC FIELD OF A CONDUCTOR In the following section we shall derive expressions for the calculation of the magnetic ﬁeld of a currentcarrying conductor by the application of Maxwell’s equations.5) . 6.
Transmission Line Electric and Magnetic Fields 95 z j R ϕ r FIGURE 6.10) from (6.6) (6.10) we have. Field Outside of Conductor from (6.7) 1 ∂Hz ∂Hϕ – =0 r ∂ϕ ∂z 1 ∂Hr ∂Hz – =0 r ∂z ∂r (6.9) 1 ∂Hϕ ∂Hr – Hϕ + =0 r ∂r ∂ϕ (6.8) we obtain. 1 ∂Hϕ Hϕ + =0 r ∂r (6.11) .6) and (6. ∂Hz = 0 or Hz = constant = 0 ∂r Since ∂ =0 ∂ϕ from (6.1 Magnetic ﬁeld of a conductor carrying dc current j.1)–(6.4) r jo = 0 ∇×H=0 (6.8) (∇ × H ) r = (∇ × H ) ϕ = (∇ × H ) z = Since ∂ =0 ∂z (6.
4) we have. Hr = where B is a constant. 1 ∂Hr Hr + =0 r ∂r and obtain the following solution. From (6.16) (6. from (6. ∂ = 0. ∇⋅H = 0 ∂Hr 1 ∂Hϕ ∂Hz 1 ∇ ⋅ H = Hr + + + =0 ∂r r ∂z r ∂ϕ since. ∂ϕ 1 ∂Hϕ Hϕ + = jz r ∂r (6.14) (6. inside the conductor has the following components. Jr = 0 Jϕ = 0 Jz ≠ 0 Since ∂ = 0 . and Applications and obtain. Hϕ = where A is a constant.12) .10) we have.17) B r (6. Field Inside the Conductor r The current. J .96 Powerline Ampacity System: Theory.13) A r (6. ∂ϕ we have. Modeling.15) ∂ =0 ∂z (6.
H (A/m).25) (6. Hϕ → ∞ or K= 0 substituting.Transmission Line Electric and Magnetic Fields 97 which has for solution.23) Hz = 0 Outside the conductor: r>R Hr = 0 Hφ = I 2π ⋅ r (6. is found as follows: Inside the conductor: 0≤r≤R Hr = 0 Hφ = Ir 2π ⋅ R 2 (6. inside and outside a conductor with current.26) Hz = 0 .24) (6.21) (6.22) (6.18) where K and L are two constants which are determined from the following boundary conditions.19) r The magnetic intensity. we have ∂Hϕ K =– 2 +L ∂r r Lr + L = jz r and obtain. as r → 0. Hϕ = Summary of Equations I⋅r 2 ⋅ π ⋅ R2 (6. Hϕ = K + Lr r (6. I.20) (6.
and Applications For environmental impact studies. and horizontal component.2. and at a distance 30 m from the line shown in Figure 6. H 3 are the individual contributions of the magnetic ﬁeld of each phase conductor at the same point in space. M. for a threephase transmission line having one conductor per phase. The line is loaded to its summer maximum thermal rating of 1500 A per phase. Example 6. H 2 .98 Powerline Ampacity System: Theory. It is seen that the magnetic ﬁeld at ground level increases with transmission line ampacity (I) and by lowering of the distance (r) from conductor to ground. 6. we are interested in the ﬁeld outside the conductor in free space.1 Calculate the magnetic ﬁeld of a threephase single circuit 750 kV transmission line at a point.27) Therefore. Equation (6. r r r r H r = H1 = H 2 + H 3 (6. Modeling.25) by vector addition of the magnetic ﬁeld of each conductor as follows: r Hr = ∑H i =1 n r i (6.25) gives the magnetic ﬁeld. outside the conductor. Hφ.2. y 15 m 1 20 m 2 15 m 3 30 m 1m 0 M x FIGURE 6. Hr.1). Hz.2 THE MAGNETIC FIELD OF A THREEPHASE POWERLINE The magnetic ﬁeld of a polyphase transmission line at a point in space can be calculated from (6.28) r H r is the resultant magnetic ﬁeld (A/m) of the transmission line at a point in space r r r and H1 . 1 m above ground. of the magnetic ﬁeld outside the conductor is zero. the magnetic ﬁeld is. The distance (r) is a function of conductor sag.2 Phase conﬁguration of 750 kV line (Example 6. The radial. .
9u x + 4. A i =1.m = = 1500∠ − 120 2 π 19 2 + 30 2 1500∠ – 120 19 r 30 r u + u 2 π35. x1 x2 x3 y1 y2 y3 = = = = = = –15 m 0m 15 m 20 m 20 m 20 m Phase current. I1 = 1500∠0 I 2 = 1500∠–120 I 3 = 1500∠120 The magnetic ﬁeld at point M is calculated from (6.8 48.8 x 48.5 35.m = = 1500∠0 2 π 19 2 + 452 1500∠0 (20 – 1) r (15 + 30) r u + uy 2 π 48.25).8 r r r r = 1. H i .m I = phase current.09)ux + (2.88)uy . u y are unit vectors in x and y direction ( ) H 2 .5 y r r = (1.m = Ii 2πri. = distance from phase conductor i to point m.79 + j3.5u y u x .2.3 phase rim.Transmission Line Electric and Magnetic Fields 99 Solution Transmission Line Conﬁguration Phase distance.5 x 35. m H1.84 + j4.
A/m I = positive sequence current. and a new conductor conﬁguration is shown in Figure 6.m = = 1500∠120 2 π 19 2 + 152 1500∠120 19 r 15 r ux + uy 2 π35.87 – j6. and the magnetic ﬁeld at 30 m distance from the axis of the transmission line for each line conﬁguration is shown in Table 6. April 1998.100 Powerline Ampacity System: Theory. The geometry of Figure 6. Tesla µ0 = permeability of free space H = magnetic ﬁeld intensity.10. and 6. Modeling.5 24. Pn B = A ⋅ n +1 ⋅ I r B = µ0H is the magnetic ﬁeld strength. The magnetic ﬁeld of each conﬁguration is shown in Figure 6. 6.9.2 24.” IEEE Transactions on Power Delivery.S.m r H M = 5.38∠15.3 THE MAGNETIC FIELD LINE GEOMETRY OF DIFFERENT TRANSMISSION In this section results are presented to show the magnetic ﬁelds of highvoltage transmission lines having different geometry. Lordan. Rashkes. * V. (6. R.65)ux + (3.29) 6. No. . and Applications H 3.25)uy The magnetic ﬁeld at point M is obtained by. “Magnetic Field Reduction Methods: Efﬁciency and Costs. threephase AC. Conventional transmission line conductor conﬁgurations are shown in Figures 6. 13.6. m A = numerical coefﬁcient depending upon line design (geometry) n = number of subphases for a split phase line The above equation is consistent with Equation (6. A P = phase to phase distance. and the transmission line geometry.5 A m The magnetic ﬁeld of a three phase powerline may be obtained approximately* by. The calculations are based upon very optimistic line ampacities that would only be possible by adopting a dynamic line rating system. 2. Vol.2 r r = (3. m r = distance from the axis of the line to the point of measurement.2.1.9 provides a compact transmission line design with reduced magnetic ﬁeld.8.25) after taking into consideration the effect of multiple conductors.m + H 3.m + H 2. r H M = H1.05 + j5.7.
Ph. 1998) indicates 100 microTesla as the upper limit.8 Double circuit vertical conﬁguration FIGURE 6. as well as static ﬁelds and MW/RF frequencies. and the Medical College of Wisconsin. A Cigré 1998 paper (Bohme et al.Transmission Line Electric and Magnetic Fields 101 FIGURE 6. Copyright©.10 are based upon high transmission line * Restriction on human exposures to static and timevarying EM ﬁelds and radiation. by John E. It must also be mentioned that transmission line magnetic ﬁelds shown in Figure 6.330 micro T for the magnetic ﬁeld. Exposure limits for powerfrequency ﬁelds. 1993–1998. Moulder.6 Single circuit horizontal conﬁguration FIGURE 6.9 Compact line with phase splitting The above values are well within the acceptable limit* of 1330 microTeslas for continuous exposure. For 60Hz the limits recommended are 10 kV/m for the electric ﬁeld and 1. . 1993. the standards apply to both residential and occupational exposure.7 Single circuit delta conﬁguration FIGURE 6.D. Documents of the NRPB 4(5): 1–69.
µT 2. Passive Shielding Passive shielding of overhead lines is accomplished by the addition of auxiliary shield wires connected in a loop at certain critical sections of the line where * “Compacting Overhead Transmission Lines. These approaches have resulted in lowering the magnetic ﬁeld to about 0. USSR. double circuit. During normal operating conditions the line current will be less than 1000 A. 1991.8 Figure 6..6 Figure 6.8 Compact Star. ACSR Zebra 1000 A TABLE 6. m FIGURE 6. and Applications Transmission Line Magnetic Field 30 25 20 15 10 5 0 10 20 30 40 50 Figure 6.7 Vertical.2. 6. Figure 6. Figure 6.9 Magnetic Field at 30 m. and are recommended for areas such as schools. Figure 6.06 0.9 MicroTesla Distance.6 Delta. even tighter control over the level of EMF radiated from a line can be achieved by EMF mitigation measures.10 Transmission line magnetic ﬁeld.94 0.* Active and passive shielding of lines by the addition of auxiliary conductors on certain sections of the line are also used to lower the magnetic ﬁeld at critical locations.” Cigré Symposium. Compact line designs having low levels of EMF are now used extensively.4 EMF MITIGATION Even though overhead transmission lines are designed with low levels of EMF. 1998). single circuit. phase splitting.92 2.1 Conﬁguration Horizontal. . 1000 A for an ACSR Zebra conductor. and other areas where the public may be exposed to EMF continuously (Bohme et al. hospitals.102 Powerline Ampacity System: Theory. therefore. 3–5 June. and.2 µT at the edge of transmission line rightofway.01 conductor ampacity.7 Figure 6. single circuit. Leningrad. Figure 6. the magnetic ﬁeld will be lower. Modeling.
In this method. current ﬂows through the auxiliary conductor by induction from the powerline conductor.2.4 750 kV line showing shielding angles (Example 6. Example 6.Transmission Line Electric and Magnetic Fields 103 controlling line EMF is important. 1 m above ground and 30 m from the line shown in Figure 6. The magnitude and phase of the current in the auxiliary conductor is managed by controlling loop impedance. M1 and M2. Two auxiliary conductors. an external power supply is used to circulate a current through the shield wires. thereby lowering the total ﬁeld from the line.3. y 15m 15m 1 M1 2 3 M2 30m M 1m 20 m 13m 0 x FIGURE 6. but instead of induced current in the auxiliary conductors. Active Shielding Active shielding is similar to passive shielding. The following example will help in the analysis of magnetic ﬁeld shielding of overhead powerlines. In this manner the auxiliary conductor generates a ﬁeld that opposes the ﬁeld produced by the power conductor. are used for magnetic ﬁeld shielding by forming a 1 km current loop. The line is loaded to its summer maximum thermal rating of 1500 A per phase.2) 30m 15m 1 2 15m 3 7m Ë1 Ë Ë2 Ò 30m M1 M2 FIGURE 6.2) .2 Calculate the magnetic ﬁeld of a threephase 750 kV single circuit transmission line at a point M. shown in Figure 6.3 750 kV line with auxiliary shield wires (Example 6. In this manner even greater control over the ﬁeld generated by a line is possible by controlling the amplitude and phase of the current through the shield wires.
and Applications 2 Magnetic Field 750 kV Line 1 A/m Hyi 0 1 2 6 3 0 Hxi A/m 3 6 FIGURE 6. V1 = voltage induced in the loop Z1 = loop impedance The induced voltage is calculated from V1 = jωφ where. I1 = where. m2 B = total ﬂux density.104 Powerline Ampacity System: Theory. Tesla . ω = angular frequency.5 Magnetic ﬁeld of 750 kV 3 phase transmission line with auxiliary shield wires. radian/s φ = total ﬂux penetrating the loop. Solution The current Il ﬂowing through the loop is calculated from. wb/m = length of the loop The total ﬂux φ is calculated by φ= V1 Z1 ∫ B. Modeling.dS s where S = area of the loop.
φ2 . dx = h cot(θ)dθ dz = 1 we ﬁnd the ﬂux φ1 due to current I1 .Transmission Line Electric and Magnetic Fields 105 Since only the y component of the ﬂux density vector B contributes to the above surface integral. B3y are the ﬂux density components due to current I1 . φ1 = ∫ B dn ⋅ dz ly s r Substituting B = µoH H from (6. cos(θ) dx = h cot(θ and.25) we have. θ1 = 0 θ 2 = 76. respectively. φ1 = 2. r r r r B = Bly + B2 y + B3y r r r Bly .80 and obtain the solution to the integral. I 3 in the phase conductors respectively and produce the ﬂux φ1 . φ3 . B2 y . µ I φ1 = 0 1 2π From ﬁgure we have. I 2 . r µ I sin θ Bly = 0 1 2 πρ from. Calculating φ1 from the integral of equation.35 ⋅ 10 –4 ∠0 wb θ2 ∫ sin(θ) cos(θ) cot(θ)dθ θ1 .
I 3 = 1500∠120 θ1 = –76. Modeling.6 0 θ2 = 0 0 µ I φ3 = 0 3 2π θ2 = 0 θ1 = 76.2 ∠ – 30 . Vl = j ⋅ φ ⋅ ω ⋅ l = 1000 m (1 km loop) ω = 2⋅π⋅60 rad/s Vl = j(2.35 ⋅ 10 –4 ∠120 wb The total ﬂux is.6 ∫ sin(θ) cos(θ) cot(θ)dθ φ3 = 2.2 · 10–4 ∠ – 120) · 377 · 103 Vl = 83.6 0 µ I φ2 = 0 2 2π θ2 = 64.2 ⋅ 10 –4 ∠ – 120 wb The induced loop voltage Vl is.6 ∫ sin(θ) cos(θ) cot(θ)dθ φ2 = 4.6 0 θ 2 = 64.55 ⋅ 10 –4 ∠ – 120 wb Flux φ3 due to current I3. φ = φ1 + φ2 + φ3 φ = 2. the ﬂux φ2 due to current I 2 is calculated I 2 = 1500∠ – 120 θ1 = 64.106 Powerline Ampacity System: Theory. and Applications Similarly.6 θ1 = 64.
2∠ – 30 0. at point M. from.57 r r = –0.3 12 r 15 r 19. Il = 83.m Ia = current in auxiliary conductors.Transmission Line Electric and Magnetic Fields 107 The auxiliary conductor loop impedance Za is selected as.m = = 277.24 u x – 0.71uy . conductor i to point M Ha 1.2 ri. Ha.m = Ia i 2πri.2 r r = –1. Za = 0. is calculated. Ha i.92 u y H 2 .x = – (0.3 A Magnetic Field with Shielding The magnetic ﬁeld.1.92 + 1.3 (13 – 1) r (15 + 30) r 46.57 u y 2 π 46. A i =1. as in Example 6. Ω The loop current is.m = distance from auxiliary conductors.m = = 277.67ux r r r H a .3∠ – 30 I l = 277. due to the two auxiliary conductors.79)uy = –2.79uy r r r H a .43)ux = –1.y = – (0.3 2 π 12 2 + 452 277.57 u x + 46.3 2 π 12 2 + 452 277.2 uy 2 π19.2 ux + 19.24 + 1.3∠ – 30.43ux – 1.
56) – 1.31 – j0. ( ) Electric ﬁeld calculation The various methods of calculation of the electric ﬁeld by numerical and analytical methods are given in a Cigré report (Cigré.71}uy = (1.67}ux = (2.x = {(3.108 Powerline Ampacity System: Theory.3 6. 1980). and. r. the gradient of the voltage is effectively zero.3 x y H y Angle = tan –1 = 18.31) r r r In a perfect conductor the potential difference ( V1 – V2 ) is zero. such that. of the transmission line at any point in space is a function of line voltage and the distance of the point from the transmission line conductor.30 Hx r H = 4. E.3∠18. the electric ﬁeld inside a perfect conductor is also zero.56)ux r r r H y = H M. r r E = – grad V V m ( ) (6.y + H a . the electric ﬁeld at ground level is affected by conductor sag since an increase in conductor temperature will increase sag and result in lowering the distance of the powerline conductor to the ground.93 – j0.37)uy r The magnetic ﬁeld H is an ellipse r r r H = H 2 + H 2 = 4. Modeling. The electric ﬁeld strength E may be deﬁned as gradient of the potential V given as. r r r V1 – V2 grad V = (6. Therefore.y = {(1. In this section we shall study the method of calculation of the electric ﬁeld of a transmission line and determine the effect of conductor sag on the electric ﬁeld at ground level.x + H a .3 TRANSMISSION LINE ELECTRIC FIELD The electric ﬁeld.37) – 2.77 – j3. and Applications r r r H x = H M. hence.08 – j3.30) There exists an electric ﬁeld if there is a potential difference between two points having potential V1 and V2 separated by a distance. In this section we shall apply the . The effect of conductor sag due to higher transmission line ampacity was studied in Chapter 5.
and [V] a column vector of phase voltages. The capacitance matrix [C] is calculated from Maxwell’s potential coefﬁcients [λ] deﬁned as the ratio of the voltage to charge.j is the distance between conductors i and j. 2D′ 1 ij ln D ij 2 πε 0 nr R 2h 1 ln i 2 πε 0 ri (6. from a charge. req = R ⋅ n r = subconductor radius n = number of subconductors in bundle R = geometric radius of the bundle The elements λij of the matrix of potential coefﬁcients are calculated by. of the matrix of potential coefﬁcients are calculated by. [C] is the capacitance matrix of the multiconductor circuit. E= q 2 πε 0 r (6. λii. and D′i. at a distance.33) where [q] is a column vector of charges. r.j is the distance between image conductors i’ and j’.34) (6. λ ii = where. q. E.Transmission Line Electric and Magnetic Fields 109 analytical method of equivalent charges to calculate the electric ﬁeld of threephase transmission lines. The electric ﬁeld. The elements.35) λ ij = (6.36) where Di. .32) The q charges carried by transmission line conductors is calculated by. [q] = [C] · [V] (6. hi = height of the conductor i above ground ri = radius of conductor i For bundle conductor system an equivalent radius is calculated as. is calculated by Gauss’s law.
110 Powerline Ampacity System: Theory. Modeling. Example 6. FIGURE 6.6 Threephase conﬁguration of a 750 kV transmission line Solution The height of conductor above ground is given as. we calculate [q] from. [C] = [λ]–1 Knowing [C] and [V].39) where. r Ei = q 2 πε 0 ri (6. 1 m above ground.38) (6.40) The following example illustrates the important concepts presented in this section by showing the calculation of the electric ﬁeld of the transmission line in Example 1. .3 Calculate the electric ﬁeld of a three phase single circuit 750 kV transmission line at a point. h1 = h2 = h3 = 20m The equivalent radius of four conductor bundle is. and Applications The matrix [C] is calculated by inversion.1. E. [q] = [C] · [V] (6. The line is loaded to its summer maximum thermal rating of 1500 A per phase. is then calculated by the application of Gauss’s law by vector summation of the individual ﬁelds due to the charge on each conductor r E= ∑E i =1 i=n r i (6. and at a distance 30 m away from the line as shown in Figure 6.37) The electric ﬁeld. M.
195 ⋅ 10 9 1.7 ⋅ 10 –13 –3.j ← 2⋅π⋅ε0 A For a threephase transmission line. n = m = 3. 7. λ ( n. 3) = 1.884 ⋅ 1010 7.126 ⋅ 10 –12 –8. is obtained by substitution of [C] and [V] .76 ⋅ 1010 λ(3.Transmission Line Electric and Magnetic Fields 111 req = R ⋅ n nr R 4 ⋅ 30 ⋅ 10 –3 150 ⋅ 10 –3 req = 150 ⋅ 10 –3 ⋅ 4 = 0..126 ⋅ 10 –12 –8.126 ⋅ 10 –12 1.375 ⋅ 10 –11 The charge.884 ⋅ 1010 9..126 ⋅ 10 –12 1. .884 ⋅ 1010 7.j ← if i = j 2⋅π⋅ε0 D′ i.n for j ∈ l.76 ⋅ 1010 ( ) And we obtain the capacitance matrix [C] from.76 ⋅ 1010 1.375 ⋅ 10 –11 C = –3.884 ⋅ 1010 9.m h ln 2 ⋅ i ri A i. m ) : = for i ∈ l.44 ⋅ 10 –11 −3. j ln D i. q. resulting in the following matrix of potential coefﬁcients.195 ⋅ 10 9 1.54 m Maxwell’s potential coefﬁcients are easily calculated in Mathcad® as. j otherwise A i. [C ] = [ λ ] –1 1.7 ⋅ 10 –13 –3.
126 ⋅ 10 –12 –8. is obtained by the application of Gauss’s law. (y i – h i ) (y i + h i ) Ey i : = E i ⋅ ρi ⋅ 2 2 – 2 2 (x i – d i ) + (y i – h i ) (x i – d i ) + (y i + h i ) (x i – d i ) (x i + d i ) Ex i : = E i ⋅ ρi ⋅ 2 2 2 – 2 (x i – d i ) + (y i – h i ) (x i – d i ) + (y i + h i ) –3. E.126 ⋅ 10 –12 –8. q1 ρ1 1 q2 [E] = 2 πε 0 ρ2 q 3 ρ3 The resultant E ﬁeld at M is calculated by vector summation by adding X and Y components of the individual elements of [E].45 ⋅ 10 –6 [q] = –6.126 ⋅ 10 –12 1. The X and Y components are obtained as.636 ⋅ 10 –6 – j1.44 ⋅ 10 –11 −3.088 ⋅ 10 –5 The electric ﬁeld. 1. Modeling.375 ⋅ 10 –11 [q] = –3.112 Powerline Ampacity System: Theory.7 ⋅ 10 –13 giving.375 ⋅ 10 –11 750 ⋅ 10 3 ∠240 .18 ⋅ 10 –5 – j1.7 ⋅ 10 –13 750 ⋅ 10 3 ∠0 –3.574 ⋅ 10 –6 + j1.131 ⋅ 10 –5 –4.126 ⋅ 10 –12 750 ⋅ 10 3 ∠120 1. and Applications [q] = [C][V] 1.
7 The electric ﬁeld of 750 kV threephase transmission line. As shown by the equations developed in this chapter. The magnetic ﬁeld at ground level also increases with higher sag. Consequently the distance from conductor to ground will become less than normal which will raise the electric ﬁeld at ground level. 6. If conductor temperature is higher than normal due to higher current through the line. Since both electric and magnetic ﬁelds of a transmission line depend upon conductor temperature.Transmission Line Electric and Magnetic Fields 113 ∑ Ey i =1 3 3 i = 9.475 ⋅ 10 3 + 7. for EMF considerations also. A numerical example is provided in this chapter for the calculation of the magnetic ﬁeld of threephase transmission line. The magnetic ﬁeld inside and outside of a currentcarrying conductor has been developed from Maxwell’s equation and Ampere’s law. Methods of reducing magnetic ﬁelds by active and passive shielding are also presented in this chapter. If conductor temperature is higher than the maximum allowed for the line. Therefore. it is very important that increasing line ampacity does not exceed the maximum design temperature of the conductor.271 + 1. it is important to follow a dynamic line rating system that will maintain normal conductor temperature within a speciﬁed limit.181i10 3 The resultant electric ﬁeld is an ellipse as seen in the Figure 6. FIGURE 6. then sag will increase. The electric ﬁeld from a transmission line at ground level is indirectly affected by line ampacity only if an increase in line ampacity raises the maximum design temperature of the transmission line conductor. A method of calculation of electric ﬁeld at ground level due to higher transmission line ampacity is given in this chapter with a numerical example.4 CHAPTER SUMMARY The magnetic ﬁeld of a transmission line at ground level is a function of line current and the distance of phase conductors from ground.602i ⋅ 10 3 ∑ Ex i =1 i = –127. . then the magnetic ﬁeld at ground level will become more signiﬁcant due to the combined effect of high current and reduced distance of conductor to ground. an increase in line current increases the magnetic ﬁeld at ground level.7.
The prediction of transmission line ampacity several hours in advance has become more important today due to competition in the electric power supply industry. are now becoming possible due to developments in weather forecasting. 3. 1997. Cibulka. and an analytical model of solar radiation shall be developed from time series data. Hourly values of future meteorological conditions from 1 to 24 hours ahead. IEEE Transactions on Power Systems. the above meteorological data can be obtained by measurement from weather stations. Waight. and Deb. Neural network models are also presented for forecasting hourly values of meteorological conditions as well as for weather pattern recognition. 1991. 1998. wind speed.7 Weather Modeling for Forecasting Transmission Line Ampacity 7. Numerical examples * M. In this chapter. and transient states requires the knowledge of the following meteorological variables: • • • • Ambient Temperature Wind Speed Wind Direction Solar Radiation When transmission line ampacity is required for the present time. Foss and Maraio.1 INTRODUCTION Since weather is an important parameter in the determination of transmission line ampacity. Vol. 1988c. 1995. and wind direction. wind direction. Douglass. H. wind speed. 13. For the prediction of line ampacity several hours in advance. K. No. and greater need for the advance planning of electricity generation and transmission capacity* (Deb. Aganagic. The solution of the differential equations for the heating of a conductor by current in the steady. AbdulRahman. dynamic. the development of weather models of ambient temperature. 1989). August 1998 115 . Hall and Deb.G. a weather model is required. or even up to 1 week in advance. Williams. and solar radiation are presented in this chapter. J. Spot pricing of capacities for generation and transmission of reserve in an extended Poolco model. These are statistical weather models based upon timeseries analysis and National Weather Service forecasts. 1986. stochastic and deterministic models of ambient temperature.
. The LINEAMPS program uses a weather model based on historical weather data as well as weather forecast data prepared by the National Weather Service.116 Powerline Ampacity System: Theory. The Fourier series model is given by. Ci.2) . Ws(t)} A0. a Kalman ﬁltertype algorithm is developed for the recursive estimation of weather variables for realtime prediction of transmission line ampacity. In the future we expect an increase in the number of power producers requiring access to utility transmission lines for the supply of electricity. Bi for i = 1. 1999. Yf (t max ) – Yf (t min ) F(t max ) – F(t min ) ∑ i =1 n C i Sin(ω i t ) + ∑ B Cos(ω t) i i i =1 n (7.n are coefﬁcients of the model ω = 2π/24 = fundamental frequency k = factor used to adjust Fourier series to National Weather Service forecast. In the ﬁrst approach. When continuous input of realtime meteorological data is available.1) k= (7. 1990). Having precise knowledge of future transmission line capacity will greatly facilitate the purchase of competitively priced electricity from remote locations. 7. and Applications of weather forecasting are presented in this chapter. In the second approach. Y( t ) = A 0 + k Description of symbols Y(t) ∈ {Ta(t). Modeling. Hourly values of future ambient temperature and wind speed data are then generated from these patterns as described in the following section. These patterns are then adjusted to forecast weather data available from the National Weather Service or other weather service companies. a transmission line ampacity program with forecast capability is essential. followed by an example of line ampacity forecast generated by a program.2 FOURIER SERIES MODEL Hourly values of ambient temperature (Ta) and wind speed (Ws) at time (t) are generated by AmbientGen and WindGen methods in the weather station object of the program by ﬁtting Fourier series to historical weather data of the region. Eberhart and Dobbins. weather patterns are recognized by training an unsupervised neural network using Kohonen’s learning algorithm (Haykin. It is calculated by. For these reasons. Two alternative approaches to weather modeling are developed.. hourly values of historical weather data for different seasons of the year are ﬁtted by a Fourier series.
and 7.3) e–jωt = A(ω) +jB(ω) (7. K. 7. It is appropriate to mention here that these patterns are applicable to the region of the San Francisco Bay area only. as we should expect for the region of San Francisco.Weather Modeling for Forecasting Transmission Line Ampacity 117 Yf(tmax) = daily maximum value of ambient temperature or wind speed forecast by the National Weather Service. The unknown parameters of the model [A0. A similar analysis is required for transmission lines in other regions. the fundamental period is found to be approximately equal to 24 hours. n.14.6 where.4.2. 2 N rN (ω ) = ∑X ⋅e i t =1 N – jωt (7.4. The same phenomenon is observed in all of the meteorological variables comprising ambient temperature. The development of a weather model (Figure 7.2. when t = tmax and t = tmin respectively.042.6. Since the period T = 1/f . which is also the fundamental frequency. F( tmax) and F( tmin) are found from the Fourier series (7. ω] are determined by least square regression analysis. 7. wind speed. Ai.1). C 30 Ta j 20 10 0 500 1000 j 1500 2000 2500 Hour FIGURE 7. The fundamental frequency ω and the coefﬁcients rN(ω) of the frequency spectra are also determined by spectral analysis (Priestley 1981) as shown in the Figures 7.2 we see that the dominant frequency is equal to 0. Examples of Fourier series patterns of hourly ambient temperature and wind speed that were developed for the San Francisco Bay area during summer time are shown in Figure 7.4) From Figure 7. and solar radiation in this region as seen in Figures 7. wind direction.1) requires determining the parameters of the model from historical weather data of each of the meteorological variables. and 7. Bi.1 Hourly averaged ambient temperature data during summer time in San Francisco. Signal: Ambient Temperature 40 Temperature Deg. Yf(tmin)= daily minimum value of ambient temperature or wind speed forecast by the National Weather Service. .
2 Ambient temperature spectrum.17. . The autocorrelation rk is calculated by. Modeling.1 0. and Applications 100 Ambient Temperature Spectrum 80 Coefficient 60 r j 40 20 0 0 0.1 0.118 Powerline Ampacity System: Theory.4 0. 20 Signal: Wind Speed 15 Wind Speed.5 FIGURE 7.4 0. The daily cyclical behavior of the meteorological variables is further supported by the autocorrelations that were calculated from the hourly averaged values of each time series as shown in the Figures 7.3 0.5 FIGURE 7. m/s Wsj 10 5 0 0 500 1000 j Hour 1500 2000 FIGURE 7.157.2 j 2048 Frequency 0.2 j 2048 Freqency 0.4 Wind speed spectrum. 40 Wind Speed Spectrum 30 Coefficient r j 20 10 0 0 0.3 Hourly averaged wind speed data during summer time in San Francisco.3 0.
5 Hourly averaged wind direction data during summer time in San Francisco.6 Wind direction spectrum.4 0. rk = ck c0 (7. 600 Wind Direction Spectrum 400 Coefficient r j 200 0 0 0.1 0.3 0. June.Weather Modeling for Forecasting Transmission Line Ampacity 119 400 Signal: Wind Direction Wind Direction Degree 300 Wd j 200 100 0 0 500 1000 j Hour 1500 2000 FIGURE 7.2 j 2048 Frequency 0. Ambient Temperature Pattern June 35 30 25 20 15 10 5 Ambient o Temperature.5 FIGURE 7.7 Ambient temperature pattern. C 0 0 4 8 12 16 20 Hour FIGURE 7.5) .
C 0 0 4 8 12 16 20 Hour FIGURE 7.120 Powerline Ampacity System: Theory. The autocovariance ck is given by. Ambient Temperature Pattern August 35 30 25 20 15 10 5 Ambient Temperature. 1. C 0 0 4 8 12 16 20 Hour FIGURE 7. and Applications Ambient Temperature Pattern July 35 30 25 20 15 10 5 Ambient Temperature. July. ck = 1 N ∑ (z – z)(z t t =1 N−k t+k – z ) k = 0. Ambient Temperature Pattern August 35 30 25 20 15 10 5 Ambient Temperatur e. C 0 0 4 8 12 16 20 Hour FIGURE 7. Modeling. August.9 Ambient temperature pattern.6) zt = average hourly value of the meteorological variable at time t .10 Ambient temperature pattern. September. 2 K n lags (7.8 Ambient temperature pattern.
7) N = number of observations in the time series .13 Wind speed pattern. z= ∑z t =1 N t N (7.Weather Modeling for Forecasting Transmission Line Ampacity 121 Wind Speed Pattern June 10 Wind Speed. m/s 8 6 4 2 0 0 4 8 12 16 20 Hour FIGURE 7.11 Wind speed pattern. July. The process mean is.12 Wind speed pattern. Wind Speed Pattern August 10 Wind Speed. Wind Speed Pattern July 10 Wind Speed. m/s 8 6 4 2 0 0 4 8 12 16 20 Hour FIGURE 7. August. m/s 8 6 4 2 0 0 4 8 12 16 20 Hour FIGURE 7. June.
Ambient Temperature Autocorrelations 1. The daily maximum and minimum values forecast by the weather . Wind Speed Autocorrelations 1. m/s 8 6 4 2 0 0 4 8 12 16 20 Hour FIGURE 7.4 0. Hourly values of future weather data are generated by the model by adjusting the coefﬁcients with general purpose weather forecast data.16 Wind speed autocorrelations.6 0.2 Autocorrelation 1 0. Modeling.6 0.2 0 0 12 24 36 48 Lag.2 0 0 12 24 36 48 Lag.1) for the prediction of hourly values of ambient temperature and wind speed is that it does not require continuous input of weather data.2 Autocorrelation 1 0.4 0. and Applications Wind Speed Pattern September 10 Wind Speed. hr FIGURE 7.8 0.8 0. One of the important advantages of a Fourier series model (Figure 7.122 Powerline Ampacity System: Theory. September.15 Ambient temperature autocorrelations. hr FIGURE 7.14 Wind speed pattern.
1 0 0.1 0. hr FIGURE 7.2. Transmission line capacities are predicted in advance by the LINEAMPS program by taking into account weather forecast data and the weather models developed in the previous section. The model is also useful for simulation purpose as shown in Figure 7.18 Selection of number of harmonics in the Fourier series model of ambient temperature. Utilities and power producers are making advance arrangements for the purchase and sale of electricity.20. Due to deregulation in the electric utility industry there is even greater competition for the supply of electric energy.4 0 24 48 Lag. Figure 7. . and as discussed in Chapter 5.3 REALTIME FORECASTING As stated earlier.19 shows hourly values of ambient temperature generated by the Fourier series model in comparison with measured data for one week during summer in the San Francisco Bay area. 7.17 Ambient temperature and wind speed crosscorrelations.3 0.2 0. forecasting of transmission line ampacity several hours in advance is beneﬁcial for the advance planning of generation and transmission resources. service are used to adjust model coefﬁcients using Figure 7.2 0.4 CrossCorrelations 0.3 0. Selection of Numbers of Harmonics Ambient Temperature Model 20000 Sum of Squares 15000 10000 5000 0 0 1 2 3 4 5 Numbers of Harmonics FIGURE 7.Weather Modeling for Forecasting Transmission Line Ampacity 123 Ambient Temperature and Wind Speed CrossCorrelations 0. which requires ensuring adequate transmission capacity.
wind speed. The following recursive algorithm is developed for the prediction of hourly values of ambient temperature.3. Modeling. It is a recursive algorithm that calculates future values of the meteorological variables consisting of ambient temperature.3.124 Powerline Ampacity System: Theory. and Applications Forecasting Ambient Temperature from Daily Max/Min Forecast Weather Data 40 35 30 25 Temperature C . C 50 40 30 20 10 0 0 24 48 72 Hour 96 120 144 168 FIGURE 7.31. Ambient Temperature Simulations Temperature. 7. wind speed. . and wind direction based on previous measurements of these variables. The predicted values of meteorological variables are then entered into the transmission line heat balance equation to calculate line ampacity. An example of forecasting hourly values of line ampacity up to seven days in advance by the LINEAMPS program is given in the Figure 7.19 Forecasting hourly values of ambient temperature.1 FORECASTING AMPACITY FROM WEATHER PATTERNS The algorithm for forecasting line ampacity by Fourier series weather patterns is given in the ﬂow chart of Figure 7. Deg. which is suitable for realtime predictions.20 Simulation of hourly values of ambient temperature by Fourier series model and a second order autoregressive stochastic model. 7. 20 15 10 5 0 5 10 0 24 48 72 96 Hour 120 144 168 Measured Predicted Error FIGURE 7.32. and wind direction. it is possible to forecast hourly values of weather data on an hourbyhour basis by the application of the Kalman ﬁlter algorithm.2 REALTIME FORECASTING OF TRANSMISSION LINE AMPACITY When realtime weather data is available continuously.
9) where . The Auto Regressive Moving Average process with exogenous Variables (ARMAV) was selected to model z(t) as given below.t)S(t) (7. This is carried out recursively by the Kalman ﬁlter algorithm.10) the problem now becomes that of estimating xT(t) at each instant (t) based on the measurement y(t). p(t) = m(t) + fi(t) sin(iωt) + gi(t) cos(iωt) where. z(t). m(t). gi(t) … gnh(t)} (7. u(t). y(t) = p(t) + z(t) (7. p(t). wind speed. fi(t) … fnh(t). and s(t) represent the output. m(t) = process mean fi. comprising ambient temperature.2 … nh are the coefﬁcients of the model nh = number of harmonics ω = 2π/24 = fundamental period Writing the parameter vector compactly.11) (7. and a stochastic component.t) = 1 + b1(t) q–1 + b2(t) q–1 + … + bnb(t) q–1 C(q–1.t)U(td) + C(q–1. and wind direction are considered to be composed of a periodic component. xT(t) = {a1(t) … ana(t). is represented by a Fourier series given by. z(t).t) = 1 + a1(t) q–1 + a2(t) q–1 + … + ana(t) q–1 B(q–1.Weather Modeling for Forecasting Transmission Line Ampacity 125 The measurements. b1(t) … bnb(t). A(q–1.t) = 1 + b1(t) q–1 + b2(t) q–1 + … + cnb(t) q–1 In the above equations. input.t).8) where y(t) represents the hourly values of measurement of ambient temperature.t). . C(q–1. The periodic term. wind speed. respectively. y(t).t) are timevariable polynomials in the backward shift operator q–1: A(q–1. gi i = 1. and white noise sequence. c1(t) … cnp(t). and wind direction. A(q–1.t)z(t) = B(q–1. B(q–1. p(t). Hc. or a coefﬁcient of heat transfer.
sin(2ωt). … u(tdnw).16) (7. … cos(nhωt)} (7. cos(ωt). u(td2). x(t+1) = x(t) + v(t) where v(t) is a sequence of independent gaussien random vector. and. respectively. HT = { y(t1).7) – (7.3 KALMAN FILTER ALGORITHM State Update Equation x(t) = x(t1) + n(t)k(t) Innovations n(t) = y(t) .n(tnf). sin(ωt).15) . and Applications State Equation The true value of the parameter vector is assumed to vary according to. y(t) = HT(t) x(t) + e(t) where the matrix H (actually a row vector) is given by. … y(tnp). the problem of parameter estimation is reduced to the problem of state estimation.126 Powerline Ampacity System: Theory. … . 1.HTx(t1) Kalman Gain k(t) = p(t1)H(t)[R2(t) + HT(t)p(t1)H(t)]1 (7. Modeling. (70) constitute the state and measurement equation. u(td1). 1. y(t2). n(t1). cos(2ωt). The Kalman ﬁlter algorithm can now be applied to estimate the state vector x(t).14) (7.13) (7. … sin(nhωt).12) Equations (69). n(t2). therefore. 7.17) (7.11) we may write the measurement equation as. Measurement Equation From equations ( 7.3.
21 Recursive estimation of San Francisco Bay area ambient temperature during summer time by the application of Kalman ﬁlter algorithm. RECURSIVEESTIMATIONOFAMBIENT TEMPERATURE 35 30 25 Temperature.24. ˚C 20 15 10 Measured 5 Predicted 0 1 21 41 Hour 61 81 101 FIGURE 7. For example. Results obtained by the application of the above neural networks are presented in Figures 7. 7.21. published by PrenticeHall in 1999 recommends the Kalman ﬁlter for realtime recurrent learning. and an unsupervised neural network using Kohonen’s learning algorithm. a recent book by Simon Haykin. Neural networks consist of many simple elements called neurons that are linked by connections of varying strengths as shown in Figure 7.4 ARTIFICIAL NEURAL NETWORK MODEL Neural network is an important subject of research in artiﬁcial intelligence where computations are based on mimicking the functions of a human brain.Weather Modeling for Forecasting Transmission Line Ampacity 127 where the error covariance matrix p(t) is given by. An unsupervised neural network is also developed for weather pattern recognition by using Kohonen’s learning algorithm.22. The neural networks used in numerical analysis today are gross abstractions of the human brain. A neural network is trained to forecast hourly values of ambient temperature and wind speed by using the back propagation algorithm. are the two types of neural networks that were used in weather modeling for line ampacity predictions. The brain consists of very large numbers of far more complex neurons that are interconnected with far more complex and structured couplings (Haykin. 1999). .18) The results obtained by the application of the above algorithm* to predict hourly values of ambient temperature are shown in the Figure 7. Neural Networks: A Comprehensive Foundation. p(t ) = p(t – 1) + [R (t) – p(t – 1)H (t)p(t – 1)] [R (t) + H (t)p(t – 1)H(t)] T 1 T 2 (7.23 and 7. * There is recent interest in the application of the Kalman ﬁlter algorithm for the efﬁcient solution of nonlinear recurrent neural networks for realtime prediction. A supervised neural network using the back propagation algorithm.
. NY. (1994). 2. 25 20 15 10 5 0 1 48 Hours 95 Measured ANN S/D Model FIGURE 7. Macmillan. The network trains by supervised learning using the back propagation algorithm. p. S. * Haykin. Knowledge is acquired by the network through a learning process.* a neural network is a massively parallel distributed processor that has a natural propensity for storing experiential knowledge and making it available for use. Modeling. Neural Networks: A Comprehensive Foundation. 22 A three layer Artificial Neural Network with three inputs and three outputs and a hidden layer. and Applications Input Layer Hidden Layer Output Layer Input 1 Output 1 Input 2 Output 2 Input 3 Output 3 Figure 7. Neural network results are compared to a statistical forecasting model and actual data.22 A threelayer artiﬁcial neural network with three inputs and three outputs and a hidden layer. FIGURE 7. 2.128 Powerline Ampacity System: Theory. It resembles the brain in two respects: 1.23 Example of neural network application to forecast next hour ambient temperature in the San Francisco Bay area. According to Haykin. TEMPERATURE FORECASTING BY ARTIFICIAL NEURAL NETWORK (ANN) 40 35 30 Temperature. Interneuron connection strengths known as synaptic weights are used to store the knowledge.
there is no target response with which to compare output.24 Application of unsupervised neural network for ambient temperature pattern recognition in the San Francisco Bay area. Unsupervised Learning In unsupervised learning there is no teacher.23. where network output is compared to a target and the difference is used to adjust the strength of the connections.1 TRAINING OF THE NEURAL NETWORK A neural network is ﬁrst trained by feeding it with data. If the training set is sufﬁciently large. the network provides the correct output from a set of input data. Unlike a lookup table. 7. It is therefore clear that a neural network is particularly useful to predict the outcome of a phenomenon that cannot be formulated otherwise. The back propagation algorithm is presented below. Unsupervised learning is carried out by Kohonen’s learning algorithm. Training is completed when the sum of squares of errors is minimized. in other words.2 SUPERVISED AND UNSUPERVISED LEARNING Learning is the key to AI—the artiﬁcial neural network learns from data and demonstrates intelligent capability. from which it learns the inputoutput relationship of a system. The network organizes by itself (selforganizing neural network) and learns to recognize patterns within data. learning or training is carried out by a back propagation algorithm. then the neural net will provide the correct output from a set of input data that is different from the training set. Learning in an artiﬁcial neural network is either supervised or unsupervised.4. the dynamics of the system are represented by a trained neural network. and an application of this algorithm to forecast hourly ambient air temperature in the San Francisco Bay area is shown in Figure 7. 7. C 20 o 15 10 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Hour of Day FIGURE 7. Supervised Learning In a supervised neural network. A neural network is different from a lookup table. which is used for .Weather Modeling for Forecasting Transmission Line Ampacity 129 Pattern Recognition By Neural Network 30 25 Temperature.4. Once trained.
130
Powerline Ampacity System: Theory, Modeling, and Applications
pattern recognition and data classiﬁcation. In Kohonen’s learning algorithm the strength of interconnections is adjusted by minimizing the Euclidean distance of output neuron. The output neuron having the minimum Euclidean distance is declared the winner and set to 1; all others are set to 0. For the above reasons, a selforganizing neural network is also called a “winnertakeall” algorithm, because when the network is trained only a certain output will go high, depending upon the characteristics of the input vector. During training, the weights of the connections are adjusted until subsequent iterations do not change weights. A winnertakeall selforganizing neural network due to Kohonen’s learning algorithm is presented below, and its application to pattern recognition of daily ambient temperature is shown in Figure 7.24. The above types of neural networks are examples of nonrecurrent networks. Another important type of network is the recurrent network due to Hopﬁeld, where there is continuous feedback from output to input. Recurrent networks ﬁnd applications in nonlinear optimization problems whose solutions are difﬁcult by conventional means.
7.4.3
BACK PROPAGATION ALGORITHM*
The back propagation algorithm is composed of the following steps: 1. Apply an input vector x 2. Calculate the error e between the output vector y and a known target vector z e=
∑ [z(1) – y(1)]
l =1
n
2
n = length of training vector
3. Minimize errors ∂e ∂e ∂z = ⋅ = δ( l ) ∂w( j, i) ∂y ∂w(l, j) 4. Calculate error signal of output layer δ(l) δ(l) =[z(l) – y(l)] · y(l) · [l – y(l)] 5. Calculate error signal of input layer δ(j) δ( j) = y( j) ⋅ 1 – y( j)
[
]∑ w(l, j)δ(l)
l=0
l = nl
* Rumelhart, David E., McClelland, James, L., Parallel Distributed Processing, Volume 1, M.I.T. Press, Cambridge, MA, 1988.
Weather Modeling for Forecasting Transmission Line Ampacity
131
6. Update weights by the learning rule w(j, i) new = w(j, i) old + δ(j) · y(i) + α[∆w(j, i)old] w(l, i) new = w(l, i) old + δ(l) · y(l) + α[∆w(l, i)old] Steps 1 through 6 are repeated until the sum of squares of errors is minimized. A neural network for the prediction of hourly values of ambient temperature is developed by using the back propagation algorithm and the results are presented in Figure 7.23.
7.4.4
UNSUPERVISED NEURAL NETWORK TRAINING ALGORITHM*
The unsupervised neural network training algorithm is due to Kohonen and is composed of the following steps: 1. Apply an input vector x 2. Calculate the Euclidean space d(j) between x and the weight vector w of each neuron as follows:
d( j) =
∑ [(x(i) – w(i, j))]
i=l
n
2
n = number of training vector x w(i, j) = weight from input i to neuron j 3. The neuron that has the weight vector closest to x is declared the winner. This weight vector, called wc, becomes the center of a group of weight vectors that lie within a distance d from wc. 4. Update nearby weight vectors as follows: w(i, j) new = w(i, j) old + α[x – w(i, j) old] where, α is a timevarying learning coefﬁcient normally in the range 0.1 < α < 1. It starts with a low value of 0.1 and gradually increases to 1 as learning takes place. Steps 1 through 4 are repeated until weight change between subsequent iterations becomes negligible. A selforganizing neural network is also developed for ambient temperature pattern recognition by using Kohonen’s learning algorithm, and the results are presented in Figure 7.25.
* Wasserman, Philip D. 1989 Neural Computing, Van Nostrand, New York.
132
Powerline Ampacity System: Theory, Modeling, and Applications
7.5 MODELING BY FUZZY SETS
Fuzzy Set Theory was introduced by Professor Lotﬁ Zadeh* of the University of California, Berkeley during the 1960s. Fuzzy set theory accepts many valued logic and departs from the classical logic of Aristotle which allows a proposition to be either true or false. The idea of manyvalued logic was developed by Jan Lukasiewicz, a Polish logician in the 1920s and applied by Max Black in 1937. Zadeh formally developed multivalued set theory in 1965 and called it fuzzy set theory.
µ (wind speed)
Very Low 1.0
Low
Medium
High
Very High
0.5
0
1
2
3
4
5
6
7
wind speed, m/s
FIGURE 7.25 A fuzzy set of wind speed is represented by four fuzzy levels: “very low,” “low,” “medium,” “high,” “very high.” By allowing varying degrees of membership, fuzzy sets enables the process of decision making better under uncertainty.
The main idea of fuzzy sets is that they allow partial membership of an element in a set, as shown in Figure 7.25. A fuzzy set F in a universe of discourse U is deﬁned to be a set of ordered pairs, F = {(u, m F(u)) u ∈ U} (7.19)
where m F(u) is called the membership function of u in U. When U is continuous, F can be written as, F=
∫µ
u n
F
( u) u
(7.20)
and when U is discrete, F is represented as, F=
∑µ
i=l
F
u
(7.21)
where n is the number of elements in the fuzzy set F.
* Zadeh, L. A., “Fuzzy sets as a basis for a theory of possibility.” Fuzzy Sets and Systems, 10, (3), 243–260, 1978.
Weather Modeling for Forecasting Transmission Line Ampacity
133
7.5.1
LINGUISTIC VARIABLES
Fuzzy set theory enables modeling a system in the natural language by making use of linguistic variables. A linguistic variable is characterized by a quintuple, (x, T(x), U, G, M) where, x = name of variable [Example: wind speed] T(x) = Term set of x [Example: T(wind speed) = {very low, low, medium, high, very high}] U = Universe of discourse [Example: U(wind speed) = (0, 7) m/s] G = Syntactic rule for generating the name of values of x M = Semantic rule for associating a meaning with each value The terms Very Low, Low, Medium, High, and Very High wind speeds represent fuzzy sets whose membership functions are shown in of Figure 7.25. If A and B are fuzzy sets with membership function µA(u) and µB(u), respectively, then the membership function of the union, intersection, and complement, and the fuzzy relation involving the two sets, are as follows: Union (AND) Example: IF A AND B THEN C µC(u) = µA∪B = max(µA(u), µB(u)) Intersection (OR) Example: IF A OR B THEN C µC(u) = µA∩B = min(µA(u), µB(u)) Complement (NOT) Example: NOT C µ c( u) = 1 – µc( u) u∈U u∈U
FUZZY RELATION
Two or more fuzzy IF/THEN rules of the form, Y is Bi IF X is Ai, i = 1,2,… n can be connected by the fuzzy relation R,
134
Powerline Ampacity System: Theory, Modeling, and Applications
R=
∑ (A xB )
i i i
(7.22)
where the membership function of the cartesian product (Ai x Bi) is given by, µ A xKKl A n ( u1 , KK u n ) = min µ A1 ( u l ), KK,µ A1
{
(7.23)
One of the most successful applications of fuzzy sets is in the design of Fuzzy Logic Controller (FLC). A fuzzy logic controller may be developed to control tranmission line ampacity. The steps in FLC design follow: • Deﬁne input and control variables. For example, in ampacity calculation, the input variables are wind speed and direction, solar radiation, and air temperature. The control variable is ampacity. • Describe the input and control variables as fuzzy sets (fuzziﬁcation). • Design the rule base (fuzzy control rules). • Develop the fuzzy computational algorithm and fuzzy output. • Transform fuzzy outputs to crisp control actions (defuzziﬁcation). An important step in the design of FLC is the selection of membership function and fuzzy IF/THEN rules. Generally, they are obtained by experimentation with data. More recently, neural networks have been used to learn membership function and the rules. Figure 7.26 is a schematic representation of a fuzzy logic controller with learning by neural networks.
FLC Membership Function and Rules
Transmission line
Output
Neural Network
FIGURE 7.26 Schematic representation of fuzzy logic controller that learns from a Neural network.
An example of a fuzzy logic system for the calculation of transmission line ampacity is given in Figure 7.27. Only two rules are shown in the ﬁgure for illustration of main concepts, and a calculation to obtain a crisp value of line ampacity from fuzzy sets is given below. This is an excellent example of fuzzy logic because the meteorological variables comprising wind speed and ambient temperature, are best described by fuzzy sets. In the example, wind speed, ambient temperature and ampacity are represented by fuzzy sets having four fuzzy levels (very low, low, medium, high, very high). In this manner, we can represent weather parameters by linguistic variables and consider the uncertainty in weather data comprising wind speed and ambient temperature accurately. Furthermore, weather forecast data from
From Rule 1. the membership of input T denoted as mVL(T).Weather Modeling for Forecasting Transmission Line Ampacity 135 the National Weather Service or other sources are normally presented in a similar manner.27 Fuzzy logic system of calculation of transmission line ampacity. then Ampacity (A) is high. then Ampacity is very high. . We are required to calculate ampacity from the above data. The following two rules are activated: Rule 1: Rule 2: If Ambient Temperature (T) is very low and Wind Speed (W) is medium. is obtained from Figure 7. If Ambient Temperature is low and Wind Speed is high. Temperature (T) Wind Speed (W) Ampacity (A) Very Low (T) Input (T) Input (W) Medium (W) High (A) Rule 1 Input (T) Low (T) High (W) Input (W) Very High (A) Rule 2 Final value of Ampacity (A) Rule 1: If Ambient Temperature (T) is Very Low and Wind Speed (W) is Medium then Ampacity (A) is High Rule 2: If Ambient Temperature is Low and Wind Speed (W) is High then Ampacity (A) is Very High Fuzzy centroid (A) FIGURE 7. Example of Ampacity Calculation by Fuzzy Sets Typical input data for ampacity calculation may be as follows: Sunny. The degree of membership of fuzzy variables T and W in the four fuzzy levels is given in Table 1. and the membership of input W denoted as mM(W).27. wind northwest 8–12 km/h. which can be directly utilized to calculate transmission line ampacity as shown in Figure 7. temperature low 2–15°C.27.
0.75.27. . mL(T) = 0. and the membership of input W (mH(W)).0.6.5.5. p A = Sf ∑ y ⋅ m (y ) j o j j= l ∑ m (y ) o j j= l p y = (1.55 Similarly. mo(yj ) = (0. Rule 2 activates the consequent fuzzy set VH of A to degree mVH(A) = min(0. from Rule 2. 3.5.0. 5. are obtained from Figure 7.136 Powerline Ampacity System: Theory. 0.55.0.5) Where the elements yj of vector y are the mean value of each fuzzy level. and Applications TABLE 1 Var/Level T (2–15) °C W (2.6 0 M 0 0. the combined output fuzzy set of A.6 Therefore.75) = 0. Rule 1 activates the consequent fuzzy set H of A to degree mH(A) = min(0.55 H 0 0. the membership of input T (mL(T)).55) = 0.6).75 0 L 0. A crisp value of ampacity is obtained by centroid defuzziﬁcation of the combined output fuzzy set.2) m/s VL 0.75 mM(W) = 0.55 Therefore. Modeling.75 VH 0 0 mVL(T) = 0. 4.5–3.75 Therefore.6 mM(W) = 0.0. 2.
55 + 5.6 A = Sf x 5.27) (7. Sf = 400.5 × 0 + 1. Sn = the component of solar radiation that is normal to the surface of the earth θ = Angle of deviation from the normal θz = Zenith angle.6 SOLAR RADIATION MODEL During daytime. Sb = Sn · cos(θ) The diffused radiation Sd is calculated as.55 + 0.5 × 0 + 2. cos(θz) = sin(φ) sin(δ) + cos(φ) cos(δ) cos(ω) (7. 7. Es = αs(Sb + Sd) Where. αs = solar absorption coefﬁcient. the temperature of the conductor may increase by 1–10°C by solar heating alone.26) (7. given by.25) (7. the transmission line conductor is heated by the energy received from the sun. Depending upon the condition of the sky and the position of the sun with respect to the conductor.02 Ampacity = 2008 A Since the membership functions given in Figure 7. the energy received on the surface of the conductor from the sun (Es) is obtained as.5 × 0.26 are based on scaled values the actual value of ampacity for a transmission line having an ACSR Cardinal conductor is obtained by multiplying with a scaling factor. 0 < αs < 1 Sb = Solar energy received by conductor due to beamed radiation Sd = Solar energy received by conductor due to diffused radiation The beamed radiation Sb is calculated as.Weather Modeling for Forecasting Transmission Line Ampacity 137 A = Sf 0. Sd = Sn · cos(θz) Where.5 × 0. To calculate conductor heating by solar radiation.24) The normal component of the beamed solar radiation inside the earth’s atmosphere is obtained as. .6 0.5 × 0 + 4.
0033 ⋅ Cos 365 (7. John A.4237 – 0.365 day number Sc = Solar constant = 1353 W/m2 (measured value outside the earth’s atmosphere) τb = atmospheric transmittance of beamed radiation.138 Powerline Ampacity System: Theory. 1980 Solar Engineering of Thermal Processes.. William A.5055 – 0.31) * Dufﬁe.00595(6.28) The normal component of the diffused solar radiation inside the earth’s atmosphere is obtained as.2710 – 0. Modeling.5Alt)2 Alt = Altitude of the transmission line above MSL. Snd = Sn(o) · τd (7.2.29) Sn(o) = Normal component of the solar radiation outside the earth’s atmosphere which is obtained as.2939τb a0 = 0.5Alt)2 k = 0. New York.00821(6Alt)2 a1 = 0.* –k τ b = a 0 + a 1 exp Cos(θ z ) τd = atmospheric transmittance of diffused radiation given by. It takes into account attenuation by the earth’s atmosphere of the extraterrestrial radiation and is given as.30) Jd = 1. 360 ⋅ J d S n ( o ) = Sc 1 + 0.33) (7. cos(θ) = sin(δ)sin(φ)cos(β)sin(δ)cos(φ)sin(β)cos(γ) – sin(δ)cos(φ)sin(β)cos(γ) + cos(δ)cos(φ)cos(β)cos(ω) + cos(δ)sin(φ)sin(β)cos(γ)cos(ω) + cos(δ)sin(φ)sin(β)cos(γ)cos(ω) + cos(δ)sin(β)sin(γ)sin(ω) (7. John Wiley & Sons.2711 – 0. and Beckman.01858(2. τd = 0. . and Applications Snb = Sn(o) · τb (7. km The angle of deviation θ of the beamed radiation with respect to the normal to surface of the conductor is given by the following formula.32) (7.
weather modeling for the prediction of transmission line ampacity is presented ﬁrstly by Fourier analysis of weather data. 7. north +ve: –90° =< φ =< 90° δ = Declination. East ve. South zero.36) Lstd = Longitude of Standard Meridian (Example: Lstd San Francisco = 120°) Lloc = Longitude of location Eqt = Equation of time = 9. afternoon +ve. Ambient temperature and .34) β = Slope. Angular displacement of the sun east or west of the local meridian due to rotation of the earth on its axis at the rate of 15°/hr. Angle between north or south of the equator. The declination angle is calculated as. West +ve: –180° =< γ=< 180° ω = Hour angle.29. The hour angle is given by.5°.87sin(2B) –7. north +ve: –23.30. Angle between the conductor axis and the horizontal: 0 =< β =< 180° γ = Line orientation angle (azimuth). (284 + J d ) δ = 23. and SolarGen.45360 365 (7.38) θ = Angle of incidence.5sin(B) 360( J d – 81) 364 (7.31 is a ﬂow chart showing the line ampacity forecasting procedure from weather models AmbientGen.7 CHAPTER SUMMARY In this chapter.28 and 7.35) (7. Angular position of the sun at solar noon with respect to the plane of the equator. Figure 7. These angles are shown in Figures 7.Weather Modeling for Forecasting Transmission Line Ampacity 139 φ = Latitude. The result of solar radiation calculation by the program is for one day during the month of July in the region of San Francisco and is shown in Figure 7.37) B= (7.53cos(B) –1. morning –ve. WindGen. The angle between the beam radiation on a surface and the normal to the surface. ω = (12 – SolarTime) ⋅ 15 SolarTime = StandardTime + 4(Lstd–Lloc) + Eqt (7.5° =< δ =< 23.
July 1400 1200 1000 800 2 W/m 600 400 200 0 0 4 8 12 16 20 24 Time of Day hr . and Applications Atmosphere Sun Beamed Radiation 0 90 Normal Conductor Tower Equator Earth 0 FIGURE 7. FIGURE 7. calculated by program for one day during the month of July. and a realtime forecasting algorithm is presented that uses a recursive estimation procedure for the prediction of ambient temperature and wind speed. .29 Slope angle between conductor and horizontal.28 Transmission line solar angles. Modeling. Global Solar Radiation San Francisco. wind speed models were developed by ﬁtting Fourier series to hourly weather data available from the National Weather Service. The forecasts are adapted to the daily high and low values of ambient temperature that are forecast daily by the National Weather Service.140 Powerline Ampacity System: Theory. FIGURE 7.30 Global Solar Radiation (Direct Beam + Diffused Radiation) on a transmission line conductor surface in San Francisco. A Kalman ﬁltertype algorithm is then used to model the uncertainty in the Fourier series. CA. Transmission Line EW direction and the slope angle is 5 °.
Neural networks offer an alternative method of forecasting weather variables and pattern recognition. and a selforganizing neural network is developed using Kohonen’s learning algorithm. Neural network models for prediction and weather pattern recognition are developed by using the back propagation algorithm. Analytical expressions for the calculation of hourly values of solar radiation are also developed which take into account transmission line location and line geometry. Sr Method AmbientGen generates hourly vcalues a(t) T Method WinGen generates hourly values Ws(t) Method SolarGen generates hourly values Sr(t) Method Steady Static Ampacity generates hourly values of Transmission Line Ampacity at each line segment (Ij. FIGURE 7. New Zealand.Weather Modeling for Forecasting Transmission Line Ampacity 141 Input for ecast weather data: Ta.32 Forecasting hourly ampacity values of a 220 kV transmission line seven days in advance in the region of North Island.t ) FIGURE 7.t ) Calculate the minimum value of ampacity of all line segments Min (Ij. Ws. . Basic fuzzy logic concepts are presented and a system for developing a fuzzy logic controller of transmission line ampacity is proposed for further research.31 Flow chart for forecasting transmission line ampacity from weather models.
Examples of weather models developed by the program are shown for the region of San Francisco. and Applications For realtime prediction of transmission line ampacity. Modeling. .142 Powerline Ampacity System: Theory. a recursive estimation algorithm for weather forecasting is developed based on Kalman ﬁlter equations.
forests. Modeling such a system is not easy. IEEE Computer Application in Power. and water. mountains. The end product is an intelligent line ampacity system resembling a human expert. wind. structures that support conductors in air. 143 . Line ampacity is calculated from steady. This is the object of computer modeling of line ampacity system as described in this chapter. and rain. Number 3. Some sections of a line may be exposed to industrial pollution as well. A systematic approach using practical knowledge and simplifying assumptions is required to achieve a realistic line ampacity system with sufﬁcient accuracy. Objects are used to model the transmission line system and its environment.2 THE LINEAMPS EXPERT SYSTEM LINEAMPS* is an expert system computer program developed by the author based on a systematic approach of objectoriented modeling and expert rules. and other hardware necessary for the attachment of conductors to the tower. It is also a humble contribution and a * Anjan K.1. July 1995. insulators to safely protect transmission tower structures from the high voltage conductor. This is the object of the line ampacity expert system program. deserts. In this chapter the LINEAMPS (Line Ampacity System) transmission line expert system computer program is developed by the integration of theory and practical knowledge of the transmission line system. 8. temperature. Expert system rules are used to incorporate practical knowledge. They are also exposed to varying atmospheric conditions—sun. and transient thermal models by developing an object model and expert rules of the transmission line system. 8. The transmission line environment is vast as they traverse different kinds of terrain—plains.8 Computer Modeling 8. Deb. Examples of objectoriented modeling and expert system rules are presented here to demonstrate how practical knowledge is embedded in program. All of these environmental factors affect line capacity to a certain degree.1 INTRODUCTION For the proper evaluation of transmission line capacity. connectors for the splicing of conductors.1. it is necessary to support the theory developed in Chapter 4 by a practical knowledge of the transmission system and its environment. Volume 8.1 FROM THEORY TO PRACTICE A transmission line is composed of the conductors that carry current. dynamic. Object oriented expert system estimates transmission line ampacity. which will enable a transmission line engineer to easily evaluate power line capacity in any geographic region.
several lines may be produced by inheritance. weather station objects are created. weather station object. longitude. the Internet. there is a system of objects for the presentation of data. or even the daily newspaper. * Lawrence Stevens. once a transmission line object is created. data stored in weather objects of the line ampacity system program are environmental data comprising weather conditions. In addition. and elevation. In the following sections the objectmodel and expert system design of the line ampacity system are presented in greater detail. whether it is temporary or permanent. Hayden Book Company. ** G.8. hourly values of transmission line ampacity are forecast several hours in advance.2. which enables them to perform a required action. .1. 1991. 1997.** Its attributes and behavior deﬁne an object. The Search for the Perfect Machine. Methods have the ability to use data contained in their own objects as well as other objects. Intellicorp.7. Once an object is created it is easier to create newer instances of the same object by inheritance.3. CA. and cartograph object. For example. and 8. and the hierarchical structure of transmission line. Booch.4. 1993. By using forecast weather data. ObjectOriented Design with Applications. Data relating to the electrical and mechanical properties of the transmission line are contained in the transmission line object. Objects inherit data as well as methods—this is an important property of all objects. and conductor object of the transmission line ampacity system is shown in the Figures 8.2. Similarly. as well as the stored procedures and weather patterns of the region. for example. Modeling. The object model of the line ampacity system is given in Figure 8.144 Powerline Ampacity System: Theory. and a userfriendly graphical user interface. Artiﬁcial Intelligence.* where a computer system demonstrates intelligent behavior.2 OBJECT MODEL OF TRANSMISSION LINE AMPACITY SYSTEM System modeling by object orientation is a new way of data representation and programming. 8. 8.1 is comprised of transmission line object. and Applications practical demonstration of research in the ﬁeld of artiﬁcial intelligence. the National Weather Service. The KappaPC*** object oriented modeling tool is used to implement the object model. These objects have methods to receive weather forecast data from external sources. *** KappaPC® objectoriented development software v 2. weather station. For example. conductor object. latitudes. terrain. Mountainview. Methods or stored procedures in the objects give them behavior. Objects can store data. The following sections describe the object model and expert rules of the line ampacity system.1 LINEAMPS OBJECT MODEL The object model of the LINEAMPS system shown in Figure 8. Benjamin Cummings Publishing Co.
Powerline objects are classiﬁed by categories of transmission and distribution lines. In each class there are subclasses of lines by voltage levels. 345kV.Computer Modeling 145 FIGURE 8. The distribution line class comprises 15kV. The line object has all the data and methods pertaining to the overhead line that are necessary for the evaluation of powerline ampacity.1 Line ampacity system objectmodel. and other line instances created by the user. 480 V. In each subclass there are several instances of powerlines.2. and methods perform the action of evaluating ampacity. The subclasses of line voltages in the line object classes are deﬁned by the user. Each of these transmission line objects derives its attributes and behavior from the general class of lines. and 230kV lines.2. Following is a complete list of attributes of the line object: Line object attributes: LineName LineVoltage LineLength ConductorCodeName ConductorType ConductorDiameter . Data are stored in slots.2 TRANSMISSION LINE OBJECT The Line object class is shown in the Figure 8. for example. Table 8. 8. Only a partial list of attributes and methods are shown for the purpose of illustration.1 shows an example of the data in one instance of a transmission line object. the transmission line object class is comprised of 500kV.
146
Powerline Ampacity System: Theory, Modeling, and Applications
ConductorArea ConductorAlpha ResistanceAtTemperature Conductor DC Resistance ConductorEmissivity ConductorAbsorbtivity ConductorSpeciﬁcHeat ConductorMass ConductorAluminumMass ConductorSteelMass NumberOfSites SiteList (List) AssociatedWeatherSites(List) Site#x(List), where x =1,2..NumberOfSites AmbientTemperatureSite#x(List) WindSpeedSite#x(List) WindDirectionSite#x(List) SkyConditionSite#x(List) NormalAmpacityOneDay(List) NormalAmpacitySevenDays(List) EmergencyAmpacityOneDay(List) EmergencyAmpacitySevenDays(List) TimeOfDayEnergyPrice(List) Each Site#x comprises a list having the following values: elevation, slope, latitude, longitude, standard longitude, line orientation at the location, and the type of terrain. Line object comprise the methods shown in Table 8.2.
TABLE 8.1 Transmission Line Object: 350kV_Line10
Attributes Name Distance Conductor Duration Region Values San Francisco, Berkeley 50 km acsr cardinal 30 min Coastal Method Ampacity Steady_State_Ampacity Dynamic_Ampacity Transient_Ampacity Draw Line
A graphical representation of transmission line by voltage category and by the type of line, transmission or distribution, greatly facilitates the task of a transmission line engineer to view and modify line data by simply clicking on a transmission line object shown in Figure 8.2. The following is an example of creating a transmission line object, assigning attributes and giving them behavior.
Computer Modeling
147
TABLE 8.2
Method: Amp7Days SteadyStateCurrent SteadyStateTemperature DynamicAmpacity TransientAmpacity DrawLine CheckInput WeatherData AdjustWeather MakeNewLine UpdateLineList MakeSiteData EnergyDeliveryCost Function: Calculates hourly values of line ampacity up to seven days in advance. Calculates steady state current. Calculates steady state temperature. Calculates conductor temperature response due to step change in line current. Calculates conductor temperature response due to short circuit and lightning current. Draws the line in the cartogram window. Veriﬁes the correctness of input data. Obtains data from the associated weather stations. Weather data is adjusted in AmbientTemperatureSite#x slot and WindSpeedSite#x slot based on terrain in Site#x slot. Makes an instance of a new line. Updates the list of lines when a new line is created. Makes virtual weather sites along the route of the line. Calculates hourly values of energy delivery cost based on time of day energy price and forecast ampacity.
FIGURE 8.2 Classiﬁcation of transmission line objects.
148
Powerline Ampacity System: Theory, Modeling, and Applications
Creating the object MakeClass(Transmission, Lines); MakeInstance(350kV_Line10, Transmission); Attributes MakeSlot(350kV_Line10, Name, SanFrancisco_Danville, 50); MakeSlot(350kV_Line10, Distance, 50); MakeSlot(350kV_Line10, Conductor, Cardinal); Behavior MakeMethod(Transmission, Ampacity, [time, ambient, sun, wind, interval]); Action SendMessage(350kV_Line10, Ampacity, [12:00, 20, C, 2, 60]); Result: 1000 A
8.2.3
WEATHER STATION OBJECT
Since LINEAMPS calculates transmission ampacity from weather data, modeling of weather by developing an object model is an important aspect of this program. The purpose of the weather object is to reproduce, as closely as possible by software, the behavior of an actual weather station. This is the main objective of the LINEAMPS weather station object. The behavior of a weather station object is obtained by modeling weather patterns of the region by Fourier analysis from historical weather data and regional weather forecasts prepared daily by the National Weather Service. To enable modeling of a weather station object, it is divided into subclasses of regions and region types so that weather station instances inherit class attributes. Station objects have all of the meteorological data and geographic information required in the calculation of transmission line ampacity. The weather station object hierarchy is shown in Figure 8.3 and is comprised of: 1. Subclass of regions. Example: Region1, Region2… Region#x. 2. Subclass of region types. Example: Coastal, Interior, Mountain, Desert. 3. Instances of weather stations. Example: San Francisco, Oakland, Livermore. Weather station objects have the following attributes. Attributes of weather station object • StationName • AmbientMax(List)
Computer Modeling
149
• • • • • • • • • • • • • • • •
AmbientMin(List) HourAmbientMax HourAmbientMin AmbientPattern#x(List), where x =1,2...12 months WindSpeedMax(List) WindSpeedMin(List) HourWindSpeedMax HourWindSpeedMin WindSpeedPattern#x(List), where x =1,2...12 months WindDirection(List) SkyCondition(List) ForecastTemperature(List) ForecastWind(List) ForecastSolarRadiation(List) Latitude Longitude
A “List” inside parentheses is used to indicate that the attribute has a list of values. Weather station objects comprise the following methods:
TABLE 8.3
Method AmbientGen WindGen SolarGen SelectPattern DisplayAmbient DisplayWind OnLineData MakeNewStation Function Generates hourly values of ambient temperature, (Figure 8.4). Generates hourly values of wind speed, Figure 8.5. Generates hourly values of solar radiation, Figure 7.3 Selects ambient temperature and wind speed pattern of the month Display ambient temperature in a line plot and a transcript image Display wind speed in a line plot and a transcript image Reads weather data downloaded from AmericaOnLine. Makes an instance of a new weather station.
NEW ZEALAND EXAMPLE*
To ﬁx ideas, an example of New Zealand weather station object is presented in Figure 8.3. The object has North and South Island subclasses. North Island is Region 1, and South Island Region 2. Each region is further divided into subclasses of Coastal, Interior, Mountain, and Desert. In each subclass there are instances of weather stations. These instances derive their attributes and behaviors from the general class of weather stations, and their characteristics are reﬁned by the properties of each region Table 8.4 shows data in one instance of a weather station object.
* Deb, Anjan K., LINEAMPS for New Zealand, A Software User’s Guide, 1996.
150
Powerline Ampacity System: Theory, Modeling, and Applications
FIGURE 8.3 Classiﬁcation of weather station objects.
TABLE 8.4 Weather Station Object: Wellington Example
Attribute Name Latitude Longitude Elevation Region Value Wellington 41° 18 ‘ S 174° 47’ E 100 Coastal Method Ambient_Gen Wind_Gen Solar_Gen Draw_on_Map Show_Value
Examples of Ambient_Gen and Wind_Gen methods used to generate hourly values of ambient temperature and wind speed are shown in Figures 8.4 and 8.5. Seven days’ forecast weather data from the National Weather Service are shown in Figure 8.6.
8.2.4
CONDUCTOR OBJECT
A conductor object class is shown in Figure 8.7. It is comprised of the following subclasses of conductor types: AAAC, AAC, ACAR, SSAC, ACSR_AW, ACSR, ACSR_TW, COPPER, COPPERWELD and ACSR_INVAR. Each conductor type has plurality of conductor instances. The user may also create other subclasses of conductor types and new instances of conductor objects.
4 Temperature modeling.Computer Modeling 151 FIGURE 8. FIGURE 8. FIGURE 8.5 Wind speed modeling. .6 National Weather Service seven day weather forecast.
2.7 Classiﬁcation of transmission line conductor objects.8. Modeling. 8. and Applications FIGURE 8.9. as seen in Figure 8.152 Powerline Ampacity System: Theory. . Conductor object has following attributes: Attributes of Conductor object: Conductor code name Conductor type Conductor diameter Conductor area DC resistance of conductor Emissivity of conductor Absorptivity of conductor Speciﬁc heat of conductor Conductor mass Aluminum mass Steel mass Conductor object has following methods: Data pertaining to one instance of a transmission line conductor is shown in Figure 8.5 CARTOGRAPH OBJECT A cartograph window is used to show the location of weather stations and the transmission line route in a geographic map of the region.
and by specifying the type of terrain through which the line passes.Computer Modeling 153 TABLE 8. Similarly.8 Data in a transmission line conductor object. one obtains a better picture of the transmission line route and its environment.9. A transmission line appears on the map when the line is selected from the database. Makes an instance of a new conductor FIGURE 8.5 Methods SpeciﬁcHeat SteadyStateCurrent SteadyStateTemperature DynamicAmpacity DynamicTemperature TransientAmpacity MakeNewConductor Function Calculates the speciﬁc heat of conductor Calculates steady state current Calculates steady state temperature Calculates conductor ampacity n the dynamic state. The maximum and minimum ambient temperatures of the day are also displayed at the location of each weather station. weather station objects are also created. Calculates conductor temperature versus time in the dynamic state. A unique feature of the program realized by objectoriented modeling is the ability to create new lines and weather stations by inheritance. It is generated automatically by the program with a DrawLine method using data stored in the transmission line object shown in Figure 8. Calculates conductor temperature versus time in the transient state. . By displaying the transmission line in a map. A transmission line object is created by entering the latitudes and longitudes of the line at discrete intervals.
. and Applications FIGURE 8. This declarative style of rule based programming offers an alternative to the traditional procedural programming method of solving problems.154 Powerline Ampacity System: Theory. Reasoning is then carried out automatically by an inference engine.. California USA is shown in the cartograph window. Modeling. The cartograph shows the geographic boundary of the region that was created by program.* An expert system is generally composed of the following: • • • • • • Goals. MA. facts. A Guide to Expert Systems. Reading. The rules and the data in the objects describe the problem. which ﬁnds a solution by using a backward or forward chaining mechanism. 8.9 A cartograph is shown in the right window of the LINEAMPS program and the corresponding weather station objects are shown in the left window.3 EXPERT SYSTEM DESIGN The line ampacity expert system is accomplished by a system of rules and goals to be achieved by the program. Donald A. Expert systems are capable of ﬁnding solutions to a problem by a description of the problem only. database Rules or knowledge base Inference engine (reasoning capability) Explanation facility Man machine interface Learning capability * Waterman. to solve a problem by rules. daily maximum and minimum values of ambient temperature (max/min) and the trace of a selected 230 kV transmission line from Sacramento to Livermore. The location of weather stations. For example. AddisonWesley. we specify what rules to apply and a goal. 1986.
IEEE Computer Application in Power. rules. Result: In the above example. K. }. In the KappaPC objectoriented development environment.3. These features were used in the program to check user input data and explain error messages. An example of a man machine interface was given in a previous IEEE publication.1 GOALORIENTED PROGRAMMING Goaloriented programming by rules greatly facilitates the task of computer programming as the programmer is not required to code a detailed logic to solve a problem. July 1995. a programmer must precisely code the logic. rules are easily understood and maintained. In addition. inference engine. The SteadyStateGoal was satisﬁed and the program correctly evaluated steadystate conductor temperature to be equal to 60°C. The object of the above action statement is to satisfy a SteadyState Goal by verifying all of the steadystate rules stored in the Object:Slot pair Global:SteadyStateRules. Following is a simple example of programming by rules and a goal in the line ampacity system. Number 3.Computer Modeling 155 In the following section. If ( Transmission:Problem # = N ) Then CalcSteadyTemperature( ). The expert system correctly detected the problem and * Deb. Volume 8. Anjan. of a mathematical equation for example. the transmission line expert system is described by presenting an example of goaloriented programming. ForwardChain( [ ASSERT ].. SteadyStateGoal. When new facts are generated by the ﬁring of rules. The program proceeds with the calculation of transmission line conductor temperature only if there are no problems detected in the data entered by the user. the user entered a value of conductor temperature less than ambient temperature. [ASSERT] ensures that the new facts are automatically inserted into a fact database.* Learning by artiﬁcial neural network is described later in this chapter. Global:SteadyStateRules ). the user input data were checked by the expert system rules. . to solve a problem. Goal: SteadyStateGoal Action: { SetExplainMode( ON ). and explanation facility. In the following example. In traditional programming by procedures. Object oriented expert system estimates transmission line ampacity. 8. but also provide a practical method of incorporating practical and imprecise knowledge such as “rules of thumb” that are not easily amenable to formal mathematical treatment. Setting the explain mode to ON enables the user to receive explanations of expert system generated error messages. Rules not only facilitate a declarative style of programming. slots are provided to store data of an object.
10 Example of error message given by program when user entered incorrect value of conductor temperature.* FIGURE 8. the expert system generated the required explanations. Modeling. and Applications generated an error message.11 Explanation of error message given by program when user clicks on the explain button.11). .156 Powerline Ampacity System: Theory. as shown in Figure 8. Richard Forsyth. By clicking on the Explain button.10. and the story of the transmission line ampacity problem started to unfold (Figure 8. FIGURE 8. * Towards the Learning Machine.
if post overload current results in a higher than maximum allowable conductor temperature. and the coefﬁcient of solar absorbtivity is low and/or the emissivity of the conductor is high. If user input. Rule 6. Rule 10. Rule 4. If ambient temperature is greater than conductor temperature.3. then advise user. Rule 2.2 EXPERT SYSTEM RULES The expert system knowledge base is comprised of the abovementioned system of objects and rules. or to caution the user during speciﬁc operating conditions.Computer Modeling 157 8. If the line passes through urban areas with highrise buildings or where wind is restricted by tall structures or trees and line ampacity is high. In LINEAMPS. then advise user. is low 2 or 4 ft/s wind speed. rules are used to offer expert advice to users in the event of erroneous input or conﬂicting data. Some examples of rules used in the program are: Rule 1. Rule 8. In the transient state. In the dynamic state. Rule 9. then advise user. In the dynamic state. If the age of the conductor is old or the line passes through areas of industrial pollution. In the dynamic state. then advise user. If the age of the conductor is old and the conductor temperature is high. then advise user. In the transient state. then advise user. then advise user. then advise user. If the temperature of the selected conductor is greater than the allowable maximum for the conductor type. Rule 5. if the transient current is high and the line is old. then advise user. and the National Weather Service forecast is high wind speed. . Rule 3. if preload current results in a higher than maximum allowable conductor temperature. then advise user. then advise user. Rule 11. if the duration of transient current is greater than the speciﬁed maximum. Rule 7. if the user speciﬁed preload current results in a conductor temperature that is higher than the allowable maximum.
After receiving an explanation. the value of conductor temperature entered by the user is 35°C. In the above steadystate analysis window of the LINEAMPS program. } ). then reduce current ***********************************************/ MakeRule( ConductorRule. One of the rules in the expert system detects this problem and displays the error message shown in Figure 8. ***************************************************/ MakeRule( SteadyStateTemperature.158 Powerline Ampacity System: Theory. Steady:ConductorType ) ) Or Steady:Temperature > Steady:ConductorType:MaxTemperature. and Applications Rule 12. then reduce current” ). []. []. Steady:Calculate #= Current And Steady:ShowTemperature <= Steady:AmbientTemperature. “If conductor temperature is greater than the allowed maximum for the conductor type. If the value of conductor emissivity is less than or equal to 0 or greater than 1. . /*************************************************** **** RULE 1: Conductor Rule **** If conductor temperature is greater than the allowed **** maximum for the conductor type. Rule 13. Modeling. SetRuleComment( SteadyStateTemperature. /*************************************************** **** RULE 2: SteadyState Temperature **** Conductor temperature must be greater than **** ambient temperature.10. and solar radiation are given. SetRuleComment( ConductorRule. The following example shows a listing of two rules used by the program. { Transmission_Line_Ampacity:Problem = “High conductor temperature”. emissivity. then advise user. 40°C. Not( Member?( ConductorTypes:AllowableValues. which is inconsistent with the value of the ambient temperature. users also have the ability to request other relevant information regarding various input data required in this session window.11 explanations of temperature. “Conductor temperature must be greater than ambient temperature”). } ). PostMessage( “Check ambient temperature” ). PostMessage( "Please check valid conductor type and conductor temperature" ). { Line_Ampacity:Problem = “Conductor temperature”. In Figure 8. If the value of conductor absorbtivity is less than or equal to 0 or greater than 1 then advise user.
a brief description of each icon shown in the Control Panel window (Figure 8. While full details of the program is presented in the user manual. 1998. weather modeling. .1 LINEAMPS WINDOWS In this section.12) is presented here. These windows may be opened by clicking icons. conductor modeling. LINEAMPS is designed as a system of windows where users conduct various sessions with the program on different aspects of power line ampacity.2 MODELING TRANSMISSION LINE AND ENVIRONMENT The following additional session windows are available for the modeling of a transmission line and its environment. • • • • • • • • LINEAMPS Control Panel SteadyState Analysis Dynamic Analysis Transient Analysis Forecast Ampacity Power Lines Conductors Weather 8. Following are the main session windows that are presently available in the program for the analysis and planning of transmission line ampacity.4 PROGRAM DESCRIPTION 8.4. each representing a unique function. the main features and functions of the LINEAMPS program are brieﬂy described. • • • • • • • • • Power Lines Conductor Static Rating Cartograph Ambient Temperature Wind Speed Daily Weather Forecast Extended Weather Forecast Object Browser The functions and operations of the LINEAMPS program in each window are described in detail in the user manual.4.3 LINEAMPS CONTROL PANEL The program is operated by a system of icons.Computer Modeling 159 8. and cartograph. or by selection from the LINEAMPS Control Panel Window.* 8. The control panel window has the following icons: * LINEAMPS User Manual.4.
It is used for selecting weather patterns and for generating hourly values of ambient temperature data from forecast weather data. Forecast Ampacity Clicking this button opens the Forecast Ampacity window. and Applications LINEAMPS: Control Panel Database Weather Analysis Browser Help E E $ ? Exit FIGURE 8. It is used for the calculation and display of hourly values of transmission line ampacity up to seven days in advance. Transient Analysis Clicking this button opens the Transient Analysis window.160 Powerline Ampacity System: Theory. Modeling.12 LINEAMPS Control Panel window. It is used for the analysis and display of transmission line ampacity and conductor temperature in the steady state. It is used for the analysis and display of transmission line ampacity and conductor temperature in the dynamic state. It is used for the analysis and display of transmission line ampacity and conductor temperature in the transient state. Extended Weather Forecast The Extended Weather Forecast window is opened by clicking this button. . as well as for input of forecast weather data. It is used to view the forecast weather data of the region. Wind Speed Model E E Clicking this button opens the Wind Speed Model window. SteadyState Analysis Clicking this button opens the SteadyState Analysis window. Ambient Temperature Model Clicking this button opens the Ambient Temperature Model window. Dynamic Analysis Clicking this button opens the Dynamic Analysis window. It is used for selecting weather patterns and for generating hourly values of wind speed data from forecast weather data.
Conductor Clicking this button opens the Conductor window. and forecast hourly values of ampacity for seven days in advance. the location of weather stations. This window is used to generate hourly values of transmission costs for seven days in advance. and transient analysis. It is used to print LINEAMPSgenerated reports on steadystate analysis.Computer Modeling 161 Cartograph Clicking this button opens the Cartograph window. This window is used for generating help text on userrequested topics. It is used for conductor selection. Clicking this button opens the Print window. Exit Exit the program by clicking this button. and viewing and updating data in existing line objects. It is used to view a geographic map of the region. It is used for navigating the LINEAMPS system of windows. Help ? Print Clicking this button opens the Help window. and viewing data in conductor objects. This window is used for creating new lines. Exit . The Welcome window is opened by clicking this button. Powerlines Clicking this button opens the Powerlines window. Transmission Cost $ Welcome Clicking this button opens the hourly Transmission Cost window. dynamic analysis. creating new transmission line conductors. and the transmission line route.
and rules are used to ﬁnd a solution to the powerline ampacity problem. A combination of procedures. and methods and the sending of messages between objects are presented to show how an elegant system of objects having messaging capability is realized in the LINEAMPS program. Procedures or methods are used in objects when a mathematical model is available. Modeling. thus demonstrating the intelligent behavior like a true expert. The theory developed in the previous chapter and practical knowledge of the transmission line system are implemented in the LINEAMPS program by objectoriented modeling and expert rules. goals. checks user input. conductor object. The object model of the complete line ampacity system is presented. . and Applications 8. Rules are used to incorporate practical knowledge. followed by the component object models consisting of transmission line object. Decisions based on rules are generated automatically by an inference engine. and explains error messages to the user.162 Powerline Ampacity System: Theory. Examples are presented to show how the system calculates powerline ampacity by objects and rules. A brief description of the various functions and features of the program and the graphical user interface are presented to demonstrate the realization of a complete line ampacity system that is suitable for all geographic regions.5 CHAPTER SUMMARY This chapter describes computer modeling of the transmission line ampacity system by objects and rules. and a cartograph object. weather station object. The creation of objects.
This chapter presents an overview of FACTS and the various power semiconductor devices it uses. FACTS devices are also used in lowvoltage distribution networks. a power transmission and distribution line engineer will make better decisions regarding the selection of FACTS technology best suited for his or her requirements. Therefore. Installation of FACTS devices may be required for better control of the current and voltage in a transmission line.1 INTRODUCTION AC transmission circuits are mostly composed of passive elements having very little controllability. and voltage regulation. FACTS devices are installed at all voltage levels up to 800 kV. used in power distribution networks to maintain power quality. and to lower harmonics and minimize voltage ﬂicker. therefore. new types of power electronic devices called FACTS (Flexible AC Transmission System) are used to control existing T&D networks. FACTS devices are. certain lines are more heavily loaded and stability margins are reduced. and other similar devices in homes and businesses. 9. consumers are paying greater attention to power quality in distribution circuits. When transmission capacity increases are planned by dynamic thermal ratings and reconductoring of existing circuits with special or highersize conductors. it will be necessary to properly evaluate their impact on power system performance and load ﬂows. and increasing use of high technology electronic equipment like computers. An important application of power electronic semiconductor devices in electric networks 163 . As a result. Due to the difﬁculty of controlling power ﬂows by existing methods. existing network control methods are not sufﬁcient to properly accommodate increased power ﬂows. Due to greater utilization of electrotechnologies in the industry. TV.9 New Methods of Increasing Transmission Capacity 9. It is hoped that by gaining knowledge of the various FACTS technologies presented in this chapter. when the load increases.2 ADVANCEMENT IN POWER SEMICONDUCTOR DEVICES Recent advances in power electronics semiconductor devices have made it possible to develop equipment with sufﬁcient current and voltage ratings to enable their utilization in electric power circuits for better control of voltage and current.
Modeling. but unlike a diode it also requires a pulse having a certain voltage and current to be applied to the gate to turn on the thyristor for conduction to begin from anode to cathode. Cathode Gate Cathode N1 P1 N2 P2 Anode Silicon Base Plate FIGURE 9.164 Powerline Ampacity System: Theory. R. Like a diode. Similar to a diode.” shown in Figure 9.1 Thyristor is to control the amount of current that can pass through the device in response to a control action.6. It is blocking during the negative half of the AC cycle when the anodetocathode voltage becomes negative.5. This is accomplished by supplying a current pulse to the gate at a desired triggering time. of a thyristor is illustrated by a simple circuit in Figure 9. when the anodetocathode voltage is positive. THYRISTOR A thyristor is a threeterminal device with an anode. the instantaneous load current. θ. The most common semiconductor device used for the control of current ﬂowing through a circuit is a siliconcontrolled rectiﬁer device called a “thyristor. and gate. Considering a resistive load. It is a special type of diode.29. conduction takes place during the positive half of the AC cycle.2 Thyristor Construction The current through a thyristor is regulated by controlling the gatetriggering time. is . a thyristor requires a certain positive anodetocathode voltage. cathode.1. and Applications FIGURE 9. The ability to control current in a circuit by controlling the ﬁring angle. ir. as shown in Figures 9.
ir = V sin(ωt ) R 0 ≤ ωt ≤ π The average halfwave rectiﬁed DC load current.4 Two transistor model of Thyristor Anode Gate Cathode FIGURE 9. is .3 Schematic Anode P2 T2 Gate T1 P1 N1 N2 Cathode FIGURE 9.5 Thyristor Symbol Ir V 5kV/60Hz/0Deg R 40 FIGURE 9.New Methods of Increasing Transmission Capacity 165 Anode P2 N2 P1 Gate N1 Cathode FIGURE 9. Ir.6 Current control by Thristor in a 5 kV line to ground distribution circuit.
and is turned off by applying a reverse gate signal. This limitation is overcome by the newer type of semiconductor switching devices like the IGBT discussed later. and is therefore used in lowvoltage FACTS control circuits. Being a low voltage device.A 30 20 10 0 0 20 40 60 80 0 0 0 14 16 0 18 0 12 10 Firing angle FIGURE 9. The main disadvantage of a thyristor is that it cannot be turned off easily by applying a control signal at the gate. it is not suitable for direct connection to a highvoltage network. for example. The main disadvantage of a GTO is the high current requirement to turn off current. there are several types of semiconductor devices that are suitable for various power system applications. Modeling. It has a short turnoff time in the order of tens of nanoseconds.7. Other than the thyristor. which is much faster than a thyristor.166 Powerline Ampacity System: Theory. The variation of load current as a function of the ﬁring angle. to control a GTO device. θ.1 and are brieﬂy discussed here. GTO The Gate Turn Off Thyristor (GTO) is turned on by a short pulse of gate current. These are primarily electronic switching devices derived from transistor technology and adapted for highcurrent and highvoltage applications. It has a .6 is shown in Figure 9. MOSFET The MOSFET is a voltagedriven source used mainly in lowvoltage applications. It can be turned on or off rapidly and is suitable for high switchingfrequency operations. and Applications π 1 Ir = 2π Ir = ∫ θ V ⋅ Sin(ωt ) dωt R V ⋅ (1 + Cos(θ)) 2 πR where θ is the angle at which a pulse is applied to the thyristor gate. Their important characteristics are listed in the Table 9. Thyristor Current Control 50 40 Current.7 Control of circuit current as a function of thyristor ﬁring angle. in the circuit of Figure 9.
it is not widely used due to the complex electronics required to control it. It is therefore suitable for highvoltage applications. This feature is particularly useful in highvoltage FACTS applications because the control signal is transmitted from ground potential to the gate of the IGBT device at very high voltage. several of these devices are required to be connected in series to withstand high commutation voltage for highvoltage applications. generally in the range of 3 to 5. IGBT The IGBT takes advantage of the high commutation speed of a power MOSFET and the low resistance offered by a bipolar transistor. By connecting these active devices at . Highvoltage DC convertors and SVC stations are generally located indoors. The MCT device under development is expected to offer high commutation power similar to a GTO. but is available with sufﬁcient power capability that is comparable to a thyristor. and an electrical ﬁlter for the elimination of noise generated by switching action. A schematic of a thyristor valve assembly is shown in Figure 9. The term “insulated gate” refers to the metal oxide insulated gate which requires very little control power. Therefore a thyristor valve assembly is made modular in structure for ease of installation and maintenance. Several of these modules are then connected in series to develop the full transmissionlinetoground voltage. MCT The MOSControlled Thyristor is considered the power semiconductor switching device of the future. a unit for cooling the thyristor. The switch can be turned on or off quite rapidly by the application of a control signal at the gate. A number of such devices have been developed which are widely known as FACTS (Flexible AC Transmission System) devices. where several of these devices can be connected in series. At the individual thyristor level it is comprised of a control unit for controlling the ﬁring angle of the thyristor.8. FACTS Semiconductor Valve Assembly Since the present state of the art in semiconductor assembly allows a thyristor or an IGBT device to be built with voltage rating up to 10 kV. A module may consist of four to six thyristors connected in series with a reactor.New Methods of Increasing Transmission Capacity 167 low ratio of commutation current to turn off current. where a stack of thyristor modules forming a highvoltage valve is either ﬂoor standing or suspended from a ceiling. and a system for the distribution of cooling ﬂuid to the various units and support insulators. GBTR The Giant Bipolar Transistor (GBTR) is older than the IGBT. with control capability similar to an IGBT. This reduces the power that can be commuted by a GTO in a FACTS application. Even though the cost of the GBTR is somewhat less than IGBT and power MOSFET. A thyristor valve assembly is comprised of a system for the communication and control of the various thyristor units.
1 Semiconductor Device Properties Semiconductor Device MOSFET IGBT GBTR GTO Thyristor MCT Rated Voltage 200 V 1200 V 1200 V 4000 V 5000 V 3000 V Rated Current 100 A 300 A 300 A 3000 A 4000 A 30 A Speed 200 ns 1 µs 5 µs 40 µs – 40 µs Switching Frequency 100 kHz 10 kHz 3 kHz 1 kHz 300 Hz 1 kHz Adapted from “Technologie des FACTS. A converter station is generally composed of 12 or 24 pulse thyristorcontrolled electronic bridge circuits. 9. These devices will also enable greater utilization of economic energy sources (including distributed energy resources) and facilitate competition in the electric supply business. a wide variety of control is possible in an HVDC transmission system by controlling the ﬁring angle of thyristors. greater operational ﬂexibility. and Applications 3 2 2 2 2 4 4 4 4 4 1 1 1 1 FIGURE 9. and faster recovery from system disturbances. it is now possible to implement a wide range of control over an AC network in a manner that was not possible before.168 Powerline Ampacity System: Theory. By having better control over the ﬂow of electricity in a T&D network by FACTS devices. and an inverter station is used to convert DC back to AC. Unlike an AC transmission circuit. it is expected to lower electricity costs by efﬁcient utilization of existing assets. better outage management. Modeling. SEE Conference 1994. Lataire. TABLE 9.3 FLEXIBLE AC TRANSMISSION HVDC The development of FACTS technology has evolved from the early days of HighVoltage DC Transmission (HVDC) where converter stations are used to rectify AC to DC.8 Construction of a thyristor valve module. .” Ph. suitable locations in the T&D network.
consisting of anode. SVC systems are active devices whose output can be controlled continuously to match system requirements very precisely. enhance power ﬂow. The reactive power absorbed by the TCR device is controlled by regulating the current ﬂow through the reactor by directing the ﬁring angle of the thyristor between 90° and 180°. basically a fourlayer pnpn device. In an uncompensated transmission line. There is full conduction at 90° ﬁring . When an SVC device is connected. cathode. greater utilization of FACTS devices are also expected in the distribution systems for better voltage regulation.9. and voltage support. These devices are used to maintain constant voltage levels. HVDC offers several advantages over AC transmission. STATIC VAR COMPENSATOR (SVC) Static VAR compensators are reactive power devices that generate or absorb reactive power as required by the electric power system.New Methods of Increasing Transmission Capacity 169 There are several HVDC transmission systems operational in the USA. the power transmission capacity of the line may be increased up to the full thermal rating of the transmission line conductor as shown in Figure 9. improve stability. Passive shunt capacitors and reactors have been used for a long time for reactive power supply. The thyristor is turned on by applying a short pulse across the gate and cathode. minimum voltage ﬂicker. asynchronous operation of two electrical systems. If an SVC is connected to the transmission line. The most common types of SVC systems used at present consist of a thyristorcontrolled reactor (TCR) and a thyristorswitched capacitor (TSC). and gate. Without an SVC device connected on this line. or as a backup resource for local generation. The performance of the SVC device to augment power ﬂow in a transmission line is seen in Figure 9.11. and greater reliability of service. including lower transmission cost. A TCR consists of a reactor in series with a bidirectional thyristor pair. and a TSC is shown in Figure 9. the transmission capacity is limited to 750 MW.9. such that any further increase in power transfer results in voltage collapse. up to 1000 MW or more. The use of FACTS devices in distribution systems will also lead to greater integration of smaller generation systems with lower environmental impact. lower harmonics. SVC is widely used in the electric power system to increase the transmission capacity of existing lines. The Paciﬁc Intertie is one example of an HVDC transmission system that is used to bring lowcost hydroelectricity from the Paciﬁc Northwest to California by means of a bipolar ± 500kV HVDC transmission line. power ﬂow is limited by voltage drop on the line.10. As consumers require better power quality. This example is given for a 132 kV transmission line having an ACSR Cardinal conductor. it is possible to utilize the full thermal capacity of the line. The basic elements of a TCR are shown in Figure 9. and turned off by applying a reverse voltage across the anode and cathode. power factor correction. and provide various other improvements listed at the end of this section. A thyristor is a fastacting electronic switch. HVDC LIGHT is another new development for DC power distribution by underground cables to remote locations where there is no local generation available. but were limited by their ability to provide continuously variable output to match system requirements. and support to AC systems when required.
A continuously variable lagging reactor power. and Applications Power Transmission Capacity 120. Q = I (α)2X where.kV 74.10 Thyristor Controlled Reactor (TCR) I Vc c V V L L FIGURE 9. QL.MW 1000 WithoutSVC With SVCand LINEAMPS FIGURE 9. Increasing the ﬁring angle decreases current.170 Powerline Ampacity System: Theory. is made possible by controlling the ﬁring angle of the thyristor given by.9 4. IL (a) V L FIGURE 9.9 Increasing power transmission line capacity by FACTS using a Static Var Compensator (SVC).0 27.3 Voltage.11 Thyristor Switched Capacitors (TSC) angle. Q = reactive power (lagging) I = reactor current α = ﬁring angle of thyristor X = inductive reactance . Modeling.7 0 500 Power. and conduction is blocked when the angle is 180°.4 97.1 51.
Qc = leading reactive power delivered to the system Vc = capacitor voltage Xc = capacitive reactance A combination of TCR and TCS is often connected in parallel to offer a continuously variable SVC system providing leading or lagging reactive power.New Methods of Increasing Transmission Capacity 171 Thyristors having voltage and current ratings up to 5 kV and 4000 A are presently available for switching frequency up to 300 Hz. STATCOM The development of Gate TurnOff (GTO) Thyristor technology has recently led to the making of an all solidstate VAR generator called STATCOM (Static Synchronous Compensator). Unlike thyristors. GTOs are capable of turning off by changing their state from conduction to nonconduction by pulsing the gate. obtained as follows: Qr = Ql – Qc. Qr.12.13. Hydro Quebec Canada. A TSC is used to deliver leading reactive power.12 A 735 kV SVC substation. Qc. FIGURE 9. An example of an SVC system having a combination of TCR and TCS in an actual 735 kV substation is shown in Figure 9. by switching on thyristors and is given by. GTO . A basic model of a thyristorswitched capacitor (TSC) system is shown in the Figure 9. At present.11. Q c = v2 X c c Where. There are more than 100 SVC installations worldwide operating at voltages up to 500kV with capacity in the range of ± 50 MVA to ± 400 MVA. and the singleline diagram is shown in Figure 9.
14.13 Singleline diagram of the 735 kV SVC system with TCR and TSC voltage and current rating is 4 kV and 3000 A. 6 pulse GTO converter and equivalent circuit . The basic operating feature of a STATCOM is similar to the rotating synchronous condenser.172 Powerline Ampacity System: Theory. The STATCOM consists of a DC to AC converter circuit. VL Transformer VO GTO Diode Vdc Ia Va Ia + Ib Ib Vb FIGURE 9. having no moving parts and with a high degree of reliability. respectively. but has the advantage of solidstate technology. Modeling. and 40 µs speed. 1 kHz commutation frequency. and Applications 735 kV line to line voltage 330 MVA 3 phase Transformer 330 MVA 3 phase Transformer 110 MVAR TCR 110 MVAR TSC 110 MVAR TSC 110 MVAR TSC 110 MVAR TSC 110 MVAR TSC 110 MVAR TSC 110 MVAR TCR FIGURE 9. As a result. Unlike the TCS and TCR devices described previously.14 STATCOM. the size and cost of static VAR generating equipment is reduced considerably. This makes the GTO more suitable for faster turning off. requiring less power than a conventional thyristor. STATCOM offers continuously variable reactive power without the use of capacitors or reactors. which provides a threephase output voltage in phase with the AC system voltage as shown in Figure 9.
delivers leading VARS to the AC system.15 Series compensated line of Example 1. SERIES COMPENSATION Series compensation has been used for a long time to increase the power transmission capacity of long transmission lines by connecting capacitors in series at suitable locations along the line. using GTO. Early series compensation devices used mechanical switches to connect capacitor banks in series with a transmission line to reduce line impedance and increase transmission capacity. Thus.New Methods of Increasing Transmission Capacity 173 By controlling the magnitude of the converter output voltages. VL. the reactive power delivered by STATCOM can be controlled from full leading to full lagging. Vs j250 Vr 1000 km FIGURE 9. Increasing the magnitude of converter output voltages above AC system voltage. Some examples of possible series capacitor locations in long transmission lines are shown in Figures 9.17.17 Controllable series capacitor . as well to maximize the ampacity of the line.15 and 9. controllable series capacitors with GTO thyristors are used to control the degree of reactive power compensation more precisely as a function of network operating conditions.16 Location of series capacitors on long transmission lines. A reduction in converter output voltage delivers lagging VARS. (A) (B) (C) FIGURE 9. (TCSC) it is now possible to control the current ﬂowing through a speciﬁc highvoltage line.16. Today. 6 1 2 3 5 1 = Capacitor 2 = Zno varistor 3 = GTO thyristors 4 = Bypass breaker 5 = Damping reactor 6 = Transmission line 4 FIGURE 9. with a thyristorcontrolled series capacitors. Vo. and a thyristorcontrolled series capacitor with a protection device is shown in Figure 9.
Transmission line data Receiving end voltage = 765 kV Rdc = 0. A series capacitor having a reactance of j250 ohm is installed at the middle of the line. A GTO circuit is used for SSSC.174 Powerline Ampacity System: Theory. A complex voltage Vseries is generated by connecting the GTO converter to a transformer connected in series with the transmission line. which is similar to a STATCOM.041 ohm/km L = 8. This complex voltage V series is then added to the AC voltage of the transmission line to control transmission line voltage.78 x 10–9 F/km G=0 . Example 1 A 1000 km 765 kV transmission line delivers natural load. Modeling. as shown in Figure 9.18 Static Synchronous Series Compensator (SSSC) and equivalent circuit Static Synchronous Series Compensators (SSSC) are also developed that offer the faster response time required for the damping of power system oscillations caused by faults and other system disturbances. It is proposed to increase the transmission capacity by two times the natural load.35 x 10–4 H/km C = 12. as shown in Figure 9.18.18. Determine the voltage regulation of the line. and Applications I Vseries VL Transformer GTO Diode Vds Vseries Va Va + Vseries FIGURE 9.
. Ir = SIL 3 ⋅ 765 ⋅ 10 3 Ir = 1728 A The propagation constant and the characteristic impedance are obtained from (Example 4. Ir. the load current. will be used. B. km Vx = Voltage at a point. is. Section 10.2) x = distance from receiving end. derived in Appendix 10. x.5).New Methods of Increasing Transmission Capacity 175 Solution For a 1000 km line. The values can be represented by the following matrix equation utilizing the wellknown A.1 at the end of Chapter 10. in the line γ = propagation constant Zc = characteristic impedance of the line The above equation is derived in the appendix at the end of this chapter. C. x. SIL = 2289 MVA Therefore. D constants of the line: Vs A = Is C B Vr D Ir The natural load of the line is given by the Surge Impedance Loading (SIL). Section 10. in the line Vr = Voltage at the receiving end of the line I r = Current at the receiving end of the line I x = Current at a point.5). (See Example 4. Vx = cosh( γx)Vr + Zc ⋅ sinh( γx)I r Ix = sinh( γx)Vr + cosh( γx)I r Zc (9.1) (9. the following longline equations.
Modeling.79 ⋅ 10 –3 {( } ( ) 100 Vr ( ) 100 441. A series capacitor bank having a total reactance of j250 ohm is now added at the middle of the line. The ABCD constants of a 500 km section of line are: A = cosh( γ ⋅ l) A = D = cosh 2 ⋅ 10 –5 + j1. Vs = cosh 2 ⋅ 10 –5 + j1.16) ⋅ sinh 2 ⋅ 10 –5 + j1.8 ⋅ 10 3 Regulation = Vs – Vr ⋅ = 2. Vs = (166.4) ⋅ 10 3 Vs = 857.5.1% The sending ending voltage is too high and the regulation is unacceptable.7 ⋅ 10 3 – 441.6 – j4.7 ⋅ 10 3 ⋅ = 94.23 ⋅ 10 –5 ⋅ 10 3 ⋅ 441 ⋅ 10 3 {( ) } ) + (255.9) ⋅ 10 3 Vs = 452.23 ⋅ 10 –3 ⋅ 500 = 0.6 − j4.15 The sending end voltage Vs is obtained from Section 10.2 ⋅ 10 –3 Zo = 255.7 ⋅ 10 3 ( ) .23 ⋅ 10 –3 ⋅ 10 3 ⋅ 1728 Vs = (156.7 ⋅ 10 3 Regulation = 857.176 Powerline Ampacity System: Theory.7 + j424.4 + j841.816 + j5. and Applications γ = 2 ⋅ 10 –5 + j1.52% Sending end voltage when load is 2 times SIL.
023) = 5 –3 – 3.49 + j147.79) (0.935 ⋅ 10 – j4.816 + j5.23 ⋅ 10 –3 ⋅ 500 {( )} C− {( (255.023) .6 – j4.15) )} ⋅ C = −4. 816 + j5.77 ⋅ 10 – j2.79) 0 (4.New Methods of Increasing Transmission Capacity 177 B = Zo ⋅ sinh( γ ⋅ l) B = (255.6) (0.794 + j0.993 + j74.816 + j5.15) ⋅ sinh 2 ⋅ 10 –5 + j1.26 ⋅ 10 –3 The ABCD constants of the series capacitor are: A′ = 1 B ′ = j250 C′ = 0 C′ = 1 The modiﬁed ABCD constants of the series compensated transmission line are: A ′′ C ′′ B ′′ A = D ′′ C B A ′ DC ′ B ′ A D ′ C B D – j250 1 (0.77 ⋅ 10 –6 + j2.6 C= sinh( γ ⋅ l) Zo sinh 2 ⋅ 10 –5 + j1.26 ⋅ 10 ( ) ) (4.515) (0.794 + j0.816 + j5.6 – j4.6) 1 (0.49 + j147.23 ⋅ 10 –3 ⋅ 500 B = 4.49 + j147.966 ⋅ 10 ( ) (7.77 ⋅ 10 – j2.79) –6 –3 – 4.26 ⋅ 10 ( (0.79) = –6 –3 – 4.
( ) 100 441. the reactance.A.19.4 ⋅ 10 3 Regulation = 463. .B.515) ⋅ 3455 Vs = 3. and Applications Vs = A ′′ ⋅ Vr + B ′′ ⋅ I r Vs = (. Having a GTO thyristorcontrolled series capacitor as shown in Figure 9.178 Powerline Ampacity System: Theory.794 + j0. Modeling. et al. which is twice the natural load of the line.D constants of the line according to network operating conditions.023) ⋅ 441. Rena. control domain. The above example shows that by adding a series capacitor at the middle of a line.993 + j74.91% The above sending end voltage and regulation is acceptable. CUSTOM POWER Custom Power concept covers a number of power electronics devices suitable for connection at the distribution system level. A block diagram of the UPFC converter.7 ⋅ 10 3 + (7.7 ⋅ 10 3 UNIFIED POWER FLOW CONTROLLER (UPFC) The UPFC is a new development that offers multiple compensation functions by providing independent control of the following transmission line parameters: • Bus voltage • Active power • Reactive power The UPFC is made of two GTObased converters connected by a common DC link.C. Cigré 1998. and equivalent circuit is shown in Figure 9. Recent UPFC installations worldwide are in the range ± 100 to ± 200 MVA at transmission voltage levels. hence.7 ⋅ 10 3 ⋅ = 4. and prevent plant shutdown during adverse voltage con* World’s ﬁrst Uniﬁed Power Flow Controller on the AEP system.677 ⋅ 10 5 Vs = 463. These devices typically have a range from 1 to 10 MVA and may be connected at customer point of connection to provide better voltage regulation. the maximum power delivered by the line is increased from 2289 MVA natural load to 4578 MVA. They can also operate independently with Converter 1 acting as a STATCOM and Converter 2 like an SSSC. B.782 ⋅ 10 5 + j2.17 allows control of the current passing through the capacitor and. The AEP* installation (1997–1998) is rated at ± 160 MVA and is located on a 138 kV transmission line. by modifying the A.4 ⋅ 10 3 – 441.
A distribution system STATCOM is used at the distribution level for power factor correction as more current is passed through the distribution circuit. Power electronics devices also enable greater integration of smaller generating systems located closer to the loads.19c Uniﬁed Power Flow Controller (UPFC).19a Uniﬁed Power Flow Controller (UPFC). damping of oscillations. 1999). L. the installation of power electronic devices offers faster connection to the main supply in the event of a failure. fuel cells.. and photovoltaic and other renewable energy sources.19b Uniﬁed Power Flow Controller (UPFC). Small generation technologies include microturbines. (Buckles et al. (Borgard. Converter schematic Re Vseries Va Vb Im FIGURE 9. damping of power system oscillations. 1997). and for protecting the distribution system from the effects of nonlinear loads and ﬂicker reduction applications. and smoothing out the voltage delivered by the small generation systems. SMES The Superconducting Magnetic Energy Storage (SMES) device is another FACTS device (Feak. wind turbines.New Methods of Increasing Transmission Capacity 179 Qin Pin Va Converter 1 Converter 2 Vdc Vseries Qout Pout Vb Control Qin Qout Pin Pout Va Vb FIGURE 9. 2000) that is used for voltage support. . In this scenario. Equivalent circuit1 ditions such as voltage dips and power surges. and improvement of power system stability. Control domain Ia Vseries + Ib Va + Ic Vb FIGURE 9.
and Applications During normal operating conditions.20. Modeling. 4 and 5. and distributed energy sources Reduce temporary overvoltages Damp subsynchronous resonance Provide reactive power to ACDC converters FACTS devices for distribution system and custom power MANUFACTURERS There are several manufacturers of FACTS devices in the US and worldwide.20 SMES device connected to a high voltage transmission line LIST OF FACTS APPLICATIONS A number of FACTS device systems are in use worldwide to solve various power system problems: • • • • • • • • • • Voltage control Load balancing Increase active power transmission capacity of existing and future lines Increase transient stability margin Increase damping of power oscillations caused by a disturbance in the power system Facilitate greater use of dispersed. 3. small. are required for conversion of DC power supplied by the superconducting coil to 60 Hz ac. energy is stored in a superconducting coil. The stored energy released to the network when there is a disturbance is used to provide voltage support. The AC output voltage of the converters is stepped up to the transmission line voltage by a threewinding transformer. ABB GE Power Electronics Division GEC Alsthom T&D Power Electronics System Hitachi SIEMENS Westinghouse Electric Company . 6. A simpliﬁed SMES device is shown in Figure 9. Following is a list of important manufacturers. The two GTO thyristor converters. comprise the SMES device.180 Powerline Ampacity System: Theory. as well as for damping power system oscillations. 3 1 2 5 4 6 1 = Transmission line 2 = Harmonic filters 3 = Three winding transformer 4 = GTO thyristor converter 1 5 = GTO thyristor converter 2 6 = SMES FIGURE 9.
Further development of semiconductor power electronic devices is required to obtain greater current and voltage ratings with faster response time and higher switching frequency.4 CHAPTER SUMMARY Power electronics semiconductor device technology for FACTS has evolved considerably during the last decade. the main requirements are fast turnon and turnoff times with high current and voltage ratings. Superconducting Magnetic Energy Storage (SMES) systems are developed for application in all voltage levels for faster response time. SMES provides continuous VAR support and maintains the stability of the interconnected transmission grid. For power semiconductor devices. For HV and EHV network applications. Lowvoltage DC converters are used to tap off connections from HV and EHV AC lines and convert AC to lowvoltage DC. The lowvoltage DC power supply is then distributed by DC cables to remote areas. and HVDC LIGHT for lowvoltage .New Methods of Increasing Transmission Capacity 181 FUTURE RESEARCH AND DEVELOPMENT The FACTS devices used at present are mostly independent units and are not yet widely integrated with other FACTS devices in the network. Another interesting application of FACTS for distribution systems is HVDC LIGHT. This chapter mainly focused on FACTS technology for transmission systems. Further R&D is required to develop integrated control of several FACTS devices by communication between devices in order to exercise much wider and distributed control of the power system. Underground DC is also more acceptable for environmental reasons. there is a need to increase the current and voltage rating of power semiconductor devices as well as to increase the switching frequency for higher commutation speed and fewer losses. The development and application of FACTS devices are largely dependent upon semiconductor switching devices. lower harmonics. which includes both static series and shuntcompensated devices. As consumers require better power quality. greater utilization of FACTS devices is expected for better voltage regulation. Since DC voltage is used for distribution it does not have the problem of the high capacitive reactance of AC underground cables. This technology is providing competition for distributed generation systems. and greater reliability of service. Therefore. minimum voltage ﬂicker. and several applications of the different technologies are presented in this chapter. The IGBT device appears to be promising for highvoltage applications requiring fast turnon and turnoff times and having high switching frequency. and maximum current ratings in the range of 1000 A to 3000 A. At the present time these devices are developed with voltage ratings up to 10 kV. 9. GTO thyristors are also available with sufﬁcient voltage and current rating and fast switching capability. The use of FACTS devices in distribution systems will also lead to greater integration of smaller generation systems with lower environmental impact. it appears that lowvoltage underground DC transmission can supply power for longer distances to remote areas. HVDC LIGHT applies FACTS technology for low voltage DC transmission by underground cables. with a brief discussion on distribution system applications. A wide variety of FACTS devices are presented in this chapter.
Thyristor controlled series capacitors are particularly useful for increasing the power transmission capacity of long lines without causing synchronous resonance problems associated with generators. . This will lead to increased reliability of service. and Applications distribution to remote locations. and will maintain generator stability. and better power quality. these new FACTS devices will respond quickly in the event of a system disturbance by providing voltage support. When lines are operated close to thermal ratings. the lowering of electricity cost. It is expected that these developments will enable maximum utilization of existing transmission and distribution system resources by making use of the line ampacity system.182 Powerline Ampacity System: Theory. The development and application of FACTS technology is essential for the control of current and voltage in transmission and distribution circuits. Modeling.
The steady. For this reason. power system stability. the present work on generation and transmission system cost optimization by dynamic line rating is an extension of previous work on this subject. is compared to the increased cost of losses due to higher transmission line current. followed by a study of economic conductor temperatures. dynamic. actual PG&E transmission and generation system costs were presented to show cost savings by dynamic line ratings. and transient stabilities of generators are then related to the steady. Therefore. are discussed. the effect of increasing line ampacity on generator stability is presented.1 INTRODUCTION The application of the powerline ampacity system to the economic operation of transmission network. mainly from the point of view of transmission line ampacity. 183 . due consideration must be given to utilizing renewable energy sources that are generally located far away from major metropolitan areas and industrial load centers.2 ECONOMIC OPERATION In an earlier study. and transient line ratings. In this study. For transmission planning purposes. The maximum power transmission capacity of certain overhead powerlines is sometimes limited by the stability of generators connected to transmission lines. it is necessary to clearly understand the factors affecting the choice of conductors for overhead transmission line applications. For future transmission system planning. For this purpose. Extra High Voltage (EHV) transmission lines are required to transport electric energy economically over long distances. 10. which clearly demonstrates the economic beneﬁt of operating a power system by utilizing a dynamic powerline ampacity system. It is shown that the construction of a new line can be deferred by at least ten years by adopting a dynamic line rating system in this region. Results are presented by example of a hypothetical utility generation and transmission network. 1988a). dynamic. and transmission planning is presented in this chapter. for example. Based on the above considerations. Therefore.10 Applications 10. the cost of adding a new line in the region of the San Francisco Bay area. some important factors affecting longdistance transmission. the author performed economic evaluation of an interconnected electricity generation and transmission system by a dynamic line rating system (Hall. a formulation of the economic sizing of conductors is presented. Deb. We presented a case study without performing a formal network optimization analysis. The economic load dispatch problem and the optimal power ﬂow problem are presented by consideration of variable transmission line ratings.
The objective function is given as. Nodal Balance of Active and Reactive Powers (LoadFlow Equations) Active Power Pi = ∑ V V [g i k k =1 n i .k Cos(θ i – θ k ) + b i.4) . that represents the total cost of producing electricity.184 Powerline Ampacity System: Theory. Modeling. F= where.1) is subject to the following constraints: Power Balance This requires that the sum of all power generated be equal to the sum of demand and transmission loss. PL = ∑B P i =1 n 2 i i (10.k Sin(θ i – θ k ) ] i = 1.3) Bi is the transmission loss coefﬁcient. ci. including generation and transmission cost for a given load demand and operating system constraints. the problem is how to allocate the required load demand among the available generation units on an hourbyhour basis. This is carried out by minimizing an objective function.1) ∑P – P i i=l n D – PL = 0 (10. = cost coefﬁcients of the ith generator n = number of generator units Pi = power output of the ith generator The objective function (10. F.1 FORMULATION OPTIMIZATION PROBLEM It is required to minimize the total cost of electricity production. ai. and Applications OF THE 10.2. bi.2) PD = demand PL = transmission loss given by. K n nodes (10. ∑ (a + b ⋅ P + c ⋅ P ) i i i i 2 i i =1 n (10. In an interconnected electric utility system comprising dispersed generation sources.
Ng (10.i ≤ Pi < Pmax.k ≤ I i.i i = 1.i i = 1.i < Qi ≤ Qmax.10) The above equations describe the optimal powerﬂow problem.6) I i.k = z i .Ng Ng = Number of generator nodes Bus Voltage Limits Vmin.k (Max) I i .9) (10. K n nodes (10. artiﬁcial neural .…Ntap (10.…N nodes Transformer Tap Limits VTmin.k Cos(θ i – θ k ) + b i.k (Max) = Dynamic line rating Generator Capacity Limits The active (P) and reactive (Q) power output of the ith generator should be between minimum and maximum generation limits: Pmin. A.k = (10..5) θ ≅ ∠Vi V = node voltage Transmission Line Capacity Limits (Line Ratings) Line current between nodes i and k: Vi ∠θ i – Vk ∠θ k z i .….i ≤ VTi ≤ VTmax.i ≤ Vi ≤ Vmax.7) (10.i i= 1. More recently. A solution to the above nonlinear optimization problem is available in a modern power system analysis textbook (Bergen.k + jb i.k 1 g i.8) Qmin.k Sin(θ i – θ k ) ] i = 1.Applications 185 Reactive Power Qi = ∑ V V [g i k k =1 n i . Additional constraints may be added to consider emission levels in fossil fuelbased generation units.…. 1986).i i= 1.
4 for a static and dynamic line rating system.03 j0.2 ELECTRICITY GENERATION COST SAVING IN INTERCONNECTED TRANSMISSION NETWORK To illustrate the main concepts presented in the previous section.Y. A simpliﬁed solution of the above problem is presented in Table 2. 10.02 0. 1998) and genetic algorithm GA (Wong.32 j0.4.05 0.025 j0. 1998) solutions have also been proposed.10 Ys 2 = jωC 2 j0. 1998).186 Powerline Ampacity System: Theory. The generation sources G1 and G3 are hydroelectric. and all other generation sources are thermal. et al.016 + + + + + j0. The transmission network receives power from ﬁve generation sources (G1G5). Electricity Generation Cost Using Static Line Rating System Electric power companies generally follow a static line rating system by assuming constant line capacity based on conservative assumptions of the meteorological parameters that affect line capacity. TABLE 10.016 0.13 j0. Modeling. It is shown that substantial economy in electricity generation cost can be achieved by adopting a dynamic line rating system. and Applications network (Lee. transmission lines are operated more easily ..3 and 10.015 A comparison of electricity generation costs is made in the following sections using static and dynamic line rating systems. An ACSR Cardinal conductor is used in each phase of the line. K. (Yalcinov.1.02 j0. There are several reasons for operating lines based on static line ratings.3 and 10.02 j0. Short.1. For example. The object of this study is to show the signiﬁcance of dynamic transmission line ratings in the economic operation of an interconnected power system having diverse generation sources. The transmission network is comprised of ﬁve 230 kV double circuit lines and the network data is shown in the Table 10.1 230 kV Transmission network data with ACSR Cardinal Conductor HalfLine Charging Capacitance From Bus 1 1 2 3 4 To Bus 2 5 3 4 5 Line # 1 2 3 4 5 Impedance Z1 = R1 + jX1 0.13 j0. Many electric power companies have seasonal line ratings for summer and winter. The dimensions of the 230kV double circuit line are given in Figure 10.02 0.2.10 j0.3 and Figures 10. The transmission line electrical π equivalent model shown in Figure 10. a hypothetical electric utility generation and transmission system was created as shown in Figures 10. Yuryevich.2 was used to obtain network loadﬂow solutions.
5 m 5.6 m 30.5 m FIGURE 10. the static line rating of each transmission line phase conductor is assumed to be 849 A for normal operating conditions. An ACSR . In the example of Figure 10.5 m 7. R jXl Ys/2 Ys/2 FIGURE 10. conductor tension.9 m 4. as well as monitoring transmission line conductor temperatures. by following a static line rating system. close monitoring of weather conditions all along the transmission line route.3 Electricity generation and transmission line current using static line rating.2 Transmission Line Electrical PI model.6 m 7. 1022MW G1 400MW G2 600MW 500MW G3 500MW 1244A 1697A 1109A 829A 320A 1000MW G5 600MW 800MW G4 400MW FIGURE 10. The transmission line protection scheme does not have to be changed because line current limits are held constant. Also.Applications 187 10.3. and other parameters are not required.1 230 kV Double Circuit Transmission Line Tower.
5 = 2127 A) offered by LINEAMPS. and the total cost of electricity generation for meeting the demand for one hour is $ 48.2. ..4. As stated in the previous chapters.32.877. and Applications 1549 MW G1 400 MW G2 600 MW 400 MW G3 500 MW 2127 A 1731 A 378 A 1544 A 1136 A 1000 MW G5 200 MW 800 MW G4 400 MW FIGURE 10.188 Powerline Ampacity System: Theory. In this section the economic beneﬁts of a dynamic line rating system are demonstrated by an example of the transmission network considered in the previous section. Electricity Generation Cost Using LINEAMPS Rating Due to environmental and economic considerations. It is shown that by utilizing a dynamic line rating system the network is able to utilize a greater percentage of hydroelectric generation from the generation source (G1). the capacity of Line 1 reached its maximum static ampacity. (Steeley et al. depending upon actual transmission line operating conditions (Soto et al. A similar result was obtained in a previous study (Hall. many electric power companies have now started to use a system of rating transmission lines that is variable..* Cost Savings by LINEAMPS The power ﬂow solution presented in Table 10. As seen in this ﬁgure the network was able to receive 1549 MW of cheaper hydroelectric energy from generator source (G1) due to the higher transmission line ampacity (2 x 1063. the optimum powerﬂow solution of the network is shown in Figure 10. The optimum powerﬂow solution achieved by dynamic line rating is presented in Figure 10. the variable system of line ratings is commonly known as “dynamic line rating system.4 Electricity generation and line current using dynamic line rating. 1991). 1998). Cardinal conductor is used in all lines. Based on the above assumptions. In the optimum powerﬂow solution of Figure 10. (Wook et al. Modeling. which has resulted in an electricity cost saving of $ 4020.87 for one hour. 1988a).3 and Table 10. Such favorable conditions exist often during the life of a transmission line when it is required to * Assuming typical electricity generation cost at PG&E.3 utilizes dynamic line rating. which limited further addition of cheaper hydroelectricity available from the generator (G1)..” where transmission line ampacity is adapted to actual and forecast weather conditions. In Figure 10.3 it is seen that the ampacity of Line 1 has reached its maximum static rating (2 × 848 = 1996 A).3. 1997). Deb.
it is also possible to save the capital investment required for the construction of new lines or the reconductoring of existing lines.45 In addition to operational cost savings by facilitating economy energy transfer as mentioned above. $/hr Generation Cost $/hr 10527.00 48877.17 400 400 400 200 Load MW 0 600 500 800 1000 Generation Cost $/MWh 10. these cheaper sources of electricity can be utilized in the Bay Area with substantial cost savings as shown in Figure 10.30 20.45 9680.07 400 500 400 600 Load MW 0 600 500 800 1000 Generation Cost $/MWh 10.30 24. $/hr Generation Cost $/hr 15956.00 44856.5 that capital investment for new line construction can be deferred for at least ten years by increasing the capacity of existing lines for different overload conditions. the network of Figure 10.50 25. TABLE 10.20 12.4.00 10120.32 TABLE 10.3.50 25. the generation source (G3) could well represent hydroelectric energy supplied from the Sierras.2 Electricity Production Cost: Static Line Rating Bus Number 1 2 3 4 5 Generation MW 1022.20 12.00 4100.3 may be considered a simpliﬁed transmission network of the San Francisco region. By following the economic analysis presented in Section 10.00 6250. Using LINEAMPS ratings. .30 20.4 and Table 10.50 Total Generation Cost.Applications 189 transfer power greater than static line rating. Similarly. The generation source (G1) may represent the cheaper source of hydroelectric energy from the Paciﬁc Northwest supplying power to San Francisco Bay area.50 Total Generation Cost. it is shown in Figure 10.32 9680. For a more realistic example.30 24.00 5000.00 12300. Sometimes there is surplus electric energy available from these sources due to greater than normal rainfall or snow in these areas.00 10120.3 Electricity Production Cost: Dynamic Line Rating Bus Number 1 2 3 4 5 Generation MW 1549.
5 Cost of adding a new line is compared to the capitalized cost of increased transmission line losses for a range of overload conditions.3 STABILITY The maximum power transmission capacity of certain overhead powerlines are limited by voltage drop and the stability of generators connected to transmission lines. The following types of stability (Nasar. The above example is for a 230 kV line with ACSR Cardinal conductor having a static line rating of 849 A. The analysis is carried out for a period of ten years.190 Powerline Ampacity System: Theory.000 /km. For example. In Chapter 9 the various methods of increasing transmission capacity by the installation of modern power electronics devices were presented. 1998) problems are discussed: • Steadystate stability • Dynamic stability • Transient stability We are concerned with the steadystate stability of a generator when the power transferred over a line is increased slowly. $x1000 350 300 250 200 150 100 50 0 1000A 1200A 1400A 0 1 2 3 4 5 6 7 8 9 10 11 12 Overload hr/Day FIGURE 10. Trutt. The steadystate stability limit of a single generator supplying a load through a transmission line is given by. Modeling. if we consider 1000 A overload for 2 hr/Day we can see from this graph that it is more economical to allow the overload current than to build a new line having cost greater than $75. 10. In this section the method of calculation of generation stability when supplying power over a transmission line is presented brieﬂy. Ps = where.11) . and Applications Cost Saving by LINEAMPS Deferment of Investment for New Line Cost of New Line. Ps = steady state power transfer Vs = sending end voltage Vr = receiving end voltage Vs Vr Sin(δ ) X (10.
1998): .1 DYNAMIC STABILITY Dynamic stability is concerned with generator oscillations due to step changes in load or other small disturbances. Calculate the maximum power transfer capability of the line. Generally.9 ohm 1000 1 sin 90 0. is obtained when the rotor angle δ = 90°. ultimately. 3 phase Selecting transmission line base kV = 230 kV Transmission line base impedance = Line impedance = Pmax 50 ⋅ 0. This was discussed in Chapter 9 in Series Compensation. Increasing δ > 90° will result in lower power transfer and. t. of the line. The system is dynamically stable if the oscillations diminish with time. Small changes in generator output due to load variations result in generator rotor oscillations. with respect to time. X. loss of steadystate stability as Ps approaches zero. one possible means of increasing steadystate stability is to add capacitors in series with the line to lower the reactance.5 ohm/km Transmission line sending and receiving end voltage magnitude = 230kV Solution Selecting generator base MVA = 1000 MVA.47 pu 52. the dynamic condition oscillations remain for several seconds until steadystate conditions are reached. The transmission line data is given below. in time the system becomes unstable.9 230 2 = 52.47 Pmax = 2.12 pu Pmax = 2.Applications 191 X = line reactance δ = generator rotor angle Maximum steadystate power.5 = 0. If oscillations increase.11.3. Example 10. is obtained by (Saadat et al. Pmax. due to a small increase in power. From Equation 10. ∆P.12x1000 = 2120 MVA 10. The differential equation governing rotor motion.1 A generator is supplying a load through a 50 km 230 kV double circuit line. Conductor = ACSR Cardinal Reactance of line = 0.. and the generators return to a stable state. ∆δ.
2 The transmission line of Example 10.38° δ 0 = sin –1 2. ∆δ = small deviation in power angle from initial operating point δ0 D = damping constant H = Inertia constant f0 = frequency ωn = natural frequency of oscillation ζ = damping ratio given by. Assume generator frequency is 60 Hz during normal operating conditions.1 delivers a load of 1000 MVA under steadystate conditions. and the inertia constant is H = 6 pu. The solution of the above differential equation is ( Saadat et al 1998): π ⋅ f0 ⋅ ∆P 1 1 – ⋅ e – ςω n t sin(ω d t + θ) 2 H ⋅ ωn 1 – ς2 ∆δ = (10.192 Powerline Ampacity System: Theory. Solution Pm Initial operating angle δ 0 = sin –1 P max 0.138 pu.128 Ps = P max · cos(δ0) = 2.04 pu .15) Example 10.14) The above differential equation is obtained by linearization of the swing equation (Bergen 1986) and is applicable for small disturbances only. Show that the system will remain stable if the load is suddenly increased to 1200 MVA. Modeling. and Applications H d 2 ∆δ d∆δ +D + Ps ∆δ = ∆P 2 πfo dt dt πf ∆P d 2 ∆δ d∆δ + 2ςω n + ω 2 ∆δ = 0 n 2 dt dt H where.13) (10.6 = 16. The damping constant is D = 0.12) (10. ς= D 2 π ⋅ f0 H ⋅ Ps (10.
5 3 FIGURE 10. or generator or transmission line outage.2 1 – e ( –0.29) 0.2 TRANSIENT STABILITY Transient stability is concerned with generator oscillations due to sudden changes in power transfer levels caused by large disturbances which are due to shortcircuit.6 Generator rotor angle oscillations due to sudden increase in transmission line load current from 1000 MVA to 1200 MVA. .2712 The solution of the above equation for a period of 3 s is shown in Figure 10.Applications 193 ωn = π 60 H ⋅ Ps ζ= D⋅ π ⋅ 60 H ⋅ Ps = 0. Dynamic Stability High Ampacity Line 38 Generator Angle. s 2 2.709 θ = cos–1(ζ) δ(t ) = 0. Since we are dealing with large disturbances.271⋅8⋅t ) ⋅ 2 6⋅8 1 – 0. Deg 36 34 32 30 28 26 24 22 20 0 0.707 ⋅ t + 1.286 + π ⋅ 60 ⋅ sin(7. A simpliﬁed swing equation neglecting damping for transient stability studies is given by.271 2 ω d = ω n ⋅ 1 – ζ 2 = 7. large scale loadshedding. As seen in the ﬁgure. the system returns to a stable state after a sudden increase in transmission line load from 1000 MVA to 1200 MVA.5 1 1.3.6. Numerical solution of the nonlinear differential equation is obtained by Euler’s or the Runge Kutte method.5 Time. 10. linearization of the swing equation is not possible.
The power delivered by the line is 1200 MVA after the fault. The fault is cleared in 0. 1.96 Starting with t = 0 and δ = 36° and time interval ∆t = 0. during a small interval. Modeling. ∆t: ∆δ n = δ n – δ n –1 ∆δ n – ∆δ n −1 = ∆t ⋅ ω n −1 2 – ω n −3 2 = ∆t ⋅ ∆t π ⋅ f0 ⋅ ∆Pn −1 H 2 ( ) π ⋅ f0 ⋅ ∆Pn −1 ⋅ ( ∆t ) ∆δ n = ∆δ n −1 + H Example 10.194 Powerline Ampacity System: Theory.3 The transmission line of Example 10.2 P δ 0 = sin –1 o = sin –1 2.16) Following Euler’s method. ∆δn.4 sin δ0 = 0.04 P max = 36° The accelerating power is ∆P = 0 before the fault. ∆P. we obtain the change in angle.125s. Determine if the system will remain stable and attain a steadystate operating condition. after the fault is: ∆P = P0 – 0. Solution Initial operating angle. and Applications d 2 δ π ⋅ f0 ⋅ ∆P – =0 dt 2 H (10.05 we ﬁnd.2 is delivering 1200 MVA through both circuits when a shortcircuit occurs on one transmission line that lowers the power to 400 MVA. Acceleration power. ∆δ n = ∆δ n +1 + 180 ⋅ 60 ⋅ ( ∆t ) ⋅ ∆Pn −1 H ⋅ P0 2 .
s FIGURE 10.125 s. the application of the powerline ampacity system to power system operations was presented. This study includes transmission system cost analysis.4 sin(δ n −1 ) 2 ∆δ n = ∆δ n −1 + and after the fault.00 Time. 180 ⋅ 60 ⋅ ( ∆t ) ⋅ P0 – 0. The swing curve is stable as shown in Figure 10 7.00 0. H ⋅ P0 [ ] ∆δ n = ∆δ n −1 + 180 ⋅ 60 ⋅ ( ∆t ) ⋅ P0 – 1.4 TRANSMISSION PLANNING In the previous section.00 1. Transient Stability High Ampacity Line 100 90 Generator Angle.50 1. Deg 80 70 60 50 40 30 20 10 0 0. $/MWH Load factor Conductivity of conductor material .2 sin(δ n −1 ) 2 H ⋅ P0 [ ] The above equation is solved for a period of 2 s. optimum sizing of transmission line conductors.50 2. applications to transmission system planning and design are discussed. and the evaluation of optimum conductor temperatures. ∆t. 10. The following factors are considered: • • • • • Capital cost of line Cost of capital (interest rate) Cost of energy.Applications 195 During fault we obtain the change in rotor ∆δ angle during a small interval. In this section.7 Swing curve for a transmission line fault cleared in 0. by.
Cp = a1 + a2S + a3V (10.196 Powerline Ampacity System: Theory. Cp. Co = k(Clj + Cmj) j = 1. and Applications Speciﬁc Transmission Cost The economic evaluation of transmission lines is carried out by comparing the speciﬁc transmission cost (Vp) of different alternatives.18) Where.20) I = conductor current. The speciﬁc transmission cost is deﬁned as the cost per MVA/Km of power delivered by the line given as follows (Hall. $/MWH . a1. 1988a): Vp = Wp MVA (10. Deb. C1 = 3 ⋅ I 2 ⋅ R ac ⋅ L s ⋅ d ⋅ 8760 S (10.19) k = capitalization factor = n = life of the line. Capitalized Cost of Line Operation The capitalized cost of line operations (Co) includes the cost of losses (Cl) and the cost of line maintenance (Cm).17) Wp = Present worth of line = Cp + Co Cp = Capital cost of line Co = Capitalized cost of line operation Capital Cost of Line For estimation purposes the capital cost of line. years i = interest rate ∑ (1 + i) j=1 n –j The annual cost of line losses Cl is obtained by. ohm/km Ls = load loss factor d = cost of energy. A Rac = ac resistance of conductor.2…n years (10. a3 are the coefﬁcients of the line cost model obtained by statistical ﬁtting of historical data of line costs at different transmission voltage (V) and conductor size (S). may be obtained by. Modeling. a2. It is calculated as follows.
21) Optimum Size of Conductor When planning a new transmission line.4 adapted from (Anand et al.23 we obtain.22) (10.24) With the help of Equation 10. k2 are constants (Hall. it is required to select the optimum size of conductor for a given maximum power transfer. The optimum size of the conductor is obtained by minimizing the present worth (Wp) of the total transmission cost as follows: Min( Wp) = From equations 10. k1. mm2 Differentiating Wp with respect to S and setting size of conductor. 1988a).25) . 1985). we obtain the optimum dS (10.23) (10. Deb. Soptimum = I 3 ⋅ k ⋅ L s ⋅ d ⋅ r ⋅ 8760 a2 dWp = 0 . L s = k 1 ⋅ L f + k 2 ⋅ L2f Where. Lf = Energy supplied by the line in a year Maximum demand × 8760 (10.l S r = resistivity of conductor.24. 3 ⋅ I 2 ⋅ r ⋅ l ⋅ L s ⋅ d ⋅ 8760 Wp = k + Cm + (a 1 + a 2 S + a 3 V) S dWp =0 dS (10.Applications 197 The load loss factor Ls is related to the load factor Lf by. R ac = r. The load factor Lf is deﬁned as.17–10. we can perform transmission cost evaluation studies with alternative conductor designs for transmission planning purposes as shown in the Table 10. ohm⋅mm2 ⋅ m1 l = length of line = 1 km S = sectional area of conductor.
06 0. J optimum = a2 3 ⋅ k ⋅ L s ⋅ d ⋅ r ⋅ 8760 (10.198 Powerline Ampacity System: Theory.4 Line Data MVA capacity Economic span.97 Compact 54/7 800 425 43 0.5 LONGDISTANCE TRANSMISSION Extra High Voltage (EHV) transmission lines are used for the transportation of electric energy over long distances economically. The following distributed parameter equations of the transmission line are used for the analysis of transmission line voltage and current along the length of the line: Vx = cosh( γx)Vr + Zc ⋅ sinh( γx)I r (10.26 shows that optimum current density depends upon the factor a2.99 0. Transmission line voltage up to 800 kV is operational in many countries. and a 1150 kV EHV AC line is operating in Russia (Alexandrov et al.27) . The value of r is important — as we approach superconductivity.0 AAAC(1) 61 790 475 48 0. a lumped parameter equivalent of a line is no longer accurate.96 0. and the resistivity of conductor material. which represents that portion of the capital cost of line that depends upon conductor size. interest rate. For longdistance transmission line analysis.01 ACSR/AS20 54/7 805 425 43 0.95 0.96 ACAR 54/7 815 400 36 1. m Tension. Load factor.97 1.0 1.97 0. (Lf ∝ Ls). S. r. 1998). the optimum current density. J. and Applications and the optimum current density. (k ∝ i) . TABLE 10.97 0.0 1. d. will become very high. there are some experimental lines capable of reaching voltages up to 1100 kV but they are not in operation. and a distributed parameter representation of the line is used for transmission line analysis. energy cost.98 0.97 AAAC(2) 61 810 425 38 0. At the present time the highest transmission line voltage in North America is 765 kV. Modeling. kN Loss1 Line cost1 Transmission cost1 1 ACSR 54/7 780 425 43 1.97 0.98 0.26) Equation 10.97 Cost with respect to ACSR Conductor diameter = 31 mm AAAC (1) = 53% IACS AAAC (2) = 56% IACS 10.
78 × 10–9 F/km Conductor Type: ACSR Diameter: 35. D constants of the line.28) x = distance from receiving end. and 2 SIL.29) The following example will illustrate some interesting features of longdistance transmission.1 mm Rdc @ 20°C: 0.04 ohm/km .5 SIL. Find the following: 1. and 2 SIL.35 × 10–4 H/km G=0 C = 12.Applications 199 Ix = sinh( γx)Vr + cosh( γx)I r Zc (10. C. Example 10.01 ohm/km L = 8. Transmission line current as a function of line distance for power transmission equal to 0.5 SIL. Surge Impedance Loading for this line 2. The following line constants are assumed for the 765kV line: R = 0. B. 4.4 It is proposed to supply large amounts of cheap hydroelectricity by Extra High Voltage (EHV) transmission line from a location 2500 km away from the load center. km Vx = Voltage at a point x in line Vr = Voltage at the receiving end of line I r = Current at the receiving end of line I x = Current at a point x in line γ = propagation constant Zc = characteristic impedance of the line The above equation is derived from Appendix 10 at the end of this chapter. Line ampacity and maximum power transmission capacity 3. Vs A Is = C B Vr D Ir (10. It can be represented by the following matrix equation utilizing the well known A. 1 SIL. 1 SIL. Transmission line voltage as a function of line distance for power transmission equal to 0. The transmission line voltage is 765 kV AC.
Zc = (R + jL ⋅ ω ) (G + jC ⋅ ω ) Zc = 255. Ampacity/subconductor = 1920 A Line Ampacity = 4 × 1920 = 7680 A . γ = Z⋅Y γ= (R + jL ⋅ ω )(G + jC ⋅ ω ) γ = 2 ⋅ 10 –5 + j1.15 (765 ⋅ 10 ) SIL = 255.6 3 2 ⋅ 10 –6 SIL = 2289 MVA 2. Line ampacity is calculated by the program by following the procedure described in Chapter 3 for the speciﬁed transmission line conductor. Modeling. Surge Impedance Loading (SIL) Propagation constant is calculated.6 − j4.200 Powerline Ampacity System: Theory. and Applications Number of subconductors / phase: 4 Number of circuit: 1 Frequency: 60 Hz Latitude: 54°N Longitude: 77°W Time of day: 2 pm Day: Dec 12 Meteorological Conditions: Ambient temperature: 0°C Wind speed: 1 m/s Wind direction: 90° with respect to conductor Sky condition: Clear sky Solution 1.2 ⋅ 10 –3 The characteristic impedance is calculated. and by consideration of four subconductors per transmission line phase. meteorological conditions.
Ir = 2289 3 ⋅ 765 ⋅ 10 3 Ir = 3455 Line to ground voltage Vr.6 – j4.2 ⋅ 10 –3 ⋅ x ⋅ 441 ⋅ 10 3 255. It is interesting to observe the variation of line current as a function of the distance. 10.28. From this ﬁgure we can see that there is substantial increase in line voltage at midpoint when power transfer is increased beyond the surge impedance load of the line.13.10 ⋅–5 + j1.15 Ix = {( ) } + cosh 2. From this ﬁgure we can see that the thermal limit of the line is not exceeded even at two times the surge impedance loading (2 SIL) of the line.27 and is shown in Figure 10.2 ⋅ 10 –3 ⋅ x ⋅ 3455 {( ) } The value of line current as a function of the distance from receiving end is shown in Figure 10. is. Transmission and distribution systems are generally provided with overcurrent and earth fault relays.Applications 201 Maximum power transmission capacity of the line = 3 · 765 · 7680 · 10–3 = 5875. this power is sufﬁcient for many metropolitan cities. Vr = 765 ⋅ 10 3 3 Vr = 441 kV The transmission line current as a function of distance is obtained from Equation 10. Transmission line current as a function of line distance The receiving end currentm Ir. and voltage and underfrequency relays. These protective devices are designed to offer .2 MVA = 5875. 4.10 –5 + j1. sinh 2. 3.2 MVA To give an idea.12. differential relays.6 PROTECTION The effect of variable transmission line ratings on system protection requires careful evaluation to ensure proper functioning of the protective relaying system for both transmission and distribution lines. impedance relays. It is interesting to observe the variation of line voltage as a function of line distance. Transmission line voltage as a function of line distance is calculated similarly from Equation 10.
Energy Cost 100 90 80 70 60 50 40 50 30 20 10 20 30 40 50 60 70 80 90 100 Temperature. As seen in this ﬁgure the optimum value of transmission line conductor temperature increases with lower electric energy cost. Optimum Conductor Ampacity 230 kV Line ACSR Cardinal Transmission Cost.9 Optimum value of transmission line ampacity is 848 A. Modeling. $/MVA/Km 2500 2000 1500 1000 500 0 84 0 10 57 12 24 13 61 14 78 15 83 16 78 18 00 50 0 Ampacity. and Applications Optimum Conductor Temperature 230 kV Line ACSR Cardinal 2500 Transmission Cost. Optimum Conductor Temperature vs. C o Energy Cost.202 Powerline Ampacity System: Theory. C FIGURE 10.10 Optimum conductor temperature as a function of energy cost. .8 Optimum conductor temperature is 80°C as seen in the above ﬁgure. A FIGURE 10. $/MWh FIGURE 10. $/MVA/Km 2000 1500 1000 500 0 50 70 90 110 130 150 170 190 210 o Conductor Temperature.
$/MWh FIGURE 10. A 2 SIL 6000 5000 4000 3000 2000 1000 0 0 500 1000 1500 Distance.12 Variation of transmission line current as a function of distance for different power transmission levels.5 SIL 1 0.5 0 0 500 1000 1500 2000 2500 Transmission Distance.11 The optimum value of transmission line ampacity increases with lower electric energy cost. km 2000 2500 SIL 0. .5 SIL 0.13 Variation of transmission line voltage as a function of distance for different power transmission levels.5 2 SIL 2 1.5 SIL FIGURE 10. Energy Cost 1200 Ampacity. Transmission Line Current vs Distance 8000 7000 Line Current. Sending and Receiving Voltage Ratio Transmission Line Voltage vs Distance 2. km FIGURE 10.Applications 203 Optimum Transmission Line Ampacity vs. A 1000 800 600 400 200 0 20 30 40 50 60 70 80 90 100 Energy Cost.
a complete formulation of transmission line economics was presented. preferably on a realtime basis. the system must be able to distinguish between permissible overload current and a fault current to avoid faulty tripping during an acceptable overload condition. transformers. Automatic reclosures are also provided on most circuits for automatic recovery from temporary faults. loadﬂow. transformers. the protective relaying settings will have to be updated continuously. 10. Electricity production costs were evaluated with static and dynamic line rating in a transmission network having diverse generation sources. generator stability. resulting in winter and summer ratings. This is the subject of adaptive relaying and beyond the scope of this book. or the reconductoring of existing wires. For all of the above protection schemes. Examples were provided in the chapter to show the impact of high transmission line ampacity on steadystate generator stability. optimum current density. A formulation of the optimal power ﬂow problem was given to show the signiﬁcance of transmission line dynamic thermal ratings in the economic operation of an interconnected electric power system having diverse generation sources. with substantial cost savings. and the evaluation of alternative conductor designs. In a static system. The application of a powerline ampacity system in the planning and design of new overhead lines includes the selection of optimum conductor size. Cigré. For networks having dynamic line ratings. and dynamic and transient stability. protection relays are timecoordinated so that the circuit breaker closest to the fault opens ﬁrst.7 CHAPTER SUMMARY Applications of powerline ampacity system to power system economic operation. To ensure fault discrimination. may be postponed in many cases by dynamic line ratings. and Applications protection to transmission lines. It was shown that power system stability limits are enhanced by dynamic line ratings. and similar conferences. as well as the loads they serve. Backup protection is provided so that if the breaker closest to the fault fails to operate. and design considerations of overhead powerlines in view of powerline ampacity were presented in this chapter. the relay pickup current setting for overload protection was determined by static line ratings. substation busbars. the ratings of transmission lines. and other substation currentcarrying equipment are usually considered to be constant during a season. distribution feeders. For satisfactory functioning of the protection system.1988) and other excellent technical papers in IEEE. Modeling.204 Powerline Ampacity System: Theory. Traditionally. . For this purpose. the magnitude of the overcurrent and relay operation times are determined from network shortcircuit studies with proper prefault and postfault line operating conditions. The interested reader is referred to an excellent book on the subject of adaptive computer relaying (Phadke et al. Reclosurers are circuit breakers that close automatically at predetermined intervals after opening a circuit to eliminate faults that are temporary in nature. the next breaker will open. and generators. Results were presented to show the savings in electricity production cost achieved by the dynamic rating of transmission lines using LINEAMPS. Economic analysis of existing lines show that the construction of new lines. transmission line planning.
dVx = Vx – Vx′ = dZ I x – dI x ≈ dZ ⋅ I x = Z ⋅ I x dx dI x = I x – I x = dY ⋅ Vx = Y ⋅ Vx dx ′ ( ) (A10.5) 205 .3) (A10.2) where the product of differential quantities are neglected. From above.1) (A10.Appendix 10 Transmission Line Equations I1 Ix dZ I x = Ix – dI x V1 Vx dY Vx = Vx – dVx V2 x 0 FIGURE A10.4) (A10.1 Long transmission line model. Voltage. Applying Kirchoff’s law to a small section dx (dx << λ) of the line we have. the following secondorder linear differential equations are obtained. d 2 Ix = Z ⋅ Y ⋅ Ix = γ 2 Ix dx (A10. d 2 Vx = Z ⋅ Y ⋅ Vx = γ 2 Vx dx d 2 Vx − γ 2 Vx = 0 dx Current.
7) (A10.6) are.9) (A10. Modeling.17) substituting x = .11) (A10.206 Powerline Ampacity System: Theory.15) Vx = cosh( γx)V2 + Zc ⋅ sinh( γx)I 2 Ix = sinh( γx)V2 + cosh( γx)I 2 Zc (A10.6) (A10.13) Vx = V2 e γx + e – γx 2 ( )+Z I )+I 2 c 2 (e γx − e – γx ) (A10. we obtain the sending end voltage current as follows. .10) (A10.4). γ = Z⋅Y The solutions of differential equations (10.16) (A10. (10. V2 = k i + k r Zc I2 = k i − k r ki = V2 + Z c I 2 2 V2 − Z c I 2 2 k i γx k r – γx e − e Zc Zc (A10.8) (A10. Vx = k i e γx + k r e – γx Ix = at x = 0 we have.14) Ix = V2 e γx + e – γx 2 Zc ( 2 (e γx − e – γx ) (A10. and Applications d2Ix − γ 2 Ix = 0 dx where the propagation constant γ is given by.12) kr = (A10.
and current. is. Vs A = Is C where.18) and (A10.Appendix 10 Transmission Line Equations 207 Vs = cosh( γ ⋅ l)V2 + Zc ⋅ sinh( γ ⋅ l)I 2 Is = sinh( γ ⋅ l)V2 + cosh( γ ⋅ l)I 2 Zc (A10. and current. Vs. Ir. Is. A = cosh( γ ⋅ l) B = sinh( γ ⋅ l)I r C= sinh( γ ⋅ l) Zc B Vr D Ir (A10.19) Equations (A10. to receiving end voltage.19) in matrix form relating sending end voltage.18) (A10. Vr.20) D = cosh( γ ⋅ l) .
and who offered solutions to increase line capacity. By developing a system of rating overhead lines as a function of forecasting weather conditions by an objectmodel and expert system. Utility line rating practices. a threedimensional conductor thermal model is developed.1 SUMMARY As the demand for electricity grows in all regions of the world. It is stated that voltage and stability limits can be improved by control of reactive power and/or boosting voltage levels by transformer action. Therefore. Chapter 1 introduced the subject of transmission line ampacity and presented the line ampacity problem. and improve land use. and a methodology suitable for implementation in a computer program. The line ampacity system described in this book offers an integrated line ampacity system comprising a transmission line model. forecast ratings. and several researchers from different countries who were concerned with the problem of transmission line ampacity. and a design tool for the construction of future T&D facilities. when demand increased. and transient thermal rating models are developed from it. electric utilities added new T&D capacity by the construction of new lines and substations. algorithms. Faraday. probabilistic ratings. a conductor model. Historically. a complete system of rating overhead powerlines is developed giving theory. Chapter 2 described line rating methods from the early works of Ampere. public authorities are paying greater attention to properly locating electric power facilities and proper land use. The concept of 209 . Future Plans and Conclusion 11. and then steadystate. a very userfriendly program is realized that is easily implemented in all geographic regions. and a weather model for the ﬁrst time. there is greater need to develop better ways of operating existing Transmission and Distribution (T&D) networks for more efﬁcient utilization of existing facilities. and to population growth.11 Summary. and online and ofﬂine methods are discussed. but is also a valuable planning tool for line maintenance. The program is not only an operational tool enabling better utilization of existing transmission and distribution facilities. including realtime ratings. First. In this book. increase energy efﬁciency. there is greater R&D effort by power companies and concerned government agencies to minimize environmental impact. As a result. and static line ratings. in many cases the transport capacity of overhead powerlines is limited only by the thermal rating of powerline conductors. Due to public concern about protecting the environment. dynamic. Chapter 3 provided a complete theory of conductor thermal ratings.
Examples of hourly values of future meteorological conditions generated from the models are given. Then. wind direction. the development of weather models of ambient temperature. different types of powerline conductors. and Applications steady. This aspect of transmission line ampacity is signiﬁcant because there is little previous work carried out in this direction. and transient ratings are introduced for the ﬁrst time in this chapter and are related to machine stability in Chapter 10. and solar radiation were presented. planning. A signiﬁcant contribution made by the program is that it not only predicts line ampacity. dynamic. measures are suggested for the reduction of electric and magnetic ﬁelds from transmission lines by new conductor conﬁgurations.210 Powerline Ampacity System: Theory. Application of fuzzy set modeling of transmission line ampacity is also given. Results were compared with data from industry standards and transmission line ﬁeld data with excellent agreement. on outdoor transmission line test span. wind speed. and operation of overhead powerlines. The database is comprised of powerlines. and by active and passive shielding. Modeling. Since weather is an important parameter in transmission line ampacity calculations. A complete theory with examples of the method of calculation of the electric and magnetic ﬁelds from a line were presented to form a clear understanding of the subject. First. and weather stations for any geographic region. Chapter 5 presented elevated temperature effects of different types of transmission line conductors. Chapter 7 described various approaches to weather modeling for the prediction of line ampacity. This chapter presents uptodate knowledge in the ﬁeld of line ampacity calculations by comparing different line ampacity calculation methods. Even though there is no evidence of any signiﬁcant environmental impact by EMF due to increased transmission line current. The probability distribution of conductor temperature in service was obtained by a MonteCarlo simulation of time series stochastic models. neural network modeling was presented for forecasting and pattern recognition. . The program offers a userconﬁgurable transmission line database for the ﬁrst time. Experimentallydetermined empirical models of loss of tensile strength of conductors and permanent elongation due to elevated temperature operations were presented with results from each model. as well as ﬁeld measurements on real transmission line circuits. but also offers a complete environment for the management. A probability method of calculation of transmission line sag and tension was presented for the ﬁrst time. Chapter 6 presented a study of the electric and magnetic ﬁelds from highvoltage power transmission lines. Computer algorithms were presented for the recursive estimation of loss of strength and permanent elongation. Chapter 8 described the computer modeling method of the line ampacity system and its implementation in a computer program called LINEAMPS. statistical weather models based upon timeseries analysis of National Weather Service forecasts were developed. thanks to recent developments in object oriented modeling technology (Cox 2000). Chapter 4 described experimental research in the laboratory. Theoretical results that were obtained by calculation from the transmission line conductor thermal models developed in Chapter 3 are compared to IEEE and Cigré standards with excellent agreement.
and loadﬂow and generator stability issues. new conductors and new weather stations. For this purpose. An important feature of the new line ampacity system in a competitive electricity supply market is the ability to forecast hourly values of line ampacity up to seven days in advance. last but not least. It is a userfriendly program created by the application of artiﬁcial intelligence using objectmodel and expert system rules. describes modeling transmission lines and weather stations for the region of New Zealand. It shows the suitability of the program in all geographic regions of the world. STATCOM. Cost saving by the implementation of a variable line rating system in an interconnected electric network was presented by example. North and South Island. Thyristor Controlled Series Compensation (TCSC). Paris. economic power system operation. 1996.3) Hydro Quebec. LINEAMPS is being used or evaluated in the following power companies in the different regions of the world. An example of an actual FACTS installation in a highvoltage substation is also provided. As of this writing. there is a lowering of environmental impact. Chapter 10 presented applications of the line ampacity system to transmission system planning. optimum current density. LINEAMPS for Korea User Manual. enabling advance purchase and sale of electricity. * LINEAMPS for New Zealand User Manual. The various types of FACTS devices are described in this chapter. including Static Var Compensating (SVC) devices. South Korea (Figure 11. a complete formulation of transmission line economics was presented. Real examples* of computer modeling of the line ampacity system in two different geographic regions of the world are presented in Chapter 8.4) Electricité de France (EDF). . The cost of electricity may be reduced by greater utilization of economy energy sources and. • • • • TransPower. both visual and ecological. and Superconducting Magnetic Energy Storage (SMES).5) Korea Electric Power Company. New Zealand (Figure 11. Economic analysis of existing lines shows that the construction of new lines or the reconductoring of existing wires can be postponed in many cases by dynamic line ratings. and the evaluation of alternative conductor designs. It is also a contribution to the ﬁeld of cognitive science applied to electric power transmission where a computer program resembles a human expert. France (Figure 11. A formulation of the optimal power ﬂow problem was presented in Chapter 10 to show the signiﬁcance of transmission line dynamic thermal ratings to the economic operation of an electric power system with diverse generation sources.2) Chapter 9 discussed the state of the art in the development of power semiconductor devices for FACTS applications. with substantial cost savings.Summary. Users of the program have less chance of making errors in data input because expert rules check user input and explain error messages. Future Plans and Conclusion 211 LINEAMPS provides the ability to create new line. The application of the powerline ampacity system in the planning and design of new overhead lines includes the selection of optimum conductor size. 1995. Canada (Figure 11. describes modeling transmission lines and weather stations for the region of South Korea. UPFC.
212
Powerline Ampacity System: Theory, Modeling, and Applications
11.2 MAIN CONTRIBUTIONS
The Concept of Steady, Dynamic and Transient Line Rating Most of the literature on transmission line conductor thermal rating considers the shortterm thermal rating of a conductor as its transient thermal rating. In effect, it would be more appropriate to call shortterm rating “dynamic line rating,” because the time span is of the order of several seconds, which corresponds well with the time span considered in dynamic stability studies on generators. The transient thermal rating of the conductor is considered when there is a shortcircuit or lightning currents. Therefore, the time span of transients is much shorter, generally in the order of milliseconds, and corresponds well with the time span considered in transient stability studies on generators. Thus, the concept of steadystate rating, dynamic rating,* and transient line rating were introduced for the ﬁrst time in relation to generation stability as presented in Chapter 10. The above terminology of the different transmission line ratings is consistent with other areas of the electric power system relating to the steadystate stability, dynamic stability, and transient stability of electrical generators. The steadystate transmission line conductor thermal rating is applicable to the condition of steadystate operation of a power system; dynamic thermal rating is applicable during power system dynamic operations; and transient thermal rating is applicable during power system transient operating conditions. ThreeDimensional Conductor Thermal Model A threedimensional conductor thermal model was used in a previous wind tunnel study conducted by the author at PG&E to determine radial temperature distribution inside a transmission line conductor (Hall, Savoullis, Deb, 1988). A threedimensional conductor thermal model was developed in a recent report (Cigré, 1997) for the calculation of transmission line ampacity. In Chapter 4, the differential equation of conductor temperature was developed from a threedimensional conductor thermal model. The main reason for selecting this model is that it enables the calculation of radial temperature distribution and the average conductor temperature within the conductor. As we may recall from Chapter 5, the sag and tension of a transmission line conductor is calculated from the average conductor temperature. Transmission Line Risk Evaluation By evaluating weather conditions and transmission line temperatures in the region of Detroit (Davis, 1977), it was shown that transmission line capacity was greater than static line capacity for a large percentage of the time in that region. Davis also showed that static line rating is not risk free, and there exists a small percentage of time when maximum conductor temperature is exceeded even by static line rating.
* For a discussion on dynamic rating terminology, please see the discussion contribution in the references (Hall, Deb, 1988a).
Summary, Future Plans and Conclusion
213
Similarly, the research carried out at PG&E in California also showed that actual transmission line ampacity is signiﬁcantly greater than static line ratings (Hall and Deb, 1988b) for a large percentage of time. A transmission line risk analysis study was carried out at PG&E which showed there is minimum risk to a dynamic line rating system (PG&E Report. 1989).* The results presented in this book show that there is minimum risk as conductor sag; loss of strength and EMF were calculated and found to be within acceptable limits. RealTime Rating Realtime line rating systems (Davis, 1977), (Seppa, T.O., 1998), (Soto et al., 1998), (Deb, 1998) calculate line ampacity by the measurement of weather variables, conductor temperature, and/or conductor tension. In Davis’s system, realtime line ratings were calculated by monitoring conductor temperature with sensors installed at different sections of the line. In the PG&E system (Mauldin et al., 1988), (Steeley et al., 1991), (Cibulka et al., 1992), line ampacity is calculated by monitoring ambient temperature only. The PG&E line rating system also offered one to 24 hours ahead forecast rating capability. For this reason, realtime stochastic and deterministic models were developed by the author at PG&E to forecast transmission line ampacity up to 24 hours in advance (Hall, Deb, 1988b), (Steeley et al., 1991). Another realtime line rating system calculates ampacity by the measurement of conductor tension (Seppa, T.O., 1998). The relationship between transmission line conductor temperature and tension is presented in the conductor change of state equation in Chapter 5. By the application of the change of state equation we can easily calculate conductor temperature if conductor tension is measured. Ampacity is then calculated from conductor temperature by the application of the conductor thermal models presented in the Chapter 3. It is difﬁcult to obtain forecast ratings by monitoring conductor tension or conductor temperature alone. It is easier to forecast line ratings by forecasting weather conditions. Installation of temperature sensors or tension monitors requires taking the line out of service for periodic maintenance. It also requires communication with sensors installed on the line to the utility power control center computer where line ampacity is calculated by a program. The prediction of line ampacity by monitoring weather conditions does not require any new hardware to be installed on a transmission line. Weather conditions are generally monitored in electric utility power systems for other reasons, such as load forecasting, etc. Therefore, realtime weather data is available in most electric power company control centers at no additional cost. It is also expensive to install temperature or tension monitoring devices on all transmission lines. Thus, at the present time, these devices are installed on a limited number of heavilyloaded critical transmission circuits.** For the above reasons, a line monitoring system is expected to complement the more generalpurpose line rating system developed in this book. The line rating
* The report presents a quantitative analysis of the risk of an ambient adjusted line rating system developed at PG&E. ** (Waldorf, Stephen, P., Engelhardt, John S., 1998). This paper describes realtime ratings of critical transmission circuits by monitoring temperature of overhead lines, switchgear, and power transformers.
214
Powerline Ampacity System: Theory, Modeling, and Applications
methodology presented in this book offers a more economical system that can be easily implemented for the rating of all lines in all geographic regions.* (See Figures 11.2–11.5). The Spanish system (Soto et al., 1998) is already beginning to follow this approach. However, their system lacks forecast rating capability, weather modeling, and the ability to make new conductors, new weather stations, and new transmission lines, which are the objects of the author’s patent (Deb. 1999).**
FIGURE 11.2 LINEAMPS for Hydro Quebec. Map shows the geographic region of Quebec, Canada and weather stations created by program. A 765 kV transmission line from La Grande hydroelectric stations to Montreal, Quebec is shown in the ﬁgure.
* LINEAMPS for HydroQuebec, Canada, Software Users Guide, 1998, describes the line campacity system developed for HydroQuebec in Canada. LINEAMPS for EDF, France, Software Users Guide, 1998, describes the line ampacity system developed for EDF, France. LINEAMPS for TransPower, New Zealand, Software Users Guide, 1996, describes the line ampacity system developed for TransPower, NZ. LINEAMPS for KEPCO, S. Korea, Software Users Guide, 1996, describes the line ampacity system developed for TransPower, NZ. ** Anjan K. Deb, 1999. “Objectoriented line ampacity expert system.” U.S. Patent.
Summary, Future Plans and Conclusion
215
Forecast Rating In the line ampacity system developed in this book, ampacity is forecast up to seven days in advance by adjusting the LINEAMPS weather model to National Weather Service forecasts. Hourly values of ambient temperature, wind speed, and solar radiation are generated by AmbientGen, WindGen, and SolarGen methods in each weather station object as described in Chapter 7. At the present time, line ampacity forecasts are limited to seven days in advance because weather forecasts are less accurate beyond that period.*
FIGURE 11.3 LINEAMPS for EDF, France. EDF is the national electric company of France. Weather stations are created by the LINEAMPS program. A 400 kV transmission line from Paris to Bordeaux created by program is also shown in the ﬁgure.
Weather and Line Current Modeling A timeseries stochastic model of ambient temperature for the prediction of transmission line conductor temperatures was ﬁrst presented by the author (Deb, 1985) at the (Cigré Symposium 1985)** on High Currents and (Hall, Deb, 1988b). The BoxJenkins forecasting model is quite accurate, but the model coefﬁcients could not be easily adapted in realtime. Therefore, a recursive least square estimation model was developed at PG&E suitable for realtime calculation (Steeley, Norris, and Deb, 1991), (Cibulka, Steeley, Deb, 1992). Statistical models based on hourly differences of ambient temperature were also proposed by others (Douglass, 1986), (Foss, Maraio,
* Weather section, USA Today, November 23, 1998. ** The Cigré Symposium was exclusively devoted to the subject of transmission line ampacity.
216
Powerline Ampacity System: Theory, Modeling, and Applications
FIGURE 11.4 LINEAMPS for KEPCO, South Korea. KEPCO is the national electric company of South Korea. Weather stations are created by the LINEAMPS program. A 345 kV transmission line from Seoul to Wonju, also created by program, is shown in the ﬁgure.
1989) but it was shown in a discussion contribution* that realtime recursive estimation was more accurate. The following forecasting models were evaluated: • • • • Fourier series model Recursive least square estimation algorithm Kalman Filter Neural network
The results presented in the Chapter 3 and Chapter 4 show the accuracy of each model. A transmission line current model is useful for the prediction of conductor temperature, and may be developed by using the forecasting models mentioned above (Deb et al., 1985). The Fourier series model of ambient temperature and wind speed is used in the LINEAMPS program. This model is recommended for its simplicity, and is suitable when a generalpurpose weather forecast of a region is available.** A neural network is useful for weather pattern recognition, as shown in Section 4.3. A recursive
* Discussion contribution by Anjan K. Deb and J.F. Hall to the IEEE paper (Douglass, D.A., 1988). ** Generalpurpose weather forecast issued by the National Weather Service includes daily maximum and minimum values of ambient temperature for three to seven days in advance. The data is generally published in daily newspapers as well as in several Internet web sites: www.intellicast.com, www.nws.noaa.gov.
Transmission lines are classiﬁed by voltage levels. 1998). communication by message.. estimation algorithm and Kalman ﬁlter are suitable for realtime forecasting of transmission line ampacity on an hourly basis. polymorphism. 1993). E. New Zealand. By using the objectmodel approach. Future Plans and Conclusion 217 FIGURE 11. and is seriously considered by electric power companies (MPS Review Article. . weather station data * KappaPC ver 2. weather stations and transmission line conductors in the LINEAMPS program (Deb. 1995. and delegation of messages (Booch. and other substation equipment.* The objectmodel is particularly suitable for electric power system applications for the modeling of transmission lines. G. 2000). (Cox. 1998). TransPower operates the transmission grid in New Zealand. The objectmodel approach is used to model transmission lines. Examples of the objectmodel that were created for the regions of New Zealand and South Korea are presented in Chapter 5. It is seen in this chapter how a systematic method of transmission line ampacity system data classiﬁcation is developed. (KappaPC). generators.Summary. which emables the LINEAMPS program to access transmission line and weather data.4 provides connection to industry standard databases by ODBC. A 375 kV transmission line from Auckland to Wellington in the North Island of New Zealand is also shown in the ﬁgure. 1997..5 LINEAMPS for TransPower. it is shown in Chapter 5 how transmission line objects are easily created with attributes and behavior by class inheritance. The weather stations shown in the above map are created by the LINEAMPS program. Transmission Line ObjectModel Objectoriented modeling is a new way of developing computer programs that makes use of inheritance.
Transmission Line Thermal Rating Studies. 1978) gives the static line rating for summer and winter of all conductors used by PG&E for normal and emergency conditions. . showed that LINEAMPS ratings never exceeded realtime ratings at any time. Kevin M. and Applications is classiﬁed by geographic regions. 1998). Chicago. **** Thanks are due to Mr. * (Taylor et al. 2000).m. Technical Expert. As shown in Chapter 8. and the corresponding assumptions of meteorological conditions. Nandi. Experimental Veriﬁcation of Transmission Line Ampacity Experimental work carried out at the PG&E wind tunnel to verify the conductor thermal model (Hall. 1998). there are other expert systems developed for the power industry for fault diagnosis (Taylor et al.S. and conductor data is classiﬁed by conductor types.. checks user input data. Hourly values of an actual 345kV transmission line ampacity was available from Commonwealth Edison Company**** (ComEd). 1988) is described in Chapter 6. 1998).. *** (Southwire. Results of steady state ampacity and dynamic and transient ampacity obtained from the LINEAMPS program are presented in a table and compared to wind tunnel data and other data compiled from various power companies (Urbain J. M. Modeling. ** (PG&E Standard. 1978)** and a conductor manufacturer’s catalog (Southwire. LINEAMPS also accurately predicted the lowest rating at 12:00 noon. S. expert systems use an inference engine to make decisions. 1994) provides the ampacity of all commonly used conductors in the U. Deb. Line Ampacity ExpertSystem The declarative style of programming by rules was used to develop the LINEAMPS program.. 1995).218 Powerline Ampacity System: Theory. while in a procedural language the programmer writes the code for decision making. (PG&E Standard. and the corresponding assumptions of meteorological conditions.. The LINEAMPS program uses an expert system to check user input data. Tribble.N. A comparison of this data from 8:00 a. B. 1998). Commonwealth Edison. 1998) provides a list of expert systems developed for the power industry worldwide. The line ampacity data was obtained from a realtime line rating system operated by ComEd.. Chicago. A new equation was developed from wind tunnel data to determine the Nusselt number as a function of the Reynolds number (Ozisik. rules of thumb and practical knowledge are easily implemented by an expert system. and explains error messages like a true expert. Savoullis.m. IL. These are implemented in the LINEAMPS program to enable users to easily create their own objects.P. 1994)*** with excellent agreement.* intelligent tutoring systems (Negnevitsky. to 4:00 p.. Alan C. The values of steadystate ampacity and dynamic ampacity calculated by the LINEAMPS program also compared well with the IEEE standard and a recent Cigré report.. 1985) for the calculation of convection cooling by wind. Similarly. There are several advantages of programming by rules instead of purely by procedures (Kronfeld. and power quality (Kennedy. For example. LINEAMPS is the ﬁrst transmission line expert system using objectoriented modeling and rules (Deb. Examples are presented to show how the program calculates powerline ampacity by objects and rules.
1978). A uniﬁed approach to the modeling and evaluation of the effects of higher transmission line ampacity is presented in Chapter 5. conductor sag and tension are also affected by elevated temperature operation due to high currents (Cigré. Future Plans and Conclusion 219 The above example showed that LINEAMPS ratings have minimum risk of exceeding allowable maximum conductor temperatures. 1998. The accuracy of sagtension program was tested with ALCOA program (Lankford. The magnetic ﬁelds of conductors with dynamic ampacity are presented in Chapter 6.Summary. 1980). Deb. Line design and EMF mitigation methods are suggested to lower transmission line magnetic ﬁelds in sensitive areas (Böhme et al. These effects are evaluated recursively from the probability distribution of conductor temperature by using Morgan’s equation (Morgan. Typical powerline conﬁgurations are evaluated to show the magnetic ﬁelds with high currents. In addition to the loss of tensile strength. Choi. 2000). The magnetic ﬁeld at ground level of a transmission line increases with line ampacity and conductor sag (Rashkes. Lordan. The same method was recently used to calculate the thermal deterioration of powerline conductors in service in Japan (Mizuno et al. and KEPCO’s transmission line ﬁeld data (Wook. The electric ﬁeld of a highvoltage line at ground level does not depend upon transmission line ampacity if the maximum design temperature of the conductor is not exceeded and minimum conductortoground distance is maintained. . They also showed that the frequency distribution of a conductor should consider the relationship between weather and line current.. In their study. They showed by simulation that conductor loss of strength is less than 1% by static line rating. In Chapter 5. 1997). 1998).. Morgan calculated the loss of tensile strength of conductor by a percentile method. 1978). the reduction in tensile strength of the conductor was used as the index of thermal deterioration of the conductor. Effects of Higher Transmission Line Ampacity When transmission line ampacity is increased. A new method is developed to determine the sag and tension of overhead line conductors with elevated temperature effects. 1998). The thermal effects include conductor loss of strength and permanent elongation of the transmission line conductor. Results are presented which compare well with all of the above data. the probability distribution of conductor temperature is generated by a MonteCarlo simulation of weather data from timeseries stochastic models and transmission line currents which consider the correlation between weather and load current. it is necessary to properly evaluate the thermal effects of the powerline conductor as well as the electric and magnetic ﬁelds of the transmission line. It is shown that the magnetic ﬁelds of overhead transmission lines with dynamic line ampacity are within acceptable limits. 1989). STESS program (CEA Report. Most studies on magnetic ﬁelds from powerlines are conducted with typical transmission line currents equal to or less than static line ratings.
Yuryevich. and Applications Line Ampacity Applications The application of a dynamic line rating system in the economic operation of an electric power system was ﬁrst presented by (Hall. In addition to the solution of the optimal power ﬂow problem by a classical solution of the nonlinear optimization problem (Bergen. the ampacity of existing overhead powerlines could not be fully utilized because existing Alternating Current (AC) circuits are mostly composed of passive elements having very little controllability. F.3 SUGGESTIONS FOR FUTURE WORK Overhead powerlines constitute the single most important component of the electrical power system. 1994). when demand increases.. artiﬁcial neural network (Lee. J. existing network control methods are not sufﬁcient to properly accommodate increased power ﬂows. Deb developed the economic analysis method of dynamic line rating system with adaptive forecasting to demonstrate power system operational cost savings by dynamic thermal rating. supply. certain lines are more heavily loaded and stability margins are reduced.. K. and later on by (Deb. Deb. Anjan K. 11. Due to the difﬁculty of controlling power ﬂow by existing methods. 1986). Substantial capital cost savings by the deferment of capital investment required for the construction of new lines and environmental beneﬁts were shown. A recent study (Yalcinov. Energy Transfer. and distribution environment is also proposed. and genetic algorithm GA (Wong. . and the development of new transmission line technologies with greater emphasis on renewable energy sources and minimum environmental impact. Economy.. Therefore. and Short. The power system will continue to evolve as our demand for electricity increases in a fair and competitive environment based on free market principles. A plan to develop the line ampacity system further for application in a deregulated electricity production. 1988a). Power plants and transmission lines require substantial investments. The theory and mathematical model of a hypothetical utility system are presented in this book to demonstrate electric power system economy achieved by the new line ampacity system.Y. the contributions made in this book to increase line capacity of existing overhead power transmission and distribution lines will have a signiﬁcant impact in the improvement of the overall performance of the electrical power system. T. It is hoped that these studies will lead to the optimum utilization of all resources. As a result. Overcoming Limitations of Existing AC Networks by the Improvement of Power System Stability. which must be carefully evaluated before new facilities are added. Therefore. and Reliability In the past.. 1998) were developed in the industry to obtain faster and more efﬁcient solutions to the optimal power ﬂow problem. M. 1998).220 Powerline Ampacity System: Theory. A. J. Modeling. as fewer lines are required.. 1998) on generation system cost optimization by neural network now conﬁrms that substantial cost saving is possible by increasing transmission line capacity. et al. Following is a list of areas that require further research for optimum utilization of existing assets.
Since transmission lines constitute the most important component of a power system. and in the Amazon in South America. With public concern about global warming. The above examples demonstrate the importance of developing a dynamic transmission line thermal rating system with FACTS for the improvement of power system stability. and increasing the power transfer capability of existing lines to achieve greater economy. Similarly. Greater Utilization of Renewable Energy Sources It has been said that the electric power system is the most complex system ever created by human beings. and to maintain stability. there exists a large potential for solar energy development in the Sahara desert. and hydroelectricity. it is now possible to exercise a range of control over the AC network in a manner that was not possible before. By locating FACTS devices at suitable locations on existing networks.Summary. Future Plans and Conclusion 221 FACTS devices are being developed and are used to control existing T&D networks. Similarly. there will be even tighter control on the emission levels of electricity generated by burning fossil fuels. there are other new developments such as UPFC. The world has an abundance of renewable energy sources. wind. are yet to be developed. electricity could be transmitted efﬁciently to a distance of about 2500 km. an interconnected transmission network is essential to connect all of these sources of energy. There is now greater emphasis on producing electricity from renewable energy sources including solar. and Superconducting Magnetic Energy Storage that will enable even greater control over power ﬂow through transmission lines. Thyristor controlled shunt and series capacitors are now widely used in the electric power system to increase the transmission capacity of existing lines. hydroelectric reserves in Alaska and Canada in North America. Research from the International Conference on Large High Voltage . SMES devices are installed for faster response and voltage support. In the past. with varying capacities. Therefore. These devices are presently installed at a number of locations worldwide. including dynamic and transient stability. the powerline ampacity system presented in this book will continue to develop further and is expected to play a vital role in the planning and operation of existing and future powerlines in all regions. it is also beneﬁcial for the enhancement of power system reliability. FACTS Technology Due to the development of FACTS technology it is now possible to fully exploit existing transmission line capacities. As these sources of energy are generally at remote locations. In addition to the SVC system. Increasing transmission line capacity by dynamic thermal rating adjusted to actual weather conditions is not only useful for economic energy transfer. When renewable energy sources are fully developed. For example. at all voltage levels up to 800 kV. PowerFormer. it is expected to meet all of our energy requirements for this planet. Extra High Voltage (EHV) transmission lines will be required to bring electricity from remote locations to major metropolitan areas and industrial centers.
with minimum environmental impact. 1998). it is hoped this will lead to greater cooperation between nations. Interconnection of a highvoltage power transmission network will also enable the ﬂow of electricity across time zones. the powerline ampacity system will become more useful for the precise control of the current ﬂowing through transmission line wires. It will send signals through the powerline communication network to remote power electronics (FACTS) devices for the precise control of power ﬂow through the lines. Development of New Transmission Line Technology New transmission line technology is required to achieve greater energy efﬁciency and reliability of supply by interconnection. and Applications Networks (Cigré) now shows that it is technically and economically feasible to transport electricity to about 7000 km by UltraHigh Voltage (UHV) lines (IEEE Power Engineering Review.** Similarly. * IEEE Power Engineering Review. Conductors with ﬁberoptic communication technology will be required for longdistance communication with minimum number of booster stations. O.P. Choi. 101 College Rd.222 Powerline Ampacity System: Theory.. and interconnection with neighboring countries. and compact conductors to lower transmission losses (JeanLuc Bousquet et al. . As transmission lines extend beyond political boundaries. ** North American Electric Reliability (NERC) Transmission Map. EHV lines are presently transporting low cost electricity to New York generated from renewable energy sources in Canada. Furthermore. 1997). Modern highvoltage transmission networks will be used to carry electric energy and data for the precise control of power ﬂow through the wires. EHV and HVDC lines are planned to bring lowcost hydroelectricity from Siberia to Japan. Here.* This modern trend in electricity transmission is already beginning to happen. Modeling. presents articles from several authors in different countries for the development of renewable energy sources and UHV transmission to meet the energy requirements of this planet in the 21st century. NERC. Tatat. Bonicel. The ﬁber optic core can be easily used as a continuous wire temperature sensor. thus eliminating the uncertainty of presentday line ratings. more research is required to develop high temperature conductors using advanced aluminum alloys (Wook. and a utility transmission line network could become a part of the Internet for the reliable transmission of data. Deb. thereby taking advantage of the difference in timeofday demand for electricity. With further development of the highvoltage transmission network to accommodate the diverse sources of energy. again. in addition to other business and commercial uses. NJ 08540. which is free from electromagnetic disturbances. there is greater opportunity for the line ampacity system to develop further. the Internet will be used extensively for utility data communication. Princeton. 1998. bringing greater peace and prosperity. Advanced systems of communication are being developed using powerline communication by ﬁber optics (J. 1998). 1997) for higher transmission capacity. In the ﬁeld of overhead transmission line conductor technology.
et al. 1998). Underground transmission lines have no electric ﬁelds on ground surfaces since the outer layer of the cable is generally connected to ground and remains at ground potential. The space required for cable terminations may be around 2. AC underground lines by cable also have limitations on transmission distance due to the high capacitive reactance of cables. underground cables have to be well spaced to allow for natural cooling and to avoid overheating.. and the emission level of fossil fuel generators. .Summary. Whereas the surrounding air directly cools overhead lines. the electric and magnetic ﬁelds of typical transmission line conﬁgurations is calculated by the application of Maxwell’s equation. HVDC Light is a new underground power transmission and distribution system technology that is being developed for electricity distribution to remote areas. (Awad et al. UK. In the future. The space is needed because the highvoltage cable generates a great deal of heat — equivalent to a onebar electric ﬁre every two meters. Future Plans and Conclusion 223 Environmental Impact The world is paying greater attention to the environment (Alexandrov. land use and ecological effects. and active shielding to minimize EMF near sensitive areas. 1998. the powerline ampacity system methodology developed in this book will be extended to include underground cables and other currentcarrying transmission and distribution line equipment. ** This discussion is based on a report on the World Wide Web by the National Grid Company (NGC). Egypt. * There were other papers on this subject at Cigre 1998.N. there is a magnetic ﬁeld from the cable at ground surface. G. and well below acceptable limits. Since they are underground. To place a 400. modiﬁed phase conﬁguration. There are certain disadvantages of underground cables. Russia. and other transmission equipment..** Underground cables take up a much greater area of land than that needed for overhead lines of the same capacity. a small building.000volt line underground would typically involve digging a trench the width of a threelane road and 11/2 meters in depth to accommodate up to 12 separate cables. EMF mitigation measures are also suggested by compact line design. It is shown that the magnetic ﬁeld of a line is minimum at ground level. Additional land is needed at cable terminals where underground cables are joined to overhead lines. 1998)* by greater consideration of factors such as EMF from powerlines. This problem is overcome by HVDC transmission. Underground Transmission Underground lines are recommended for aesthetic reasons in cities and scenic areas.000 m2 in order to place a terminal tower (pylon) somewhat heavier in appearance than a normal suspension tower. Unless properly shielded. they have no visual impact on the surrounding environment. In Chapter 6.
Modeling.. a thermal model of the device is required along with a knowledge of the weather conditions at that location.224 Powerline Ampacity System: Theory. The calculation of equipment ampacity is much easier because. it is necessary to ensure that adjacent transmission circuits will have sufﬁcient line ampacity to safely carry the additional load of the line taken out of service. and Applications Dynamic Rating of Substation Equipment The powerline rating system methodology may be easily adapted to determine the ampacity of other transmission and distribution line equipment. A simple equipment rating model (Douglass and Edris. Korea Electric Power Company (KEPCO) engineers used the LINEAMPS program (Choi et al... 1996). τ = Thermal time constant of substation equipment t = time (for steady state rating t = ∞) Ta = Ambient temperature (actual or forecast) a = experimental constant If device ratings limit line capacity it is relatively less expensive to replace a device with a higher rating device. enabling equipment ratings to be easily determined from this data. Line Maintenance A line ampacity system having forecast capability is required for the planning of transmission line maintenance. Conductor replacement was carried out one circuit at a time on a doublecircuit line by leaving the other circuit energized. For the planning of conductor replacement work they used LINEAMPS ratings to ensure that the energized circuit had sufﬁcient capacity to carry the load of both circuits. (Soto et al. it is situated at one location in a substation. 1997) to safely reconductor existing transmission circuits by replacing ACSR conductors with highampacity Zirconium aluminum alloy conductor and Invar core. etc. Weather conditions are generally monitored in utility substations. For the calculation of equipment rating. When taking a line out of service to examine joints. . inspect conductors. clean insulators. unlike a transmission line. 1998) for implementation in the LINEAMPS program is given below: T –T max a – exp( – t τ) Tmax( r ) – Ta ( r ) I = Ir ⋅ 1 − exp( – t τ) I = Equipment ampacity at ambient temperature Ta Ir = Nameplate rating of equipment Tmax = Maximum equipment operating temperature Tmax(r) = Rated maximum temperature of equipment speciﬁed for nameplate rating at rated ambient temperature Ta(r).
however. An independent system operator (ISO) would ensure the reliability of the system as well as act as a clearinghouse for the purchase and sale of electricity through a competitive bidding process. Independent System Operator and Power Pool Operations The deregulation of the electric supply business is under active consideration worldwide (Cigré Panel Session.. 1998). the concept of electricity deregulation is accepted almost universally. ISO. In this scenario. 1998). This is not an easy task and much remains to be done in this area.1 Operation of an Open Electricity Market. (Cigré Regional Meeting. Similarly. and Canada (Electricity Today. D. Future Plans and Conclusion 225 As transmission line ampacity is increased by dynamic line ratings.1 is a simpliﬁed diagram of an unbundled system to illustrate the operation of a deregulated electric supply business. a line ampacity system will provide assistance to the ISO in determining transmission line capacity and pricing for the transmission of power from (G) to (D). 1998).S. and the transmission network is operated by a transmission company (T). but the rules governing a competitive electric energy marketplace.A. and power pool operations are not yet well established. . Deregulation. Independent generation companies operate power generation (G). B. 1996) and is already implemented in certain states (Barkovitch. Hawk. G G1 G2 G3 T D PD1 PD2 PD3 FIGURE 11. better techniques will be required to ensure that lines operate safely. Line maintenance using infrared imaging from a helicopter is becoming routine for many utilities (Burchette. that the rules should result in optimal system operations and encourage optimum system expansion plans. According to (Cigré Panel Session.) of the U. devices for the robotic maintenance of conductors and line hardware are being developed so that long transmission circuits can be inspected rapidly without human intervention in areas of high electric and magnetic ﬁelds. The application of the LINEAMPS program in the ideal operation of a power system as described in Chapter 8 is a contribution in this direction.. Sam N. There is general consensus. Figure 11. 1989).Summary. distribution of electricity to consumers is done by independent distribution companies (D).
In this scenario.gov/ .noaa. and near realtime data is available on the Internet.6 A proposal for LINEAMPS for North America (including USA. These data are adequate for LINEAMPS to determine powerline ampacity in all regions of the U./St. have revealed that many companies are still using a static line rating.4 A PLAN TO DEVELOP LINEAMPS FOR AMERICA Discussions with various electric power companies in the U.S. it will enable the development of a more competitive business environment for the purchase and sale of electricity. Canada.226 Powerline Ampacity System: Theory. LINEAMPS for America Thursday December 3.S. and even Mexico.* In addition.S.. Canada and Mexico). Modeling. As deregulation of the electricity supply business is embraced by all states in the U. determining transmission capacity will become more important. Canadian and Mexican weather data is easily available from various agencies. 1998 41 Seattle o o 40 o o Great Falls 57 40 Mpls. or longer) will enable * US WEATHER SERVICE NOAA Internet Web Site: http://www.S. Paul Boise 70 Denver 69 Phoenix o o o 57 Boston 65 o o o 56 San Francisco 65 Los Angeles o Chicago Wichita 66 o 63 New York o St Louis 73 Dallas o 67 74 Atlanta o o 69 Washington o 76 New Orleans 1 Fairbanks 81 Honolulu o o 82 Miami o Weather Data G G1 G2 G3 LINEAMPS T D PD1 PD2 PD3 FIGURE 11. and Applications 11. The NWS has an extensive network of weather stations connecting most regions of the U. there are several private weather service companies (Figure 11. daily. weekly. Advance knowledge of transmission capacity (hourly. Similarly. The LINEAMPS program can be easily adapted for implementation in these utilities with minimum cost.6) that deliver custom weather data.nws.. Near realtime weather data is available from the National Weather Service (NWS).
As the demand for electricity grows. Unpublished report. the need for higher transmission capacity will increase. With public concern about environmental protection. The implementation of this proposal will result in the development of a more efﬁcient transmission network in America. for support in the implementation of this program. and Virginia Power. . and new methods are required to maximize the utilization of existing lines. Anjan K. Oklahoma City. Future Plans and Conclusion 227 more efﬁcient electric energy futures trading by competitive bidding and spot pricing of electricity (Schweppe. In fact there is no reason why an electricity user should not be able to shop around in the electric energy market for the best rates. * LINEAMPS Dynamic Data Exchange (DDE) for Real Time Ampacity Calculations. new transmission lines were constructed to meet the needs of the electricity supply industry. and maintenance of lines. As seen in Figure 11. 11. A plan to develop LINEAMPS for America is shown in Figure 11. Washington DC. As of this writ