Distribution
System
Modeling
and
Analysis
The ELECTRIC POWER ENGINEERING Series
series editor Leo Grigsy
Published Titles
Electromechanical Systems, Electric Machines,
and Applied Mechatronics
Sergey E. Lyshevski
Electrical Energy Systems
Mohamed E. ElHawary
Electric Drives
Ion Boldea and Syed Nasar
Distribution System Modeling and Analysis
William H. Kersting
Linear Synchronous Motors:
Transportation and Automation Systems
Jacek Gieras and Jerry Piech
The ELECTRIC POWER ENGINEERING Series
series editor Leo Grigsby
Forthcoming Titles
Induction Machine Handbook
Ion Boldea and Syed Nasar
Power System Operations
in a Restructured Business Environment
Fred I. Denny and David E. Dismukes
Power Quality
C. Sankaran
Distribution
System
Modeling
and
Analysis
William H. Kersting
New Mexico State University
Las Cruces, New Mexico
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.
Visit the CRC Press Web site at www.crcpress.com
© 2002 by CRC Press LLC
No claim to original U.S. Government works
International Standard Book Number 0849308127
Library of Congress Card Number 2001035681
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
Kersting, William H.
Distribution system modeling and analysis / William H. Kersting
p. cm.  (Electric power engineering series)
Includes bibliographical references and index.
ISBN 0849308127 (alk. paper)
1. Electric power distribution–Mathematical models. I. Title. II. Series.
TK3001 .K423 2001
621.31—dc21 2001035681
CIP
0812_frame_FM.fm Page iv Tuesday, July 31, 2001 10:49 AM
Contents
1
Introduction to Distribution Systems
........................................... 1
1.1 The Distribution System...................................................................... 2
1.2 Distribution Substations ...................................................................... 2
1.3 Radial Feeders....................................................................................... 5
1.4 Distribution Feeder Map..................................................................... 6
1.5 Distribution Feeder Electrical Characteristics.................................. 8
1.6 Summary................................................................................................ 9
2
The Nature of Loads
..................................................................... 11
2.1 Deﬁnitions............................................................................................ 11
2.2 Individual Customer Load................................................................ 13
2.2.1 Demand ................................................................................... 13
2.2.2 Maximum Demand................................................................ 13
2.2.3 Average Demand.................................................................... 14
2.2.4 Load Factor ............................................................................. 14
2.3 Distribution Transformer Loading................................................... 15
2.3.1 Diversiﬁed Demand............................................................... 16
2.3.2 Maximum Diversiﬁed Demand........................................... 17
2.3.3 Load Duration Curve ............................................................ 17
2.3.4 Maximum Noncoincident Demand .................................... 17
2.3.5 Diversity Factor ...................................................................... 18
2.3.6 Demand Factor ....................................................................... 19
2.3.7 Utilization Factor.................................................................... 19
2.3.8 Load Diversity........................................................................ 20
2.4 Feeder Load......................................................................................... 20
2.4.1 Load Allocation ...................................................................... 20
2.4.1.1 Application of Diversity Factors........................... 21
2.4.1.2 Load Survey.............................................................. 21
2.4.1.3 Transformer Load Management............................ 25
2.4.1.4 Metered Feeder Maximum Demand.................... 25
2.4.1.5 What Method to Use? ............................................. 27
2.4.2 VoltageDrop Calculations Using Allocated Loads.......... 27
2.4.2.1 Application of Diversity Factors........................... 27
2.4.2.2 Load Allocation Based upon
Transformer Ratings................................................ 31
2.5 Summary.............................................................................................. 32
Problems.......................................................................................... 33
0812_frame_FM.fm Page v Tuesday, July 31, 2001 10:49 AM
3
Approximate Methods of Analysis
............................................. 39
3.1 Voltage Drop........................................................................................ 39
3.2 Line Impedance................................................................................... 41
3.3 “K” Factors .......................................................................................... 43
3.3.1 The
K
drop
Factor....................................................................... 43
3.3.2 The
K
rise
Factor........................................................................ 46
3.4 Uniformly Distributed Loads ........................................................... 47
3.4.1 Voltage Drop........................................................................... 48
3.4.2 Power Loss .............................................................................. 50
3.4.3 The Exact Lumped Load Model.......................................... 52
3.5 Lumping Loads in Geometric Conﬁgurations .............................. 55
3.5.1 The Rectangle.......................................................................... 55
3.5.2 The Triangle ............................................................................ 60
3.5.3 The Trapezoid......................................................................... 65
3.6 Summary.............................................................................................. 71
References ..................................................................................................... 71
Problems.......................................................................................... 71
4
Series Impedance of Overhead and Underground Lines
......... 77
4.1 Series Impedance of Overhead Lines.............................................. 77
4.1.1 Transposed ThreePhase Lines............................................. 78
4.1.2 Untransposed Distribution Lines ........................................ 79
4.1.3 Carson’s Equations ................................................................ 81
4.1.4 Modiﬁed Carson’s Equations............................................... 83
4.1.5 Primitive Impedance Matrix for Overhead Lines ............ 85
4.1.6 Phase Impedance Matrix for Overhead Lines .................. 86
4.1.7 Sequence Impedances............................................................ 89
4.2 Series Impedance of Underground Lines....................................... 95
4.2.1 Concentric Neutral Cable ..................................................... 96
4.2.2 TapeShielded Cables........................................................... 101
4.3 Summary............................................................................................ 105
References ................................................................................................... 105
Problems........................................................................................ 105
5
Shunt Admittance of Overhead and Underground Lines
...... 109
5.1 The General VoltageDrop Equation............................................. 110
5.2 Overhead Lines................................................................................. 111
5.3 Concentric Neutral Cable Underground Lines ........................... 115
5.4 TapeShielded Cable Underground Lines .................................... 119
5.5 Sequence Admittance....................................................................... 121
5.6 Summary............................................................................................ 122
References ................................................................................................... 122
Problems........................................................................................ 122
0812_frame_FM.fm Page vi Tuesday, July 31, 2001 10:49 AM
6
Distribution System Line Models
............................................. 125
6.1 Exact Line Segment Model ............................................................. 125
6.2 The Modiﬁed Line Model ............................................................... 132
6.3 The Approximate Line Segment Model ....................................... 136
6.4 Summary............................................................................................ 141
References ................................................................................................... 141
Problems........................................................................................ 141
7
Regulation of Voltages
............................................................... 145
7.1 Standard Voltage Ratings ................................................................ 145
7.2 TwoWinding Transformer Theory................................................ 147
7.3 The TwoWinding Autotransformer.............................................. 152
7.3.1 Autotransformer Ratings .................................................... 156
7.3.2 PerUnit Impedance............................................................. 158
7.4 StepVoltage Regulators................................................................... 162
7.4.1 SinglePhase StepVoltage Regulators .............................. 163
7.4.1.1 Type A StepVoltage Regulator............................ 163
7.4.1.2 Type B StepVoltage Regulator............................ 164
7.4.1.3 Generalized Constants.......................................... 167
7.4.1.4 The Line Drop Compensator............................... 168
7.4.2 ThreePhase StepVoltage Regulators ............................... 174
7.4.2.1 WyeConnected Regulators.................................. 175
7.4.2.2 Closed DeltaConnected Regulators................... 180
7.4.2.3 Open DeltaConnected Regulators ..................... 183
7.5 Summary............................................................................................ 193
References ................................................................................................... 194
Problems........................................................................................ 194
8
ThreePhase Transformer Models
............................................. 199
8.1 Introduction....................................................................................... 199
8.2 Generalized Matrices ....................................................................... 200
8.3 The Delta–Grounded Wye StepDown Connection.................... 201
8.3.1 Voltages.................................................................................. 202
8.3.2 Currents ................................................................................. 206
8.4 The Ungrounded Wye–Delta StepDown Connection............... 212
8.5 The Grounded Wye–Grounded Wye Connection....................... 222
8.6 The Delta–Delta Connection........................................................... 224
8.7 The Open Wye–Open Delta Connection ...................................... 236
8.8 The Thevenin Equivalent Circuit................................................... 242
8.9 Summary............................................................................................ 245
Problems........................................................................................ 245
0812_frame_FM.fm Page vii Tuesday, July 31, 2001 10:49 AM
9
Load Models
................................................................................. 251
9.1 WyeConnected Loads..................................................................... 252
9.1.1 Constant Real and Reactive Power Loads....................... 252
9.1.2 Constant Impedance Loads................................................ 253
9.1.3 Constant Current Loads...................................................... 253
9.1.4 Combination Loads.............................................................. 254
9.2 DeltaConnected Loads ................................................................... 257
9.2.1 Constant Real and Reactive Power Loads....................... 257
9.2.2 Constant Impedance Loads................................................ 258
9.2.3 Constant Current Loads...................................................... 258
9.2.4 Combination Loads.............................................................. 258
9.2.5 Line Currents Serving a DeltaConnected Load............. 259
9.3 TwoPhase and SinglePhase Loads .............................................. 259
9.4 Shunt Capacitors............................................................................... 259
9.4.1 WyeConnected Capacitor Bank........................................ 259
9.4.2 DeltaConnected Capacitor Bank ...................................... 260
9.5 The ThreePhase Induction Motor................................................. 261
References ................................................................................................... 266
Problems........................................................................................ 266
10
Distribution Feeder Analysis
..................................................... 269
10.1 PowerFlow Analysis ..................................................................... 269
10.1.1 The Ladder Iterative Technique ..................................... 270
10.1.1.1 Linear Network................................................ 270
10.1.1.2 Nonlinear Network.......................................... 271
10.1.2 The General Feeder .......................................................... 274
10.1.3 The Unbalanced ThreePhase Distribution Feeder ..... 276
10.1.3.1 Series Components .......................................... 276
10.1.3.2 Shunt Components .......................................... 278
10.1.4 Applying the Ladder Iterative Technique .................... 279
10.1.5 Putting It All Together ..................................................... 279
10.1.6 Load Allocation................................................................. 289
10.1.7 Summary of PowerFlow Studies .................................. 289
10.2 ShortCircuit Studies ...................................................................... 290
10.2.1 General Theory.................................................................. 290
10.2.2 Speciﬁc Short Circuits ...................................................... 293
10.3 Summary.......................................................................................... 298
References ................................................................................................... 299
Problems........................................................................................ 299
Appendix A
.......................................................................................... 303
Appendix B
.......................................................................................... 307
Index
..................................................................................................... 309
0812_frame_FM.fm Page viii Tuesday, July 31, 2001 10:49 AM
Preface
In the last 40 years many papers and textbooks devoted to the computer
modeling and analysis of large power system networks have been written.
For the most part the models and analysis techniques have been developed
for large interconnected transmission systems and synchronous generators.
Little, if any, attention was devoted to the distribution system and its major
components. As a result, the distribution engineer has not had the same
number of tools as the systems engineer to analyze the distribution system
under steadystate (powerﬂow) and fault (shortcircuit) conditions. Without
these tools the distribution engineer has been left in the dark (no pun inten
ded) as to the operating characteristics of distribution feeders. A lot of “seat
of the pants” engineering has had to take place in order to keep the lights on.
In recent years more attention has been devoted to the computer modeling
and analysis of distribution systems. Computer programs are now available
so that the distribution engineer can develop a real feel for how the distri
bution system is operating. With the tools, powerﬂow studies can be run to
simulate present loading conditions and to help with the longrange plan
ning of new facilities. The tools also provide an opportunity for the distri
bution engineer to do such things as optimize capacitor placement in order
to minimize losses. Different switching scenarios for normal and emergency
conditions can be simulated, and shortcircuit studies provide the necessary
data for the development of a reliable coordinated protection plan for fuses,
reclosers, and relay/circuit breakers. In short, the distribution engineer now
has the needed tools.
So what is the problem? “Garbage in, garbage out” is the answer. Armed
with a commercially available computer program, it is possible for a user to
prepare incorrect data that will lead to results that do not make any sense.
Without an understanding of the models and a general “feel” for the operating
characteristics of a distribution system, serious design errors and operational
procedures may result. The user must fully understand the models and
analysis techniques of the program.
Most power systems textbooks and courses are limited to the modeling
and analysis of balanced threephase systems. The models and analyses
assume a balance so that only a singlephase equivalent model is required.
While this works ﬁne for interconnected systems, it is not sufﬁcient for the
modeling and analysis of a distribution system. A distribution system is
inherently unbalanced, and therefore threephase models of all the compo
nents must be employed. There is a signiﬁcant difference between the com
puter programs developed for interconnected system studies and the
programs developed for distribution systems. The data requirements for the
0812_frame_FM.fm Page ix Tuesday, July 31, 2001 10:49 AM
distribution system models are more extensive. In fact, much of the necessary
data may not be readily available.
For many years there has been a need for a textbook to assist the student
and distribution engineer in developing a basic understanding of the mod
eling and operating characteristics of the major components of a distribution
system. With this knowledge it will be possible to prevent the “garbage in,
garbage out” scenario.
This textbook assumes that the student has a basic understanding of trans
formers, electric machines, transmission lines, and symmetrical components.
In many universities all of these topics are crammed into a onesemester
course. For that reason a quick review of the theory is presented as needed.
There are many example problems throughout the text. These examples
are intended to not only demonstrate the application of the models, but to
also teach a “feel” for what the answers should be. The example problems
should be studied very carefully. Each chapter will have a series of home
work problems that will assist the student in applying the models and
developing a better understanding of the operating characteristics of the
component being modeled. A word of warning: most of the problems are very
number intensive, intensive to the point that most of them cannot be worked
easily without using a computing tool such as Mathcad
TM
. Students are urged
to learn how to use this very powerful program. They are also encouraged to
write their own simple computer programs for many of the problems. A
summary of the intent of each chapter follows.
Chapter 1 introduces the basic components of a distribution system.
Included is an introduction to the type of data that is necessary to model a
distribution system.
Chapter 2 is a discussion of “load.” The attempt here is to make the student
understand that the load on a distribution system is constantly changing,
and that this must be taken into account in all studies.
Chapter 3 presents some helpful approximate analysis techniques that will
help the student know what “ballpark” answers to look for when more
precise studies are made.
Chapter 4 is a very important chapter in developing the exact model of
line segments. How to take into account the unbalanced loading and unsym
metrical conﬁgurations in the calculation of line impedances is presented in
great detail. Both overhead and underground lines are included.
Chapter 5 is in many ways a continuation of Chapter 4, except that it is
limited to shunt admittance calculations.
Chapter 6 develops the ﬁrst of the generalized matrices that will be used
to model the major components of a distribution system. This chapter is
limited to the threephase, unbalanced line model.
Chapter 7 addresses voltage regulation. Starting with a review of basic
transformer theory, the chapter moves to the development of threephase
models of stepvoltage regulators and their control. The models developed
are in the form of generalized matrices similar to those developed for line
segments.
0812_frame_FM.fm Page x Tuesday, July 31, 2001 10:49 AM
Chapter 8 develops comprehensive models of several of the standard
threephase transformer connections that are common on a distribution
system. The models, again, are in the form of generalized matrices.
Chapter 9 develops the models for the various types of loads on a distri
bution system.
Chapter 10 puts it all together. All of the component models developed in
earlier chapters are put together to form a model of a distribution feeder.
The ladder iterative technique is developed and demonstrated. Also, the
threephase model for shortcircuit studies is developed and demonstrated.
Two student version software packages are available. Students and pro
fessors are encouraged to acquire one or both. The packages available are
1.
Radial Distribution Analysis Package
(
RDAP
)
W. H. Power Consultants
P.O. Box 3903
Las Cruces, NM 88003
(505) 6462434
Email: wkerstin@nmsu.edu
Homepage: www.zianet.com/whpower
or www.nmsu.edu/~wkerstin/
2.
Windmil
Milsoft Integrated Solutions, Inc.
P.O. Box 7526
Abilene, TX 79608
Email: support@milsoft.com
Homepage: www.milsoft.com
0812_frame_FM.fm Page xi Tuesday, July 31, 2001 10:49 AM
The Author
William H. Kersting
received his BSEE degree from New Mexico State
University (NMSU), Las Cruces, and his MSEE degree from the Illinois
Institute of Technology. He joined the faculty at New Mexico State University
(NMSU) in 1962 and is currently Professor of Electrical Engineering and
Director of the Electric Utility Management Program. He is also a partner in
W. H. Power Consultants.
Professor Kersting is a Fellow of the Institute of Electrical and Electronics
Engineers; he received the Edison Electric Institute’s Power Engineering
Educator Award in 1979 and the NMSU Westhafer Award for Excellence
in Teaching in 1977. Prior to joining NMSU, he was employed as a distribution
engineer by the El Paso Electric Company. Professor Kersting has been an
active member of the IEEE Power Engineering Education and Power Engi
neering Committees.
0812_frame_FM.fm Page xii Tuesday, July 31, 2001 10:49 AM
Acknowledgments
I would be remiss if I didn’t acknowledge the patience that my students
have displayed over the past many years, as I have taught this material
without the aid of a textbook. The students have had to live with taking
notes and/or deciphering lastminute notes distributed in class. Their posi
tive attitudes toward the material and what I was trying to accomplish have
gone a long way toward making this text possible.
I want to thank Dr. Leonard Bohmann and his students at Michigan Tech
for reviewing and correcting the manuscript. Their suggestions were very
helpful. My thanks also to Dr. Leo Grigsby for his encouragement and review
of the manuscript.
I would like to dedicate this book to my loving wife Joanne for her encour
agement and love that has made all of this possible. She spent many lonely
evenings practicing the piano as I sat pounding out the text and/or yelling
at the computer.
0812_frame_FM.fm Page xiii Tuesday, July 31, 2001 10:49 AM
0812_frame_FM.fm Page xiv Tuesday, July 31, 2001 10:49 AM
1
1
Introduction to Distribution Systems
The major components of an electric power system are shown in Figure 1.1.
Of these components, the distribution system has traditionally been char
acterized as the most unglamorous component. In the last half of the
twentieth century the design and operation of the generation and trans
mission components presented many challenges to practicing engineers
and researchers. Power plants became larger and larger. Transmission lines
crisscrossed the land forming large interconnected networks. The opera
tion of these large interconnected networks required the development of
new analysis and operational techniques. Meanwhile, the distribution
systems continued to deliver power to the ultimate user’s meter with little
or no analysis. As a direct result, distribution systems were typically
overdesigned.
Times have changed. It has become very important and necessary to oper
ate a distribution system at its maximum capacity. Some of the questions
that need to be answered are
1. What is the maximum capacity?
2. How do we determine this capacity?
3. What are the operating limits that must be satisﬁed?
4. What can be done to operate the distribution system within the
operating limits?
5. What can be done to make the distribution system operate more
efﬁciently?
All of these questions can be answered only if the distribution system can
be modeled very accurately.
The purpose of this text is to develop accurate models for all of the major
components of a distribution system. Once the models have been developed,
analysis techniques for steadystate and shortcircuit conditions will be
developed.
0812_frame_C01.fm Page 1 Friday, July 20, 2001 2:18 PM
2
Distribution System Modeling and Analysis
1.1 The Distribution System
The distribution system typically starts with the distribution substation that
is fed by one or more subtransmission lines. In some cases the distribution
substation is fed directly from a highvoltage transmission line, in which
case there is likely no subtransmission system. This varies from company to
company. Each distribution substation will serve one or more primary feeders.
With a rare exception, the feeders are radial, which means that there is only
one path for power to ﬂow from the distribution substation to the user.
1.2 Distribution Substations
A diagram of a very simple oneline distribution substation is shown in
Figure 1.2. Although Figure 1.2 displays the simplest distribution substation,
it illustrates the major components that will be found in all substations.
FIGURE 1.1
Major power system components.
FIGURE 1.2
Simple distribution substation.
Interconnected
Transmission
System
Substation
Bulk Power
Subtransmission
Network
Distribution
Substation
Primary
Feeders
Generation
Subtransmission Line
Disconnect Switch
Fuse
Transformer
Voltage Regulator
Circuit Breakers
Primary Feeders
Meters
0812_frame_C01.fm Page 2 Friday, July 20, 2001 2:18 PM
Introduction to Distribution Systems
3
1. Highside and lowside switching: in Figure 1.2 the highvoltage
switching is done with a simple switch. More extensive substations
may use highvoltage circuit breakers in a variety of highvoltage
bus designs. The lowvoltage switching in the ﬁgure is accom
plished with relaycontrolled circuit breakers. In many cases reclos
ers will be used in place of the relay/circuit breaker combination.
Some substation designs will include a lowvoltage bus circuit
breaker in addition to the circuit breakers for each feeder. As is the
case with the highvoltage bus, the lowvoltage bus can take on a
variety of designs.
2. Voltage transformation: the primary function of a distribution sub
station is to reduce the voltage to the distribution voltage level.
In Figure 1.2 only one transformer is shown. Other substation
designs will call for two or more threephase transformers. The
substation transformers can be threephase units or three single
phase units connected in a standard connection. There are many
“standard” distribution voltage levels. Some of the common ones
are 34.5 kV, 23.9 kV, 14.4 kV, 13.2 kV, 12.47 kV, and, in older
systems, 4.16 kV.
3. Voltage regulation: as the load on the feeders varies, the voltage drop
between the substation and the user will vary. In order to maintain
the user’s voltages within an acceptable range, the voltage at the
substation needs to change as the load changes. In Figure 1.2 the
voltage is regulated by a “steptype” regulator that will alter the volt
age plus or minus 10% on the lowside bus. Sometimes this function
is accomplished with a “load tap changing” (LTC) transformer. The
LTC changes the taps on the lowvoltage windings of the trans
former as the load varies. Many substation transformers will have
“ﬁxed taps” on the highvoltage winding. These are used when
the source voltage is always either above or below the nominal
voltage. The ﬁxed tap settings can alter the voltage plus or minus
5%. Many times, instead of a bus regulator, each feeder will have
its own regulator. This can be in the form of a threephase gang
operated regulator or individual phase regulators that operate
independently.
4. Protection: the substation must be protected against the occurrence
of short circuits. In the simple design of Figure 1.2, the only auto
matic protection against short circuits inside the substation is by way
of the highside fuses on the transformer. As substation designs
become more complex, more extensive protective schemes will be
employed to protect the transformer, the high and lowvoltage
buses, and any other piece of equipment. Individual feeder circuit
breakers or reclosers are used to provide interruption of short
circuits that occur outside the substation.
0812_frame_C01.fm Page 3 Friday, July 20, 2001 2:18 PM
4
Distribution System Modeling and Analysis
5. Metering: every substation has some form of metering. This may
be as simple as an analog ammeter displaying the present value of
substation current, as well as the minimum and maximum currents
that have occurred over a speciﬁc time period. Digital recording
meters are becoming very common. These meters record the min
imum, average, and maximum values of current, voltage, power,
power factor, etc. over a speciﬁed time range. Typical time ranges
are 15 minutes, 30 minutes, and 1 hour. The digital meters may
monitor the output of each substation transformer and/or the out
put of each feeder.
A more comprehensive substation layout is shown in Figure 1.3. The sub
station in Figure 1.3 has two loadtap changing transformers, serves four
distribution feeders, and is fed from two subtransmission lines. Under nor
mal conditions the circuit breakers (CB) are in the following positions:
Circuit breakers closed: X, Y, 1,3,4,6
Circuit breakers open: Z, 2,5
With the breakers in their normal positions, each transformer is served
from a different subtransmission line and serves two feeders. Should one
of the subtransmission lines go out of service, then breaker X or Y is opened
and breaker Z is closed. Now both transformers are served from the same
subtransmission line. The transformers are sized such that each trans
former can supply all four feeders under an emergency operating condition.
For example, if Transformer T1 is out of service, then breakers X, 1, and 4
are opened and breakers 2 and 5 are closed. With that breaker arrangement,
all four feeders are served by transformer T2. The lowvoltage bus arrangement
FIGURE 1.3
Twotransformer substation with breakerandahalf scheme.
T1 T2
FD1
2
FD3
FD2 FD4
X
Z
Y
N.C.
N.O.
N.C.
3 1
N.C. N.O. N.C.
4
N.C.
6
N.O.
5
N.C.
Line 1
Line 2
0812_frame_C01.fm Page 4 Friday, July 20, 2001 2:18 PM
Introduction to Distribution Systems
5
is referred to as a “breakerandahalf scheme” since three breakers are
required to serve two feeders.
There is an unlimited number of substation conﬁgurations possible. It is
up to the substation design engineer to create a design that provides the ﬁve
basic functions and yields the most reliable service economically possible.
1.3 Radial Feeders
Radial distribution feeders are characterized by having only one path for
power to ﬂow from the source (distribution substation) to each customer. A
typical distribution system will be composed of one or more distribution
substations consisting of one or more feeders. Components of the feeder may
consist of the following:
1. Threephase primary “main” feeder
2. Threephase, twophase (“V” phase), and singlephase laterals
3. Steptype voltage regulators
4. Inline transformers
5. Shunt capacitor banks
6. Distribution transformers
7. Secondaries
8. Threephase, twophase, and singlephase loads
The loading of a distribution feeder is inherently unbalanced because of the
large number of unequal singlephase loads that must be served. An addi
tional unbalance is introduced by the nonequilateral conductor spacings of
threephase overhead and underground line segments.
Because of the nature of the distribution system, conventional powerﬂow
and shortcircuit programs used for transmission system studies are not
adequate. Such programs display poor convergence characteristics for radial
systems. The programs also assume a perfectly balanced system so that a
singlephase equivalent system is used.
If a distribution engineer is to be able to perform accurate powerﬂow and
shortcircuit studies, it is imperative that the distribution feeder be modeled
as accurately as possible. This means that threephase models of the major
components must be utilized. Threephase models for the major components
will be developed in the following chapters. They will be developed in the
“phase frame” rather than applying the method of symmetrical components.
Figure 1.4 shows a simple “oneline” diagram of a threephase feeder.
Figure 1.4 illustrates the major components of a distribution system. The
connecting points of the components will be referred to as “nodes.” Note that
0812_frame_C01.fm Page 5 Friday, July 20, 2001 2:18 PM
6
Distribution System Modeling and Analysis
the phasing of the line segments is shown. This is important if the most
accurate models are to be developed.
1.4 Distribution Feeder Map
The analysis of a distribution feeder is important to an engineer in order to
determine the existing operating conditions of a feeder, and to be able to play
the “what if” scenarios of future changes to the feeder. Before the engineer
can perform the analysis of a feeder, a detailed map of the feeder must be
available. A sample of such a map is shown in Figure 1.5. The map of Figure 1.5
contains most of the following information:
1. Lines (overhead and underground)
a. Where
b. Distances
c. Details
i. Conductor sizes (not shown on this map)
ii. Phasing
FIGURE 1.4
Simple distribution feeder.
Node
b
c
a
c
a
b
Voltage Regulator
Singlephase lateral
"V" phase lateral
Threephase lateral
Underground cables
a
b
c
c
b
a b c
Fuse
Distribution
transformer
Secondary
Customers
Inline transformer
Capacitor bank
Transformer
Substation
b
0812_frame_C01.fm Page 6 Friday, July 20, 2001 2:18 PM
Introduction to Distribution Systems
7
FIGURE 1.5
123node test feeder.
25
25
25
25
25
25
25
25
25
25
25
25
25
50
50
50
50
50
50
50
50
50
50
50
25
50
50
50
50
50
25
50
25
50
25
50
25
50
50
50
50
50
50
50
50
50
50
50
50
25
50
50
50
25
25
25
25
25
25
25
50
50
100
100
M
50
50
abc
b
400'
250'
325'
250'
200'
300'
300'
25
200'
225'
50
250'
250'
c
a
b
b
a
225'
425'
175'
150'
100'
375'
c
200'
50
400'
b
b
350'
250'
250'
b
a
c
b
825'
b
250'
325'
300'
b
525'
250'
c
550'
275'
350'
275'
500'
a
25
300'
225'
c
50
50
200'
350'
c
b
a
3100
350
250'
150'
350
250'
200'
250'
250'
300'
200'
a
25
500'
b
325'
650'
375'
b
250'
325'
a
300'
250'
1
0
0
0
'
450'
50
300'
575'
525'
325'
a
b
225'
575'
25
c
225'
325'
700'
275'
b c a
550'
300'
800'
a
200'
275'
325'
275'
3
2
5
'
2
7
5
'
250'
250'
275'
200'
50
400'
350'
275'
c
100'
225'
a
c
b
475'
175'
c
475'
675'
250'
250'
a
c
b
a
b
700'
25
50
450'
b
b
a
225'
200'
b
300'
650'
275'
a
c
300'
225'
b 250'
275'
175'
275'
b
b
275'
a
c
b
200'
300'
c a
350'
3100
400'
a c b
c
a
250'
c b a
c b a
800'
c
c
b
a
350
325'
25
350'
175'
250'
425'
b a
Substation
ThreePhase OH
ThreePhase UG
TwoPhase OH
OnePhase OH
b c a
c
a
b
350
a
b
c
a
c
b
a c b
125'
350'
750'
550'
1Phase Transformer kVA 50
3Phase Transformer Bank
350
Voltage Regulator
0812_frame_C01.fm Page 7 Friday, July 20, 2001 2:18 PM
8
Distribution System Modeling and Analysis
2. Distribution transformers
a. Location
b. kVA rating
c. Phase connection
3. Inline transformers
a. Location
b. kVA rating
c. Connection
4. Shunt capacitors
a. Location
b. kvar rating
c. Phase connection
5. Voltage regulators
a. Location
b. Phase connection
c. Type (not shown on this map)
i. Singlephase
ii. Threephase
6. Switches
a. Location
b. Normal open/close status
1.5 Distribution Feeder Electrical Characteristics
Information from the map will deﬁne the physical location of the various
devices. Electrical characteristics for each device will have to be determined
before the analysis of the feeder can commence. In order to determine the
electrical characteristics, the following data must be available:
1. Overhead and underground spacings
2. Conductor tables
a. Geometric mean radius (GMR) (ft.)
b. Diameter (inches)
c. Resistance (
Ω
/mile)
3. Voltage regulators
a. Potential transformer ratios
b. Current transformer ratios
0812_frame_C01.fm Page 8 Friday, July 20, 2001 2:18 PM
Introduction to Distribution Systems
9
c. Compensator settings
i. Voltage level
ii. Bandwidth
iii. R and X settings, in volts (V)
4. Transformers
a. kVA rating
b. Voltage ratings
c. Impedance (R and X)
d. Noload power loss
1.6 Summary
It is becoming increasingly more important to be able to accurately model
and analyze distribution systems. There are many different substation
designs possible, but, for the most part, the substation serves one or more
radial feeders. Each feeder must be modeled as accurately as possible in order
for the analysis to have meaning. Sometimes the most difﬁcult task for the
engineer is to acquire all of the necessary data. Feeder maps will contain
most of the needed data. Additional data such as standard pole conﬁgura
tions, speciﬁc conductors used on each line segment, threephase transformer
connections, and voltage regulator settings must come from stored records.
Once all of the data has been acquired, the analysis can begin utilizing models
of the various devices that will be developed in later chapters.
0812_frame_C01.fm Page 9 Friday, July 20, 2001 2:18 PM
0812_frame_C01.fm Page 10 Friday, July 20, 2001 2:18 PM
11
2
The Nature of Loads
The modeling and analysis of a power system depend upon the load. What
is load? The answer to that question depends upon what type of an analysis
is desired. For example, the steadystate analysis (powerﬂow study) of an
interconnected transmission system will require a different deﬁnition of
load than that used in the analysis of a secondary in a distribution feeder.
The problem is that the load on a power system is constantly changing. The
closer you are to the customer, the more pronounced will be the ever
changing load. There is no such thing as a “steadystate” load. In order to come
to grips with load, it is ﬁrst necessary to look at the load of an individual
customer.
2.1 Deﬁnitions
The load that an individual customer or a group of customers presents to
the distribution system is constantly changing. Every time a light bulb or an
electrical appliance is switched on or off, the load seen by the distribution
feeder changes. In order to describe the changing load, the following terms
are deﬁned:
1. Demand
• Load averaged over a speciﬁc period of time
• Load can be kW, kvar, kVA, or A
• Must include the time interval
• Example: the 15minute kW demand is 100 kW
2. Maximum Demand
• Greatest of all demands that occur during a speciﬁc time
• Must include demand interval, period, and units
• Example: the 15minute maximum kW demand for the week
was 150 kW
0812_Frame_C02.fm Page 11 Saturday, July 21, 2001 2:53 PM
12
Distribution System Modeling and Analysis
3. Average Demand
• The average of the demands over a speciﬁed period (day, week,
month, etc.)
• Must include demand interval, period, and units
• Example: the 15minute average kW demand for the month was
350 kW
4. Diversiﬁed Demand
• Sum of demands imposed by a group of loads over a particular
period
• Must include demand interval, period, and units
• Example: the 15minute diversiﬁed kW demand in the period
ending at 9:30 was 200 kW
5. Maximum Diversiﬁed Demand
• Maximum of the sum of the demands imposed by a group of
loads over a particular period
• Must include demand interval, period, and units
• Example: the 15minute maximum diversiﬁed kW demand for
the week was 500 kW
6. Maximum Noncoincident Demand
• For a group of loads, the sum of the individual maximum de
mands without any restriction that they occur at the same time
• Must include demand interval, period, and units
• Example: the maximum noncoincident 15minute kW demand
for the week was 700 kW
7. Demand Factor
• Ratio of maximum demand to connected load
8. Utilization Factor
• Ratio of the maximum demand to rated capacity
9. Load Factor
• Ratio of the average demand of any individual customer or
group of customers over a period to the maximum demand over
the same period
10. Diversity Factor
• Ratio of the maximum noncoincident demand to the maximum
diversiﬁed demand
11. Load Diversity
• Difference between maximum noncoincident demand and the
maximum diversiﬁed demand
0812_Frame_C02.fm Page 12 Saturday, July 21, 2001 2:53 PM
The Nature of Loads
13
2.2 Individual Customer Load
Figure 2.1 illustrates how the instantaneous kW load of a customer changes
during two 15minute intervals.
2.2.1 Demand
In order to deﬁne the load, the demand curve is broken into equal time
intervals. In Figure 2.1 the selected time interval is 15 minutes. In each interval
the average value of the demand is determined. In Figure 2.1 the straight
lines represent the average load in a time interval. The shorter the time
interval, the more accurate will be the value of the load. This process is very
similar to numerical integration. The average value of the load in an interval
is deﬁned as the
15minute kW demand
.
The 24hour 15minute kW demand curve for a customer is shown in
Figure 2.2. This curve is developed from a spreadsheet that gives the 15minute
kW demand for a period of 24 hours.
2.2.2 Maximum Demand
The demand curve shown in Figure 2.2 represents a typical residential cus
tomer. Each bar depicts the 15minute kW demand. Note that during the
24hour period there is a great variation in the demand. This particular
customer has three periods in which the kW demand exceeds 6.0 kW. The
greatest of these is the
15minute maximum kW demand
. For this customer
the 15minute maximum kW demand occurs at 13:15 and has a value of
6.18 kW.
FIGURE 2.1
Customer demand curve.
6:15 6:30
1.0
2.0
3.0
4.0
5.0
6.0
1
5
M
i
n
u
t
e
k
W
D
e
m
a
n
d
6:45
Time of Day
Instantaneous
0812_Frame_C02.fm Page 13 Saturday, July 21, 2001 2:53 PM
14
Distribution System Modeling and Analysis
2.2.3 Average Demand
During the 24hour period, energy (kWh) will be consumed. The energy in
kWh used during each 15minute time interval is computed by:
(2.1)
The total energy consumed during the day is the summation of all of the
15minute interval consumptions. From the spreadsheet, the total energy
consumed during the period by Customer #1 is 58.96 kWh. The
15minute
average kW demand
is computed by:
(2.2)
2.2.4 Load Factor
“Load factor” is a term that is often used when describing a load. It is deﬁned
as the ratio of the average demand to the maximum demand. In many ways
load factor gives an indication of how well the utility’s facilities are being
utilized. From the utility’s standpoint, the optimal load factor would be 1.00
since the system has to be designed to handle the maximum demand. Some
times utility companies will encourage industrial customers to improve their
load factors. One method of encouragement is to penalize the customer on
the electric bill for having a low load factor.
For Customer #1 in Figure 2.2 the load factor is computed to be
(2.3)
FIGURE 2.2
24hour demand curve for Customer #1.
kWh 15min kW demand ( ) =
1
4
 hour ⋅
Average demand
Total energy
Hours

58.96
24
 2.46 kW = = =
Load factor
Average 15min kW demand
Max. 15min kW demand

2.46
6.18
 0.40 = = =
0812_Frame_C02.fm Page 14 Saturday, July 21, 2001 2:53 PM
The Nature of Loads
15
2.3 Distribution Transformer Loading
A distribution transformer will provide service to one or more customers.
Each customer will have a demand curve similar to that in Figure 2.2.
However, the peaks and valleys and maximum demands will be different
for each customer. Figures 2.3, 2.4, and 2.5 give the demand curves for the
three additional customers connected to the same distribution transformer.
The load curves for the four customers show that each customer has his
unique loading characteristic. The customers’ individual maximum kW
demand occurs at different times of the day. Customer #3 is the only one
who will have a high load factor. A summary of individual loads is given
in Table 2.1. These four customers demonstrate that there is great diversity
among their loads.
FIGURE 2.3
24hour demand curve for Customer #2.
FIGURE 2.4
24hour demand curve for Customer #3.
0812_Frame_C02.fm Page 15 Saturday, July 21, 2001 2:53 PM
16
Distribution System Modeling and Analysis
2.3.1 Diversiﬁed Demand
It is assumed that the same distribution transformer serves the four custom
ers discussed previously. The sum of the four 15 kW demands for each time
interval is the
diversiﬁed demand
for the group in that time interval, and, in
this case, the distribution transformer. The 15minute diversiﬁed kW demand
of the transformer for the day is shown in Figure 2.6. Note how the demand
curve is beginning to smooth out. There are not as many signiﬁcant changes
as in some of the individual customer curves.
FIGURE 2.5
24hour demand curve for Customer #4.
TABLE 2.1
Individual Customer Load Characteristics
Cust. #1 Cust. #2 Cust. #3 Cust. #4
Energy Usage (kWh) 58.57 36.46 95.64 42.75
Maximum kW Demand 6.18 6.82 4.93 7.05
Time of Max. kW Demand 13:15 11:30 6:45 20:30
Average kW Demand 2.44 1.52 3.98 1.78
Load Factor 0.40 0.22 0.81 0.25
FIGURE 2.6
Transformer diversiﬁed demand curve.
0812_Frame_C02.fm Page 16 Saturday, July 21, 2001 2:53 PM
The Nature of Loads
17
2.3.2 Maximum Diversiﬁed Demand
The transformer demand curve of Figure 2.6 demonstrates how the com
bined customer loads begin to smooth out the extreme changes of the indi
vidual loads. For the transformer, the 15minute kW demand exceeds 16 kW
twice. The greater of these is the
15minute maximum diversiﬁed kW demand
of the transformer. It occurs at 17:30 and has a value of 16.16 kW. Note that
this maximum demand does not occur at the same time as any one of the
individual demands, nor is this maximum demand the sum of the individual
maximum demands.
2.3.3 Load Duration Curve
A
load duration curve
can be developed for the transformer serving the four
customers. Sorting in descending order, the kW demand of the transformer
develops the load duration curve shown in Figure 2.7. The load duration
curve plots the 15minute kW demand versus the percent of time the
transformer operates at or above the speciﬁc kW demand. For example,
the load duration curve shows the transformer operates with a 15minute
kW demand of 12 kW or greater 22% of the time. This curve can be used
to determine whether a transformer needs to be replaced due to an over
loading condition.
2.3.4 Maximum Noncoincident Demand
The
15minute maximum noncoincident kW demand
for the day is the sum of
the individual customer 15minute maximum kW demands. For the trans
former in question, the sum of the individual maximums is
Max. noncoincident demand
=
6.18
+
6.82
+
4.93
+ 7.05 =
24.98 kW (2.4)
FIGURE 2.7
Transformer load duration curve.
0812_Frame_C02.fm Page 17 Saturday, July 21, 2001 2:53 PM
18
Distribution System Modeling and Analysis
2.3.5 Diversity Factor
By deﬁnition,
diversity factor
is the ratio of the maximum noncoincident
demand of a group of customers to the maximum diversiﬁed demand of the
group. With reference to the transformer serving four customers, the diver
sity factor for the four customers would be
(2.5)
The
idea behind the diversity
factor is that when the maximum demands
of the customers are known, then the maximum diversiﬁed demand of a
group of customers can be computed. There will be a different value of the
diversity factor for different numbers of customers. The value computed
above would apply for four customers. If there are ﬁve customers, then a
load survey would have to be set up to determine the diversity factor for
ﬁve customers. This process would have to be repeated for all practical
numbers of customers. Table 2.2 is an example of the diversity factors for
the number of customers ranging from one to 70. The table was developed
from a different database than the four customers discussed previously. A
graph of the diversity factors is shown in Figure 2.8. Note in Table 2.2 and
Figure 2.8 that the value of the diversity factor basically leveled out when
the number of customers reached 70. This is an important observation because
it means, at least for the system from which these diversity factors were
determined, that the diversity factor will remain constant at 3.20 from 70
customers up. In other words, as viewed from the substation, the maximum
diversiﬁed demand of a feeder can be predicted by computing the total
noncoincident maximum demand of all of the customers served by the
feeder and dividing by 3.2.
TABLE 2.2
Diversity Factors
N DF N DF N DF N DF N DF N DF N DF
1 1.0 11 2.67 21 2.90 31 3.05 41 3.13 51 3.15 61 3.18
2 1.60 12 2.70 22 2.92 32 3.06 42 3.13 52 3.15 62 3.18
3 1.80 13 2.74 23 2.94 33 3.08 43 3.14 53 3.16 63 3.18
4 2.10 14 2.78 24 2.96 34 3.09 44 3.14 54 3.16 64 3.19
5 2.20 15 2.80 25 2.98 35 3.10 45 3.14 55 3.16 65 3.19
6 2.30 16 2.82 26 3.00 36 3.10 46 3.14 56 3.17 66 3.19
7 2.40 17 2.84 27 3.01 37 3.11 47 3.15 57 3.17 67 3.19
8 2.55 18 2.86 28 3.02 38 3.12 48 3.15 58 3.17 68 3.19
9 2.60 19 2.88 29 3.04 39 3.12 49 3.15 59 3.18 69 3.20
10 2.65 20 2.90 30 3.05 40 3.13 50 3.15 60 3.18 70 3.20
Diversity factor
Maximum noncoincident demand
Maximum diversified demand

24.98
16.16
 1.5458 = = =
0812_Frame_C02.fm Page 18 Saturday, July 21, 2001 2:53 PM
The Nature of Loads
19
2.3.6 Demand Factor
The demand factor can be deﬁned for an individual customer. For example,
the 15minute maximum kW demand of Customer #1 was found to be
6.18 kW. In order to determine the demand factor, the total connected load
of the customer needs to be known. The total connected load will be the sum
of the ratings of all of the electrical devices at the customer’s location.
Assume that this total comes to 35 kW; then, the demand factor is computed
to be
(2.6)
The demand factor gives an indication of the percentage of electrical devices
that are on when the maximum demand occurs. The demand factor can be
computed for an individual customer but not for a distribution transformer
or the total feeder.
2.3.7 Utilization Factor
The utilization factor gives an indication of how well the capacity of an
electrical device is being utilized. For example, the transformer serving the
four loads is rated 15 kVA. Using the 16.16kW maximum diversiﬁed demand
and assuming a power factor of 0.9, the 15minute maximum kVA demand
on the transformer is computed by dividing the 16.16kW maximum kW
demand by the power factor, and would be 17.96 kVA. The utilization factor
is computed to be
(2.7)
FIGURE 2.8
Diversity factors.
Demand factor
Maximum demand
Total connected load

6.18
35
 0.1766 = = =
Utilization factor
Maximum kVA demand
Transformer kVA rating

17.96
15
 1.197 = = =
0812_Frame_C02.fm Page 19 Saturday, July 21, 2001 2:53 PM
20
Distribution System Modeling and Analysis
2.3.8 Load Diversity
Load diversity is deﬁned as the difference between the noncoincident max
imum demand and the maximum diversiﬁed demand. For the transformer
in question, the load diversity is computed to be
Load diversity
=
24.97
−
16.16
=
8.81 kW (2.8)
2.4 Feeder Load
The load that a feeder serves will display a smoothedout demand curve
as shown in Figure 2.9. The feeder demand curve does not display any of
the abrupt changes in demand of an individual customer demand curve or
the semiabrupt changes in the demand curve of a transformer. The simple
explanation for this is that with several hundred customers served by the
feeder, the odds are good that as one customer is turning off a light bulb
another customer will be turning a light bulb on. The feeder load therefore
does not experience a jump as would be seen in the individual customer’s
demand curve.
2.4.1 Load Allocation
In the analysis of a distribution feeder load, data will have to be speciﬁed.
The data provided will depend upon how detailed the feeder is to be mod
eled, and the availability of customer load data. The most comprehensive
model of a feeder will represent every distribution transformer. When this
is the case, the load allocated to
each transformer needs to be determined.
FIGURE 2.9
Feeder demand curve.
0812_Frame_C02.fm Page 20 Saturday, July 21, 2001 2:53 PM
The Nature of Loads
21
2.4.1.1 Application of Diversity Factors
The deﬁnition of the diversity factor (DF) is the ratio of the maximum
noncoincident demand to the maximum diversiﬁed demand. Diversity fac
tors are shown in Table 2.2. When such a table is available, then it is possible
to determine the maximum diversiﬁed demand of a group of customers such
as those served by a distribution transformer; that is, the maximum diver
siﬁed demand can be computed by:
(2.9)
This maximum diversiﬁed demand becomes the allocated load for the
transformer.
2.4.1.2 Load Survey
Many times the maximum demand of individual customers will be known,
either from metering or from a knowledge of the energy (kWh) consumed
by the customer. Some utility companies will perform a load survey of similar
customers in order to determine the relationship between the energy con
sumption in kWh and the maximum kW demand. Such a load survey requires
the installation of a demand meter at each customer’s location. The meter
can be the same type used to develop the demand curves previously discussed,
or it can be a simple meter that only records the maximum demand during
the period. At the end of the survey period the maximum demand vs. kWh
for each customer can be plotted on a common graph. Linear regression is
used to determine the equation of a straight line that gives the kW demand
as a function of kWh. The plot of points for 15 customers, along with the
resulting equation derived from a linear regression algorithm, is shown in
Figure 2.10. The straightline equation derived is
Max. kW demand
=
0.1058
+
0.005014
⋅
kWh (2.10)
Knowing the maximum demand for each customer is the ﬁrst step in devel
oping a table of diversity factors as shown in Table 2.2. The next step is to
perform a load survey where the maximum diversiﬁed demand of groups of
customers is metered. This will involve selecting a series of locations where
demand meters can be placed that will record the maximum demand for
groups of customers ranging from at least 2 to 70. At each meter location
the maximum demand of all downstream customers must also be known.
With that data, the diversity factor can be computed for the given number
of downstream customers.
Example 2.1
A singlephase lateral provides service to three distribution transformers as
shown in Figure 2.11. The energy in kWh consumed by each customer during
Max. diversified demand
Max noncoincident demand
DF
n
 =
0812_Frame_C02.fm Page 21 Saturday, July 21, 2001 2:53 PM
22
Distribution System Modeling and Analysis
a month is known. A load survey has been conducted for customers in this
class, and it has been found that the customer 15minute maximum kW
demand is given by the equation:
kW
demand
=
0.2
+
0.008
⋅
kWh
The kWh consumed by Customer #1 is 1523 kWh. The 15minute maximum
kW demand for Customer #1 is then computed as:
kW
1
=
0.2
+
0.008
⋅
1523
=
12.4
FIGURE 2.10
kW demand vs. kWh for residential customers.
FIGURE 2.11
Singlephase lateral.
400 600 800 1000 1200 1400 1600 1800 2000
12
10
8
6
4
2
0
Energy (kWh)
1
5

M
i
n
u
t
e
M
a
x
i
m
u
m
k
W
D
e
m
a
n
d
(
k
W
)
kW
i
kWl
i
kWh
i
T1 T2 T3
N1 N2 N3
N4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
0812_Frame_C02.fm Page 22 Saturday, July 21, 2001 2:53 PM
The Nature of Loads
23
The results of this calculation for the remainder of the customers is summa
rized below by transformer.
1. Determine for each transformer the 15minute noncoincident max
imum kW demand and, using the Table of Diversity Factors in
Table 2.2, determine the 15minute maximum diversiﬁed kW
demand.
T1
: Noncoin. max.
=
12.4
+
13.4
+
16.1
+
12.9
+
11.9
=
66.7 kW
T2
: Noncoin. max.
=
12.9
+
13.8
+
14.2
+
16.3
+
14.3
+
17.0
=
81.6 kW
T3: Noncoin. max. = 17.0 + 15.1 + 16.7 + 18.3 + 17.3 + 16.1 + 17.0
= 117.5 kW
Based upon the 15minute maximum kW diversiﬁed demand
on each transformer and an assumed power factor of 0.9,
TRANSFORMER T1
Customer #1 #2 #3 #4 #5
kWh 1523 1645 1984 1590 1456
kW 12.4 13.4 16.1 12.9 11.9
TRANSFORMER T2
Customer #6 #7 #8 #9
kWh 1235 1587 1698 1745 2015 1765
kW 10.1 12.9 13.8 14.2 16.3 14.3
TRANSFORMER T3
Customer #12
kWh 2098 1856 2058 2265 2135 1985 2103
kW 17.0 15.1 16.7 18.3 17.3 16.1 17.0
#10 #11
#13 #14 #15 #16 #17 #18
Max. div. demand
Noncoincident max.
Diversity factor for 5

66.7
2.20
 30.3 kW = = =
Max. div. demand
Noncoincident max.
Diversity factor for 6

81.6
2.30
 35.5 kW = = =
Max. div. demand
Noncoincident max.
Diversity factor for 7

117.5
2.40
 48.9 kW = = =
0812_Frame_C02.fm Page 23 Saturday, July 21, 2001 2:53 PM
24 Distribution System Modeling and Analysis
the 15minute maximum kVA diversiﬁed demand on each
transformer would be
The kVA ratings selected for the three transformers would be
25 kVA, 37.5 kV, and 50 kVA, respectively. With those selec
tions, only transformer T1 would experience a signiﬁcant
maximum kVA demand greater than its rating (135%).
2. Determine the 15minute noncoincident maximum kW demand
and 15minute maximum diversiﬁed kW demand for each of the
line segments.
Segment N1 to N2:
The maximum noncoincident kW demand is the sum of the max
imum demands of all 18 customers.
Noncoin. max. demand = 66.7 + 81.6 + 117.5 = 265.5 kW
The maximum diversiﬁed kW demand is the computed by using
the diversity factor for 18 customers.
Max. div. demand = = 92.8 kW
Segment N2 to N3:
This line segment “sees” 13 customers. The noncoincident max
imum demand is the sum of customers number 6 through
18. The diversity factor for 13 (2.74) is used to compute the
maximum diversiﬁed kW demand.
Segment N3 to N4:
This line segment sees the same noncoincident demand and
diversiﬁed demand as that of transformer T3.
Max. kVA
T1
demand
30.3
.9
 33.6 = =
Max. kVA
T2
demand
35.5
.9
 39.4 = =
Max. kVA
T3
demand
49.0
.9
 54.4 = =
265.8
2.86

Noncoin. demand 81.6 117.5 + 199.0 kW = =
Max. div. demand
199.1
2.74
 72.6 kW = =
Noncoin. demand 117.4 kW =
Max. div. demand 48.9 kW =
0812_Frame_C02.fm Page 24 Saturday, July 21, 2001 2:53 PM
The Nature of Loads 25
Example 2.1 demonstrates that Kirchhoff’s current law (KCL) is not obeyed
when the maximum diversiﬁed demands are used as the load ﬂowing through
the line segments and through the transformers. For example, at node N1
the maximum diversiﬁed demand ﬂowing down the line segment N1N2 is
92.8 kW, and the maximum diversiﬁed demand ﬂowing through transformer
T1 is 30.3 kW. KCL would then predict that the maximum diversiﬁed demand
ﬂowing down line segment N2N3 would be the difference of these, or 62.5 kW.
However, the calculations for the maximum diversiﬁed demand in that seg
ment was computed to be 72.6 kW. The explanation is that the maximum
diversiﬁed demands for the line segments and transformers don’t necessarily
occur at the same time. At the time that line segment N2N3 is experiencing
its maximum diversiﬁed demand, line segment N1N2 and transformer T1 are
not at their maximum values. All that can be said is that, at the time segment
N2N3 is experiencing its maximum diversiﬁed demand, the difference between
the actual demand on line segment N1N2 and the demand of transformer
T1 will be 72.6 kW. There will be an inﬁnite amount of combinations of line
ﬂow down N1N2 and through transformer T1 that will produce the maximum
diversiﬁed demand of 72.6 kW on line N2N3.
2.4.1.3 Transformer Load Management
A transformer load management program is used by utilities to determine
the loading on distribution transformers based upon a knowledge of the
kWh supplied by the transformer during a peak loading month. The program
is primarily used to determine when a distribution transformer needs to be
changed out due to a projected overloading condition. The results of the
program can also be used to allocate loads to transformers for feeder analysis
purposes.
The transformer load management program relates the maximum diver
siﬁed demand of a distribution transformer to the total kWh supplied by
the transformer during a speciﬁc month. The usual relationship is the equa
tion of a straight line. Such an equation is determined from a load survey.
This type of load survey meters the maximum demand on the transformer
in addition to the total energy in kWh of all of the customers connected to
the transformer. With the information available from several sample trans
formers, a curve similar to that shown in Figure 2.10 can be developed, and
the constants of the straightline equation can be computed. This method
has an advantage because the utility will have in the billing database the
kWh consumed by each customer every month. As long as the utility knows
which customers are connected to each transformer by using the developed
equation, the maximum diversiﬁed demand (allocated load) on each trans
former on a feeder can be determined for each billing period.
2.4.1.4 Metered Feeder Maximum Demand
The major disadvantage of allocating load using the diversity factors is that
most utilities will not have a table of diversity factors. The process of devel
oping such a table is generally not cost effective. The major disadvantage of
0812_Frame_C02.fm Page 25 Saturday, July 21, 2001 2:53 PM
26 Distribution System Modeling and Analysis
the transformer load management method is that a database is required that
speciﬁes which transformers serve which customers. Again, this database is
not always available.
Allocating load based upon the metered readings in the substation requires
the least amount of data. Most feeders will have metering in the substation
that will, at minimum, give either the total threephase maximum diversiﬁed
kW or kVA demand and/or the maximum current per phase during a month.
The kVA ratings of all distribution transformers is always known for a feeder.
The metered readings can be allocated to each transformer based upon the
transformer rating. An “allocation factor” (AF) can be determined based
upon the metered threephase kW or kVA demand and the total connected
distribution transformer kVA.
(2.12)
where Metered demand can be either kW or kVA, and
kVA
total
= sum of the kVA ratings of all distribution transformers
The allocated load per transformer is then determined by:
Transformer demand = AF ⋅ kVA
transformer
(2.13)
The transformer demand will be either kW or kVA depending upon the
metered quantity.
When the kW or kVA is metered by phase, then the load can be allocated
by phase where it will be necessary to know the phasing of each distribution
transformer. When the maximum current per phase is metered, the load
allocated to each distribution transformer can be done by assuming nominal
voltage at the substation and then computing the resulting kVA. The load
allocation will now follow the same procedure as outlined above. If there is
no metered information on the reactive power or power factor of the feeder,
a power factor will have to be assumed for each transformer load.
Modern substations will have microprocessorbased metering that will
provide kW, kvar, kVA, power factor, and current per phase. With this data,
the reactive power can also be allocated. Since the metered data at the
substation will include losses, an iterative process will have to be followed
so that the allocated load plus losses will equal the metered readings.
Example 2.2
Assume that the metered maximum diversiﬁed kW demand for the system
of Example 2.1 is 92.9 kW. Allocate this load according to the kVA ratings
of the three transformers.
AF
Metered demand
kVA
total
 =
kVA
total
25 37.5 50 + + 112.5 = =
AF
92.8
112.5
 0.8249 kW/kVA = =
0812_Frame_C02.fm Page 26 Saturday, July 21, 2001 2:53 PM
The Nature of Loads 27
The allocated kW for each transformer becomes:
2.4.1.5 What Method to Use?
Four different methods have been presented for allocating load to distribu
tion transformers:
• Application of diversity factors
• Load survey
• Transformer load management
• Metered feeder maximum demand
Which method to use depends upon the purpose of the analysis. If the pur
pose is to determine as closely as possible the maximum demand on a dis
tribution transformer, then either the diversity factor or the transformer load
management method can be used. Neither of these methods should be
employed when the analysis of the total feeder is to be performed. The
problem is that using those methods will result in a much larger maximum
diversiﬁed demand at the substation than actually exists. When the total
feeder is to be analyzed, the only method that gives good results is that of
allocating load based upon the kVA ratings of the transformers.
2.4.2 VoltageDrop Calculations Using Allocated Loads
The voltage drops down line segments and through distribution transformers
are of interest to the distribution engineer. Four different methods of allocat
ing loads have been presented. The various voltage drops will be computed
using the loads allocated by two of the methods in the following examples.
For these studies it is assumed that the allocated loads will be modeled as
constant real power and reactive power.
2.4.2.1 Application of Diversity Factors
The loads allocated to a line segment or a distribution transformer using
diversity factors are a function of the total number of customers downstream
from the line segment or distribution transformer. The application of the
diversity factors was demonstrated in Example 2.1. With a knowledge of the
allocated loads ﬂowing in the line segments and through the transformers
and the impedances, the voltage drops can be computed. The assumption
is that the allocated loads will be constant real power and reactive power.
T1: kW
1
0.8249 25 ⋅ 20.62 kW = =
T2: kW
1
0.8249 37.5 ⋅ 30.93 kW = =
T3: kW
1
0.8249 50 ⋅ 41.24 kW = =
0812_Frame_C02.fm Page 27 Saturday, July 21, 2001 2:53 PM
28 Distribution System Modeling and Analysis
In order to avoid an iterative solution, the voltage at the source is assumed
and the voltage drops calculated from that point to the last transformer.
Example 2.3 demonstrates how the method of load allocation using diversity
factors is applied. The same system and allocated loads from Example 2.1
are used.
Example 2.3
For the system of Example 2.1, assume the voltage at N1 is 2400 volts and
compute the secondary voltages on the three transformers using the diversity
factors. The system of Example 2.1, including segment distances, is shown
in Figure 2.12.
Assume that the power factor of the loads is 0.9 lagging.
The impedance of the lines are: z = 0.3 + j0.6 Ω/mile
The ratings of the transformers are
From Example 2.1 the maximum diversiﬁed kW demands were computed.
Using the 0.9 lagging power factor, the maximum diversiﬁed kW and kVA
demands for the line segments and transformers are
Segment N1N2: P
12
= 92.8 kW S
12
= 92.8 + j45.0 kVA
Segment N2N3: P
23
= 72.6 kW S
23
= 72.6 + j35.2 kVA
Segment N3N4: P
34
= 48.9 kW S
34
= 48.9 + j23.7 kVA
Transformer T1: P
T1
= 30.3 kW S
T1
= 30.3 + j14.7 kVA
Transformer T2: P
T2
= 35.5 kW S
T2
= 35.5 + j17.2 kVA
Transformer T3: P
T3
= 48.9 kW S
T3
= 48.9 + j23.7 kVA
FIGURE 2.12
Singlephase lateral with distances.
T1 T2 T3
N1 N2 N3
N4
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
750' 500' 5000'
T1: 25 kVA, 2400240 volts, Z 1.8/40% =
T2: 37.5 kVA, 2400240 volts, Z 1.9/45% =
T3: 50 kVA, 2400240 volts, Z 2.0/50% =
0812_Frame_C02.fm Page 28 Saturday, July 21, 2001 2:53 PM
The Nature of Loads 29
Convert transformer impedances to ohms referred to the highvoltage side
Compute the line impedances:
Calculate the current ﬂowing in segment N1N2:
Calculate the voltage at N2:
Calculate the current ﬂowing into T1:
Calculate the secondary voltage referred to the high side:
T1: Z
base
kV
2
1000 ⋅
kVA

2.4
2
1000 ⋅
25
 230.4 Ω = = =
Z
T1
0.018/40 ( ) 230.4 ⋅ 3.18 j2.67 Ω + = =
T2: Z
base
kV
2
1000 ⋅
kVA

2.4
2
1000 ⋅
37.5
 153.6 Ω = = =
Z
T2
0.019/45 ( ) 153.6 ⋅ 2.06 j2.06 Ω + = =
T3: Z
base
kV
2
1000 ⋅
kVA

2.4
2
1000 ⋅
50
 115.2 Ω = = =
Z
T3
0.02/50 ( ) 115.2 ⋅ 1.48 j1.77 Ω + = =
N1N2: Z
12
0.3 j0.6 + ( )
5000
5280
 ⋅ 0.2841 j0.5682 Ω + = =
N2N3: Z
23
0.3 j0.6 + ( )
500
5280
 ⋅ 0.0284 j0.0568 Ω + = =
N3N4: Z
34
0.3 j0.6 + ( )
750
5280
 ⋅ 0.0426 j0.0852 Ω + = =
I
12
kW
jk var
+
kV

∗
92.9
j
45.0 +
2.4/0

∗
43.0/ 25.84 – A = = =
V
2
V
1
Z
12
I
12
⋅ – =
V
2
2400/0 0.2841 j0.5682 + ( ) 43.0/ 25.84 – ⋅ – 2378.4/ 0.4 – V = =
I
T1
kW
jk
var +
kV

∗
30.3 j14.7 +
2.378/ 0.4 –

∗
14.16/ 26.24 – A = = =
V
T1
V
2
Z
T2
I
T1
⋅ – =
V
T1
2378.4/ 0.4 – 3.18 j2.67 + ( ) 14.16/ 26.24 – ⋅ – 2321.5/ 0.8 – V = =
0812_Frame_C02.fm Page 29 Saturday, July 21, 2001 2:53 PM
30
Distribution System Modeling and Analysis
Compute the secondary voltage by dividing by the turns ratio of 10:
Calculate the current ﬂowing in line section N2N3:
Calculate the voltage at N3:
Calculate the current ﬂowing into T2:
Calculate the secondary voltage referred to the high side:
Compute the secondary voltage by dividing by the turns ratio of 10:
Calculate the current ﬂowing in line section N3N4:
Calculate the voltage at N4:
Vlow
T1
2321.5/ 0.8 –
10
 232.15/ 0.8 – V = =
I
23
kW jk var +
kV

∗
72.6 j35.2 +
2.378/ 0.4 –

∗
33.9/ 26.24 – A = = =
V
3
V
2
Z
23
I
23
⋅ – =
V
3
2378.4/ 0.4 – 0.0284 j0.0568 + ( ) 33.9/ 26.24 – ⋅ – 2376.7/ 0.4 – V = =
I
T2
kW jk var +
kV

∗
35.5 j17.2 +
2.3767/ 0.4 –

∗
16.58/ 26.27 – = = =
V
T2
V
3
Z
T2
I
T2
⋅ – =
V
T2
2376.7/ 0.4 – 2.06 j2.06 + ( ) 16.58/ 26.27 – ⋅ – 2331/1/ 0.8 – V = =
Vlow
T2
2331.1/ 0.8 –
10
 233.1/ 0.8 – V = =
I
34
kW jk var +
kV

∗
49.0 j23.7 +
2.3767/ 0.4 –

∗
22.9/ 26.27 – A = = =
V
4
V
3
Z
34
I
34
⋅ – =
V
4
2376.7/ 0.4 – 0.0426 0.0852 + ( ) 22.9/ 26.27 – ⋅ – 2375.0/ 0.5 – V = =
0812_Frame_C02.fm Page 30 Monday, October 28, 2002 11:11 AM
The Nature of Loads 31
The current ﬂowing into T3 is the same as the current from N3 to N4:
Calculate the secondary voltage referred to the high side:
Compute the secondary voltage by dividing by the turns ratio of 10:
Calculate the percent voltage drop to the secondary of transformer T3. Use
the secondary voltage referred to the high side:
2.4.2.2 Load Allocation Based upon Transformer Ratings
When only the ratings of the distribution transformers are known, the feeder
can be allocated based upon the metered demand and the transformer kVA
ratings. This method was discussed in Section 2.3.3. Example 2.4 demons
trates this method.
Example 2.4
For the system of Example 2.1, assume the voltage at N1 is 2400 volts and
compute the secondary voltages on the three transformers, allocating the
loads based upon the transformer ratings. Assume that the metered kW
demand at N1 is 92.9 kW.
The impedances of the line segments and transformers are the same as in
Example 2.3. Assume the load power factor is 0.9 lagging; compute the kVA
demand at N1 from the metered demand:
Calculate the allocation factor:
I
T3
22.91/ 26.30 – A =
V
T3
V
4
Z
T3
I
T3
⋅ – =
V
T3
2375.0/ 0.5 – 1.48 j1.77 + ( ) 22.9/ 26.27 – ⋅ – 2326.9/ 1.0 – V = =
Vlow
T3
2326.9/ 1.0 –
10
 232.7/ 1.0 – V = =
V
drop
V
1
V
T3
–
V
1
 100 ⋅
2400 2326.11 –
2400
 100 ⋅ 3.0789% = = =
S
12
92.9
0.9
/cos
1 –
0.9 ( ) 92.8 j45.0 + 103.2/25.84 kVA = = =
AF
103.2/25.84
25 37.5 50 + +
 0.9175/25.84 = =
0812_Frame_C02.fm Page 31 Saturday, July 21, 2001 2:53 PM
32
Distribution System Modeling and Analysis
Allocate the loads to each transformer:
Calculate the line ﬂows:
Using these values of line ﬂows and ﬂows into transformers, the procedure
for computing the transformer secondary voltages is exactly the same as in
Example 2.3. When this procedure is followed, the node and secondary
transformer voltages are
The percent voltage drop for this case is
2.5 Summary
This chapter has demonstrated the nature of the loads on a distribution
feeder. There is great diversity between individual customer demands, but
as the demand is monitored on line segments working back toward the
substation, the effect of the diversity between demands becomes very slight.
It was shown that the effect of diversity between customer demands must
be taken into account when the demand on a distribution transformer is
computed. The effect of diversity for short laterals can be taken into account
in determining the maximum ﬂow on the lateral. For the diversity factors
of Table 2.2, it was shown that when the number of customers exceeds 70 the
effect of diversity has pretty much disappeared. This is evidenced by the fact
that the diversity factor became almost constant as the number of customers
S
T1
AF kVA
T1
⋅ 0.9175/25.84 ( ) 25 ⋅ 20.6 j10.0 kVA + = = =
S
T2
AF kVA
T2
⋅ 0.9175/25.84 ( ) 37.5 ⋅ 31.0 j15.0 kVA + = = =
S
T3
AF kVA
T3
⋅ 0.9175/25.84 ( ) 50 ⋅ 41.3 j20.0 kVA + = = =
S
12
S
T1
S
T2
S
T3
+ + 92.9 j45.0 kVA + = =
S
23
S
T2
S
T3
+ 72.3 j35 kVA + = =
S
34
S
T3
41.3 j20.0 kVA + = =
V
2
2378.4/ 0.4 – V Vlow
T1
234.0/ 0.4 – V = =
V
3
2376.7/0.4 V Vlow
T2
233.7/ 0.8 – V = =
V
4
2375.3/ 0.4 – V Vlow
T3
233.5/ 0.9 – V = =
V
drop
V
1
V
T3
–
V
1
 100 ⋅
2400 2334.8 –
2400
 100 ⋅ 2.7179% = = =
0812_Frame_C02.fm Page 32 Monday, October 28, 2002 9:47 AM
The Nature of Loads 33
approached 70. It must be understood that the number 70 will apply only to
the diversity factors of Table 2.2. If a utility is going to use diversity factors, that
utility must perform a comprehensive load survey in order to develop the
table of diversity factors that apply to that particular system.
Examples 2.3 and 2.4 show that the ﬁnal node and transformer voltages
are approximately the same. There was very little difference between the
voltages when the loads were allocated using the diversity factors and when
the loads were allocated based upon the transformer kVA ratings.
Problems
2.1 Shown below are the 15minute kW demands for four customers
between the hours of 17:00 and 21:00. A 25kVA singlephase transformer
serves the four customers.
1. For each of the customers determine:
(a) Maximum 15minute kW demand
(b) Average 15minute kW demand
(c) Total kWh usage in the time period
(d) Load factor
Time Cust #1 Cust #2 Cust #3 Cust #4
kW kW kW kW
17:00 8.81 4.96 11.04 1.44
17:15 2.12 3.16 7.04 1.62
17:30 9.48 7.08 7.68 2.46
17:45 7.16 5.08 6.08 0.84
18:00 6.04 3.12 4.32 1.12
18:15 9.88 6.56 5.12 2.24
18:30 4.68 6.88 6.56 1.12
18:45 5.12 3.84 8.48 2.24
19:00 10.44 4.44 4.12 1.12
19:15 3.72 8.52 3.68 0.96
19:30 8.72 4.52 0.32 2.56
19:45 10.84 2.92 3.04 1.28
20:00 6.96 2.08 2.72 1.92
20:15 6.62 1.48 3.24 1.12
20:30 7.04 2.33 4.16 1.76
20:45 6.69 1.89 4.96 2.72
21:00 1.88 1.64 4.32 2.41
0812_Frame_C02.fm Page 33 Saturday, July 21, 2001 2:53 PM
34 Distribution System Modeling and Analysis
2. For the 25kVA transformer determine:
(a) Maximum 15minute diversiﬁed demand
(b) Maximum 15minute noncoincident demand
(c) Utilization factor (assume unity power factor)
(d) Diversity factor
(e) Load diversity
3. Plot the load duration curve for the transformer
2.2 Two transformers each serving four customers are shown in Figure 2.13.
The following table gives the time interval and kVA demand of the four
customer demands during the peak load period of the year. Assume a power
factor of 0.9 lagging.
1. For each transformer determine the following:
(a) 30minute maximum kVA demand
(b) Noncoincident maximum kVA demand
(c) Load factor
(d) Diversity factor
(e) Suggested transformer rating (50, 75, 100, 167)
FIGURE 2.13
System for Problem 2.2.
Time #1 #2 #3 #4 #5 #6 #7 #8
3:00–3:30 10 0 10 5 15 10 50 30
3:30–4:00 20 25 15 20 25 20 30 40
4:00–4:30 5 30 30 15 10 30 10 10
4:30–5:00 0 10 20 10 13 40 25 50
5:00–5:30 15 5 5 25 30 30 15 5
5:30–6:00 15 15 10 10 5 20 30 25
6:00–6:30 5 25 25 15 10 10 30 25
6:30–7:00 10 50 15 30 15 5 10 30
#3 #4 #1 #2 #5 #6 #7 #8
Tap
0812_Frame_C02.fm Page 34 Saturday, July 21, 2001 2:53 PM
The Nature of Loads 35
(f) Utilization factor
(g) Energy (kWh) during the 4hour period
2. Determine the maximum diversiﬁed 30minute kVA demand at the
Tap
2.3 Two singlephase transformers serving 12 customers are shown in
Figure 2.14.
The 15minute kW demands for the 12 customers between the hours of
5:00 p.m. and 9:00 p.m. are given in the tables that follow. Assume a load
power factor of 0.95 lagging. The impedance of the lines are z = 0.306 +
j0.6272 Ω/mile. The voltage at node N1 is 2500 V. Transformer ratings:
1. Determine the maximum kW demand for each customer.
2. Determine the average kW demand for each customer.
3. Determine the kWH consumed by each customer in this time period.
4. Determine the load factor for each customer.
5. Determine the maximum diversiﬁed demand for each transformer.
6. Determine the maximum noncoincident demand for each transformer.
7. Determine the utilization factor (assume 1.0 power factor) for each
transformer.
8. Determine the diversity factor of the load for each transformer.
9. Determine the maximum diversiﬁed demand at Node N1.
10. Compute the secondary voltage for each transformer, taking diver
sity into account.
FIGURE 2.14
Circuit for Problem 2.3.
T1 T2
N1 N2 N3
1 2 3 4 5 6 7 8 9 10 11 12
2500' 5000'
/0
T1: 25 kVA 2400240 V Z
pu
0.018/40 =
T2: 37.5 kVA 2400240 V Z
pu
0.020/50 =
0812_Frame_C02.fm Page 35 Saturday, July 21, 2001 2:53 PM
36 Distribution System Modeling and Analysis
2.4 On a different day, the metered 15minute kW demand at node N1 for
the system of Problem 2.3 is 72.43 kW. Assume a power factor of 0.95 lagging.
Allocate the metered demand to each transformer based upon the trans
former kVA rating. Assume the loads are constant current and compute the
secondary voltage for each transformer.
TRANSFORMER #125 kVA
Time #1 #2 #3 #4 #5
kW kW kW kW kW
05:00 2.13 0.19 4.11 8.68 0.39
05:15 2.09 0.52 4.11 9.26 0.36
05:30 2.15 0.24 4.24 8.55 0.43
05:45 2.52 1.80 4.04 9.09 0.33
06:00 3.25 0.69 4.22 9.34 0.46
06:15 3.26 0.24 4.27 8.22 0.34
06:30 3.22 0.54 4.29 9.57 0.44
06:45 2.27 5.34 4.93 8.45 0.36
07:00 2.24 5.81 3.72 10.29 0.38
07:15 2.20 5.22 3.64 11.26 0.39
07:30 2.08 2.12 3.35 9.25 5.66
07:45 2.13 0.86 2.89 10.21 6.37
08:00 2.12 0.39 2.55 10.41 4.17
08:15 2.08 0.29 3.00 8.31 0.85
08:30 2.10 2.57 2.76 9.09 1.67
08:45 3.81 0.37 2.53 9.58 1.30
09:00 2.04 0.21 2.40 7.88 2.70
TRANSFORMER #237.5 kVA
Time #6 #7 #8 #9 #10 #11 #12
kW kW kW kW kW kW kW
05:00 0.87 2.75 0.63 8.73 0.48 9.62 2.55
05:15 0.91 5.35 1.62 0.19 0.40 7.98 1.72
05:30 1.56 13.39 0.19 5.72 0.70 8.72 2.25
05:45 0.97 13.38 0.05 3.28 0.42 8.82 2.38
06:00 0.76 13.23 1.51 1.26 3.01 7.47 1.73
06:15 1.10 13.48 0.05 7.99 4.92 11.60 2.42
06:30 0.79 2.94 0.66 0.22 3.58 11.78 2.24
06:45 0.60 2.78 0.52 8.97 6.58 8.83 1.74
07:00 0.60 2.89 1.80 0.11 7.96 9.21 2.18
07:15 0.87 2.75 0.07 7.93 6.80 7.65 1.98
07:30 0.47 2.60 0.16 1.07 7.42 7.78 2.19
07:45 0.72 2.71 0.12 1.35 8.99 6.27 2.63
08:00 1.00 3.04 1.39 6.51 8.98 10.92 1.59
08:15 0.47 1.65 0.46 0.18 7.99 5.60 1.81
08:30 0.44 2.16 0.53 2.24 8.01 7.74 2.13
08:45 0.95 0.88 0.56 0.11 7.75 11.72 1.63
09:00 0.79 1.58 1.36 0.95 8.19 12.23 1.68
0812_Frame_C02.fm Page 36 Saturday, July 21, 2001 2:53 PM
The Nature of Loads
37
2.5
A singlephase lateral serves four transformers as shown in Figure 2.15.
Assume that each customer’s maximum demand is 15.5 kW +
j
7.5 kvar. The
impedance of the singlephase lateral is
z
=
0.4421
+
j
0.3213
Ω
/1000 ft. The
four transformers are rated as:
T1 and T2: 37.5 kVA, 2400240 V,
Z
=
0.01
+
j
0.03 perunit
T3 and T4: 50 kVA, 2400240 V,
Z
=
0.015
+
j
0.035 perunit
Use the diversity factors found in Table 2.2 and determine:
(1) The 15minute maximum diversiﬁed kW and kvar demands on
each transformer.
(2) The 15minute maximum diversiﬁed kW and kvar demands for
each line section.
(3) If the voltage at node 1 is 2600 V, determine the voltage at nodes
2, 3, 4, 5, 6, 7, 8, and 9. In calculating the voltages, take into account
diversity using the answers from (1) and (2) above.
(4) Use the 15minute maximum diversiﬁed demands at the lateral tap
(Section 12) from Part (2) above. Divide these maximum demands
by 18 (number of customers) and assign that as the “instantaneous
load” for each customer. Now calculate the voltages at all of the
nodes listed in Part (3) using the instantaneous loads.
(5) Repeat Part (4) above, but assume the loads are “constant current.”
To do this, take the current ﬂowing from node 1 to node 2 from
Part (4) above, divide by 18 (number of customers), and assign that
as the instantaneous constant current load for each customer.
Again, calculate all of the voltages.
FIGURE 2.15
System for Problem 2.5.
1
2
4 6
8
7 5 3 9
380' 470' 750' 820'
T1 T2 T3 T4
/0
0812_Frame_C02.fm Page 37 Monday, October 28, 2002 9:48 AM
38 Distribution System Modeling and Analysis
(6) Take the maximum diversiﬁed demand from node 1 to node 2 and
allocate that to each of the four transformers based upon their kVA
ratings. To do this, take the maximum diversiﬁed demand and
divide by 175 (total kVA of the four transformers). Now multiply
each transformer kVA rating by that number to give how much of
the total diversiﬁed demand is being served by each transformer.
Again, calculate all of the voltages.
(7) Compute the percent differences in the voltages for Parts (4), (5),
and (6) at each of the nodes using the Part (3) answer as the base.
0812_Frame_C02.fm Page 38 Saturday, July 21, 2001 2:53 PM
39
3
Approximate Methods of Analysis
A distribution feeder provides service to unbalanced threephase, twophase,
and singlephase loads over untransposed threephase, twophase, and single
phase line segments. This combination leads to threephase line currents and
line voltages being unbalanced. In order to analyze these conditions as pre
cisely as possible, it will be necessary to model all three phases of the feeder
accurately, however, many times only a “ballpark” answer is needed. When
this is the case, some approximate methods of modeling and analysis can be
employed. It is the purpose of this chapter to develop some of the approximate
methods and leave for later chapters the exact models and analysis.
All of the approximate methods of modeling and analysis will assume
perfectly balanced threephase systems. It will be assumed that all loads are
balanced threephase, and all line segments will be threephase and perfectly
transposed. With these assumptions, a single linetoneutral equivalent cir
cuit for the feeder will be used.
3.1 Voltage Drop
A linetoneutral equivalent circuit of a threephase line segment serving a
balanced threephase load is shown in Figure 3.1. Kirchhoff’s voltage law
applied to the circuit of Figure 3.1 gives:
(3.1)
The phasor diagram for Equation 3.1 is shown in Figure 3.2. In Figure 3.2
the phasor for the voltage drop through the line resistance (RI) is shown in
phase with the current phasor, and the phasor for the voltage drop through
the reactance is shown leading the current phasor by 90 degrees. The dashed
lines represent the real and imaginary parts of the impedance (ZI) drop. The
voltage drop down the line is deﬁned as the difference between the magni
tudes of the source and the load voltages.
(3.2)
V
S
V
L
R jX + ( ) I ⋅ + V
L
R I jX I ⋅ + ⋅ +
V
drop
V
S
V
L
–
0812_frame_C03.fm Page 39 Saturday, July 21, 2001 3:00 PM
40
Distribution System Modeling and Analysis
The angle between the source voltage and the load voltage (
δ
) is very small.
Because of that, the voltage drop between the source and load voltage is
approximately equal to the real part of the impedance drop. That is
(3.3)
For the purposes of this chapter, Equation 3.3 will be used as the deﬁnition
of voltage drop.
Example 3.1
In Example 2.3, the impedance of the ﬁrst line segment is
The current ﬂowing through the line segment is
FIGURE 3.1
Linetoneutral equivalent circuit.
FIGURE 3.2
Phasor diagram.
R jX
Load
+

V
S
+
V
L

I
I
RI
V
S
V
L
jXI
ZI
0
Im(ZI)
Real(ZI)
V
drop
Re Z I ⋅ ( ) ≅
Z
12
0.2841 j0.5682 Ω +
I
12
43.0093/ 25.8419 – A
0812_frame_C03.fm Page 40 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis
41
The voltage at node N1 is
The exact voltage at node N2 is computed to be
The voltage drop between the nodes is then:
Computing the voltage drop according to Equation 3.3 gives:
This example demonstrates the very small error in computing voltage drop
when using the approximate equation given by Equation 3.3.
3.2 Line Impedance
For the approximate modeling of a line segment, it will be assumed that the line
segment is transposed. With this assumption, only the positive sequence im
pedance of the line segment needs to be determined. A typical threephase line
conﬁguration is shown in Figure 3.3. The equation for the positive sequence
impedance for the conﬁguration shown in Figure 3.3 is given by:
(3.4)
where
r
conductor resistance (from tables)
Ω
/mile
(3.5)
GMR
conductor geometric mean radius (from tables) (ft.)
V
1
2400/0.0 V
V
2
2400/0.0 0.2841 j0.5682 + ( ) 43.0093/ 25.8419 – ⋅ –
2378.4098/ 0.4015 – V
V
drop
2400.0000 2378.4098 – 21.5902 V
V
drop
Re 0.2841 j0.5682 + ( ) 43.0093/ 25.8419 – ⋅ [ ] 21.6486 V
Error
21.5902 21.6486 –
21.5902
 100 ⋅ 0.27% –
z
positive
r j0.12134 ln
D
eq
GMR

( J
 ·
Ω/mile ⋅ +
D
eq
D
ab
D
bc
D
ca
⋅ ⋅
3
(ft.)
0812_frame_C03.fm Page 41 Saturday, July 21, 2001 3:00 PM
42
Distribution System Modeling and Analysis
Example 3.2
A threephase line segment has the conﬁguration as shown in Figure 3.3. The
spacings between conductors are
The conductors of the line are 336,400 26/7 ACSR. Determine the positive
sequence impedance of the line in ohms/mile:
SOLUTION
From the table of conductor data in Appendix A:
Compute the equivalent spacing:
Using Equation 3.4:
FIGURE 3.3
Threephase line conﬁguration.
Dab
Dbc
a b c
n
Dca
D
ab
2.5 ft., D
bc
4.5 ft., D
ca
7.0 ft.
r 0.306 Ω/mile
GMR 0.0244 ft
D
eq
2.5 4.5 7.0 ⋅ ⋅
3
4.2863 ft
z
positive
0.306 j0.12134 ln
4.2863
0.0244

( J
 ·
⋅ + 0.306 j0.6272 Ω/mile +
0812_frame_C03.fm Page 42 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis
43
3.3 “K” Factors
A ﬁrst approximation for calculating the voltage drop along a line segment
is given by Equation 3.3. Another approximation is made by employing a
‘‘K” factor. There will be two types of K factors: one for voltage drop and
the other for voltage rise calculations.
3.3.1 The
K
drop
Factor
The
K
drop
factor is deﬁned as:
(3.5)
The
K
drop
factor is determined by computing the percent voltage drop down
a line that is one mile long and serving a balanced threephase load of 1 kVA.
The percent voltage drop is referenced to the nominal voltage of the line. In
order to calculate this factor, the power factor of the load must be assumed.
Example 3.3
For the line of Example 3.2, compute the
K
drop
factor assuming a load power
factor of 0.9 lagging and a nominal voltage of 12.47 kV (linetoline).
SOLUTION
The impedance of one mile of line was computed to be
The current taken by 1 kVA at 0.9 lagging power factor is given by:
The voltage drop is computed to be
The nominal linetoneutral voltage is
K
drop
Percent voltage drop
kVA · mile

Z 0.306 j0.6272 Ω +
I
1 kVA
3 kV
LL
⋅
/ cos
1 –
PF ( ) –
1
3 12.47 ⋅
/ cos
1 –
0.9 ( ) – 0.046299/ 25.84 – A
V
drop
Re Z I ⋅ [ ] Re 0.306 j0.6272 + ( ) 0.046299/ 25.84 – ⋅ [ ] 0.025408 V
V
LN
12470
3
 7199.6 V
0812_frame_C03.fm Page 43 Saturday, July 21, 2001 3:00 PM
44 Distribution System Modeling and Analysis
The K
drop
factor is then:
The K
drop
factor computed in Example 3.3 is for the 336,400 26/7 ACSR
conductor with the conductor spacings deﬁned in Example 3.2, a nominal
voltage of 12.47 kV, and a load power factor of 0.9 lagging. Unique K
drop
factors can be determined for all standard conductors, spacings, and volt
ages. Fortunately, most utilities will have a set of standard conductors, stan
dard conductor spacings, and one or two standard distribution voltages.
Because of this, a simple spreadsheet program can be written that will
compute the K
drop
factors for the standard conﬁgurations. The assumed
power factor of 0.9 lagging is a good approximation for a feeder serving a
predominately residential load.
The K
drop
factor can be used to quickly compute the approximate voltage
drop down a line section. For example, assume a load of 7500 kVA is to be
served at a point 1.5 miles from the substation. Using the K
drop
factor com
puted in Example 3.3, the percent voltage drop down the line segment is
computed to be
This example demonstrates that a load of 7500 kVA can be served 1.5 miles
from the substation with a resulting voltage drop of 3.97%. Suppose now
that the utility has a maximum allowable voltage drop of 3.0%. How much
load can be served 1.5 miles from the substation?
The application of the K
drop
factor is not limited to computing the percent
voltage drop down just one line segment. When line segments are in cascade,
the total percent voltage drop from the source to the end of the last line segment
is the sum of the percent drops in each line segment. This seems logical, but
it must be understood that in all cases the percent drop is in reference to the
nominal linetoneutral voltage. That is, the percent voltage drop in a line
segment is not referenced to the source end voltage, but rather the nominal
linetoneutral voltage, as would be the usual case. Example 3.4 will demons
trate this application.
Example 3.4
A threesegment feeder is shown in Figure 3.4. The K
drop
factor for the line
segments is
Determine the percent voltage drop from N0 to N3.
K
drop
0.025408
7199.6
 100 ⋅ 0.00035291% drop/kVAmile
V
drop
K
drop
kVA mile ⋅ ⋅ 0.00035291 7500 1.5 ⋅ ⋅ 3.9702%
kVA
load
3.0%
0.00035291 1.5 ⋅
 5667.2 kVA
K
drop
0.00035291
0812_frame_C03.fm Page 44 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 45
SOLUTION
The total kVA ﬂowing in segment N0 to N1 is
The percent voltage drop from N0 to N1 is
The total kVA ﬂowing in segment N1 to N2 is
The percent voltage drop from N1 to N2 is
The kVA ﬂowing in segment N2 to N3 is
The percent voltage drop in the last line segment is
The total percent voltage drop from N0 to N3 is
The application of the K
drop
factor provides an easy way of computing the
approximate percent voltage drop from a source to a load. It should be kept
in mind that the assumption has been a perfectly balanced threephase load,
FIGURE 3.4
Threesegment feeder.
1.5 mile 0.75 mile 0.5 mile
750 kVA 500 kVA
300 kVA
N2 N3 N0 N1
kVA
01
300 750 500 + + 1550 kVA
Vdrop
01
0.00035291 1550 1.5 ⋅ ⋅ 0.8205%
kVA
12
750 500 + 1250 kVA
Vdrop
12
0.00035291 1250 0.75 ⋅ ⋅ 0.3308%
kVA
23
500
Vdrop
23
0.00035291 500 0.5 ⋅ ⋅ 0.0882%
Vdrop
total
0.8205 0.3308 0.0882 + + 1.2396%
0812_frame_C03.fm Page 45 Saturday, July 21, 2001 3:00 PM
46 Distribution System Modeling and Analysis
an assumed load power factor, and transposed line segments. Even with
these assumptions the results will always provide a ‘‘ballpark” result that
can be used to verify the results of more sophisticated methods of computing
voltage drop.
3.3.2 The K
rise
Factor
The K
rise
factor is similar to the K
drop
factor except that now the load is a
shunt capacitor. When a leading current ﬂows through an inductive reac
tance there will be a voltage rise across the reactance rather than a voltage
drop. This is illustrated by the phasor diagram of Figure 3.5. Referring to
Figure 3.5, the voltage rise is deﬁned as
(3.6)
In Equation 3.6 it is necessary to take the magnitude of the real part of ZI
so that the voltage rise is a positive number. The K
rise
factor is deﬁned exactly
the same as for the K
drop
factor:
(3.7)
Example 3.5
1. Calculate the K
rise
factor for the line of Example 3.3.
2. Determine the rating of a threephase capacitor bank to limit the
voltage drop in Example 3.3 to 2.5%.
SOLUTION
1. The impedance of one mile of line was computed to be
FIGURE 3.5
Voltage rise phasor diagram.
V
L
I
cap
RI
V
S
cap
jXI
cap
ZI
Real(ZI)
Im(ZI)
V
rise
Re ZI
cap
( ) X I
cap
⋅
K
rise
Percent voltage rise
kvar mile.
 =
Z 0.306 j0.6272 Ω +
0812_frame_C03.fm Page 46 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 47
The current taken by a 1kvar threephase capacitor bank is given by:
The voltage rise per kvar mile is computed to be
The nominal linetoneutral voltage is
The K
rise
factor is then
2. The percent voltage drop in Example 3.3 was computed to be 3.9702%.
To limit the total voltage drop to 2.5%, the required voltage rise
due to a shunt capacitor bank is
The required rating of the shunt capacitor is
3.4 Uniformly Distributed Loads
Many times it can be assumed that loads are uniformly distributed along a
line where the line can be a threephase, twophase, or singlephase feeder
or lateral. This is certainly the case on singlephase laterals where the same
rating transformers are spaced uniformly over the length of the lateral.
When the loads are uniformly distributed, it is not necessary to model each
load in order to determine the total voltage drop from the source end to the
last load. Figure 3.6 shows a generalized line with n uniformly distributed
loads.
I
cap
1 kvar
3 kV
LL
⋅
/90
1
3 12.47 ⋅
/90 0.046299/90 A
V
rise
Re Z I
cap
⋅ [ ] Re 0.306 j0.6272 + ( ) 0.046299/90 ⋅ [ ]
0.029037 V
V
LN
12,470
3
 7199.6 V
K
rise
0.029037
7199.6
 100 ⋅ 0.00040331% rise/kvar mile
V
rise
3.9702 2.5 – 1.4702%
kvar
V
rise
K
rise
mile ⋅

1.4702
0.00040331 1.5 ⋅
 2430.18 kvar
0812_frame_C03.fm Page 47 Saturday, July 21, 2001 3:00 PM
48 Distribution System Modeling and Analysis
3.4.1 Voltage Drop
Figure 3.6 shows n uniformly spaced loads dx miles apart. The loads are all
equal and will be treated as constant current loads with a value of di. The
total current into the feeder is I
T
. It is desired to determine the total voltage
drop from the source node (S) to the last node n.
Let
l = length of the feeder
z = r + jx = impedance of the line in Ω/mile
dx = length of each line section
di = load currents at each node
n = number of nodes and number of line sections
I
T
= total current into the feeder
The load currents are given by:
(3.8)
The voltage drop in the ﬁrst line segment is given by:
(3.9)
The voltage drop in the second line segment is given by:
(3.10)
The total voltage drop from the source node to the last node is then given by:
(3.11)
FIGURE 3.6
Uniformly distributed loads.
1 2 3 4 5
di di di di di di
dx dx dx dx dx
S
I
T
n
length
di
I
T
n

Vdrop
1
Re z dx n di ⋅ ( ) ⋅ ⋅ { ¦
Vdrop
2
Re z dx n 1 – ( ) di ⋅ [ ] ⋅ ⋅ { ¦
Vdrop
total
Vdrop
1
Vdrop
2
…
Vdrop
n
+ + +
Vdrop
total
Re z dx di n n 1 – ( ) n 2 – ( )
…
1 ( ) + + + + [ ] ⋅ ⋅ ⋅ { ¦
0812_frame_C03.fm Page 48 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 49
Equation 3.11 can be reduced by recognizing the series expansion:
(3.12)
Using the expansion, Equation 3.11 becomes:
(3.13)
The incremental distance is
(3.14)
The incremental current is
(3.15)
Substituting Equations 3.14 and 3.15 into Equation 3.13 results in:
(3.16)
where Z z ⋅ l
Equation 3.16 gives the general equation for computing the total voltage
drop from the source to the last node n for a line of length l. In the limiting
case where n goes to inﬁnity, the ﬁnal equation becomes:
(3.17)
1 2 3
…
n + + + +
n n 1 + ( )
2

Vdrop
total
Re z dx di
n n 1 + ( ) ⋅
2
 ⋅ ⋅ ⋅
 
¦ 
 
dx
l
n

di
I
T
n

Vdrop
total
Re z
l
n

I
T
n

n n 1 + ( ) ⋅
2
 ⋅ ⋅ ⋅
 
¦ 
 
Vdrop
total
Re z l I
T
1
2

n 1 +
n

( J
 ·
⋅ ⋅ ⋅ ⋅
 
¦ 
 
Vdrop
total
Re
1
2
 Z I
T
1
1
n
 +
( J
 ·
⋅ ⋅ ⋅
 
¦ 
 
Vdrop
total
Re
1
2
 Z I
T
⋅ ⋅
 
¦ 
 
0812_frame_C03.fm Page 49 Saturday, July 21, 2001 3:00 PM
50 Distribution System Modeling and Analysis
In Equation 3.17, Z represents the total impedance from the source to the
end of the line. The voltage drop is the total from the source to the end of
the line. The equation can be interpreted in two ways. The ﬁrst is to recognize
that the total line distributed load can be lumped at the midpoint of the
lateral as shown in Figure 3.7. A second interpretation of Equation 3.17 is to
lump onehalf of the total line load at the end of the line (node n). This model
is shown in Figure 3.8. Figures 3.7 and 3.8 give two different models that
can be used to calculate the total voltage drop from the source to the end of
a line with uniformly distributed loads.
3.4.2 Power Loss
Of equal importance in the analysis of a distribution feeder is the power
loss. If the model of Figure 3.7 is used to compute the total threephase power
loss down the line, the result is
(3.18)
When the model of Figure 3.8 is used to compute the total threephase power
loss, the result is
(3.19)
FIGURE 3.7
Load lumped at the midpoint.
FIGURE 3.8
Onehalf load lumped at the end.
S
I
T
n
length
I
T
length/2
S
I
T
n
length
I
T
/2
P
loss
3 I
T
2 R
2
 ⋅ ⋅
3
2
 I
T
2
R ⋅ ⋅
P
loss
3
I
T
2

2
R ⋅ ⋅
3
4
 I
T
2
R ⋅ ⋅
0812_frame_C03.fm Page 50 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 51
It is obvious that the two models give different results for power loss. The
question is, which one is correct? The answer is neither one.
To derive the correct model for power loss, reference is made to Figure 3.6
and the deﬁnitions for the parameters in that ﬁgure. The total threephase
power loss down the line will be the sum of the power losses in each short
segment of the line. For example, the threephase power loss in the ﬁrst seg
ment is
(3.20)
The power loss in the second segment is given by:
(3.21)
The total power loss over the length of the line is then given by:
(3.22)
The series inside the brackets of Equation 3.22 is the sum of the squares of
n numbers and is equal to:
(3.23)
Substituting Equations 3.14, 3.15, and 3.23 into Equation 3.22 gives:
(3.24)
Simplifying Equation 3.24:
(3.25)
Where R r ⋅ l, the total resistance per phase of the line segment, Equation
3.25 gives the total threephase power loss for a discrete number of nodes
and line segments. For a truly uniformly distributed load, the number of
Ploss
1
3 r dx ⋅ ( ) n di ⋅ ( )
2
⋅ ⋅
Ploss
2
3 r dx ⋅ ( ) n 1 – ( ) di ⋅ [ ]
2
⋅ ⋅
Ploss
total
3 r dx ⋅ ( ) di
2
n
2
n 1 – ( )
2
n 2 – ( )
2
…
1
2
+ + + + [ ] ⋅ ⋅
1
2
2
2
3
2
…
n
2
+ + + +
n n 1 + ( ) 2n 1 + ( ) ⋅ ⋅
6
 =
Ploss
total
3 r
l
n
 ⋅
( J
 ·
I
T
n

( J
 ·
2
n n 1 + ( ) 2n 1 + ( ) ⋅ ⋅
6
 ⋅ ⋅ ⋅
Ploss
total
3 R I
T
2 n 1 + ( ) 2n 1 + ( ) ⋅
6 n
2
⋅
 ⋅ ⋅ ⋅
Ploss
total
3 R I
T
2 2 n
2
3 n 1 + ⋅ + ⋅
6 n
2
⋅
 ⋅ ⋅ ⋅
Ploss
total
3 R I
T
2 1
3

1
2 n ⋅

1
6 n
2
⋅
 + + ⋅ ⋅ ⋅
0812_frame_C03.fm Page 51 Saturday, July 21, 2001 3:00 PM
52 Distribution System Modeling and Analysis
nodes goes to inﬁnity. When that limiting case is taken in Equation 3.25, the
ﬁnal equation for computing the total threephase power loss down the line
is given by:
(3.26)
A circuit model for Equation 3.26 is given in Figure 3.9. From a comparison
of Figures 3.7 and 3.8, used for voltage drop calculations, to Figure 3.9, used
for power loss calculations, it is obvious that the same model cannot be used
for both voltage drop and power loss calculations.
3.4.3 The Exact Lumped Load Model
In the previous sections lumped load models were developed. The ﬁrst mod
els of Section 3.4.1 can be used for the computation of the total voltage drop
down the line. It was shown that the same models cannot be used for the
computation of the total power loss down the line. Section 3.4.2 developed a
model that will give the correct power loss of the line. What is needed is one
model that will work for both voltage drop and power loss calculations.
Figure 3.10 shows the general conﬁguration of the exact model that will give
correct results for voltage drop and power loss. In Figure 3.10 a portion (I
x
) of
FIGURE 3.9
Power loss model.
FIGURE 3.10
General exact lumped load model.
S
I
T
n
length
length/3
I
T
l
I
l
l k (1k)
n
c
T
I
T
I
X
S
Ploss
total
3
1
3
 R I
T
2
⋅ ⋅ ⋅
0812_frame_C03.fm Page 52 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 53
the total line current (I
T
) will be modeled kl miles from the source end, and
the remaining current (cI
T
) will be modeled at the end of the line. The values
of k and c need to be derived.
In Figure 3.10 the total voltage drop down the line is given by:
(3.27)
where
Z total line impedance in ohms
k factor of the total line length where the ﬁrst part of the load current
is modeled
c factor of the total current to place at the end of the line such that
In Section 3.4.1 it was shown that the total voltage drop down the line is
given by:
(3.28)
Set Equation 3.17 equal to Equation 3.27:
(3.29)
Equation 3.29 shows that the terms inside the brackets on both sides of the
equal sign need to be set equal, that is
(3.30)
Simplify Equation 3.30 by dividing both side of the equation by ZI
T
:
(3.31)
Solve Equation 3.31 for k:
(3.32)
The same procedure can be followed for the power loss model. The total
threephase power loss in Figure 3.10 is given by:
(3.33)
Vdrop
total
Re k Z I
T
1 k – ( ) Z c I
T
⋅ ⋅ ⋅ + ⋅ ⋅ [ ]
I
T
I
x
c I
T
⋅ + =
Vdrop
total
Re
1
2
 Z I
T
⋅ ⋅
Vdrop
total
Re
1
2
 Z I
T
⋅ ⋅ Re k Z I
T
1 k – ( ) Z c I
T
⋅ ⋅ ⋅ + ⋅ ⋅ [ ]
1
2
 Z I
T
⋅ ⋅ k Z I
T
1 k – ( ) Z c I
T
⋅ ⋅ ⋅ + ⋅ ⋅ [ ]
1
2
 k 1 k – ( ) + c ⋅ [ ]
k
0.5 c –
1 c –

Ploss
total
3 k R I
T
2
1 k – ( ) R c I
T
⋅ ( )
2
⋅ ⋅ + ⋅ ⋅ [ ] ⋅
0812_frame_C03.fm Page 53 Saturday, July 21, 2001 3:00 PM
54 Distribution System Modeling and Analysis
The model for the power loss of Figure 3.9 gives the total threephase power
loss as:
(3.34)
Equate the terms inside the brackets of Equations 3.33 and 3.34 and simplify:
(3.35)
Substitute Equation 3.32 into Equation 3.35:
(3.36)
Solving Equation 3.36 for c results in:
(3.37)
Substitute Equation 3.37 into Equation 3.32 and solve for k:
(3.38)
The interpretation of Equations 3.37 and 3.38 is that onethird of the load
should be placed at the end of the line, and twothirds of the load placed
onefourth of the way from the source end. Figure 3.11 gives the ﬁnal exact
lumped load model.
FIGURE 3.11
Exact lumped load model.
Ploss
total
3
1
3
 R I
T
2
⋅ ⋅ ⋅ =
1
3
 R I
T
2
⋅ ⋅ k R I
T
2
1 k – ( ) R c I
T
⋅ ( )
2
⋅ ⋅ + ⋅ ⋅ [ ]
1
3
 k 1 k – ( ) c ( )
2
⋅ + [ ]
1
3
 k c
2
k c
2
⋅ – + [ ] k 1 c
2
– ( ) c
2
+ ⋅ [ ]
1
3

0.5 c –
1 c –
 1 c
2
– ( ) c
2
+ ⋅
c
1
3

k
1
4

0812_frame_C03.fm Page 54 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 55
3.5 Lumping Loads in Geometric Conﬁgurations
Many times feeder areas can be represented by geometric conﬁgurations
such as rectangles, triangles, and trapezoids. By assuming a constant load
density in the conﬁgurations, approximate calculations can be made for
computing the voltage drop and total power losses. The approximate calcu
lations can aid in the determination of the maximum load that can be served
in a speciﬁed area at a given voltage level and conductor size. For all of the
geographical areas to be evaluated, the following deﬁnitions will apply:
D load density in
PF assumed lagging power factor
z line impedance in Ω/mile
l length of the area
w width of the area
kV
LL
nominal linetoline voltage in kV
It will also be assumed that the loads are modeled as constant current loads.
3.5.1 The Rectangle
A rectangular area of length l and width w is to be served by a primary main
feeder. The feeder area is assumed to have a constant load density with three
phase laterals uniformly tapped off of the primary main. Figure 3.12 is a
model for the rectangular area. Figure 3.12 represents a rectangular area of
constant load density being served by a threephase main running from node
n to node m. It is desired to determine the total voltage drop and the total
threephase power loss down the primary main from node n to node m.
FIGURE 3.12
Constant load density rectangular area.
kVA
mile
2

n
.5di
.5 di
l
w
dx
m
x
i
I
T
0812_frame_C03.fm Page 55 Saturday, July 21, 2001 3:00 PM
56 Distribution System Modeling and Analysis
The total current entering the area is given by:
(3.39)
An incremental segment is located x miles from node n. The incremental
current serving the load in the incremental segment is given by:
(3.40)
The current in the incremental segment is given by:
(3.41)
The voltage drop in the incremental segment is
(3.42)
The total voltage drop down the primary main feeder is
Evaluating the integral and simplifying:
(3.43)
where Z z ⋅ l
Equation 3.43 gives the same result as that of Equation 3.17, which was
derived for loads uniformly distributed along a feeder. The only difference
is the manner in which the total current (I
T
) is determined. The bottom line
is that the total load of a rectangular area can be modeled at the centroid of
the rectangle as shown in Figure 3.13. It must be understood that in Figure 3.13,
with the load modeled at the centroid, the voltage drop computed to the
load point will represent the total voltage drop from node n to node m.
I
T
D l w ⋅ ⋅
3 kV
LL
⋅
/ cos
1 –
PF ( ) –
di
I
T
l
 A/mile
i I
T
x di ⋅ – I
T
x
I
T
l
 ⋅ – I
T
1
x
l
 –
( J
 ·
⋅
dV Re z i dx ⋅ ⋅ ( ) Re z I
T
1
x
l
 –
( J
 ·
dx ⋅ ⋅ ⋅
V
drop
dV
0
l
∫
Re z I
T
1
x
l
 –
( J
 ·
0
l
∫
dx ⋅ ⋅ ⋅
V
drop
Re z I
T
1
2
 l ⋅ ⋅ ⋅
( J
 ·
Re
1
2
 Z I
T
⋅ ⋅
0812_frame_C03.fm Page 56 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 57
A similar derivation can be done in order to determine the total three
phase power loss down the feeder main. The power loss in the incremental
length is
The total threephase power loss down the primary main is
Evaluating the integral and simplifying:
(3.44)
where R r ⋅ l
Equation 3.44 gives the same result as that of Equation 3.26. The only dif
ference, again, is the manner in which the total current I
T
is determined. The
model for computing the total threephase power loss of the primary main
feeder is shown in Figure 3.14. Once again, it must be understood that the
power loss computed using the model of Figure 3.14 represents the total
power loss from node n to node m.
FIGURE 3.13
Rectangle voltage drop model.
n
l
w
1
2
l
I
T
I
T
m
dp 3 i
2
r dx ⋅ ⋅ ⋅ 3 I
T
2
1
x
l
 –
( J
 ·
2
r dx ⋅ ⋅ ⋅ ⋅
3 r I
T
2
1 2
x
l

x
2
l
2
 + ⋅ –
( J
 ·
dx ⋅ ⋅ ⋅ ⋅
P
loss
dp
0
l
∫
3 r I
T
2
1 2
x
l

x
2
l
2
 + ⋅ –
( J
 ·
0
l
∫
dx ⋅ ⋅ ⋅ ⋅
P
loss
3
1
3
 r l I
T
2
⋅ ⋅ ⋅ ⋅ 3
1
3
 R I
T
2
⋅ ⋅ ⋅
0812_frame_C03.fm Page 57 Saturday, July 21, 2001 3:00 PM
58 Distribution System Modeling and Analysis
Example 3.6
It is proposed to serve a rectangular area of length 10,000 ft. and width of
6000 ft. The load density of the area is 2500 kVA/mile
2
with a power factor
of 0.9 lagging. The primary main feeder uses 336,400 26/7 ACSR on a pole
conﬁgured as shown in Example 3.2, Figure 3.3. The question at hand is
what minimum standard nominal voltage level can be used to serve this
area without exceeding a voltage drop of 3% down the primary main? The
choices of nominal voltages are 4.16 kV and 12.47 kV. Compute, also, the
total threephase power loss.
The area to be served is shown in Figure 3.15. From Example 3.2, the
impedance of the line was computed to be
z 0.306 + j0.6272 Ω/mile
The length and width of the area in miles are
The total area of the rectangular area is
A l ⋅ w 2.1522 miles
2
The total load of the area is
kVA D ⋅ A 2500 ⋅ 2.1522 5380.6 kVA
The total impedance of the line segment is
For a nominal voltage of 4.16 kV, the total area current is
FIGURE 3.14
Rectangle power loss model.
n
l
w
I
T
m
I
T
1
3
l
l
10,000
5280
 1.8939 miles and w
6000
5280
 1.1364 miles
Z z l ⋅ 0.306 j0.6272 + ( ) 1.8939 ⋅ 0.5795 j1.1879 Ω +
I
T
kVA
3 kV
LL
⋅

5380.6
3 4.16 ⋅
/ cos
1 –
0.9 ( ) – 746.7/ 25.84 – A
0812_frame_C03.fm Page 58 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 59
The total voltage drop down the primary main is
The nominal linetoneutral voltage is
The percent voltage drop is
It is clear that the nominal voltage of 4.16 kV will not meet the criteria of a
voltage drop less than 3.0%. For a nominal voltage of 12.47 kV, the total area
current is
The total voltage drop down the primary main is
FIGURE 3.15
Example 3.8 rectangular area.
6,000'
10,000'
T
I
V
drop
Re
1
2
 Z I
T
⋅ ⋅ Re
1
2
 0.5795 j1.1879 + ( ) 746.7/ 25.84 – ⋅ ⋅
388.1 V
V
LN
4160
3
 2401.8 V
V
%
V
drop
V
LN
 100% ⋅
388.1
2401.8
 100% ⋅ 16.16%
I
T
kVA
3 kV
LL
⋅

5380.6
3 12.47 ⋅
/ cos
1 –
0.9 ( ) – 249.1/ 25.84 – A
V
drop
Re
1
2
 Z I
T
⋅ ⋅ Re
1
2
 0.5795 j1.1879 + ( ) 249.1/ 25.84 – ⋅ ⋅
129.5 V
0812_frame_C03.fm Page 59 Saturday, July 21, 2001 3:00 PM
60 Distribution System Modeling and Analysis
The nominal linetoneutral voltage is
The percent voltage drop is
The nominal voltage of 12.47 kV is more than adequate to serve this load.
It would be possible at this point to determine how much larger the area
could be and still satisfy the 3.0% voltage drop constraint.
For the 12.47 kV selection, the total threephase power loss down the primary
main is
3.5.2 The Triangle
Figure 3.16 represents a triangular area of constant load density being served
by a threephase main running from node n to node m. It is desired to
determine the total voltage drop and the total threephase power loss down
the primary main from node n to node m.
FIGURE 3.16
Constant load density triangular area.
V
LN
12,470
3
 7199.6 V
V
%
V
drop
V
LN
 100% ⋅
129.5
7199.6
 100% ⋅ 1.80%
P
loss
3
1
3

R I
T
2
⋅ ⋅
1000
 ⋅ 3
1
3

0.5795 249.1
2
⋅ ⋅
1000
 ⋅ 35.965 kW
w
dx
x
l
.5di
.5di
I
T
n
m
1
w
0812_frame_C03.fm Page 60 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 61
The area of the triangle is
(3.45)
The total current entering the area is given by:
(3.46)
Let (3.47)
The current entering the incremental line segment is
(3.48)
where A
1
area of triangle up to the incremental line segment.
By similar triangles, (3.49)
The area of the small triangle up to the incremental line segment is
(3.50)
Substituting Equations 3.47 and 3.50 into Equation 3.48:
(3.51)
The voltage drop in the incremental line segment is given by:
(3.52)
The total voltage drop from node n to node m is
Area
1
2
 l w ⋅ ⋅
I
T
D Area ⋅
3 kV
LL
⋅
/ cos
1 –
PF ( ) – A
di
I
T
Area

I
T
1
2
 l w ⋅ ⋅

2 I
T
⋅
l w ⋅
 A/mile
2
i I
T
A
1
di ⋅ –
w
1
x
w
l
 ⋅
A
1
1
2
 x w
1
⋅ ⋅
1
2
 x x
w
l
 ⋅
( J
 ·
⋅ ⋅
1
2

w
l
 x
2
⋅ ⋅
i I
T
1
2

w
l
 x
2
⋅ ⋅
( J
 ·
2
l w ⋅
 I
T
⋅
( J
 ·
⋅ – I
T
1
x
2
l
2
 –
( J
 ·
⋅
dv Re i z dx ⋅ ⋅ [ ] Re z I
T
1
x
2
l
2
 –
( J
 ·
dx ⋅ ⋅ ⋅
V
drop
dv
0
l
∫
Re z I
T
1
x
2
l
2
 –
( J
 ·
dx ⋅
0
l
∫
⋅ ⋅
0812_frame_C03.fm Page 61 Saturday, July 21, 2001 3:00 PM
62 Distribution System Modeling and Analysis
Evaluating the integral and simplifying:
(3.53)
where Z
T
z ⋅ l
Equation 3.53 shows that the total voltage drop from the vertex to the base
of the triangular area can be computed by modeling the total triangle load
twothirds of the distance between the vertex and the base of the triangle.
The model for the voltage drop calculation is shown in Figure 3.17. A similar
derivation can be made for the power loss model. The power loss in the
incremental line segment is
(3.54)
Substituting Equation 3.51 into Equation 3.54:
The total threephase power loss from node n to node m becomes:
FIGURE 3.17
Triangle voltage drop model.
w
l
I
T
n
m
I
T
2
3
l
V
drop
Re z I
T
2
3
 l ⋅ ⋅ ⋅ Re
2
3
 Z
T
I
T
⋅ ⋅
dp 3 r i
2
dx ⋅ ⋅ [ ] ⋅
dp 3 r I
T
2
1
x
2
l
2
 –
( J
 ·
2
dx ⋅ ⋅ ⋅ ⋅ 3 r I
T
2
1 2
x
2
l
2

x
4
l
4
 + ⋅ –
( J
 ·
dx ⋅ ⋅ ⋅ ⋅
P
loss
dp
0
l
∫
3 r I
T
2
1 2
x
2
l
2

x
4
l
4
 + ⋅ –
( J
 ·
dx ⋅
0
l
∫
⋅ ⋅ ⋅
0812_frame_C03.fm Page 62 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 63
Evaluating the integral and simplifying:
(3.55)
Equation 3.55 gives the total threephase power loss down the primary main
from node n to node m. The model for the power loss is given in Figure 3.18.
Example 3.7
The triangular area shown in Figure 3.19 is to be served by a feeder of nominal
voltage 12.47 kV. The load density of the area is 3500 kVA/mile
2
at a power
factor of 0.9 lagging. The conductor on the primary main is 336,400 26/7 ACSR,
and the conﬁguration of the pole is that of Example 3.2 in Figure 3.3.
Use the K
drop
factor from the line of Example 3.2 and determine the percent
voltage drop from node n to node m.
From Example 3.3 the K
drop
factor was computed to be
The length and width of the triangle in miles is
The area of the triangle is
FIGURE 3.18
Triangle power loss model.
w
l
I
T
n
m
15
l
I
T
8
P
loss
3
8
15
 R I
T
2
⋅ ⋅ ⋅
K
drop
0.00035291% drop/kVAmile
l
15,000
5280
 2.8409 miles and w
6000
5280
 1.1364 miles
Area
1
2
 2.8509 1.1364 ⋅ ⋅ 1.6142 miles
2
0812_frame_C03.fm Page 63 Saturday, July 21, 2001 3:00 PM
64 Distribution System Modeling and Analysis
The total load of the triangular area is
kVA 3500 ⋅ 1.6142 5649.5 kVA
The total complex power of the triangular area is
Using the K
drop
factor and lumping the total load at the twothirds point, the
percent drop to node m is
Suppose now that a shunt capacitor bank is to be installed somewhere along
the primary main in order to limit the percent voltage drop to node m to
3.0%. Two decisions must be made:
1. Threephase rating of the capacitor bank
2. Location of the capacitor bank
The total reactive power of the area was computed to be 2462.6 kvar. That
means that a capacitor bank rated up to 2462.6 can be used without causing
the feeder to go into a leading power factor condition. Since this is assumed
FIGURE 3.19
Example 3.9 triangular area.
I
T
n
m
15,000'
6,000'
S kVA / cos
1 –
PF ( ) – 5649.5/ 25.84 – kVA
5084.6 j2462.6 +
V
drop
2
3
 K
drop
kVA miles ⋅ ⋅ ⋅
2
3
 0.00035291 5649.5 2.8409 ⋅ ⋅ ⋅ 3.7761%
0812_frame_C03.fm Page 64 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 65
to be the peak load, a capacitor bank rated at 1800 kvar (threephase) will
be used in order to prevent a leading power factor condition for a smaller
load. Depending upon the load curve during the day, this bank may or may
not have to be switched.
Use the K
rise
factor from Example 3.5 and determine how far from node n
the capacitor bank should be installed in order to limit the voltage drop to
3.0%. From Example 3.5:
The needed voltage rise due to the capacitor is
The distance from node n is determined by:
The total threephase power loss down the primary main before the shunt
capacitor is added is computed by lumping the total triangular load at:
The total load current is
The total resistance of the primary main is
The total threephase power loss down the primary main is
3.5.3 The Trapezoid
The ﬁnal geometric conﬁguration to consider is the trapezoid. As before, it
is assumed that the load density is constant throughout the trapezoid. The
general model of the trapezoid is shown in Figure 3.20.
K
rise
0.00040331% rise/kvarmile
V
rise
V
drop
3.0 – 3.7761 3.0 – 0.7761
dist
V
rise
K
rise
kvar ⋅

0.7761
0.00040331 1800 ⋅
 1.0691 miles
l
Load
8
15
 l ⋅ 1.5151 miles from node n
I
T
kVA
3 kV
LL
⋅

5649.5
3 12.47 ⋅
/ cos
1 –
PF ( ) – 261.6/ 25.84 – A
R r l ⋅ 0.306 2.8409 ⋅ 0.8693 Ω
P
loss
3
1000

8
15
 R I
T
2
⋅ ⋅ ⋅
3
1000

8
15
 0.8693 261.6
2
⋅ ⋅ ⋅ 95.16 kW
0812_frame_C03.fm Page 65 Saturday, July 21, 2001 3:00 PM
66 Distribution System Modeling and Analysis
Figure 3.20 represents a trapezoidal area of constant load density being
served by a threephase primary running from node n to node m. It is desired
to determine the total voltage drop and the total threephase power loss down
the primary main from node n to node m.
It is necessary to determine the value of the current entering the incremen
tal line segment as a function of the total current and the known dimensions
of the trapezoid. The known dimensions will be the length l and the widths
w
1
and w
2
.
The total current entering the trapezoid is
(3.56)
where Area
T
total area of the trapezoid.
(3.57)
The current that is delivered to the trapezoid abef is
(3.58)
where Area
x
area of the trapezoid abef.
(3.59)
Solving Equation 3.56 for D:
(3.60)
FIGURE 3.20
General trapezoid.
l
w
dx
m
x
i
I
T
w
w
1
x
2
.5 di
.5 di
n
a
b
c
d
e
f
I
T
D Area
T
⋅
3 kV
LL
⋅

Area
T
1
2
 w
2
w
1
+ ( ) l ⋅ ⋅
I
x
D Area
x
⋅
3 kV
LL
⋅

Area
x
1
2
 w
x
w
1
+ ( ) x ⋅ ⋅
D
3 kV
LL
I
T
⋅ ⋅
Area
T

0812_frame_C03.fm Page 66 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 67
Substitute Equation 3.60 into Equation 3.58:
(3.61)
The current entering the incremental line segment is
(3.62)
The only problem at this point is that the area of the small trapezoid cannot
be determined since the width w
x
is not known. Figure 3.21 will be used to
establish the relationship between the unknown width and the known
dimensions. Referring to Figure 3.21:
(3.63)
From similar triangles:
(3.64)
But
(3.65)
FIGURE 3.21
Trapezoid dimensions.
l
w
x
2
w
1
y
x
i
w
x
y
2
a
b
d
e
f
I
x
3 kV
LL
I
T
⋅ ⋅
Area
T

( J
 ·
Area
x
3 kV
LL
⋅

( J
 ·
⋅
Area
x
Area
T
 I
T
⋅
i I
T
I
x
– I
T
1
Area
x
Area
T
 –
( J
 ·
⋅
w
x
w
1
2 y
x
⋅ +
y
x
x
l
 y
2
⋅
y
2
1
2
 w
2
w
1
– ( ) ⋅
0812_frame_C03.fm Page 67 Saturday, July 21, 2001 3:00 PM
68 Distribution System Modeling and Analysis
Substitute Equation 3.65 into Equation 3.64:
(3.66)
Substitute Equation 3.66 into Equation 3.63:
(3.67)
Substitute Equation 3.67 into Equation 3.59:
(3.68)
Substitute Equations 3.57 and 3.68 into Equation 3.62:
(3.69)
The current entering the incremental line segment of Figure 3.20 is given in
Equation 3.69 and will be used to compute the voltage drop and power loss in
the incremental line segment. The voltage drop in the incremental line segment
is given by:
(3.70)
Substitute Equation 3.69 into Equation 3.70:
(3.71)
y
x
x
l

1
2
 w
2
w
1
– ( ) ⋅ ⋅
w
x
w
1
2
x
l

1
2
 ⋅ ⋅ + w
2
w
1
– ( ) ⋅ w
1
x
l
 w
2
w
1
– ( ) ⋅ + w
1
1
x
l
 –
( J
 ·
x
l
 w
2
⋅ + ⋅
Area
x
1
2
 w
1
1
x
l
 –
( J
 ·
x
l
 w
2
⋅ + ⋅
( J
 ·
w
1
+ x ⋅ ⋅
i I
T
1
1
2
 w
1
2
x
l
 – ( )
x
l
 w
2
⋅ + ⋅ ( ) [ ] x ⋅ ⋅
1
2
 w
2
w
1
+ ( ) l ⋅ ⋅
 – ⋅
i
I
T
w
1
w
2
+ ( ) l ⋅
 w
1
w
2
+ ( ) l ⋅ 2 w
1
x ⋅ ⋅ w
1
x
2
l
 w
2
x
2
l
 ⋅ – ⋅ + – ⋅
i
I
T
w
1
w
2
+ ( ) l ⋅
 l 2 x
x
2
l
 + ⋅ –
( J
 ·
w
1
⋅ l
x
2
l
 –
( J
 ·
w
2
⋅ + ⋅
dv Re z i dx ⋅ ⋅ [ ]
dv Re z
I
T
w
1
w
2
+ ( ) l ⋅
 l 2 x
x
2
l
 + ⋅ –
( J
 ·
w
1
⋅ l
x
2
l
 –
( J
 ·
w
2
⋅ + dx ⋅ ⋅ ⋅
 
¦ 
 
0812_frame_C03.fm Page 68 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 69
The total voltage drop down the primary from node n to node m is given by:
Evaluating the integral and simplifying results in:
(3.72)
Equation 3.72 is very general and can be used to determine the models for
the rectangular and triangular areas.
THE RECTANGLE
For a rectangular area the two widths w
1
and w
2
will be equal.
Let (3.73)
Substitute Equation 3.73 into Equation 3.72:
(3.74)
Equation 3.74 is the same that was initially derived for the rectangular area.
THE TRIANGLE
For a triangular area the width w
1
will be zero.
Let (3.75)
Substitute Equation 3.75 into Equation 3.72:
(3.76)
Equation 3.76 is the same as was derived for the triangular area.
The total threephase power loss down the line segment can be developed
by starting with the derived current in the incremental segment as given by
V
drop
v d
0
l
∫
Re
z I
T
⋅
w
1
w
2
+ ( ) l ⋅
 l 2 x
x
2
l
 + ⋅ –
( J
 ·
w
1
l
x
2
l
 –
( J
 ·
w
2
⋅ + ⋅ x d ⋅
0
l
∫
⋅
 
¦ 
 
V
drop
Re Z I
T
w
1
2 w
2
⋅ +
3 w
1
w
2
+ ( ) ⋅

( J
 ·
⋅ ⋅
w
1
w
2
w
V
drop
Re Z I
T
w 2 w ⋅ +
3 w w + ( ) ⋅

( J
 ·
⋅ ⋅ Re Z I
T
3 w ⋅
6 w ⋅
 ⋅ ⋅
V
drop
Re
1
2
 Z I
T
⋅ ⋅
w
1
0
V
drop
Re Z I
T
0
1
2 w
2
⋅ +
3 0 w
2
+ ( ) ⋅

( J
 ·
⋅ ⋅ Re
2
3
 Z I
T
⋅ ⋅
0812_frame_C03.fm Page 69 Saturday, July 21, 2001 3:00 PM
70 Distribution System Modeling and Analysis
Equation 3.69. The threephase power loss in the incremental segment is
(3.77)
The total threephase power loss down the line segment is then:
(3.78)
Substitute Equation 3.69 into Equation 3.78 and simplify:
(3.79)
Evaluating the integral and simplifying results in:
(3.80)
where R r · l
The rectangular and triangular areas are special cases of Equation 3.80.
RECTANGLE
For the rectangle, the two widths w
1
and w
2
are equal.
Let
Substitute into Equation 3.79:
(3.81)
Equation 3.81 is the same as the Equation 3.44 that was previously derived
for the rectangular area.
TRIANGLE
For the triangular area, the width w
1
is zero.
Let
dp 3 r i
2
dx ⋅ ⋅
P
loss
3 r i
2
dx
0
l
∫
⋅ ⋅
P
loss
3
r I
T
2
⋅
w
1
w
2
+ ( )
2
l
2
⋅
 l 2 x
x
2
l
 + ⋅ –
( J
 ·
w
1
l
x
2
l
 –
( J
 ·
w
2
⋅ + ⋅
2
dx
0
l
∫
⋅ ⋅
P
loss
3 R I
T
2
8 w
2
2
9 w
1
w
2
3 w
1
2
⋅ + ⋅ ⋅ + ⋅
15 w
1
w
2
+ ( )
2
⋅
 ⋅ ⋅
 
¦ 
 
⋅
w w
1
w
2
P
loss
3 R I
T
2 8 w
2
9 w w 3 w
2
⋅ + ⋅ ⋅ + ⋅
15 w w + ( )
2
⋅
 ⋅ ⋅
 
¦ 
 
⋅ 3 R I
T
2 8 9 3 + +
15 2 ( )
2
⋅
 ⋅ ⋅
 
¦ 
 
⋅
P
loss
3
1
3
 R I
T
2
⋅ ⋅ ⋅
w
1
0
0812_frame_C03.fm Page 70 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 71
Substitute into Equation 3.80:
(3.82)
Equation 3.82 is the same as Equation 3.55, which was previously derived
for the total power loss in a triangular area.
3.6 Summary
This chapter has been devoted to the development of some useful techniques
for computing the voltage drop and power loss of line segments with uni
formly distributed loads, and for geometric areas with constant load densities.
These techniques are very useful for making quick calculations that will be
ballpark values. Many times only a ballpark value is needed. More often than
not, once inside the ballpark more precise values of voltage drop and power
loss are needed. This will be especially true when the unbalanced nature of
a distribution feeder is taken into account. The remainder of this text will be
devoted to the more precise methods for analyzing a distribution feeder
under balanced and unbalanced, steadystate, and shortcircuit conditions.
References
1. Glover, J.D. and Sarma, M., Power System Analysis and Design, 2nd edition, PWS
Publishing Co., Boston, 1994.
Problems
3.1 Shown in Figure 3.22 is the pole conﬁguration of conductors for a three
phase primary feeder. The conductors are 250,000 cm, CON Lay, AA. The
nominal linetoline voltage of the feeder is 14.4 kV.
1. Determine the series impedance per mile of this line.
2. Determine the K
drop
factor assuming a power factor of 0.88 lag.
3. Determine the K
rise
factor.
P
loss
3 R I
T
2
8 w
2
2
9 0 w
2
3 0
2
⋅ + ⋅ ⋅ + ⋅
15 0 w
2
+ ( )
2
⋅
 ⋅ ⋅
 
¦ 
 
⋅ 3
8
15
 R I
T
2
⋅ ⋅ ⋅
0812_frame_C03.fm Page 71 Saturday, July 21, 2001 3:00 PM
72 Distribution System Modeling and Analysis
3.2 A 4.16 threephase primary feeder is shown in Figure 3.23.
The K
drop
0.00298639% drop/kVAmile
The K
rise
0.00334353% rise/kvarmile
1. Determine the percent voltage drop to node E4.
2. Determine the rating of a threephase shunt capacitor bank to be
placed at E3 to limit the voltage drop to E4 to 5.0%.
3.3 A 4160V, threephase feeder is shown in Figure 3.24.
The phase conductors are 4/0 ACSR and are conﬁgured on an 8ft. crossarm
with phase spacings of: 2.5 ft., 4.5 ft., and 7.0 ft.
1. Determine the series impedance of the line segment in Ω/mile.
2. Determine the K
drop
and K
rise
factors assuming a load power factor
of 0.9 lagging.
3. Determine the total percent voltage drop to node 6.
FIGURE 3.22
Problem 3.1 conﬁguration.
FIGURE 3.23
System for Problem 3.2.
FIGURE 3.24
System for Problem 3.3.
a
b
c
n
2'
2'
2'
25'
E2 E3
E4
500 kVA
1200 kVA
750 kVA
E1
0.50 miles
0.65 miles
0.9 miles
0.15 mile 0.175 mile 0.2 mile 0.125 mile 0.225 mile 0.125 mile
0 1 2 3 4 5 6
200 kVA 150 kVA 100 kVA 300 kVA 425 kVA 500 kVA
D
ab
D
bc
D
ca
0812_frame_C03.fm Page 72 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 73
4. Determine the threephase kvar rating of a shunt capacitor to be
placed at node 4 to limit the total percent voltage drop to node 6
to 3.0%.
3.4 Flash Thunderbolt, junior engineer for Tortugas Power and Light, has
been given an assignment to design a new 4.16kV, threephase feeder that
will have the following characteristics:
Total length of feeder 5000 ft.
Load: 10–500 kVA (threephase), 0.9 lagging power spaced every
500 ft. with the ﬁrst load 500 ft. from the substation.
Voltage drop: not to exceed 5% from the sub to the last load.
Figure 3.25 illustrates the new feeder.
Flash has decided that he will use 336,400 26/7 ACSR (Linnet) conductors
constructed on 45ft. poles with 8ft. crossarms. The spacings of the conductors
on the crossarms are 2.5 ft., 4.5 ft., and 7.0 ft.
1. Determine the percent voltage drop to the last load point and the
total threephase power loss for the feeder shown in Figure 3.25.
2. Lump the total feeder load at the midpoint of the feeder and com
pute the percent voltage drop to the end of the feeder.
3. Use the exact lumped load model of Figure 3.11 and compute the
percent voltage drop to the end of the line, and the total three
phase power loss down the line.
3.5 The rectangular area in Figure 3.26 has a uniform load density of 2000
kVA/mile
2
at 0.9 lagging power factor. The nominal voltage of the area being
served is 4.16 kV. The threephase primary main conductors are 556,500 26/7
ACSR, while the threephase lateral conductors are 266,800 26/7 ACSR. The
primary main and the laterals are constructed so that the equivalent spacing
(D
eq
) is 3.5 ft. Determine:
1. The % voltage drop to the last customer in the ﬁrst lateral (point A).
2. The % voltage drop to the last customer in the last lateral (point B).
3. The total threephase power loss for the total area.
FIGURE 3.25
System for Problem 3.4.
500' 500' 500' 500' 500' 500' 500' 500' 500' 500'
SUB
0812_frame_C03.fm Page 73 Saturday, July 21, 2001 3:00 PM
74 Distribution System Modeling and Analysis
3.6 Shown in Figure 3.27 is a rectangletriangle area that is being fed from
a source at point X. Both areas have a load density of 6000 kVA/mile
2
, with
loads being uniformly distributed as denoted by the dashed laterals. In addi
tion to the uniformly distributed loads, there is a “spot load” at point Z that
is 2000 kVA. The K
drop
factor for the primary main conductors is 0.00022626%
drop/kVAmile, and the K
rise
factor for the primary main conductors is
0.00028436% rise/kvarmile.
1. Determine the percent drop to point Z.
2. Determine the kVAr rating (to the nearest 300 kVAr/phase) for a
capacitor bank to be placed at point Y in order to limit the voltage
drop to Z to 3%.
3. With the capacitor in place, what now is the percent drop to point Z?
3.7 A square area of 20,000 ft. on a side has a load density of 2000 kVA/
mile
2
, and 0.9 lagging power factor is to be served from a 12.47kV substation
that is located in the center of the square. Two different plans are being
considered for serving the area. The two plans are shown in Figure 3.28.
PlanA proposes to break the area into four square areas and serve it as
shown. The big black line will be the threephase primary main consisting
FIGURE 3.26
Rectangular area for Problem 3.5.
FIGURE 3.27
Rectangulartriangular area of Problem 3.6.
12,000'
2,500'
2,500'
Source
B
A
X
Y Z
1.5 mile
1.5 mile
1 mile
2000 kVA
0812_frame_C03.fm Page 74 Saturday, July 21, 2001 3:00 PM
Approximate Methods of Analysis 75
of 336/400 26/7 ACSR conductors, and the dotted lines will be the three
phase laterals consisting of 4/0 ACSR conductors. Both the main and laterals
are constructed such that D
eq
4.3795 ft. The threephase laterals will be
spaced every 500 ft.
PlanB proposes to serve the area with four triangularly shaped feeders.
Again, the primary main is shown in the dark black line, and the laterals are
spaced every 500 ft. and are shown as the dotted lines. The same conductors
and D
eq
will be used in this plan. Determine the percent voltage drop to the
“last customer” (points A and B) for the two plans.
3.8 Shown in Figure 3.29 are the areas normally served by two feeders.
Under an emergency condition the switch at b is closed so that the feeder
normally serving the triangle area must now serve both areas. Assume both
FIGURE 3.28
Two plans for Problem 3.7.
FIGURE 3.29
Areas for Problem 3.8.
A
B
Plan  A Plan  B
1.5 mi
0.5 mile
0.5 mile
0.5 mile
0.5 mile
S
a
e
b
c
d
S'
0.75 mi
0812_frame_C03.fm Page 75 Saturday, July 21, 2001 3:00 PM
76 Distribution System Modeling and Analysis
areas have a uniform load density of 2.5 MVA/square mile, and 0.9 lagging
power factor. The primary feeder voltage is 13.8 kV. Laterals are uniformly
tapped off of the primary main from S to a. No loads are tapped off the feed
from a to b to c, and laterals are tapped off from c to d and from c to S’. The
primary main conductors are 2/0 ACSR and are placed on a pole such that
D
eq
4.3795 ft.
1. Determine the K
drop
and K
rise
factors.
2. Determine the voltage drop to point d.
3. Determine the threephase kVAr rating of a shunt capacitor bank
placed at c in order to limit the voltage drop to point d to 3.0%.
4. Determine the voltage drop to e with the capacitor bank at c.
5. Determine the voltage drop to e with the source at S’ and the
capacitor at c.
0812_frame_C03.fm Page 76 Saturday, July 21, 2001 3:00 PM
77
4
Series Impedance of Overhead
and Underground Lines
The determination of the series impedance for overhead and underground
lines is a critical step before the analysis of a distribution feeder can begin.
The series impedance of a singlephase, twophase (Vphase), or threephase
distribution line consists of the resistance of the conductors and the self and
mutual inductive reactances resulting from the magnetic ﬁelds surrounding
the conductors. The resistance component for the conductors will typically
come from a table of conductor data such as found in Appendix A.
4.1 Series Impedance of Overhead Lines
The inductive reactance (self and mutual) component of the impedance is a
function of the total magnetic ﬁelds surrounding a conductor. Figure 4.1
shows conductors 1 through
n
with the magnetic ﬂux lines created by currents
ﬂowing in each of the conductors. The currents in all conductors are assumed
to be ﬂowing out of the page. It is further assumed that the sum of the currents
will be zero. That is
I
1
+
I
2
+
· ·
I
i
+
· ·
I
n
=
0 (4.1)
The total ﬂux linking conductor
i
is given by:
Wbt/m
(4.2)
where
D
i n
=
Distance between conductor
i
and conductor
n
(ft.)
GMR
i
=
Geometric Mean Radius of conductor
i
(ft.)
λ
i
2 10
7 –
I
1
1
D
i1
 ln ⋅ I
2
1
D
i2

· ·
I
i
1
GMR
i
 ln
· ·
I
n
1
D
in
 ln ⋅ + ⋅ + ln ⋅ +
⋅ ⋅ =
0812_frame_C04.fm Page 77 Friday, July 20, 2001 2:55 PM
78
Distribution System Modeling and Analysis
The inductance of conductor
i
consists of the self inductance of conductor
i
and the mutual inductance between conductor
i
and all of the other
n
−
1
conductors. By deﬁnition:
H/m (4.3)
H/m (4.4)
4.1.1 Transposed ThreePhase Lines
Highvoltage transmission lines are usually assumed to be transposed (each
phase occupies the same physical position on the structure for onethird of
the length of the line). In addition to the assumption of transposition, it is
assumed that the phases are equally loaded (balanced loading). With these
two assumptions it is possible to combine the “self” and “mutual” terms
into one “phase” inductance.
1
H/m (4.5)
where (4.6)
FIGURE 4.1
Magnetic ﬁelds.
D
in
2
D
i2
D
i1
O
1
O
n
O
2
O
i
O
n
O
2
O
1
1
i
n
Self inductance: L
ii
λ
ii
I
i
 2 10
7 – 1
GMR
i
 ln ⋅ ⋅ = =
Mutual inductance: L
in
λ
in
I
n
 2 10
7 – 1
D
in
 ln ⋅ ⋅ = =
Phase inductance: L
i
2 10
7 –
D
eq
GMR
i
 ln ⋅ ⋅ =
D
eq
D
ab
D
bc
D
ca
⋅ ⋅
3
= ft.
D
ab
, D
bc
, and D
ca
distance between phases =
0812_frame_C04.fm Page 78 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines
79
Assuming a frequency of 60 Hz, the phase inductive reactance is given by:
Ω
/mile (4.7)
The series impedance per phase of a transposed threephase line consisting
of one conductor per phase is given by:
Ω
/mile (4.8)
4.1.2 Untransposed Distribution Lines
Because distribution systems consist of singlephase, twophase, and untrans
posed threephase lines serving unbalanced loads, it is necessary to retain
the identity of the self and mutual impedance terms of the conductors and
take into account the ground return path for the unbalanced currents. The
ac resistance of the conductors is taken directly from a table of conductor data
(Appendix A). Equations 4.3 and 4.4 are used to compute the self and mutual
inductive reactances of the conductors. The inductive reactance will be
assumed to be at a frequency of 60 Hz, and the length of the conductor will
be assumed to be one mile. With those assumptions the self and mutual
impedances are given by:
Ω
/mile (4.9)
Ω
/mile (4.10)
In 1926 John Carson published a paper in which he developed a set of equa
tions for computing the self and mutual impedances of lines, taking into
account the return path of current through ground.
2
Carson’s approach was
to represent a line with the conductors connected to a source at one end and
grounded at the remote end. Figure 4.2 illustrates a line consisting of two
conductors (
i
and
j
) carrying currents (
I
i
and
I
j
) with the remote ends of the
conductors tied to ground. A ﬁctitious “dirt” conductor carrying current
I
d
is
used to represent the return path for the currents. In Figure 4.2, Kirchhoff’s
voltage law (KVL) is used to write the equation for the voltage between
conductor
i
and ground.
(4.11)
Collect terms in Equation 4.11:
(4.12)
Phase reactance: x
i
ω L
i
⋅ 0.12134
D
eq
GMR
i
 ln ⋅ = =
Series impedance: z
i
r
i
j 0.12134
D
eq
GMR
i
 ln ⋅ ⋅ + =
z
ii
r
i
j0.12134
1
GMR
i
 ln ⋅ + =
z
ij
j0.12134
1
D
ij
 ln ⋅ =
V
ig
z
ii
I
i
z
ij
I
j
z
id
I
d
z
dd
I
d
z
di
I
i
z
dj
I
j
⋅ + ⋅ + ⋅ ( ) – ⋅ + ⋅ + ⋅ =
V
ig
z
ii
z
di
– ( ) I
i
z
ij
z
dj
– ( ) I
j
z
id
z
dd
– ( ) I
d
⋅ + ⋅ + ⋅ =
0812_frame_C04.fm Page 79 Friday, July 20, 2001 2:55 PM
80 Distribution System Modeling and Analysis
From Kirchhoff’s current law:
(4.13)
Substitute Equation 4.13 into Equation 4.12 and collect terms:
(4.14)
Equation 4.14 is of the general form:
(4.15)
where (4.16)
(4.17)
In Equations 4.16 and 4.17, the “over bar” impedances are given by Equations
4.9 and 4.10. Note that in these two equations the effect of the ground return
path is being folded into what will now be referred to as the “primitive” self
and mutual impedances of the line. The equivalent primitive circuit is shown
in Figure 4.3. Substituting Equations 4.9 and 4.10 into Equations 4.16 and
4.17, the primitive self impedance is given by:
(4.18)
FIGURE 4.2
Two conductors with dirt return path.
I
i
I
d
I
j
+

V
ig
+

V
jg
ground
z
ii
z
jj
z
ij
z
id
z
jd z
dd
I
i
I
j
I
d
+ + 0 =
I
d
I
i
I
j
– – =
V
ig
z
ii
z
dd
z
di
z
id
– – + ( ) I
i
z
ij
z
dd
z
dj
z
id
– – + ( ) I
j
⋅ + ⋅ =
V
ig
zˆ
ii
I
i
zˆ
ij
I
j
⋅ + ⋅ =
zˆ
ii
z
ii
z
dd
z
di
– z
id
– + =
zˆ
ij
z
ij
z
dd
z
dj
z
id
– – + =
zˆ
ii
r
i
jx
ii
r
d
jx
dd
jx
id
jx
di
– – + + + =
zˆ
ii
r
i
r
d
j0.12134
1
GMR
i
 ln
1
GMR
d
 ln
1
D
id
 ln
1
D
di
 ln – – +
⋅ + + =
zˆ
ii
r
i
r
d
j0.12134
1
GMR
i
 ln
D
id
D
di
⋅
GMR
d
 ln +
⋅ + + =
0812_frame_C04.fm Page 80 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines
81
In similar manner, the primitive mutual impedance can be expanded:
(4.19)
The obvious problem in using Equations 4.18 and 4.19 is that we do not know
the values of the resistance of dirt (
r
d
), the Geometric Mean Radius of dirt
(GMR
d
), and the distances from the conductors to dirt (
D
id
,
D
di
,
D
jd
,
D
dj
). This
is where John Carson’s work bails us out.
4.1.3 Carson’s Equations
Since a distribution feeder is inherently unbalanced, the most accurate analysis
should not make any assumptions regarding the spacing between conductors,
conductor sizes, and transposition. In Carson’s 1926 paper he developed a
technique whereby the self and mutual impedances for an arbitrary number
of overhead conductors can be determined. The equations can also be applied
to underground cables. The technique was not met with a lot of enthusiasm
because of the tedious calculations that would have to be done on the slide
rule and by hand. With the advent of the digital computer, Carson’s equations
have become widely used.
In his paper, Carson assumes the earth is an inﬁnite, uniform solid with a
ﬂat uniform upper surface and a constant resistivity. Any end effects intro
duced at the neutral grounding points are not large at power frequencies, and
are therefore neglected.
Carson made use of conductor images; that is, every conductor at a given
distance above ground has an image conductor the same distance below ground.
FIGURE 4.3
Equivalent primitive circuit.
I
i
I
j
+

V
ig
+

V
jg
ground
z
ii
z
jj
z
ij
V'
ig
+
+
V'
jg
 
zˆ
ij
jx
ij
r
d
jx
dd
jx
dj
jx
id
– – + + =
zˆ
ij
r
d
j0.12134
1
D
ij
 ln
1
GMR
d

1
D
dj
 ln
1
D
id
 ln – – ln +
⋅ + =
zˆ
ij
r
d
j0.12134
1
D
ij
 ln
D
dj
D
id
⋅
GMR
d
 ln +
+ =
0812_frame_C04.fm Page 81 Monday, October 28, 2002 9:49 AM
82 Distribution System Modeling and Analysis
This is illustrated in Figure 4.4. Referring to Figure 4.4, the original Carson
equations are given in Equations 4.20 and 4.21.
Self Impedance of Conductor i:
Ω/mile (4.20)
Mutual Impedance between Conductor i and j:
Ω/mile (4.21)
where
= self impedance of conductor i in Ω/mile
= mutual impedance between conductors i and j in Ω/mile
r
i
= resistance of conductor i in Ω/mile
ω = 2πf = system angular frequency in radians per second
G = 0.1609344 × 10
−3
Ω/mile
RD
i
= radius of conductor i in feet
GMR
i
= Geometric Mean Radius of conductor i in feet
f = system frequency in Hertz
ρ = resistivity of earth in Ωmeters
D
i j
= distance between conductors i and j in feet (see Figure 4.4)
FIGURE 4.4
Conductors and images.
zˆ
ii
r
i
4ωP
ii
G j X
i
2ωG
S
ii
RD
i
 ln 4ωQ
ii
G + ⋅ +
+ + =
zˆ
ij
4ωP
ij
G j 2ωG
S
ij
D
ij
 ln 4ωQ
ij
G + ⋅
+ =
zˆ
ii
zˆ
ij
0812_frame_C04.fm Page 82 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 83
S
i j
= distance between conductor i and image j in feet (see Figure 4.4)
θ
i j
= angle between a pair of lines drawn from conductor i to its own
image and to the image of conductor j (see Figure 4.4)
Ω/mile (4.22)
(4.23)
(4.24)
(4.25)
4.1.4 Modiﬁed Carson’s Equations
Only two approximations are made in deriving the modiﬁed Carson’s equa
tions. These approximations involve the terms associated with P
i j
and Q
i j
by
using only the ﬁrst term of the variable P
i j
and the ﬁrst two terms of Q
i j
.
(4.26)
(4.27)
Substitute X
i
(Equation 4.22) into Equation 4.20
(4.28)
Combine terms and simplify:
(4.29)
Simplify Equation 4.21:
(4.30)
X
i
2ωG
RD
i
GMR
i
 ln ⋅ =
P
ij
π
8

1
3 2
k
ij
θ
ij
( )
k
ij
2
16
 2θ
ij
( ) 0.6728
2
k
ij
 ln +
⋅ cos + cos – =
Q
ij
0.0386 –
1
2

2
k
ij
 ln
1
3 2
k
ij
θ
ij
( ) cos + ⋅ + =
k
ij
8.565 10
4 –
S
ij
f
ρ
 ⋅ ⋅ × =
P
ij
π
8
 =
Q
ij
0.03860 –
1
2

2
k
ij
 ln + =
zˆ
ii
r
i
4ωP
ii
G j 2ωG
RD
i
GMR
i
 2ωG
S
ii
RD
i
 4ωQ
ii
G + ln ⋅ + ln ⋅
+ + =
zˆ
ii
r
i
4ωP
ii
G j2ωG
S
ii
GMR
i
 ln 2Q
ii
+
+ + =
zˆ
ij
4ωP
ij
G j2ωG
S
ij
D
ij
 ln 2Q
ij
+
+ =
0812_frame_C04.fm Page 83 Friday, July 20, 2001 2:55 PM
84 Distribution System Modeling and Analysis
Substitute expressions for P (Equation 4.27) and :
(4.31)
(4.32)
Substitute the expression for k
ij
(Equation 4.25) into the approximate expres
sion for Q
ij
(Equation 4.27):
(4.33)
Expand:
(4.34)
Equation 4.34 can be reduced to
(4.35)
or (4.36)
Substitute Equation 4.36 into Equation 4.31 and simplify:
(4.37)
Substitute Equation 4.36 into Equation 4.32 and simplify:
(4.38)
ω 2 π f ⋅ ⋅ ( )
zˆ
ii
r
i
π
2
fG j4πfG
S
ii
GMR
i
 ln 2Q
ii
+
+ + =
zˆ
ij
π
2
fG j4πfG
S
ij
D
ij
 ln 2Q
ij
+
+ =
Q
ij
0.03860
1
2
 + –
2
8.565 10
4 –
S
ij
f
ρ
 ⋅ ⋅ ⋅

ln =
Q
ij
0.03860 –
1
2

2
8.565 10
4 –
⋅

1
2

1
S
ij

1
2

ρ
f
 ln + ln + ln + =
Q
ij
3.8393
1
2
 S
ij
1
4

ρ
f
 ln + ln – =
2Q
ij
7.6786 S
ij
ln
1
2

ρ
f
 ln + – =
zˆ
ii
r
i
π
2
fG j4πfG
S
ii
GMR
i
 7.6786 S
ii
1
2

ρ
f
 ln + ln – + ln
+ + =
zˆ
ii
r
i
π
2
fG 4πfG
1
GMR
i
 ln 7.6786
1
2

ρ
f
 ln + +
+ + =
zˆ
ij
π
2
fG j4πfG
S
ij
D
ij
 ln 7.6786 S
ij
1
2

ρ
f
 ln + ln – +
+ =
zˆ
ij
π
2
fG j4πfG
1
D
ij
 7.6786
1
2

ρ
f
 ln + + ln
+ =
0812_frame_C04.fm Page 84 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 85
Substitute in the values of π and G:
(4.39)
(4.40)
It is now assumed:
f = Frequency = 60 Hertz
ρ = Earth resistivity = 100 Ohmmeter
Using these approximations and assumptions, the modiﬁed Carson’s equa
tions are
Ω/mile (4.41)
Ω/mile. (4.42)
Recall that Equations 4.18 and 4.19 could not be used because the resistance
of dirt, the GMR of dirt, and the various distances from conductors to dirt
were not known. A comparison of Equations 4.18 and 4.19 to Equations 4.41
and 4.42 shows that the modiﬁed Carson’s equations have deﬁned the
missing parameters. A comparison of the two sets of equations shows that:
Ω/mile (4.43)
(4.44)
The modiﬁed Carson’s equations will be used to compute the primitive self
and mutual impedances of overhead and underground lines.
4.1.5 Primitive Impedance Matrix for Overhead Lines
Equations 4.41 and 4.42 are used to compute the elements of an ncond x ncond
primitive impedance matrix. An overhead fourwire grounded wye distri
bution line segment will result in a 4 × 4 matrix. For an underground
grounded wye line segment consisting of three concentric neutral cables, the
zˆ
ii
r
i
0.00158836 f j0.00202237 f
1
GMR
i
 7.6786
1
2

ρ
f
 ln + + ln
⋅ + ⋅ + =
zˆ
ij
0.00158836 f j0.00202237 f
1
D
ij
 ln 7.6786
1
2

ρ
f
 ln + +
⋅ + ⋅ =
zˆ
ii
r
i
0.09530 j0.12134
1
GMR
i
 ln 7.93402 +
+ + =
zˆ
ij
0.09530 j0.12134
1
D
ij
 7.93402 + ln
+ =
r
d
0.09530 =
D
id
D
di
⋅
GMR
d
 ln
D
dj
D
id
⋅
GMR
d
 ln 7.93402 = =
0812_frame_C04.fm Page 85 Friday, July 20, 2001 2:55 PM
86 Distribution System Modeling and Analysis
resulting matrix will be 6 × 6. The primitive impedance matrix for a three
phase line with m neutrals will be of the form
(4.45)
In partitioned form, Equation 4.45 becomes
(4.46)
4.1.6 Phase Impedance Matrix for Overhead Lines
For most applications the primitive impedance matrix needs to be reduced
to a 3 × 3 phase frame matrix consisting of the self and mutual equivalent
impedances for the three phases. Figure 4.5 shows a fourwire grounded
neutral line segment. One standard method of reduction is the Kron reduc
tion.
3
The assumption is made that the line has a multigrounded neutral as
FIGURE 4.5
Fourwire grounded wye line segment.
zˆ
primitive
[ ]
zˆ
aa
zˆ
ab
zˆ
ac
 zˆ
an1
zˆ
an2
zˆ
anm
zˆ
ba
zˆ
bb
zˆ
bc
 zˆ
bn1
zˆ
bn2
zˆ
bnm
zˆ
ca
zˆ
ca
zˆ
cc
 zˆ
cn1
zˆ
cn2
zˆ
cnm
      
zˆ
n1a
zˆ
n1b
zˆ
n1c
 zˆ
n1n1
zˆ
n1n2
zˆ
n1nm
zˆ
n2a
zˆ
n2b
zˆ
n2c
 zˆ
n2n1
zˆ
n2n2
zˆ
n2nm
zˆ
nma
zˆ
nmb
zˆ
nmc
 zˆ
nmn1
zˆ
nmn2
zˆ
nmnm
=
zˆ
primitive
[ ]
zˆ
ij
[ ] zˆ
in
[ ]
zˆ
nj
[ ] zˆ
nn
[ ]
=
V'
bg
V'
ag
V'
cg
V
ag
V
bg
V
cg
+
+
+
+
+
+
V
ng
+
   
V'
ng
+

  
z
aa
z
bb
z
cc
z
nn
z
ab
z
bn
z
an
z
bc
z
cn
z
ac
I
a
I
b
I
n
I
c
0812_frame_C04.fm Page 86 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 87
shown in Figure 4.5. The Kron reduction method applies Kirchhoff’s voltage
law to the circuit.
(4.47)
In partitioned form, Equation 4.47 becomes
(4.48)
Because the neutral is grounded, the voltages V
ng
and are equal to zero.
Substituting those values into Equation 4.48 and expanding results in:
(4.49)
(4.50)
Solve Equation 4.50 for [I
n
]:
(4.51)
Substitute Equation 4.51 into Equation 4.49:
(4.52)
where (4.53)
Equation 4.53 is the ﬁnal form of the Kron reduction technique. The ﬁnal
phase impedance matrix becomes:
Ω/mile (4.54)
V
ag
V
bg
V
cg
V
ng
V′
ag
V′
bg
V′
cg
V′
ng
zˆ
aa
zˆ
ba
zˆ
ca
zˆ
na
zˆ
ab
zˆ
bb
zˆ
cb
zˆ
nb
zˆ
ac
zˆ
bc
zˆ
cc
zˆ
nc
zˆ
an
zˆ
bn
zˆ
cn
zˆ
nn
I
a
I
b
I
c
I
n
⋅ + =
V
abc
[ ]
V
ng
[ ]
V′
abc
[ ]
V′
ng
[ ]
z
ij
[ ] z
in
[ ]
z
nj
[ ] z
nn
[ ]
I
abc
[ ]
I
n
[ ]
⋅ + =
V′
ng
V
abc
[ ] V′
abc
[ ] z
ij
[ ] I
abc
[ ] z
in
[ ] I
n
[ ] ⋅ + ⋅ + =
0 [ ] 0 [ ] z
nj
[ ] I
abc
[ ] z
nn
[ ] I
n
[ ] ⋅ + ⋅ + =
I
n
[ ] z
nn
[ ]
−1
– z
nj
[ ] I
abc
[ ] ⋅ ⋅ =
V
abc
[ ] V′
abc
[ ] z
ij
[ ] z
in
[ ] z
nn
[ ]
1 –
z
nj
[ ] ⋅ ⋅ – ( ) + I
abc
[ ] ⋅ =
V
abc
[ ] V′
abc
[ ] z
abc
[ ] I
abc
[ ] ⋅ + =
z
abc
[ ] z
ij
[ ] z
in
[ ] z
nn
[ ]
1 –
z
nj
[ ] ⋅ ⋅ – =
z
abc
[ ]
z
aa
z
ab
z
ac
z
ba
z
bb
z
bc
z
ca
z
cb
z
cc
=
0812_frame_C04.fm Page 87 Friday, July 20, 2001 2:55 PM
88 Distribution System Modeling and Analysis
For a distribution line that is not transposed, the diagonal terms of Equation 4.54
will not be equal to each other, and the offdiagonal terms will not be equal to
each other. However, the matrix will be symmetrical.
For twophase (Vphase) and singlephase lines in grounded wye systems,
the modiﬁed Carson’s equations can be applied which will lead to initial
3 × 3 and 2 × 2 primitive impedance matrices. Kron reduction will reduce
the matrices to 2 × 2 and a single element. These matrices can be expanded
to 3 × 3 phase frame matrices by the addition of rows and columns consisting
of zero elements for the missing phases. For example, a Vphase line con
sisting of phases a and c, the phase impedance matrix would be
Ω/mile (4.55)
The phase impedance matrix for a phase b singlephase line would be
Ω/mile (4.56)
The phase impedance matrix for a threewire delta line is determined by the
application of Carson’s equations without the Kron reduction step.
The phase impedance matrix can be used to accurately determine the volt
age drops on the feeder line segments once the currents have been deter
mined. Since no approximations (transposition, for example) have been made
regarding the spacing between conductors, the effect of the mutual coupling
between phases is accurately taken into account. The application of the mod
iﬁed Carson’s equations and the phase frame matrix leads to the most accurate
model of a line segment. Figure 4.6 shows the general threephase model of
FIGURE 4.6
Threephase line segment model.
z
abc
[ ]
z
ab
0 z
ac
0 0 0
z
ca
0 z
cc
=
z
abc
[ ]
0 0 0
0 z
bb
0
0 0 0
=
Zca
Zbc
Zab
Zaa
Zbb
Zcc
Ia
Ib
Ic
+
+
  
+
Vbg
Vcg
Vag
+
+
+
  
Vcg
Vbg
Vag
Node n
Node m
n
n
n
m
m
m
0812_frame_C04.fm Page 88 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 89
a line segment. Bear in mind that for Vphase and singlephase lines some of
the impedance values will be zero. The voltage equation in matrix form for
the line segment is
(4.57)
where Z
i j
= z
i j
⋅ length.
Equation 4.57 can be written in condensed form as:
(4.58)
4.1.7 Sequence Impedances
Many times the analysis of a feeder will use only the positive and zero
sequence impedances for the line segments. There are two methods for
obtaining these impedances. The ﬁrst incorporates the application of the
modiﬁed Carson’s equations and the Kron reduction to obtain the phase
impedance matrix.
The deﬁnition of linetoground phase voltages as a function of the line
toground sequence voltages is given by
2
:
(4.59)
where a
s
=
In condensed form, Equation 4.59 becomes:
(4.60)
where (4.61)
The phase line currents are deﬁned in the same manner:
(4.62)
V
ag
V
bg
V
cg
n
V
ag
V
bg
V
cg
m
=
z
aa
z
ab
z
ac
z
ba
z
bb
z
bc
z
ca
z
cb
z
cc
I
a
I
b
I
c
⋅ +
VLG
abc
[ ]
n
VLG
abc
[ ]
m
Z
abc
[ ] I
abc
[ ] ⋅ + =
V
ag
V
bg
V
cg
1 1 1
1 a
s
2
a
s
1 a
s
a
s
2
=
VLG
0
VLG
1
VLG
2
⋅
1.0/120.
VLG
abc
[ ] A
s
[ ] VLG
012
[ ] ⋅ =
A
s
[ ]
1 1 1
1 a
s
2
a
s
1 a
s
a
s
2
=
I
abc
[ ] A
s
[ ] I
012
[ ] ⋅ =
0812_frame_C04.fm Page 89 Friday, July 20, 2001 2:55 PM
90 Distribution System Modeling and Analysis
Equation 4.60 can be used to solve for the sequence linetoground voltages
as a function of the phase linetoground voltages:
(4.63)
where (4.64)
Equation 4.58 can be transformed to the sequence domain by multiplying
both sides by [A
s
]
−1
and substituting in the deﬁnition of the phase currents
as given by Equation 4.62.
(4.65)
where (4.66)
Equation 4.65 in expanded form is given by:
(4.67)
Equation 4.66 is the deﬁning equation for converting phase impedances to
sequence impedances. In Equation 4.66 the diagonal terms of the matrix are
the sequence impedances of the line such that
Z
00
= zero sequence impedance,
Z
11
= positive sequence impedance, and
Z
22
= negative sequence impedance.
The offdiagonal terms of Equation 4.66 represent the mutual coupling
between sequences. In the idealized state these offdiagonal terms would
VLG
012
[ ] A
s
[ ]
1 –
VLG
abc
[ ] ⋅ =
A
s
[ ]
1 – 1
3

1 1 1
1 a
s
a
s
2
1 a
s
2
a
s
⋅ =
VLG
012
[ ]
n
A
s
[ ]
1 –
VLG
abc
[ ]
n
⋅ =
VLG
012
[ ]
n
A
s
[ ]
1 –
VLG
abn
[ ]
m
A
s
[ ]
1 –
Z
abc
[ ] A
s
[ ] I
012
[ ] ⋅ ⋅ ⋅ + ⋅ =
VLG
012
[ ]
n
VLG
012
[ ]
m
Z
012
[ ] I
012
[ ] ⋅ + =
Z
012
[ ] A
s
[ ]
1 –
Z
abc
[ ] A
s
[ ] ⋅ ⋅
Z
00
Z
01
Z
02
Z
10
Z
11
Z
12
Z
20
Z
21
Z
22
= =
VLG
0
VLG
1
VLG
2
n
VLG
0
VLG
1
VLG
2
m
Z
00
Z
01
Z
02
Z
10
Z
11
Z
12
Z
20
Z
21
Z
22
I
0
I
1
I
2
⋅ + =
0812_frame_C04.fm Page 90 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 91
be zero. In order for this to happen it must be assumed that the line has
been transposed. For highvoltage transmission lines this will sometimes be
the case. When the lines are transposed the mutual coupling between phases
(offdiagonal terms) are equal and, consequently, the offdiagonal terms of
the sequence impedance matrix become zero. Since distribution lines are
rarely if ever transposed, the mutual coupling between phases is not equal
and, as a result, the offdiagonal terms of the sequence impedance matrix
will not be zero.
If a line is assumed to be transposed, the phase impedance matrix is modiﬁed
so the three diagonal terms are equal and all of the offdiagonal terms are
equal. The usual procedure is to set the three diagonal terms of the phase
impedance matrix equal to the average of the diagonal terms of Equation 4.54,
and the offdiagonal terms equal to the average of the offdiagonal terms
of Equation 4.54. When this is done the self and mutual impedances are
deﬁned as:
Ω/mile (4.68)
Ω/mile (4.69)
The phase impedance matrix is now deﬁned as:
Ω/mile (4.70)
When Equation 4.66 is used with this phase impedance matrix the resulting
sequence matrix is diagonal (offdiagonal terms are zero). The sequence
impedances can be determined directly as:
Ω/mile (4.71)
Ω/mile (4.72)
A second method commonly used to determine the sequence impedances
directly is to employ the concept of Geometric Mean Distances (GMD). The
GMD between phases is deﬁned as:
ft (4.73)
The GMD between phases and neutral is deﬁned as:
ft (4.74)
z
s
1
3
 z
aa
z
bb
z
cc
+ + ( ) ⋅ =
z
m
1
3
 z
ab
z
bc
z
ca
+ + ( ) ⋅ =
z
abc
[ ]
z
s
z
m
z
m
z
m
z
s
z
m
z
m
z
m
z
s
=
z
00
z
s
2 z
m
⋅ + =
z
11
z
22
z
s
z
m
– = =
D
ij
GMD
ij
D
ab
D
bc
D
ca
⋅ ⋅
3
= =
D
in
GMD
in
D
an
D
bn
D
cn
⋅ ⋅
3
= =
0812_frame_C04.fm Page 91 Friday, July 20, 2001 2:55 PM
92 Distribution System Modeling and Analysis
The GMDs as deﬁned above are used in Equations 4.41 and 4.42 to determine
the various self and mutual impedances of the line, resulting in:
Ω/mile (4.75)
Ω/mile (4.76)
Ω/mile (4.77)
Ω/mile (4.78)
Equations 4.75 through 4.78 will deﬁne a matrix of order ncond × ncond,
where ncond is the number of conductors (phases plus neutrals) in the line
segment. Application of the Kron reduction (Equation 4.53) and the sequence
impedance transformation (Equation 4.66) leads to the following expressions
for the zero, positive, and negative sequence impedances:
Ω/mile (4.79)
(4.80)
Ω/mile
Equation 4.80 is recognized as the standard equation for the calculation of the
line impedances when a balanced threephase system and transposition are
assumed.
Example 4.1
An overhead threephase distribution line is constructed as shown in Figure 4.7.
Determine the phase impedance matrix and the positive and zero sequence
impedances of the line. The phase conductors are 336,400 26/7 ACSR (Linnet),
and the neutral conductor is 4/0 6/1 ACSR.
SOLUTION
From the table of standard conductor data (Appendix A) it is found that
336,400 26/7 ACSR: GMR = 0.0244 ft
Resistance = 0.306 Ω/mile
zˆ
ii
r
i
0.0953 j0.12134
1
GMR
i

7.93402 + ln ⋅ + + =
zˆ
nn
r
n
0.0953 j0.12134
1
GMR
n

7.93402 + ln ⋅ + + =
zˆ
ij
0.0953 j0.12134
1
D
ij

7.93402 + ln ⋅ + =
zˆ
in
0.0953 j0.12134
1
D
in

7.93402 + ln ⋅ + =
z
00
zˆ
ii
2 zˆ
ij
3
zˆ
in
2
zˆ
nn

⋅ – ⋅ + =
z
11
z
22
zˆ
ii
zˆ
ij
– = =
z
11
z
22
r
i
j0.12134
D
ij
GMR
i

ln ⋅ + = =
0812_frame_C04.fm Page 92 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 93
4/0 6/1 ACSR: GMR = 0.00814 ft.
Resistance = 0.5920 Ω/mile
From Figure 4.7, the following distances between conductors can be
determined:
D
ab
= 2.5 ft. D
bc
= 4.5 ft. D
ca
= 7.0 ft.
D
an
= 5.6569 ft. D
bn
= 4.272 ft. D
cn
= 5.0 ft.
Applying the modiﬁed Carson’s equation for self impedance (Eq. 4.41), the
self impedance for phase a is
Ω/mile
Applying Equation 4.42 for the mutual impedance between phases a and b:
Ω/mile
Applying the equations for the other self and mutual impedance terms results
in the primitive impedance matrix:
Ω/mile
FIGURE 4.7
Threephase distribution line spacings.
4.0'
2.5' 4.5'
3.0'
a b c
n
25.0'
zˆ
aa
0.0953 0.306 j0.12134
1
0.0244
 7.93402 + ln
⋅ + + =
0.4013 j1.4133 + =
zˆ
ab
0.0953 j0.12134
1
2.5
 7.93402 + ln
⋅ + =
0.0953 j0.8515 + =
zˆ [ ]
0.4013 j1.4133 + 0.0953 j0.8515 + 0. 0953 j0.7266 + 0.0953 j0.7524 +
0.0953 j0.8515 + 0.4013 j1.4133 + 0.0953 j0.7802 + 0.0953 j0.7865 +
0.0953 j0.7266 + 0.0953 j0.7802 + 0.4013 j1.4133 + 0.0953 j0.7674 +
0.0953 j0.7524 + 0.0953 j0.7865 + 0.0953 j0.7674 + 0.6873 j1.5465 +
=
0812_frame_C04.fm Page 93 Friday, July 20, 2001 2:55 PM
94 Distribution System Modeling and Analysis
The primitive impedance matrix in partitioned form is
Ω/mile
Ω/mile
Ω/mile
Ω/mile
The Kron reduction of Equation 4.53 results in the phase impedance matrix:
Ω/mile
The phase impedance matrix can be transformed into the sequence imped
ance matrix with the application of Equation 4.66:
Ω/mile
In the sequence impedance matrix the 1,1 term is the zero sequence imped
ance; the 2,2 term is the positive sequence impedance; and the 3,3 term is
the negative sequence impedance. The 2,2 and 3,3 terms are equal, which
demonstrates that for line segments, the positive and negative sequence
impedances are equal. Note that the offdiagonal terms are not zero. This
implies that there is mutual coupling between sequences. This is a result of
the nonsymmetrical spacing between phases. With the offdiagonal terms
zˆ
ij
[ ]
0.4013 j1.4133 + 0.0953 j0.8515 + 0.0953 j0.7266 +
0.0953 j0.8515 + 0.4013 j1.4133 + j0.0943 j0.7802 +
0.0953 j0.7266 + 0.0953 j0.7802 + 0.4013 j1.4133 +
=
zˆ
in
[ ]
0.0953 j0.7524 +
0.0953 j0.7865 +
0.0953 j0.7674 +
=
zˆ
nn
[ ] 0.6873 j1.5465 + [ ] =
zˆ
nj
[ ]
0.0953 j0.7524 + 0.0953 j0.7865 + 0.0953 j0.7674 +
=
z
abc
[ ] zˆ
ij
[ ] zˆ
in
[ ] zˆ
nn
[ ]
1 –
zˆ
nj
[ ] ⋅ ⋅ – =
z
abc
[ ]
0.4576 j1.0780 + 0.1560 j.5017 + 0.1535 j0.3849 +
0.1560 j0.5017 + 0.4666 j1.0482 + 0.1580 j0.4236 +
0.1535 j0.3849 + 0.1580 j0.4236 + 0.4615 j1.0651 +
=
z
012
[ ] A
s
[ ]
1 –
z
abc
[ ] A
s
[ ] ⋅ ⋅ =
z
012
[ ]
0.7735 j1.9373 + 0.0256 j0.0115 + 0.0321 – j0.0159 +
−0.0321 j0.0159 + 0.3061 j0.6270 + 0.0723 – j – 0.0060
0.0256 j0.0115 + 0.0723 j0.0059 – 0.3061 j0.6270 +
=
0812_frame_C04.fm Page 94 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 95
nonzero, the three sequence networks representing the line will not be inde
pendent. However, it is noted that the offdiagonal terms are small relative
to the diagonal terms.
In highvoltage transmission lines, it is usually assumed that the lines are
transposed and that the phase currents represent a balanced threephase
set. The transposition can be simulated in Example 4.1 by replacing the
diagonal terms of the phase impedance matrix with the average value of
the diagonal terms (0.4619 + j1.0638), and replacing each offdiagonal term
with the average of the offdiagonal terms (0.1558 + j0.4368). This modiﬁed
phase impedance matrix becomes:
Ω/mile
Using this modiﬁed phase impedance matrix in the symmetrical compo
nent transformation equation results in the modiﬁed sequence impedance
matrix
Ω/mile
Note that now the offdiagonal terms are all equal to zero, meaning there is
no mutual coupling between sequence networks. It should also be noted
that the modiﬁed zero, positive, and negative sequence impedances are
exactly equal to the exact sequence impedances that were ﬁrst computed.
The results of this example should not be interpreted to mean that a three
phase distribution line can be assumed to have been transposed. The original
phase impedance matrix must be used if the correct effect of the mutual
coupling between phases is to be modeled.
4.2 Series Impedance of Underground Lines
Figure 4.8 shows the general conﬁguration of three underground cables (con
centric neutral or tape shielded) with an additional neutral conductor. The
modiﬁed Carson’s equations can be applied to underground cables in much
the same manner as for overhead lines. The circuit of Figure 4.8 will result
in a 7 × 7 primitive impedance matrix. For underground circuits that do not
have the additional neutral conductor, the primitive impedance matrix will
be 6 × 6.
z1
abc
[ ]
0.4619 j1.0638 + 0.1558 j0.4368 + 0.1558 j0.4368 +
0.1558 j0.4368 + 0.4619 j1.0638 + 0.1558 j0.4368 +
0.1558 j0.4368 + 0.1558 j0.4368 + 0.4619 j1.0638 +
=
z1
012
[ ]
0.7735 j1.9373 + 0 0
0 0.3061 j0.6270 + 0
0 0 0.3061 j0.6270 +
=
0812_frame_C04.fm Page 95 Friday, July 20, 2001 2:55 PM
96 Distribution System Modeling and Analysis
Two popular types of underground cables are the concentric neutral cable
and the tapeshielded cable. To apply the modiﬁed Carson’s equations, the
resistance and GMR of the phase conductor and the equivalent neutral must
be known.
4.2.1 Concentric Neutral Cable
Figure 4.9 shows a simple detail of a concentric neutral cable. The cable
consists of a central phase conductor covered by a thin layer of nonmetallic
semiconducting screen, to which is bonded the insulating material. The
insulation is then covered by a semiconducting insulation screen. The solid
strands of concentric neutral are spiraled around the semiconducting screen
with a uniform spacing between strands. Some cables will also have an
insulating jacket encircling the neutral strands. In order to apply Carson’s
equations to this cable, the following data needs to be extracted from a table
of underground cables (Appendix B):
d
c
= phase conductor diameter (inches)
d
od
= nominal diameter over the concentric neutrals of the cable (inches)
d
s
= diameter of a concentric neutral strand (inches)
GMR
c
= geometric mean radius of the phase conductor (ft.)
FIGURE 4.8
Threephase underground with additional neutral.
FIGURE 4.9
Concentric neutral cable.
D14
D13
D12 D23 D34
a b c n
d
c
d
od
d
s
R
Phase Conductor
Insulation
Concentric Neutral Strand
Insulation Screen
Jacket
0812_frame_C04.fm Page 96 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 97
GMR
s
= geometric mean radius of a neutral strand (ft.)
r
c
= resistance of the phase conductor (Ω/mile)
r
s
= resistance of a solid neutral strand (Ω/mile)
k = number of concentric neutral strands
The geometric mean radii of the phase conductor and a neutral strand are
obtained from a standard table of conductor data (Appendix A). The equiv
alent geometric mean radius of the concentric neutral is computed using the
equation for the geometric mean radius of bundled conductors used in high
voltage transmission lines.
1
ft (4.81)
where
R = radius of a circle passing through the center of the concentric neutral
strands
ft (4.82)
The equivalent resistance of the concentric neutral is
Ω/mile (4.83)
The various spacings between a concentric neutral and the phase conductors
and other concentric neutrals are as follows:
Concentric neutral to its own phase conductor
D
i j
= R (Equation 4.82 above)
Concentric neutral to an adjacent concentric neutral
D
i j
= centertocenter distance of the phase conductors
Concentric neutral to an adjacent phase conductor
Figure 4.10 shows the relationship between the distance between centers of
concentric neutral cables and the radius of a circle passing through the centers
of the neutral strands. The geometric mean distance between a concentric
neutral and an adjacent phase conductor is given by:
ft (4.84)
where D
nm
= centertocenter distance between phase conductors. For cables
buried in a trench the distance between cables will be much greater than the
GMR
cn
GMR
s
k R
k−1
⋅ ⋅
k
=
R
d
od
d
s
–
24
 =
r
cn
r
s
k
 =
D
ij
D
nm
k
R
k
–
k
=
0812_frame_C04.fm Page 97 Friday, July 20, 2001 2:55 PM
98 Distribution System Modeling and Analysis
radius R, and therefore it may be assumed that D
i j
in Equation 4.84 is equal
to D
nm
. For cables in conduit that assumption is not valid.
In applying the modiﬁed Carson’s equations, the numbering of conductors
and neutrals is important. For example, a threephase underground circuit
with an additional neutral conductor must be numbered as:
1 = phase conductor #1
2 = phase conductor #2
3 = phase conductor #3
4 = neutral of conductor #1
5 = neutral of conductor #2
6 = neutral of conductor #3
7 = additional neutral conductor (if present)
Example 4.2
Three concentric neutral cables are buried in a trench with spacings as shown
in Figure 4.11. The cables are 15 kV, 250,000 CM stranded allaluminum with
13 strands of #14 annealed, coated copper wires (1/3 neutral). The outside
diameter of the cable over the neutral strands is 1.29 inches (Appendix B).
Determine the phase impedance matrix and the sequence impedance matrix.
FIGURE 4.10
Distances between concentric neutral cables.
FIGURE 4.11
Threephase concentric neutral cable spacing.
R R
D
nm
6" 6"
0812_frame_C04.fm Page 98 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 99
SOLUTION
The data for the phase conductor and neutral strands from a conductor data
table (Appendix A) are
250,000 AA phase conductor:
GMR = 0.0171 ft.
Diameter = 0.567 inches
Resistance = 0.4100 Ω/mile
#14 copper neutral strands:
GMR
s
= 0.00208 ft.
Resistance = 14.8722 Ω/mile
Diameter (d
s
) = 0.0641 inches
The radius of the circle passing through the center of the strands (Equation 4.82)
is
The equivalent GMR of the concentric neutral is computed by:
The equivalent resistance of the concentric neutral is
Ω/mile
The phase conductors are numbered 1, 2, and 3. The concentric neutrals are
numbered 4, 5, and 6. The conductortoconductor and concentric neutral
toconcentric neutral spacings are
The spacings between conductors and their concentric neutrals are
Since the radius R is much smaller than the spacings between cables, the
distances between concentric neutrals and adjacent phase conductors are
R
d
od
d
s
–
24
 0.0511 ft = =
GMR
cn
GMR
s
k R
k−1
⋅ ⋅
k
0.00208 13 0.0511
13−1
⋅ ⋅
13
0.0486 ft = = =
r
cn
r
s
k

14.8722
13
 1.1438 = = =
D
12
D
21
D
45
D
54
0.5 ft = = = =
D
23
D
32
D
56
D
65
0.5 ft = = = =
D
31
D
13
D
64
D
46
1.0 ft = = = =
D
14
D
25
D
36
R 0.0511 ft = = = =
0812_frame_C04.fm Page 99 Friday, July 20, 2001 2:55 PM
100
Distribution System Modeling and Analysis
just the centertocenter distances between conductors:
ft
The self impedance for the cable in position 1 is
The self impedance for the concentric neutral for Cable #1 is
Ω
/mile
The mutual impedance between Cable #1 and Cable #2 is
Ω
/mile
The mutual impedance between Cable #1 and its concentric neutral is
Ω
/mile
The mutual impedance between the concentric neutral of Cable #1 and the
concentric neutral of Cable #2 is
Ω
/mile
Continuing the application of the modiﬁed Carson’s equations results in a
6
×
6 primitive impedance matrix. This matrix in partitioned (Equation 4.33)
form is
D
15
D
51
0.5 = =
D
26
D
62
0.5 = =
D
61
D
16
1.0. = =
zˆ
11
0.0953 0.41 j0.12134 ln
1
0.0171
 7.93402 +
⋅ + + =
0.5053 j1.4564 Ω/mile + =
zˆ
44
0.0953 1.144 j0.12134 ln
1
0.0486
 7.93402 +
⋅ + + =
1.2393 j1.3296 + =
zˆ
12
0.0953 j0.12134 ln
1
0.5
 7.93402 +
⋅ + 0.0953 j1.0468 + = =
zˆ
14
0.0953 j0.12134 ln
1
0.0511
 7.93402 +
⋅ + 0.0953 j1.3236 + = =
zˆ
45
0.0953 j0.12134 ln
1
0.5
 7.93402 +
⋅ + 0.0953 j1.0468 + = =
zˆ
ij
[ ]
.5053 j1.4564 .0953 j1.0468 .0953 j.9627 + + +
.0953 j1.0468 .5053 j1.4564 .0953 j1.0468 + + +
.0953 j.9627 + .0953 j1.0468 .5053 j1.4564 + +
Ω/mile =
zˆ
in
[ ]
.0953 j1.3236 .0953 j1.0468 .0953 j.9627 + + +
.0953 j1.0468 .0953 j1.3236 .0953 j1.0468 + + +
.0953 j.9627 .0953 j1.0468 .0953 j1.3236 + + +
Ω/mile =
0812_frame_C04.fm Page 100 Monday, October 28, 2002 9:50 AM
Series Impedance of Overhead and Underground Lines 101
Using the Kron reduction results in the phase impedance matrix:
Ω/mile
The sequence impedance matrix for the concentric neutral threephase line
is determined using Equation 4.66:
Ω/mile
4.2.2 TapeShielded Cables
Figure 4.12 shows a simple detail of a tapeshielded cable. The cable consists
of a central phase conductor covered by a thin layer of nonmetallic semi
conducting screen to which is bonded the insulating material. The insulation
is covered by a semiconducting insulation screen. The shield is bare copper
FIGURE 4.12
Tapeshielded cable.
zˆ
nj
[ ] zˆ
in
[ ]
T
=
zˆ
nn
[ ]
1.2391 j1.3296 + .0953 j1.0468 + .0953 j.9627 +
.0953 j1.0468 + 1.2391 j1.3296 + .0953 j1.0468 +
.0953 j.9627 + .0953 j1.0468 + 1.2391 j1.3296 +
Ω/mile =
z
abc
[ ] zˆ
ij
[ ] zˆ
in
[ ] zˆ
nn
[ ]
1 –
zˆ
nj
[ ] ⋅ ⋅ – =
z
abc
[ ]
0.7981 j0.4463 0.3191 j0.0328 0.2849 j0.0143 – + +
0.3191 j0.0328 0.7891 j0.4041 0.3191 j0.0328 + + +
0.2849 j0.0143 0.3191 – j0.0328 0.7981 j0.4463 + +
=
z
012
[ ] A
s
[ ]
1 –
z
abc
[ ] A
s
[ ] ⋅ ⋅ =
z
012
[ ]
1.4105 j0.4665 + 0.0028 – j0.0081 – 0.0056 – j0.0065 +
0.0056 – j0.0065 + 0.4874 j0.4151 + 0.0264 – j0.0451 +
0.0028 – j0.0081 – 0.0523 j0.0003 + 0.4874 j0.4151 +
=
AL or CU Phase
Conductor
Insulation
Jacket
CU Tape Shield
T
d d d
od s
c
0812_frame_C04.fm Page 101 Friday, July 20, 2001 2:55 PM
102 Distribution System Modeling and Analysis
tape helically applied around the insulation screen. An insulating jacket
encircles the tape shield. Parameters of the taped shielded cable are
d
c
= diameter of phase conductor (inches)
d
s
= outside diameter of the tape shield (inches)
d
od
= outside diameter over jacket (inches)
T = thickness of copper tape shield in mils
Once again, the modiﬁed Carson’s equations will be applied to calculate the self
impedances of the phase conductor and the tape shield, as well as the mutual
impedance between the phase conductor and the tape shield. The resistance and
GMR of the phase conductor are found in a standard table of conductor data.
The resistance of the tape shield is given by:
Ω/mile (4.85)
The resistivity (ρ) in Equation 4.85 must be expressed in Ωmeters at 50°C.
The outside diameter of the tape shield (d
s
) is given in inches and the
thickness of the tape shield (T) is in mils.
The GMR of the tape shield is the radius of a circle passing through the
middle of the shield and is given by:
ft. (4.86)
The various spacings between a tape shield and the conductors and other
tape shields are as follows:
Tape shield to its own phase conductor
D
i j
= GMR
shield
= radius to midpoint of the shield (ft.) (4.87)
Tape shield to an adjacent tape shield
D
i j
= centertocenter distance of the phase conductors (ft.) (4.88)
Tape shield to an adjacent phase or neutral conductor
D
i j
= centertocenter distance between phase conductors (ft.)
(4.89)
Example 4.3
A singlephase circuit consists of a 1/0 AA, 220mil insulation tape
shielded cable and a 1/0 Cu neutral conductor, as shown in Figure 4.13.
The singlephase line is connected to phase b. Determine the phase imped
ance matrix. Assume p = 2.3715 × 10
−8
Ωmeter.
r
shield
7.9385×10
8
ρ
d
s
· T
 =
GMR
shield
d
s
T
1000
 –
24
 =
0812_frame_C04.fm Page 102 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 103
Cable Data: 1/0 AA
Outside diameter of the tape shield = d
s
= 0.88 inches
Resistance = 0.97 Ω/mile
GMR
p
= 0.0111 ft.
Tape shield thickness = T = 5 mils
Neutral Data: 1/0 Copper, 7 strand
Resistance = 0.607 Ω/mile
GMR
n
= 0.01113 ft.
Distance between cable and neutral = D
nm
= 3 inches
The resistance of the tape shield is computed according to Equation 4.85:
Ω/mile
The GMR of the tape shield is computed according to Equation 4.86:
ft
The conductors are numbered such that:
#1 = 1/0 AA conductor
#2 = tape shield
#3 = 1/0 copper ground
FIGURE 4.13
Singlephase tape shield with neutral.
3"
r
shield
7.9385×10
8
·
ρ
d
s
· T
 7.9385×10
8
·
2.3715×10
8 –
0.88 · 5
 4.2785 = = =
GMR
shield
d
s
T
1000
 –
24

0.88
5
1000
 –
24
 0.0365 = = =
0812_frame_C04.fm Page 103 Friday, July 20, 2001 2:55 PM
104 Distribution System Modeling and Analysis
The spacings used in the modiﬁed Carson’s equations are
The self impedance of Conductor #1 is
The mutual impedance between Conductor #1 and the tape shield (Conductor #2)
is
The self impedance of the tape shield (Conductor #2) is
Continuing on, the ﬁnal primitive impedance matrix is
Ω/mile
In partitioned form, the primitive impedance matrix is
D
12
GMR
shield
0.0365 = =
D
13
3
12
 0.25 = =
z
ˆ
11
0.0953 = 0.97 j0.12134 + +
1
0.0111
 7.93402 + ln
⋅
1.0653 j1.5088 Ω/mile + =
zˆ
12
0.0953 = j0.12134
1
0.0365
 ln 7.93402 +
⋅ +
0.0953 j1.3645 + = Ω/mile
z
ˆ
22
0.0953 = 4.2786 j0.12134
1
0.0365
 ln 7.93402 +
⋅ + +
4.3739 j1.3645 + = Ω/mile
z
ˆ [ ]
1.0653 j1.5088 + 0.0953 j1.3645 + 0.0953 j1.1309 +
0.0953 j1.3645 + 4.3739 j1.3645 + 0.0953 j1.1309 +
0.0953 j1.1309 + 0.0953 j1.1309 + 0.7023 j1.5085 +
=
z
ˆ
ij
[ ] 1.0653 j1.5088 + [ ] =
z
ˆ
in
[ ] 0.0953 j1.3645 + 0.0953 j1.1309 + [ ] =
z
ˆ
nj
[ ]
0.0953 j1.3645 +
0.0953 j1.1309 +
=
z
ˆ
nn
[ ]
4.3739 j1.3645 + 0.0953 j1.1309 +
0.0953 j1.1309 + 0.7023 j1.5085 +
=
0812_frame_C04.fm Page 104 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines
105
Applying Kron’s reduction method will result in a single impedance which
represents the equivalent singlephase impedance of the tape shield cable
and the neutral conductor:
Since the singlephase line is on
phase b
, then the phase impedance matrix
for the line is
Ω
/mile
4.3 Summary
This chapter has been devoted to presenting methods for computing the phase
impedances and sequence impedances of overhead lines and underground
cables. Carson’s equations have been modiﬁed in order to simplify the com
putation of the phase impedances. When using the modiﬁed Carson’s equa
tions there is no need to make any assumptions, such as transposition of the
lines. By assuming an untransposed line and including the actual phasing of
the line, the most accurate values of the phase impedances, self and mutual,
are determined. Since voltage drop is a primary concern on a distribution
line, the impedances used for the line must be as accurate as possible.
References
1. Glover, J. D. and Sarma, M.,
Power System Analysis and Design,
2nd
ed.,
PWSKent
Publishing, Boston, 1994.
2. Carson, John R., Wave propagation in overhead wires with ground return,
Bell
System Technical Journal
, Vol. 5, New York, 1926.
3. Kron, G., Tensorial analysis of integrated transmission systems, Part I, the six
basic reference frames,
AIEE Trans.,
Vol. 71, 1952.
Problems
4.1
Determine the phase impedance matrix [
Z
abc
] and the sequence imped
ance matrix [
Z
012
] in
Ω
/mile for the threephase conﬁguration shown in
Figure 4.14.
z
1p
z
ˆ
ij
[ ] z
in
[ ] z
ˆ
nn
[ ]
1 –
z
ˆ
nj
[ ] ⋅ ⋅ – =
z
1p
1.3219 = j0.6743 + Ω/mile.
z
abc
[ ]
0 0 0
0 1.3219 j0.6743 + 0
0 0 0
=
0812_frame_C04.fm Page 105 Monday, October 28, 2002 9:50 AM
106 Distribution System Modeling and Analysis
Phase Conductors: 556,500 26/7 ACSR
Neutral Conductor: 4/0 ACSR
4.2 Determine the phase impedance [Z
abc
] matrix in for the two
phase conﬁguration in Figure 4.15.
Phase Conductors: 336,400 26/7 ACSR
Neutral Conductor: 4/0 6/1 ACSR
4.3 Determine the phase impedance [Z
abc
] matrix in Ω/mile for the single
phase conﬁguration shown in Figure 4.16.
Phase and Neutral Conductors: 1/0 6/1 ACSR
FIGURE 4.14
Threephase conﬁguration for Problem 4.1.
FIGURE 4.15
Twophase conﬁguration for Problem 4.2.
4.0'
2.5' 4.5'
3.0'
a c b
n
25.0'
Ω/mile
4.0'
3.0'
a c
n
25.0'
7.0'
0812_frame_C04.fm Page 106 Friday, July 20, 2001 2:55 PM
Series Impedance of Overhead and Underground Lines 107
4.4 Create the spacings and conﬁgurations of Problems 4.1, 4.2, and 4.3 in
the Radial Distribution Analysis Package (RDAP). Compare the phase
impedance matrices to those computed in the previous problems.
4.5 Determine the phase impedance matrix and sequence impedance
matrix [Z
012
] in Ω/mile for the threephase pole conﬁguration in Figure 4.17.
The phase and neutral conductors are 250,000 allaluminum.
4.6 Compute the positive, negative, and zero sequence impedances in
Ω/1000 ft. using the Geometric Mean Distance (GMD) method for the pole
conﬁguration shown in Figure 4.17.
FIGURE 4.16
Singlephase pole conﬁguration for Problem 4.3.
FIGURE 4.17
Threephase pole conﬁguration for Problem 4.5.
5.0'
n
25.0'
b
0.5'
Z
abc
[ ]
4'
a
b
c
n
2
2
2
'
'
'
25’
0812_frame_C04.fm Page 107 Friday, July 20, 2001 2:55 PM
108 Distribution System Modeling and Analysis
4.7 Determine the and [Z
012
] matrices in Ω/mile for the threephase
conﬁguration shown in Figure 4.18. The phase conductors are 350,000 all
aluminum, and the neutral conductor is 250,000 allaluminum.
4.8 Compute the positive, negative, and zero sequence impedances in
Ω/1000 ft. for the line of Figure 4.18 using the average self and mutual
impedances deﬁned in Equations 4.68 and 4.69.
4.9 A 4/0 aluminum concentric neutral cable is to be used for a singlephase
lateral. The cable has a full neutral (see Appendix B). Determine the imped
ance of the cable and the resulting phase impedance matrix in Ω/mile,
assuming the cable is connected to phase b.
4.10 Three 250,000 CM aluminum concentric cables with onethird neutrals
are buried in a trench in a horizontal conﬁguration (see Figure 4.11). Deter
mine the [Z
abc
] and [Z
012
] matrices in Ω/1000 ft., assuming phasing of cab.
4.11 Create the spacings and conﬁgurations of Problems 4.9 and 4.10 in
RDAP. Compare the values of the phase impedance matrices to those com
puted in the previous problems.
4.12 A singlephase underground line is composed of a 350,000 CM alu
minum tapeshielded cable. A 4/0 copper conductor is used as the neutral.
The cable and neutral are separated by 4 inches. Determine the phase imped
ance matrix in Ω/mile for this singlephase cable line, assuming phase c.
4.13 Three onethird neutral, 2/0 aluminumjacketed, concentric neutral
cables are installed in a 6inch conduit. Assume the cable jacket has a
thickness of 0.2inch and the cables lie in a triangular conﬁguration inside
the conduit. Compute the phase impedance matrix in Ω/mile for this cabled
line.
4.14 Create the spacing and conﬁguration of Problem 4.13 in RDAP. Com
pare results.
FIGURE 4.18
Threephase pole conﬁguration for Problem 4.7.
Z
abc
[ ]
a
b
c
2'
2'
n
2'
2'
25'
0812_frame_C04.fm Page 108 Friday, July 20, 2001 2:55 PM
109
5
Shunt Admittance of Overhead
and Underground Lines
The shunt admittance of a line consists of the conductance and the capacitive
susceptance. The conductance is usually ignored because it is very small
compared to the capacitive susceptance. The capacitance of a line is the result
of the potential difference between conductors. A charged conductor creates
an electric ﬁeld that emanates outward from the center. Lines of equipotential
are created that are concentric to the charged conductor, as illustrated in
Figure 5.1.
In Figure 5.1 a difference of potential between two points (
P
1
and
P
2
) is a
result of the electric ﬁeld of the charged conductor. When the potential
difference between the two points is known, then the capacitance between
the two points can be computed. If there are other charged conductors
nearby, the potential difference between the two points will be a function of
FIGURE 5.1
Electric ﬁeld of a charged round conductor.
D
2
P
2
P
1
+
+
+
+
+
D
1
RD
0812_frame_C05.fm Page 109 Friday, July 20, 2001 3:06 PM
110
Distribution System Modeling and Analysis
the distance to the other conductors and the charge on each conductor. The
principle of superposition is used to compute the total voltage drop between
two points, and then the resulting capacitance between the points. The points
can be points in space, the surface of two conductors, or the surface of a
conductor and ground.
5.1 General VoltageDrop Equation
Figure 5.2 shows an array of
N
positively charged solid, round conductors.
Each conductor has a unique uniform charge density of
q
C meter. The
voltage drop between Conductor
i
and Conductor
j
as a result of all of the
charged conductors is given by:
(5.1)
Equation 5.1 can be written in general form as:
(5.2)
where
ε
=
ε
0
ε
r
= permittivity of the medium
ε
0
=
permittivity of free space = 8.85
×
10
−
12
µ
F/meter
ε
r
=
relative permittivity of the medium
q
n
=
charge density on Conductor
n
cb/meter
D
ni
=
distance between Conductor
n
and Conductor
i
(ft.)
D
nj
=
distance between Conductor
n
and Conductor
j
(ft.)
RD
n
=
radius of Conductor
n
FIGURE 5.2
Array of round conductors.
V
ij
1
2πε
 q
1
D
1j
D
1i

…
q
i
D
ij
RD
i

…
q
j
RD
j
D
ij

…
q
N
D
Nj
D
Ni
 ln + + ln + + ln + + ln
=
V
ij
1
2πε
 q
n
D
nj
D
ni
 ln
n=1
N
∑
=
1
i
j
+

V
ij
+
+
+
+
+
n
N
0812_frame_C05.fm Page 110 Friday, July 20, 2001 3:06 PM
Shunt Admittance of Overhead and Underground Lines
111
5.2 Overhead Lines
The method of conductors and their images is employed in the calculation
of the shunt capacitance of overhead lines. This is the same concept that was
used in Chapter 4 in the general application of Carson’s equations. Figure 5.3
illustrates the conductors and their images, and will be used to develop a
general voltagedrop equation for overhead lines. In Figure 5.3 it is assumed
that:
=
−
and
=
−
(5.3)
Applying Equation 5.2 to Figure 5.3:
(5.4)
Because of the assumptions of Equation 5.3, Equation 5.4 can be simpliﬁed:
(5.5)
FIGURE 5.3
Conductors and images.
q′
i
q
i
q′
j
q
j
V
ii
1
2πε
 q
i
S
ii
RD
i
 q
i
′
RD
i
S
ii
 q
j
S
ij
D
ij
 q
j
′
D
ij
S
ij
 ln + ln + ln + ln
=
V
ii
1
2πε
 q
i
S
ii
RD
i
 q
i
RD
i
S
ii
 q
j
S
ij
D
ij
 q
j
D
ij
S
ij
 ln – ln + ln – ln
=
V
ii
1
2πε
 q
i
S
ii
RD
i
 q
i
S
ii
RD
i
 q
j
S
ij
D
ij
 q
j
S
ij
D
ij
 ln + ln + ln + ln
=
V
ii
1
2πε
 2 q ⋅
i
S
ii
RD
i
 2 q
j
S
ij
D
ij
 ln ⋅ + ln
=
0812_frame_C05.fm Page 111 Friday, July 20, 2001 3:06 PM
112 Distribution System Modeling and Analysis
where
S
ii
= distance from Conductor i to its image i’ (ft.)
S
ij
= distance from Conductor i to the image of Conductor j (ft.)
D
ij
= distance from Conductor i to Conductor j (ft.)
= radius of Conductor i in ft.
Equation 5.5 gives the total voltage drop between Conductor i and its image.
The voltage drop between Conductor i and ground will be onehalf of that
given in Equation 5.5:
(5.6)
Equation 5.6 can be written in general form as:
(5.7)
where and are the self and mutual potential coefﬁcients. For overhead
lines the relative permittivity of air is assumed to be 1.0 so that:
ε
air
= 1.0 × 8.85 × 10
−12
F/meter
ε
air
= 1.4240 × 10
−2
µF/mile
Using the value of permittivity in µF/mile, the self and mutual potential
coefﬁcients are deﬁned as:
mile/µF (5.9)
mile/µF (5.10)
NOTE In applying Equations 5.9 and 5.10, the values of , S
ii
, S
ij
, and
D
ij
must all be in the same units. For overhead lines the distances between
conductors are typically speciﬁed in feet, while the value of the conductor
diameter from a table will typically be in inches. Care must be taken to assure
that the radius in feet is used in applying the two equations.
For an overhead line of ncond conductors, the primitive potential coefﬁ
cient matrix [ ] can be constructed. The primitive potential coefﬁcient
RD
i
V
ig
1
2πε
 q
i
S
ii
RD
i
 q
j
S
ij
D
ij
 ln + ln
=
V
ig
P
ˆ
ii
q
i
P
ˆ
ij
q
j
⋅ + ⋅ =
P
ˆ
ii
P
ˆ
ij
P
ˆ
ii
11.17689
S
ii
RD
i
 ln ⋅ =
P
ˆ
ij
11.17689
S
ij
D
ij
 ln ⋅ =
RD
i
P
ˆ
primitive
(5.8)
0812_frame_C05.fm Page 112 Friday, July 20, 2001 3:06 PM
Shunt Admittance of Overhead and Underground Lines 113
matrix will be an ncond x ncond matrix. For a fourwire grounded wye line,
the primitive coefﬁcient matrix will be of the form:
(5.11)
The dots (•) in Equation 5.11 are partitioning the matrix between the third
and fourth rows and columns. In partitioned form, Equation 5.11 becomes:
(5.12)
Because the neutral conductor is grounded, the matrix can be reduced using
the Kron reduction method to an nphase x nphase phase potential coefﬁcient
matrix :
(5.13)
The inverse of the potential coefﬁcient matrix will give the nphase x nphase
capacitance matrix :
(5.14)
For a twophase line, the capacitance matrix of Equation 5.14 will be 2 × 2.
A row and column of zeros must be inserted for the missing phase. For a
singlephase line, Equation 5.14 will result in a single element. Again, rows
and columns of zero must be inserted for the missing phase. In the case
of the singlephase line, the only nonzero term will be that of the phase in
use.
Neglecting the shunt conductance, the phase shunt admittance matrix is
given by:
µS/mile (5.15)
where
P
ˆ
primitive
[ ]
P
ˆ
aa
P
ˆ
ab
P
ˆ
ac
• P
ˆ
an
P
ˆ
ba
P
ˆ
bb
P
ˆ
bc
• P
ˆ
bn
P
ˆ
ca
P
ˆ
cb
P
ˆ
cc
• P
ˆ
cn
• • • • •
P
ˆ
na
P
ˆ
nb
P
ˆ
nc
• P
ˆ
nn
=
P
ˆ
primitive
[ ]
P
ˆ
ij [ ] P
ˆ
in
[ ]
P
ˆ
nj [ ] P
ˆ
nn
[ ]
=
P
abc
[ ]
P
abc
[ ] P
ˆ
ij
[ ] P
ˆ
in
[ ] P
ˆ
nn
[ ]
1 –
P
ˆ
nj
[ ] ⋅ ⋅ – =
C
abc
[ ]
C
abc
[ ] P
abc
[ ]
1 –
=
y
abc
[ ] 0 j ω C
abc
[ ] ⋅ ⋅ + =
ω 2 π f ⋅ ⋅ 2 π 60 ⋅ ⋅ 376.9911 = = =
0812_frame_C05.fm Page 113 Friday, July 20, 2001 3:06 PM
114 Distribution System Modeling and Analysis
Example 5.1
Determine the shunt admittance matrix for the overhead line in Example 4.1.
Assume that the neutral conductor is 25 ft. above ground.
The diameters of the phase and neutral conductors from the conductor
table (Appendix A) are
Conductor: 336,400 26/7 ACSR: d
c
= 0.721 inches, RD
c
= 0.03004 ft.,
4/0 6/1 ACSR: d
s
= 0.563 inches, RD
s
= 0.02346 ft.
For the conﬁguration, the distances between conductors and images in matrix
form are:
The selfprimitive potential coefﬁcient for phase a and the mutual primitive
potential coefﬁcient between phases a and b are
mile/µF
mile/µF
Using Equations 5.9 and 5.10, the total primitive potential coefﬁcient matrix
is computed to be
mile/µF
Since the fourth conductor (neutral) is grounded, the Kron reduction method
is used to compute the phase potential coefﬁcient matrix. Because only one
row and column need to be eliminated, the term is a single element, so
the Kron reduction equation for this case can be modiﬁed to:
where i = 1, 2, 3 and j = 1, 2, 3.
S [ ]
58 58.0539 58.4209 54.1479
58.0539 58 58.1743 54.0208
58.4209 58.1743 58 54.0833
54.1479 54.0208 54.0833 50
ft =
P
ˆ
aa
11.17689 ln
58
0.03004
 84.5600 = =
P
ˆ
ab
11.17689 ln
58.0539
2.5
 35.1522 = =
P
ˆ
primitive
[ ]
84.5600 35.1522 23.7147 25.2469
35.4522 84.5600 28.6058 28.359
23.7147 28.6058 84.5600 26.6131
25.2469 28.359 26.6131 85.6659
=
P
ˆ
44
[ ]
P
ij
P
ˆ
ij
P
ˆ
i4
P
ˆ
4j
⋅
P
ˆ
44
 – =
0812_frame_C05.fm Page 114 Friday, July 20, 2001 3:06 PM
Shunt Admittance of Overhead and Underground Lines 115
For example, the value of P
cb
is computed to be
Following the Kron reduction, the phase potential coefﬁcient matrix is
mile/µF
Invert to determine the shunt capacitance matrix:
µf/mile
Multiply by the radian frequency to determine the ﬁnal threephase
shunt admittance matrix:
µS/mile
5.3 Concentric Neutral Cable Underground Lines
Most underground distribution lines consist of one or more concentric neutral
cables. Figure 5.4 illustrates a basic concentric neutral cable with center con
ductor (black) being the phase conductor and the concentric neutral strands
(gray) displaced equally around a circle of radius R
b
. Referring to Figure 5.4
the following deﬁnitions apply:
R
b
= radius of a circle passing through the centers of the neutral
strands
d
c
= diameter of the phase conductor
d
s
= diameter of a neutral strand
k = total number of neutral strands
P
cb
P
ˆ
3,2
P
ˆ
3,4
P
ˆ
4,2
⋅
P
ˆ
4,4
 – 28.6058 = =
26.6131 28.359 ⋅
85.6659
 – 19.7957 =
P
abc
[ ]
77.1194 26.7944 15.8714
26.7944 75.172 19.7957
15.8714 19.7957 76.2923
=
P
abc
[ ]
C
abc
[ ] P [ ]
1 –
0.015 0.0049 – 0.0019 –
0.0019 – 0.0159 0.0031 –
0.0019 – 0.0031 – 0.0143
= =
C
abc
[ ]
y
abc
[ ] j 376.9911 C
abc
[ ] ⋅ ⋅
j5.6712 j1.8362 – j0.7034 –
j1.8362 – j5.9774 j1.169 –
j0.7034 – j1.169 – j5.3911
= =
0812_frame_C05.fm Page 115 Friday, July 20, 2001 3:06 PM
116 Distribution System Modeling and Analysis
The concentric neutral strands are grounded so that they are all at the same
potential. Because of the stranding, it is assumed that the electric ﬁeld created
by the charge on the phase conductor will be conﬁned to the boundary of
the concentric neutral strands. In order to compute the capacitance between
the phase conductor and ground, the general voltage drop of Equation 5.2
will be applied. Since all of the neutral strands are at the same potential, it
is only necessary to determine the potential difference between phase con
ductor p and Strand 1.
(5.16)
where
It is assumed that each of the neutral strands carries the same charge such
that:
(5.17)
Equation 5.16 can be simpliﬁed:
(5.18)
FIGURE 5.4
Basic concentric neutral cable.
D
12
R
b
i
5
j
4
3
1
2 k
0
12
d
c
R
b
ds
V
p1
1
2πε
 q
p
ln
R
b
RD
c
 q
1
ln
RD
s
R
b
 q
2
ln
D
12
R
b

…
q
i
+ + ln
D
1i
R
b

…
q
k
ln + +
D
k1
R
b
 + +
=
RD
c
d
c
2
 =
RD
s
d
s
2
 =
q
1
q
2
q
i
q
k
q
p
k
 – = = = =
V
p1
1
2πε
 q
p
ln
R
b
RD
c

q
p
k
 ln
RD
s
R
b
 ln
D
12
R
b

…
ln
D
1i
R
b

…
ln
D
1k
R
b
 + + + + +
– =
V
p1
q
p
2πε
 ln
R
b
RD
c

1
k
 ln
RD
s
D
12
D
1i
…
D
1k
⋅ ⋅ ⋅ ⋅
R
b
k

– =
0812_frame_C05.fm Page 116 Friday, July 20, 2001 3:06 PM
Shunt Admittance of Overhead and Underground Lines
117
The numerator of the second
ln
term in Equation 5.18 needs to be expanded.
The numerator represents the product of the radius and the distances
between
Strand
i
and all of the other strands. Referring to Figure 5.4, the
following relations apply:
In general, the angle between
Strand
1 and any other
Strand
i
is given by:
(5.19)
The distances between the various strands are given by:
The distance between
Strand
1 and any other
Strand
i
is given by:
(5.20)
Equation 5.20 can be used to expand the numerator of the second
ln
term
of Equation 5.18:
(5.21)
The term inside the bracket in Equation 5.21 is a trigonometric identity that
is merely equal to the number of strands
k
.
1
Using that identity, Equation 5.18
becomes:
(5.22)
θ
12
2 π ⋅
k
 =
θ
13
2 θ
12
⋅
4 π ⋅
k
 = =
θ
1i
i 1 – ( ) θ
12
⋅
i 1 – ( ) 2π ⋅
k
 = =
D
12
2 R
b
sin
θ
12
2

\ !
( \
⋅ ⋅ 2 R
b
sin
π
k

\ !
( \
⋅ ⋅ = =
D
13
2 R
b
sin
θ
13
2

\ !
( \
⋅ ⋅ 2 R
b
sin
2π
k

\ !
( \
⋅ ⋅ = =
D
1i
2 R
b
sin
θ
1i
2

\ !
( \
⋅ ⋅ 2 R
b
sin
i 1 – ( ) π ⋅
k

\ !
( \
⋅ ⋅ = =
RD
s
D
12
…
D
1i
…
D
1k
⋅ ⋅ ⋅ ⋅ ⋅
RD =
s
R
b
k−1
2 sin
π
k

\ !
( \
2sin
2π
k

\ !
( \
…
2
i 1 – ( )π
k

\ !
( \
 
¦ ¦
 
…
2sin
k 1 – ( )
k

\ !
( \
 
¦ ¦
 
⋅ ⋅ sin ⋅ ⋅ ⋅
⋅
V
p1
q
p
2πε

ln
R
b
RD
c

1
k
 ln
k RD
s
R
b
k−1
⋅ ⋅
R
b
k

\ !
 
( \
–
=
V
p1
q
p
2πε

ln
R
b
RD
c

1
k
 ln
k RD
s
⋅
R
b

\ !
( \
–
=
0812_frame_C05.fm Page 117 Monday, October 28, 2002 9:51 AM
118 Distribution System Modeling and Analysis
Equation 5.22 gives the voltage drop from the phase conductor to Neutral
Strand 1. Care must be taken that the units for the various radii are the same.
Typically, underground spacings are given in inches, so the radii of the phase
conductor (RD
c
) and the strand conductor (RD
s
) should be speciﬁed in inches.
Since the neutral strands are all grounded, Equation 5.22 gives the voltage
drop between the phase conductor and ground. Therefore, the capacitance
from phase to ground for a concentric neutral cable is given by:
(5.23)
where
ε = ε
0
ε
r
= permittivity of the medium
ε
0
= permittivity of free space = 0.01420 µF/mile
ε
r
= relative permittivity of the medium.
The electric ﬁeld of a cable is conﬁned to the insulation material. Various
types of insulation materials are used and each will have a range of values
for the relative permittivity. Table 5.1 gives the range of values of relative
permittivity for four common insulation materials.
2
Crosslinked polyethly
ene is a very popular insulation material. If the minimum value of relative
permittivity is assumed (2.3), the equation for the shunt admittance of the
concentric neutral cable is given by:
µS/mile (5.24)
Example 5.2
Determine the threephase shunt admittance matrix for the concentric neu
tral line of Example 4.2 in Chapter 4.
TABLE 5.1
Typical Values of Relative Permittivity (ε
r
)
Material
Range of Values
of Relative Permittivity
Polyvinyl Chloride (PVC) 3.4–8.0
EthylenePropylene Rubber (EPR) 2.5–3.5
Polyethylene (PE) 2.5–2.6
CrossLinked Polyethlyene (XLPE) 2.3–6.0
C
pg
q
p
V
p1

2πε
ln
R
b
RD
c

1
k

k RD
s
⋅
R
b
 ln –
 = =
y
ag
0 j + =
77.3619
ln
R
b
RD
c

1
k
ln
k RD
s
⋅
R
b
 –

0812_frame_C05.fm Page 118 Friday, July 20, 2001 3:06 PM
Shunt Admittance of Overhead and Underground Lines 119
From Example 4.2:
R
b
= R = 0.0511 ft. = 0.6132 inch
Diameter of the 250,000 AA phase conductor = 0.567 inch,
Diameter of the #14 CU concentric neutral strand = 0.0641 inch,
inch
Substitute into Equation 5.24:
The phase admittance for this threephase underground line is
µS/mile
5.4 TapeShielded Cable Underground Lines
A tapeshielded cable is shown in Figure 5.5. Referring to Figure 5.5, R
b
is
the radius of a circle passing through the center of the tape shield. As with
the concentric neutral cable, the electric ﬁeld is conﬁned to the insulation
so that the relative permittivities of Table 5.1 will apply.
The tapeshielded conductor can be visualized as a concentric neutral cable
where the number of strands k has become inﬁnite. When k in Equation 5.24
RD
c
0.567
2
 0.2835 = = inch
RD
s
0.0641
2
 0.03205 = =
y
ag
j
77.3619
ln
R
b
RD
c

1
k
 ln ⋅
k RD
s
⋅
R
b

–
 =
y
ag
j =
77.3619
ln
0.6132
0.2835

1
13
 ln
13 0.03205 ⋅
0.6132

⋅ –
 j96.5569 =
y
abc
[ ]
j96.5569 0 0
0 j96.5569 0
0 0 j96.5569
=
µS/mile
0812_frame_C05.fm Page 119 Friday, July 20, 2001 3:06 PM
120 Distribution System Modeling and Analysis
approaches inﬁnity, the second term in the denominator approaches zero.
Therefore, the equation for the shunt admittance of a tapeshielded conduc
tor becomes:
µS/mile (5.25)
Example 5.3
Determine the shunt admittance of the singlephase tapeshielded cable of
Example 4.3 in Chapter 4. From Example 4.3, the outside diameter of the
tape shield is 0.88 inch. The thickness of the tape shield (T) is 5 mils. The
radius of a circle passing through the center of the tape shield is
The diameter of the 1/0 AA phase conductor = 0.368 inch
inch
Substitute into Equation 5.25:
µS/mile
FIGURE 5.5
Tapeshielded conductor.
AL or CU Phase
Conductor
Insulation
Jacket
CU Tape Shield
R
b
y
ag
0 j
77.3619
ln
R
b
RD
c

 + =
T
5
1000
 0.005 = =
R
b
d
s
T –
2

0.88 0.005 –
2
 0.4375 = = = inch
RD
c
d
p
2

0.368
2
 0.1840 = = =
y
bg
j
77.3619
ln
R
b
RD
c

 j
77.3619
ln
0.4375
0.184

 j89.3179 = = =
0812_frame_C05.fm Page 120 Friday, July 20, 2001 3:06 PM
Shunt Admittance of Overhead and Underground Lines 121
The line is on phase b so that the phase admittance matrix becomes:
µS/mile
5.5 Sequence Admittance
The sequence admittances of a threephase line can be determined in much
the same manner as the sequence impedances were determined in Chapter 4.
Assume that the 3 × 3 admittance matrix is given in S/mile. Then the three
phase capacitance currents as a function of the linetoground voltages are
given by:
(5.26)
(5.27)
Applying the symmetrical component transformations:
(5.28)
From Equation 5.28 the sequence admittance matrix is given by:
(5.29)
For a threephase overhead line with unsymmetrical spacing, the sequence
admittance matrix will be full. That is, the offdiagonal terms will be nonzero.
However, a threephase underground line with three identical cables will
only have the diagonal terms since there is no mutual capacitance between
phases. In fact, the sequence admittances will be exactly the same as the
phase admittances.
y
abc
[ ]
0 0 0
0 j89.3179 0
0 0 0
=
Icap
a
Icap
b
Icap
c
y
aa
y
ab
y
ac
y
ba
y
bb
y
bc
y
ca
y
cb
y
cc
V
ag
V
bg
V
cg
⋅ =
Icap
abc
[ ] y
abc
[ ] VLG
abc
[ ] ⋅ =
Icap
012
[ ] A
s
[ ]
1 –
Icap
abc
[ ] ⋅ A
s
[ ]
1 –
y
abc
[ ] A
s
[ ] VLG
012
[ ] ⋅ ⋅ ⋅ = =
y
012
[ ] A
s
[ ]
1 –
y
abc
[ ] A
s
[ ] ⋅ ⋅
y
00
y
01
y
02
y
10
y
11
y
12
y
20
y
21
y
22
= =
0812_frame_C05.fm Page 121 Friday, July 20, 2001 3:06 PM
122 Distribution System Modeling and Analysis
5.6 Summary
Methods for computing the shunt capacitive admittance for overhead and
underground lines have been presented in this chapter. Distribution lines are
typically so short that the shunt admittance can be ignored. However, there are
cases of long, lightly loaded overhead lines where the shunt admittance should
be included. Underground cables have a much higher shunt admittance per
mile than overhead lines. Again, there will be cases where the shunt admittance
of an underground cable should be included in the analysis process. When the
analysis is being done using a computer, the approach to take is to go ahead
and model the shunt admittance for both overhead and underground lines.
Why make a simplifying assumption when it is not necessary?
References
1. Glover, J.D. and Sarma, M., Power System Analysis and Design, 2nd edition,
PWSKent Publishing, Boston, 1995.
2. Arnold, T.P. and Mercier, C.D., Power Cable Manual, 2nd ed., Southwire Company,
Carrollton, GA, 1997.
Problems
5.1 Determine the phase admittance matrix and sequence admittance
matrix [Y
012
] in µS/mile for the threephase overhead line of Problem 4.1.
5.2 Determine the phase admittance matrix in µS/mile for the twophase
line of Problem 4.2.
5.3 Determine the phase admittance matrix in µS/mile for the singlephase
line of Problem 4.3.
5.4 Verify the results of Problems 5.1, 5.2, and 5.3 using RDAP.
5.5 Determine the phase admittance matrix and sequence admittance
matrix in µS/mile for the threephase line of Problem 4.5.
5.6 Determine the phase admittance matrix in µS/mile for the singlephase
concentric neutral cable of Problem 4.9.
5.7 Determine the phase admittance matrix and sequence admittance
matrix for the threephase concentric neutral line of Problem 4.10.
Y
abc
[ ]
0812_frame_C05.fm Page 122 Friday, July 20, 2001 3:06 PM
Shunt Admittance of Overhead and Underground Lines 123
5.8 Verify the results of Problems 5.6 and 5.7 using RDAP.
5.9 Determine the phase admittance matrix in µS/mile for the singlephase
tapeshielded cable line of Problem 4.12.
5.10 Determine the phase admittance for the threephase tapeshielded
cable line of Problem 4.13.
5.11 Verify the results of Problems 5.9 and 5.10 using RDAP.
0812_frame_C05.fm Page 123 Friday, July 20, 2001 3:06 PM
0812_frame_C05.fm Page 124 Friday, July 20, 2001 3:06 PM
125
6
Distribution System Line Models
The modeling of distribution overhead and underground line segments is a
critical step in the analysis of a distribution feeder. It is important to include
the actual phasing of the line and the correct spacing between conductors.
Chapters 4 and 5 developed the method for the computation of the phase
impedance and phase admittance matrices without assuming transposition
of the lines. Those matrices will be used in the models for overhead and
underground line segments.
6.1 Exact Line Segment Model
The exact model of a threephase, twophase, or singlephase overhead or
underground line is shown in Figure 6.1. When a line segment is twophase
(Vphase) or singlephase, some of the impedance and admittance values
will be zero. Recall in Chapters 4 and 5 that in all cases the phase impedance
and phase admittance matrices were 3
×
3. Rows and columns of zeros for
the missing phases represent twophase and singlephase lines. Therefore,
one set of equations can be developed to model all overhead and under
ground line segments. The values of the impedances and admittances in
Figure 6.1 represent the total impedances and admittances for the line. That is,
the phase impedance matrix derived in Chapter 4 has been multiplied by the
length of the line segment. The phase admittance matrix derived in Chapter 5
has also been multiplied by the length of the line segment.
For the line segment of Figure 6.1, the equations relating the input (
Node
n
)
voltages and currents to the output (
Node m
) voltages and currents are deve
loped as follows:
Kirchhoff’s current law applied at
Node m
:
(6.1)
Iline
a
Iline
b
Iline
c
n
I
a
I
b
I
c
m
1
2
 ·
Y
aa
Y
ab
Y
ac
Y
ba
Y
bb
Y
bc
Y
ca
Y
cb
Y
cc
·
V
ag
V
bg
V
cg
m
+
0812_frame_C06.fm Page 125 Friday, July 20, 2001 3:11 PM
126
Distribution System Modeling and Analysis
In condensed form Equation 6.1 becomes:
(6.2)
Kirchhoff’s voltage law applied to the model gives:
(6.3)
In condensed form Equation 6.3 becomes:
(6.4)
Substituting Equation 6.2 into Equation 6.4:
(6.5)
Collecting terms:
(6.6)
(6.7)
FIGURE 6.1
Threephase line segment model.
Vag
+
Vbg
+
Vcg
+
+
+
Vcg
+
Vag
Ic
Ib
Ia
Ib
Ia
Ic
  
[Yabc]
2
1
[Yabc]
2
1
  
[ICabc] [ICabc]
n
n
n
n
Vbg
n
n
n
m
m
m
m
m
m
m
Node  m Node  n
Z
bb
Z
aa
Z
cc
Z
ab
Z
bc
Z
ca
a
Iline
c
Iline
b
Iline
Iline
abc
[ ]
n
I
abc
[ ]
m
1
2
 Y
abc
[ ] · VLG
abc
[ ]
m
+
Vag
Vbg
Vcg
n
Vag
Vbg
Vcg
m
Z
aa
Z
ab
Z
ac
Z
ba
Z
bb
Z
bc
Z
ca
Z
cb
Z
cc
·
Iline
a
Iline
b
Iline
c
m
+
VLG
abc
[ ]
n
VLG
abc
[ ]
m
Z
abc
[ ] · Iline
abc
[ ]
m
+
VLG
abc
[ ]
n
VLG
abc
[ ]
m
Z
abc
[ ] · I
abc
[ ]
m
1
2
 Y
abc
[ ] · VLG
abc
[ ]
m
+
 
¦ 
 
+
VLG
abc
[ ]
n
U [ ]
1
2
 · Z
abc
[ ] · Y
abc
[ ] +
 
¦ 
 
· VLG
abc
[ ]
m
Z
abc
[ ] · I
abc
[ ]
m
+
where U [ ]
1 0 0
0 1 0
0 0 1
0812_frame_C06.fm Page 126 Friday, July 20, 2001 3:11 PM
Distribution System Line Models
127
Equation 6.6 is of the general form:
(6.8)
(6.9)
(6.10)
The input current to the line segment at
Node n
is
(6.11)
In condensed form Equation 6.11 becomes:
(6.12)
Substitute Equation 6.2 into Equation 6.12:
(6.13)
Substitute Equation 6.6 into Equation 6.13:
(6.14)
Collecting terms in Equation 6.14:
(6.15)
VLG
abc
[ ]
n
a [ ] · VLG
abc
[ ]
m
b [ ] · I
abc
[ ]
m
+
where a [ ] U [ ]
1
2
 · Z
abc
[ ] · Y
abc
[ ] +
b [ ] Z
abc
[ ]
I
a
I
b
I
c
n
Iline
a
Iline
b
Iline
c
1
2
 ·
Y
aa
Y
ab
Y
ac
Y
ba
Y
bb
Y
bc
Y
ca
Y
cb
Y
cc
·
Vag
Vbg
Vcg
n
+
I
abc
[ ]
n
Iline
abc
[ ]
m
1
2
 · Y
abc
[ ] · VLG
abc
[ ]
n
+
I
abc
[ ]
n
I
abc
[ ]
m
1
2
 Y
abc
[ ] · VLG
abc
[ ]
m
1
2
 · Y
abc
[ ] · VLG
abc
[ ]
n
+ +
I
abc
[ ]
n
I
abc
[ ]
m
1
2
 Y
abc
[ ] · VLG
abc
[ ]
m
1
2
 · Y
abc
[ ] + +
U [ ]
1
2
 · Z
abc
[ ] · Y
abc
[ ] +
 
¦ 
 
· VLG
abc
[ ]
m
Z
abc
[ ] · I
abc
[ ]
m
+
( J
' J
 ·
×
I
abc
[ ]
n
Y
abc
[ ]
1
4
 · Y
abc
[ ] Z
abc
[ ] ⋅ · Y
abc
[ ] +
 
¦ 
 
· VLG
abc
[ ]
m
U [ ]
1
2
 · Y
abc
[ ] · Z
abc
[ ] +
 
¦ 
 
I
abc
[ ]
m
+
0812_frame_C06.fm Page 127 Friday, July 20, 2001 3:11 PM
128 Distribution System Modeling and Analysis
Equation 6.15 is of the form:
(6.16)
(6.17)
(6.18)
Equations 6.8 and 6.16 can be put into partitioned matrix form:
(6.19)
Equation 6.19 is very similar to the equation used in transmission line analy
sis when the ABCD parameters have been deﬁned.
1
In the case here, the
abcd parameters are 3 × 3 matrices rather than single variables and will be
referred to as the “generalized line matrices.”
Equation 6.19 can be turned around to solve for the voltages and currents
at Node m in terms of the voltages and currents at Node n:
(6.20)
The inverse of the abcd matrix is simple because the determinant is
(6.21)
Using the relationship of Equation 6.21, Equation 6.20 becomes:
(6.22)
Since the matrix [a] is equal to the matrix [d], Equation 6.22 in expanded
form becomes:
(6.23)
(6.24)
Sometimes it is necessary to compute the voltages at Node m as a function
of the voltages at Node n and the currents entering Node m. This is true in
the iterative technique that is developed in Chapter 10.
I
abc
[ ]
n
c [ ] · VLG
abc
[ ]
m
d [ ] · I
abc
[ ]
m
+
where c [ ] Y
abc
[ ]
1
4
 · Y
abc
[ ] Z
abc
[ ] ⋅ · Y
abc
[ ] +
d [ ] U [ ]
1
2
 · Z
abc
[ ] · Y
abc
[ ] +
VLG
abc
[ ]
n
I
abc
[ ]
n
a [ ] b [ ]
c [ ] d [ ]
·
VLG
abc
[ ]
m
I
abc
[ ]
m
VLG
abc
[ ]
m
I
abc
[ ]
m
a [ ] b [ ]
c [ ] d [ ]
1 –
·
VLG
abc
[ ]
n
I
abc
[ ]
n
a [ ] · d [ ] b [ ] · c [ ] – U [ ]
VLG
abc
[ ]
m
I
abc
[ ]
m
d [ ] b [ ] –
c [ ] – a [ ]
·
VLG
abc
[ ]
n
I
abc
[ ]
n
VLG
abc
[ ]
m
a [ ] · VLG
abc
[ ]
n
b [ ] · I
abc
[ ]
n
–
I
abc
[ ]
m
c [ ] – · VLG
abc
[ ]
n
d [ ] · I
abc
[ ]
n
+
0812_frame_C06.fm Page 128 Friday, July 20, 2001 3:11 PM
Distribution System Line Models 129
Solving Equation 6.8 for the Bus m voltages gives:
(6.25)
Equation 6.25 is of the form:
(6.26)
(6.27)
(6.28)
Because the mutual coupling between phases on the line segments are not
equal, there will be different values of voltage drop on each of the three phases.
As a result, the voltages on a distribution feeder become unbalanced even
when the loads are balanced. A common method of describing the degree
of unbalance is to use the National Electrical Manufacturers Association
(NEMA) deﬁnition of voltage unbalance as given in Equation 6.29
2
:
(6.29)
Example 6.1
A balanced threephase load of 6000 kVA, 12.47 kV, 0.9 lagging power factor
is being served at Node m of a 10,000ft. threephase line segment. The
conﬁguration and conductors of the line segment are those of Example 4.1.
Determine the generalized line constant matrices [a], [b], [c], and [d]. Using
the generalized matrices, determine the linetoground voltages and line
currents at the source end (Node n) of the line segment.
SOLUTION
The phase impedance matrix and the shunt admittance matrix for the line
segment as computed in Examples 4.1 and 5.1 are
Ω/mile
VLG
abc
[ ]
m
a [ ]
1 –
· VLG
abc
[ ]
n
b [ ] · I
abc
[ ]
m
– { ¦
VLG
abc
[ ]
m
A [ ] · VLG
abc
[ ]
n
B [ ] · I
abc
[ ]
m
–
where A [ ] a [ ]
1 –
B [ ] a [ ]
1 –
· b [ ]
V
unbalance
Maximum deviation from average
V
average
 · 100%
z
abc
[ ]
0.4576 j1.0780 + 0.1560 j.5017 + 0.1535 j0.3849 +
0.1560 j0.5017 + 0.4666 j1.0482 + 0.1580 j0.4236 +
0.1535 j0.3849 + 0.1580 j0.4236 + 0.4615 j1.0651 +
y
abc
[ ] j · 376.9911 · C
abc
[ ]
j5.6712 j1.8362 – j0.7034 –
j1.8362 – j5.9774 j1.169 –
j0.7034 – j1.169 – j5.3911
µ S/mile
0812_frame_C06.fm Page 129 Friday, July 20, 2001 3:11 PM
130 Distribution System Modeling and Analysis
For the 10,000ft. line segment, the total phase impedance matrix and shunt
admittance matrix are
Ω
It should be noted that the elements of the phase admittance matrix are very
small. The generalized matrices computed according to Equations 6.9, 6.10,
6.17, and 6.18 are
Because the elements of the phase admittance matrix are so small, the [a] and
[d] matrices appear to be the unity matrix. If more signiﬁcant ﬁgures are dis
played, the (1,1) element of these matrices is
Also, the elements of the [c] matrix appear to be zero. Again, if more signiﬁ
cant ﬁgures are displayed, the 1,1 term is
Z
abc
[ ]
0.8667 j2.0417 + 0.2955 j0.9502 + 0.2907 j0.7290 +
0.2955 j0.9502 + 0.8837 j1.9852 + 0.2992 j0.8023 +
0.2907 j0.7290 + 0.2992 j0.8023 + 0.8741 j2.0172 +
Y
abc
[ ]
j10.7409 j3.4777 – j1.3322 –
j3.4777 – j11.3208 j2.2140 –
j1.3322 – j2.2140 – j10.2104
µS
a [ ] U [ ]
1
2
 · Z
abc
[ ] · Y
abc
[ ] +
1.0 0 0
0 1.0 0
0 0 1.0
b [ ] Z
abc
[ ]
0.8667 j2.0417 + 0.2955 j0.9502 + 0.2907 j0.7290 +
0.2955 j0.9502 + 0.8837 j1.9852 + 0.2992 j0.8023 +
0.2907 j0.7290 + 0.2992 j0.8023 + 0.8741 j2.0172 +
c [ ]
0 0 0
0 0 0
0 0 0
d [ ]
1.0 0 0
0 1.0 0
0 0 1.0
a
1,1
0.99999117 j0.00000395 +
c
1,1
0.0000044134 – j0.0000127144 +
0812_frame_C06.fm Page 130 Friday, July 20, 2001 3:11 PM
Distribution System Line Models 131
The point here is that for all practical purposes, the phase admittance matrix
can be neglected.
The magnitude of the linetoground voltage at the load is
Selecting the phaseatoground voltage as reference, the linetoground voltage
matrix at the load is
The magnitude of the load currents is
For a 0.9 lagging power factor, the load current matrix is
The linetoground voltages at Node n are computed to be
It is important to note that the voltages at Node n are unbalanced even though
the voltages and currents at the load (Node m) are perfectly balanced. This
is a result of the unequal mutual coupling between phases. The degree of
voltage unbalance is of concern since, for example, the operating character
istics of a threephase induction motor are very sensitive to voltage unbal
ance. Using the NEMA deﬁnition for voltage unbalance (Equation 6.29), the
VLG
12470
3
 7199.56
V
ag
V
bg
V
cg
m
7199.56/0
7199.56/ 120 –
7199.56/120
V
I
m
6000
3 · 12.47
 277.79
I
abc
[ ]
m
277.79/ 25.84 –
277.79/ 145.84 –
277.79/94.16
A
VLG
abc
[ ]
n
a [ ] · VLG
abc
[ ]
m
b [ ] · I
abc
[ ]
m
+
7538.70/1.57
7451.25/ 118.30 –
7485.11/121.93
V
0812_frame_C06.fm Page 131 Friday, July 20, 2001 3:11 PM
132 Distribution System Modeling and Analysis
voltage unbalance is
Although this may not seem like a large unbalance, it does give an indication
of how the unequal mutual coupling can generate an unbalance. It is impor
tant to know that NEMA standards require that induction motors be derated
when the voltage unbalance exceeds 1.0%.
Selecting rated linetoground voltage as base (7199.56), the perunit volt
ages at Bus n are
By converting the voltages to perunit it is easy to see that the voltage drop
by phase is 4.71% for Phase a, 3.50% for Phase b, and 3.97% for Phase c.
The line currents at Node n are computed to be
Comparing the computed line currents at Node n to the balanced load cur
rents at Node m, a very slight difference is noted that is another result of the
unbalanced voltages at Node n and the shunt admittance of the line segment.
6.2 The Modiﬁed Line Model
It was demonstrated in Example 6.1 that the shunt admittance of a line is so
small that it can be neglected. Figure 6.2 shows the modiﬁed line segment
model with the shunt admittance neglected.
V
average
V
ag n
V
bg n
V
cg n
+ +
3

7538.70 7451.25 7485.11 + +
3
 7491.69
Vdeviation
max
7538.70 7491.69 – 47.01
V
unbalance
47.01
7491.70
 · 100% 0.6275%
V
ag
V
bg
V
cg
n
1
7199.56

7538.70/1.577
7451.25/ 118.30 –
7485.11/121.93
1.0471/1.57
1.0350/ 118.30 –
1.0397/121.93
perunit
I
abc
[ ]
n
c [ ] · VLG
abc
[ ]
m
d [ ] · I
abc
[ ]
m
+
277.71/ 25.83 –
277.73/ 148.82 –
277.73/94.17
A
0812_frame_C06.fm Page 132 Friday, July 20, 2001 3:11 PM
Distribution System Line Models 133
When the shunt admittance is neglected, the generalized matrices become:
(6.30)
(6.31)
(6.32)
(6.33)
(6.34)
(6.35)
If the line is a threewire delta, the voltage drops down the line must be in
terms of the linetoline voltages and line currents. However, it is possible
to use “equivalent” linetoneutral voltages so that the equations derived to
this point will still apply. Writing the voltage drop equations in terms of line
toline voltages for the line in Figure 6.2 results in:
(6.36)
where (6.37)
FIGURE 6.2
Modiﬁed line segment model.
Vag
+
Vbg
+
Vcg
+
+
+
Vcg
+
Vag
Ic
Ib
Ia
Ib
Ia
Ic
  
  
n
n
n
n
Vbg
n
n
m
m
m
m
m
m
Node  m Node  n
Z
bb
Z
aa
Z
cc
Z
ab
Z
bc
Z
ca
a
Iline
c
Iline
b
Iline
a [ ] U [ ]
b [ ] Z
abc
[ ]
c [ ] 0 [ ]
d [ ] U [ ]
A [ ] U [ ]
B [ ] Z
abc
[ ]
V
ab
V
bc
V
ca
n
V
ab
V
bc
V
ca
m
vdrop
a
vdrop
b
vdrop
c
vdrop
b
vdrop
c
vdrop
a
– +
vdrop
a
vdrop
b
vdrop
c
Z
aa
Z
ab
Z
ac
Z
ba
Z
bb
Z
bc
Z
ca
Z
cb
Z
cc
·
Iline
a
Iline
b
Iline
c
0812_frame_C06.fm Page 133 Friday, July 20, 2001 3:11 PM
134 Distribution System Modeling and Analysis
Expanding Equation 6.36 for Phase ab:
(6.38)
but (6.39)
Substitute Equations 6.39 into Equation 6.38:
(6.40)
Equation 6.40 can be broken into two parts in terms of equivalent lineto
neutral voltages:
(6.41)
The conclusion here is that it is possible to work with equivalent lineto
neutral voltages in a threewire delta line. This is very important since it
makes the development of general analyses techniques the same for four
wire wye and threewire delta systems.
Example 6.2
The line of Example 6.1 will be used to supply an unbalanced load at Node m.
Assume that the voltages at the source end (Node n) are balanced threephase
at 12.47 kV linetoline. The balanced linetoground voltages are
The unbalanced currents measured at the source end are given by:
Determine the linetoground and linetoline voltages at the load end (Node m)
using the modiﬁed line model. Determine also the voltage unbalance and the
complex powers of the load.
Vab
n
Vab
m
vdrop
a
vdrop
b
– +
Vab
n
Van
n
Vbn
n
–
Vab
m
Van
m
Vbn
m
–
Van
n
Vbn
n
– Van
m
Vbn
m
– vdrop
a
vdrop
b
– +
Van
n
Van
m
vdrop
a
+
Vbn
n
Vbn
m
vdrop
b
+
VLG
abc
[ ]
n
7199.56/0
7199.56/ 120 –
7199.56/120
V
I
a
I
b
I
c
n
249.97/ 24.5 –
277.56/ 145.8 –
305.54/95.2
A
0812_frame_C06.fm Page 134 Friday, July 20, 2001 3:11 PM
Distribution System Line Models 135
SOLUTION
The [A] and [B] matrices for the modiﬁed line model are
Since this is the approximate model, is equal to Therefore:
The linetoground voltages at the load end are
For this condition the average load voltage is
The maximum deviation from the average is on Phase c so that:
The linetoline voltages at the load can be computed by:
A [ ] U [ ]
1 0 0
0 1 0
0 0 1
B [ ] Z
abc
[ ]
0.8666 j2.0417 + 0.2954 j0.9501 + 0.2907 j0.7290 +
0.2954 j0.9501 + 0.8838 j1.9852 + 0.2993 j0.8024 +
0.2907 j0.7290 + 0.2993 j0.8024 + 0.8740 j2.0172 +
Ω
I
abc
[ ]
m
I
abc
[ ]
n
.
I
a
I
b
I
c
m
249.97/ 24.5 –
277.56/ 145.8 –
305.54/95.2
A
VLG
abc
[ ]
m
A [ ] · VLG
abc
[ ]
n
B [ ] · I
abc
[ ]
m
–
6942.53/ 1.47 –
6918.35/ 121.55 –
6887.71/117.31
V
V
average
6942.53 6918.35 6887.71 + +
3
 6916.20 V
Vdeviation
max
6887.71 6916.20 – 28.49
V
unbalance
28.49
6916.20
 · 100 0.4119%
V
ab
V
bc
V
ca
1 1 – 0
0 1 1 –
1 – 0 1
·
6942.53/ 1.47 –
6918.35/ 121.55 –
6887.71/117.31
12008.43/28.43
12024.62/ 92.19 –
11903.23/148.05
V
0812_frame_C06.fm Page 135 Friday, July 20, 2001 3:11 PM
136 Distribution System Modeling and Analysis
The complex powers of the load are
6.3 The Approximate Line Segment Model
Many times the only data available for a line segment will be the positive
and zero sequence impedances. The approximate line model can be devel
oped by applying the reverse impedance transformation from symmetrical
component theory.
Using the known positive and zero sequence impedances, the sequence
impedance matrix is given by:
(6.42)
The reverse impedance transformation results in the following approximate
phase impedance matrix:
(6.43)
(6.44)
Notice that the approximate impedance matrix is characterized by the three
diagonal terms being equal and all mutual terms being equal. This is the
same result that is achieved if the line is assumed to be transposed. Applying
the approximate impedance matrix, the voltage at Node n is computed to be
(6.45)
S
a
S
b
S
c
1
1000
 ·
V
ag
· I
a
ء
V
bg
· I
b
ء
V
cg
· I
c
ء
1735.42/23.03
1920.26/24.25
2104.47/22.11
kVA
Z
seq
[ ]
Z
0
0 0
0 Z
+
0
0 0 Z
+
Z
approx
[ ] A
s
[ ] · Z
seq
[ ] · A
s
[ ]
1 –
Z
approx
[ ]
1
3
 ·
2 · Z
+
Z
0
+ ( ) Z
0
Z
+
– ( ) Z
0
Z
+
– ( )
Z
0
Z
+
– ( ) 2 · Z
+
Z
0
+ ( ) Z
0
Z
+
– ( )
Z
0
Z
+
– ( ) Z
0
Z
+
– ( ) 2 · Z
+
Z
0
+ ( )
V
ag
V
bg
V
cg
n
V
ag
V
bg
V
cg
m
1
3
 ·
2 · Z
+
Z
0
+ ( ) Z
0
Z
+
– ( ) Z
0
Z
+
– ( )
Z
0
Z
+
– ( ) 2 · Z
+
Z
0
+ ( ) Z
0
Z
+
– ( )
Z
0
Z
+
– ( ) Z
0
Z
+
– ( ) 2 · Z
+
Z
0
+ ( )
·
I
a
I
b
I
c
m
+
0812_frame_C06.fm Page 136 Friday, July 20, 2001 3:11 PM
Distribution System Line Models 137
In condensed form, Equation 6.45 becomes:
(6.46)
Note that Equation 6.46 is of the form:
(6.47)
where [a] unity matrix
[b] [Z
approx
]
Equation 6.45 can be expanded and an equivalent circuit for the approximate
line segment model can be developed. Solving Equation 6.45 for the Phase a
voltage at Node n results in:
(6.48)
Modify Equation 6.48 by adding and subtracting the term and
then combining terms and simplifying:
(6.49)
The same process can be followed in expanding Equation 6.45 for Phases b
and c. The ﬁnal results are
(6.50)
(6.51)
Figure 6.3 illustrates the approximate line segment model. Figure 6.3 is a
simple equivalent circuit for the line segment since no mutual coupling has
VLG
abc
[ ]
n
VLG
abc
[ ]
m
Z
approx
[ ] · I
abc
[ ]
m
+
VLG
abc
[ ]
n
a [ ] VLG
abc
[ ]
m
b [ ] · I
abc
[ ]
m
+
Vag
n
Vag
m
1
3
 2Z
+
Z
0
+ ( )I
a
Z
0
Z
+
– ( )I
b
Z
0
Z
+
+ ( )I
c
+ + { ¦ +
Z
0
Z
+
– ( ) I
a
Vag
n
Vag
m
1
3

2Z
+
Z
0
+ ( )I
a
Z
0
Z
+
– ( )I
b
Z
0
Z
+
– ( )I
c
+ +
Z
0
Z
+
– ( )I
a
Z
0
Z
+
– ( )I
a
– +
 
¦ 
 
+
Vag
n
Vag
m
1
3
 3Z
+
( )I
a
Z
0
Z
+
– ( ) I
a
I
b
I
c
+ + ( ) + { ¦ +
Vag
n
Vag
m
Z
+
· I
a
Z
0
Z
+
– ( )
3
 · I
a
I
b
I
c
+ + ( ) + +
Vbg
n
Vbg
m
Z
+
· I
b
Z
0
Z
+
– ( )
3
 · I
a
I
b
I
c
+ + ( ) + +
Vcg
n
Vcg
m
Z
+
· I
c
Z
0
Z
+
– ( )
3
 · I
a
I
b
I
c
+ + ( ) + +
0812_frame_C06.fm Page 137 Friday, July 20, 2001 3:11 PM
138 Distribution System Modeling and Analysis
to be modeled. It must be understood, however, that the equivalent circuit can
only be used when transposition of the line segment has been assumed.
Example 6.3
The line segment of Example 4.1 is to be analyzed assuming that the line
has been transposed. In Example 4.1 the positive and zero sequence imped
ances were computed to be
Ω/mile
Assume that the load at Node m is the same as in Example 6.1, that is
kVA 6000, kVLL 12.47, Power factor 0.8 lagging
Determine the voltages and currents at the source end (Node n) for this
loading condition.
SOLUTION
The sequence impedance matrix is
Ω/mile
Performing the reverse impedance transformation results in the approximate
phase impedance matrix:
Ω/mile
FIGURE 6.3
Approximate line segment model.
z
+
0.3061 j0.6270 +
z
0
0.7735 j1.9373 +
z
seq
[ ]
0.7735 j1.9373 + 0 0
0 0.3061 j0.6270 + 0
0 0 0.3061 j0.6270 +
z
approx
[ ] A
s
[ ] · z
seq
[ ] · A
s
[ ]
1 –
Z
approx
[ ]
0.4619 j1.0638 + 0.1558 j0.4368 + 0.1558 j0.4368 +
0.1558 j0.4368 + 0.4619 j1.0638 + 0.1558 j0.4368 +
0.1558 j0.4368 + 0.1558 j0.4368 + 0.4619 j1.0638 +
0812_frame_C06.fm Page 138 Friday, July 20, 2001 3:11 PM
Distribution System Line Models 139
For the 10,000ft. line, the phase impedance matrix and the [b] matrix are
Note in the approximate phase impedance matrix that the three diagonal terms
are equal and all of the mutual terms are equal. Again, this is an indication of
the transposition assumption.
From Example 6.1 the voltages and currents at Node m are
Using Equation 6.47:
Note that the computed voltages are balanced. In Example 6.1 it was shown
that when the line is modeled accurately, there is a voltage unbalance of
0.6275%. It should also be noted that the average value of the voltages at
Node n in Example 6.1 was 7491.69 volts.
The at Node n can also be computed using Equation 6.48:
Since the currents are balanced, this equation reduces to:
b [ ] Z
approx
[ ] z
approx
[ ] ·
10,000
5280

b [ ]
0.8748 j2.0147 + 0.2951 j0.8272 + 0.2951 j0.8272 +
0.2951 j0.8272 + 0.8748 j2.0147 + 0.2951 j0.8272 +
0.2951 j0.8272 + 0.2951 j0.8272 + 0.8748 j2.0147 +
Ω
VLG
abc
[ ]
m
7199.56/0
7199.56/ 120 –
7199.56/120
V
I
abc
[ ]
m
277.79/ 25.84 –
277.79/ 145.84 –
277.79/94.16
A
VLG
abc
[ ]
n
a [ ] · VLG
abc
[ ]
m
b [ ] · I
abc
[ ]
m
+
7491.65/ 1.73 –
7491.65/ 118.27 –
7491.65/ 121.73 –
V
V
ag
Vag
n
Vag
m
Z
+
· I
a
Z
0
Z
+
– ( )
3
 + · I
a
I
b
I
c
+ + ( ) +
Vag
n
Vag
m
Z
+
· I
a
+
7199.56/0 0.5797 j1.1875 + ( ) · 277.79/ 25.84 – + 7491.65/−1.73 V
0812_frame_C06.fm Page 139 Friday, July 20, 2001 3:11 PM
140 Distribution System Modeling and Analysis
It can be noted that when the loads are balanced and transposition has been
assumed, the threephase line can be analyzed as a simple singlephase equi
valent, as was done in the calculation above.
Example 6.4
Use the balanced voltages and unbalanced currents at Node n in Example 6.2
and the approximate line model to compute the voltages and currents at
Node m.
SOLUTION
From Example 6.2 the voltages and currents at Node n are given as:
The [A] and [B] matrices for the approximate line model are
where [A] Unity matrix
[B] [Z
approx
]
The voltages at Node m are determined by:
The voltage unbalance for this case is computed by:
Note that the approximate model has led to a higher voltage unbalance than
the exact model.
VLG
abc
[ ]
n
7199.56/0
7199.56/ 120 –
7199.56/120
V
I
a
I
b
I
c
n
249.97/ 24.5 –
277.56/ 145.8 –
305.54/95.2
A
VLG
abc
[ ]
m
A [ ] · VLG
abc
[ ]
n
B [ ] – · I
abc
[ ]
n
6993.12/ 1.93 –
6881.16/ 121.61 –
6880.24/117.50
V
V
average
6993.12 6881.16 6880.24 + +
3
 6918.17
Vdeviation
max
6993.12 6918.17 – 74.94
V
unbalance
74.94
6918.17
 · 100 1.0833%
0812_frame_C06.fm Page 140 Friday, July 20, 2001 3:11 PM
Distribution System Line Models 141
6.4 Summary
This chapter has developed the exact, modiﬁed, and approximate line seg
ment models. The exact model uses no approximations; that is, the phase
impedance matrix, assuming no transposition is used, as well as the shunt
admittance matrix. The modiﬁed model ignores the shunt admittance. The
approximate line model ignores the shunt admittance and assumes that the
positive and zero sequence impedances of the line are the known parameters.
This is paramount to assuming the line is transposed. For the three line models,
generalized matrix equations have been developed. The equations utilize
the generalized matrices [a], [b], [c], [d], [A], and [B]. The example problems
demonstrate that, because the shunt admittance is very small, the general
ized matrices can be computed neglecting the shunt admittance with very
little if any error. In most cases the shunt admittance can be neglected;
however, there are situations where the shunt admittances should not be
neglected. This is particularly true for long, rural, lightly loaded lines, and
for many underground lines.
References
1. Glover, J.D. and Sarma, M., Power System Analysis and Design, 2nd ed., PWS
Kent Publishing, Boston, 1995.
2. ANSI/NEMA Standard Publication No. MG11978, National Electrical Manufac
turers Association, Washington, D.C.
Problems
6.1 A twomilelong, threephase line uses the conﬁguration of Problem 4.1.
The phase impedance matrix and shunt admittance matrix for the conﬁgu
ration are
z
abc
[ ]
0.3375 j1.0478 + 0.1560 j0.5017 + 0.1535 j0.3849 +
0.1560 j0.5017 + 0.3465 j1.0179 + 0.1580 j0.4236 +
0.1535 j0.3849 + 0.1580 j0.4236 + 0.3414 j1.0348 +
Ω/mile
y
abc
[ ]
j5.9540 j2.0030 – j0.7471 –
j2.0030 – j6.3962 j1.2641 –
j0.7471 – j1.2641 – j5.6322
µS/mile
0812_frame_C06.fm Page 141 Friday, July 20, 2001 3:11 PM
142 Distribution System Modeling and Analysis
The line is serving a balanced threephase load of 10,000 kVA, with balanced
voltages of 13.2 kV linetoline and a power factor of 0.85 lagging.
(1) Determine the generalized matrices.
(2) For the given load, compute the linetoline and linetoneutral vol
tages at the source end of the line.
(3) Compute the voltage unbalance at the source end.
(4) Compute the source end complex power per phase.
(5) Compute the power loss by phase over the line. (Hint: Power loss
is deﬁned as power in minus power out.)
6.2 The positive and zero sequence impedances for the line of Problem 6.1
are
z
+
0.186 + j0.5968 Ω/mile z
0
0.6534 + j1.907 Ω/mile
Repeat Problem 6.1 using the approximate line model.
6.3 The line of Problem 6.1 serves an unbalanced, grounded wye, connected
constant impedance load of:
The line is connected to a balanced threephase 13.2kV source.
(1) Determine the load currents.
(2) Determine the load linetoground voltages.
(3) Determine the complex power of the load by phase.
(4) Determine the source complex power by phase.
(5) Determine the power loss by phase and the total threephase power
loss.
6.4 Repeat Problem 6.3, but change the impedance on Phase b to Ω
6.5 The twophase line of Problem 4.2 has the following phase impedance
matrix:
The line is two miles long and serves a twophase load such that:
Z
ag
15/30 Ω, Z
bg
17/36.87 Ω, Z
cg
20/25.84 Ω
50/36.87
z
abc
[ ]
0.4576 j1.0780 + 0 0.1535 j0.3849 +
0 0 0
0.1535 j0.3849 + 0 0.4615 j1.0651 +
Ω/mile
S
ag
2000 kVA at 0.9 lagging power factor and voltage of 7620/0 V
S
cg
1500 kVA at 0.95 lagging power factor and voltage of 7620/120 V
0812_frame_C06.fm Page 142 Friday, July 20, 2001 3:11 PM
Distribution System Line Models 143
Neglect the shunt admittance and determine the following:
(1) The source linetoground voltages using the generalized matrices.
(Hint: Even though Phase b is physically not present, assume that
it is with a value of and is serving a 0 kVA load.)
(2) The complex power by phase at the source.
(3) The power loss by phase on the line.
6.6 The singlephase line of Problem 4.3 has the following phase impedance
matrix:
Ω/mile
The line is one mile long and is serving a singlephase load of 2000 kVA,
0.95 lagging power factor at a voltage of . Determine the source
voltage and power loss on the line. (Hint: As in the previous problem, even
though Phases a and c are not physically present, assume they are, and, along
with Phase b, make up a balanced threephase set of voltages.)
6.7 The threephase concentric neutral cable conﬁguration of Problem 4.10
is two miles long and serves a balanced threephase load of 10,000 kVA,
13.2 kV, 0.85 lagging power factor. The phase impedance and shunt admit
tance matrices for the cable line are
(1) Determine the generalized matrices.
(2) For the given load, compute the linetoline and linetoneutral
voltages at the source end of the line.
(3) Compute the voltage unbalance at the source end.
(4) Compute the source end complex power per phase.
(5) Compute the power loss by phase over the line. (Hint: Power loss
is deﬁned as power in minus power out.)
7620/ 120 – V
z
abc
[ ]
0 0 0
0 1.3292 j1.3475 + 0
0 0 0
7500/ 120 – V
z
abc
[ ]
0.8040 j0.4381 + 0.3176 j0.0276 + 0.2824 j0.0184 –
0.3176 j0.0276 + 0.7939 j0.3966 + 0.3176 j0.0276 +
0.2824 j0.0184 – 0.2824 j0.0184 – 0.8040 j0.4381 +
Ω/mile
y
abc
[ ]
j117.52 0 0
0 j117.52 0
0 0 j117.52
µS/mile
0812_frame_C06.fm Page 143 Friday, July 20, 2001 3:11 PM
144 Distribution System Modeling and Analysis
6.8 The line of Problem 6.7 serves an unbalanced grounded wye connected
constant impedance load of:
The line is connected to a balanced threephase 13.2 kV source.
(1) Determine the load currents.
(2) Determine the load linetoground voltages.
(3) Determine the complex power of the load by phase.
(4) Determine the source complex power by phase.
(5) Determine the power loss by phase and the total threephase power
loss.
6.9 The tapeshielded cable singlephase line of Problem 4.12 is two miles
long and serves a singlephase load of 3000 kVA, at 8.0 kV and 0.9 lagging
power factor. The phase impedance and shunt admittances for the line are
Determine the source voltage and the power loss for the loading condition.
Z
ag
15/30 Ω, Z
bg
50/36.87 Ω, Z
cg
20/25.84 Ω
z
abc
[ ]
0 0 0
0 0 0
0 0 0.5287 j0.5717 +
Ω/mile
y
abc
[ ]
0 0 0
0 0 0
0 0 j140.39
µS/mile
0812_frame_C06.fm Page 144 Friday, July 20, 2001 3:11 PM
145
7
Regulation of Voltages
The regulation of voltages is an important function on a distribution feeder.
As the loads on the feeders vary, there must be some means of regulating
the voltage so that every customer’s voltage remains within an acceptable
level. Common methods of regulating the voltage are the application of step
type voltage regulators, load tap changing transformers (LTC), and shunt
capacitors.
7.1 Standard Voltage Ratings
The American National Standards Institute (ANSI) standard ANSI C84.1
1995 for “Electric Power Systems and Equipment Voltage Ratings (60 Hertz)”
provides the following deﬁnitions for system voltage terms:
1
•
System Voltage
: the root mean square (rms) phasor voltage of a
portion of an alternatingcurrent electric system. Each system volt
age pertains to a portion of the system that is bounded by trans
formers or utilization equipment.
•
Nominal System Voltage
: the voltage by which a portion of the sys
tem is designated, and to which certain operating characteristics
of the system are related. Each nominal system voltage pertains to
a portion of the system bounded by transformers or utilization
equipment.
•
Maximum System Voltage
: the highest system voltage that occurs
under normal operating conditions, and the highest system voltage
for which equipment and other components are designed for sat
isfactory continuous operation without derating of any kind.
•
Service Voltage
: the voltage at the point where the electrical system
of the supplier and the electrical system of the user are connected.
•
Utilization Voltage
: the voltage at the line terminals of utilization
equipment.
0812_frame_C07.fm Page 145 Saturday, July 21, 2001 3:11 PM
146
Distribution System Modeling and Analysis
•
Nominal Utilization Voltage
: the voltage rating of certain utilization
equipment used on the system.
The ANSI standard speciﬁes two voltage ranges. An oversimpliﬁcation of
the voltage ranges is
•
Range A
: Electric supply systems shall be so designated and oper
ated such that most service voltages will be within the limits spec
iﬁed for Range A. The occurrence of voltages outside of these limits
should be infrequent.
•
Range B
: Voltages above and below Range A. When these voltages
occur, corrective measures shall be undertaken within a reasonable
time to improve voltages to meet Range A.
For a normal threewire 120/240 volt service to a user, the Range A and
Range B voltages are
• Range A
– Nominal utilization voltage
=
115 V
– Maximum utilization and service voltage
=
126 V
– Minimum service voltage
=
114 V
– Minimum utilization voltage
=
110 V
• Range B
– Nominal utilization voltage
=
115 V
– Maximum utilization and service voltage
=
127 V
– Minimum service voltage
=
110 V
– Minimum utilization voltage
=
107 V
These ANSI standards give the distribution engineer a range of normal
steadystate voltages (Range A) and a range of emergency steadystate volt
ages (Range B) that must be supplied to all users.
In addition to the acceptable voltage magnitude ranges, the ANSI standard
recommends that the “electric supply systems should be designed and oper
ated to limit the maximum voltage unbalance to 3% when measured at the
electricutility revenue meter under a noload condition.” Voltage unbalance
is deﬁned as:
(7.1)
The task for the distribution engineer is to design and operate the distribu
tion system so that under normal steadystate conditions the voltages at the
Voltage
unbalance
Max. deviation from average voltage
Average voltage
 100% ⋅ =
0812_frame_C07.fm Page 146 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages
147
meters of all users will lie within Range A, and that the voltage unbalance
will not exceed 3%.
A common device used to maintain system voltages is the stepvoltage
regulator. Stepvoltage regulators can be singlephase or threephase. Single
phase regulators can be connected in wye, delta, or open delta, in addition
to operating as singlephase devices. The regulators and their controls allow
the voltage output to vary as the load varies.
A stepvoltage regulator is basically an autotransformer with a load tap
changing mechanism on the series winding. The voltage change is obtained
by changing the number of turns (tap changes) of the series winding of the
autotransformer.
An autotransformer can be visualized as a twowinding transformer with
a solid connection between a terminal of the primary side of the transformer
to a terminal on the secondary. Before proceeding to the autotransformer, a
review of twowinding transformer theory and the development of gener
alized constants will be presented.
7.2 TwoWinding Transformer Theory
The exact equivalent circuit for a twowinding transformer is shown in
Figure 7.1. In Figure 7.1 the highvoltage transformer terminals are denoted
by
H
1
and
H
2
, and the lowvoltage terminals are denoted by
X
1
and
X
2
. The
standards for these markings are such that at no load, the voltage between
H
1
and
H
2
will be in phase with the voltage between
X
1
and
X
2
. Under a
steadystate load condition, the currents
I
1
and
I
2
will be in phase.
Without introducing a signiﬁcant error, the exact equivalent circuit of
Figure 7.1 is modiﬁed by referring the primary impedance (
Z
1
) to the sec
ondary side as shown in Figure 7.2. Referring to Figure 7.2, the total leakage
FIGURE 7.1
Twowinding transformer exact equivalent circuit.
E
1
E
2
+ +
N
2
N
1
:
I
1
Z
2
I
2
V
L
+
  
+

V
S
I
S
Y
m
I
ex
Z
1
H
1
H
2
X
1
X
2
0812_frame_C07.fm Page 147 Saturday, July 21, 2001 3:11 PM
148
Distribution System Modeling and Analysis
impedance of the transformer is given by:
Z
t
=
(7.2)
where (7.3)
In order to better understand the model for the stepregulator, a model for
the twowinding transformer will ﬁrst be developed. Referring to Figure 7.2,
the equations for the ideal transformer become:
(7.4)
(7.5)
Applying KVL in the secondary circuit:
(7.6)
In general form, Equation 7.6 can be written as:
(7.7)
where (7.8)
(7.9)
FIGURE 7.2
Twowinding transformer approximate equivalent circuit.
E
1
E
2
+ +
N
2
N
1
:
I
1
Z
t
I
2
V
L
+
  
+

V
S
Y
m
ex
I
I
S
X
2
X
1
H
1
H
2
n
t
2
Z
1
Z
2
+ ⋅
n
t
N
2
N
1
 =
E
2
N
2
N
1
 E
1
⋅ n
t
E
1
⋅ = =
I
1
N
2
N
1
 I
2
⋅ n
t
I
2
⋅ = =
E
2
V
L
Z
t
I
2
⋅ + =
V
S
E
1
1
n
t
 E
2
⋅
1
n
t
 V
L
Z
t
n
t
 I
2
⋅ + ⋅ = = =
V
S
a V
L
b I
2
⋅ + ⋅ =
a
1
n
t
 =
b
Z
t
n
t
 =
0812_frame_C07.fm Page 148 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages
149
The input current to the twowinding transformer is given by:
I
S
=
Y
m
⋅
V
S
+
I
1
(7.10)
Substitute Equations 7.6 and 7.5 into Equation 7.10:
(7.11)
In general form, Equation 7.11 can be written as:
I
S
=
c
⋅
V
L
+
d
⋅
I
2
(7.12)
where (7.13)
(7.14)
Equations 7.7 and 7.12 are used to compute the input voltage and current
to a twowinding transformer when the load voltage and current are known.
These two equations are of the same form as Equations 6.8 and 6.16 that
were derived in Chapter 6 for the threephase line models. The only differ
ence at this point is that only a singlephase twowinding transformer is
being modeled. Later in this chapter the terms
a
,
b
,
c
, and
d
will be expanded
to 3
×
3 matrices for all possible threephase regulator connections.
Sometimes, particularly in an iterative process, the output voltage needs
to be computed knowing the input voltage and the load current. Solving
Equation 7.7 for the load voltage yields:
(7.15)
Substituting Equations 7.8 and 7.9 into Equation 7.15 results in:
V
L
=
A
⋅
V
S
−
B ⋅ I
2
(7.16)
where A = n
t
(7.17)
B = Z
t
(7.18)
I
S
Y
m
1
n
t
 V
L
Y
m
Z
t
n
t
 I
2
n
t
I
2
⋅ + ⋅ ⋅ + ⋅ ⋅ =
I
S
Y
m
n
t
 V
L
Y
m
Z
t
⋅
n
t
 n
t
+
I
2
⋅ + ⋅ =
c
Y
m
n
t
 =
d
Y
m
Z
t
⋅
n
t
 n
t
+ =
V
L
1
a
 V
S
b
a
 I
2
⋅ – ⋅ =
0812_frame_C07.fm Page 149 Saturday, July 21, 2001 3:11 PM
150 Distribution System Modeling and Analysis
Again, Equation 7.16 is of the same form as Equation 6.26. Later in this
chapter the expressions for A and B will be expanded to 3 × 3 matrices for
all possible threephase transformer connections.
Example 7.1
A singlephase transformer is rated 75 kVA, 2400240 volts. The transformer
has the following impedances and shunt admittance:
Z
1
= 0.612 + j1.2 Ω (highvoltage winding impedance)
Z
2
= 0.0061 + j0.0115 Ω (lowvoltage winding impedance)
Y
m
= 1.92 × 10
−4
− j8.52 × 10
−4
S (referred to the highvoltage winding)
Determine the generalized a, b, c, d constants and the A and B constants.
The transformer “turns ratio” is
The equivalent transformer impedance referred to the lowvoltage side:
Z
t
= Z
2
+ ⋅ Z
1
= 0.0122 + j0.0235
The generalized constants are
A = n
t
= 0.1
B = Z
t
= 0.0122 + j0.0235
n
t
N
2
N
1

V
rated 2
V
rated 1

240
2400
 0.1 = = = =
n
t
2
a
1
n
t

1
0.1
 10 = = =
b
Z
t
0.1
 0.1222 j0.235 + = =
c
Y
m
n
t
 0.0019 j0.0085 – = =
d
Y
m
Z
t
⋅
n
t
 n
t
+ 0.1002 j0.0001 – = =
0812_frame_C07.fm Page 150 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 151
Assume that the transformer is operated at rated load (75 kVA) and rated
voltage (240 V) with a power factor of 0.9 lagging. Determine the source
voltage and current using the generalized constants.
Applying the values of the a, b, c, and d parameters computed above:
V
S
= a ⋅ V
L
+ b ⋅ I
2
=
I
S
= c ⋅ V
L
+ d ⋅ I
2
=
Using the computed source voltage and the load current, determine the load
voltage.
For future reference, the perunit impedance of the transformer is computed
by:
The perunit shunt admittance is computed by:
Example 7.1 demonstrates that the generalized constants provide a quick
method for analyzing the operating characteristics of a twowinding
transformer.
V
L
240/0 =
I
2
75 1000 ⋅
240
/ cos
1 –
0.9 ( ) – 312.5/ 25.84 – = =
2466.9/1.15 V
32.67/ 28.75 – A
V
L
A V
S
B I
S
⋅ – ⋅ 0.1 ( ) 2466.9/1.15 ( ) ⋅ = =
0.0122 j0.0235 + ( ) 312.5/−25.84 ( ) ⋅ –
V
L
240.0/0 V =
Z
base
2
V
rated 2
2
kVA 1000 ⋅

240
2
75,000
 0.768 Ω = = =
Z
pu
Z
t
Z
base
2

0.0122
j
0.0115 +
0.768
 0.0345/62.5 = = = perunit
Y
base
1
kVA
kV
1
2
1000 ⋅
 0.013 S = =
Y
pu
Y
m
Y
base

1.92 10
4 –
⋅
j
8.52 ·
4 –
–
0.013
 0.0148 j0.0655 perunit – = = =
0812_frame_C07.fm Page 151 Saturday, July 21, 2001 3:11 PM
152 Distribution System Modeling and Analysis
7.3 The TwoWinding Autotransformer
A twowinding transformer can be connected as an autotransformer. Con
necting the highvoltage terminal H
1
to the lowvoltage terminal X
2
as shown
in Figure 7.3 can create a “stepup” autotransformer. The source is connected
to terminals H
1
and H
2
, while the load is connected between the X
1
terminal
and the extension of H
2
. In Figure 7.3, V
S
is the source voltage and V
L
is the
load voltage. The lowvoltage winding of the twowinding transformer will
be referred to as the “series” winding of the autotransformer, and the high
voltage winding will be referred to as the “shunt” winding.
Generalized constants similar to those of the twowinding transformer can
be developed for the autotransformer. The total equivalent transformer imp
edance is referred to the “series” winding. The ideal transformer equations of
7.4 and 7.5 still apply.
Applying KVL in the secondary circuit:
E
1
+ E
2
= V
L
+ Z
t
⋅ I
2
(7.19)
Using the ideal transformer relationship of Equation 7.5:
E
1
+ n
t
⋅ E
1
= (1 + n
t
) ⋅ E
1
= V
L
+ Z
t
⋅ I
2
(7.20)
Since the source voltage V
S
is equal to E
1
, and I
2
is equal to I
L
, Equation 7.20
can be modiﬁed to:
(7.21)
FIGURE 7.3
Stepup autotransformer.
V
S
1
1 n
t
+
 V
L
Z
t
1 n
t
+
 I
L
⋅ + ⋅ =
E
1
+
N
1
I
1

E
2
+
N
2
Z
t
I
2

V
S
+

I
S
V
L
+

Y
m
I
ex
H
2
X
2
H
1
X
1
0812_frame_C07.fm Page 152 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 153
(7.22)
where (7.23)
(7.24)
Applying KCL at input node H
1
:
I
S
= I
1
+ I
2
+
(7.25)
I
S
= (1 + n
t
) ⋅ I
2
+ Y
m
⋅ V
S
Substitute Equation 7.21 into Equation 7.25:
(7.26)
I
S
= c ⋅ V
L
+ d ⋅ I
2
where (7.27)
(7.28)
Equations 7.23, 7.24, 7.27, and 7.28 deﬁne the generalized constants relating
the source voltage and current as functions of the output voltage and current
for the stepup autotransformer.
The twowinding transformer can also be connected in the stepdown
connection by reversing the connection between the shunt and series wind
ing as shown in Figure 7.4. Generalized constants can be developed for the
stepdown connection following the same procedure as that for the stepup
connection.
Applying KVL in the secondary circuit:
E
1
− E
2
= V
L
+ Z
t
⋅ I
2
(7.29)
V
S
a V
L
b I
L
⋅ + ⋅ =
a
1
1 n
t
+
 =
b
Z
t
1 n
t
+
. =
I
ex
I
S
1 n
t
+ ( ) I
2
⋅ Y
m
1
1 n
t
+
 V
L
Z
t
1 n
t
+
 I
2
⋅ + ⋅
+ =
I
S
Y
m
1 n +
t
 V
L
Y
m
Z
t
⋅
1 n
t
+
 n
t
1 + +
I
2
⋅ + ⋅ =
c
Y
m
1 n +
t
 =
d
Y
m
Z
t
⋅
1 n
t
+
 n
t
1 + + =
0812_frame_C07.fm Page 153 Saturday, July 21, 2001 3:11 PM
154
Distribution System Modeling and Analysis
Using the ideal transformer relationship of Equation 7.5:
E
1
−
n
t
⋅
E
1
=
(1
−
n
t
)
⋅
E
1
=
V
L
+
Z
t
⋅
I
2
(7.30)
Since the source voltage
V
S
is equal to
E
1
, and
I
2
is equal to
I
L
, Equation 7.30
can be modiﬁed to:
(7.31)
=
a
⋅
V
L
+
b
⋅
I
L
(7.32)
where (7.33)
(7.34)
It is observed at this point that the only difference between the
a
and
b
constants of Equations 7.23 and 7.24 for the stepup connection, and Equa
tions 7.33 and 7.44 for the stepdown connection, is the sign in front of the
turns ratio (
n
t
). This will also be the case for the
c
and
d
constants. Therefore,
for the stepdown connection, the
c
and
d
constants are deﬁned by:
FIGURE 7.4
Stepdown autotransformer.
E
1
+
N
1
I
1

E
2
+
I
2

V
S
+

I
S
V
L
+

Y
m
I
ex
H
H
1
X
2
X
1
V
S
1
1 n
t
–
 V
L
Z
t
1 n
t
–
 I
L
⋅ + ⋅ =
a
1
1 n
t
–
 =
b
Z
t
1 n
t
–
 =
0812_frame_C07.fm Page 154 Monday, October 28, 2002 9:52 AM
Regulation of Voltages
155
(7.35)
(7.36)
The only difference between the deﬁnitions of the generalized constants is
the sign of the turns ratio
n
t
. In general, then, the generalized constants can
be deﬁned by:
(7.37)
(7.38)
(7.39)
(7.40)
In Equations 7.37 through 7.40, the sign in the equations will be positive for
the stepup connection, and negative for the stepdown connection.
As with the twowinding transformer, it is sometimes necessary to relate
the output voltage as a function of the source voltage and the output current.
Solving Equation 7.32 for the output voltage:
(7.41)
(7.42)
where (7.43)
(7.44)
The generalized equations for the stepup and stepdown autotransformers
have been developed. They are of exactly the same form as was derived for
c
Y
m
1 n
t
–
 =
d
Y
m
Z
t
⋅
1 n
t
–
 1 n –
t
+ =
a
1
1 n
t
±
 =
b
Z
t
1 n
t
±
 =
c
Y
m
1 n
t
±
 =
d
Y
m
Z
t
⋅
1 n
t
±
1 n
t
± =
V
L
1
a
 V
S
b
a
 I
2
⋅ – ⋅ =
V
L
A V
S
B I
2
⋅ – ⋅ =
A
1
a
 1 n
t
± = =
B
b
a
 Z
t
= =
0812_frame_C07.fm Page 155 Monday, October 28, 2002 9:53 AM
156 Distribution System Modeling and Analysis
The generalized equations for the stepup and stepdown autotransformers
have been developed. They are of exactly the same form as was derived for
the twowinding transformer and for the line segment in Chapter 6. For the
singlephase autotransformer the generalized constants are single values,
but will be expanded later to 3 × 3 matrices for threephase autotransformers.
7.3.1 Autotransformer Ratings
The kVA rating of the autotransformer is the product of the rated input
voltage V
S
times the rated input current I
S
, or the rated load voltage V
L
times
the rated load current I
L
. Deﬁne the rated kVA and rated voltages of the two
winding transformer and autotransformer as:
= kVA rating of the twowinding transformer
kVA
auto
= kVA rating of the autotransformer
V
rated 1
= E
1
= rated source voltage of the twowinding transformer
V
rated 2
= E
2
= rated load voltage of the twowinding transformer
V
auto S
= rated source voltage of the autotransformer
V
auto L
= rated load voltage of the autotransformer
For the following derivation, neglect the voltage drop through the series wind
ing impedance:
V
auto L
= E
1
± E
2
= (1 ± n
t
) ⋅ E
1
(7.45)
The rated output kVA is then:
(7.46)
but
therefore
(7.47)
but
therefore
(7.48)
kVA
xfm
kVA
auto
V
auto L
I
2
⋅ 1 n
t
± ( ) E
1
I
2
⋅ ⋅ = =
I
2
I
1
n
t
 =
kVA
auto
1 n
t
± ( )
n
t
 E
1
I
1
⋅ ⋅ =
E
1
I
1
⋅ kVA
xfm
=
kVA
auto
1 n
t
± ( )
n
t
 kVA
xfm
⋅ =
0812_frame_C07.fm Page 156 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages
157
Equation 7.48 gives the kVA rating of a twowinding transformer when con
nected as an autotransformer. For the stepup connection, the sign of
n
t
will
be positive while the stepdown will use the negative sign. In general, the
turns ratio
n
t
will be a relatively small value, so the kVA rating of the autotrans
former will be considerably greater than the kVA rating of the twowinding
transformer.
Example 7.2
The twowinding transformer of Example 7.1 is connected as a stepup
autotransformer. Determine the kVA and voltage ratings of the autotransformer.
From Example 7.1 the turns ratio was determined to be
n
t
=
0.1. The rated
kVA of the autotransformer using Equation 7.48 is given by:
The voltage ratings are
Therefore, the autotransformer would be rated as 825 kVA, 24002640 V.
Suppose now that the autotransformer is supplying rated kVA at rated
voltage with a power factor of 0.9 lagging. Determine the source voltage and
current:
Determine the generalized constants:
kVA
auto
1 0.1 +
0.1
 75 ⋅ 825 kVA = =
V
auto S
V
rated 1
2400 V = =
V
auto L
V
rated 1
V
rated 2
+ 2400 240 + 2640 V = = =
V
L
V
auto L
2640/0 = = V
I
2
kVA
auto
1000 ⋅
V
auto L

825,000
2640
/ cos
1 –
0.9 ( ) – 312.5/ 25.84 – = = = A
a
1
1 0.1 +
 0.9091 = =
b
0.0122 j0.0235 +
1 0.1 +
 0.0111 j0.0214 + = =
c
1.92 j8.52 – ( ) 10
4 –
⋅
1 0.1 +
 1.7364 j7.7455 – ( ) 10
4 –
⋅ = =
d
1.92 j8.52 – ( ) 10
4 –
0.0122 j0.0235 + ( ) ⋅ ⋅
1 0.1 +
 1 0.1 + + 1.1002 j0.000005 – = =
0812_frame_C07.fm Page 157 Monday, October 28, 2002 9:53 AM
158 Distribution System Modeling and Analysis
Applying the generalized constants:
When the load side voltage is determined knowing the source voltage and
load current, the A and B parameters are needed:
The load voltage is then:
7.3.2 PerUnit Impedance
The perunit impedance of the autotransformer based upon the autotrans
former kVA and kV ratings can be developed as a function of the perunit
impedance of the twowinding transformer based upon the twowinding
transformer ratings.
Let = Perunit impedance of the twowinding transformer based
upon the twowinding kVA and kV ratings,
= Rated load voltage of the twowinding transformer.
The base impedance of the twowinding transformer referred to the low
voltage winding (series winding of the autotransformer) is
(7.49)
The actual impedance of the transformer referred to the lowvoltage (series)
winding is
(7.50)
Assume that the rated source voltage of the autotransformer is the nominal
voltage of the system; that is
(7.51)
V
S
a 2640/0 b 312.5/ 25.84 – ⋅ + ⋅ 2406.0/0.1 = = V
I
S
c 2640/0 d 312.5/ 25.84 – ⋅ + ⋅ 345.06/ 26.11 – = = A
A 1 n
t
+ 1.1 = =
B Z
t
0.0111 j0.0235 + = =
V
L
A 2406.04/0.107 B 312.5/ 25.84 – ⋅ – ⋅ 2640.00/0 = = V
Zpu
xfm
V
rated 2
Zbase
xfm
V
rated 2
2
kVA
xfm
1000 ⋅
 =
Zt
actual
Zt
pu
Zbase
xfm
⋅ Zt
pu
V
auto series
2
kVA
xfm
1000 ⋅
 ⋅ = =
V
nominal
V
rated 1
V
rated 2
n
t
 = =
0812_frame_C07.fm Page 158 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 159
The base impedance for the autotransformer referenced to the nominal sys
tem voltage is
(7.52)
Substitute Equations 7.48 and 7.51 into Equation 7.52:
(7.53)
The perunit impedance of the autotransformer based upon the rating of the
autotransformer is
(7.54)
Substitute Equations 7.50 and 7.53 into Equation 7.54:
(7.55)
Equation 7.55 gives the relationship between the perunit impedance of the
autotransformer and the perunit impedance of the twowinding transformer.
The point is that the perunit impedance of the autotransformer is very small
compared to that of the twowinding transformer. When the autotransformer
is connected to boost the voltage 10%, the value of n
t
is 0.1, and Equation 7.57
becomes:
(7.56)
The perunit shunt admittance of the autotransformer can be developed as
a function of the perunit shunt admittance of the twowinding transformer.
Recall that the shunt admittance is represented on the source side of the two
winding transformer.
Zbase
auto
V
nominal
2
kVA
auto
1000 ⋅
 =
Zbase
auto
V
nominal
2
kVA
auto
1000 ⋅

Vrated 2
n
t

2
1 n
t
±
n
t
 kVA
xfm
1000 ⋅ ⋅
 = =
Zbase
auto
V
rating 2
2
n
t
1 n
t
± ( ) kVA
xfm
1000 ⋅ ⋅ ⋅
 =
Zauto
pu
Zt
actual
Zbase
auto
 =
Zauto
pu
Zt
pu
V
rating 2
2
kVA
xfm
1000 ⋅

V
rating 2
2
n
t
1 n
t
± ( ) kVA
xfm
1000 ⋅ ⋅ ⋅

 ⋅ n
t
1 n
t
± ( ) Zt
pu
⋅ ⋅ = =
Zauto
pu
0.1 1 0.1 + ( ) Zt
pu
⋅ ⋅ 0.11 Zt
pu
⋅ = =
0812_frame_C07.fm Page 159 Saturday, July 21, 2001 3:11 PM
160 Distribution System Modeling and Analysis
Let
Yt
pu
= Ym
pu
= perunit admittance of the twowinding transformer
based upon the transformer ratings,
Yauto
pu
= perunit admittance of the autotransformer based upon
the autotransformer ratings.
The base admittance of the twowinding transformer referenced to the source
side is given by:
(7.57)
The actual shunt admittance referred to the source side of the twowinding
transformer is
(7.58)
The perunit shunt admittance for the autotransformer is given by:
(7.59)
Substitute Equation 7.58 into 7.59:
(7.60)
Equation 7.60 shows that the perunit admittance based upon the autotrans
former ratings is much smaller than the perunit admittance of the twowinding
transformer. For an autotransformer in the raise connection with n
t
= 0.1,
Equation 7.60 becomes:
It has been shown that the perunit impedance and admittance values based
upon the autotransformer kVA rating and nominal voltage are approximately
onetenth that of the values for the twowinding transformer.
Ybase
source
kVA
xfm
1000 ⋅
V
rating 1
2
 =
Yt
source
Yt
pu
Ybase
source
⋅ Yt
pu
kVA
xfm
1000 ⋅
V
rating 1
2
 ⋅ = =
Yauto
pu
Yt
source
Ybase
auto
 Yt
source
V
rated 1
2
kVA
auto
1000 ⋅
 ⋅ = =
Yauto
pu
Yt
pu
kVA
xfm
1000 ⋅
V
rated 1
2

V
rated 1
2
kVA
auto
1000 ⋅
 ⋅ ⋅ =
Yauto
pu
Yt
pu
kVA
xfm
kVA
auto
 ⋅ Yt
pu
kVA
xfm
1 n
t
± ( )
n
t
 kVA
xfm
⋅
 ⋅
n
t
1 n
t
± ( )
 Yt
pu
⋅ = = =
Ya
pu
0.1
1 0.1 +
 Yt
pu
⋅ 0.0909 Yt
pu
⋅ = =
0812_frame_C07.fm Page 160 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 161
Example 7.3
The shunt admittance referred to the source side of the twowinding trans
former of Example 7.2 is
Y
t
= Y
m
= 1.92 ⋅ 10
−4
− j8.52 ⋅ 10
−4
S
1. Determine the perunit shunt admittance based upon the two
winding transformer ratings:
2. In Example 7.2 the kVA rating of the twowinding transformer
connected as an autotransformer was computed to be 825 kVA,
and the voltage ratings 2400−2640 V. Determine the perunit admit
tance based upon the autotransformer kVA rating and a nominal
voltage of 2400 V, and determine the ratio of the perunit admittance
of the autotransformer to the perunit admittance of the twowinding
transformer:
In this section the equivalent circuit of an autotransformer has been devel
oped for the “raise” and “lower” connections. These equivalent circuits included
the series impedance and shunt admittance. If a detailed analysis of the
autotransformer is desired, the series impedance and shunt admittance
should be included. However, it has been shown that these values are very
small, and when the autotransformer is to be a component of a system, very
little error will be made by neglecting both the series impedance and shunt
admittance of the equivalent circuit.
Ybase
source
75 1000 ⋅
2400
2
 0.013 = =
Yt
pu
1.92 10
4 –
j
8.52 10
4 –
⋅ – ⋅
0.013
 0.014746 j0.065434 – = =
Ybase
auto
825 1000 ⋅
2400
2
 0.1432 = =
Yauto
pu
1.92 10
4 –
j
8.52 10
4 –
⋅ – ⋅
0.1432
 0.001341 j0.005949 – = =
Ratio
0.001341
j
0.005949 –
0.014746 j0.065434 –
 0.0909 = =
0812_frame_C07.fm Page 161 Saturday, July 21, 2001 3:11 PM
162 Distribution System Modeling and Analysis
7.4 StepVoltage Regulators
A stepvoltage regulator consists of an autotransformer and a load tap
changing mechanism. The voltage change is obtained by changing the taps
of the series winding of the autotransformer. The position of the tap is
determined by a control circuit (line drop compensator). Standard step
regulators contain a reversing switch enabling a ±10% regulator range,
usually in 32 steps. This amounts to a 5/8% change per step, or 0.75V
change per step, on a 120V base. Step regulators can be connected in a
Type A or Type B connection according to the ANSI/IEEE C57.151986 stan
dard.
2
The more common Type B connection is shown in Figure 7.5. The
block diagram circuit shown in Figure 7.6 controls tap changing on a step
voltage regulator. The stepvoltage regulator control circuit requires the
following settings:
1. Voltage Level: the desired voltage (on a 120V base) to be held at
the load center. The load center may be the output terminal of the
regulator or a remote node on the feeder.
2. Bandwidth: the allowed variance of the load center voltage from the
set voltage level. The voltage held at the load center will be ± one
half the bandwidth. For example, if the voltage level is set to 122 V
FIGURE 7.5
Type B stepvoltage regulator.
Reversing
Switch
Control
CT
Control
VT
V
Load
Series
Winding
Shunt
Winding
Source
Preventive
Autotransformer
0812_frame_C07.fm Page 162 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 163
and the bandwidth is set to 2 V, the regulator will change taps until
the load center voltage lies between 121 and 123 V.
3. Time Delay: length of time that a raise or lower operation is called
for before the actual execution of the command. This prevents taps
changing during a transient or short time change in current.
4. Line Drop Compensator: set to compensate for the voltage drop (line
drop) between the regulator and the load center. The settings con
sist of R and X settings in volts corresponding to the equivalent
impedance between the regulator and the load center. This setting
may be zero if the regulator output terminals are the load center.
The required rating of a stepvoltage regulator is based upon the kVA
transformed, not the kVA rating of the line. In general, this will be 10%
of the line rating since rated current ﬂows through the series winding
which represents the ±10% voltage change. The kVA rating of the step
voltage regulator is determined in the same manner as that of the previ
ously discussed autotransformer.
7.4.1 SinglePhase StepVoltage Regulators
Because the series impedance and shunt admittance values of stepvoltage
regulators are so small, they will be neglected in the following equivalent
circuits. It should be pointed out, however, that if it is desired to include the
impedance and admittance, they can be incorporated into the following equiv
alent circuits in the same way they were originally modeled in the autotrans
former equivalent circuit.
7.4.1.1 Type A StepVoltage Regulator
The detailed equivalent circuit and abbreviated equivalent circuit of a
Type A stepvoltage regulator in the raise position is shown in Figure 7.7.
FIGURE 7.6
Stepvoltage regulator control circuit.
0812_frame_C07.fm Page 163 Saturday, July 21, 2001 3:11 PM
164 Distribution System Modeling and Analysis
As shown in Figure 7.7, the primary circuit of the system is connected directly
to the shunt winding of the Type A regulator. The series winding is connected
to the shunt winding and, in turn, via taps, to the regulated circuit. In this
connection the core excitation varies because the shunt winding is connected
directly across the primary circuit.
When the Type A connection is in the lower position, the reversing switch
is connected to the L terminal. The effect of this reversal is to reverse the
direction of the currents in the series and shunt windings. Figure 7.8 shows
the equivalent circuit and abbreviated circuit of the Type A regulator in the
lower position.
7.4.1.2 Type B StepVoltage Regulator
The more common connection for stepvoltage regulators is the Type B. Since
this is the more common connection, the deﬁning voltage and current equations
for the voltage regulator will be developed only for the Type B connection.
The detailed and abbreviated equivalent circuits of a Type B stepvoltage
regulator in the raise position is shown in Figure 7.9. The primary circuit of
the system is connected, via taps, to the series winding of the regulator in
the Type B connection. The series winding is connected to the shunt winding,
which is connected directly to the regulated circuit. In a Type B regulator
the core excitation is constant because the shunt winding is connected across
the regulated circuit.
The deﬁning voltage and current equations for the regulator in the raise
position are as follows:
FIGURE 7.7
Type A stepvoltage regulator in the raise position.
SL
SL
E
2
N
2

V
L
+
+
+


N
1
E
1
L
S +

V
S
I
S
L
I
L
R
I
2
I
1
S
I
S
S
V
+

L
L
I
+

V
L
0812_frame_C07.fm Page 164 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 165
FIGURE 7.8
Type A stepvoltage regulator in the lower position.
FIGURE 7.9
Type B stepvoltage regulator in the raise position.
SL
SL
E
2
N
2

V
L
+
+
+


N
1
E
1
L
S +

V
S
I
S
L
I
L
R
S
I
S
S
V
+

L
L
I
+

V
L
I
2
I
1
+

V
S
SL
+

V
R
S
L
+

I
1
E
1
+

E
I
2
2
I
S
I
S
I
L
S
+
S
V

SL
L
V
L
L
I
I
S
+

L
N
2
N
1
L
0812_frame_C07.fm Page 165 Saturday, July 21, 2001 3:11 PM
166 Distribution System Modeling and Analysis
VOLTAGE EQUATIONS CURRENT EQUATIONS
(7.61)
V
S
= E
1
− E
2
I
L
= I
S
− I
1
(7.62)
V
L
= E
1
I
2
= I
S
(7.63)
(7.64)
(7.65)
V
S
= a
R
⋅ V
L
I
L
= a
R
⋅ I
S
(7.66)
(7.67)
Equations 7.66 and 7.67 are the necessary deﬁning equations for modeling
a regulator in the raise position.
The Type B stepvoltage connection in the lower position is shown in Figure
7.10. As in the Type A, connection note that the direction of the currents
through the series and shunt windings change, but the voltage polarity of the
two windings remain the same.
The deﬁning voltage and current equations for the Type B stepvoltage
regulator in the lower position are as follows:
VOLTAGE EQUATIONS CURRENT EQUATIONS
(7.68)
V
S
= E
1
+ E
2
I
L
= I
S
+ I
1
(7.69)
V
L
= E
1
I
2
= I
S
(7.70)
(7.71)
(7.72)
V
S
= a
R
⋅ V
L
I
L
= a
R
⋅ I
S
(7.73)
(7.74)
E
1
N
1

E
2
N
2
 = N
1
I
1
⋅ N
2
I
2
⋅ =
E
2
N
2
N
1
 E
1
⋅
N
2
N
1
 V
L
⋅ = = I
1
N
2
N
1
 I
2
⋅
N
2
N
1
 I
S
⋅ = =
V
S
1
N
2
N
1
 –
V
L
⋅ = I
L
1
N
2
N
1
 –
I
S
⋅ =
a
R
1
N
2
N
1
 – =
E
1
N
1

E
2
N
2
 = N
1
I
1
⋅ N
2
I
2
⋅ =
E
2
N
2
N
1
 E
1
⋅
N
2
N
1
 V
L
⋅ = = I
1
N
2
N
1
 I
2
⋅
N
2
N
1
 I
S
⋅ = =
V
S
1
N
2
N
1
 +
V
L
⋅ = I
L
1
N
2
N
1
 +
I
S
⋅ =
a
R
1
N
2
N
1
 + =
0812_frame_C07.fm Page 166 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 167
Equations 7.67 and 7.74 give the value of the effective regulator ratio as a
function of the ratio of the number of turns on the series winding (N
2
) to the
number of turns on the shunt winding (N
1
).
In the ﬁnal analysis, the only difference between the voltage and current
equations for the Type B regulator in the raise and lower positions is the
sign of the turns ratio (N
2
/N
1
). The actual turns ratio of the windings is not
known; however, the particular tap position will be known. Equations 7.67
and 7.74 can be modiﬁed to give the effective regulator ratio as a function
of the tap position. Each tap changes the voltage by 5/8% or 0.00625 perunit.
Therefore, the effective regulator ratio can be given by:
(7.75)
In Equation 7.75, the minus sign applies for the raise position and the plus
sign for the lower position.
7.4.1.3 Generalized Constants
In previous chapters and sections of this text generalized abcd constants
have been developed for various devices. It can now be shown that the
generalized abcd constants can also be applied to the stepvoltage regu
lator. For both the Type A and Type B regulators, the relationship be
tween the source voltage and current to the load voltage and current are
FIGURE 7.10
Type B stepvoltage regulator in the lower position.
I
I
I
I
I
I
I
a
R
1 0.00625 Tap ⋅
+
−
=
0812_frame_C07.fm Page 167 Saturday, July 21, 2001 3:11 PM
168
Distribution System Modeling and Analysis
of the form:
Type A: (7.76)
Type B: (7.77)
Therefore, the generalized constants for a singlephase stepvoltage regulator
become:
Type A: (7.78a)
Type B: (7.78b
where
a
R
is given by Equation 7.75 and the sign convention is given in
Table 7.1.
7.4.1.4 The Line Drop Compensator
The changing of taps on a regulator is controlled by the line drop compen
sator. Figure 7.11 shows a simpliﬁed sketch of the compensator circuit and
how it is connected to the distribution line through a potential transformer
and a current transformer. The purpose of the line drop compensator is to
model the voltage drop of the distribution line from the regulator to the load
center. The compensator is an analog circuit that is a scale model of the line
circuit. The compensator input voltage is typically 120 volts, which requires
the voltage transformer in Figure 7.11 to reduce the rated voltage to 120 volts.
For a regulator connected linetoground the rated voltage is the nominal line
toneutral voltage, while for a regulator connected linetoline the rated voltage
is the linetoline voltage. The current transformer turns ratio is speciﬁed as
CT
p
:
CT
s
, where the primary rating (
CT
p
) will typically be the rated current of
the feeder. The setting that is most critical is that of
R
′
and
X
′
calibrated in
volts. These values must represent the equivalent impedance from the regu
lator to the load center. The basic requirement is to force the perunit line
impedance to be equal to the perunit compensator impedance. In order to
cause this to happen, it is essential that a consistent set of base values be
developed wherein the perunit voltage and currents in the line and in the
compensator are equal. The consistent set of base values is determined by
TABLE 7.1
Sign Convention Table for Equation 7.75
Type A Type B
Raise
+ −
Lower
− +
V
S
1
a
R
 V
L
⋅ = I
S
a
R
I
L
⋅ =
V
S
a
R
V
L
⋅ = I
S
1
a
R
 I
L
⋅ =
a
1
a
R
 = b 0 = c 0 = d a
R
=
a a
R
= b 0 = c 0 = d
1
a
R
 =
0812_frame_C07.fm Page 168 Monday, October 28, 2002 10:40 AM
Regulation of Voltages 169
selecting a base voltage and current for the line circuit, and then computing
the base voltage and current in the compensator by dividing the system
base values by the voltage transformer ratio and current transformer ratio,
respectively. For regulators connected linetoground, the base system voltage
is selected as the rated linetoneutral voltage (V
LN
) and the base system current
is selected as the rating of the primary winding of the current transformer
(CT
P
). Table 7.2 gives a table of base values and employs these rules for a
regulator connected linetoground. With the table of base values developed,
the compensator R and X settings in ohms can be computed by ﬁrst com
puting the perunit line impedance:
(7.79)
FIGURE 7.11
Line drop compensator circuit.
TABLE 7.2
Table of Base Values
Base Line Circuit Compensator Circuit
Voltage V
LN
Current CT
P
CT
S
Impedance Zbase
line
= Zbase
comp
=
Voltage
Relay
R line + jX line
Load
Center
MVA rating
kV hi  kV lo
R' X'
+  V drop
I line
I comp
Npt:1
CTp:CTs
+

V reg
+

V R
1:1
V
LN
N
PT

V
LN
CT
P

V
LN
N
PT
CT
S
⋅

R
pu
jX
pu
+
Rline
Ω
jXline
Ω
+
Zbase
line
 =
R
pu
jX
pu
+ Rline
Ω
jXline
Ω
+ ( )
CT
P
V
LN
 ⋅ =
0812_frame_C07.fm Page 169 Saturday, July 21, 2001 3:11 PM
170 Distribution System Modeling and Analysis
The perunit impedance of Equation 7.79 must be the same in the line
and in the compensator. The compensator impedance in ohms is com
puted by multiplying the perunit impedance by the base compensator
impedance:
Rcomp
Ω
+ jXcomp
Ω
=
Rcomp
Ω
+ jXcomp
Ω
= (Rline
Ω
+ jXline
Ω
) ⋅ (7.80)
Rcomp
Ω
+ jXcomp
Ω
= (Rline
Ω
+ jXline
Ω
) ⋅
Equation 7.80 gives the value of the compensator R and X settings in ohms.
The compensator R and X settings in volts are determined by multiplying
the compensator R and X in ohms times the rated secondary current in amps
(CT
S
) of the current transformer:
R′ + jX′ = (Rcomp
Ω
+ jXcomp
Ω
) ⋅ CT
S
R′ + jX′ = (Rline
Ω
+ jXline
Ω
) ⋅ (7.81)
R′ + jX′ = (Rline
Ω
+ jXline
Ω
) ⋅
Knowing the equivalent impedance in ohms from the regulator to the load
center, the required value for the compensator settings in volts is determined
by using Equation 7.81. This is demonstrated in Example 7.4.
Example 7.4
Refer to Figure 7.11.
The substation transformer is rated 5000 kVA, 115 delta − 4.16 grounded
wye, and the equivalent line impedance from the regulator to the load center
is 0.3 + j0.9 Ω.
(1) Determine the voltage transformer and current transformer ratings
for the compensator circuit.
The rated linetoground voltage of the substation transformer is
R
pu
jX
pu
+ ( ) Zbase
comp
⋅
CT
P
V
LN

V
LN
N
PT
CT
S
⋅
 ⋅
CT
P
N
PT
CT
S
⋅
 Ω
CT
P
N
PT
CT
S
⋅
 CT
S
⋅
CT
P
N
PT
 V
V
S
4160
3
 2401.8 = =
0812_frame_C07.fm Page 170 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 171
In order to provide approximately 120 V to the compensator, the
voltage transformer ratio is
The rated current of the substation transformer is
The primary rating of the CT is selected as 700 A, and if the com
pensator current is reduced to 5 A, the CT ratio is
(2) Determine the R and X settings of the compensator in ohms and
volts.
Applying Equation 7.78 to determine the settings in volts:
The R and X settings in ohms are determined by dividing the
settings in volts by the rated secondary current of the current
transformer:
Understand that the R and X settings on the compensator control
board are calibrated in volts.
Example 7.5
The substation transformer in Example 7.4 is supplying 2500 kVA at 4.16 kV
and 0.9 power factor lag. The regulator has been set so that:
R′ + jX′ = 10.5 + j31.5 V
Voltage Level = 120 V (desired voltage to be held at the load center)
Bandwidth = 2 V
N
PT
2400
120
 20 = =
I
rated
5000
3 4.16 ⋅
 693.9 = =
CT
CT
P
CT
S

700
5
 140 = = =
R′ jX′ + 0.3 j0.9 + ( )
700
20
 ⋅ 10.5 j31.5 V + = =
R
ohms
jX
ohms
+
10.5 j31.5 +
5
 2.1 j6.3 Ω + = =
0812_frame_C07.fm Page 171 Saturday, July 21, 2001 3:11 PM
172 Distribution System Modeling and Analysis
Determine the tap position of the regulator that will hold the load center
voltage at the desired voltage level and within the bandwidth. This means
that the tap on the regulator needs to be set so that the voltage at the load
center lies between 119 and 121 V.
The ﬁrst step is to calculate the actual line current:
The current in the compensator is then:
The input voltage to the compensator is
The voltage drop in the compensator circuit is equal to the compen
sator current times the compensator R and X values in ohms:
The voltage across the voltage relay is
The voltage across the voltage relay represents the voltage at the load center.
Since this is well below the minimum voltage level of 119, the voltage
regulator will have to change taps in the raise position to bring the load
center voltage up to the required level. Recall that on a 120V base, one step
change on the regulator changes the voltage 0.75 V. The number of required
tap changes can then be approximated by
This shows that the ﬁnal tap position of the regulator will be “raise 13.”
With the tap set at +13, the effective regulator ratio assuming a Type B
regulator is
a
R
= 1 − 0.00625 ⋅ 13 = 0.9188
I
line
2500
3 4.16 ⋅
/cos
1 –
0.9 ( ) 346.97/ 25.84 – A = =
I
comp
346.97/ 25.84 –
140
 2.4783/ 25.84 – = = A
V
reg
2401.8/0
20
 120.09/0 V = =
V
drop
2.1 j6.3 + ( ) 2.4783/–25.84 ⋅ 16.458/45.72 V = =
V
R
V
reg
V
drop
– 120.09/0 16.458/45.72 – 109.24/ 6.19 – = = = V
Tap
119 109.24 –
0.75
 13.02 = =
0812_frame_C07.fm Page 172 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 173
The generalized constants for modeling the regulator for this operating
condition are
Example 7.6
Using the results of Examples 7.5, calculate the actual voltage at the load
center assuming the 2500 kVA at 4.16 kV is measured at the substation trans
former lowvoltage terminals.
The actual linetoground voltage and line current at the load side
terminals of the regulator are
The actual linetoground voltage at the load center is
On a 120V base, the load center voltage is
Therefore, the +13 tap on the regulator has provided the desired
voltage at the load center.
As an exercise, the student should go back and, using the output voltage and
current of the regulator on the +13 tap, calculate the voltage across the voltage
relay in the compensator circuit. That value will be computed to be exactly
the same as the load center voltage on the 120V base.
It is important to understand that the value of equivalent line impedance is
not the actual impedance of the line between the regulator and the load center.
a a
R
0.9188 = =
b 0 =
c 0 =
d
1
0.9188
 1.0884 = =
V
L
V
S
a

2401.8/0
0.9188
 2614.2/0 = = = V
I
L
I
S
d

346.97/ 25.84 –
1.0884
 318.77/ 25.84 – = = = A
V
LC
V
L
Z
line
I
L
⋅ – 2614.2/0 0.3 j0.9 + ( ) 318.77/ 25.84 – ⋅ – = =
2412.8/ 5.15 – = V
VLC
120
V
LC
N
pt

2412.8/ 5.15 –
20
 120.6/ 5.15 – V = = =
0812_frame_C07.fm Page 173 Saturday, July 21, 2001 3:11 PM
174 Distribution System Modeling and Analysis
Typically, the load center is located down the primary main feeder after several
laterals have been tapped. As a result, the current measured by the CT of
the regulator is not the current that ﬂows all the way from the regulator to
the load center. The only way to determine the equivalent line impedance
value is to run a powerﬂow program of the feeder without the regulator
operating. From the output of the program the voltages at the regulator output
and the load center are known. Now the equivalent line impedance can be
computed as:
(7.82)
In Equation 7.82 the voltages must be speciﬁed in system volts and the
current in system amperes.
This section has developed the model and generalized constants for Type
A and Type B singlephase stepvoltage regulators. The compensator control
circuit has been developed, and it has been demonstrated how this circuit
controls the tap changing of the regulator. The next section will discuss the
various threephase steptype voltage regulators.
7.4.2 ThreePhase StepVoltage Regulators
Three singlephase stepvoltage regulators can be connected externally to
form a threephase regulator. When three singlephase regulators are con
nected together, each regulator has its own compensator circuit and, there
fore, the taps on each regulator are changed separately. Typical connections
for singlephase stepregulators are
1. Singlephase
2. Two regulators connected in “openwye” (sometimes referred to
as “V” phase)
3. Three regulators connected in grounded wye
4. Two regulators connected in open delta
5. Three regulators connected in closed delta
A threephase regulator has the connections between the singlephase wind
ings internal to the regulator housing. The threephase regulator is gang
operated so that the taps on all windings change the same and, as a result,
only one compensator circuit is required. For this case it is up to the engineer
to determine which phase current and voltage will be sampled by the com
pensator circuit. Threephase regulators will only be connected in a three
phase wye or closed delta.
Rline
Ω
jXline
Ω
+
V
regulator output
V
load center
–
I
line
 Ω =
0812_frame_C07.fm Page 174 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 175
In the regulator models to be developed in the next sections, the phasing
on the source side of the regulator will use capital letters A, B, and C. The
load side phasing will use lowercase letters a, b, and c.
7.4.2.1 WyeConnected Regulators
Three Type B singlephase regulators connected in wye are shown in Figure 7.12.
In Figure 7.12 the polarities of the windings are shown in the raise position.
When the regulator is in the lower position a reversing switch will have
reconnected the series winding so that the polarity on the series winding is
now at the output terminal. Regardless of whether the regulator is raising
or lowering the voltage, the following equations apply:
VOLTAGE EQUATIONS
(7.83)
where a
Ra
, a
Rb
, and a
Rc
represent the effective turns ratios for the three single
phase regulators. Equation 7.83 is of the form:
(7.84)
FIGURE 7.12
Wyeconnected type B regulators.
B
A
C I
C
a
b
c
I
a
I
b
I
c
+

V
an
+

V
An
V
An
V
Bn
V
Cn
a
R a
0 0
0 a
R b
0
0 0 a
R c
V
an
V
bn
V
cn
⋅ =
VLN
ABC
[ ] a [ ] VLN
abc
[ ] b [ ] I
abc
[ ] ⋅ + ⋅ =
0812_frame_C07.fm Page 175 Saturday, July 21, 2001 3:11 PM
176 Distribution System Modeling and Analysis
CURRENT EQUATIONS
(7.85)
or (7.86)
Equations 7.84 and 7.86 are of the same form as the generalized equations that
were developed for the threephase line segment of Chapter 6. For a three
phase wyeconnected stepvoltage regulator, neglecting the series impedance
and shunt admittance, the generalized matrices are deﬁned as:
(7.87)
(7.88)
(7.89)
(7.90)
In Equations 7.87 and 7.90, the effective turns ratio for each regulator must
satisfy:
0.9 ≤ a
R abc
≤ 1.1 in 32 steps of 0.625%/step (0.75 V/step on 120V base)
The effective turn ratios (a
Ra
, a
Rb
, and a
Rc
) can take on different values when
three singlephase regulators are connected in wye. It is also possible to have
a threephase regulator connected in wye where the voltage and current are
I
A
I
B
I
C
1
a
R a
 0 0
0
1
a
R b
 0
0 0
1
a
R c

I
a
I
b
I
c
= =
I
ABC
[ ] c [ ] VLG
abc
[ ] d [ ] I
abc
[ ] + ⋅ =
a [ ]
a
R a
0 0
0 a
R b
0
0 0 a
R c
=
b [ ]
0 0 0
0 0 0
0 0 0
=
c [ ]
0 0 0
0 0 0
0 0 0
=
d [ ]
1
a
R a
 0 0
0
1
a
R b
 0
0 0
1
a
R c

=
0812_frame_C07.fm Page 176 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 177
sampled on only one phase, and then all three phases are changed by the
same number of taps.
Example 7.7
An unbalanced threephase load is served at the end of a 10,000ft., 12.47kV
distribution line segment. The phase generalized matrices for the line seg
ment were computed in Example 6.1 as
For this line the A and B matrices are deﬁned as:
The linetoneutral voltages at the substation are balanced threephase:
The line currents at the substation for the unbalanced loading are
For the measured substation voltages and currents, the linetoneutral volt
ages at the load are computed as:
The load voltages on a 120V base are determined by dividing by the
potential transformer ratio that transforms rated linetoneutral voltage
a [ ]
1 0 0
0 1 0
0 0 1
=
b [ ]
0.8667 j2.0417 + 0.2955 j0.9502 + 0.2907 j0.7290 +
0.2955 j0.9502 + 0.8837 j1.9852 + 0.2992 j0.8023 +
0.2907 j0.7290 + 0.2992 j0.8023 + 0.8741 j2.0172 +
=
A [ ] a [ ]
1 –
=
B [ ] a [ ]
1 –
b [ ] ⋅ Z
abc
[ ] = =
VLN
ABC
[ ]
7200/0
7200/ 120 –
7200/120
V =
I
abc
[ ] I
ABC
[ ]
258/ 20 –
288/ 147 –
324/86
A = =
Vload
abc
[ ] A [ ] VLN
ABC
[ ] ⋅ B [ ] I
abc
[ ] ⋅ –
6965.1/ 2.1 –
6943.1/ 121.2 –
6776.7/117.8
V = =
0812_frame_C07.fm Page 177 Saturday, July 21, 2001 3:11 PM
178 Distribution System Modeling and Analysis
down to 120 V:
The load voltages on a 120V base are
Three singlephase Type B stepvoltage regulators are connected in wye and
installed in the substation. The regulators are to be set so that each lineto
neutral load voltage on a 120V base will lie between 119 and 121 volts.
The current transformers of the regulators are rated:
The equivalent line impedance for each phase can be determined by applying
Equation 7.79:
Even though the three regulators will change taps independently, it is the usual
practice to set the R and X settings of the three regulators the same. The average
value of the three line impedances above can be used for this purpose:
Zline
average
= 0.5787 + j1.1763
The compensator R and X settings are computed according to Equation 7.78:
R′ + jX′ = (Rline
Ω
+ jXline
Ω
) ⋅ = (0.5787 + j1.1763) ⋅
R′ + jX′ = 5.787 + j11.763 V
The compensator controls are not calibrated to that many signiﬁcant ﬁgures,
so the values set are
R′ + jX′ = 6 + j12 V
N
pt
7200
120
 60 = =
V
120
[ ]
1
60
 Vload
abc
[ ] ⋅
116.1/ 2.1 –
115.7/ 121.2 –
112.9/117.8
V = =
CT
600
5

CT
P
CT
S
 120 = = =
Zline
a
7200/0 6965.7/ 2.1 – –
258/ 20 –
 0.5346 j1.2385 Ω + = =
Zline
b
7200/ 12 – 0 6943.1/ 121.2 – –
288/ 147 –
 0.5628 j0.8723 Ω + = =
Zline
c
7200/120 6776.7/117.8 –
324/86
 0.6386 j1.418 Ω + = =
CT
P
N
PT

600
60

0812_frame_C07.fm Page 178 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 179
The compensator control will be set such that the voltage level = 120 V with
a bandwidth of 2 V.
For the same unbalanced loading, and with the threephase wyeconnected
regulators in service, the approximate tap settings are
Since the taps must be integers, the actual tap settings will be
The effective turns ratio for the three regulators and the resulting generalized
matrices are determined by applying Equation 7.75 for each phase:
The metered substation voltages and currents are the inputs to the three
voltage regulators. The output voltages and currents of the regulators are
Tap
a
119 Vload
a
–
0.75

119 116.1 –
0.75
 3.8736 = = =
Tap
b
119 Vload
b
–
0.75

119 115.7 –
0.75
 4.3746 = = =
Tap
a
119 Vload
c
–
0.75

119 112.9 –
0.75
 8.0723 = = =
Tap
a
+4 =
Tap
b
+5 =
Tap
c
+9 =
a [ ]
1 0.00625 4 ⋅ – 0 0
0 1 0.00625 5 ⋅ – 0
0 0 1 0.00625 8 ⋅ –
=
0.975 0 0
0 0.9688 0
0 0 0.9500
=
d [ ] a [ ]
1 –
1.0256 0 0
0 1.0323 0
0 0 1.0526
= =
Vreg
abc
[ ] a [ ]
1 –
VLN
ABC
[ ] ⋅
7384.6/0
7432.3/ 120 –
7578/120
V = =
0812_frame_C07.fm Page 179 Saturday, July 21, 2001 3:11 PM
180 Distribution System Modeling and Analysis
The output currents of the regulators are
With the regulators adjusted, the load voltages can be computed to be
On a 120volt base the load voltages are
With the given regulator taps, the load voltages all now lie between the
desired voltage limits of 119 and 121 V per the voltage level and bandwidth
settings of the compensator circuit.
Example 7.7 is a long example intended to demonstrate how the engineer
can determine the correct compensator R and X settings knowing the sub
station and load voltages and the currents leaving the substation. Generally,
it will be necessary to run a powerﬂow study to determine these values.
The example demonstrates that with the regulator tap settings, the load
voltages lie within the desired limits. The regulator has automatically set the
taps for this load condition and, as the load changes, the taps will continue
to change in order to hold the load voltages within the desired limits.
7.4.2.2 Closed DeltaConnected Regulators
Three singlephase Type B regulators can be connected in a closed delta as
shown in Figure 7.13. In the ﬁgure, the regulators are shown in the raise
position. The closed delta connection is typically used in threewire delta
feeders. Note that the voltage transformers for this connection are monitor
ing the load side linetoline voltages, and the current transformers are not
monitoring the load side line currents.
Ireg
abc
[ ] d [ ]
1 –
I
sub
[ ] ⋅
251.6/ 20 –
279.0/ 147 –
307.8/86
A = =
Vload
abc
[ ] A [ ] Vreg
abc
[ ] B [ ] Ireg
abc
[ ] ⋅ – ⋅
7150.7/ 2.0 –
7185.5/121.2
7179.1/118.1
V = =
V
120
[ ]
1
60

7150.7/ 2.0 –
7185.5/ 121.2 –
7179.1/118.2
⋅
119.2/ 2.0 –
119.8/ 121.2 –
119.7/118.1
V = =
0812_frame_C07.fm Page 180 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 181
The relationships between the source side and currents and voltages are
needed. Equations 7.64 through 7.67 deﬁne the relationships between the
series and shunt winding voltages, and currents for a stepvoltage regulator.
These must be satisﬁed no matter how the regulators are connected.
Kirchhoff’s voltage law is ﬁrst applied around a closed loop, starting with
the linetoline voltage between phases A and B on the source side. Refer to
Figure 7.13.
V
AB
+ V
Bb
+ V
ba
+ V
aA
= 0 (7.91)
but (7.92)
(7.93)
V
ba
= −V
ab
(7.94)
Substitute Equations 7.92, 7.93, and 7.94 into Equation 7.91 and simplify:
(7.95)
FIGURE 7.13
Closed deltaconnected regulators.
L
SL
S
L
S
L
S
L
SL S
A
C
B
I
I
A
C
I
B
I
ca
I
C
I
c'
I
bc
I
b'
I
B
I
I
A
ab
I
a'
I
a
I
b
I
c
a
b
c
V
Bb
N
2
N
1
 – V
bc
⋅ =
V
aA
N
2
N
1
 V
ab
⋅ =
V
AB
1
N
2
N
1
 –
V
ab
N
2
N
1
 V
bc
⋅ + ⋅ a
R ab
V
ab
1 a
R bc
– ( ) V
bc
⋅ + ⋅ = =
0812_frame_C07.fm Page 181 Saturday, July 21, 2001 3:11 PM
182 Distribution System Modeling and Analysis
The same procedure can be followed to determine the relationships between
the other linetoline voltages. The ﬁnal threephase equation is
(7.96)
Equation 7.96 is of the generalized form:
(7.97)
The relationship between source and load line currents starts with applying
KCL at the load side terminal a:
I
a
= + I
ca
= I
A
− I
ab
+ I
ca
(7.98)
but
(7.99)
(7.100)
Substitute Equations 7.96 and 7.97 into Equation 7.95 and simplify:
(7.101)
The same procedure can be followed at the other two load side terminals.
The resulting threephase equation is
(7.102)
Equation 7.102 is of the general form:
(7.103)
V
AB
V
BC
V
CA
a
R ab
1 a
R bc
– 0
0 a
R bc
1 a
R ca
–
1 a
R ab
– 0 a
R ca
V
ab
V
bc
V
ca
⋅ =
VLL
ABC
[ ] a [ ] VLL
abc
[ ] ⋅ b [ ] I
abc
[ ] ⋅ + =
I′
a
I
ab
N
2
N
1
 I
A
⋅ =
I
ca
N
2
N
1
 I
C
⋅ =
I
a
1
N
2
N
1
 –
I
A
N
2
N
1
I
C
+ ⋅ a
R ab
I
A
1 a
R ca
– ( ) I
C
⋅ + ⋅ = =
I
a
I
b
I
c
a
R ab
0 1 a
R ca
–
1 a
R ab
– a
R bc
0
0 1 a
R bc
– a
R ca
I
A
I
B
I
C
⋅ =
I
abc
[ ] d [ ]
1 –
I
ABC
[ ] ⋅ =
0812_frame_C07.fm Page 182 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 183
where
With the exception of the case when all three regulators are in the neutral
position (a
R
= 1), [d]
−1
does not give a simple expression for each of the
elements. In general, all of the elements of [d] will be nonzero. However,
when the tap positions of each regulator are known, the elements of the
inverse matrix can be determined. The inverse of the resulting matrix gives
the matrix [d] so that the generalized form of the current equations can still
be applied.
(7.104)
As with the wyeconnected regulators, the matrices [b] and [c] are zero as
long as the series impedance and shunt admittance of each regulator are
neglected.
The closed delta connection can be difﬁcult to apply. Note that in both the
voltage and current equations, a change of the tap position in one regulator
will affect voltages and currents in two phases. As a result, increasing the
tap in one regulator will affect the tap position of the second regulator.
Therefore, in most cases the bandwidth setting for the closed delta connec
tion will have to be wider than that for wyeconnected regulators.
7.4.2.3 Open DeltaConnected Regulators
Two Type B singlephase regulators can be connected in the open delta
connection. Shown in Figure 7.14 is an open delta connection where two
singlephase regulators have been connected between phases AB and CB.
Two additional open connections can be made by connecting the single
phase regulators between phases BC and AC, and also between phases CA
and BA.
The open delta connection is typically applied to threewire delta feeders.
Note that the potential transformers monitor the linetoline voltages and
the current transformers monitor the line currents. Once again, the basic
voltage and current relations of the individual regulators are used to deter
mine the relationships between the source side and load side voltages and
currents. The connection shown in Figure 7.14 will be used to derive the
relationships, and then the relationships of the other two possible connec
tions will merely be stated.
d [ ]
1 –
a
R ab
0 1 a
R ca
–
1 a
R ab
– a
R bc
0
0 1 a
R bc
– a
R ca
=
I
ABC
[ ] c [ ] VLL
ABC
[ ] d [ ] I
abc
[ ] ⋅ + ⋅ =
0812_frame_C07.fm Page 183 Saturday, July 21, 2001 3:11 PM
184 Distribution System Modeling and Analysis
The voltage drop V
AB
across the ﬁrst regulator consists of the drop across
the series winding plus the drop across the shunt winding:
V
AB
= V
AL
+ V
ab
(7.105)
where V
AL
= drop across the series winding.
Paying attention to the polarity marks on the series and shunt windings, the
drop across the series winding is
(7.106)
Substituting Equation 7.106 into Equation 7.105 yields:
(7.107)
Following the same procedure for the regulator connected across V
BC
, the
voltage equation is
(7.108)
Kirchhoff’s voltage law must be satisﬁed so that:
V
CA
= −(V
AB
+ V
BC
) = −a
R ab
⋅ V
ab
− a
R cb
⋅ V
bc
(7.109)
FIGURE 7.14
Open delta connection.
C
I
cb
I
b
a
c
I
b
I
a
I
ab
I
A
I
B
I
L
SL S
L
S
L
S
V
V
V
+


CA
AB
C
+

A

BC
B
+
+

V
bc
V
ca
V
ab
+ 
+
c
V
AL
N
2
N
1
 V
ab
⋅ – =
V
AB
1
N
2
N
1
 –
V
ab
⋅ a
R ab
V
ab
⋅ = =
V
BC
1
N
2
N
1
 –
V
bc
⋅ a
R cb
V
bc
⋅ = =
0812_frame_C07.fm Page 184 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 185
Equations 7.107, 7.108, and 7.109 can be put into matrix form:
(7.110)
Equation 7.110 in generalized form is
(7.111)
where (7.112)
The effective turns ratio of each regulator is given by Equation 7.75. Again, as
long as the series impedance and shunt admittance of the regulators are
neglected, [b
LL
] is zero. Equation 7.111 gives the linetoline voltages on the
source side as functions of the linetoline voltages on the load side of the open
delta using the generalized matrices. Up to this point the relationships between
the voltages have been in terms of linetoneutral voltages. In Chapter 10 it
will be shown how to convert this equation using equivalent linetoneutral
voltages.
When the load side linetoline voltages are needed as functions of the
source side linetoline voltages, the necessary equation is
(7.113)
(7.114)
where (7.115)
V
AB
V
BC
V
CA
a
R ab
0 0
0 a
R cb
0
a
R ab
– a
R cb
– 0
V
ab
V
bc
V
ca
⋅ =
VLL
ABC
[ ] a
LL
[ ] VLL
abc
[ ] b
LL
[ ] I
abc
[ ] ⋅ + ⋅ =
a
LL
[ ]
a
R ab
0 0
0 a
R cb
0
a
R ab
– a
R cb
– 0
=
V
ab
V
bc
V
ca
1
a
R ab
 0 0
0
1
a
R cb
 0
1
a
R ab
 –
1
a
R cb
 – 0
V
AB
V
BC
V
CA
⋅ =
VLL
abc
[ ] A
LL
[ ] VLL
ABC
[ ] ⋅ =
A
LL
[ ]
1
a
R ab
 0 0
0
1
a
R cb
 0
1
a
R ab
 –
1
a
R cb
 – 0
=
0812_frame_C07.fm Page 185 Saturday, July 21, 2001 3:11 PM
186 Distribution System Modeling and Analysis
The current equations are derived by applying KCL at the L node of each
regulator:
I
A
= I
a
+ (7.116)
but
Therefore, Equation 7.116 becomes:
(7.117)
therefore (7.118)
In a similar manner, the current equation for the second regulator is given by:
(7.119)
Because this is a threewire delta line, then:
(7.120)
In matrix form, the current equations become:
(7.121)
In generalized form Equation 7.121 becomes:
(7.122)
I
ab
I
ab
N
2
N
1
 I
A
⋅ =
1
N
2
N
1
 –
I
A
I
a
=
I
A
1
a
R ab
 I
a
⋅ =
I
C
1
a
R cb
 I
c
⋅ =
I
B
I
A
I
C
+ ( ) –
1
a
R ab
 I
a
1
a
R cb
 I
c
⋅ – ⋅ – = =
I
A
I
B
I
C
1
a
R ab
 0 0
1
a
R ab
 – 0
1
a
R cb
 –
0 0
1
a
R cb

I
a
I
b
I
c
⋅ =
I
ABC
[ ] c
LL
[ ] VLL
ABC
[ ] d
LL
[ ] I
abc
[ ] ⋅ + ⋅ =
0812_frame_C07.fm Page 186 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 187
where (7.123)
When the series impedances and shunt admittances are neglected, the con
stant matrix [c
LL
] will be zero.
The load side line currents as functions of the source line currents are given by:
(7.124)
(7.125)
where (7.126)
The determination of the R and X compensator settings for the open delta
follows the same procedure as that of the wyeconnected regulators. However,
care must be taken to recognize that in the open delta connection, the voltages
applied to the compensator are linetoline and the currents are line currents.
The open deltaconnected regulators will maintain only two of the linetoline
voltages at the load center within deﬁned limits. The third linetoline voltage
will be dictated by the other two (Kirchhoff’s voltage law). Therefore, it is
possible that the third voltage may not be within the deﬁned limits. With
reference to Figure 7.15, an equivalent impedance between the regulators and
the load center must be computed. Since each regulator is sampling lineto
line voltages and a line current, the equivalent impedance is computed by
taking the appropriate linetoline voltage drop and dividing by the sampled
line current. For the open delta connection shown in Figure 7.15, the equivalent
impedances are computed as:
(7.127)
(7.128)
d
LL
[ ]
1
a
R ab
 0 0
1
a
R ab
 – 0
1
a
R cb
 –
0 0
1
a
R cb

=
I
a
I
b
I
c
a
R ab
0 0
a
R ab
– 0 a
R cb
–
0 0 a
R cb
I
A
I
B
I
C
⋅ =
I
abc
[ ] D
LL
[ ] I
ABC
[ ] ⋅ =
D
LL
[ ]
a
R ab
0 0
a
R ab
– 0 a
R cb
–
0 0 a
R cb
=
Zeq
a
VR
ab
VL
ab
–
I
a
 =
Zeq
c
VR
cb
VL
cb
–
I
c
 =
0812_frame_C07.fm Page 187 Saturday, July 21, 2001 3:11 PM
188 Distribution System Modeling and Analysis
The units of these impedances will be in system ohms. They must be con
verted to compensator volts by applying Equation 7.78. For the open delta
connection, the potential transformer will transform the system linetoline
rated voltage down to 120 V. Example 7.8 demonstrates how the compensator
R and X settings are determined knowing the linetoline voltages at the
regulator and at the load center.
Example 7.8
A powerﬂow study has been run on a system prior to the installation of an
open delta regulator bank at Node R. The load center is at Node L, as shown
in Figure 7.16. The results of the powerﬂow study are
Node R: = = =
I
a
= I
b
= I
c
=
Node L: = = =
FIGURE 7.15
Open delta connected to a load center.
FIGURE 7.16
Circuit for Example 7.8.
C
I
cb
I
c
I
b
I
a
I
ab
I
A
I
B
I
L
SL S
L
S
L
S
V
V
V
+


CA
AB
C
+

A

BC
B
+
+
a
c
b
VR
ab
v
a
v
b
v
c
+
VR
cb

+

VL
ab

+
VL
cb
Load
Center
VR
ab
12,470/0 VR
bc
12,470/ 120 – VR
ca
12,470/120
308.2/ 58.0 – 264.2/ 176.1 – 297.0/70.3
VL
ab
11,911/ 1.4 – VL
bc
12,117/ 122.3 – VL
ca
11,859/117.3
L R
0812_frame_C07.fm Page 188 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages
189
For this connection, the potential transformer ratio and current transformer
ratios are selected to be
On a 120V base, the load center voltages are
Two singlephase Type B regulators are to be installed in an open delta con
nection. The regulators are to be connected between phases
A

B
and
B

C
as
shown in Figure 7.15. The voltage level will be set at 120 V with a bandwidth
of 2 V. As computed above, the load center voltages are not within the desired
limits of 120
±
1 V.
The compensator
R
and
X
settings must ﬁrst be determined using the
results of the powerﬂow study. The ﬁrst regulator monitors the voltage
V
ab
and the line current
I
a
. The equivalent line impedance for this regulator is
The second regulator monitors the voltage and the line current
I
c
. In the
computation of the equivalent line impedance, it is necessary to use the
cb
voltages, which are the negatives of the given
b

c
voltages:
Unlike the wyeconnected regulators, the compensator settings for the two
regulators will be different. The settings calibrated in volts are
N
pt
12,470
120
 103.92 = =
CT
500
5
 100 = =
V120
ab
V120
bc
V120
ca
1
103.92

11,911/ 1.4 –
12,117/ 122.3 –
11,859/117.3
⋅
114.6/ 1.4 –
116.6/ 122.3 –
114.3/117.5
V = =
Zeq
a
VR
ab
VL
ab
–
I
a

12,470/0 11,911/ 1.4 – –
308.2/ 58.0 –
 0.1665 j2.0483 Ω + = = =
V
cb
Zeq
c
VR
cb
VL
cb
–
I
c

12,470/60 12,117/ 57.7 – –
297.0/70.3
 1.4945 j1.3925 Ω + = = =
R′
ab
jX′
ab
+ Z
a
CT
P
N
PT
 ⋅ 0.1665 j2.0483 + ( )
500
103.92
 ⋅ 0.8012 j9.8555 V + = = =
R′
cb
jX′
cb
+ Z
c
CT
P
N
PT
 ⋅ 0.1495 j1.3925 + ( )
500
103.92
 ⋅ 7.1908 j6.7002 V + = = =
0812_frame_C07.fm Page 189 Monday, October 28, 2002 10:42 AM
190
Distribution System Modeling and Analysis
The compensator settings will be
With regulators installed and in the neutral position, and with the same
loading, the currents and voltages in the compensator circuits are
The compensator impedances in ohms are determined by dividing the set
tings in volts by the secondary rating of the current transformer:
The voltages across the voltage relays in the two compensator circuits are
Since the voltages are below the lower limit of 119, the control circuit will
send “raise” commands to change the taps on both regulators. The number
of tap changes necessary to bring the load center voltage into the lower limit
of the bandwidth for each regulator will be
R′
ab
X′
ab
+ 0.8 j9.9 + =
R′
cb
jX′
cb
+ 7.2 j6.7 + =
Vcomp
ab
VR
ab
N
pt

12,470/0
103.92
 120/0 = = = V
Vcomp
cb
VR
bc
–
N
pt

12,470/60
103.92
 120/60 = = = V
Icomp
a
I
a
CT

308.2/ 58.0 –
100
 3.082/ 58.0 – = = = A
Icomp
c
I
c
CT

297.0/70.3
100
 2.97/70.3 A = = =
R
ab
jX
ab
+
R′
ab
jX′
ab
+
CT
secondary

0.8
j
9.9 +
5
 0.16 j1.98 + = = =
R
cb
jX
cb
+
R′
cb
jX′
cb
+
CT
secondary

7.2
j
6.7 +
5
 1.44 j1.34 + = = =
Vrelay
ab
Vcomp
ab
R
ab
jX
ab
+ ( ) Icomp
a
⋅ – 114.6/ 1.4 – = = V
Vrelay
cb
Vcomp
cb
R
cb
jX
cb
+ ( ) Icomp
c
⋅ – 116.6/57.7 = = V
Tap
ab
119 114.6 –
0.75
 5.47 6 ≈ = =
Tap
cb
119 116.6 –
0.75
 3.20 4 ≈ = =
0812_frame_C07.fm Page 190 Monday, October 28, 2002 10:42 AM
Regulation of Voltages 191
With the taps set at 6 and 4, a check can be made to determine if the voltages
at the load center are now within the limits.
With the taps adjusted, the regulator ratios are
In order to determine the load side regulator voltages and currents, the
matrices [A
LL
] (Equation 7.112) and [D
LL
] (Equation 7.123), must be deﬁned:
The output voltages from the regulators are
The output currents from the regulators are
There are two ways to test if the voltages at the load center are within the
limits. The ﬁrst method is to compute the relay voltages in the compensator
circuits. The procedure is the same as was done initially to determine the
load center voltages. First the voltages and currents in the compensator
a
R ab
1.0 0.00625 Tap
ab
⋅ – 0.9625 = =
a
R cb
1.0 0.00625 Tap
cb
⋅ – 0.975 = =
A
LL
[ ]
1
0.9625
 0 0
0
1
0.975
 0
1
0.9625
 –
1
0.975
 – 0
1.039 0 0
0 1.0256 0
1.039 – 1.0256 – 0
= =
D
LL
[ ]
0.9625 0 0
0.9625 – 0 0.975 –
0 0 0.975
=
Vreg
abc
[ ] A
LL
[ ] VLL
ABC
[ ] ⋅
12,956/0
12,790/ 120 –
12,874/120
V = =
I
abc
[ ] D
LL
[ ] I
ABC
[ ] ⋅
296.6/ 58 –
255.7/ 175.3 –
289.6/70.3
A = =
0812_frame_C07.fm Page 191 Saturday, July 21, 2001 3:11 PM
192 Distribution System Modeling and Analysis
circuits are computed:
The voltages across the voltage relays are computed to be
Since both voltages are within the bandwidth, no further tap changing will
be necessary.
The actual voltages at the load center can be computed using the output
voltages and currents from the regulator and then computing the voltage
drop to the load center.
The phase impedance matrix for the line between the regulator and the
load center is
With reference to Figure 7.15, the voltage drops per phase are
The load center linetoline voltages are
Vcomp
ab
VR
ab
N
pt

12,956/0
103.92
 124.7/0 V = = =
Vcomp
cb
VR
bc
–
N
pt

12,790/60
103.92
 123.1/60 = = = V
Icomp
a
I
a
CT

296.6/ 58.0 –
100
 2.966/ 58.0 – = = = A
Icomp
c
I
c
CT

289.6/70.3
100
 2.896/70.3 = = = A
Vrelay
ab
Vcomp
ab
R
ab
jX
ab
+ ( ) Icomp
a
⋅ – 119.5/ 1.3 – = = V
Vrelay
cb
Vcomp
cb
R
cb
jX
cb
+ ( ) Icomp
a
⋅ – 119.8/57.8 = = V
Z
abc
[ ]
0.7604 j2.6762 + 0.1804 j1.6125 + 0.1804 j1.2762 +
0.1804 j1.6125 + 0.7604 j2.6762 + 0.1804 j1.4773 +
0.1804 j1.2762 + 0.1804 j1.4773 + 0.7604 j2.6762 +
Ω =
v
abc
[ ] Z
abc
[ ] I
abc
[ ] ⋅
450.6/ 1.5 –
309.4/ 106.6 –
402.8/142.8
V = =
VL
ab
Vreg
ab
v
a
v
b
+ – 12,420/ 1.3 – V = =
VL
bc
Vreg
bc
v
b
v
c
+ – 12,447/ 122.2 – V = =
VL
ca
Vreg
ca
v
c
v
a
+ – 12,273/ 118.1 – V = =
0812_frame_C07.fm Page 192 Saturday, July 21, 2001 3:11 PM
Regulation of Voltages 193
Dividing the load center linetoline voltages by the potential transformer
ratio gives the voltages on the 120V base as:
The desired voltages are being held on the two phases that have the voltage
regulators. The third linetoline voltage is below the limit. This cannot be
helped since that voltage is being dictated by the other two linetoline volt
ages. The only way to bring that voltage up is to set a higher voltage level on
the two regulators.
This example is very long but has been included to demonstrate how the
compensator circuit is set, and then how it will adjust taps so that the voltages
at a remote load center node will be held within the set limits. In actual
practice, the only responsibilities of the engineer will be to correctly deter
mine the R and X settings of the compensator circuit, and to determine the
desired voltage level and bandwidth.
The open delta regulator connection using phases AB and CB has been
presented. There are two other possible open delta connections using phases
BC and AC, and then CA and BA. Generalized matrices for these addi
tional two connections can be developed using the procedures presented in
this section.
7.5 Summary
It has been shown that all possible connections for Type B stepvoltage
regulators can be modeled using the generalized matrices. The derivations
in this chapter were limited to threephase connections. If a singlephase
regulator is connected linetoneutral or two regulators connected in open
wye, then the [a] and [d] matrices will be of the same form as that of the wye
connected regulators, only the terms in the rows and columns associated with
the missing phases will be zero. The same can be said for a singlephase
regulator connected linetoline. Again, the rows and columns associated with
the missing phases would be set to zero in the matrices developed for the
open delta connection.
The generalized matrices developed in this chapter are of exactly the same
form as those developed for the threephase line segments. In the next
chapter, the generalized matrices for all threephase transformers will be
developed.
V120
ab
119.5/ 1.3 – V =
V120
bc
119.8/ 122.2 V – =
V120
ca
118.1/118.1 V =
0812_frame_C07.fm Page 193 Saturday, July 21, 2001 3:11 PM
194
Distribution System Modeling and Analysis
References
1.
American National Standard for Electric Power – Systems and Equipment Voltage
Ratings (60 Hertz),
ANSI C84.11995, National Electrical Manufacturers Asso
ciation, Rosslyn, VA, 1996.
2.
IEEE Standard Requirements, Terminology, and Test Code for StepVoltage and
InductionVoltage Regulators,
ANSI/IEEE C57.151986, Institute of Electrical and
Electronic Engineers, New York, 1988.
Problems
7.1
A singlephase transformer is rated 100 kVA, 2400240 V. The imped
ances and admittance of the transformer are
Z
1
=
0.65
+
j
0.95
Ω
(highvoltage winding impedance)
Z
2
=
0.0052
+
j
0.0078
Ω
(lowvoltage winding impedance)
Y
m
=
2.56
×
10
−
4
−
j
11.37
×
10
−
4
S (referred to the highvoltage winding)
(1) Determine the
a
,
b
,
c
,
d
constants and the
A
and
B
constants.
(2) The transformer is serving an 80kW, 0.85 lagging power factor
load at 230 V. Determine the primary voltage, current, complex
power, and percent voltage drop.
(3) Determine the perunit transformer impedance and shunt admit
tance based upon the transformer ratings.
7.2
The singlephase transformer of Problem 7.1 is to be connected as a
stepdown autotransformer to transform the voltage from 2400 V to 2160 V.
(1) Draw the connection diagram, including the series impedance and
shunt admittance.
(2) Determine the autotransformer kVA rating.
(3) Determine the
a
,
b
,
c
,
d
,
A
, and
B
generalized constants.
(4) The autotransformer is serving a load of 800 kVA, 0.95 lagging
power factor at a voltage of 2000 V. Including the impedance and
shunt admittance, determine the input voltage, current, complex
power, and percent voltage drop.
(5) Determine the perunit impedance and shunt admittance based
upon the autotransformer rating. How do these values compare to
the perunit values of Problem 7.1?
0812_frame_C07.fm Page 194 Monday, October 28, 2002 10:44 AM
Regulation of Voltages
195
7.3
A Type B stepvoltage regulator is installed to regulate the voltage on
a 7200V singlephase lateral. The potential transformer and current trans
former ratios connected to the compensator circuit are
Potential transformer: 7200120 V
Current transformer: 500:5 A
The
R
and
X
settings in the compensator circuit are:
R
=
5 V and
X
=
10 V.
The regulator taps are set on the
+
10 position when the voltage and current
on the source side of the regulator are
V
source
=
7200 V and
I
source
=
375 at a power factor of 0.866 lagging power factor.
(1) Determine the voltage at the load center.
(2) Determine the equivalent line impedance between the regulator
and the load center.
(3) Assuming that the voltage level on the regulator has been set at
120 V with a bandwidth of 2 V, what tap will the regulator move
to?
7.4
Refer to Figure 7.11. The substation transformer is rated 24 MVA, 230 kV
delta13.8 kV wye. Three singlephase Type B regulators are connected in
wye. The equivalent line impedance between the regulators and the load
center node is
Z
line
=
0.264 +
j
0.58
Ω
/mile
The distance to the load center node is 10,000 ft.
(1) Determine the appropriate
PT
and
CT
ratios.
(2) Determine the
R
′
and
X
′
settings in ohms and volts for the com
pensator circuit.
(3) The substation is serving a balanced threephase load of 16 MVA,
0.9 lagging power factor when the output linetoline voltages of
the substation are balanced 13.8 kV and the regulators are set in
the neutral position. Assume the voltage level is set at 121 V and
a bandwidth of 2 V. Determine the ﬁnal tap position for each regu
lator (they will be the same). The regulators have 32 – 5/8% taps
(16 raise and 16 lower).
(4) What would be the regulator tap settings for a load of 24 MVA, 0.9
lagging power factor with the output voltages of the substation
transformer balanced threephase 13.8 kV?
(5) What would be the load center voltages for the load of Part 4 above?
7.5
Three Type B stepvoltage regulators are connected in wye and located
on the secondary bus of a 12.47kV substation. The feeder is serving an
0812_frame_C07.fm Page 195 Monday, October 28, 2002 10:46 AM
196
Distribution System Modeling and Analysis
unbalanced load. A powerﬂow study has been run and the voltages at the
substation and the load center node are
The currents at the substation are
The regulator potential transformer ratio is 7200120 and the current trans
former ratio is 500:5. The voltage level of the regulators is set at 121 V and
the bandwidth at 2 V.
(1) Determine the equivalent line impedance per phase between the
regulator and the load center.
(2) The compensators on each regulator are to be set with the same
R
and
X
values. Specify these values in volts and in ohms.
7.6
The impedance compensator settings for the three step regulators of
Problem 7.5 have been set as:
R
′
=
3.0 V
X
′
=
9.3 V
The voltages and currents at the substation bus are
Determine the ﬁnal tap settings for each regulator.
Vsub
abc
[ ]
7200/0
7200/ 120 –
7200/120
V =
VLC
abc
[ ]
6890.6/ 1.49 –
6825.9/ 122.90 –
6990.5/117.05
V =
I
abc
[ ]
362.8/ 27.3 –
395.4/ 154.7 –
329.0/98.9
A =
Vsub
abc
[ ]
7200/0
7200/ 120 –
7200/120
V =
I
abc
[ ]
320.6/ 27.4 –
409.0/ 155.1 –
331.5/98.2
A =
0812_frame_C07.fm Page 196 Monday, October 28, 2002 10:46 AM
Regulation of Voltages 197
7.7 For a different load condition for the system of Problem 7.5, the taps
on the regulators have been automatically set by the compensator circuit to:
Tap
a
= +8 Tap
b
= +11 Tap
c
= +6
The load reduces so that the voltages and currents at the substation bus are
Determine the new ﬁnal tap settings for each regulator.
7.8 The load center node for the regulators described in Problem 7.5 is
located 1.5 miles from the substation. There are no lateral taps between the
substation and the load center. The phase impedance matrix of the line
segment is
A wyeconnected, unbalanced constant impedance load is located at the load
center node. The load impedances are
ZL
a
= 19 + j11 Ω, ZL
b
= 22 + j12 Ω, ZL
c
= 18 + j10 Ω
The voltages at the substation are balanced threephase of 7200 V lineto
neutral. The regulators are set on neutral.
(1) Determine the linetoneutral voltages at the load center.
(2) Determine the R and X settings in volts for the compensator.
(3) Determine the required tap settings in order to hold the load center
voltages within the desired limits.
Vsub
abc
[ ]
7200/0
7200/ 120 –
7200/120
V =
I
abc
[ ]
177.1/ 28.5 –
213.4/ 156.4 –
146.8/98.3
A =
z
abc
[ ]
0.3465 j1.0179 + 0.1560 j0.5017 + 0.1580 j0.4236 +
0.1560 j0.5017 + 0.3375 j1.0478 + 0.1535 j0.3849 +
0.1580 j0.4236 + 0.1535 j0.3849 + 0.3414 j1.0348 +
Ω/mile =
0812_frame_C07.fm Page 197 Saturday, July 21, 2001 3:11 PM
198 Distribution System Modeling and Analysis
7.9 The phase impedance matrix for a threewire line segment is
The line is two miles long and serving an unbalanced load, with the substa
tion transformer linetoline voltages and output currents:
Two Type B stepvoltage regulators are connected in open delta at the sub
station using phases AB and CB. The potential transformer ratios are
12,470/120, and the current transformer ratios are 500:5. The voltage level
is set at 121 V with a 2V bandwidth.
(1) Determine the linetoline voltages at the load center.
(2) Determine the R and X compensator settings in volts. For the open
delta connection, the R and X settings will be different on each
regulator.
(3) Determine the ﬁnal tap positions of the two voltage regulators.
7.10 The regulator in Problem 7.9 has gone to the +9 tap on both regulators
for a particular load. The load is reduced so that the currents leaving the
substation transformer with the regulators in the +9 position are
Determine the ﬁnal tap settings on each regulator for this new load condition.
z
abc
[ ]
0.4013 j1.4133 + 0.0953 j0.8515 + 0.0953 j0.7802 +
0.0953 j0.8515 + 0.4013 j1.4133 + 0.0953 j0.7266 +
0.0953 j0.7802 + 0.0953 j0.7266 + 0.4013 j1.4133 +
Ω/mile =
VLL
abc
[ ]
12,470/0
12,470/ 120 –
12,470/120
V =
I
abc
[ ]
307.9/ 54.6 –
290.6/178.6
268.2/65.3
A =
I
abc
[ ]
144.3/ 53.5 –
136.3/179.6
125.7/66.3
A =
0812_frame_C07.fm Page 198 Saturday, July 21, 2001 3:11 PM
199
8
ThreePhase Transformer Models
Threephase transformer banks are found in the distribution substation
where the voltage is transformed from the transmission or subtransmission
level to the distribution feeder level. In most cases the substation transformer
will be a threephase unit, perhaps with highvoltage noload taps and, per
haps, lowvoltage load tap changing (LTC). For a fourwire wye feeder, the
most common substation transformer connection is the delta–grounded wye.
A threewire delta feeder will typically have a delta–delta transformer con
nection in the substation. Threephase transformer banks out on the feeder
will provide the ﬁnal voltage transformation to the customer’s load. A vari
ety of transformer connections can be applied. The load can be pure three
phase or a combination of singlephase lighting load and a threephase load
such as an induction motor. In the analysis of a distribution feeder, it is
important that the various threephase transformer connections be modeled
correctly.
Unique models of threephase transformer banks applicable to radial dis
tribution feeders will be developed in this chapter. Models for the following
threephase connections are included:
• Delta–Grounded Wye
• Ungrounded Wye–Delta
• Grounded Wye–Grounded Wye
• Delta–Delta
• Open Wye–Open Delta
8.1 Introduction
Figure 8.1 deﬁnes the various voltages and currents for all transformer banks
connected between the source side
Node n
and the load side
Node m
. In
Figure 8.1 the models can represent a stepdown (source side to load side)
or a stepup (source side to load side) transformer bank. The notation is such
0812_frame_C08.fm Page 199 Friday, July 20, 2001 4:47 PM
200
Distribution System Modeling and Analysis
that the capital letters A, B, C, N will always refer to the source side (
Node n
)
of the bank and the lower case letters a, b, c, n will always refer to the load
side (
Node m
) of the bank. It is assumed that all variations of the wye–delta
connections are connected in the “American Standard ThirtyDegree” con
nection. With the described phase notation, the standard phase shifts for
positive sequence voltages and currents are
StepDown Connection
V
AB
leads
V
ab
by 30 degrees (8.1)
I
A
leads
I
a
by 30 degrees (8.2)
StepUp Connection
V
ab
leads
V
AB
by 30 degrees (8.3)
I
a
leads
I
A
by 30 degrees (8.4)
8.2 Generalized Matrices
The models to be used in powerﬂow and shortcircuit studies are general
ized for the connections in the same form as have been developed for line
segments (Chapter 6) and voltage regulators Chapter 7). The matrix equa
tions for computing the voltages and currents at
Node n
as a function of the
voltages and currents at
Node m
are given by:
(8.5)
(8.6)
FIGURE 8.1
General threephase transformer bank.
I
I
I
V
V
V
V
V
AN
BN
CN
AB
BC
A
B
C
N
+
+
+
  
+
+
+



V
CA
V
V
V
V
V
V
an
bn
cn
ab
bc
ca
I
I
a
b
c
I
n
+
  
+
+
+
+
+



H0
H1 X1
H2
H3
X2
X3
X0
Source Side Load Side
VLN
ABC
[ ] a
t
[ ] · VLN
abc
[ ] b
t
[ ] · I
abc
[ ] + =
I
ABC
[ ] c
t
[ ] · VLN
abc
[ ] d
t
[ ] · I
abc
[ ] + =
0812_frame_C08.fm Page 200 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models
201
The ladder iterative technique described in Chapter 10 requires that the
voltages at
Node m
to be a function of the voltages at
Node n
and the currents
at
Node m
. The required equation is
(8.7)
In Equations 8.5, 8.6, and 8.7, the matrices and represent
the linetoneutral voltages for an ungrounded wye connection, or the line
toground voltages for a grounded wye connection. For a delta connection the
voltage matrices represent equivalent linetoneutral voltages. The current
matrices represent the line currents regardless of the transformer winding
connection.
8.3 The Delta–Grounded Wye StepDown Connection
The delta–grounded wye stepdown connection is a popular connection that
is typically used in a distribution substation serving a fourwire wye feeder
system. Another application of the connection is to provide service to a load
that is primarily singlephase. Because of the wye connection, three single
phase circuits are available, thereby making it possible to balance the single
phase loading on the transformer bank.
Three singlephase transformers can be connected to a delta–grounded wye
in a standard thirtydegree stepdown connection as shown in Figure 8.2.
FIGURE 8.2
Standard delta–grounded wye connection with voltages.
VLN
abc
[ ] A
t
[ ] · VLN
ABC
[ ] B
t
[ ] · I
abc
[ ] – =
VLN
ABC
[ ] VLN
abc
[ ]
Zt
b
Zt
c
+ + +
_ _ _
Zt
a
V
CA
V
AB
V
BC
Vt
a
Vt
b
Vt
c
V
ab
V
bc
+
_
+
_
t
n
V
AB
V
CA
V
g
+
_ V
ag
X3c X2b
V
AN
1
1
V
BC
1
Vt
b
1
_
+
_
+
_
+
Vt
a
1
Vt
c
1
I
a
I
b
I
c
0812_frame_C08.fm Page 201 Friday, July 20, 2001 4:47 PM
202
Distribution System Modeling and Analysis
8.3.1 Voltages
The positive sequence phasor diagrams of the voltages in Figure 8.2 show
the relationships between the various positive sequence voltages. Care must
be taken to observe the polarity marks on the individual transformer wind
ings. In order to simplify the notation it is necessary to label the “ideal”
voltages with voltage polarity markings as shown in Figure 8.2. Observing
the polarity markings of the transformer windings, the voltage
Vt
a
will be
180 degrees out of phase with the voltage
V
CA
, and the voltage
Vt
b
will be
180 degrees out of phase with the voltage
V
AB
. Kirchhoff’s voltage law gives
the linetoline voltage between phases
a
and
b
as:
V
ab
=
Vt
a
–
Vt
b
(8.8)
The phasors of the positive sequence voltages of Equation 8.8 are shown in
Figure 8.2.
The magnitude changes between the voltages can be deﬁned in terms of
the actual winding turns ratio (
n
t
) or the transformer ratio (
a
t
). With reference
to Figure 8.2, these ratios are deﬁned as follows:
(8.9)
Applying Equation 8.9, the magnitude of the linetoline voltage relative to
the ideal transformer voltage is
(8.10)
The magnitude of the positive sequence equivalent linetoneutral voltage
on the highvoltage side is given by:
(8.11)
where
(8.12)
With reference to Figure 8.2, the linetoline voltages on the primary side of
the transformer connection as a function of the ideal secondary side voltages
n
t
VLL
Rated High Side
VLN
Rated Low Side
 =
VLL n
t
· Vt =
VLN
VLL
3

n
t
3
 · Vt a
t
Vt = = =
a
t
n
t
3

VLL
Rated High Side
3 · VLN
Rated Low Side
 = =
a
t
VLL
Rated High Side
VLL
Rated Low Side
 =
0812_frame_C08.fm Page 202 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models
203
is given by:
(8.13)
In condensed form, Equation 8.13 is
(8.14)
where (8.15)
Equation 8.14 gives the primary linetoline voltages at
Node n
as functions
of the ideal secondary voltages. However, what is needed is a relationship
between equivalent linetoneutral voltages at
Node n
and the ideal secondary
voltages. The question is how is the equivalent linetoneutral voltages deter
mined knowing the linetoline voltages? One approach is to apply the theory
of symmetrical components.
The known linetoline voltages are transformed to their sequence voltages
by:
(8.16)
where (8.17)
By deﬁnition, the zero sequence linetoline voltage is always zero. The rela
tionship between the positive and negative sequence linetoneutral and line
toline voltages is known. These relationships in matrix form are given by:
(8.18)
(8.19)
where (8.20)
V
AB
V
BC
V
CA
0 n
t
– 0
0 0 n
t
–
n
t
– 0 0
·
Vt
a
Vt
b
Vt
c
=
VLL
ABC
[ ] AV [ ] · Vt
abc
[ ] =
AV [ ]
0 n
t
– 0
0 0 n
t
–
n
t
– 0 0
=
VLL
012
[ ] A
s
[ ]
1 –
· VLL
ABC
[ ] =
A
s
[ ]
1 1 1
1 a
s
2
a
s
1 a
s
a
s
2
=
a
s
1.0/120 =
VLN
0
VLN
1
VLN
2
1 0 0
0 t
s
ء
0
0 0 t
s
·
VLL
0
VLL
1
VLL
2
=
VLN
012
[ ] T [ ] · VLL
012
[ ] =
t
s
1
3
/30 =
0812_frame_C08.fm Page 203 Friday, July 20, 2001 4:47 PM
204 Distribution System Modeling and Analysis
Since the zero sequence linetoline voltage is zero, the (1,1) term of the
matrix [T] can be any value. For the purposes here, the (1,1) term is chosen
to have a value of 1.0. Knowing the sequence linetoneutral voltages, the
equivalent linetoneutral voltages can be determined.
The equivalent linetoneutral voltages as functions of the sequence line
toneutral voltages are
(8.21)
Substitute Equation 8.19 into Equation 8.21:
(8.22)
Substitute Equation 8.16 into Equation 8.22:
(8.23)
where (8.24)
Equation 8.23 provides a method of computing the equivalent linetoneutral
voltages from knowledge of the linetoline voltages. This is an important
relationship that will be used in a variety of ways as other threephase trans
former connections are studied.
To continue, Equation 8.14 can now be substituted into Equation 8.23:
(8.25)
where (8.26)
Equation 8.25 deﬁnes the generalized [a] matrix for the delta–grounded wye
stepdown connection.
The ideal secondary voltages as functions of the secondary linetoground
voltages and the secondary line currents are
(8.27)
where (8.28)
VLN
ABC
[ ] A
s
[ ] · VLN
012
[ ] =
VLN
ABC
[ ] A
s
[ ] · T [ ] · VLL
012
[ ] =
VLN
ABC
[ ] W [ ] · VLL
ABC
[ ] =
W [ ] A
s
[ ] · T [ ] · A
s
[ ]
1 – 1
3
 ·
2 1 0
0 2 1
1 0 2
= =
VLN
ABC
[ ] W [ ] · AV [ ] · Vt
abc
[ ] a
t
[ ] · Vt
abc
[ ] = =
a
t
[ ] W [ ] · AV [ ]
n
t
–
3
 ·
0 2 1
1 0 2
2 1 0
= =
Vt
abc
[ ] VLG
abc
[ ] Zt
abc
[ ] · I
abc
[ ] + =
Zt
abc
[ ]
Zt
a
0 0
0 Zt
b
0
0 0 Zt
c
=
0812_frame_C08.fm Page 204 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 205
Notice in Equation 8.28 there is no restriction that the impedances of the
three transformers be equal.
Substitute Equation 8.27 into Equation 8.25:
(8.29)
where (8.30)
The generalized matrices [a] and [b] have now been deﬁned. The derivation
of the generalized matrices [A] and [B] begins with solving Equation 8.14
for the ideal secondary voltages:
(8.31)
The linetoline voltages as functions of the equivalent linetoneutral volt
ages are
(8.32)
where (8.33)
Substitute Equation 8.32 into Equation 8.31:
(8.34)
where (8.35)
Substitute Equation 8.27 into Equation 8.34:
(8.36)
VLN
ABC
[ ] a
t
[ ] · VLG
abc
[ ] Zt
abc
[ ] · I
abc
[ ] + ( ) =
VLN
ABC
[ ] a
t
[ ] · VLG
abc
[ ] b
t
[ ] · I
abc
[ ] + =
b
t
[ ] a
t
[ ] · Zt
abc
[ ]
n
t
–
3
 ·
0 2 · Zt
b
Zt
c
Zt
a
0 2 · Zt
c
2 · Zt
a
Zt
b
0
= =
Vt
abc
[ ] AV [ ]
1 –
· VLL
ABC
[ ] =
VLL
ABC
[ ] D [ ] · VLN
ABC
[ ] =
D [ ]
1 1 – 0
0 1 1 –
1 – 0 1
=
Vt
abc
[ ] AV [ ]
1 –
· D [ ] · VLN
ABC
[ ] A
t
[ ] · VLN
ABC
[ ] = =
A
t
[ ] AV [ ]
1 –
· D [ ]
1
n
t
 ·
1 0 1 –
1 – 1 0
0 1 – 1
= =
VLG
abc
[ ] Zt
abc
[ ] · I
abc
[ ] + A
t
[ ] · VLN
ABC
[ ] =
0812_frame_C08.fm Page 205 Friday, July 20, 2001 4:47 PM
206 Distribution System Modeling and Analysis
Rearrange Equation 8.36:
(8.37)
where (8.38)
Equations 8.29 and 8.37 are the generalized voltage equations for the step
down delta–grounded wye transformer. These equations are in exactly the
same form as those derived in earlier chapters for line segments and step
voltage regulators.
8.3.2 Currents
The 30degree connection speciﬁes that the positive sequence current enter
ing the H1 terminal will lead the positive sequence current leaving the X1
terminal by 30 degrees. Figure 8.3 shows the same connection as Figure 8.2, but
with the currents instead of the voltages displayed. As with the voltages,
the polarity marks on the transformer windings must be observed for the
currents. For example, in Figure 8.3 the current I
a
is entering the polarity
mark on the lowvoltage winding, so the current I
AC
on the highvoltage
FIGURE 8.3
Delta–grounded wye connection with currents.
VLG
abc
[ ] A
t
[ ] · VLN
ABC
[ ] B
t
[ ] · I
abc
[ ] – =
B
t
[ ] Zt
abc
[ ]
Zt
a
0 0
0 Zt
b
0
0 0 Zt
c
= =
Zt
b
Zt
c
g
Zt
a
t
n
I
AC
I
BA
I
CB
I
a
I
b
I
c
+
_
V
cg
V
bg
_
+
_
V
ag
I
A
I
B
I
C
I
a
I
b
I
c
I
AC
I
A
X3c
I
BA
1
I
CB
1
1
1
1
X1a
X2a
0812_frame_C08.fm Page 206 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 207
winding will be in phase with I
a
. This relationship is shown in the phasor
diagrams for positive sequence currents in Figure 8.3.
The line currents can be determined as functions of the delta currents by
applying Kirchhoff’s current law:
(8.39)
In condensed form, Equation 8.39 is
(8.40)
The matrix equation relating the delta primary currents to the secondary
line currents is given by:
(8.41)
(8.42)
Substitute Equation 8.42 into Equation 8.40:
(8.43)
where (8.44)
(8.45)
Equation 8.43 provides a direct method of computing the phase line currents
at Node n knowing the phase line currents at Node m. Again, this equation
I
A
I
B
I
C
1 1 – 0
0 1 1 –
1 – 0 1
·
I
AC
I
BA
I
CB
=
I
ABC
[ ] D [ ] · ID
ABC
[ ] =
I
AC
I
BA
I
CB
1
n
t
 ·
1 0 0
0 1 0
0 0 1
·
I
a
I
b
I
c
=
ID
ABC
[ ] AI [ ] · I
abc
[ ] =
I
ABC
[ ] D [ ] · AI [ ] · I
abc
[ ] c
t
[ ] · VLG
abc
[ ] d
t
[ ] · I
abc
[ ] + = =
d
t
[ ] D [ ] · AI [ ]
1
n
t
 ·
1 1 – 0
0 1 1 –
1 – 0 1
= =
c
t
[ ]
0 0 0
0 0 0
0 0 0
=
0812_frame_C08.fm Page 207 Friday, July 20, 2001 4:47 PM
208 Distribution System Modeling and Analysis
is in the same form as that previously derived for threephase line segments
and threephase stepvoltage regulators.
The equations derived in this section are for the stepdown connection. If
the transformer bank is a stepup, the connection between transformer wind
ings will be different, as will the deﬁnitions of the generalized matrices. The
procedure for the derivation of the generalized matrices for the stepup
connection will be the same as that developed in this section.
Example 8.1
In the Example System of Figure 8.4, an unbalanced constant impedance
load is being served at the end of a onemile section of a threephase line.
The line is being fed from a substation transformer rated 5000 kVA, 138 kV
delta–12.47 kV grounded wye with a perunit impedance of The
phase conductors of the line are 336,400 26/7 ACSR with a neutral conductor
4/0 ACSR. The conﬁguration and computation of the phase impedance matrix
is given in Example 4.1 From that example, the phase impedance matrix was
computed to be
The transformer impedance needs to be converted to perunit referenced to
the lowvoltage side of the transformer. The base impedance is
The transformer impedance referenced to the lowvoltage side is
FIGURE 8.4
Example system.
1
3
Equivalent
System
0.085/85.
Zline
abc
[ ]
0.4576 j1.0780 + 0.1560 j0.5017 + 0.1535 j0.3849 +
0.1560 j0.5017 + 0.4666 j1.0482 + 0.1580 j0.4236 +
0.1535 j0.3849 + 0.1580 j0.4236 + 0.4615 j1.0651 +
Ω/mile =
Z
base
12.47
2
· 1000
5000
 31.1 = =
Zt 0.085/85 ( ) · 31.3 0.2304 j2.6335 Ω + = =
0812_frame_C08.fm Page 208 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 209
The transformer phase impedance matrix is
The unbalanced constant impedance load is connected in grounded wye.
The load impedance matrix is speciﬁed to be
The unbalanced linetoline voltages at Node 1 serving the substation trans
former are given as:
(1) Determine the generalized matrices for the transformer.
From Equation 8.26:
Zt
abc
[ ]
0.2304 j2.6335 + 0 0
0 0.2304 j2.6335 + 0
0 0 0.2304 j2.6335 +
Ω =
Zload
abc
[ ]
12 j6 + 0 0
0 13 j4 + 0
0 0 14 j5 +
Ω =
VLL
ABC
[ ]
138,000/0
135,500/ 115.5 –
145,959/123.1
V =
The transformer turns ratio is n
t
kVLL
high
kVLN
low

138
12.47
3

 19.1678 = = =
The transformer ratio is a
t
kVLL
high
kVLL
low

138
12.47
 11.0666 = = =
a
t
[ ]
19.1678 –
3
 = ·
0 2 1
1 0 2
2 1 0
0 12.7786 – 6.3893 –
6.3893 – 0 12.7786 –
12.7786 – 6.3893 – 0
=
0812_frame_C08.fm Page 209 Friday, July 20, 2001 4:47 PM
210
Distribution System Modeling and Analysis
From Equation 8.30:
From Equation 8.44:
From Equation 8.35:
From Equation 8.38:
(2) Given the linetoline voltages at
Node 1
, determine the ideal trans
former voltages. From Equation 8.15:
b
t
[ ]
19.1678 –
3
 ·
0 2 · 0.2304 j2.6335 + ( ) 0.2304 j2.6335 + ( )
0.2304 j2.6335 + ( ) 0 2 · 0.2304 j2.6335 + ( )
2 · 0.2304 j2.6335 + ( ) 0.2304 j2.6335 + ( ) 0
=
b
t
[ ]
0 −2.9441 j33.6518 – −1.4721 j16.8259 –
−1.4721 j16.8259 – 0 −2.9441 j33.6518 –
−2.9441 j33.6518 – −1.4721 j16.8259 – 0
=
d
t
[ ]
1
19.1678
 ·
1 1 – 0
0 1 1 –
1 – 0 1
0.0522 0.0522 – 0
0 0.0522 0.0522 –
0.0522 – 0 0.0522
= =
A
t
[ ]
1
19.1678
 ·
1 0 1 –
1 – 1 0
0 1 – 1
0.0522 0 0.0522 –
0.0522 – 0.0522 0
0 0.0522 – 0.0522
= =
B
t
[ ] Zt
abc
[ ]
0.2304 j2.6335 + 0 0
0 0.2304 j2.6335 + 0
0 0 0.2304 j2.6335 +
= =
AV [ ]
0 n
t
– 0
0 0 n
t
–
n
t
– 0 0
0 19.1678 – 0
0 0 19.1678 –
19.1678 – 0 0
= =
Vt
abc
[ ] AV [ ]
1 –
= · VLL
ABC
[ ]
7614.8/ 56.9 –
7199.6/180
7069/64.5
= V
0812_frame_C08.fm Page 210 Monday, October 28, 2002 10:47 AM
ThreePhase Transformer Models 211
(3) Determine the load currents.
Kirchhoff’s voltage law gives:
The line currents can now be computed:
(4) Determine the linetoground voltages at the load and at Node 2.
(5) Using the generalized matrices, determine the equivalent lineto
neutral voltages and the linetoline voltages at Node 1.
Vt
abc
[ ] Zt
abc
[ ] Zline
abc
[ ] Zload
abc
[ ] + + ( ) · I
abc
[ ] Ztotal
abc
[ ] · I
abc
[ ] = =
Ztotal
abc
[ ]
12.688 j9.7115 + 0.156 j0.5017 + 0.1535 j0.3849 +
0.156 j0.5017 + 13.697 j7.6817 + 0.158 j0.4236 +
0.1535 j0.3849 + 0.158 j0.4236 + 14.6919 j8.6986 +
Ω =
I
abc
[ ] Ztotal
abc
[ ]
1 –
· Vt
abc
[ ]
484.1/ 93.0 –
470.7/151.5
425.4/34.8
A = =
Vload
abc
[ ] Zload
abc
[ ] · I
abc
[ ]
6494.8/ 66.4 –
6401.6/168.6
6323.5/54.4
V = =
VLG
abc
[ ] Vload
abc
[ ] Zline
abc
[ ] · I
abc
[ ] +
6842.2/ 65.0 –
6594.5/171.0
6594.9/56.3
V = =
VLN
ABC
[ ] a
t
[ ] · VLG
abc
[ ] b
t
[ ] · I
abc
[ ] +
83,224/ 29.3 –
77,103/ 148.1 –
81,843/95.0
V = =
VLL
ABC
[ ] D [ ] · VLN
ABC
[ ]
138,000/0
135,500/ –115.5
145,959/123.1
V = =
0812_frame_C08.fm Page 211 Friday, July 20, 2001 4:47 PM
212 Distribution System Modeling and Analysis
It is always comforting to be able to work back and compute what
was initially given. In this case, the linetoline voltages at Node 1
have been computed and the same values result that were given
at the start of the problem.
(6) Use the reverse equation to verify that the linetoground voltages
at Node 2 can be computed knowing the equivalent linetoneutral
voltages at Node 1 and the currents leaving Node 2.
These are the same values for the linetoground voltages at Node
2 that were determined working from the load towards the source.
Example 8.1 has demonstrated the application of the generalized constants.
The example also provides veriﬁcation that the same voltages and currents
result when working from the load toward the source or from the source
toward the load. This will become very important for the iterative technique
to be developed in the Chapter 10.
8.4 The Ungrounded Wye–Delta StepDown Connection
Three singlephase transformers can be connected in a wye–delta connection.
The neutral of the wye can be grounded or ungrounded. The grounded wye
connection is rarely used because:
• The grounded wye provides a path for zero sequence currents for
linetoground faults upstream from the transformer bank. This
causes the transformers to be susceptible to burnouts on the
upstream faults.
• If one phase of the primary circuit is opened, the transformer bank
will continue to provide threephase service by operating as an open
wye–open delta bank. However, the two remaining transformers
may be subject to an overload condition leading to burnout.
The most common connection is the ungrounded wye–delta. This connection
is typically used to provide service to a combination singlephase “lighting”
load and a threephase “power” load such as an induction motor. The gener
alized constants for only the ungrounded wye–delta transformer connection
VLG
abc
[ ] A
t
[ ] · VLN
ABC
[ ] B
t
[ ] · I
abc
[ ] –
6842.2/ 65.0 –
6594.5/171.0
6494.9/56.3
V = =
0812_frame_C08.fm Page 212 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 213
will be developed following the same procedure as was used for the
delta–grounded wye.
Three singlephase transformers can be connected in an ungroundedwye
standard 30degree connection as shown in Figure 8.5. The voltage phasor
diagrams in Figure 8.5 illustrate that the high side positive sequence line
toline voltage leads the low side positive sequence linetoline voltage by
30 degrees. Also, the same phase shift occurs between the high side lineto
neutral voltage and the low side equivalent linetoneutral voltage. The
negative sequence phase shift is such that the high side negative sequence
voltage will lag the low side negative sequence voltage by 30 degrees.
The positive sequence current phasor diagrams for the connection in
Figure 8.5 are shown in Figure 8.6. Figure 8.6 illustrates that the positive
FIGURE 8.5
Standard ungrounded wye–delta connection.
FIGURE 8.6
Positive sequence current phasors.
V
AN
V
BN
V
CN
Vt
ab
Vt
bc
Vt
ca
t
n
_
+
_
+
_
+
+
_
+
_
+
_
V
CN
1
V
BN
1
V
AN
1
V
ca
1
V
ab
1
V
bc
1
V
an
1
V
AB
1
+
_
X3c
+
_
X2b
_
V
ab
+
V
bc
I
b
I
c
I
a
I
ba I
cb
I
ac
I
A
I
B
V
+
AB
+
V
BC
I
C
Zt
ab
Zt
bc
Zt
ca
I
A
1
I
ac
1
I
cb
1
I
ba
1
0812_frame_C08.fm Page 213 Friday, July 20, 2001 4:47 PM
214 Distribution System Modeling and Analysis
sequence line current on the high side of the transformer (Node n) leads the
low side line current (Node m) by 30 degrees. It can also be shown that the
negative sequence high side line current will lag the negative sequence low
side line current by 30 degrees.
The deﬁnition for the “turns ratio n
t
” will be the same as Equation 8.9, with
the exception that the numerator will be the linetoneutral voltage and the
denominator will be the linetoline voltage. The “transformer ratio a
t
,” as
given in Equation 8.12, will apply for this connection. It should be noted in
Figure 8.5 that the ideal low side transformer voltages for this connection will
be linetoline voltages. Also, the ideal low side currents are the currents
ﬂowing inside the delta.
The basic ideal transformer voltage and current equations as functions of
the turns ratio are
(8.46)
where
(8.47)
(8.48)
(8.49)
Solving Equation 8.47 for the ideal delta transformer voltages:
(8.50)
The linetoline voltages at Node m as functions of the ideal transformer
voltages and the delta currents are given by:
(8.51)
(8.52)
V
AN
V
BN
V
CN
n
t
0 0
0 n
t
0
0 0 n
t
·
Vt
ab
Vt
bc
Vt
ca
=
n
t
VLN
Rated High Side
VLL
Rated Low Side
 =
VLN
ABC
[ ] AV [ ] · Vt
abc
[ ] =
I
ba
I
cb
I
ac
n
t
0 0
0 n
t
0
0 0 n
t
·
I
A
I
B
I
C
=
ID
abc
[ ] AI [ ] · I
ABC
[ ] =
Vt
abc
[ ] AV [ ]
1 –
· VLN
ABC
[ ] =
V
ab
V
bc
V
ca
Vt
ab
Vt
bc
Vt
ca
Zt
ab
0 0
0 Zt
ac
0
0 0 Zt
ca
·
ID
ba
ID
cb
ID
ac
– =
VLL
abc
[ ] Vt
abc
[ ] Zt
abc
[ ] · ID
abc
[ ] – =
0812_frame_C08.fm Page 214 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models
215
Substitute Equations 8.49 and 8.50 into Equation 8.52:
(8.53)
where (8.54)
The line currents on the delta side of the transformer bank as functions of
the wye transformer currents are given by:
(8.55)
where (8.56)
Substitute Equation 8.49 into Equation 8.55:
(8.57)
where (8.58)
Because the matrix [
DY
] is singular, it is not possible to use Equation 8.57
to develop an equation relating the wye side line currents at
Node n
to the
delta side line currents at
Node m
. In order to develop the necessary matrix
equation, three independent equations must be written. Two independent
KCL equations at the vertices of the delta can be used. Because there is no
path for the high side currents to ﬂow to ground, they must sum to zero
and, therefore, so must the delta currents sum to zero. This provides the
third independent equation. The resulting three independent equations in
matrix form are given by:
(8.59)
VLL
abc
[ ] AV [ ]
1 –
· VLN
ABC
[ ] ZNt
abc
[ ] · I
ABC
[ ] – =
ZNt
abc
[ ] Zt
abc
[ ] · AI [ ]
n
t
· Zt
ab
0 0
0 n
t
· Zt
bc
0
0 0 n
t
· Zt
ca
= =
I
abc
[ ] DI [ ] · ID
abc
[ ] =
DI [ ]
1 0 1 –
1 – 1 0
0 1 – 1
=
I
abc
[ ] DI [ ] · AI [ ] · I
ABC
[ ] DY [ ] · I
ABC
[ ] = =
DY [ ] DI [ ] · AI [ ]
n
t
0 n
t
–
n
t
– n
t
0
0 n
t
– n
t
= =
I
a
I
b
0
1 0 1 –
1 – 1 0
1 1 1
·
I
ba
I
cb
I
ac
=
0812_frame_C08.fm Page 215 Monday, October 28, 2002 10:47 AM
216 Distribution System Modeling and Analysis
Solving Equation 8.59 for the delta currents:
(8.60)
(8.61)
Equation 8.61 can be modiﬁed to include the Phase c current by setting the
third column of the [L0] matrix to zero:
(8.62)
(8.63)
Solve Equation 8.49 for and substitute into Equation 7.63:
(8.64)
where (8.65)
Equation 8.65 deﬁnes the generalized constant matrix [d
t
] for the
ungrounded wye–delta transformer connection. In the process of the deri
vation, a very convenient equation (8.62) evolved that can be used anytime
the currents in a delta need to be determined when the line currents are
known. However, it must be understood that this equation will only work
when the delta currents sum to zero, which means an ungrounded neutral
on the primary. The generalized matrices [a
t
] and [b
t
] can now be developed.
Solve Equation 8.53 for [VLN
ABC
]:
(8.66)
I
ba
I
cb
I
ac
1 0 1 –
1 – 1 0
1 1 1
1 –
·
I
a
I
b
0
1
3
 ·
1 1 – 1
1 2 1
2 – 1 – 1
·
I
a
I
b
0
= =
ID
abc
[ ] L0 [ ] · I
ab0
[ ] =
I
ba
I
cb
I
ac
1
3
 ·
1 1 – 0
1 2 0
2 – 1 – 0
·
I
a
I
b
I
c
=
ID
abc
[ ] L [ ] · I
abc
[ ] =
I
ABC
[ ]
I
ABC
[ ] AI [ ]
1 –
· L [ ] · I
abc
[ ] d
t
[ ] · I
abc
[ ] = =
d
t
[ ] AI [ ]
1 –
· L [ ]
1
3 · n
T
 ·
1 1 – 0
1 2 0
2 – 1 – 0
= =
VLN
ABC
[ ] AV [ ] · VLL
abc
[ ] AV [ ] · ZNt
abc
[ ] · I
ABC
[ ] + =
0812_frame_C08.fm Page 216 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 217
Substitute Equation 8.64 into Equation 8.66:
(8.67)
where (8.68)
(8.69)
The generalized constant matrices have been developed for computing volt
ages and currents from the load toward the source. The reverse generalized
matrices can be developed by referring back to Equation 8.53, which is
repeated as Equation 8.70 for convenience:
(8.70)
Equation 8.23 is used to compute the equivalent linetoneutral voltages as
functions of the linetoline voltages:
(8.71)
Substitute Equations 8.70 and 8.64 into Equation 8.71:
(8.72)
where (8.73)
(8.74)
VLN
ABC
[ ] AV [ ] · VLL
abc
[ ] AV [ ] · ZNt
abc
[ ] · d
t
[ ] · I
abc
[ ] + =
but VLL
abc
[ ] D [ ] · VLN
abc
[ ] =
VLN
ABC
[ ] AV [ ] · D [ ] · VLN
abc
[ ] AV [ ] · ZNt
abc
[ ] · d
t
[ ] · I
abc
[ ] + =
VLN
ABC
[ ] a
t
[ ] · VLN
abc
[ ] b
t
[ ] · I
abc
[ ] + =
a
t
[ ] AV [ ] · D [ ] n
t
·
1 1 – 0
0 1 1 –
1 – 0 1
= =
b
t
[ ] AV [ ] · ZNt
abc
[ ] · d
t
[ ]
n
t
3
 ·
Zt
ab
Zt
ab
– 0
Zt
bc
2 · Zt
bc
0
2 · Zt
ca
– Zt
ca
– 0
= =
VLL
abc
[ ] AV [ ]
1 –
· VLN
ABC
[ ] ZNt
abc
[ ] · I
ABC
[ ] – =
VLN
abc
[ ] W [ ] · VLL
abc
[ ] =
VLN
abc
[ ] W [ ] · AV [ ]
1 –
· VLN
ABC
[ ] W [ ] · ZNt
abc
[ ] · d
t
[ ] · I
abc
[ ] – =
VLN
abc
[ ] A
t
[ ] · VLN
ABC
[ ] B
t
[ ] · I
abc
[ ] – =
A
t
[ ] W [ ] · AV [ ]
1 – 1
3 · n
t
 ·
2 1 0
0 2 1
1 0 2
= =
B
t
[ ] W [ ] · ZNt
abc
[ ] · d
t
[ ]
1
9
 ·
2 · Zt
ab
Zt
bc
+ 2 · Zt
bc
2 · Zt
ab
– 0
2 · Zt
bc
2 · Zt
ca
– 4 · Zt
bc
Zt
ca
– 0
Zt
ab
4 · Zt
ca
– Zt
ab
– 2 · Zt
ca
– 0
= =
0812_frame_C08.fm Page 217 Friday, July 20, 2001 4:47 PM
218 Distribution System Modeling and Analysis
The generalized matrices have been developed for the ungrounded
wye–delta transformer connection. The derivation has applied basic circuit
theory and the basic theories of transformers. The end result of the deriva
tions is to provide an easy method of analyzing the operating characteristics
of the transformer connection. Example 8.2 will demonstrate the application
of the generalized matrices for this transformer connection.
Example 8.2
Figure 8.7 shows three singlephase transformers in an ungrounded wye
delta connection serving a combination singlephase and threephase load
in a delta connection. The voltages at the load are balanced threephase of
240 V linetoline. The net loading by phase is
S
ab
= 100 kVA at 0.9 lagging power factor
S
bc
= S
ca
= 50 kVA at 0.8 lagging power factor
In sizing the three transformers it will be assumed that the lighting trans
former serves all of the singlephase load and a third of the threephase load,
while the two power transformers will each serve a third of the threephase
load. With those assumptions, the selected ratings of the transformers are
Lighting: 100 kVA, 7200240 V, Z = 0.01 + j0.04 perunit
Power: 50 kVA, 7200240 V, Z = 0.015 + j0.035 perunit
Determine the following:
1. The currents in the load
2. The secondary line currents
3. The equivalent linetoneutral secondary voltages
4. The primary linetoneutral and linetoline voltages
5. The primary line currents
FIGURE 8.7
Ungrounded wye–delta serving an unbalanced load.
+
+
+
+
+
+
+
+
+
+
+




 




VAG
VBG
IB
IN
VBN
VCN
VAN
VCG
IC
Iba
Icb
Iac
Ic
Ib
Vca
Vbc
Vtca
Vtbc
Vtab
Ia
Ica
Ibc
Sca
Sbc
Sab
Z
t
c
a
Z
t
b
c
Z
t
a
b
0812_frame_C08.fm Page 218 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 219
Before the analysis can start, the transformer impedances must be converted
to actual values in ohms and located inside the deltaconnected secondary
windings.
Lighting transformer:
Power transformers:
The transformer impedance matrix can now be deﬁned:
The turns ratio of the transformers is:
Deﬁne all of the matrices:
Z
base
0.24
2
· 1000
100
 0.576 = =
Zt
ab
0.01 j0.4 + ( ) · 0.576 0.0058 j 0.23 Ω + = =
Z
base
0.24
2
· 1000
50
 1.152 = =
Zt
bc
Zt
ca
0.015 j0.35 + ( ) · 1.152 0.0173 j0.0403 Ω + = = =
Zt
abc
[ ]
0.0058 j0.023 + 0 0
0 0.017 j0.0403 + 0
0 0 0.0173 j0.0403 +
= Ω
n
t
7200
240
 30 = =
W [ ]
1
3
 ·
2 1 0
0 2 1
1 0 2
D [ ]
1 1 – 0
0 1 1 –
1 – 0 1
DI [ ]
1 0 1 –
1 – 1 0
0 1 – 1
= = =
a
t
[ ] n
t
·
1 1 – 0
0 1 1 –
1 – 0 1
30 30 – 0
0 30 30 –
30 – 0 30
= =
b
t
[ ]
n
t
3
 ·
Zt
ab
Zt
ab
– 0
Zt
bc
2 · Zt
bc
0
2 · Zt
ca
– Zt
ca
– 0
0.576 j0.2304 + 0.576 – j0.2304 – 0
0.1728 j0.4032 + 0.3456 j0.8064 + 0
0.3456 j0.8064 + – 0.1728 – j0.4032 –
= =
c
t
[ ]
0 0 0
0 0 0
0 0 0
=
0812_frame_C08.fm Page 219 Friday, July 20, 2001 4:47 PM
220 Distribution System Modeling and Analysis
Deﬁne the linetoline load voltages:
Deﬁne the loads:
Calculate the delta load currents:
d
t
[ ]
1
3 · n
T
 ·
1 1 – 0
1 2 0
2 – 1 – 0
0.0111 0.0111 – 0
0.0111 0.0222 0
0.0222 – 0.0111 –
= =
A
t
[ ]
1
3 · n
T
 ·
2 1 0
0 2 1
1 0 2
0.0222 0.0111 0
0 0.0222 0.0111
0.0111 0 0.0222
= =
B
t
[ ]
1
9
 ·
2 · Zt
ab
Zt
bc
+ 2 · Zt
bc
2 · Zt
ab
– 0
2 · Zt
bc
2 · Zt
ca
– 4 · Zt
bc
Zt
ca
– 0
Zt
ab
4 · Zt
ca
– Zt
ab
2 · Zt
ca
– – 0
=
B
t
[ ]
0.0032 j0.0096 + 0.0026 j0.0038 + 0
0 0.0058 j0.0134 + 0
−0.007 j0.0154 – −0.0045 j0.0115 – 0
=
VLL
abc
[ ]
240/0
240/ 120 –
240/120
V =
SD
abc
[ ]
100/cos
1 –
0.9 ( )
50/cos
1 –
0.8 ( )
50/cos
1 –
0.8 ( )
90 j43.589 +
40 j30 +
40 j30 +
kVA = =
ID
i
SD
i
· 1000
VLL
abc
i

∗
A =
ID
abc
[ ]
I
ab
I
bc
I
ca
416.7/ 25.84 –
208.3/ 156.87 –
208.3/83.13
A = =
0812_frame_C08.fm Page 220 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 221
Compute the secondary line currents:
Compute the equivalent secondary linetoneutral voltages:
Use the generalized constant matrices to compute the primary linetoneutral
voltages and linetoline voltages:
The high primary line currents are
It is interesting to compute the operating kVA of the three transformers.
Taking the product of the transformer voltage times the conjugate of the
current gives the operating kVA of each transformer:
I
abc
[ ] DI [ ] · ID
abc
[ ]
522.9/ 47.97 –
575.3/ 119.06 –
360.8/53.13
A = =
VLN
abc
[ ] W [ ] · VLL
abc
[ ]
138.56/ 30 –
138.56/ 150 –
138.56/90
A = =
VLN
ABC
[ ] a
t
[ ] · VLN
abc
[ ] b
t
[ ] · I
abc
[ ] +
7367.6/1.4
7532.3/ 119.1 –
7406.2/121.7
V = =
VLL
ABC
[ ]
D [ ] · VLN
ABC
[ ]
1000

12.94/31.54
12.88/ 88.95 –
12.81/151.50
kV = =
I
ABC
[ ] d
t
[ ] · I
abc
[ ]
11.54/ 28.04 –
8.95/ 166.43 –
7.68/101.16
A = =
ST
i
VLN
ABC
i
· I
ABC
i
( )
ء
1000

85.02/29.46
67.42/47.37
56.80/20.58
kVA = =
0812_frame_C08.fm Page 221 Friday, July 20, 2001 4:47 PM
222 Distribution System Modeling and Analysis
The operating power factors of the three transformers are
Note that the operating kVAs do not match very closely the rated kVAs
of the three transformers. In particular, the lighting transformer did not serve
the total load of 100 kVA that is directly connected its terminals. The lighting
transformer is operating below rated kVA while the two power transformers
are overloaded. In fact, the transformer connected to Phase B is operating
35% above rated kVA. Because of this overload, the ratings of the three
transformers should be changed so that the two power transformers are
rated 75 kVA. Finally, the operating power factors of the three transformers
bear little resemblance to the load power factors.
Example 8.2 demonstrates how the generalized constant matrices can be
used to determine the operating characteristics of the transformers. In addi
tion, the example shows that the obvious selection of transformer ratings
will lead to an overload condition on the two power transformers. The
beauty in this is that if the generalized constant matrices have been applied
in a computer program, it is a simple task to change the transformer kVA
ratings and be assured that none of the transformers will be operating in an
overload condition.
8.5 The Grounded Wye–Grounded Wye Connection
The grounded wye–grounded wye connection is primarily used to supply
singlephase and threephase loads on fourwire multigrounded systems.
The grounded wye–grounded wye connection is shown in Figure 8.8. Unlike
the delta–wye and wye–delta connections, there is no phase shift between
the voltages and the currents on the two sides of the bank. This makes the
derivation of the generalized constant matrices much easier.
With reference to Figure 8.8, the ideal transformer voltages on the second
ary windings can be computed by:
(8.75)
(8.76)
PF [ ]
cos(29.46)
cos(47.37)
cos(20.58)
87.1
67.7
93.6
% lagging = =
Vt
a
Vt
b
Vt
c
V
ag
V
bg
V
cg
Zt
a
0 0
0 Zt
b
0
0 0 Zt
c
·
I
a
I
b
I
c
+ =
Vt
abc
[ ] VLG
abc
[ ] Zt
abc
[ ] + · I
abc
[ ] =
0812_frame_C08.fm Page 222 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models
223
The linetoground voltages on the primary side are related to the ideal
transformer voltages by the turns ratio:
(8.77)
(8.78)
Substitute Equation 8.76 into Equation 8.78:
(8.79)
Equation 8.79 is in the generalized form with the [
a
] and [
b
] matrices deﬁned
by:
(8.80)
(8.81)
FIGURE 8.8
Grounded wye–grounded wye connection.
Zt
b
Zt
c
+ + +
_ _ _
Zt
a
V
AG
V
BG
V
CG
Vt
a
Vt
b
Vt
c
+ +
t
n
_
+
_
+
_
+
I
a
I
b
I
c
I
A
I
B
I
C
X2b
X3c V
cg
_ _ _
V
bg
+
V
ag
V
AG
V
BG
V
CG
n
t
0 0
0 n
t
0
0 0 n
t
·
Vt
a
Vt
b
Vt
c
=
VLG
ABC
[ ] AV [ ] · Vt
abc
[ ] =
VLG
ABC
[ ] AV [ ] · VLG
abc
[ ] AV [ ] · Zt
abc
[ ] · I
abc
[ ] + =
a
t
[ ] AV [ ]
n
t
0 0
0 n
t
0
0 0 n
t
= =
b
t
[ ] AV [ ] · Zt
abc
[ ]
n
t
· Zt
a
0 0
0 n
t
· Zt
b
0
0 0 n
t
· Zt
c
= =
0812_frame_C08.fm Page 223 Monday, October 28, 2002 10:48 AM
224 Distribution System Modeling and Analysis
The primary line currents as a function of the secondary line currents are
given by:
(8.82)
where (8.83)
The reverse equation is determined solving Equation 8.79 for the secondary
linetoground voltages:
(8.84)
Constant matrices for the reverse equation are given by:
(8.85)
(8.86)
The modeling and analysis of the grounded wye–grounded wye connection
does not present any problems. Without the phase shift and a closed delta
connection, there is a direct relationship between the primary and secondary
voltages and currents, as has been demonstrated in the derivation of the
generalized constant matrices.
8.6 The Delta–Delta Connection
The delta–delta connection is primarily used on threewire delta systems to
provide service to a threephase load or a combination of threephase and
singlephase loads. Three singlephase transformers connected in a delta–delta
I
ABC
[ ] d
t
[ ] · I
abc
[ ] =
d
t
[ ]
1
n
t
 0 0
0
1
n
t
 0
0 0
1
n
t

=
VLG
abc
[ ] AV [ ]
1 –
· VLG
ABC
[ ] Zt
abc
[ ] – · I
abc
[ ] =
A
t
[ ] AV [ ]
1 –
1
n
t
 0 0
0
1
n
t
 0
0 0
1
n
t

= =
B
t
[ ]
Zt
a
0 0
0 Zt
b
0
0 0 Zt
c
=
0812_frame_C08.fm Page 224 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 225
are shown in Figure 8.9. The basic ideal transformer voltage and current
equations as functions of the turns ratio are
(8.87)
(8.88)
(8.89)
(8.90)
where (8.91)
Solve Equation 8.90 for the source side delta currents:
(8.92)
FIGURE 8.9
Delta–delta connection.
Vt
ab
Vt
bc
Vt
ca
t
n
_
+
_
+
_
+
+
_
X3c
+
_
V
X2b
_
V
ab
+
V
bc
I
b
I
c
I
a
I
ba I
cb
I
ac
Zt
ab
Zt
bc
Zt
ca
I
A
I
B
I
C
_
+
V
AB
_
+
V
BC
+
_
V
CA
BC
I
AB
I
CA
I
VLL
AB
VLL
BC
VLL
CA
n
t
0 0
0 n
t
0
0 0 n
t
·
Vt
ab
Vt
bc
Vt
ca
=
VLL
ABC
[ ] AV [ ] · Vt
abc
[ ] =
I
ba
I
cb
I
ca
n
t
0 0
0 n
t
0
0 0 n
t
·
I
AB
I
BC
I
CA
=
ID
abc
[ ] AI [ ] · ID
ABC
[ ] =
n
t
VLL
Rated High Side
VLL
Rated Low Side
 =
ID
ABC
[ ] AI [ ]
1 –
· ID
abc
[ ] =
0812_frame_C08.fm Page 225 Friday, July 20, 2001 4:47 PM
226 Distribution System Modeling and Analysis
The line currents as functions of the delta currents on the source side are
given by:
(8.93)
(8.94)
Substitute Equation 8.92 into Equation 8.94:
(8.95)
Since [AI] is a diagonal matrix, Equation 8.95 can be rewritten as:
(8.96)
Determine the load side line currents as functions of the load side delta
currents:
(8.97)
(8.98)
Applying Equation 8.98, Equation 8.96 becomes:
(8.99)
Turn Equation 8.99 around to solve for the load side line currents as functions
of the source side line currents:
(8.100)
Equations 8.99 and 8.100 merely demonstrate that the line currents on the
two sides of the transformer are in phase and differ only by the turns ratio
of the transformer windings. In the perunit system, the perunit line currents
on the two sides of the transformer are exactly equal.
I
A
I
B
I
C
1 0 1 –
1 – 1 0
0 1 – 1
·
I
AB
I
BC
I
CA
=
I
ABC
[ ] DI [ ] · ID
ABC
[ ] =
I
ABC
[ ] DI [ ] · AI [ ]
1 –
· ID
abc
[ ] =
I
ABC
[ ] AI [ ]
1 –
· DI [ ] · ID
abc
[ ] =
I
a
I
b
I
c
1 0 1 –
1 – 1 0
0 1 – 1
·
I
ba
I
cb
I
ac
=
I
abc
[ ] DI [ ] · ID
abc
[ ] =
I
ABC
[ ] AI [ ]
1 –
· I
abc
[ ] =
I
abc
[ ] AI [ ] · I
ABC
[ ] =
0812_frame_C08.fm Page 226 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 227
The ideal delta voltages on the load side as functions of the linetoline
voltages, the delta currents, and the transformer impedances are given by:
(8.101)
where (8.102)
Substitute Equation 8.101 into Equation 8.88:
(8.103)
Solve Equation 8.103 for the load side linetoline voltages:
(8.104)
The delta currents [ID
abc
] in Equations 8.103 and 8.104 need to be replaced
by the load side line currents [I
abc
]. In order to develop the needed relation
ship, three independent equations are needed. The ﬁrst two come from
applying KCL at two vertices of the deltaconnected secondary.
(8.105)
(8.106)
The third equation comes from recognizing that the sum of the primary line
toline voltages and, therefore, the secondary ideal transformer voltages must
sum to zero. KVL around the delta windings gives:
(8.107)
Replacing the ideal delta voltages with the source side linetoline voltages:
(8.108)
Since the sum of the linetoline voltages must equal zero (KVL) and the
turns ratio of the three transformers are equal, Equation 8.108 is simpliﬁed to:
(8.109)
Vt
abc
[ ] VLL
abc
[ ] Zt
abc
[ ] · ID
abc
[ ] + =
Zt
abc
[ ]
Zt
ab
0 0
0 Zt
bc
0
0 0 Zt
ca
=
VLL
ABC
[ ] AV [ ] · VLL
abc
[ ] AV [ ] · Zt
abc
[ ] · ID
abc
[ ] + =
VLL
abc
[ ] AV [ ]
1 –
· VLL
ABC
[ ] Zt
abc
[ ] · ID
abc
[ ] – =
I
a
I
ba
I
ac
– =
I
b
I
cb
I
ba
– =
Vt
ab
Zt
ab
· I
ba
Vt
bc
Zt
bc
· I
cb
– + – Vt
ca
Zt
ca
· I
ac
– + 0 =
V
AB
n
t

V
BC
n
t

V
CA
n
t
 + + Zt
ab
· I
ba
Zt
bc
· I
cb
Zt
ca
· I
ac
+ + =
0 Zt
a
· I
ba
Zt
b
· I
cb
Zt
c
· I
ac
+ + =
0812_frame_C08.fm Page 227 Friday, July 20, 2001 4:47 PM
228 Distribution System Modeling and Analysis
Note in Equation 8.109 that if the three transformer impedances are equal,
then the sum of the delta currents will add to zero, meaning that the zero
sequence delta currents will be zero.
Equations 8.105, 8.106, and 8.109 can be put into matrix form:
(8.110)
(8.111)
where (8.112)
(8.113)
Solve Equation 8.111 for the load side delta currents:
(8.114)
where
(8.115)
Writing Equation 8.114 in matrix form gives:
(8.116)
From Equati ons 8.115 and 8.116 it is seen that the delta currents are functions
of the transformer impedances and just the line currents in Phases a and b.
Equation 8.116 can be modiﬁed to include the line current in Phase c by
I
a
I
b
0
1 0 1 –
1 – 1 0
Zt
ab
Zt
bc
Zt
ca
·
I
ba
I
cb
I
ac
=
I0
abc
[ ] F [ ] · ID
abc
[ ] =
I0
abc
[ ]
I
a
I
b
0
=
F [ ]
1 0 1 –
1 – 1 0
Zt
ab
Zt
bc
Zt
ca
=
ID
abc
[ ] F [ ]
1 –
· I0
abc
[ ] G [ ] · I0
abc
[ ] = =
G [ ] F [ ]
1 – 1
Zt
ab
Zt
bc
Zt
ca
+ +
 ·
Zt
ca
Zt
bc
– 1
Zt
ca
Zt
ab
Zt
ca
+ 1
Zt
ab
– Zt
bc
– Zt
bc
– 1
= =
I
ba
I
cb
I
ac
G
11
G
12
G
13
G
21
G
22
G
23
G
31
G
32
G
33
·
I
a
I
b
0
=
0812_frame_C08.fm Page 228 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 229
setting the last column of the [G] matrix to zeros:
(8.117)
The ﬁnal form of Equation 8.117 can be written as:
(8.118)
where (8.119)
If the three transformers have equal impedances, Equation 8.117 becomes:
(8.120)
Note in Equation 8.120 that the sum of the delta currents will be zero, meaning
that there is no circulating zero sequence current in the delta windings.
It is a common practice to serve a combination of singlephase and three
phase loads from the delta–delta connection. The singlephase loads are
served by center tapping one of the three transformers (lighting transformer),
thereby giving the standard 120/240 V service. The kVA rating, and thus the
impedance of the lighting transformer, will be different from the two power
transformers. Therefore, there will be a circulating zero sequence in the delta
windings.
The general form of Equation 8.118 will be used to deﬁne the generalized
constant matrices for the delta–delta connection. Substitute Equation 8.118
into Equation 8.103:
(8.121)
The generalized matrices are deﬁned in terms of the linetoneutral voltages
on the two sides of the transformer bank. Equation 8.121 is modiﬁed to be
in terms of equivalent linetoneutral voltages:
(8.122)
I
ba
I
cb
I
ac
G
11
G
12
0
G
21
G
22
0
G
31
G
32
0
·
I
a
I
b
I
c
=
ID
abc
[ ] G1 [ ] · I
abc
[ ] =
G1 [ ]
1
Zt
ab
Zt
bc
Zt
ca
+ +
 ·
Zt
ca
Zt
bc
– 0
Zt
ca
Zt
ab
Zt
ca
+ 0
−Zt
ab
Zt
bc
– Zt
bc
– 0
=
I
ba
I
cb
I
ac
1
3

1 1 – 0
1 2 0
2 – 1 – 0
I
a
I
b
I
c
⋅ ⋅
1
3

I
a
I
b
–
I
a
2 I
b
⋅ +
2 – I
a
I
b
– ⋅
⋅ = =
VLL
ABC
[ ] AV [ ] VLL
abc
[ ] AV [ ] Zt
abc
[ ] G1 [ ] I
abc
[ ] ⋅ ⋅ + ⋅ =
VLN
ABC
[ ] W [ ] VLL
ABC
[ ] ⋅ =
VLN
ABC
[ ] W [ ] AV [ ] D [ ] VLN
abc
[ ] W [ ] AV [ ] Zt
abc
[ ] G1 [ ] I
abc
[ ] ⋅ ⋅ ⋅ ⋅ + ⋅ ⋅ =
0812_frame_C08.fm Page 229 Friday, July 20, 2001 4:47 PM
230 Distribution System Modeling and Analysis
Equation 8.122 is in the general form:
(8.123)
where (8.124)
(8.125)
Equation 8.125 does not reduce to a simple form, so the quadruple matrix
multiplication must be used to compute the matrix [b
t
].
Equation 8.99 gives the generalized equation for currents:
(8.126)
where (8.127)
The reverse generalized equation can be derived by modifying Equation
8.104 in terms of equivalent linetoneutral voltages:
(8.128)
The generalized equation is
(8.129)
where (8.130)
(8.131)
VLN
ABC
[ ] a
t
[ ] VLN
abc
[ ] ⋅ b
t
[ ] I
abc
[ ] ⋅ + =
a
t
[ ] W [ ] AV [ ] D [ ] ⋅ ⋅
n
t
3

2 1 – 1 –
1 – 2 1 –
1 – 1 – 2
⋅ = =
b
t
[ ] W [ ] AV [ ] Zt
abc
[ ] G1 [ ] ⋅ ⋅ ⋅ =
I
ABC
[ ] AI [ ]
1 –
I
abc
[ ] ⋅ d
t
[ ] I
abc
[ ] ⋅ = =
d
t
[ ]
1
n
t
 0 0
0
1
n
t
 0
0 0
1
n
t

=
VLN
abc
[ ] W [ ] VLL
abc
[ ] ⋅ =
VLN
abc
[ ] W [ ] AV [ ]
1 –
D [ ] VLN
ABC
[ ] ⋅ ⋅ ⋅ W [ ] Zt
abc
[ ] G1 [ ] I
abc
[ ] ⋅ ⋅ ⋅ – =
VLN
abc
[ ] A
t
[ ] VLN
ABC
[ ] ⋅ B
t
[ ] I
abc
[ ] ⋅ – =
A
t
[ ] W [ ] AV [ ]
1 –
D [ ] ⋅ ⋅
1
3 n
t
⋅

2 1 – 1 –
1 – 2 1 –
1 – 2 1 –
⋅ = =
B
t
[ ] W [ ] Zt
abc
[ ] G1 [ ] ⋅ ⋅ =
0812_frame_C08.fm Page 230 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 231
The generalized matrices for the delta–delta connection have been derived.
Once again it has been a long process to get to the ﬁnal six equations that
deﬁne the matrices. The derivation provides an excellent exercise in the
application of basic transformer theory and circuit theory. Once the matrices
have been deﬁned for a particular transformer connection the analysis of the
connection is a relatively simple task. Example 8.3 will demonstrate the
analysis of this connection using the generalized matrices.
Example 8.3
Figure 8.10 shows three singlephase transformers in a delta–delta connection
serving a combination singlephase and threephase load connected in delta.
The voltages at the load are balanced threephase of 240 V linetoline:
The net loading by phase is
S
ab
= 100 kVA at 0.9 lagging power factor
S
bc
= S
ca
= 50 kVA at 0.8 lagging power factor.
As in Example 8.2, it will be assumed that the lighting transformer serves
all of the singlephase load and a third of the threephase load, while the
two power transformers will each serve a third of the threephase load. With
those assumptions, the selected ratings of the transformers are
Lighting: 100 kVA, 12,470240 V, Z = 0.01 + j0.04 perunit
Power: 50 kVA, 12,470240 V, Z = 0.015 + j0.035 perunit
FIGURE 8.10
Delta–delta bank serving and unbalanced deltaconnected load.
VLL
abc
[ ]
240/0
240/ 120 –
240/120
V =
+
+
+
+
+
+ +
+
+









IA
IAB
IB
IC
IBC
ICA
VCA
VAB
VBC
Iac
Vca
Ic
Iba
Vtab
Vtca
Vtbc
Ztca
Ztbc
Ztab
Icb
Ib
Sca
Vbc
Sbc
Sab
0812_frame_C08.fm Page 231 Friday, July 20, 2001 4:47 PM
232
Distribution System Modeling and Analysis
Determine the following:
1. The currents in the load
2. The secondary line currents
3. The equivalent linetoneutral secondary voltages
4. The primary linetoneutral and linetoline voltages
5. The primary line currents
6. The delta currents in the primary and secondary windings
Before the analysis can start, the transformer impedances must be converted
to actual values in ohms and located inside the deltaconnected secondary
windings:
Lighting transformer:
Power transformers:
The transformer impedance matrix can now be deﬁned:
The turns ratio of the transformers is
Deﬁne all of the matrices:
Z
base
0.24
2
1000 ⋅
100
 0.576 Ω = =
Zt
ab
0.01 j0.4 + ( ) 0.576 ⋅ 0.0058 j0.023 Ω + = =
Z
base
0.24
2
1000 ⋅
50
 1.152 Ω = =
Zt
bc
Zt
ca
0.015 j0.35 + ( ) 1.152 ⋅ 0.0173 j0.0403 Ω + = = =
Zt
abc
[ ]
0.0058 j0.023 + 0 0
0 0.0173 j0.0403 + 0
0 0 0.0173 j0.0403 +
Ω =
n
t
12,470
240
 51.9583. = =
W [ ]
1
3

2 1 0
0 2 1
1 0 2
⋅ = D [ ]
1 1 – 0
0 1 1 –
1 – 0 1
= DI [ ]
1 0 1 –
1 – 1 0
0 1 – 1
=
0812_frame_C08.fm Page 232 Monday, October 28, 2002 11:13 AM
ThreePhase Transformer Models 233
AV [ ]
n
t
0 0
0 n
t
0
0 0 n
t
51.9583 0 0
0 51.9583 0
0 0 51.9583
= =
AI [ ]
1
n
t
 0 0
0
1
n
t
 0
0 0
1
n
t

0.0192 0 0
0 0.0192 0
0 0 0.0192
= =
G1 [ ]
1
Zt
ab
Zt
bc
Zt
ca
+ +

Zt
ca
Zt
bc
– 0
Zt
ca
Zt
ab
Zt
ca
+ 0
Zt
ab
Zt
bc
– – Zt
bc
– 0
⋅ =
G1 [ ]
0.3941 j0.0134 – −0.3941 j0.0134 + 0
0.3941 j0.0134 – 0.6059 j0.0134 + 0
−0.6059 j0.0134 – −0.3941 j0.0134 + 0
=
a
t
[ ]
51.9583
3

2 1 – 1 –
1 – 2 1 –
1 – 1 – 2
⋅
34.6389 17.3194 – 17.3194 –
17.3194 – 34.6389 17.3194 –
17.3194 – −17.3194 34.6389
= =
b
1
[ ] = AV [ ] W [ ] Zt
abc
[ ] G1 [ ] =
0.2166 j0.583 + 0.0826 j0.1153 + 0
0.0826 j0.1153 + 0.2166 j0.583 + 0
−0.2993 j0.6983 – −0.2993 j0.6983 – 0
⋅ ⋅ ⋅
A
t
[ ]
1
3 51.983 ⋅

2 1 – 1 –
1 – 2 1 –
1 – 1 – 2
⋅
0.0128 0.0064 – 0.0064 –
0.0064 – 0.0128 −0.0064
0.0064 – 0.0064 – 0.0128
= =
B
t
[ ] W [ ] Zt
abc
[ ] G1 [ ] ⋅ ⋅
0.0042 j0.0112 + 0.0016 j0.0022 + 0
0.0016 j0.0022 + 0.0042 j0.0112 + 0
−0.0058 j0.0134 – −0.0058 j0.0134 – 0
= =
0812_frame_C08.fm Page 233 Friday, July 20, 2001 4:47 PM
234 Distribution System Modeling and Analysis
With the generalized constant matrices deﬁned, the delta load currents can
ﬁrst be computed:
(1)
(2) Compute the line currents:
(3) Compute the equivalent linetoneutral secondary voltages:
(4) Use the generalized constant matrices to compute the primary
equivalent linetoneutral voltages:
The primary linetoline voltages are
(5) The primary line currents are
IL
i
SL
i
VLL
i

∗
416.67/ 25.84 –
208.33/ 156.87 –
208.33/83.13
A = =
I
abc
[ ] DI [ ] IL [ ] ⋅
522.93/ 47.97 –
575.31/170.01
360.84/53.13
A = =
VLN
abc
[ ] W [ ] VLL
abc
[ ] ⋅ =
VLN
abc
[ ]
138.56/ 30 –
138.56/ 150 –
138.56/90
V =
VLN
ABC
[ ] a
t
[ ] VLN
abc
[ ] ⋅ b
t
[ ] I
abc
[ ] ⋅ +
7381.53/ 28.63 –
7445.51/ 148.41 –
7438.40/91.05
V = =
VLL
ABC
[ ] D [ ] VLN
ABC
[ ] ⋅
12,826/1.62
12,925/ 118.69 –
12,814/121.08
V = =
I
ABC
[ ] d
t
[ ] I
abc
[ ] ⋅
10.06/ 47.97 –
11.07/170.01
6.95/53.13
A = =
0812_frame_C08.fm Page 234 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 235
(6) Equation 8.118 can be used to compute the secondary delta currents:
The primary delta currents are computed from Equation 8.92:
A check on the primary line currents can be made by applying Equation 8.94:
These are the same primary line currents previously computed. This pro
vides a method of checking on the many calculations that have taken place
to get to this point. Another check can be made on the accuracy of the
calculations and the models by applying the reverse Equation 8.129:
Again, these are the same values of the equivalent secondary linetoneutral
voltages that were used at the start of the analysis.
Last but not least, a check on the operating kVA and power factor for each
transformer can be determined:
From this example it is seen that the operating kVA is very close to the rating
of each transformer.
ID
abc
[ ] G1 [ ] I
abc
[ ] ⋅
409.54/ 29.99 –
236.43/ 153.65 –
183.81/88.56
A = =
ID
ABC
[ ] AI [ ]
1 –
ID
abc
[ ] ⋅
7.88/ 29.99 –
4.55/ 153.65 –
3.54/88.56
A = =
I
ABC
[ ] DI [ ] ID
ABC
[ ] ⋅
10.06/ 47.98 –
11.07/170.01
6.95/53.13
A = =
VLN
abc
[ ] A
t
[ ] VLN
ABC
[ ] ⋅ B
t
[ ] I
abc
[ ] ⋅ –
138.56/ 30 –
138.56/ 150 –
138.56/90
V = =
ST
i
VLL
ABC
i
ID
ABC
i
( )
∗
⋅
1000

101.1
58.8
45.3
= = kVA
PF
i
0.85
0.82
0.84
lagging =
0812_frame_C08.fm Page 235 Friday, July 20, 2001 4:47 PM
236 Distribution System Modeling and Analysis
8.7 Open Wye–Open Delta Connection
A common load to be served on a distribution feeder is a combination of a
singlephase lighting load and a threephase power load. Many times the
threephase power load will be an induction motor. This combination load
can be served by an ungrounded wye–delta connection as previously
described, or by an open wye–open delta connection. When the threephase
load is small compared to the singlephase load, the open wye–open delta
connection is commonly used. The open wye–open delta connection requires
only two transformers, but the connection will provide threephase lineto
line voltages to the combination load. Figure 8.11 shows the open wye–open
delta connection and the primary and secondary positive sequence voltage
phasors. With reference to Figure 8.11, the basic ideal transformer voltages
as functions of the turns ratio are
(8.132)
(8.133)
FIGURE 8.11
Open wye–open delta connection.
V
AG
V
BG
V
CG
n
t
0 0
0 n
t
0
0 0 0
Vt
ab
Vt
bc
Vt
ca
⋅ =
VLG
ABC
[ ] AV [ ] Vt
abc
[ ] ⋅ =
V
AG
V
BG
Vt
ab
Vt
bc
t
n
_
+
_
+
+
_
+
_
V
BG
1
V
AG
1
V
ca
1
V
ab
1
V
bc
1
V
an
1
V
AB
1
+
X1a
_
X3c
+
_
V
X2b
_
V
ab
+
V
bc
I
b
I
c
I
a
I
ba I
cb
I
A
I
B
+
V
AB
Zt
ab
Zt
bc
0812_frame_C08.fm Page 236 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 237
The currents as functions of the turns ratio are given by:
(8.134)
Equation 8.134 can be expressed in matrix form by:
(8.135)
(8.136)
The ideal voltages on the delta windings can be determined by:
(8.137)
Applying Equation 8.132 to Equations 8.137:
(8.138)
Equation 8.138 can be put into threephase matrix form as:
(8.139)
(8.140)
The secondary linetoline voltages in Equation 8.140 can be replaced by the
equivalent linetoneutral secondary voltages:
(8.141)
I
ba
n
T
I
A
⋅ I
a
= =
I
cb
n
T
I
B
⋅ I
c
– = =
I
b
I
a
I
c
+ ( ) – =
I
a
I
b
I
c
n
t
0 0
n
t
– n
t
0
0 n
t
– 0
I
A
I
B
I
C
⋅ =
I
abc
[ ] AI [ ] I
ABC
[ ] ⋅ =
Vt
ab
V
ab
Zt
ab
I
a
⋅ + =
Vt
bc
V
bc
Zt
bc
I
c
⋅ – =
V
AG
n
t
Vt
ab
⋅ n
t
V
ab
⋅ n
t
Zt
ab
I
a
⋅ ⋅ + = =
V
BG
n
t
Vt
bc
⋅ n
t
V
bc
⋅ n
t
Zt
bc
I
c
⋅ ⋅ – = =
V
AG
V
BG
V
CG
n
t
0 0
0 n
t
0
0 0 0
V
ab
V
bc
V
ca
⋅
n
t
Zt
ab
⋅ 0 0
0 0 n
t
– Zt
bc
⋅
0 0 0
I
a
I
b
I
c
⋅ + =
VLG
ABC
[ ] AV [ ] VLL
abc
[ ] ⋅ b
t
[ ] I
abc
[ ] ⋅ + =
VLG
ABC
[ ] AV [ ] D [ ] VLN
abc
[ ] ⋅ ⋅ b
t
[ ] I
abc
[ ] ⋅ + =
0812_frame_C08.fm Page 237 Friday, July 20, 2001 4:47 PM
238 Distribution System Modeling and Analysis
The generalized matrix equation becomes:
(8.142)
where (8.143)
(8.144)
The source side line currents as functions of the load side line currents are
given by:
(8.145)
(8.146)
where (8.147)
The reverse equations relating the secondary equivalent linetoneutral volt
ages as functions of the source side linetoground voltages and line currents
is needed. Solving Equation 8.138 for the two linetoline secondary voltages:
(8.148)
VLG
ABC
[ ] a
t
[ ] VLN
abc
[ ] ⋅ b
t
[ ] I
abc
[ ] ⋅ + =
a
t
[ ]
n
t
n
t
– 0
0 n
t
n
t
–
0 0 0
=
b
t
[ ]
n
t
Zt
ab
⋅ 0 0
0 0 n
t
– Zt
bc
⋅
0 0 0
=
I
A
I
B
I
C
1
n
t
 0 0
0 0
1
n
t
 –
0 0 0
I
a
I
b
I
c
⋅ =
I
ABC
[ ] d
t
[ ] I
abc
[ ] ⋅ =
d
t
[ ]
1
n
t
 0 0
0 0
1
n
t
 –
0 0 0
=
V
ab
1
n
t
 V
AG
Zt
ab
I
a
⋅ – ⋅ =
V
bc
1
n
t
 V
BG
Zt
bc
I
c
⋅ – ⋅ =
0812_frame_C08.fm Page 238 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 239
The third linetoline voltage (V
ca
) must be equal to the negative sum of the
other two linetoline voltages (KVL). In matrix form the desired equation is
(8.149)
(8.150)
The equivalent secondary linetoneutral voltages are then given by:
(8.151)
The generalized reverse equation is given by:
(8.152)
where (8.153)
(8.154)
The open wye–open delta connection derived in this section utilized Phases
A and B on the primary. This is just one of three possible connections. The
other two possible connections would use Phases B and C, and then Phases
C and A. The terms “leading” and “lagging” connection are also associated
with the open wye–open delta connection. The generalized matrices will be
different from those just derived. The same procedure can be used to derive
the matrices for the other two connections.
V
ab
V
bc
V
ca
1
n
t
 0 0
0
1
n
t
 0
1
n
t
 –
1
n
t
 – 0
V
AG
V
BG
V
CG
Zt
ab
0 0
0 0 Zt
bc
–
Zt
ab
– 0 Zt
bc
I
a
I
b
I
c
⋅ – ⋅ =
VLL
abc
[ ] BV [ ] VLG
ABC
[ ] Zt
abc
[ ] I
abc
[ ] ⋅ – ⋅ =
VLN
abc
[ ] W [ ] VLL
abc
[ ] ⋅ W [ ] BV [ ] VLG
ABC
[ ] W [ ] – Zt
abc
[ ] I
abc
[ ] ⋅ ⋅ ⋅ = =
VLN
abc
[ ] A
t
[ ] VLG
ABC
[ ] B
t
[ ] I
abc
[ ] ⋅ – ⋅ =
A
t
[ ] W [ ] BV [ ] ⋅
1
3 n
t
⋅
.
2 1 0
1 – 1 0
1 – 2 – 0
= =
B
t
[ ] W [ ] Zt
abc
[ ] ⋅
1
3
.
2 Zt
ab
⋅ 0 Zt
bc
–
Zt
ab
– 0 Zt
bc
–
Zt
ab
– 0 2 Zt
bc
⋅
= =
0812_frame_C08.fm Page 239 Friday, July 20, 2001 4:47 PM
240 Distribution System Modeling and Analysis
When the lighting transformer is connected across the leading of the
two phases, the connection is referred to as the “leading” connection.
Similarly, when the lighting transformer is connected across the lagging
of the two phases, the connection is referred to as the “lagging” connec
tion. For example, if the bank is connected to Phases A and B and the
lighting transformer is connected from Phase A to ground, this would
be referred to as the leading connection because the voltage AG leads
the voltage BG by 120 degrees. Reverse the connection and it would
now be called the lagging connection. Obviously, there is a leading and
lagging connection for each of the three possible open wye–open delta
connections.
Example 8.4
The unbalanced load of Example 8.2 is to be served by the leading open
wye–open delta connection using Phases A and B. Assume that the voltages
again are balanced threephase at the load so that the voltages and line
currents of Example 8.2 will apply for this example. It will also be assumed
that the lighting transformer is rated 100 kVA and the power transformer is
rated 50 kVA. Calculate the primary voltages, currents, and the operating kVA
of each transformer.
Compute the generalized constant matrices:
a
t
[ ]
n
t
n
t
– 0
0 n
t
n
t
–
0 0 0
30 30 – 0
0 30 30 –
0 0 0
= =
b
t
[ ]
n
t
Zt
ab
⋅ 0 0
0 0 n
t
Zt
bc
⋅ –
0 0 0
=
b
t
[ ]
0.1728 j0.6912 + 0 0
0 0 0.5184 – j1.2096 –
0 0 0
=
d
t
[ ]
1
n
t
 0 0
0 0
1
n
t
 –
0 0 0
0.0333 0 0
0 0 0.0333 –
0 0 0
= =
0812_frame_C08.fm Page 240 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 241
The equivalent linetoneutral secondary voltages and line currents from
Example 8.2 are
The primary linetoground voltages are computed to be
The voltage V
CG
has been computed to be zero above. That should not be
interpreted to mean that the actual primary voltage from C to ground is zero.
Rather, it must be interpreted to mean that there is no transformer connected
between Phase C and ground. If Phase C is present on the primary, it will
certainly have a value other than zero.
The primary transformer currents are
The operating kVA and power factor of each transformer is
VLN
abc
[ ]
138.56/ 30 –
138.56/ 150 –
138.56/90
V =
I
abc
[ ]
522.93/ 47.97 –
575.31/170.01
360.84/53.13
A =
VLG
ABC
[ ] a
t
[ ] VLN
abc
[ ] ⋅ b
t
[ ] I
abc
[ ] ⋅ +
7531/1.33
7449/ 116.84 –
0
= = V
I
ABC
[ ] d
t
[ ] I
abc
[ ] ⋅
17.43/ 47.97 –
12.03/ 126.87 –
0
= = A
ST
i
VLG
i
I
ABC
i
( )
∗
⋅
1000

131.27
89.60
0
= = kVA
PF
i
0.652
0.985
0
lagging =
0812_frame_C08.fm Page 241 Friday, July 20, 2001 4:47 PM
242
Distribution System Modeling and Analysis
Note in this case that the two transformers are operating far above their ratings.
If this load is to be served by an open wye–open delta connection, the lighting
transformer would need to be rated 167 kVA and the power transformer rated
100 kVA. Note also that the operating power factors of the two transformers
are quite diverse.
8.8 The Thevenin Equivalent Circuit
The six generalized constant matrices for each of ﬁve common threephase
transformer connections have been developed. In Chapter 10, the section for
shortcircuit analysis will require the Thevenin equivalent circuit referenced
to the secondary terminals of the transformer. This equivalent circuit must
take into account the equivalent impedance between the primary terminals
of the transformer and the feeder source. Figure 8.12 is a general circuit
showing the feeder source down to the secondary bus. The Thevenin equiv
alent circuit needs to be determined at the secondary node of the transformer
bank. This basically is the same as referring the source voltage and the source
impedance to the secondary side of the transformer. The desired Thevenin
equivalent circuit at the transformer secondary node is shown in Figure 8.13.
A general Thevenin equivalent circuit that can be used for all connections
employs the generalized matrices.
FIGURE 8.12
Equivalent system.
FIGURE 8.13
The Thevenin equivalent circuit.
[Zsys
]
ABC
Source
[ELN
ABC
abc
]
[I
ABC
]
[I
abc
]
[VLN ]
[VLN
ABC
]
[Z
th
]
[I
abc
]
[VLN
abc
[E
th
]
0812_frame_C08.fm Page 242 Monday, October 28, 2002 10:49 AM
ThreePhase Transformer Models
243
In Figure 8.12 the primary transformer equivalent linetoneutral voltages
as functions of the source voltages and the equivalent high side impedance
are given by:
(8.155)
but
therefore (8.156)
The general reverse equation gives the secondary linetoneutral voltages as
functions of the primary linetoneutral voltages and secondary currents:
(8.157)
Substitute Equation 8.156 into Equation 8.157:
(8.158)
With reference to Equation 8.158, the Thevenin equivalent voltages and
impedances can be deﬁned as:
(8.159)
(8.160)
The deﬁnitions of the Thevenin equivalent voltages and impedances as given
in Equations 8.159 and 8.160 are general and can be used for all transformer
connections. Example 8.5 is used to demonstrate the computation and appli
cation of the Thevenin equivalent circuit.
Example 8.5
The ungrounded wye–delta transformer bank of Example 8.2 is connected
to a source through a onemile section of a fourwire threephase line. The
phase impedance matrix for the one mile of line is given by:
VLN
ABC
[ ] ELN
ABC
[ ] Zsys
ABC
[ ] I
ABC
[ ] ⋅ – =
I
ABC
[ ] d
t
[ ] I
abc
[ ] ⋅ =
VLN
ABC
[ ] ELN
ABC
[ ] Zsys
ABC
[ ] d
t
[ ] I
abc
[ ] ⋅ ⋅ + =
VLN
abc
[ ] A
t
[ ] VLN
ABC
[ ] B
t
[ ] I
abc
[ ] ⋅ – ⋅ =
VLN
abc
[ ] A
t
[ ] ELN
ABC
[ ] Zsys
ABC
[ ] d
t
[ ] I
abc
[ ] ⋅ ⋅ – { } B
t
[ ] I
abc
[ ] ⋅ – ⋅ =
Eth
abc
[ ] A
t
[ ] ELN
ABC
[ ] ⋅ =
Zth
abc
[ ] A
t
[ ] Zsys
ABC
[ ] d
t
[ ] B
t
[ ] + ⋅ ⋅ =
Zsys
ABC
[ ]
0.4576 j1.0780 + 0.1559 j0.5017 + 0.1535 j0.3849 +
0.1559 j0.5017 + 0.4666 j1.0482 + 0.1580 j0.4236 +
0.1535 j0.3849 + 0.1580 j0.4236 + 0.4615 j1.0651 +
Ω =
0812_frame_C08.fm Page 243 Monday, October 28, 2002 10:49 AM
244 Distribution System Modeling and Analysis
For the unbalanced load and balanced voltage in Example 8.2, the primary
linetoneutral voltages and primary currents were computed as:
The source linetoneutral voltage for this load condition is computed to be
Now the Thevenin equivalent circuit at the transformer bank secondary ter
minals is
The load currents computed in Example 8.2 were
VLN
ABC
[ ]
7367.6/1.4
7532.3/−119.1
7406.2/121.7
= V
I
ABC
[ ] d
t
[ ] I
abc
[ ] ⋅ =
11.54/−28.04
8.95/−166.43
7.68/101.16
A =
ELN
ABC
[ ] VLN
ABC
[ ] = Zsys
ABC
[ ] I
ABC
[ ] ⋅ +
7374.8/1.45
7537.4/−119.1
7409.9/121.8
= V
Eth
abc
[ ] A
t
[ ] ELN
ABC
[ ] ⋅
141.2/ 29.3 –
146.2/ 148.5 –
142.2/91.9
V = =
Zth
abc
[ ] A
t
[ ] Zsys
ABC
[ ] d
t
[ ] ⋅ B
t
[ ] + ⋅ =
Zth
abc
[ ]
0.0033 j0.0099 + 0.0026 j0.0039 + 0
0 0.0059 j0.0137 + 0
−0.0072 j0.0156 – −0.0046 j0.0117 – 0
0 Ω =
I
abc
[ ]
522.9/ 47.97 –
575.3/170.0
360.8/53.13
A =
0812_frame_C08.fm Page 244 Friday, July 20, 2001 4:47 PM
ThreePhase Transformer Models 245
Using the Thevenin equivalent circuit and the previously computed line
currents, the equivalent linetoneutral voltages can be computed:
This example is only intended to demonstrate that it is possible to compute
the Thevenin equivalent circuit at the secondary terminals of the transformer
bank. The example shows that using the Thevenin equivalent circuit and the
original secondary line currents, the original equivalent linetoneutral load
voltages are computed. The major application of the Thevenin equivalent
circuit will be in the shortcircuit analysis of a distribution that will be devel
oped in Chapter 10.
8.9 Summary
In this chapter the generalized matrices have been developed for ﬁve common
threephase transformer bank connections. The generalized matrices can be
derived for all of the other common threephase transformer bank connec
tions using the same approach taken in this chapter.
The generalized matrices can be used for two purposes:
1. A quick computation of the operating conditions of the individual
transformers for a speciﬁed load condition.
2. With the generalized matrices in a powerﬂow iterative routine, it
is not necessary to build into the program the ability to determine
the type of transformer bank connection.
For shortcircuit calculations it is necessary to develop the threephase Thev
enin equivalent circuit at the point of the fault. Since all transformer connec
tions are deﬁned by the constant matrices, the computation of the equivalent
circuit is straightforward.
Problems
8.1 A threephase substation transformer is connected delta–grounded wye
and rated:
5000 kVA, 115 kV delta−12.47 kV grounded wye, Z = 1.0 + j7.5%
VLN
abc
[ ] Eth
abc
[ ] Zth
abc
[ ] I
abc
[ ] ⋅ –
138.56/ 30 –
138.56/−150
138.56/90
V = =
0812_frame_C08.fm Page 245 Friday, July 20, 2001 4:47 PM
246
Distribution System Modeling and Analysis
The transformer serves an unbalanced wyeconnected load of:
Phase a: 1384.5 kVA, 89.2% lagging power factor, at V
Phase b: 1691.2 kVA, 80.2% lagging power factor, at V
Phase c: 1563.0 kVA, unity power factor, at V
(1) Determine the generalized matrices for the transformer.
(2) Compute the primary equivalent linetoneutral voltages.
(3) Compute the primary linetoline voltages.
(4) Compute the primary line currents.
(5) Compute the currents ﬂowing in the high side delta windings.
(6) Compute the real power loss in the transformer for this load
condition.
8.2
Three singlephase transformers are connected in delta–grounded wye
and are serving an unbalanced load. The ratings of three transformers are
Phase AB: 100 kVA, 12,470–120 V,
Z
=
1.3
+
j
1.7%
Phase BC: 50 kVA, 12,470–120 V,
Z
=
1.1
+
j
1.4%
Phase CA: same as Phase BC transformer
The unbalanced wyeconnected loads are
Phase a: 40 kVA, 0.8 lagging power factor,
V
=
117.5 V
Phase b: 85 kVA, 0.95 lagging power factor,
V
=
115.7 V
Phase c: 50 kVA, 0.8 lagging power factor,
V
=
117.0 V
(1) Determine the generalized matrices for this connection.
(2) Compute the load currents.
(3) Compute the primary linetoneutral voltages.
(4) Compute the primary linetoline voltages.
(5) Compute the primary currents.
(6) Compute the currents in the delta primary windings.
(7) Compute the transformer bank real power loss.
8.3
The three singlephase transformers of Problem 8.2 are serving an
unbalanced constant impedance wyeconnected load of:
Phase a: 0.32
+
j
0.14
Ω
Phase b: 0.21
+
j
0.08
Ω
Phase c: 0.28
+
j
0.12
Ω
6922.5/−33.1
6776.8/−153.4
7104.7/85.9
/−32.5
/−147.3
/−95.3
0812_frame_C08.fm Page 246 Monday, October 28, 2002 10:49 AM
ThreePhase Transformer Models
247
The transformers are connected to a balanced 12.47kV source.
(1) Determine the load currents.
(2) Determine the load voltages.
(3) Compute the complex power of each load.
(4) Compute the primary currents.
(5) Compute the operating kVA of each transformer.
8.4
A threephase transformer connected wye–delta is rated:
500 kVA, 4160–240 V,
Z
=
1.1
+
j
3.0%
The primary neutral is ungrounded. The transformer is serving a balanced
load of 480 kW with balanced voltages of 235 V linetoline and a lagging
power factor of 0.9.
(1) Compute the secondary line currents.
(2) Compute the primary line currents.
(3) Compute the currents ﬂowing in the secondary delta windings.
(4) Compute the real power loss in the transformer for this load.
8.5
The transformer of Problem 8.4 is serving an unbalanced delta load of:
S
ab
=
150 kVA, 0.95 lagging power factor
S
bc
=
125 kVA, 0.90 lagging power factor
S
ca
=
160 kVA, 0.8 lagging power factor
The magnitudes of the linetoline voltages are
V
ab
=
240 V,
V
bc
=
237 V,
V
ca
=
235 V
(1) Use the Law of Cosines to compute the angles on the voltages. Set
the voltage ab as reference.
(2) Determine the delta load currents.
(3) Compute the generalized matrices for the transformer bank.
(4) Compute the primary linetoneutral and linetoline voltages.
(5) Compute the primary line currents.
(6) Compute the operating kVA of each transformer winding.
8.6
Three singlephase transformers are connected in an ungrounded
wye–delta connection and serving an unbalanced delta connected load. The
transformers are rated:
0812_frame_C08.fm Page 247 Monday, October 28, 2002 10:50 AM
248
Distribution System Modeling and Analysis
Phase A: 15 kVA, 2400–240 V,
Z
=
1.3
+
j
1.0%
Phase B: 25 kVA, 2400–240 V,
Z
=
1.1
+
j
1.1%
Phase C: Same as phase A transformer
The transformers are connected to a balanced source of 4.16 kV lineto
line. The primary currents entering the transformer are
I
A
=
4.62 A, 0.95 lagging power factor
I
B
=
6.92 A, 0.88 lagging power factor
I
C
=
5.28 A, 0.69 lagging power factor
(1) Determine the primary linetoneutral voltages.
(2) Determine the line currents entering the deltaconnected load.
(3) Determine the linetoline voltages at the load.
(4) Determine the operating kVA of each transformer.
(5) Is it possible to determine the load currents in the deltaconnected
load? If so, do it. If not, why not?
8.7
Three singlephase transformers are connected in grounded wye–
grounded wye and are serving an unbalanced constant impedance load.
The transformer bank is connected to a balanced threephase 12.47 kV line
toline voltage source. The transformers are rated:
Phase A: 100 kVA, 7200–120 V,
Z
=
0.9
+
j
1.8%
Phase B and Phase C: 37.5 kVA, 7200–120 V,
Z
=
1.1
+
j
1.4%
The constant impedance wyeconnected loads are
Phase a: 0.14
+
j
0.08
Ω
Phase b: 0.40
+
j
0.14
Ω
Phase c: 0.50
+
j
0.20
Ω
(1) Compute the generalized matrices for this transformer bank.
(2) Determine the load currents.
(3) Determine the load voltages.
(4) Determine the kVA and power factor of each load.
(5) Determine the primary line currents.
(6) Determine the operating kVA of each transformer.
8.8
Three singlephase transformers are connected in delta–delta and are
serving a balanced threephase motor rated 150 kVA, 0.8 lagging power
factor, and a singlephase lighting load of 25 kVA, 0.95 lagging power factor
connected across Phases AB. The transformers are rated:
0812_frame_C08.fm Page 248 Monday, October 28, 2002 10:50 AM
ThreePhase Transformer Models
249
Phase AB: 75 kVA, 4800–240 V,
Z
=
1.0
+
j
1.5%
Phase BC: 50 kVA, 4800–240 V,
Z
=
1.1
+
j
1.4%
Phase CA: same as Phase BC
The load voltages are balanced threephase of 240 V linetoline.
(1) Determine the generalized matrices.
(2) Compute the motor input currents.
(3) Compute the singlephase lighting load current.
(4) Compute the primary line currents.
(5) Compute the primary linetoline voltages.
(6) Compute the currents ﬂowing in the primary and secondary delta
windings.
8.9
The threephase motor and singlephase lighting loads of Problem 8.8
are being served by an open wye–open delta transformer bank. The load
voltages are balanced threephase of 240 V linetoline. The transformers are
rated:
Lighting transformer connected to Phase A to ground:
100 kVA, 7200–240 V,
Z
=
0.8
+
j
1.5%
Power transformer connected to Phase B to ground
75 kVA, 7200–240 V,
Z
=
0.8
+
j
1.2%
(1) Determine the generalized matrices for this connection.
(2) Compute the primary linetoground and linetoline voltages.
(3) Compute the primary line currents.
(4) Compute the operating kVA of each transformer. Are they over
loaded?
8.10
The unbalanced loads of Problem 8.5 are served from an open wye–
open delta connection. The linetoline load voltages are the same as in
Problem 8.5. Each transformer is rated:
167 kVA, 2400–240 V,
Z
=
1.1
+
j
1.3%
(1) Determine the operating kVA and power factor for each trans
former.
(2) What is the percentage overload on each transformer? Is this over
load okay?
0812_frame_C08.fm Page 249 Monday, October 28, 2002 10:51 AM
0812_frame_C08.fm Page 250 Friday, July 20, 2001 4:47 PM
251
9
Load Models
The loads on a distribution system are typically speciﬁed by the complex
power consumed. With reference to Chapter 2, the speciﬁed load will be the
maximum diversiﬁed demand. This demand can be speciﬁed as kVA and
power factor, kW and power factor, or kW and kvar. The voltage speciﬁed
will always be the voltage at the lowvoltage terminals of the distribution
substation. This creates a problem since the current requirement of the loads
cannot be determined without knowing the voltage. For this reason, some
form of an iterative technique must be employed. An iterative technique will
be presented in Chapter 10.
Loads on a distribution feeder can modeled as wyeconnected or delta
connected. The loads can be threephase, twophase, or singlephase with
any degree of unbalance, and can be modeled as:
• Constant real and reactive power (constant PQ)
• Constant current
• Constant impedance
• Any combination of the above
The load models developed are to be used in the iterative process of a power
ﬂow program where the load voltages are initially assumed. One of the results
of the powerﬂow analysis is to replace the assumed voltages with the actual
operating load voltages. All models are initially deﬁned by a complex power
per phase and an assumed linetoneutral voltage (wye load) or an assumed
linetoline voltage (delta load). The units of the complex power can be in
voltamperes and volts, or perunit voltamperes and perunit voltages. For
all loads the line currents entering the load are required in order to perform
the powerﬂow analysis.
0812_frame_C09.fm Page 251 Friday, July 20, 2001 4:08 PM
252
Distribution System Modeling and Analysis
9.1 WyeConnected Loads
Figure 9.1 is the model of a wyeconnected load. The notation for the spec
iﬁed complex powers and voltages are as follows:
Phase a: (9.1)
Phase b: (9.2)
Phase c: (9.3)
9.1.1 Constant Real and Reactive Power Loads
The line currents for constant real and reactive power loads (PQ loads) are
given by:
(9.4)
FIGURE 9.1
Wyeconnected load.
S
a
/θ
a
P
a
jQ
a
and V
an
/δ
a
+ =
S
b
/θ
b
P
b
jQ
b
and V
bn
/δ
b
+ =
S
c
/θ
c
P
c
jQ
c
and V
cn
/δ
c
+ =
IL
a
S
a
V
an

∗
S
a
V
an
/δ
a
θ
a
– IL
a
/α
a
= = =
IL
b
S
b
V
bn

∗
S
b
V
bn
/δ
b
θ
b
– IL
b
/α
b
= = =
IL
c
S
c
V
cn

∗
S
c
V
cn
/δ
c
θ
c
– IL
c
/α
c
= = =
S
a
c
S
b
S
+


+

+
V
an
bn
V
cn
V
IL
a
c
IL
b
IL
0812_frame_C09.fm Page 252 Friday, July 20, 2001 4:08 PM
Load Models
253
In this model the linetoneutral voltages will change during each iteration
until convergence is achieved.
9.1.2 Constant Impedance Loads
The constant load impedance is ﬁrst determined from the speciﬁed complex
power and assumed linetoneutral voltages:
(9.5)
The load currents as a function of the constant load impedances are given by:
(9.6)
In this model the linetoneutral voltages will change during each iteration,
but the impedance computed in Equation 9.5 will remain constant.
9.1.3 Constant Current Loads
In this model the magnitudes of the currents are computed according to
Equations 9.4 and are then held constant while the angle of the voltage (
δ
)
changes, resulting in a changing angle on the current so that the power factor
of the load remains constant:
(9.7)
where
=
Linetoneutral voltage angles
=
Power factor angles.
Z
a
V
an
2
S
a
∗

V
an
2
S
a
/θ
a
Z
a
/θ
a
= = =
Z
b
V
bn
2
S
b
∗

V
bn
2
S
b
/θ
b
Z
b
/θ
b
= = =
Z
c
V
cn
2
S
c
∗

V
cn
2
S
c
/θ
c
Z
c
/θ
c
= = =
IL
a
V
an
Z
a

V
an
Z
a
/δ
a
θ
a
– IL
a
/α
a
= = =
IL
b
V
bn
Z
b

V
bn
Z
b
/δ
b
θ
b
– IL
b
/α
b
= = =
IL
c
V
cn
Z
c

V
cn
Z
c
/δ
c
θ
c
– IL
c
/α
c
= = =
IL
a
IL
a
/δ
a
θ
a
– =
IL
b
IL
b
/δ
b
θ
b
– =
IL
c
IL
c
/δ
c
θ
c
– =
δ
abc
θ
abc
0812_frame_C09.fm Page 253 Friday, July 20, 2001 4:08 PM
254 Distribution System Modeling and Analysis
9.1.4 Combination Loads
Combination loads can be modeled by assigning a percentage of the total
load to each of the three above load models. The total line current entering
the load is the sum of the three components.
Example 9.1
The complex powers of a wyeconnected load are
The load is speciﬁed to be 50% constant complex power, 20% constant im
pedance, and 30% constant current. The nominal linetoline voltage of the
feeder is 12.47 kV.
(1) Assume the nominal voltage and compute the component of load
current attributed to each component of the load and the total load
current.
The assumed linetoneutral voltages at the start of the iterative
routine are
The component of currents due to the constant complex power are
The constant impedances for that part of the load are computed as
S
abc
[ ]
2236.1/26.6
2506.0/28.6
2101.4/25.3
kVA =
VLN
abc
[ ]
7200/
0
7200/ 120 –
7200/120
V =
Ipq
i
S
i
1000 ⋅
VLN
i

∗
0.5 ⋅
155.3/ 26.6 –
174.0/ 148.6 –
146.0/94.7
A = =
Z
i
VLN
i
[ ]
2
S
i
∗
1000 ⋅

20.7 j10.4 +
18.2 j9.9 +
22.3 j10.6 +
Ω = =
0812_frame_C09.fm Page 254 Friday, July 20, 2001 4:08 PM
Load Models 255
For the ﬁrst iteration, the currents due to the constant impedance
portion of the load are
The magnitudes of the constant current portion of the load are
The contribution of the load currents due to the constant current
portion of the load are
The total load current is the sum of the three components
(2) Determine the currents at the start of the second iteration. The
voltages at the load after the ﬁrst iteration are
The steps are repeated with the exceptions that the impedances of
the constant impedance portion of the load will not be changed
and the magnitude of the currents for the constant current por
tion of the load change will not change.
Iz
i
VLN
i
Z
i

0.2 ⋅
62.1/ 26.6 –
69.6/ 148.6 –
58.4/94.7
A = =
IM
i
S
i
1000 ⋅
VLN
i

∗
0.3 ⋅
93.2
104.4
87.6
A = =
II
i
IM
i
/δ
i
θ
i
–
93.2/ 26.6 –
104.4/ 148.6 –
87.6/94.7
A = =
I
abc
[ ] I
pq
[ ] I
z
[ ] I
I
[ ] + +
310.6/ 26.6 –
348.1/ 148.6 –
292.0/94.7
A = =
VLN [ ]
6850.0/ 1.9 –
6972.7/ 122.1 –
6886.1/117.5
V =
0812_frame_C09.fm Page 255 Friday, July 20, 2001 4:08 PM
256 Distribution System Modeling and Analysis
The constant complex power portion of the load currents are
The currents due to the constant impedance portion of the load are
The currents due to the constant current portion of the load are
The total load currents at the start of the second iteration will be
Observe how these currents have changed from the original currents. The
currents for the constant complex power loads have increased because the
voltages are reduced from the original assumption. The currents for the con
stant impedance portion of the load have decreased because the impedance
stayed constant but the voltages are reduced. Finally, the magnitude of
constant current portion of the load has indeed remained constant. Again,
all three components of the load have the same phase angles since the power
factor of the load has not changed.
Ipq
i
S
i
1000 ⋅
VLN
i

∗
0.5 ⋅
163.2/ 28.5 –
179.7/ 150.7 –
152.7/92.1
A = =
Iz
i
VLN
i
Z
i

0.2 ⋅
59.1/ 28.5 –
67.4/ 150.7 –
55.9/92.1
A = =
II
i
IM
i
/δ
i
θ
i
–
93.2/ 28.5 –
104.4/ 150.7 –
87.6/92.1
A = =
I
abc
[ ] I
pq
[ ] I
z
[ ] I
I
[ ] + +
315.5/ 28.5 –
351.5/ 150.7 –
296.2/92.1
A = =
0812_frame_C09.fm Page 256 Friday, July 20, 2001 4:08 PM
Load Models 257
9.2 DeltaConnected Loads
The model for a deltaconnected load is shown in Figure 9.2. The notation
for the speciﬁed complex powers and voltages in Figure 9.2 are as follows:
Phase ab: (9.8)
Phase bc: (9.9)
Phase ca: (9.10)
9.2.1 Constant Real and Reactive Power Loads
The currents in the delta connected loads are
(9.11)
In this model the linetoline voltages will change during each iteration
resulting in new current magnitudes and angles at the start of each iteration.
FIGURE 9.2
Deltaconnected load.
S
ab
/θ
ab
P
ab
jQ
ab
and V
ab
/δ
ab
+ =
S
bc
/θ
bc
P
bc
jQ
bc
and V
bc
/δ
bc
+ =
S
ca
/θ
ca
P
ca
jQ
ca
and V
ca
/δ
ca
+ =
IL
ab
S
ab
V
ab

∗
S
ab
V
ab
/δ
ab
θ
ab
– IL
ab
/α
ab
= = =
IL
bc
S
bc
V
bc

∗
S
bc
V
bc
/δ
bc
θ
bc
– IL
bc
/α
bc
= = =
IL
ca
S
ca
V
ca

∗
S
ca
V
ca
/δ
ca
θ
ca
– IL
ca
/α
ac
= = =
S
ab
ca
S
bc
S
IL
b
a
IL
c
IL
ab
IL
bc
IL
ca
IL
0812_frame_C09.fm Page 257 Friday, July 20, 2001 4:08 PM
258 Distribution System Modeling and Analysis
9.2.2 Constant Impedance Loads
The constant load impedance is ﬁrst determined from the speciﬁed complex
power and linetoline voltages:
(9.12)
The delta load currents as functions of the constant load impedances are
(9.13)
In this model the linetoline voltages will change during each iteration until
convergence is achieved.
9.2.3 Constant Current Loads
In this model the magnitudes of the currents are computed according to
Equations 9.11 and then held constant while the angle of the voltage (δ )
changes during each iteration. This keeps the power factor of the load
constant:
(9.14)
9.2.4 Combination Loads
Combination loads can be modeled by assigning a percentage of the total
load to each of the three above load models. The total delta current for each
load is the sum of the three components.
Z
ab
V
ab
2
S
ab
∗

V
ab
2
S
ab
/θ
ab
Z
ab
/θ
ab
= = =
Z
bc
VL
bc
2
S
bc
∗

V
bc
2
S
bc
/θ
bc
Z
bc
/θ
bc
= = =
Z
ca
V
ca
2
S
ca
∗

V
ca
2
S
ca
/θ
ca
Z
ca
/θ
ca
= = =
IL
ab
V
ab
Z
ab

V
anb
Z
ab
/δ
ab
θ
ab
– IL
ab
/α
ab
= = =
IL
bc
V
bc
Z
bc

V
bc
Z
bc
/δ
bc
θ
bc
– IL
bc
/α
bc
= = =
IL
ca
V
ca
Z
ca

V
ca
Z
ca
/δ
ca
θ
ca
– IL
ca
/α
ca
= = =
IL
ab
IL
ab
/δ
ab
θ
ab
– =
IL
bc
IL
bc
/δ
bc
θ
bc
– =
IL
ca
IL
bca
/δ
ca
θ
ca
– =
0812_frame_C09.fm Page 258 Friday, July 20, 2001 4:08 PM
Load Models 259
9.2.5 Line Currents Serving a DeltaConnected Load
The line currents entering the deltaconnected load are determined by apply
ing Kirchhoff’s current law at each of the nodes of the delta. In matrix form
the equations are
(9.15)
9.3 TwoPhase and SinglePhase Loads
In both the wye and deltaconnected loads, singlephase and twophase loads
are modeled by setting the currents of the missing phases to zero. The currents
in the phases present are computed using the same appropriate equations for
constant complex power, constant impedance, and constant current.
9.4 Shunt Capacitors
Shunt capacitor banks are commonly used in distribution systems to help
in voltage regulation and to provide reactive power support. The capacitor
banks are modeled as constant susceptances connected in either wye or delta.
Similar to the load model, all capacitor banks are modeled as threephase
banks with the currents of the missing phases set to zero for singlephase
and twophase banks.
9.4.1 WyeConnected Capacitor Bank
The model of a threephase wye connected shunt capacitor bank is shown in
Figure 9.3. The individual phase capacitor units are speciﬁed in kvar and kV.
The constant susceptance for each unit can be computed in Siemans. The
susceptance of a capacitor unit is computed by:
(9.16)
With the susceptance computed, the line currents serving the capacitor bank
are given by:
(9.17)
IL
a
IL
b
IL
c
1 0 1 –
1 – 1 0
0 1 – 1
IL
ab
IL
bc
IL
ca
⋅ =
B
c
kvar
kV
LN
2
1000 ⋅
 S =
IC
a
jB
a
V
an
⋅ =
IC
b
jB
b
V
bn
⋅ =
IC
c
jB
c
V
cn
⋅ =
0812_frame_C09.fm Page 259 Friday, July 20, 2001 4:08 PM
260 Distribution System Modeling and Analysis
9.4.2 DeltaConnected Capacitor Bank
The model for a deltaconnected shunt capacitor bank is shown in Figure 9.4.
The individual phase capacitor units are speciﬁed in kvar and kV. For the
deltaconnected capacitors the kV must be the linetoline voltage. The con
stant susceptance for each unit can be computed in Siemans. The susceptance
of a capacitor unit is computed by:
(9.18)
With the susceptance computed, the delta currents serving the capacitor bank
are given by:
(9.19)
FIGURE 9.3
Wyeconnected capacitor bank.
FIGURE 9.4
Deltaconnected capacitor bank.
+


+

+
V
an
bn
V
cn
V
IC
a
c
IC
b
IC
jB
a
b
jB
c
jB
B
c
kvar
kV
LL
2
1000 ⋅
 S =
IC
ab
jB
a
V
ab
⋅ =
IC
bc
jB
b
V
bc
⋅ =
IC
ca
jB
c
V
ca
⋅ =
0812_frame_C09.fm Page 260 Friday, July 20, 2001 4:08 PM
Load Models 261
The line currents ﬂowing into the deltaconnected capacitors are computed
by applying Kirchhoff’s current law at each node. In matrix form the equa
tions are
(9.20)
9.5 The ThreePhase Induction Motor
The analysis of an induction motor when operating under unbalanced voltage
conditions has traditionally been performed using the method of symmetrical
components. Using this approach, the positive and negative sequence equiv
alent circuits of the machine are developed and then, given the sequence line
toneutral voltages, the sequence currents are computed. The zero sequence
network is not required since the machines are typically connected delta or
ungrounded wye, which means there will not be any zero sequence currents
or voltages. The phase currents are determined by transforming back to the
phase line currents. The internal operating conditions are determined by the
complete analysis of the sequence networks.
The sequence linetoneutral equivalent circuit of a wyeconnected three
phase induction machine is shown in Figure 9.5. The circuit in Figure 9.5
applies to both the positive and negative sequence networks. The only dif
ference between the two is the value of the load resistance RL deﬁned by:
(9.21)
FIGURE 9.5
Sequence equivalent circuit.
IC
a
IC
b
IC
c
1 0 1 –
1 – 1 0
0 1 – 1
IC
ab
IC
bc
IC
ca
⋅ =
RL
i
1 s
i
–
s
i
 Rr
i
⋅ =
Rs jXs jXr Rr
Ir
RL
Is
Vs

+
jXm
Im
0812_frame_C09.fm Page 261 Friday, July 20, 2001 4:08 PM
262 Distribution System Modeling and Analysis
where
i = 1 for positive sequence
i = 2 for negative sequence.
Positive sequence slip: (9.22)
n
s
= Synchronous speed
n
r
= Rotor speed
Negative sequence slip: (9.23)
Note that the negative sequence load resistance RL
2
will be a negative value.
This will lead to a negative shaft power in the negative sequence. The neg
ative sequence currents are attempting to make the motor rotate in the reverse
direction. This negative power results in additional power losses and heating
in the motor. As a result, according to the ANSI C84.11995 standard, the
motor must be derated when the voltage unbalance is greater than 1%.
1
If the value of positive sequence slip (s
1
) is known, then the input sequence
impedances for the positive and negative sequence networks can be deter
mined as:
(9.24)
The input sequence impedances are converted to input sequence admittances
by taking the reciprocal of the impedances:
(9.25)
The sequence motor currents are
(9.26)
A wyeconnected induction motor will not have the neutral grounded. With
no ground, the zero sequence current must be zero. The linetoneutral zero
sequence voltage will also be zero. Therefore, it is possible to write an
equation equating the zero sequence current to the zero sequence lineto
neutral voltage:
(9.27)
s
1
n
s
n –
n
s
 =
s
2
2 s
1
– =
ZM
i
Rs
i
jXs
i
jXm
i
( ) Rr
i
RL
i
jXr
i
+ + ( )
Rr
i
RL
i
j Xm
i
Xr
i
+ ( ) + +
 + + =
YM
i
1
ZM
i
 =
IM
i
YM
i
VLN
i
⋅ =
IM
0
VLN
0
0 = =
0812_frame_C09.fm Page 262 Friday, July 20, 2001 4:08 PM
Load Models 263
Equations 9.26 and 9.27 can be put into matrix form:
(9.28)
(9.29)
With the neutral ungrounded, the only voltages that can be measured are the
linetoline voltages at the motor terminals. In symmetrical component the
ory, the relationship between sequence linetoneutral and sequence line
toline voltages is given by:
(9.30)
where (9.31)
In condensed form, Equation 9.30 becomes:
(9.32)
Substitute Equation 9.32 into Equation 9.29:
(9.33)
Equation 9.33 is used to compute the sequence motor currents when the
sequence linetoline voltages are known. What is really needed is a rela
tionship between the phase motor currents and the terminal phase lineto
line voltages. Recognize that in symmetrical component theory:
(9.34)
(9.35)
Applying Equations 9.34 and 9.35 to Equation 9.33 gives the ﬁnal desired
result:
(9.36)
IM
0
IM
1
IM
2
1 0 0
0 YM
1
0
0 0 YM
2
VLN
0
VLN
1
VLN
2
⋅ =
IM
012
[ ] YM
012
[ ] VLN
012
[ ] ⋅ =
VLN
0
VLN
1
VLN
2
1 0 0
0 t
s
∗
0
0 0 t
s
VLL
0
VLL
1
VLL
2
⋅ =
t
s
1
3
 /30 ⋅ =
VLN
012
[ ] T [ ] VLL
012
[ ] ⋅ =
IM
012
[ ] YM
012
[ ] T [ ] ⋅ VLL
012
[ ] ⋅ =
IM
abc
[ ] A
s
[ ] IM
012
[ ] ⋅ =
VLL
012
[ ] A
s
[ ]
1 –
VLL
abc
[ ] ⋅ =
IM
abc
[ ] A
s
[ ] IM
012
[ ] ⋅ A
s
[ ] YM
012
[ ] T [ ] A
s
[ ]
1 –
⋅ ⋅ VLL
abc
[ ] ⋅ ⋅ = =
YM
abc
[ ] VLL
abc
[ ] ⋅ =
0812_frame_C09.fm Page 263 Friday, July 20, 2001 4:08 PM
264
Distribution System Modeling and Analysis
where
(9.37)
(9.38)
The phase admittance matrix will be a packed matrix and can there
fore be modeled as a wyeconnected constant admittance load with mutual
admittances between phases. Even when the motor is being modeled as
connected in wye, the applied voltages must be linetoline. The constant
admittance will only be constant for a given value of slip. However, this
model allows the motor to be modeled under all load conditions from start
(slip
=
1) to full load (slip
≈
0.03).
Example 9.2
To demonstrate the analysis of an induction motor in the phase frame, the
following induction motor will be used:
25 Hp, 240 V operating with slip
=
0.035
Zs
=
0.0774
+
j
0.1843
Ω
Zm
=
0
+
j
4.8384
Ω
Zr
=
0.0908
+
j
0.1843
Ω
The load resistances are
Ω
Ω
The input sequence impedances are
YM
abc
[ ] A
s
[ ] YM
012
[ ] T [ ] A
s
[ ]
1 –
⋅ ⋅ ⋅ =
YM
abc
[ ]
YM
aa
YM
ab
YM
ac
YM
ba
YM
bb
YM
bc
YM
ca
YM
cb
YM
cc
=
YM
abc
[ ]
RL
1
1 0.035 –
0.035

0.0908 ⋅ 2.5029 = =
RL
2
1 1.965 ( ) –
1.965

0.0908 ⋅ 0.0446 – = =
ZM
1
Zs
Zm Zr RL
1
+ ( ) ⋅
Zm Zr RL
1
+ +
 + 1.9775 j1.3431 Ω + = =
ZM
2
Zs
Zm Zr RL
2
+ ( ) ⋅
Zm Zr RL
2
+ +
 + 0.1203 j0.3623 Ω + = =
0812_frame_C09.fm Page 264 Monday, October 28, 2002 10:56 AM
Load Models 265
The positive and negative sequence input admittances are
S
S
The sequence admittance matrix is
Applying Equation 9.37, the phase admittance matrix is
S
The linetoline terminal voltages are measured to be
Since the sum of the linetoline voltages must sum to zero, the law of cosines
can be used to determine the angles on the voltages. Applying the law of
cosines results in:
The phase motor currents can now be computed:
YM
1
1
ZM
1
 0.3461 j0.2350 – = =
YM
2
1
ZM
2
 0.8255 j2.4863 – = =
YM
012
[ ]
1 0 0
0 0.3461 j0.2350 – 0
0 0 0.8255 j2.4863 –
S =
YM
abc
[ ]
0.7452 j0.4074 – −0.0999 j0.0923 – 0.3547 j0.4997 +
−0.0999 j0.0923 + 0.7452 j0.4074 – −0.0999 j0.0923 –
−0.0999 j0.0923 – 0.3547 j0.4497 + 0.7452 j0.4074 –
=
V
ab
235 V, V
bc
240 V, V
ca
245 V = = =
VLL
abc
[ ]
235/
0
240/ 117.9 –
245/120.0
V =
IM
abc
[ ] YM
abc
[ ] VLL
abc
[ ] ⋅
53.15/ 71.0 –
55.15/ 175.1 –
66.6/55.6
A = =
0812_frame_C09.fm Page 265 Friday, July 20, 2001 4:08 PM
266 Distribution System Modeling and Analysis
It is obvious that the currents are very unbalanced. The measure of unbalance
for the voltages and currents can be computed as:
This example demonstrates that the current unbalance is seven times greater
than the voltage unbalance. This ratio of current unbalance to voltage unbal
ance is typical. The actual operating characteristics including stator and
rotor losses of the motor can be determined using the method developed in
Reference 2.
References
1. American National Standard for Electric Power Systems and Equipment—Voltage
Ratings (60 Hertz), ANSI C84.11995, National Electrical Manufacturers
Association, Rosslyn, VA, 1996.
2. Kersting, W.H. and Phillips, W.H., Phase frame analysis of the effects of voltage
unbalance on induction machines, IEEE Transactions on Industry Applications,
March/April 1997.
Problems
9.1 A 12.47kV feeder provides service to an unbalanced wyeconnected
load speciﬁed to be
Phase a: 1000 kVA, 0.9 lagging power factor
Phase b: 800 kVA, 0.95 lagging power factor
Phase c: 1100 kVA, 0.85 lagging power factor
(1) Compute the initial load currents assuming the loads are modeled
as constant complex power.
(2) Compute the magnitude of the load currents that will be held
constant assuming the loads are modeled as constant current.
(3) Compute the impedance of the load to be held constant assuming
the loads are modeled as constant impedance.
(4) Compute the initial load currents assuming that 60% of the load is
complex power, 25% constant current, and 15% constant impedance.
V
unbalance
Max. deviation
V
avg

100 ⋅
5
240

100 ⋅ 2.08% = = =
I
unbalance
Max. deviation
I
avg

100 ⋅
8.3232
58.31

100 ⋅ 14.27% = = =
0812_frame_C09.fm Page 266 Friday, July 20, 2001 4:08 PM
Load Models 267
9.2 Using the results of Problem 9.1, rework the problem at the start of the
second iteration if the load voltages after the ﬁrst iteration have been com
puted to be
9.3 A 12.47kV feeder provides service to an unbalanced deltaconnected
load speciﬁed to be
Phase ab: 1500 kVA, 0.95 lagging power factor
Phase bc: 1000 kVA, 0.85 lagging power factor
Phase ca: 950 kVA, 0.9 lagging power factor
(1) Compute the load and line currents if the load is modeled as
constant complex power.
(2) Compute the magnitude of the load current to be held constant if
the load is modeled as constant current.
(3) Compute the impedance to be held constant if the load is modeled
as constant impedance.
(4) Compute the line currents if the load is modeled as 25% constant
complex power, 20% constant current, and 55% constant impedance.
9.4 After the ﬁrst iteration of the system of Problem 9.5, the load voltages
are
(1) Compute the load and line currents if the load is modeled as
constant complex power.
(2) Compute the load and line currents if the load is modeled as
constant current.
(3) Compute the load and line currents if the load is modeled as
constant impedance.
(4) Compute the line currents if the load mix is 25% constant complex
power, 20% constant current, and 55% constant impedance.
VLN
abc
[ ]
6851/ 1.9 –
6973/ 122.1 –
6886/117.5
V =
VLL
abc
[ ]
11,981/28.3
12,032/ 92.5 –
11,857/147.7
V =
0812_frame_C09.fm Page 267 Friday, July 20, 2001 4:08 PM
268
Distribution System Modeling and Analysis
9.5
The motor in Example 9.2 is operating with a slip of 0.03 with balanced
voltages of 240 V linetoline. Determine the following:
(1) The input line currents and complex threephase power.
(2) The currents in the rotor circuit.
(3) The developed shaft power in Hp.
9.6
The motor in Example 9.2 is operating with a slip of 0.03, and the line
toline voltage magnitudes are
(1) Compute the angles on the linetoline voltages assuming the volt
age
ab
is referenced.
(2) For the given voltages and slip, determine the input line currents
and complex threephase power.
(3) Compute the rotor currents.
(4) Compute the developed shaft power in Hp.
9.7
For the motor in Example 9.2, determine the starting line current if the
motor terminal voltages are those given in Problem 9.6.
V
ab
240 V, V
bc
230 V, V
ca
250 V. = = =
0812_frame_C09.fm Page 268 Monday, October 28, 2002 10:51 AM
269
10
Distribution Feeder Analysis
The analysis of a distribution feeder will typically consist of a study of the
feeder under normal steadystate operating conditions (powerﬂow analy
sis), and a study under shortcircuit conditions (shortcircuit analysis). Mod
els of all of the components of a distribution feeder have been developed in
previous chapters. These models will be applied for the analysis under
steadystate and shortcircuit conditions.
10.1 PowerFlow Analysis
The powerﬂow analysis of a distribution feeder is similar to that of an
interconnected transmission system. Typically, what will be known prior to
the analysis will be the threephase voltages at the substation and the com
plex power of all of the loads and the load model (constant complex power,
constant impedance, constant current, or a combination). Sometimes the
input complex power supplied to the feeder from the substation is also
known.
In Chapters 6, 7, and 8 phase frame models were developed for the series
components of a distribution feeder. In Chapter 9 models were developed for
the shunt components (loads and capacitor banks). These models are used
in the powerﬂow analysis of a distribution feeder.
A powerﬂow analysis of a feeder can determine the following by phase
and total threephase:
• Voltage magnitudes and angles at all nodes of the feeder
• Line ﬂow in each line section speciﬁed in kW and kvar, amps and
degrees, or amps and power factor
• Power loss in each line section
• Total feeder input kW and kvar
• Total feeder power losses
• Load kW and kvar based upon the speciﬁed model for the load
0812_frame_C10.fm Page 269 Friday, July 20, 2001 3:57 PM
270
Distribution System Modeling and Analysis
10.1.1 The Ladder Iterative Technique
Because a distribution feeder is radial, iterative techniques commonly used in
transmission network powerﬂow studies are not used because of poor con
vergence characteristics.
1
Instead, an iterative technique speciﬁcally designed
for a radial system is used.
10.1.1.1 Linear Network
A modiﬁcation of the ladder network theory of linear systems provides a
robust iterative technique for powerﬂow analysis.
2
A distribution feeder
is nonlinear because most loads are assumed to be constant kW and kvar.
However, the approach taken for the linear system can be modiﬁed to take
into account the nonlinear characteristics of the distribution feeder. Figure 10.1
shows a linear ladder network. For the ladder network it is assumed that
all of the line impedances and load impedances are known along with the
voltage at the source (
V
s
). The solution for this network is to assume a
voltage at the most remote load (
V
5
). The load current
I
5
is then determined
as:
(10.1)
For this end node case, the line current
I
45
is equal to the load current
I
5
.
Applying Kirchhoff’s voltage law (KVL), the voltage at Node 4 (
V
4
) can be
determined:
(10.2)
The load current
I
4
can be determined, and then Kirchhoff’s current law
(KCL) applied to determine the line current
I
34
:
(10.3)
Kirchhoff’s voltage law is applied to determine the node voltage
V
3
. This
procedure is continued until a voltage (
V
1
) has been computed at the source.
FIGURE 10.1
Linear ladder network.
I
5
V
5
ZL
5
 =
V
4
V
5
Z
45
I
45
⋅ + =
I
34
I
45
= I
4
+
5 4 3 2 1
Z
45
Z
34
Z
23
Z
12
ZL
2
ZL
3
ZL
4
ZL
5
I
I
2
I
3
I
23
I
12
I
34
I
45
I
4 5
V
S
+

0812_frame_C10.fm Page 270 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis
271
The computed voltage
V
1
is compared to the speciﬁed voltage
V
s
. There will
be a difference between these two voltages. The ratio of the speciﬁed voltage
to the compute voltage can be determined as:
(10.4)
Since the network is linear, all of the line and load currents and node voltages
in the network can be multiplied by the
Ratio
for the ﬁnal solution to the
network.
10.1.1.2 Nonlinear Network
The linear network of Figure 10.1 is modiﬁed to a nonlinear network by
replacing all of the constant load impedances by constant complex power
loads as shown in Figure 10.2. The procedure outlined for the linear network
is applied initially to the nonlinear network. The only difference is that the
load current at each node is computed by:
(10.5)
The forward sweep will determine a computed source voltage
V
1
. As in the
linear case, this ﬁrst iteration will produce a voltage that is not equal to the
speciﬁed source voltage
V
s
. Because the network is nonlinear, multiplying
currents and voltages by the ratio of the speciﬁed voltage to the computed
voltage will not give the solution. The most direct modiﬁcation to the ladder
network theory is to perform a backward sweep. The backward sweep com
mences by using the speciﬁed source voltage and the line currents from the
forward sweep. Kirchhoff’s voltage law is used to compute the voltage at
Node 2 by:
(10.6)
FIGURE 10.2
Nonlinear ladder network.
Ratio
V
s
V
1
 =
I
n
S
n
V
n

∗
=
V
2
V
s
Z
12
I
12
⋅ – =

+
S
V
5 4
I
45
I
34
I
12
I
23
I
3
I
2
I
I
12
Z
23
Z
34
Z
45
Z
1 2 3 4 5
S
2 3
S
4
S
5
S
0812_frame_C10.fm Page 271 Friday, July 20, 2001 3:57 PM
272
Distribution System Modeling and Analysis
This procedure is repeated for each line segment until a new voltage is deter
mined at Node 5. Using the new voltage at Node 5, a second forward sweep
is started that will lead to a new computed voltage at the source.
The forward and backward sweep process is continued until the difference
between the computed and speciﬁed voltage at the source is within a given
tolerance.
Example 10.1
A singlephase lateral is shown in Figure 10.3. The line impedance is
The impedance of the line segment 12 is
The impedance of the line segment 23 is
The loads are
The source voltage at Node 1 is 7200 V.
Compute the node voltages after one full iteration.
FIGURE 10.3
Singlephase lateral.
z 0.3 j0.6 + = Ω/mile
Z
12
0.3 j0.6 + ( )
3000
5280
 ⋅ 0.1705 j0.3409 + = = Ω
Z
23
0.3 j0.6 + ( ) =
4000
5280
 ⋅ 0.2273 = j0.4545 + Ω
S
2
1500 j750 + =
S
3
900 j500 + =
kW jkvar + ( )
S
2
S
3
3000’ 4000’
0812_frame_C10.fm Page 272 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis
273
The forward sweep begins by assuming the voltage at Node 3 to be .
The load current at Node 3 is computed to be
The current ﬂowing in the line section 23 is
The voltage at Node 2 is computed to be
The load current at Node 2 is
The current in line segment 12 is
A
The computed voltage at the source Node 1 is
At this point the magnitude of the computed voltage at Node 1 is compared
to the magnitude of the speciﬁed source voltage:
If the error is less than a speciﬁed tolerance, the solution has been achieved.
If the error is greater than the tolerance, the backward sweep begins. A typical
tolerance is 0.001 per unit, which on a 7200V base is 7.2 V. Since the error
in this case is greater than the tolerance, the backward sweep begins by
setting the voltage at Node 1 to the speciﬁed source voltage:
7200/0
I
3
900
j
500 + ( ) 1000 ⋅
7200/
0

∗
143.0/ 29.0 – = = A
I
23
I
3
143.0/ 29.0 – = = A
V
2
V
3
= Z
23
I
23
⋅ + 7200/
0
0.2273 j0.4545 + ( ) 143.0/ 29.1 – ⋅ + =
7260.1/0.32 V =
I
2
1500
j
750 + ( ) 1000 ⋅
7260 1 ⋅ /0.32

∗
231.0/ 26.3 – = = A
I
12
I
23
I
2
+ 373.9/ 27.3 – = =
V
1
V
2
Z
12
I
12
⋅ + 7376.2/0.97 = = V
Error V
s
V
1
– 176.2 = = V
V
1
V
s
7200/
0
V = =
0812_frame_C10.fm Page 273 Friday, July 20, 2001 3:57 PM
274 Distribution System Modeling and Analysis
Now the voltage at Node 2 is computed using this value of the Node 1
voltage and the computed line current in the forward sweep current:
The backward sweep continues by computing the next downstream voltage.
All of the currents computed in the forward sweep are used in the backward
sweep:
This completes the ﬁrst iteration. At this point the forward sweep will be
repeated, only this time starting with the new voltage at Node 3 rather than
the initially assumed voltage.
10.1.2 The General Feeder
A typical distribution feeder will consist of the primary main, with laterals
tapped off the primary main and sublaterals tapped off the laterals, etc.
Figure 10.4 shows a typical feeder. The ladder iterative technique for the
feeder of Figure 10.4 would proceed as follows:
FIGURE 10.4
Typical distribution feeder.
V
2
V
1
Z I
12
⋅ – 7200/
0
= = 0.1705 j0.3409 + ( ) – 373.9/ 27.2 – ⋅
7085.4/ 0.68 – = V
V
3
V
2
Z I
23
⋅ – 7026.0/ 1.02 – V = =
Source Node
6
2
3
4
10 11
12
13
5
7
8
9
1
0812_frame_C10.fm Page 274 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 275
1. Assume threephase voltages at the end nodes (6, 8, 9, 11, and 13).
The usual assumption is to use the nominal voltages.
2. Starting at Node 13, compute the node current (load current plus
capacitor current if present).
3. With this current, apply Kirchhoff’s voltage law (KVL) to calculate
the node voltages at 12 and 10.
4. Node 10 is referred to as a “junction” node since laterals branch in
two directions from the node. For this feeder, go to Node 11 and
compute the node current. Use that current to compute the voltage
at Node 10. This will be referred to as “the most recent voltage at
Node 10.”
5. Using the most recent value of the voltage at Node 10, the node
current at Node 10 (if any) is computed.
6. Apply Kirchhoff’s current law (KCL) to determine the current
ﬂowing from Node 4 toward Node 10.
7. Compute the voltage at Node 4.
8. Node 4 is a junction node. An end node downstream from Node 4
is selected to start the forward sweep toward Node 4.
9. Select Node 6, compute the node current, and then compute the
voltage at Junction Node 5.
10. Go to downstream end Node 8. Compute the node current and
then the voltage at Junction Node 7.
11. Go to downstream end Node 9. Compute the node current and
then the voltage at Junction Node 7.
12. Compute the node current at Node 7 using the most recent value
of the Node 7 voltage.
13. Apply KCL at Node 7 to compute the current ﬂowing on the line
segment from Node 5 to Node 7.
14. Compute the voltage at Node 5.
15. Compute the node current at Node 5.
16. Apply KCL at Node 5 to determine the current ﬂowing from
Node 4 toward Node 5.
17. Compute the voltage at Node 4.
18. Compute the node current at Node 4.
19. Apply KCL at Node 4 to compute the current ﬂowing from Node 3
to Node 4.
20. Calculate the voltage at Node 3.
21. Compute the node current at Node 3.
22. Apply KCL at Node 3 to compute the current ﬂowing from Node 2
to Node 3.
0812_frame_C10.fm Page 275 Friday, July 20, 2001 3:57 PM
276 Distribution System Modeling and Analysis
23. Calculate the voltage at Node 2.
24. Compute the node current at Node 2.
25. Apply KCL at Node 2.
26. Calculate the voltage at Node 1.
27. Compare the calculated voltage at Node 1 to the speciﬁed source
voltage.
28. If not within tolerance, use the speciﬁed source voltage and the
forward sweep current ﬂowing from Node 1 to Node 2, and com
pute the new voltage at Node 2.
29. The backward sweep continues, using the new upstream voltage
and line segment current from the forward sweep to compute the
new downstream voltage.
30. The backward sweep is completed when new voltages at all end
nodes have been completed.
31. This completes the ﬁrst iteration.
32. Repeat the forward sweep, only now using the new end voltages
rather than the assumed voltages as was done in the ﬁrst iteration.
33. Continue the forward and backward sweeps until the calculated
voltage at the source is within a speciﬁed tolerance of the source
voltage.
34. At this point the voltages are known at all nodes, and the currents
ﬂowing in all line segments are known. An output report can be
produced giving all desired results.
10.1.3 The Unbalanced ThreePhase Distribution Feeder
The previous section outlined the general procedure for performing the
ladder iterative technique. This section will address how that procedure can
be used for an unbalanced threephase feeder.
Figure 10.5 is the oneline diagram of an unbalanced threephase feeder.
The topology of the feeder in Figure 10.5 is the same as the feeder in
Figure 10.4. Figure 10.5 shows more detail of the feeder, however. The
feeder of Figure 10.5 can be broken into “series” components and “shunt”
components.
10.1.3.1 Series Components
The series components of a distribution feeder are
• Line segments
• Transformers
• Voltage regulators
0812_frame_C10.fm Page 276 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 277
Models for each of the series components have been developed in earlier
chapters. In all cases, models (threephase, twophase, and singlephase)
were in terms of generalized matrices. Figure 10.6 shows the general model
for each of the series components. The general equations deﬁning the input
(Node n) and output (Node m) voltages and currents are given by:
(10.7)
(10.8)
FIGURE 10.5
Unbalanced threephase distribution feeder.
FIGURE 10.6
Series feeder component.
a
b
c
b
c
c b a
c
a
c
a
6
2
3
4
10 11
12
13
5
7
8
9
a
a b c
a b c
4’
3’
[Iabc]
n
[Iabc]
m
Node n
Node m
[Vabc]
n
[Vabc]
m
Series Feeder
Component
V
abc
[ ]
n
a [ ] V
abc
[ ]
m
⋅ = b [ ] I
abc
[ ]
m
⋅ +
I
abc
[ ]
n
c [ ] V
abc
[ ]
m
⋅ d [ ] I
abc
[ ]
m
⋅ + =
0812_frame_C10.fm Page 277 Friday, July 20, 2001 3:57 PM
278 Distribution System Modeling and Analysis
The general equation relating the output (Node m) and input (Node n)
voltages are given by:
(10.9)
In Equations 10.7, 10.8, and 10.9 the voltages are linetoneutral for a four
wire wye feeder and equivalent linetoneutral for a threewire delta system.
For voltage regulators, the voltages are linetoneutral for terminals that are
connected to a fourwire wye, and linetoline when connected to a three
wire delta.
10.1.3.2 Shunt Components
The shunt components of a distribution feeder are
• Spot loads
• Distributed loads
• Capacitor banks
Spot loads are located at a node and can be threephase, twophase, or single
phase, and connected in either a wye or a delta connection. The loads can
be modeled as constant complex power, constant current, constant imped
ance, or a combination of the three.
Distributed loads are modeled in accordance with Figure 3.11. A distrib
uted load is modeled when the loads on a line segment are uniformly
distributed along the length of the segment. As in the spot load, the distrib
uted load can be threephase, twophase, or singlephase, and connected in
either a wye or delta. The loads can be modeled as constant complex power,
constant current, constant impedance or a combination of the three. Two
thirds of the load is connected at a dummy node located at the onequarter
point of the line and the remaining one third of the load is connected at the
load end of the line segment.
Capacitor banks are located at a node and can be threephase, twophase,
or singlephase, and can be connected in a wye or delta. Capacitor banks are
modeled as constant admittances.
In Figure 10.5 the solid line segments represent overhead lines, while the
dashed lines represent underground lines. Note that the phasing is shown
for all of the line segments. In Chapter 4 the application of Carson’s equations
for computing the line impedances for overhead and underground lines was
presented. In that chapter it was pointed out that twophase and singlephase
lines are represented by a threebythree matrix with zeros set in the rows
and columns of the missing phases.
In Chapter 5 the method for the computation of the shunt capacitive
susceptance for overhead and underground lines was presented. Most of the
time the shunt capacitance of the line segment can be ignored, however, for
long underground line segments the shunt capacitance should be included.
V
abc
[ ]
m
A [ ] V
abc
[ ]
n
⋅ = B [ ] I
abc
[ ]
m
⋅ –
0812_frame_C10.fm Page 278 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 279
The node currents may be threephase, twophase, or singlephase and
consist of the sum of the load current at the node plus the capacitor current
(if any) at the node.
10.1.4 Applying the Ladder Iterative Technique
Section 10.1.2 outlined the steps required for the application of the ladder
iterative technique. Generalized matrices have been developed in Chapters
6, 7, and 8 for the series devices. By applying the generalized matrices, the
computation of the voltage drops along a segment will always be the same
regardless of whether the segment represents a line, voltage regulator, or
transformer.
In the preparation of data for a powerﬂow study, it is extremely important
that the impedances and admittances of the line segments be computed using
the exact spacings and phasing. Because of the unbalanced loading and result
ing unbalanced line currents, the voltage drops due to the mutual coupling
of the lines become very important. It is not unusual to observe a voltage
rise on a lightly loaded phase of a line segment that has an extreme current
unbalance.
The real power loss in a device should not be computed by using the phase
current squared times the phase resistance. In a balanced system that works,
however, in an unbalanced system, the real power losses of a line segment
must be computed as the difference (by phase) of the input power in a line
segment minus the output power of the line segment. It is possible to have
a negative power loss on a phase that is lightly loaded compared to the other
two phases. Computing power loss as the phase current squared times the
phase resistance does not give the actual real power loss in the phases.
10.1.5 Putting It All Together
At this point of the text, the models for all components of a distribution feeder
have been developed. The modiﬁed ladder interative technique has also been
developed. It is time to put them all together and demonstrate the power
ﬂow anlaysis of a very simple system. Example 10.2 will be long but will
demonstrate how the models of the components work together in applying
the modiﬁed ladder technique to achieve a ﬁnal solution of the operating
characteristics of an unbalanced feeder.
Example 10.2
A very simple distribution feeder is shown in Figure 10.7. For the system in
Figure 10.7, the inﬁnite bus voltages are balanced threephase of 12.47 kV
linetoline. The source line segment from Node 1 to Node 2 is a threewire
delta 2000 ft. long and is constructed on the pole conﬁguration of Figure 4.7.
The load line segment from Node 3 to Node 4 is 2500 ft. long and is also
constructed on the pole conﬁguration of Figure 4.7, but is a fourwire wye.
0812_frame_C10.fm Page 279 Friday, July 20, 2001 3:57 PM
280 Distribution System Modeling and Analysis
Both line segments use 336,400 26/7 ACSR phase conductors, and the neutral
conductor on the fourwire wye line is 4/0 6/1ACSR. Since the lines are short,
the shunt admittance will be neglected. The phase impedance matrices for
the two line segments are
The transformer bank is connected deltagrounded wye and consists of three
singlephase transformers each rated:
The feeder serves an unbalanced threephase wyeconnected load of:
S
a
= 750 kVA at 0.85 lagging power factor
S
b
= 1000 kVA at 0.90 lagging power factor
S
c
= 1250 kVA at 0.95 lagging power factor
Before starting the iterative solution, the generalized matrices for the three
series components must be deﬁned.
SOURCE LINE SEGMENT
Equations 6.9 and 6.18:
FIGURE 10.7
Example 10.2 feeder.
Load
3 2 4
Bus
1
[ZeqS]
[ZeqL]
ZeqS
ABC
[ ]
0.1414 j0.5353 + 0.0361 j0.3225 + 0.0361 j0.2752 +
0.0361 j0.3225 + 0.1414 j0.5353 + 0.0361 j0.2955 +
0.0361 j0.2752 + 0.0361 j0.2955 + 0.1414 j0.5353 +
Ω =
ZeqL
abc
[ ]
0.1907 j0.5035 + 0.0607 j0.2302 + 0.0598 j0.1751 +
0.0607 j0.2302 + 0.1939 j0.4885 + 0.0614 j0.1931 +
0.0598 j0.1751 + 0.0614 j0.1931 + 0.1921 j0.4970 +
= Ω
2000 kVA, 12.47 2.4 kV, – Z 1.0 j6.0% + =
a
1
[ ] d
1
[ ] U [ ]
1 0 0
0 1 0
0 0 1
= = =
0812_frame_C10.fm Page 280 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 281
Equation 6.10:
Equation 6.17:
Equation 6.27:
Equation 6.28:
LOAD LINE SEGMENT (using the same equations as for the source segment)
b
1
[ ] ZeqS
ABC
[ ]
0.1414 j0.5353 + 0.0361 j0.3225 + 0.0361 j0.2752 +
0.0361 j0.3225 + 0.1414 j0.5353 + 0.0361 j0.2955 +
0.0361 j0.2752 + 0.0361 j0.2955 + 0.1414 j0.5353 +
= =
c
1
[ ] 0 [ ] =
A
1
[ ] a
1
[ ]
1 –
1 0 0
0 1 0
0 0 1
= =
B
1
[ ] a
1
[ ]
1 –
b
1
[ ] ⋅ =
0.1414 j0.5353 + 0.0361 j0.3225 + 0.0361 j0.2752 +
0.0361 j0.3225 + 0.1414 j0.5353 + 0.0361 j0.2955 +
0.0361 j0.2752 + 0.0361 j0.2955 + 0.1414 j0.5353 +
=
a
2
[ ] d
2
[ ]
1 0 0
0 1 0
0 0 1
= =
b
2
[ ]
0.1907 j0.5035 + 0.0607 j0.2302 + 0.0598 j0.1751 +
0.0607 j0.2302 + 0.1939 j0.4885 + 0.0614 j0.1931 +
0.0598 j0.1751 + 0.0614 j0.1931 + 0.1921 j0.4970 +
=
c
2
[ ] 0 [ ] =
A
2
[ ]
1 0 0
0 1 0
0 0 1
=
B
2
[ ]
0.1907 j0.5035 + 0.0607 j0.2302 + 0.0598 j0.1751 +
0.0607 j0.2302 + 0.1939 j0.4885 + 0.0614 j0.1931 +
0.0598 j0.1751 + 0.0614 j0.1931 + 0.1921 j0.4970 +
=
0812_frame_C10.fm Page 281 Friday, July 20, 2001 3:57 PM
282 Distribution System Modeling and Analysis
TRANSFORMER
The transformer impedance must be converted to actual values in ohms
referenced to the lowvoltage windings.
The transformer phase impedance matrix is
The turns ratio:
The transformer ratio:
The generalized matrices are
Equation 8.26:
Equation 8.30:
Equation 8.45:
Z
base
2.4
2
1000 ⋅
2000
 2.88 = = Ω
Zt
low
0.01 j0.06 + ( ) 2.88 ⋅ 0.0288 j0.1728 + = = Ω
Zt
abc
[ ]
0.0288 j0.1728 + 0 0
0 0.0288 j0.1728 + 0
0 0 0.0288 j0.1728 +
= Ω
n
t
12.47
2.4
 5.1958 = =
a
t
12.47
3 2.4 ⋅
 2.9998 = =
a
t
[ ]
n
t
–
3

0 2 1
1 0 2
2 1 0
⋅
0 3.4639 – 1.7319 –
1.7319 – 0 3.4639 –
3.4639 – 1.7319 – 0
= =
b
t
[ ]
n
t
–
3

0 2 Zt ⋅ Zt
Zt 0 2 Zt ⋅
2 Zt ⋅ Zt 0
⋅ =
b
t
[ ]
0 −0.0998 j0.5986 – −0.0499 j0.2993 –
−0.0499 j0.2993 – 0 −0.0998 j0.5986 –
−0.0998 j0.5986 – −0.0499 j0.2993 – 0
=
c
t
[ ]
0 0 0
0 0 0
0 0 0
=
0812_frame_C10.fm Page 282 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 283
Equation 8.44:
Equation 8.35:
Equation 8.38:
Deﬁne inﬁnite bus linetoline and linetoneutral voltages:
Set the linetoneutral voltages at Node 4 equal to the nominal voltage shifted
by 30 degrees:
Deﬁne the Node 4 loads:
d
t
[ ]
1
n
t

1 1 – 0
0 1 1 –
1 – 0 1
⋅
0.1925 0.1925 – 0
0 0.1925 0.1925 –
0.1925 – 0 0.1925
= =
A
t
[ ]
1
n
t

1 0 1 –
1 – 1 0
0 1 – 1
⋅
0.1925 0 0.1925 –
0.1925 – 0.1925 0
0 0.1925 – 0.1925
= =
B
t
[ ] Zt
abc
[ ]
0.0288 j0.1728 + 0 0
0 0.0288 j0.1728 + 0
0 0 0.0288 j0.1728 +
= =
ELL
s
[ ]
12,470/30
12,470/−90
12,470/150
V =
ELN
s
[ ]
7199.6/0
7199.6/−120
7199.6/120
V =
V4 [ ]
2400/−30
2400/−150
2400/90
V =
S4 [ ]
750/31.79
1000/25.84
1250/18.19
kVA =
0812_frame_C10.fm Page 283 Friday, July 20, 2001 3:57 PM
284 Distribution System Modeling and Analysis
Start the forward sweep by computing the load currents at Node 4:
Compute the voltages and currents at Node 3:
Compute the voltages and currents at Node 2:
Compute the equivalent LN voltages and line currents at Node 1:
I4
i
S
i
1000 ⋅
V4
i

∗
312.5/−61.8
416.7/ 175.8 –
520.8/71.8
A = =
V3 [ ] a
2
[ ] V4 [ ] ⋅ b
2
[ ] I4 [ ] ⋅ +
2470.9/ 29.5 –
2534.4/ 148.4 –
2509.5/94.1
V = =
I3 [ ] c
2
[ ] V4 [ ] ⋅ = d
2
[ ] I4 [ ] ⋅ +
312.5/61.8
416.7/ 175.8 –
520.8/71.8
A =
V2 [ ] a
t
[ ] V3 [ ] ⋅ b
t
[ ] I3 [ ] ⋅ +
7956.4/3.3
7344.5/ 113.4 –
7643.0/120.5
V = =
I2 [ ] c
t
[ ] V3 [ ] ⋅ = d
t
[ ] I3 [ ] ⋅ +
118.2/ 23.5 –
150.3/ 137.8 –
148.3/88.9
A =
V1 [ ] a
1
[ ] V
2
[ ] ⋅ b
1
[ ] I2 [ ] ⋅ +
7985.9/3.4
7370.6/ 113.2 –
7673.6/120.7
V = =
I1 [ ] c
1
[ ] V2 [ ] ⋅ d
1
[ ] I2 [ ] ⋅ +
118.2/ 23.5 –
150.3/ 137.8 –
148.3/88.9
A = =
0812_frame_C10.fm Page 284 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis
285
The computed linetoline voltage at Node 1 is
Compute the magnitude of the linetoline voltage errors:
Since these errors are greater than the usual tolerance of 0.001 per unit, the
backward sweep begins. The backward sweep uses the equivalent lineto
neutral voltage at the source as the Node 1 voltage, and proceeds to Node 4
using the line currents from the forward sweep.
This completes the ﬁrst iteration. The second iteration begins by computing
the currents at the Node 4 load using the new values of the Node 4 voltages.
The forward sweep uses these new currents. The forward and backward
sweeps continue until the error at the source is less than the speciﬁed toler
ance of 0.001 perunit. After four iterations the solution has converged to a
tolerance of 0.0003 perunit. The resulting load voltages at Node 4 are
VLL1 [ ] D [ ] V1 [ ] ⋅
13,067.5/33.7
13,411.4/ 85.7 –
13,375.9/152.7
V = =
Error [ ]
pu
ELL
s
VLL1 – [ ]
12,470

0.0809
0.1086
0.0876
perunit = =
V2 [ ] A
l
[ ] ELN
s
[ ] ⋅ B
l
[ ] I2 [ ] ⋅ –
7171.1/ 0.1 –
7176.7/ 120.2 –
7169.3/119.8
V = =
V3 [ ] A
t
[ ] V2 [ ] ⋅ B
t
[ ] I3 [ ] ⋅ –
2354.0/ 31.2 –
2351.0/ 151.6 –
2349.9/87.8
V = =
V4 [ ] A
2
[ ] V3 [ ] ⋅ B
2
[ ] I4 [ ] ⋅ –
2283.7/ 31.7 –
2221.4/ 153.6 –
2261.0/83.2
V = =
V4
final
[ ]
2278.7/ 31.8 –
2199.8/ 153.5 –
2211.2/83.1
V =
0812_frame_C10.fm Page 285 Monday, October 28, 2002 10:52 AM
286 Distribution System Modeling and Analysis
On a 120V base the ﬁnal voltages are
The voltages at Node 4 are below the required 120 (121 1) V. These low
voltages can be corrected with the installation of three stepvoltage regula
tors connected in wye on the secondary bus (Node 3) of the transformer.
The new conﬁguration of the feeder is shown in Figure 10.8. For the regu
lator, the potential transformer ratio will be 2400120 V ( = 20), and the
CT ratio is selected to carry the rated current of the transformer bank. The
rated current is
The CT ratio is selected to be 1000:5 = CT = 200.
The equivalent phase impedance between Node 3 and Node 4 is computed
using the converged voltages at the two nodes. This is done so that the R
and X settings of the compensator can be determined:
The three regulators are to have the same R and X compensator settings.
The average value of the computed impedances will be used:
FIGURE 10.8
Voltage regulator added to the system.
Load
2 4
Bus
1
[ZeqS]
[ZeqL]
3r 3
V4
120
[ ]
113.9/ 31.8 –
110.0/ 153.5 –
110.6/83.1
V =
±
N
pt
I
rated
6000
3 2.4 ⋅
 832.7 = =
Zeq
i
V3
i
V4
i
–
I3
i

0.1414 j0.1829 +
0.2078 j0.2826 +
0.0889 j0.3833 +
Ω = =
Z
avg
1
3
 Zeq
k
k=1
3
∑
⋅ 0.1461 j0.2830 Ω + = =
0812_frame_C10.fm Page 286 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 287
The value of the compensator impedance in volts is given by Equation 7.81:
The value of the compensator settings in ohms is
With the regulator in the neutral position, the voltages being input to the
compensator circuit for the given conditions are
The compensator currents are
With the input voltages and compensator currents, the voltages across the
voltage relays in the compensator circuit are computed to be
Notice how close these compare to the actual voltages on a 120V base at
Node 4. Assume that the voltage level has been set at 121 V with a bandwidth
of 2 V. The regulators will change taps until the phase voltage is at least 120 V.
With the computed voltages across the relays, the approximate number of
steps that each regulator will move to can be computed by:
R′ jX′ + 0.1461 j0.2830 + ( )
1000
20
 ⋅ 7.3 j14.2 + = = V
R
Ω
jX
Ω
+
7.3
j
14.2 +
5
 1.46 j2.84 + = = Ω
Vreg
i
V3
i
PT

117.5/ 31.2 –
117.1/ 151.7 –
116.7/87.8
V = =
Icomp
i
I3
i
CT

1.6456/ 63.6 –
2.2730/ 179.4 –
2.8267/64.9
A = =
V
relay
[ ] V
reg
[ ] Z
comp
[ ] I
comp
[ ] ⋅ –
113.0/ 32.5 –
111.2/ 153.8 –
110.0/84.7
V = =
Tap
i
120 V4
i
–
0.75

9.3
11.7
13.4
= =
9
12
13
≈
0812_frame_C10.fm Page 287 Friday, July 20, 2001 3:57 PM
288 Distribution System Modeling and Analysis
With the taps set at 9, 12, and 13, the [a] matrix for the regulators is
The [d] matrix for the regulators is
The [b], [c], and [B] matrices are zero and the [A] matrix is given by:
The procedure now is to go back and perform the same modiﬁed ladder
iterative technique as before, only this time the regulators are an added
element between Nodes 3r and 3 as shown in Figure 10.8.
With the taps on their speciﬁed settings, the system converges after four
iterations with the voltages on a 120V base at Node 4:
The voltages on all phases are within the speciﬁed limits.
Example 10.2 has demonstrated how all the series components of a feeder
can be modeled and then used in the modiﬁed ladder technique to achieve
a ﬁnal solution. The example has also demonstrated how the compensator
circuit settings are determined, and then how the compensator circuit causes
the taps to change on the individual regulators so that the ﬁnal voltages at
the load center (Node 4) will be within the speciﬁed limits.
a
r
[ ]
1.0 0.00625 Tap
1
⋅ – 0 0
0 1.0 0.00625 Tap
2
⋅ – 0
0 0 1.0 0.00625 Tap
3
⋅ –
=
a
r
[ ]
0.9438 0 0
0 0.9250 0
0 0 0.9188
=
d
r
[ ] a
r
[ ]
1 –
1.0596 0 0
0 1.0811 0
0 0 1.0884
= =
A
r
[ ] a
r
[ ]
1 –
1.0596 0 0
0 1.0811 0
0 0 1.0884
= =
V4
120
[ ]
121.0/ 31.8 –
120.1/ 153.3 –
121.5/83.9
V =
0812_frame_C10.fm Page 288 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 289
10.1.6 Load Allocation
Many times the input complex power (kW and kVAr) to a feeder is known
because of the metering at the substation. This information can be either
total threephase or for each individual phase. In some cases the metered
data may be the current and power factor in each phase.
It is desirable to force the computed input complex power to the feeder to
match the metered input. This can be accomplished (following a converged
iterative solution) by computing the ratio of the metered input to the com
puted input. The phase loads can now be modiﬁed by multiplying the loads
by this ratio. Because the losses of the feeder will change when the loads are
changed, it is necessary to go through the ladder iterative process to deter
mine a new computed input to the feeder. This new computed input will be
closer to the metered input, but most likely not within a speciﬁed tolerance.
Again, a ratio can be determined and the loads modiﬁed. This process is
repeated until the computed input is within a speciﬁed tolerance of the
metered input.
Load allocation does not have to be limited to matching metered readings
at the substation. The same process can be performed at any point on the
feeder where metered data is available. The only difference is that only the
load at nodes downstream from the metered point will be modiﬁed.
10.1.7 Summary of PowerFlow Studies
This section has developed a method for performing powerﬂow studies on
a distribution feeder. Models for the various components of the feeder have
been developed in previous chapters. The purpose of this section has been
to develop and demonstrate the modiﬁed ladder iterative technique using
the generalized matrices for the series elements. It should be obvious that a
study of a large feeder with many laterals and sublaterals cannot be per
formed without the aid of a computer program.
The development of the models and examples in this text has used actual
values of voltage, current, impedance, and complex power. When perunit
values are used, it is imperative that all values be converted using a common
set of base values. In the usual application of perunit there will be a base
linetoline voltage and a base linetoneutral voltage; also, there will be a
base line current and a base delta current. For both the voltage and current
there is a squarerootofthree relationship between the two base values. In
all of the derivations of the models, and in particular those for the three
phase transformers, the square root of three has been used to relate the
difference in magnitudes between linetoline and linetoneutral voltages,
and between the line and delta currents. Because of this, when using the
perunit system, there should be only one base voltage, and that should be
the base linetoneutral voltage. When this is done, for example, the perunit
positive and negative sequence voltages will be the square root of three
times the perunit positive and negative sequence linetoneutral voltages.
0812_frame_C10.fm Page 289 Friday, July 20, 2001 3:57 PM
290 Distribution System Modeling and Analysis
Similarly, the positive and negative sequence perunit line currents will be
the square of three times the positive and negative sequence perunit delta
currents. By using just one base voltage and one base current, the perunit
generalized matrices for all system models can be determined.
10.2 ShortCircuit Studies
The computation of shortcircuit currents for unbalanced faults in a normally
balanced threephase system has traditionally been accomplished by the
application of symmetrical components. However, this method is not well
suited to a distribution feeder that is inherently unbalanced. The unequal
mutual coupling between phases leads to mutual coupling between
sequence networks. When this happens there is no advantage to using sym
metrical components. Another reason for not using symmetrical compo
nents is that the phases between which faults occur are limited. For example,
using symmetrical components, linetoground faults are limited to Phase a
to ground. What happens if a singlephase lateral is connected to Phase b
or c and the shortcircuit current is needed? This section will develop a
method for shortcircuit analysis of an unbalanced threephase distribution
feeder using the phase frame.
10.2.1 General Theory
Figure 10.9 shows the unbalanced feeder as modeled for shortcircuit cal
culations. Short circuits can occur at any one of the ﬁve points shown in
Figure 10.9. Point 1 is the highvoltage bus of the distribution substation
transformer. The values of the shortcircuit currents at Point 1 are normally
determined from a transmission system shortcircuit study. The results of
these studies are supplied in terms of the threephase and singlephase short
circuit MVAs. Using the shortcircuit MVAs, the positive and zero sequence
impedances of the equivalent system can be determined. These values are
needed for the shortcircuit studies at the other four points in Figure 10.9.
FIGURE 10.9
Unbalanced feeder shortcircuit analysis model.
[Zsys ]
ABC
[ZeqL ]
abc
[Zxfm ]
abc
[ZeqS ]
ABC
[Zsub ]
ABC
System
Voltage
Source
Equivalent
System
Impedance
Substation
Transformer
Total Primary
Line Segment
Impedance
InLine Feeder
Transformer
Total Secondary
Line Segment
Impedance
5 4 3 2 1
y
0812_frame_C10.fm Page 290 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 291
Given the threephase shortcircuit MVA magnitude and angle, the posi
tive sequence equivalent system impedance in ohms is determined by:
Ω (10.10)
Given the singlephase shortcircuit MVA magnitude and angle, the zero
sequence equivalent system impedance in ohms is determined by:
Ω (10.11)
In Equations 10.10 and 10.11, kVLL is the nominal linetoline voltage in kV
of the transmission system.
The computed positive and zero sequence impedances need to be con
verted into the phase impedance matrix as deﬁned in Equations 6.42 and
6.43 in Chapter 6.
For short circuits at points 2, 3, 4, and 5, it is going to be necessary to
compute the Thevenin equivalent threephase circuit at the shortcircuit
point. The Thevenin equivalent voltages will be the nominal linetoground
voltages with the appropriate angles. For example, assume the equivalent
system linetoground voltages are balanced threephase of nominal voltage
with the Phase a voltage at zero degrees. The Thevenin equivalent voltages
at Points 2 and 3 will be computed by multiplying the system voltages by
the generalized transformer matrix [A
t
] of the substation transformer. Car
rying this further, the Thevenin equivalent voltages at Points 4 and 5 will
be the voltages at Node 3 multiplied by the generalized matrix [A
t
] for the
inline transformer.
The Thevenin equivalent phase impedance matrices will be the sum of the
phase impedance matrices of each device between the system voltage source
FIGURE 10.10
Thevenin equivalent circuit.
+

+

+

+
+
+
+




[ZTOT]
Zf
Zf
Zf
Ifa
Ifb
Ifc
c
b
a
Faulted
Bus
Vax
Vbx
Vcx
Vxg
x
Ea Eb Ec
Z
+
kVLL
2
MVA
3 phase –
( )
∗
 =
Z
0
3 kVLL
2
⋅
MVA
1 phase –
( )
∗
 2.Z
+
– =
0812_frame_C10.fm Page 291 Friday, July 20, 2001 3:57 PM
292 Distribution System Modeling and Analysis
and the point of fault. Stepvoltage regulators are assumed to be set in the
neutral position so they do not enter into the shortcircuit calculations. Any
time a delta–grounded wye transformer is encountered, the total phase
impedance matrix on the primary side of the transformer must be referred
to the secondary side using Equation 8.160.
Figure 10.10 illustrates the Thevenin equivalent circuit at the faulted node.
3
In Figure 10.10, the voltage sources E
a
, E
b
, and E
c
represent the Thevenin
equivalent linetoground voltages at the faulted node. The matrix [ZTOT]
represents the Thevenin equivalent phase impedance matrix at the faulted
node. The fault impedance is represented by Z
f
in Figure 10.10.
Kirchhoff’s voltage law in matrix form can be applied to the circuit of
Figure 10.10:
(10.12)
Equation 10.12 can be written in compressed form as
(10.13)
Combine terms in Equation 10.13:
(10.14)
where (10.15)
Solve Equation 10.14 for the fault currents
(10.16)
where (10.17)
Since the matices [Y] and are known, deﬁne:
(10.18)
Substituting Equation 10.18 into Equation 10.16 and rearranging results in
(10.19)
E
a
E
b
E
c
Z
aa
Z
ab
Z
ac
Z
ba
Z
bb
Z
bc
Z
ca
Z
cb
Z
cc
If
a
If
b
If
c
Z
f
0 0
0 Z
f
0
0 0 Z
f
If
a
If
b
If
c
V
ax
V
bx
V
cx
V
xg
V
xg
V
xg
+ + ⋅ + ⋅ =
E
abc
[ ] ZTOT [ ] If
abc
[ ] ZF [ ] If
abc
[ ] V
abcx
[ ] V
xg
[ ] + + ⋅ + ⋅ =
E
abc
[ ] ZEQ [ ] If
abc
[ ] V
abcx
[ ] V
xg
[ ] + + ⋅ =
ZEQ [ ] ZTOT [ ] ZF [ ] + =
If
abc
[ ] Y [ ] E
abc
[ ] Y [ ] V
abcx
[ ] Y [ ] V
xg
[ ] ⋅ – ⋅ – ⋅ =
Y [ ] ZEQ [ ]
1 –
=
E
abc
[ ]
IP
abc
[ ] Y [ ] E
abc
[ ] ⋅ =
IP
abc
[ ] If
abc
[ ] Y [ ] + V
abcx
[ ] Y [ ] V
xg
[ ] ⋅ + ⋅ =
0812_frame_C10.fm Page 292 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 293
Expanding Equation 10.19:
(10.20)
Performing the matrix operations in Equation 10.19:
(10.21)
where (10.22)
Equations 10.21 become the general equations that are used to simulate all
types of short circuits. Basically, there are three equations and seven
unknowns (If
a
, If
b
, If
c
, , , , and ). The other three variables in the
equations (IP
a
, IP
b
, and IP
c
) are functions of the total impedance and the
Thevenin voltages and are therefore known. In order to solve Equations 10.21,
it will be necessary to specify four additional independent equations. These
equations are functions of the type of fault being simulated. The additional
required four equations for various types of faults are given below. These
values are determined by placing short circuits in Figure 10.11 to simulate
the particular type of fault. For example, a threephase fault is simulated by
placing a short circuit from Node a to x, Node b to x, and Node c to x. That
gives three voltage equations. The fourth equation comes from applying
Kirchhoff’s current law at Node x, which gives the sum of the fault currents
to be zero.
10.2.2 Speciﬁc Short Circuits
THREEPHASE FAULTS:
(10.23)
THREEPHASETOGROUND FAULTS:
(10.24)
IP
a
IP
b
IP
c
If
a
If
b
If
c
Y
aa
Y
ab
Y
ac
Y
ba
Y
bb
Y
bc
Y
ca
Y
cb
Y
cc
V
ax
V
bx
V
cx
Y
aa
Y
ab
Y
ac
Y
ba
Y
bb
Y
bc
Y
ca
Y
cb
Y
cc
V
xg
V
xg
V
xg
+ + ⋅ + =
IP
a
If
a
Y
aa
V
ax
⋅ Y
ab
V
bx
⋅ Y
ac
V
cx
⋅ + + ( ) Ys
a
V
xg
⋅ + + =
IP
b
If
b
Y
ba
V
ax
⋅ Y
bb
V
bx
⋅ Y
bc
V
cx
⋅ + + ( ) Ys
b
V
xg
⋅ + + =
IP
c
If
a
Y
ca
V
ax
⋅ Y
cb
V
bx
⋅ Y
cc
V
cx
⋅ + + ( ) Ys
c
V
xg
⋅ + + =
Ys
a
Y
aa
Y
ab
Y
ac
+ + =
Ys
b
Y
ba
Y
bb
Y
bc
+ + =
Ys
c
Y
ca
Y
cb
Y
cc
+ + =
V
ax
V
bx
V
cx
V
xg
V
ax
V
bx
V
cx
0 = = =
I
a
I
b
I
c
+ + 0 =
V
ax
V
bx
V
cx
V
xg
0 = = = =
0812_frame_C10.fm Page 293 Friday, July 20, 2001 3:57 PM
294 Distribution System Modeling and Analysis
LINETOLINE FAULTS (assume ij fault with phase k unfaulted):
(10.25)
LINETOGROUND FAULTS (assume phase k fault with phases i and j
unfaulted):
(10.26)
Notice that Equations 10.25 and 10.26 will allow the simulation of lineto
line faults and linetoground faults for all phases. There is no limitation to
bc faults for linetoline and ag for linetoground, as is the case when the
method of symmetrical components is employed.
A good way to solve the seven equations is to set them up in matrix form:
(10.27)
Equation 10.27 in condensed form:
(10.28)
Equation 10.28 can be solved for the unknowns in matrix [X]:
(10.29)
The blanks in the last four rows of the coefﬁcient matrix in Equation 10.27
are ﬁlled in with the known variables, depending upon what type of fault
is to be simulated. For example, the elements in the [C] matrix simulating a
threephase fault would be
All of the other elements in the last four rows will be set to zero.
V
ix
V
jx
0 = =
If
k
0 =
If
i
If
j
+ 0 =
V
kx
V
xg
0 = =
If
i
If
j
0 = =
IP
a
IP
b
IP
c
0
0
0
0
1 0 0 Y
1,1
Y
1,2
Y
1,3
Ys
1
0 1 0 Y
2,1
Y
2,2
Y
2,3
Ys
2
0 0 1 Y
3,1
Y
3,2
Y
3,3
Ys
3
– – – – – – –
– – – – – – –
– – – – – – –
– – – – – – –
If
a
If
b
If
c
V
ax
V
bx
V
cx
V
xg
⋅ =
IP
s
[ ] C [ ] X [ ] ⋅ =
X [ ] C [ ]
1 –
IP
s
[ ] ⋅ =
C
4,4
1 C
5,5
, 1 C
6,6
, 1 = = =
C
7,1
C
7,2
C
7,3
1 = = =
0812_frame_C10.fm Page 294 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 295
Example 10.3
Use the system of Example 10.2 and compute the shortcircuit currents for
a bolted (Z
f
= 0) linetoline fault between Phases a and b at Node 4.
The inﬁnite bus balanced linetoline voltages are 12.47 kV, which leads to
balanced linetoneutral voltages at 7.2 kV.
The linetoneutral Thevenin circuit voltages at Node 4 are determined using
Equation 8.159:
The Thevenin equivalent impedance at the secondary terminals (Node 3) of
the transformer consists of the primary line impedances referred across the
transformer plus the transformer impedances. Using Equation 8.160:
Note that the Thevenin impedance matrix is not symmetrical. This is a result,
once again, of the unequal mutual coupling between the phases of the primary
line segment.
The total Thevenin impedance at Node 4 is
ELL
s
[ ]
12,470/30
12,470/ 90 –
12,470/150
V =
ELN
s
[ ]
7199.6/0
7199.6/ 120 –
7199.6/120
V =
Eth
4
[ ] A
t
[ ] ELN
s
[ ] ⋅
2400/ 30 –
2400/ 150 –
2400/90
V = =
Zth
3
[ ] A
t
[ ] ZeqS
ABC
[ ] d
t
[ ] Zt
abc
[ ] + ⋅ ⋅ =
Zth
3
[ ]
0.0366 j0.1921 + 0.0039 j0.0086 – – 0.0039 – j0.0106 –
0.0039 j0.0086 – – 0.0366 j0.1886 + 0.0039 – j0.0071 –
−0.0039 j0.0106 – 0.0039 – j0.0071 – 0.0366 j0.1906 +
Ω =
Zth
4
[ ] ZTOT [ ] Zth
3
[ ] ZeqL
abc
[ ] + = =
ZTOT [ ]
0.2273 j0.6955 + 0.0568 j0.2216 + 0.0559 j0.1645 +
0.0568 j0.2216 + 0.2305 j0.6771 + 0.0575 j0.1860 +
0.0559 j0.1645 + 0.0575 j0.1860 + 0.2287 j0.6876 +
Ω =
0812_frame_C10.fm Page 295 Friday, July 20, 2001 3:57 PM
296
Distribution System Modeling and Analysis
The equivalent admittance matrix at Node 4 is
Using Equation 10.18, the equivalent injected currents at the point of fault are
The sums of each row of the equivalent admittance matrix are computed
according to Equation 10.22:
For the ab fault at Node 4, according to Equation 10.25:
The coefﬁcient matrix [
C
] using Equation 10.27:
Yeq
4
[ ] ZTOT [ ]
1 –
=
Yeq
4
[ ]
0.5031 j1.4771 – 0.1763 – j0.3907 + 0.0688 – j0.2510 +
−0.1763 j0.3907 + 0.5501 j1.5280 – 0.1148 – j0.3133 +
−0.0688 j0.2510 + 0.1145 – j0.3133 + 0.4843 j1.4532 –
S =
IP [ ] Yeq
4
[ ] Eth
4
[ ] ⋅
4466.8/ 96.4 –
4878.9/138.0
4440.9/16.4
A = =
Y
i
Yeq
i, k
k=1
3
∑
0.2580 j0.8353 –
0.2590 j0.8240 –
0.3007 j0.8889 –
S = =
If
a
If
b
+ 0 =
I
c
0 =
V
ax
0 =
V
bx
0 =
C [ ]
1 0 0 0.501 j1.477 – 0.176 j0.390 + – 0.069 0.252 + – 0.258 j0.835 –
0 1 0 0.176 j0.390 + – 0.550 j1.528 – 0.115 j0.314 + – 0.259 j0.824 –
0 0 1 0.069 j0.251 + – 0.115 j0.313 + – 0.484 j1.452 – 0.301 j0.889 –
1 1 0 0 0 0 0
0 0 1 0 0 0 0
0 0 0 0 1 0 0
0 0 0 1 0 0 0
=
0812_frame_C10.fm Page 296 Monday, October 28, 2002 10:57 AM
Distribution Feeder Analysis 297
The injected current matrix is
The unknowns are computed by:
The interpretation of the results are
Using the linetoground voltages at Node 4 and the shortcircuit currents
and working back to the source, using the generalized matrices will check
the validity of these results. The linetoground voltages at Node 4 are
IP
s
[ ]
4466.8/ 96.4 –
4878.9/138.0
4440.9/16.4
0
0
0
0
=
X [ ] C [ ]
1 –
IP
s
[ ] ⋅
8901.7/ 8.4 –
8901.7/171.6
0
7740.4/ 90.6 –
0
0
2587.9/89.1
= =
If
a
X
1
8901.7/ 8.4 – = =
If
b
X
2
8901.7/171.6 = =
If
c
X
3
0 = =
V
ax
X
4
7740.4/ 90.6 – = =
V
bx
X
5
0 = =
V
cx
X
6
0 = =
V
xg
X
7
2587.9/89.1 = =
VLG
4
[ ]
V
ax
V
xg
+
V
bx
V
xg
+
V
cx
Vxg +
5153.4/ 90.4 –
2587.2/89.1
2587.2/89.1
V = =
0812_frame_C10.fm Page 297 Friday, July 20, 2001 3:57 PM
298 Distribution System Modeling and Analysis
The shortcircuit currents in matrix form:
The linetoground voltages at Node 3 are
The equivalent linetoneutral voltages and line currents at the primary
terminals (Node 2) of the transformer are
Finally, the equivalent linetoneutral voltages at the inﬁnite bus can be
computed:
These are the same equivalent linetoneutral voltages that were used to start
the shortcircuit analysis.
10.3 Summary
This chapter has taken the component models of a distribution feeder and
developed techniques for powerﬂow and shortcircuit analyses. The tech
niques do not lend themselves to hand calculations, but have been developed
with a computer program in mind. In developing the models and analyses
I
4
[ ] I
3
[ ]
8901.7/ 8.4 –
8901.7/171.6
0
A = =
VLG
3
[ ] a
2
[ ] VLG
4
[ ] b
1
[ ] I
4
[ ] ⋅ + ⋅
3261.1/ 63.4 –
1544.3/161.7
2430.9/89.9
V = =
VLN
2
[ ] a
t
[ ] VLG
3
[ ] b
t
[ ] I
3
[ ] ⋅ + ⋅
6766.3/ 6.4 –
6833.7/ 119.6 –
7480.3/116.6
V = =
I
2
[ ] d
t
[ ] I
3
[ ] ⋅
3426.4/ 8.4 –
1713.2/171.6
1713.2/171.6
A = =
VLN
1
[ ] a
1
[ ] VLN
2
[ ] b
1
[ ] I
2
[ ] ⋅ + ⋅
7199.6/0
7199.6/ 120 –
7199.6/120
V = =
0812_frame_C10.fm Page 298 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 299
techniques, the importance of modeling the system components exactly
has been emphasized. Because of the unbalanced nature of a distribution
feeder, without exact models the results of the analyses are suspect.
The examples in this chapter have been very long and should be used as
learning tools. Many of the interesting operating characteristics of a feeder
can only be demonstrated through numerical examples. The examples were
designed to illustrate some of these characteristics.
Armed with a computer program, using the models and techniques of this
text provides the engineer with powerful tools for solving presentday prob
lems and performing longrange planning studies.
References
1. Trevino, C., Cases of difﬁcult convergence in loadﬂow problems, IEEE Paper
n. 7162PWR, Presented at the IEEE Summer Power Meeting, Los Angeles,
1970.
2. Kersting, W.H. and Mendive, D.L., An application of ladder network theory
to the solution of threephase radial loadﬂow problems, IEEE Conference Paper,
Presented at the IEEE Winter Power Meeting, New York, January 1976.
3. Kersting, W.H. and Phillips, W.H., Distribution system shortcircuit analysis, 25th
Intersociety Energy Conversion Engineering Conference, Reno, NV, August 12–17,
1990.
Problems
The powerﬂow problems in this set require an iterative solution. Students
should be encouraged to write their own computer program to solve the
problems.
The ﬁrst six problems of this set will be based upon the system in Figure 10.11.
FIGURE 10.11
Wye homework system.
1 2 3
4
Infinite
Bus
0812_frame_C10.fm Page 299 Friday, July 20, 2001 3:57 PM
300 Distribution System Modeling and Analysis
The substation transformer is connected to an inﬁnite bus with balanced
threephase voltages of 69 kV. The substation transformer is rated:
5000 kVA, 69 kV delta − 4.16 grounded wye, Z = 1.5 + j8.0%
The phase impedance matrix for a fourwire wye line is
The secondary voltages of the substation transformer are balanced and being
held at 4.16 kV for all powerﬂow problems.
The fourwire wye feeder is 0.75 miles long. An unbalanced wyeconnected
load is located at Node 3 and has the following values:
Phase a: 750 kVA at 0.85 lagging power factor
Phase b: 500 kVA at 0.90 lagging power factor
Phase c: 850 kVA at 0.95 lagging power factor
The load at Node 4 is zero initially.
10.1 For the system as described above and assuming that the regulators
are in the neutral position.
(1) Determine the generalized matrices for the line segment.
(2) Use the modiﬁed ladder technique to determine the linetoground
voltages at Node 3. Use a tolerance of 0.001 perunit. Give the
voltages in actual values in volts and on a 120volt base.
10.2 Three Type B stepvoltage regulators are installed in a wye connection at
the substation in order to hold the load voltages at a voltage level of 121 V and
a bandwidth of 2 V.
(1) Compute the actual equivalent line impedance between the sub
station and the load node.
(2) Use a potential transformer ratio of 2400120 V and a current trans
former ratio of 500:5 A. Determine the R and X compensator set
tings calibrated in volts and ohms. The settings must be the same
for all three regulators.
(3) For the load conditions of Problem 10.1 and with the regulators in
the neutral position, compute the voltages across the voltage relays
in the compensator circuits.
z
4 wire –
[ ]
0.4576 j1.0780 + 0.1560 j0.5017 + 0.1535 j0.3849 +
0.1560 j0.5017 + 0.4666 j1.0482 + 0.1580 j0.4236 +
0.1535 j0.3849 + 0.1580 j0.4236 + 0.4615 j1.0651 +
Ω/mile =
0812_frame_C10.fm Page 300 Friday, July 20, 2001 3:57 PM
Distribution Feeder Analysis 301
(4) Determine the appropriate tap settings for the three regulators to
hold the Node 3 voltages at 121 V in a bandwidth of 2 V.
(5) With the regulator taps set, compute the load voltages on a 120V
base.
10.3 A wyeconnected threephase shunt capacitor bank of 300 kvar per
phase is installed at Node 3. With the regulator compensator settings from
Problem 10.2, determine:
(1) The new tap settings for the three regulators.
(2) The voltages at the load on a 120V base.
(3) The voltages across the relays in the compensator circuits.
10.4 The load at Node 4 is served through an ungrounded wye–delta trans
former bank. The load is connected in delta with the following values:
Phase ab: 400 kVA at 0.9 factor power factor
Phase bc: 150 kVA at 0.8 lagging power factor
Phase ca: 150 kVA at 0.8 lagging power factor
The three singlephase transformers are rated as:
Lighting transformer: 500 kVA, 2400−240 V, Z = 0.9 + j3.0%
Power transformers: 167 kVA, 2400 −240 V, Z = 1.0 + j1.6%
Use the original loads and the shunt capacitor bank at Node 3 and this new
load at Node 4. Determine:
(1) The voltages on a 120V base at Node 3, assuming the regulators
are in the neutral position.
(2) The voltages on a 120V base at Node 4, assuming the regulators
are in the neutral position.
(3) The new tap settings for the three regulators.
(4) The Node 3 and 4 voltages on a 120V base after the regulators
have changed tap positions.
10.5 Under shortcircuit conditions the inﬁnite bus voltage is the only volt
age that is constant. The voltage regulators in the substation are in the neutral
position. Determine the shortcircuit currents and voltages at Nodes 1, 2, 3
for the following shortcircuits at Node 3:
(1) Threephase to ground
(2) Phase b to ground
(3) Linetoline fault on Phases ac
0812_frame_C10.fm Page 301 Friday, July 20, 2001 3:57 PM
302 Distribution System Modeling and Analysis
10.6 A linetoline fault occurs at Node 4. Determine the currents in the
fault and on the line segment between Nodes 2 and 3. Determine the voltages
at Nodes 1, 2, 3, and 4.
10.7 A threewire delta line of length 0.75 miles is serving an unbalanced
delta load of:
Phase ab: 600 kVA, 0.9 lagging power factor
Phase bc: 800 kVA, 0.8 lagging power factor
Phase ca: 500 kVA, 0.95 lagging power factor
The phase impedance matrix for the line is
The line is connected to a constant balanced voltage source of 4.8 kV line
toline. Determine the load voltages on a 120V base.
10.8 Add two Type B stepvoltage regulators in an open delta connection,
using phases AB and BC, to the system in Problem 10.7. The regulator
should be set to hold 121 ± 1 V. Determine the R and X settings and the ﬁnal
tap settings. For the open delta connection, the R and X settings will be
different on the two regulators.
10.9 The threewire line of Problem 10.6 is connected to a substation trans
former connected deltadelta. The substation transformer is connected to a
69kV inﬁnite bus and is rated:
10,000 kVA, 69 kV delta − 4.8 kV delta, Z = 1.6 + j7.8%
Determine the shortcircuit currents and substation transformer secondary
voltages for the following short circuits at the end of the line:
(1) Threephase
(2) Linetoline between phases ab
z
3 wire –
[ ]
0.4013 j1.4133 + 0.0953 j0.8515 + 0.0953 j0.7802 +
0.0953 j0.8515 + 0.4013 j1.4133 + 0.0953 j0.7266 +
0.0953 j0.7802 + 0.0953 j0.7266 + 0.4013 j1.4133 +
Ω/mile =
0812_frame_C10.fm Page 302 Friday, July 20, 2001 3:57 PM
303
Appendix A
Conductor Data
Size Stranding Material
DIAM
Inches
GMR
Feet
RES
ΩΩΩ Ω
/mile
Capacity
Amps
1 ACSR 0.355 0.00418 1.38 200
1 7 STRD Copper 0.328 0.00992 0.765 270
1 CLASS A AA 0.328 0.00991 1.224 177
2 6/1 ACSR 0.316 0.00418 1.69 180
2 7 STRD Copper 0.292 0.00883 0.964 230
2 7/1 ACSR 0.325 0.00504 1.65 180
2 AWG SLD Copper 0.258 0.00836 0.945 220
2 CLASS A AA 0.292 0.00883 1.541 156
3 6/1 ACSR 0.281 0.0043 2.07 160
3 AWG SLD Copper 0.229 0.00745 1.192 190
4 6/1 ACSR 0.25 0.00437 2.57 140
4 7/1 ACSR 0.257 0.00452 2.55 140
4 AWG SLD Copper 0.204 0.00663 1.503 170
4 CLASS A AA 0.232 0.007 2.453 90
5 6/1 ACSR 0.223 0.00416 3.18 120
5 AWG SLD Copper 0.1819 0.0059 1.895 140
6 6/1 ACSR 0.198 0.00394 3.98 100
6 AWG SLD Copper 0.162 0.00526 2.39 120
6 CLASS A AA 0.184 0.00555 3.903 65
7 AWG SLD Copper 0.1443 0.00468 3.01 110
8 AWG SLD Copper 0.1285 0.00416 3.8 90
9 AWG SLD Copper 0.1144 0.00371 4.6758 80
10 AWG SLD Copper 0.1019 0.00330 5.9026 75
12 AWG SLD Copper 0.0808 0.00262 9.3747 40
14 AWG SLD Copper 0.0641 0.00208 14.8722 20
16 AWG SLD Copper 0.0508 0.00164 23.7262 10
18 AWG SLD Copper 0.0403 0.00130 37.6726 5
19 AWG SLD Copper 0.0359 0.00116 47.5103 4
20 AWG SLD Copper 0.032 0.00103 59.684 3
22 AWG SLD Copper 0.0253 0.00082 95.4835 2
24 AWG SLD Copper 0.0201 0.00065 151.616 1
1/0 ACSR 0.398 0.00446 1.12 230
1/0 7 STRD Copper 0.368 0.01113 0.607 310
1/0 CLASS A AA 0.368 0.0111 0.97 202
2/0 ACSR 0.447 0.0051 0.895 270
2/0 7 STRD Copper 0.414 0.01252 0.481 360
2/0 CLASS A AA 0.414 0.0125 0.769 230
3/0 12 STRD Copper 0.492 0.01559 0.382 420
(
Continued
)
0812_App A.fm Page 303 Friday, July 20, 2001 2:15 PM
304
Distribution System Modeling and Analysis
Conductor Data (continued)
Size Stranding Material
DIAM
Inches
GMR
Feet
RES
ΩΩΩ Ω
/mile
Capacity
Amps
3/0 6/1 ACSR 0.502 0.006 0.723 300
3/0 7 STRD Copper 0.464 0.01404 0.382 420
3/0 CLASS A AA 0.464 0.014 0.611 263
3/8 INCH STE Steel 0.375 0.00001 4.3 150
4/0 12 STRD Copper 0.552 0.0175 0.303 490
4/0 19 STRD Copper 0.528 0.01668 0.303 480
4/0 6/1 ACSR 0.563 0.00814 0.592 340
4/0 7 STRD Copper 0.522 0.01579 0.303 480
4/0 CLASS A AA 0.522 0.0158 0.484 299
250,000 12 STRD Copper 0.6 0.01902 0.257 540
250,000 19 STRD Copper 0.574 0.01813 0.257 540
250,000 CON LAY AA 0.567 0.0171 0.41 329
266,800 26/7 ACSR 0.642 0.0217 0.385 460
266,800 CLASS A AA 0.586 0.0177 0.384 320
300,000 12 STRD Copper 0.657 0.0208 0.215 610
300,000 19 STRD Copper 0.629 0.01987 0.215 610
300,000 26/7 ACSR 0.68 0.023 0.342 490
300,000 30/7 ACSR 0.7 0.0241 0.342 500
300,000 CON LAY AA 0.629 0.0198 0.342 350
336,400 26/7 ACSR 0.721 0.0244 0.306 530
336,400 30/7 ACSR 0.741 0.0255 0.306 530
336,400 CLASS A AA 0.666 0.021 0.305 410
350,000 12 STRD Copper 0.71 0.0225 0.1845 670
350,000 19 STRD Copper 0.679 0.0214 0.1845 670
350,000 CON LAY AA 0.679 0.0214 0.294 399
397,500 26/7 ACSR 0.783 0.0265 0.259 590
397,500 30/7 ACSR 0.806 0.0278 0.259 600
397,500 CLASS A AA 0.724 0.0228 0.258 440
400,000 19 STRD Copper 0.726 0.0229 0.1619 730
450,000 19 STRD Copper 0.77 0.0243 0.1443 780
450,000 CON LAG AA 0.77 0.0243 0.229 450
477,000 26/7 ACSR 0.858 0.029 0.216 670
477,000 30/7 ACSR 0.883 0.0304 0.216 670
477,000 CLASS A AA 0.795 0.0254 0.216 510
500,000 19 STRD Copper 0.811 0.0256 0.1303 840
500,000 37 STRD Copper 0.814 0.026 0.1303 840
500,000 CON LAY AA 0.813 0.026 0.206 483
556,500 26/7 ACSR 0.927 0.0313 0.1859 730
556,500 30/7 ACSR 0.953 0.0328 0.1859 730
556,500 CLASS A AA 0.858 0.0275 0.186 560
600,000 37 STRD Copper 0.891 0.0285 0.1095 940
600,000 CON LAY AA 0.891 0.0285 0.172 520
605,000 26/7 ACSR 0.966 0.0327 0.172 760
605,000 54/7 ACSR 0.953 0.0321 0.1775 750
636,000 27/7 ACSR 0.99 0.0335 0.1618 780
636,000 30/19 ACSR 1.019 0.0351 0.1618 780
636,000 54/7 ACSR 0.977 0.0329 0.1688 770
636,000 CLASS A AA 0.918 0.0294 0.163 620
0812_App A.fm Page 304 Friday, July 20, 2001 2:15 PM
Appendix A
305
Conductor Data (continued)
Size Stranding Material
DIAM
Inches
GMR
Feet
RES
ΩΩΩ Ω
/mile
Capacity
Amps
666,600 54/7 ACSR 1 0.0337 0.1601 800
700,000 37 STRD Copper 0.963 0.0308 0.0947 1040
700,000 CON LAY AA 0.963 0.0308 0.148 580
715,500 26/7 ACSR 1.051 0.0355 0.1442 840
715,500 30/19 ACSR 1.081 0.0372 0.1442 840
715,500 54/7 ACSR 1.036 0.0349 0.1482 830
715,500 CLASS A AA 0.974 0.0312 0.145 680
750,000 37 STRD AA 0.997 0.0319 0.0888 1090
750,000 CON LAY AA 0.997 0.0319 0.139 602
795,000 26/7 ACSR 1.108 0.0375 0.1288 900
795,000 30/19 ACSR 1.14 0.0393 0.1288 910
795,000 54/7 ACSR 1.093 0.0368 0.1378 900
795,000 CLASS A AA 1.026 0.0328 0.131 720
0812_App A.fm Page 305 Friday, July 20, 2001 2:15 PM
0812_App A.fm Page 306 Friday, July 20, 2001 2:15 PM
307
Appendix B
Concentric Neutral 15 kV Cable
Conductor
Size
AWG or kcmil
Diameter
over
Insulation
Inches
Diameter
over
Screen
Inches
Outside
Diameter
Inches
Copper
Neutral
No.
××× ×
AWG
Ampacity
UG Duct
Amps
Full Neutral
2(7
×
) 0.78 0.85 0.98 10
×
14 120
1(19
×
) 0.81 0.89 1.02 13
×
14 135
1/0(19
×
) 0.85 0.93 1.06 16
×
14 155
2/0(19
×
) 0.90 0.97 1.13 13
×
12 175
3/0(19
×
) 0.95 1.02 1.18 16
×
12 200
4/0(19
×
) 1.01 1.08 1.28 13
×
10 230
250(37
×
) 1.06 1.16 1.37 16
×
10 255
350(37
×
) 1.17 1.27 1.47 20
×
10 300
1/3 Neutral
2(7
×
) 0.78 0.85 0.98 6
×
14 135
1(19
×
) 0.81 0.89 1.02 6
×
14 155
1/0(19
×
) 0.85 0.93 1.06 6
×
14 175
2/0(19
×
) 0.90 0.97 1.10 7
×
14 200
3/0(19
×
) 0.95 1.02 1.15 9
×
14 230
4/0(19
×
) 1.01 1.08 1.21 11
×
14 240
250(37
×
) 1.06 1.16 1.29 13
×
14 260
350(37
×
) 1.17 1.27 1.39 18
×
14 320
500(37
×
) 1.29 1.39 1.56 16
×
12 385
750(61
×
) 1.49 1.59 1.79 15
×
10 470
1000(61
×
) 1.64 1.77 1.98 20
×
10 550
0812_App B.fm Page 307 Friday, July 20, 2001 2:17 PM
308
Distribution System Modeling and Analysis
TapeShielded 15 kV Cable
Tape Thickness
=
5 mils
Conductor
Size
AWG or
kcmil
Diameter
over
Insulation
Inches
Diameter
over
Screen
Inches
Jacket
Thickness
mils
Outside
Diameter
Inches
Ampacity in
UG Duct
Amps
1/0 0.82 0.88 80 1.06 165
2/0 0.87 0.93 80 1.10 190
3/0 0.91 0.97 80 1.16 215
4/0 0.96 1.02 80 1.21 245
250 1.01 1.08 80 1.27 270
350 1.11 1.18 80 1.37 330
500 1.22 1.30 80 1.49 400
750 1.40 1.48 110 1.73 490
1000 1.56 1.66 110 1.91 565
0812_App B.fm Page 308 Friday, July 20, 2001 2:17 PM
309
Index
A
Allocation, load, 20
American National Standards Institute
(ANSI), 145
Approximate line segment model, 136–140
Autotransformers
perunit impedance, 158–161
ratings, 156–158
and voltage, 152–156
Average demand, 12, 14
C
Cables
concentric neutral, 96–101, 115–119, 307
tapeshielded, 101–105, 119–121, 308
Carson, John, 79–81
Circuit breakers and distribution substations,
4–5
Closed deltaconnected regulators, 180–183
Combination loads, 254, 258
Compensator settings, 9, 168–174
Conductors
and Carson's equations, 79–81, 83–85
and concentric neutral cables, 96–101,
115–119
and conductor resistance, 41–42
data, 303–305
and Kron reduction, 114–115
and overhead lines, 111–115
and sequence admittance, 121
and shunt admittance of lines, 109–122
tables, 8
and tapeshielded cables, 101–105,
119–121
and underground lines, 115–121
and voltagedrop, 110
Constant current loads, 253, 258
Constant impedance loads, 253, 258
Constant real and reactive power loads,
252–253, 257
Currents
and deltadelta connection, 224–235
and deltagrounded wye stepdown
connection, 206–212
and linear networks, 270–271
loads, constant, 253, 258
and nonlinear networks, 271–274
and open wyeopen delta connection,
236–242
serving a deltaconnected load, 259
shortcircuit, 290–298
and threephase induction motors,
261–266
and twophase and singlephase loads,
259
and ungrounded wyedelta stepdown
connection, 212–222
and wyeconnected loads, 254–256
Customer demand, 13–14
and load survey, 21–25
D
Deltaconnected loads
combination, 258
and constant current loads, 258
and constant impedance loads, 258
and constant real and reactive power
loads, 257
line currents serving, 259
Deltaconnected regulators, 180–193
Deltadelta connection, 224–235
Deltagrounded wye stepdown connection
currents, 206–212
voltages, 201–206
Demand, load
deﬁnitions of, 11–12, 13–14
and distribution transformer loading,
15–20
diversiﬁed, 12, 16–17
and diversity factor, 18, 21, 27–31
factor, 19
0812_Frame_IDX.fm Page 309 Tuesday, July 31, 2001 10:50 AM
310
Distribution System Modeling and Analysis
and Kirchhoff's current law (KCL), 25
and load survey, 21–25
and metered feeder maximum demand,
25–27
noncoincident, 12, 17, 24
and transformers, 23–27
Distribution feeders
approximate line segment model, 136–140
electrical characteristics, 8–9
exact line segment model, 125–132
general, 274–276
and K factors, 43–47
and ladder iterative technique, 270–274,
279–288
and linear networks, 270–271
and line impedance, 41–42
maps, 6–8
modiﬁed line model, 132–136
and nonlinear networks, 271–274
powerﬂow analysis, 269–290
and power loss, 50–52
rectangle load conﬁguration, 55–60
series components, 276–278
and shortcircuit studies, 290–298
shunt components, 278–279
unbalanced threephase, 276–279,
290–298
and uniformly distributed loads, 47–54
and voltage drop, 39–41
Distribution systems
approximate line segment model, 136–140
compensator settings, 9
conductor tables, 8
design, 1
exact line segment model, 125–132
feeders, 6–9
lines, 6–7
modiﬁed line model, 132–136
as power system components, 1–2
radial feeders, 5–6
and shunt capacitors, 8
substations, 2–5
switches, 8
transformers, 4, 8, 9, 15–20
and untransposed distribution lines,
79–81
and voltage regulators, 3, 8
Diversiﬁed demand, 12, 16–17
and Kirchhoff's current law (KCL), 25
Diversity factor, 12, 18, 21
application of, 27–31
Duration curve, load, 17
E
Exact line segment model, 125–132
Exact lumped load model, 52–54
F
Feeders, distribution, 5–8
load, 20–32
and metered feeder maximum demand,
25–27
G
Geometric conﬁgurations for lumping loads
rectangular, 55–60
trapezoidal, 65–71
triangular, 60–65
Geometric mean distances (GMD), 91–92
Grounded wyegrounded wye connection,
222–224
I
Impedance
and Carson's equations, 81–85
and concentric neutral cable, 96–101
and geometric mean distances (GMD),
91–92
and Kron reduction, 86–89, 94
loads, constant, 253, 258
phase matrix for overhead lines, 86–89,
94–95
primitive matrix for overhead lines,
85–86, 94
self and mutual, 77–82, 93
sequence, 89–95
and transposed threephase lines, 78–79
and untransposed distribution lines,
79–81
Induction motor, threephase, 261–266
Inductive reactance, 77–78
and transposed threephase lines, 78–79
Inline transformers, 8
0812_Frame_IDX.fm Page 310 Tuesday, July 31, 2001 10:50 AM
Index
311
K
K factors, 43–47, 64–65
Kirchhoff's current law (KCL), 25, 80,
125–132, 270–271
Kron reduction, 86–89, 94, 114–115
KW demand, 13–14
L
Ladder iterative technique
applying, 279–288
and linear networks, 270–271
and nonlinear networks, 271–274
and shunt components, 278–279
Linear networks, 270–271
Line currents,
See
Currents
Line drop compensators, 168–174
Line impedance, 41–42
and Carson's equations, 81–85
and exact lumped load model, 52–54
and K factors, 43–47
of overhead lines, 77–95
and phase impedance, 86–89, 94–95
and primitive impedance, 85–86, 94
rectangular load conﬁguration, 58
and sequence impedances, 89–95, 121
and transposed threephase lines, 78–79,
92–95
and untransposed distribution lines,
79–81
Load
allocation, 20, 289
allocation based upon transformer
ratings, 31–32
and average demand, 14
combination, 254, 258
constant current, 253, 258
constant impedance, 253, 258
deﬁnitions of, 11–12
deltaconnected, 257–259
demand, 11–12
and distribution transformer loading,
15–20
diversity, 20
and diversity factor, 12, 18
duration curve, 17
exact lumped model, 52–54
factor, 19
feeder, 20–32
individual customer, 13–14
and K factors, 43–47, 64–65
line currents serving deltaconnected, 259
and load factor, 12, 14
lumping in geometric conﬁgurations,
55–71
management, transformer, 25
and maximum demand, 11, 13–14
models, 251–266
and power loss, 50–52
rectangular conﬁguration, 55–60
and shunt capacitors, 259–261, 278–279
survey, 21–25
and threephase induction motors,
261–266
trapezoidal conﬁguration, 65–71
triangular conﬁguration, 60–65, 69–70
twophase and singlephase, 259
uniformly distributed, 47–54
and utilization factor, 12, 19
and voltagedrop calculations, 27, 39–41,
48–50, 61–62, 68–69
wyeconnected, 252–256
M
Maps, distribution feeder, 6–8
Maximum demand, 11, 13–14, 17
Maximum system voltage, 145
Metering of distribution substations, 4
Modiﬁed line model, 132–136
Mutual impedance of conductors, 77–82, 93
N
Nominal system voltage, 145
Nominal utilization voltage, 146
Noncoincident demand, 12, 17, 24
Nonlinear networks, 271–274
O
Open deltaconnected regulators, 183–193
Open wyeopen delta connection, 236–242
Overhead lines
and Carson's equations, 81–85
and conductors, 111–115
exact line segment model, 125–132
and geometric mean distances (GMD),
91–92
and Kron reduction, 114–115
0812_Frame_IDX.fm Page 311 Tuesday, July 31, 2001 10:50 AM
312
Distribution System Modeling and Analysis
phase impedance matrix, 86–89, 129–131
primitive impedance matrix, 85–86
sequence impedances, 89–95
and transposed threephase lines, 78–79
and untransposed distribution lines,
79–81
P
Perunit impedance, 158–161
Phase impedance matrix for overhead lines,
86–89, 94–95, 129–131, 139–141
Powerﬂow analysis
description of, 269
and distribution feeder series
components, 276–278
and general distribution feeders, 274–276
and ladder iterative technique, 270–274,
279
and linear networks, 270–271
and load allocation, 289
and nonlinear networks, 271–274
and shunt components, 278–279
studies, 289–290
and unbalanced threephase distribution
feeders, 276–279
Power loss, 50–52
exact lumped load model, 52–54
rectangular load conﬁguration, 57, 70
trapezoidal load conﬁguration, 69–70
triangular load conﬁguration, 63, 65,
70–71
Power systems components, 1–2
Primitive impedance matrix for overhead
lines, 85–86, 94
Protection of distribution substations, 3
R
Radial distribution feeders, 5–6
Ranges, voltage, 146
Rectangular load conﬁguration, 55–60, 69
Reduction, Kron, 86–89
S
Self impedance of conductors, 77–82, 93
Sequence admittance, 121
Sequence impedances, 89–95, 121, 132–136
Series impedance of overhead lines
and Carson's equations, 79–81, 83–85
and inductive reactance, 77–78
and Kron reduction, 86–89, 94
and phase impedance matrix, 86–89,
94–95
and primitive impedance matrix, 85–86,
94
sequence impedances, 89–95
and transposed threephase lines, 78–79,
92–93
and untransposed distribution lines,
79–81
Series impedance of underground lines
and concentric neutral cables, 96–101
general conﬁguration, 95–96
and tapeshielded cables, 101–105
Service voltage, 145
Shortcircuit studies
general theory, 290–293
of speciﬁc faults, 293–298
Shunt admittance of lines
concentric neutral cable underground,
115–119
and conductors, 109–110
and exact line segment model, 125–132
and modiﬁed line model, 132–136
overhead, 111–115
and perunit impedance, 158–161
and sequence admittance, 121
tapeshielded underground, 119–121
underground, 115–121
and voltagedrop, 110
Shunt capacitors, 8
components, 278–279
deltaconnected, 260–261
wyeconnected, 259–260
Singlephase stepvoltage regulators
generalized constants, 167
and line drop compensators, 168–174
type A, 163–164
type B, 164–167
Standard voltage ratings, 145–147
Stepvoltage regulators
closed deltaconnected, 180–183
description of, 162–163
generalized constants, 167
and line drop compensators, 168–174
open deltaconnected, 183–193
singlephase, 163–174
threephase, 174–193
type A, 163–164
type B, 164–167
wyeconnected, 175–180
0812_Frame_IDX.fm Page 312 Tuesday, July 31, 2001 10:50 AM
Index
313
Substations, distribution
and circuit breakers, 4–5
and distribution feeder maps, 6–8
and highside and lowside switching, 3, 8
and line drop compensators, 170–174
metering, 4
protection, 3
and radial feeders, 5–6
and voltage regulation, 3, 8
and voltage transformation, 3, 8, 9
Switching, highside and lowside, 3, 8
System voltage, 145
T
Tapeshielded cables, 101–105, 119–121, 308
Thevenin equivalent circuit, 242–245, 295
Threephase distribution lines, 78–79, 92–93
Threephase induction motors, 261–266
Threephase stepvoltage regulators
closed deltaconnected, 180–183
description of, 174–175
open deltaconnected, 183–193
wyeconnected, 175–180
Threephase transformer models
and currents, 206–212
and deltadelta connection, 224–235
and deltagrounded wye stepdown
connection, 201–212
description of, 199–200
generalized matrices, 200–201
and grounded wyegrounded wye
connection, 222–224
and open wyeopen delta connection,
236–242
and Thevenin equivalent circuit, 242–245
and ungrounded wyedelta stepdown
connection, 212–222
and voltages, 200–206
Transformers, 4, 8, 9
auto, 152–162
banks, 199–200
and currents, 206–212
and deltadelta connection, 224–235
and deltagrounded wye stepdown
connection, 201–212
distribution loading, 15–20
and diversiﬁed demand, 16–17
and diversity factor application, 27–31
and grounded wyegrounded wye
connection, 222–224
and ladder iterative technique, 279–288
and line drop compensators, 168–174
load management, 25
and load survey, 23–25
and metered feeder maximum demand,
25–27
and open wyeopen delta connection,
236–242
perunit impedance, 158–161
ratings, 31–32
and Thevenin equivalent circuit, 242–245
threephase generalized matrices,
200–201
twowinding autotransformers, 152–161
and twowinding transformer theory,
147–151
and ungrounded wyedelta stepdown
connection, 212–222
and voltagedrop calculations, 27
Transposed threephase lines, 78–79, 92–93
Trapezoidal load conﬁguration, 65–71
Triangular load conﬁguration, 60–65, 69–70
Twophase and singlephase loads, 259
Twowinding autotransformer, 152–161
Twowinding transformer theory, 147–151
Type A stepvoltage regulators, 163–164
Type B stepvoltage regulators, 164–167
U
Unbalanced threephase distribution feeders,
276–279
Underground lines
and concentric neutral cables, 96–101,
115–121
exact line segment model, 125–132
general conﬁguration, 95–96
and sequence admittance, 121
Ungrounded wyedelta stepdown
connection, 212–222
Uniformly distributed loads, and voltage
drop, 47–50
Untransposed distribution lines, 79–81
Utilization factor, 12, 19
Utilization voltage, 145
V
Voltage
and application of diversity factor, 27–31
and approximate line segment model,
139–141
and autotransformer ratings, 156–158
0812_Frame_IDX.fm Page 313 Tuesday, July 31, 2001 10:50 AM
314
Distribution System Modeling and Analysis
and closed deltaconnected regulators,
180–183
and deltadelta connection, 224–235
and deltagrounded wye stepdown
connection, 201–212
drop, 27, 39–41, 47–50, 56, 58–60, 61–62,
68–69, 110
and exact line segment model, 125–132
and exact lumped load model, 52–54
and general distribution feeders, 274–276
generalized regulator constants, 167
and grounded wyegrounded wye
connection, 222–224
and K factors, 43–47
and ladder iterative technique, 270–274,
279–288
and linear networks, 270–271
and line drop compensators, 168–174
and load allocation based upon
transformer ratings, 31–32
maximum system, 145
and modiﬁed line model, 132–136
nominal system, 145
nominal utilization, 146
and nonlinear networks, 271–274
and open deltaconnected regulators,
183–193
and open wyeopen delta connection,
236–242
and perunit impedance, 158–161
ranges, 146
and rectangular load conﬁguration, 56,
58–60
regulation, 3, 8, 145, 147, 167
regulators
singlephase step, 162–174
threestep, 174–193
service, 145
and shortcircuit studies, 290–298
and shunt admittance of lines, 110
standard ratings, 145–147
system, 145
and Thevenin equivalent circuit, 242–245
transformation, 3
and transformers, 147–151
and trapezoidal load conﬁguration, 68–69
and triangular load conﬁguration, 61–62
and twowinding transformer theory,
147–151
and type A stepvoltage regulators,
163–164
and type B stepvoltage regulators,
163–164
and ungrounded wyedelta stepdown
connection, 212–222
and uniformly distributed loads, 47–54
utilization, 145
and wyeconnected regulators, 175–180
W
Wyeconnected loads
combination, 254
and constant current loads, 253
and constant impedance loads, 253
and constant real and reactive power
loads, 252–253
currents, 254–256
Wyeconnected regulators, 175–180
0812_Frame_IDX.fm Page 314 Tuesday, July 31, 2001 10:50 AM
Distribution System Modeling and Analysis
The ELECTRIC POWER ENGINEERING Series The ELECTRIC POWER ENGINEERINGSeries series editor Leo Grigsy series editor Leo Grigsby
Published Titles
Electromechanical Systems, Electric Machines, and Applied Mechatronics Sergey E. Lyshevski Electrical Energy Systems Mohamed E. ElHawary Electric Drives Ion Boldea and Syed Nasar Distribution System Modeling and Analysis William H. Kersting Linear Synchronous Motors: Transportation and Automation Systems Jacek Gieras and Jerry Piech
Forthcoming Titles
Induction Machine Handbook Ion Boldea and Syed Nasar Power System Operations in a Restructured Business Environment Fred I. Denny and David E. Dismukes Power Quality C. Sankaran
Distribution System Modeling and Analysis
William H. Kersting
New Mexico State University Las Cruces, New Mexico
CRC Press Boca Raton London New York Washington, D.C.
Library of Congress CataloginginPublication Data
Kersting, William H. Distribution system modeling and analysis / William H. Kersting p. cm.  (Electric power engineering series) Includes bibliographical references and index. ISBN 0849308127 (alk. paper) 1. Electric power distribution–Mathematical models. I. Title. II. Series. TK3001 .K423 2001 621.31—dc21
2001035681 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.
Visit the CRC Press Web site at www.crcpress.com
© 2002 by CRC Press LLC No claim to original U.S. Government works International Standard Book Number 0849308127 Library of Congress Card Number 2001035681 Printed in the United States of America 1 2 3 4 5 6 7 8 9 0 Printed on acidfree paper
Contents
1 Introduction to Distribution Systems........................................... 1
1.1 1.2 1.3 1.4 1.5 1.6 The Distribution System......................................................................2 Distribution Substations ......................................................................2 Radial Feeders.......................................................................................5 Distribution Feeder Map .....................................................................6 Distribution Feeder Electrical Characteristics..................................8 Summary................................................................................................9
2 The Nature of Loads ..................................................................... 11
2.1 2.2 Deﬁnitions............................................................................................ 11 Individual Customer Load................................................................13 2.2.1 Demand ...................................................................................13 2.2.2 Maximum Demand................................................................13 2.2.3 Average Demand....................................................................14 2.2.4 Load Factor .............................................................................14 2.3 Distribution Transformer Loading...................................................15 2.3.1 Diversiﬁed Demand...............................................................16 2.3.2 Maximum Diversiﬁed Demand...........................................17 2.3.3 Load Duration Curve ............................................................17 2.3.4 Maximum Noncoincident Demand ....................................17 2.3.5 Diversity Factor ......................................................................18 2.3.6 Demand Factor .......................................................................19 2.3.7 Utilization Factor....................................................................19 2.3.8 Load Diversity ........................................................................20 2.4 Feeder Load .........................................................................................20 2.4.1 Load Allocation ......................................................................20 2.4.1.1 Application of Diversity Factors ...........................21 2.4.1.2 Load Survey..............................................................21 2.4.1.3 Transformer Load Management............................25 2.4.1.4 Metered Feeder Maximum Demand ....................25 2.4.1.5 What Method to Use? .............................................27 2.4.2 VoltageDrop Calculations Using Allocated Loads..........27 2.4.2.1 Application of Diversity Factors ...........................27 2.4.2.2 Load Allocation Based upon Transformer Ratings ................................................31 2.5 Summary..............................................................................................32 Problems.......................................................................................... 33
.......1 Voltage Drop ............................................................3 The Trapezoid .............................................. 115 5................................ 119 5..1 Concentric Neutral Cable .....................................................1...................................5..........................................................46 3....................60 3........................47 3..............122 References ....................................................1 The Rectangle..............................7 Sequence Impedances.............................................................................................2 Series Impedance of Underground Lines......................2 Overhead Lines .122 Problems...........3 Carson’s Equations .............................................................................95 4....................................2 The Krise Factor.....121 5...................4................................5 Lumping Loads in Geometric Conﬁgurations ..81 4.........................................................79 4...........1 Transposed ThreePhase Lines....................5 Sequence Admittance...............5......1 3...71 References ....................................................... 105 4...............................................................6 Phase Impedance Matrix for Overhead Lines ......48 3........................................5..................................................................4 TapeShielded Cable Underground Lines .......3.....................1................... 77 Series Impedance of Overhead Lines.........................1 The General VoltageDrop Equation ....3 Summary.....................1 5 Shunt Admittance of Overhead and Underground Lines ...............................................................................................105 Problems..........1................................................77 4...55 3............................2.................................1........................3 The Exact Lumped Load Model.............6 Summary..............1.............................83 4.................3 4 Series Impedance of Overhead and Underground Lines ................................89 4........3 Concentric Neutral Cable Underground Lines ............... 122 ..... 109 5...............................5 Primitive Impedance Matrix for Overhead Lines ...................................................................................................... 39 Voltage Drop...................................4..105 References .....................................................................................65 3...........................3................................................. 71 3............................1...... 111 5.............50 3...........................................................................................................2 TapeShielded Cables..............96 4...............................................................43 3..1 The Kdrop Factor...86 4..................................55 3...................4 Uniformly Distributed Loads .......................85 4...............................................39 Line Impedance..........................2 3....................................101 4..................4.41 “K” Factors ....4 Modiﬁed Carson’s Equations.......................................................... 110 5.6 Summary.......2.....................2 Power Loss ......................43 3...........71 Problems...............52 3....................................................................2 The Triangle ...78 4................................2 Untransposed Distribution Lines .....................................3 Approximate Methods of Analysis .1............
...................1............1 Autotransformer Ratings ..................................................2 ThreePhase StepVoltage Regulators ...........................245 Problems.....2 Type B StepVoltage Regulator...199 Generalized Matrices .......................................183 7.........................................156 7...163 7.............................................6 Distribution System Line Models ...........1 WyeConnected Regulators.2.............................4.................................................................4 StepVoltage Regulators......194 Problems...................................1..... 199 8................................167 7......164 7..........................3........................2 7.3 Introduction ...................2 Currents ......................224 8.................2 The Modiﬁed Line Model ............................222 8.............132 6....4.........................1 7............................................2................................145 TwoWinding Transformer Theory ..........................................4 The Ungrounded Wye–Delta StepDown Connection ...............1 SinglePhase StepVoltage Regulators ........................................6 The Delta–Delta Connection........193 References ..........9 Summary..........................2 PerUnit Impedance...............4........................2 8..5 Summary...4......................................................................................212 8...................................4 The Line Drop Compensator .......3 Standard Voltage Ratings ...........................................1 8..........................................................................................125 6.....................................3................................................................... 194 8 ThreePhase Transformer Models ...............................................................141 Problems.............1..200 The Delta–Grounded Wye StepDown Connection..........................4.3.......................................................147 The TwoWinding Autotransformer...................136 6............ 245 .........................7 The Open Wye–Open Delta Connection ........................................174 7.......8 The Thevenin Equivalent Circuit................... 141 7 Regulation of Voltages ....242 8....................................................202 8....1 Exact Line Segment Model ....2...158 7.....3 The Approximate Line Segment Model .............................175 7..201 8.4................................168 7...........3 Open DeltaConnected Regulators .1........................................................................3....2 Closed DeltaConnected Regulators.180 7......4 Summary.5 The Grounded Wye–Grounded Wye Connection.......................206 8......................................................................4...........................................................................................................................152 7..4........................ 145 7.....3 Generalized Constants ...............................236 8.......................... 125 6............................162 7................................1 Voltages.............141 References ........................................................1 Type A StepVoltage Regulator....................4.........163 7....................
..................................................261 References ....................................2........................................2........266 Problems........................................................1................................................257 9................ 299 Appendix A ...........269 10............1 Series Components ..........................1 Constant Real and Reactive Power Loads......................1 Linear Network ...1.......290 10...........................259 9.................9 Load Models.....1 General Theory....2..........................................2........279 10.3 TwoPhase and SinglePhase Loads ...5 Line Currents Serving a DeltaConnected Load.........253 9....4................2 The General Feeder .298 References ............258 9................293 10......1 Constant Real and Reactive Power Loads..........................................1 WyeConnected Loads .....................2 ShortCircuit Studies ...................1.......................................................................................257 9................3 Summary.......252 9....3.259 9.................3 The Unbalanced ThreePhase Distribution Feeder ............................253 9.......................1............4 Combination Loads......................... 269 10.....................................................................................................................3.....1.........................................5 The ThreePhase Induction Motor....290 10..............................................1.....7 Summary of PowerFlow Studies ........................................270 10.......2........................1......................................252 9................259 9.6 Load Allocation .............................1..278 10........259 9...............................279 10....................271 10.........258 9.........................................1 The Ladder Iterative Technique ..4.............1.....................................................................................289 10..........................................1.............................2 Shunt Components ...............................................1...2 Nonlinear Network..........................3 Constant Current Loads...................................................1................................1.................258 9..................2 Speciﬁc Short Circuits ..............1.......................................270 10.............................260 9......................................4 Combination Loads..2 Constant Impedance Loads..........................299 Problems.............276 10........ 303 Appendix B ......................4 Shunt Capacitors...................1 WyeConnected Capacitor Bank......4 Applying the Ladder Iterative Technique ....1.........................................................1...................... 266 10 Distribution Feeder Analysis...........................2 Constant Impedance Loads.....................276 10.........................................254 9................289 10..........................1.........5 Putting It All Together ...................2.................................2 DeltaConnected Loads . 309 ........274 10..........................2................................................................................3 Constant Current Loads...........................................................2 DeltaConnected Capacitor Bank ................................1 PowerFlow Analysis .......... 307 Index ..................... 251 9...................................
if any. the distribution engineer now has the needed tools. While this works ﬁne for interconnected systems. The tools also provide an opportunity for the distribution engineer to do such things as optimize capacitor placement in order to minimize losses. it is possible for a user to prepare incorrect data that will lead to results that do not make any sense.Preface In the last 40 years many papers and textbooks devoted to the computer modeling and analysis of large power system networks have been written. attention was devoted to the distribution system and its major components. and shortcircuit studies provide the necessary data for the development of a reliable coordinated protection plan for fuses. In recent years more attention has been devoted to the computer modeling and analysis of distribution systems. For the most part the models and analysis techniques have been developed for large interconnected transmission systems and synchronous generators. reclosers. Without these tools the distribution engineer has been left in the dark (no pun intended) as to the operating characteristics of distribution feeders. Little. garbage out” is the answer. In short. Computer programs are now available so that the distribution engineer can develop a real feel for how the distribution system is operating. and therefore threephase models of all the components must be employed. With the tools. The models and analyses assume a balance so that only a singlephase equivalent model is required. A lot of “seat of the pants” engineering has had to take place in order to keep the lights on. it is not sufﬁcient for the modeling and analysis of a distribution system. Most power systems textbooks and courses are limited to the modeling and analysis of balanced threephase systems. There is a signiﬁcant difference between the computer programs developed for interconnected system studies and the programs developed for distribution systems. The user must fully understand the models and analysis techniques of the program. So what is the problem? “Garbage in. Different switching scenarios for normal and emergency conditions can be simulated. serious design errors and operational procedures may result. The data requirements for the . As a result. Armed with a commercially available computer program. powerﬂow studies can be run to simulate present loading conditions and to help with the longrange planning of new facilities. and relay/circuit breakers. the distribution engineer has not had the same number of tools as the systems engineer to analyze the distribution system under steadystate (powerﬂow) and fault (shortcircuit) conditions. Without an understanding of the models and a general “feel” for the operating characteristics of a distribution system. A distribution system is inherently unbalanced.
garbage out” scenario. Starting with a review of basic transformer theory. Chapter 4 is a very important chapter in developing the exact model of line segments. but to also teach a “feel” for what the answers should be. A summary of the intent of each chapter follows. intensive to the point that most of them cannot be worked easily without using a computing tool such as MathcadTM. The models developed are in the form of generalized matrices similar to those developed for line segments. For many years there has been a need for a textbook to assist the student and distribution engineer in developing a basic understanding of the modeling and operating characteristics of the major components of a distribution system. unbalanced line model.distribution system models are more extensive. This textbook assumes that the student has a basic understanding of transformers. This chapter is limited to the threephase.” The attempt here is to make the student understand that the load on a distribution system is constantly changing. except that it is limited to shunt admittance calculations. Chapter 7 addresses voltage regulation. A word of warning: most of the problems are very number intensive. With this knowledge it will be possible to prevent the “garbage in. Chapter 1 introduces the basic components of a distribution system. Each chapter will have a series of homework problems that will assist the student in applying the models and developing a better understanding of the operating characteristics of the component being modeled. Chapter 2 is a discussion of “load. In fact. The example problems should be studied very carefully. For that reason a quick review of the theory is presented as needed. and symmetrical components. Both overhead and underground lines are included. . Chapter 3 presents some helpful approximate analysis techniques that will help the student know what “ballpark” answers to look for when more precise studies are made. There are many example problems throughout the text. transmission lines. Included is an introduction to the type of data that is necessary to model a distribution system. Students are urged to learn how to use this very powerful program. These examples are intended to not only demonstrate the application of the models. the chapter moves to the development of threephase models of stepvoltage regulators and their control. In many universities all of these topics are crammed into a onesemester course. Chapter 6 develops the ﬁrst of the generalized matrices that will be used to model the major components of a distribution system. much of the necessary data may not be readily available. and that this must be taken into account in all studies. electric machines. They are also encouraged to write their own simple computer programs for many of the problems. Chapter 5 is in many ways a continuation of Chapter 4. How to take into account the unbalanced loading and unsymmetrical conﬁgurations in the calculation of line impedances is presented in great detail.
Windmil Milsoft Integrated Solutions.edu/~wkerstin/ 2. the threephase model for shortcircuit studies is developed and demonstrated. Box 3903 Las Cruces. Chapter 10 puts it all together. Box 7526 Abilene. again.edu Homepage: www. The models. P. Two student version software packages are available. Power Consultants P.com .com/whpower or www. The ladder iterative technique is developed and demonstrated. are in the form of generalized matrices.O.com Homepage: www. Also.zianet. NM 88003 (505) 6462434 Email: wkerstin@nmsu. All of the component models developed in earlier chapters are put together to form a model of a distribution feeder.Chapter 8 develops comprehensive models of several of the standard threephase transformer connections that are common on a distribution system. Students and professors are encouraged to acquire one or both.milsoft. Inc. Chapter 9 develops the models for the various types of loads on a distribution system. TX 79608 Email: support@milsoft. The packages available are 1. Radial Distribution Analysis Package (RDAP) W.nmsu.O. H.
He joined the faculty at New Mexico State University (NMSU) in 1962 and is currently Professor of Electrical Engineering and Director of the Electric Utility Management Program. Prior to joining NMSU. Kersting received his BSEE degree from New Mexico State University (NMSU). Professor Kersting is a Fellow of the Institute of Electrical and Electronics Engineers. he received the Edison Electric Institute’s Power Engineering Educator Award in 1979 and the NMSU Westhafer Award for Excellence in Teaching in 1977. He is also a partner in W. Power Consultants. he was employed as a distribution engineer by the El Paso Electric Company. . Professor Kersting has been an active member of the IEEE Power Engineering Education and Power Engineering Committees. Las Cruces. H.The Author William H. and his MSEE degree from the Illinois Institute of Technology.
Their suggestions were very helpful. Leonard Bohmann and his students at Michigan Tech for reviewing and correcting the manuscript. as I have taught this material without the aid of a textbook. I want to thank Dr. Their positive attitudes toward the material and what I was trying to accomplish have gone a long way toward making this text possible.Acknowledgments I would be remiss if I didn’t acknowledge the patience that my students have displayed over the past many years. The students have had to live with taking notes and/or deciphering lastminute notes distributed in class. Leo Grigsby for his encouragement and review of the manuscript. My thanks also to Dr. I would like to dedicate this book to my loving wife Joanne for her encouragement and love that has made all of this possible. She spent many lonely evenings practicing the piano as I sat pounding out the text and/or yelling at the computer. .
.
2. Times have changed. Meanwhile. Once the models have been developed. Transmission lines crisscrossed the land forming large interconnected networks. The purpose of this text is to develop accurate models for all of the major components of a distribution system. Power plants became larger and larger. As a direct result. What can be done to make the distribution system operate more efﬁciently? All of these questions can be answered only if the distribution system can be modeled very accurately. Of these components. What is the maximum capacity? How do we determine this capacity? What are the operating limits that must be satisﬁed? What can be done to operate the distribution system within the operating limits? 5. 3. distribution systems were typically overdesigned.1 Introduction to Distribution Systems The major components of an electric power system are shown in Figure 1. 4. 1 . the distribution system has traditionally been characterized as the most unglamorous component. Some of the questions that need to be answered are 1.1. The operation of these large interconnected networks required the development of new analysis and operational techniques. It has become very important and necessary to operate a distribution system at its maximum capacity. In the last half of the twentieth century the design and operation of the generation and transmission components presented many challenges to practicing engineers and researchers. analysis techniques for steadystate and shortcircuit conditions will be developed. the distribution systems continued to deliver power to the ultimate user’s meter with little or no analysis.
Although Figure 1. in which case there is likely no subtransmission system. 1. which means that there is only one path for power to ﬂow from the distribution substation to the user. In some cases the distribution substation is fed directly from a highvoltage transmission line. the feeders are radial. 1. it illustrates the major components that will be found in all substations.2 Distribution Substations A diagram of a very simple oneline distribution substation is shown in Figure 1.1 Major power system components. Each distribution substation will serve one or more primary feeders. Subtransmission Line Disconnect Switch Fuse Transformer Voltage Regulator Meters Circuit Breakers Primary Feeders FIGURE 1.2 displays the simplest distribution substation.2.1 The Distribution System The distribution system typically starts with the distribution substation that is fed by one or more subtransmission lines. This varies from company to company.2 Distribution System Modeling and Analysis Interconnected Generation Transmission System Bulk Power Substation Subtransmission Network Distribution Substation Primary Feeders FIGURE 1.2 Simple distribution substation. With a rare exception. .
The LTC changes the taps on the lowvoltage windings of the transformer as the load varies. In Figure 1. Highside and lowside switching: in Figure 1.16 kV.4 kV.2 the highvoltage switching is done with a simple switch.47 kV. There are many “standard” distribution voltage levels. 23. These are used when the source voltage is always either above or below the nominal voltage.Introduction to Distribution Systems 1. Protection: the substation must be protected against the occurrence of short circuits.9 kV.and lowvoltage buses. Voltage transformation: the primary function of a distribution substation is to reduce the voltage to the distribution voltage level. Individual feeder circuit breakers or reclosers are used to provide interruption of short circuits that occur outside the substation. Many substation transformers will have “ﬁxed taps” on the highvoltage winding.2 the voltage is regulated by a “steptype” regulator that will alter the voltage plus or minus 10% on the lowside bus. As substation designs become more complex. Many times. More extensive substations may use highvoltage circuit breakers in a variety of highvoltage bus designs. the only automatic protection against short circuits inside the substation is by way of the highside fuses on the transformer. The lowvoltage switching in the ﬁgure is accomplished with relaycontrolled circuit breakers. each feeder will have its own regulator. Other substation designs will call for two or more threephase transformers. As is the case with the highvoltage bus. the voltage at the substation needs to change as the load changes. Some of the common ones are 34. The substation transformers can be threephase units or three singlephase units connected in a standard connection.2 only one transformer is shown.5 kV. The ﬁxed tap settings can alter the voltage plus or minus 5%. Sometimes this function is accomplished with a “load tap changing” (LTC) transformer. and.2. 4. the lowvoltage bus can take on a variety of designs. In many cases reclosers will be used in place of the relay/circuit breaker combination. the high. 12. the voltage drop between the substation and the user will vary. This can be in the form of a threephase gangoperated regulator or individual phase regulators that operate independently.2 kV. and any other piece of equipment. 13. in older systems. 4. 3. Some substation designs will include a lowvoltage bus circuit breaker in addition to the circuit breakers for each feeder. 2. In the simple design of Figure 1. instead of a bus regulator. Voltage regulation: as the load on the feeders varies. more extensive protective schemes will be employed to protect the transformer. In Figure 1. 14. In order to maintain the user’s voltages within an acceptable range. 3 .
C. N. Should one of the subtransmission lines go out of service. and 1 hour. FD3 N. . 30 minutes. 2. serves four distribution feeders.4 Distribution System Modeling and Analysis 5.3.C. A more comprehensive substation layout is shown in Figure 1. Metering: every substation has some form of metering. power. Now both transformers are served from the same subtransmission line.3 has two loadtap changing transformers. T2 N. then breakers X. and is fed from two subtransmission lines. With that breaker arrangement.4. N. These meters record the minimum. 4 FD2 5 FD4 6 FIGURE 1. all four feeders are served by transformer T2.C. average. voltage. Y. FD1 N.C. then breaker X or Y is opened and breaker Z is closed. and 4 are opened and breakers 2 and 5 are closed. if Transformer T1 is out of service. Typical time ranges are 15 minutes. Y Z T1 N.C. This may be as simple as an analog ammeter displaying the present value of substation current. 1. 1. Digital recording meters are becoming very common. Under normal conditions the circuit breakers (CB) are in the following positions: Circuit breakers closed: X.3 Twotransformer substation with breakerandahalf scheme. power factor. The substation in Figure 1.O. 1 2 3 N. The digital meters may monitor the output of each substation transformer and/or the output of each feeder.3. For example. N.C. each transformer is served from a different subtransmission line and serves two feeders. as well as the minimum and maximum currents that have occurred over a speciﬁc time period. The transformers are sized such that each transformer can supply all four feeders under an emergency operating condition.5 With the breakers in their normal positions. and maximum values of current. The lowvoltage bus arrangement Line 1 Line 2 X N. over a speciﬁed time range.O.6 Circuit breakers open: Z.O. etc.
This means that threephase models of the major components must be utilized. An additional unbalance is introduced by the nonequilateral conductor spacings of threephase overhead and underground line segments. Components of the feeder may consist of the following: 1.3 Radial Feeders Radial distribution feeders are characterized by having only one path for power to ﬂow from the source (distribution substation) to each customer. 3. The connecting points of the components will be referred to as “nodes. Threephase models for the major components will be developed in the following chapters. If a distribution engineer is to be able to perform accurate powerﬂow and shortcircuit studies. Figure 1. They will be developed in the “phase frame” rather than applying the method of symmetrical components. twophase (“V” phase).4 shows a simple “oneline” diagram of a threephase feeder. It is up to the substation design engineer to create a design that provides the ﬁve basic functions and yields the most reliable service economically possible. 2. it is imperative that the distribution feeder be modeled as accurately as possible.Introduction to Distribution Systems 5 is referred to as a “breakerandahalf scheme” since three breakers are required to serve two feeders. 6. 4. conventional powerﬂow and shortcircuit programs used for transmission system studies are not adequate. Threephase primary “main” feeder Threephase. twophase. 5. 7. 8. Such programs display poor convergence characteristics for radial systems. The programs also assume a perfectly balanced system so that a singlephase equivalent system is used. and singlephase loads The loading of a distribution feeder is inherently unbalanced because of the large number of unequal singlephase loads that must be served.” Note that . A typical distribution system will be composed of one or more distribution substations consisting of one or more feeders. and singlephase laterals Steptype voltage regulators Inline transformers Shunt capacitor banks Distribution transformers Secondaries Threephase. Figure 1.4 illustrates the major components of a distribution system. Because of the nature of the distribution system. There is an unlimited number of substation conﬁgurations possible. 1.
Phasing .4 Simple distribution feeder. and to be able to play the “what if” scenarios of future changes to the feeder. Lines (overhead and underground) a. Details i. 1.4 Distribution Feeder Map The analysis of a distribution feeder is important to an engineer in order to determine the existing operating conditions of a feeder. Where b. Conductor sizes (not shown on this map) ii. This is important if the most accurate models are to be developed.5. The map of Figure 1. Distances c. the phasing of the line segments is shown. a detailed map of the feeder must be available. Before the engineer can perform the analysis of a feeder.5 contains most of the following information: 1. A sample of such a map is shown in Figure 1.6 Distribution System Modeling and Analysis Substation Transformer Voltage Regulator Singlephase lateral b c a b c a b c abc b "V" phase lateral c a Threephase lateral Capacitor bank Fuse Distribution transformer b Node Underground cables Secondary Inline transformer Customers FIGURE 1.
.5 123node test feeder.Introduction to Distribution Systems 7 800' 25 325' 50 c 50 275' 400' 300' 50 50 700' 525' 275' 50 50 550' bca 475' 675' 50 50 25 250' 50 a 225' 475' 175' a c acb c 50 b 250' 25 325' 50 b 50 50 350' 50 a 25 c b acb 100' 50 b 400' 700' 50 275' 200' c 275' 200' 310 0 b 25 450' b 25 25 300' 575' 325' 50 a 250 ' a M 175' 50 a 275' 275' a c b 25 50 50 225' 450' 575' 5 a 32 ' ' 225' 275 c ' 250 350' 550' c 250' 175' 350 50 250' b 225' b 350' 25 50 a b 50 50 25 b 275' 750' 50 300' c 225' 25 10 00' 100 25 25 b 425' 350 325' bca 275' a 275' 50 a c b 650' 200' 300' 25 325' 100 cba 350' 25 b 250' b b 250' 125' 200' 50 b 25 25 800' a 50 25 a 25 25 250' 650' 325' 300' 50 b 400' c 150' 350' 375 ' 50 25 250' 500' 25 300' ba 50 100' 50 c 250' 50 b a 25 350 250' cba b c 300' c 25 b 250' 225' 25 325' 50 c 200' 250' c b a 250' 375' 825' 200' a 200' 250' 25 250' 225' a 200' 25 25 b 300' 50 b 300' 350 150' 3100 200' 300' 250' 50 425' b 250' 250' 25 275' c 525' 325' 250' 550' 25 b 50 a 175' 50 a ca 50 50 200' 350' abc 400' b c 350' 50 50 275' 50 c 225' 50 25 25 300' Subs a 500' tation 50 ThreePhase OH ThreePhase UG TwoPhase OH OnePhase OH 50 1Phase Transformer kVA 3Phase Transformer Bank 350 Voltage Regulator FIGURE 1.
Overhead and underground spacings 2. Phase connection 5. Resistance (Ω/mile) 3. Normal open/close status 1. Electrical characteristics for each device will have to be determined before the analysis of the feeder can commence.8 Distribution System Modeling and Analysis 2. Switches a. Inline transformers a. the following data must be available: 1. Distribution transformers a. Connection 4. In order to determine the electrical characteristics. Voltage regulators a. kVA rating c. Location b.) b. Conductor tables a. Potential transformer ratios b. Voltage regulators a. Location b. Current transformer ratios . kvar rating c. Phase connection 3. Location b. Phase connection c. Shunt capacitors a. Geometric mean radius (GMR) (ft. Singlephase ii. Diameter (inches) c. Threephase 6. Type (not shown on this map) i.5 Distribution Feeder Electrical Characteristics Information from the map will deﬁne the physical location of the various devices. Location b. kVA rating c. Location b.
Sometimes the most difﬁcult task for the engineer is to acquire all of the necessary data. Compensator settings i. Additional data such as standard pole conﬁgurations. but. speciﬁc conductors used on each line segment. Bandwidth iii. the substation serves one or more radial feeders. . in volts (V) 4. Impedance (R and X) d. Noload power loss 9 1. and voltage regulator settings must come from stored records. Once all of the data has been acquired.Introduction to Distribution Systems c. threephase transformer connections. Each feeder must be modeled as accurately as possible in order for the analysis to have meaning. Voltage level ii.6 Summary It is becoming increasingly more important to be able to accurately model and analyze distribution systems. There are many different substation designs possible. Voltage ratings c. R and X settings. Transformers a. Feeder maps will contain most of the needed data. for the most part. the analysis can begin utilizing models of the various devices that will be developed in later chapters. kVA rating b.
.
The closer you are to the customer. or A • Must include the time interval • Example: the 15minute kW demand is 100 kW 2. period. Every time a light bulb or an electrical appliance is switched on or off. Demand • Load averaged over a speciﬁc period of time • Load can be kW. the load seen by the distribution feeder changes. Maximum Demand • Greatest of all demands that occur during a speciﬁc time • Must include demand interval. and units • Example: the 15minute maximum kW demand for the week was 150 kW 11 . kVA. 2. In order to describe the changing load.1 Deﬁnitions The load that an individual customer or a group of customers presents to the distribution system is constantly changing. For example. What is load? The answer to that question depends upon what type of an analysis is desired. The problem is that the load on a power system is constantly changing. There is no such thing as a “steadystate” load. the following terms are deﬁned: 1. kvar. In order to come to grips with load. the steadystate analysis (powerﬂow study) of an interconnected transmission system will require a different deﬁnition of load than that used in the analysis of a secondary in a distribution feeder.2 The Nature of Loads The modeling and analysis of a power system depend upon the load. it is ﬁrst necessary to look at the load of an individual customer. the more pronounced will be the everchanging load.
and units • Example: the 15minute diversiﬁed kW demand in the period ending at 9:30 was 200 kW 5. and units • Example: the 15minute maximum diversiﬁed kW demand for the week was 500 kW 6.12 Distribution System Modeling and Analysis 3. and units • Example: the 15minute average kW demand for the month was 350 kW 4. Load Diversity • Difference between maximum noncoincident demand and the maximum diversiﬁed demand . Maximum Noncoincident Demand • For a group of loads. period. the sum of the individual maximum demands without any restriction that they occur at the same time • Must include demand interval. Diversity Factor • Ratio of the maximum noncoincident demand to the maximum diversiﬁed demand 11. Utilization Factor • Ratio of the maximum demand to rated capacity 9. Maximum Diversiﬁed Demand • Maximum of the sum of the demands imposed by a group of loads over a particular period • Must include demand interval. etc. month. period. period. week. Load Factor • Ratio of the average demand of any individual customer or group of customers over a period to the maximum demand over the same period 10. Diversiﬁed Demand • Sum of demands imposed by a group of loads over a particular period • Must include demand interval. Average Demand • The average of the demands over a speciﬁed period (day. and units • Example: the maximum noncoincident 15minute kW demand for the week was 700 kW 7. period.) • Must include demand interval. Demand Factor • Ratio of maximum demand to connected load 8.
For this customer the 15minute maximum kW demand occurs at 13:15 and has a value of 6.2.1 Customer demand curve.1 illustrates how the instantaneous kW load of a customer changes during two 15minute intervals. 6. 2. The 24hour 15minute kW demand curve for a customer is shown in Figure 2.2. In each interval the average value of the demand is determined. In Figure 2.0 4. the more accurate will be the value of the load. .0 Instantaneous 15 Minute kW Demand 5.0 3.0 1. This curve is developed from a spreadsheet that gives the 15minute kW demand for a period of 24 hours. 2. This particular customer has three periods in which the kW demand exceeds 6.18 kW.0 6:15 6:30 6:45 Time of Day FIGURE 2.1 Demand In order to deﬁne the load.1 the straight lines represent the average load in a time interval.2 Maximum Demand The demand curve shown in Figure 2.2 Individual Customer Load Figure 2.1 the selected time interval is 15 minutes. The greatest of these is the 15minute maximum kW demand. Note that during the 24hour period there is a great variation in the demand. In Figure 2.2. Each bar depicts the 15minute kW demand. This process is very similar to numerical integration. The average value of the load in an interval is deﬁned as the 15minute kW demand.0 kW.2 represents a typical residential customer.0 2. the demand curve is broken into equal time intervals. The shorter the time interval.The Nature of Loads 13 2.
15min kW demand 6.2. From the utility’s standpoint. The energy in kWh used during each 15minute time interval is computed by: 1 kWh = ( 15min kW demand ) ⋅ . Sometimes utility companies will encourage industrial customers to improve their load factors. the optimal load factor would be 1. the total energy consumed during the period by Customer #1 is 58.3) .= . In many ways load factor gives an indication of how well the utility’s facilities are being utilized.14 Distribution System Modeling and Analysis FIGURE 2.18 (2.2) 2. From the spreadsheet.40 Max.= .2 the load factor is computed to be Average 15min kW demand 2. 2.2 24hour demand curve for Customer #1.1) The total energy consumed during the day is the summation of all of the 15minute interval consumptions.46 Load factor = . For Customer #1 in Figure 2.hour 4 (2.3 Average Demand During the 24hour period.2.46 kW Hours 24 (2. It is deﬁned as the ratio of the average demand to the maximum demand.96 Average demand = . One method of encouragement is to penalize the customer on the electric bill for having a low load factor.00 since the system has to be designed to handle the maximum demand.4 Load Factor “Load factor” is a term that is often used when describing a load. The 15minute average kW demand is computed by: Total energy 58.= 0.96 kWh. energy (kWh) will be consumed.= 2.
2.3.3 24hour demand curve for Customer #2. the peaks and valleys and maximum demands will be different for each customer. Figures 2. FIGURE 2. . Each customer will have a demand curve similar to that in Figure 2.1. These four customers demonstrate that there is great diversity among their loads. However.3 Distribution Transformer Loading A distribution transformer will provide service to one or more customers. The load curves for the four customers show that each customer has his unique loading characteristic. 2.The Nature of Loads 15 2. A summary of individual loads is given in Table 2. FIGURE 2. Customer #3 is the only one who will have a high load factor.5 give the demand curves for the three additional customers connected to the same distribution transformer.4 24hour demand curve for Customer #3.4. The customers’ individual maximum kW demand occurs at different times of the day. and 2.
22 Cust. #1 Energy Usage (kWh) Maximum kW Demand Time of Max.44 0. The 15minute diversiﬁed kW demand of the transformer for the day is shown in Figure 2. Note how the demand curve is beginning to smooth out. #3 95. #4 42. TABLE 2.6. FIGURE 2.16 Distribution System Modeling and Analysis FIGURE 2. .57 6. #2 36.18 13:15 2.81 Cust.05 20:30 1.1 Individual Customer Load Characteristics Cust. There are not as many signiﬁcant changes as in some of the individual customer curves. kW Demand Average kW Demand Load Factor 58.46 6.6 Transformer diversiﬁed demand curve.3.25 2.1 Diversiﬁed Demand It is assumed that the same distribution transformer serves the four customers discussed previously.5 24hour demand curve for Customer #4.82 11:30 1.64 4. in this case. the distribution transformer.98 0.52 0.40 Cust. and. The sum of the four 15 kW demands for each time interval is the diversiﬁed demand for the group in that time interval.78 0.93 6:45 3.75 7.
The greater of these is the 15minute maximum diversiﬁed kW demand of the transformer.The Nature of Loads 2.3.93 + 7. the load duration curve shows the transformer operates with a 15minute kW demand of 12 kW or greater 22% of the time. Note that this maximum demand does not occur at the same time as any one of the individual demands.18 + 6.05 = 24. Sorting in descending order.2 Maximum Diversiﬁed Demand 17 The transformer demand curve of Figure 2.3.82 + 4. This curve can be used to determine whether a transformer needs to be replaced due to an overloading condition. the sum of the individual maximums is Max. For the transformer in question.3. noncoincident demand = 6. the 15minute kW demand exceeds 16 kW twice. 2. The load duration curve plots the 15minute kW demand versus the percent of time the transformer operates at or above the speciﬁc kW demand.6 demonstrates how the combined customer loads begin to smooth out the extreme changes of the individual loads. 2. It occurs at 17:30 and has a value of 16. nor is this maximum demand the sum of the individual maximum demands.4) FIGURE 2. the kW demand of the transformer develops the load duration curve shown in Figure 2. For example.3 Load Duration Curve A load duration curve can be developed for the transformer serving the four customers.16 kW.7. For the transformer.4 Maximum Noncoincident Demand The 15minute maximum noncoincident kW demand for the day is the sum of the individual customer 15minute maximum kW demands.98 kW (2. .7 Transformer load duration curve.
13 3.84 2.15 N 51 52 53 54 55 56 57 58 59 60 DF 3.12 3.18 3.13 N 41 42 43 44 45 46 47 48 49 50 DF 3. then the maximum diversiﬁed demand of a group of customers can be computed.82 2.65 N 11 12 13 14 15 16 17 18 19 20 DF 2.10 3.16 3.0 1.8 that the value of the diversity factor basically leveled out when the number of customers reached 70.16 3.20 from 70 customers up.10 2.05 3.88 2. the diversity factor for the four customers would be Maximum noncoincident demand 24.19 3. The table was developed from a different database than the four customers discussed previously.= . If there are ﬁve customers.17 3.67 2.15 3.20 .5458 Maximum diversified demand 16.15 3. as viewed from the substation.18 2.05 N 31 32 33 34 35 36 37 38 39 40 DF 3.60 1.01 3.92 2.15 3.90 N 21 22 23 24 25 26 27 28 29 30 DF 2.17 3.06 3. This is an important observation because it means.2 and Figure 2.80 2.40 2.18 N 61 62 63 64 65 66 67 68 69 70 DF 3.11 3. A graph of the diversity factors is shown in Figure 2. This process would have to be repeated for all practical numbers of customers.13 3.20 3.14 3.96 2.18 3. diversity factor is the ratio of the maximum noncoincident demand of a group of customers to the maximum diversiﬁed demand of the group.60 2.10 3. then a load survey would have to be set up to determine the diversity factor for ﬁve customers.16 (2.5 Diversity Factor Distribution System Modeling and Analysis By deﬁnition.04 3. In other words.90 2. Note in Table 2.= 1.8. With reference to the transformer serving four customers.14 3.2.86 2.5) The idea behind the diversity factor is that when the maximum demands of the customers are known.78 2. the maximum diversiﬁed demand of a feeder can be predicted by computing the total noncoincident maximum demand of all of the customers served by the feeder and dividing by 3.98 Diversity factor = .17 3.15 3.16 3.74 2.15 3.80 2.00 3.70 2.19 3. at least for the system from which these diversity factors were determined.08 3.18 3. Table 2.19 3.09 3.2 is an example of the diversity factors for the number of customers ranging from one to 70.30 2. that the diversity factor will remain constant at 3.98 3. There will be a different value of the diversity factor for different numbers of customers.12 3. TABLE 2. The value computed above would apply for four customers.14 3.55 2.19 3.14 3.2 Diversity Factors N 1 2 3 4 5 6 7 8 9 10 DF 1.18 3.20 2.3.19 3.94 2.02 3.
The demand factor can be computed for an individual customer but not for a distribution transformer or the total feeder.197 Transformer kVA rating 15 (2. then.6 Demand Factor The demand factor can be deﬁned for an individual customer.16kW maximum kW demand by the power factor.16kW maximum diversiﬁed demand and assuming a power factor of 0.96 kVA. the total connected load of the customer needs to be known. Using the 16. the transformer serving the four loads is rated 15 kVA. The utilization factor is computed to be Maximum kVA demand 17. the 15minute maximum kW demand of Customer #1 was found to be 6. and would be 17. the 15minute maximum kVA demand on the transformer is computed by dividing the 16.18 kW.96 Utilization factor = .9.18 Demand factor = . In order to determine the demand factor.7) .= .= 0.3.1766 Total connected load 35 (2. the demand factor is computed to be Maximum demand 6.6) The demand factor gives an indication of the percentage of electrical devices that are on when the maximum demand occurs. Assume that this total comes to 35 kW. The total connected load will be the sum of the ratings of all of the electrical devices at the customer’s location.= . For example.7 Utilization Factor The utilization factor gives an indication of how well the capacity of an electrical device is being utilized. 2. For example. 2.3.The Nature of Loads 19 FIGURE 2.= 1.8 Diversity factors.
the load allocated to each transformer needs to be determined.9.1 Load Allocation In the analysis of a distribution feeder load.97 − 16.8 Load Diversity Distribution System Modeling and Analysis Load diversity is deﬁned as the difference between the noncoincident maximum demand and the maximum diversiﬁed demand. When this is the case. The data provided will depend upon how detailed the feeder is to be modeled. FIGURE 2. The feeder demand curve does not display any of the abrupt changes in demand of an individual customer demand curve or the semiabrupt changes in the demand curve of a transformer. data will have to be speciﬁed.8) 2. . The simple explanation for this is that with several hundred customers served by the feeder. the odds are good that as one customer is turning off a light bulb another customer will be turning a light bulb on. For the transformer in question.81 kW (2. The most comprehensive model of a feeder will represent every distribution transformer.16 = 8.20 2.9 Feeder demand curve. The feeder load therefore does not experience a jump as would be seen in the individual customer’s demand curve. the load diversity is computed to be Load diversity = 24.4 Feeder Load The load that a feeder serves will display a smoothedout demand curve as shown in Figure 2.3. and the availability of customer load data.4. 2.
1 A singlephase lateral provides service to three distribution transformers as shown in Figure 2.2. The energy in kWh consumed by each customer during . At the end of the survey period the maximum demand vs. At each meter location the maximum demand of all downstream customers must also be known. The meter can be the same type used to develop the demand curves previously discussed. the diversity factor can be computed for the given number of downstream customers. or it can be a simple meter that only records the maximum demand during the period.11.2 Load Survey Many times the maximum demand of individual customers will be known.10) Knowing the maximum demand for each customer is the ﬁrst step in developing a table of diversity factors as shown in Table 2. Example 2.1058 + 0.4.2. Such a load survey requires the installation of a demand meter at each customer’s location. This will involve selecting a series of locations where demand meters can be placed that will record the maximum demand for groups of customers ranging from at least 2 to 70.10. 2. then it is possible to determine the maximum diversiﬁed demand of a group of customers such as those served by a distribution transformer. The plot of points for 15 customers. kW demand = 0.The Nature of Loads 21 2.005014 ⋅ kWh (2.1. the maximum diversiﬁed demand can be computed by: Max noncoincident demand Max. either from metering or from a knowledge of the energy (kWh) consumed by the customer. The straightline equation derived is Max. diversified demand = DF n (2. With that data. kWh for each customer can be plotted on a common graph. Some utility companies will perform a load survey of similar customers in order to determine the relationship between the energy consumption in kWh and the maximum kW demand. is shown in Figure 2. Diversity factors are shown in Table 2.1. When such a table is available.4.1 Application of Diversity Factors The deﬁnition of the diversity factor (DF) is the ratio of the maximum noncoincident demand to the maximum diversiﬁed demand. The next step is to perform a load survey where the maximum diversiﬁed demand of groups of customers is metered. Linear regression is used to determine the equation of a straight line that gives the kW demand as a function of kWh. along with the resulting equation derived from a linear regression algorithm. that is.9) This maximum diversiﬁed demand becomes the allocated load for the transformer.
008 ⋅ kWh The kWh consumed by Customer #1 is 1523 kWh. kWh for residential customers.4 . and it has been found that the customer 15minute maximum kW demand is given by the equation: kWdemand = 0.2 + 0. The 15minute maximum kW demand for Customer #1 is then computed as: kW1 = 0.22 12 Distribution System Modeling and Analysis 10 15Minute Maximum kW Demand (kW) 8 kWi kWli 6 4 2 0 400 600 800 1000 1200 kWhi Energy (kWh) 1400 1600 1800 2000 FIGURE 2.10 kW demand vs. A load survey has been conducted for customers in this class. a month is known.11 Singlephase lateral.008 ⋅ 1523 = 12.2 + 0. N1 N2 N3 N4 T1 T2 T3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 FIGURE 2.
= 48.= 35.0 = 81. 81.2 #10 2015 16.9 + 13.7 kW Noncoincident max.7 #15 2265 18. = 17.5 kW Diversity factor for 6 2. TRANSFORMER T1 Customer kWh kW #1 1523 12. max.2 + 16. demand = . T1: Noncoin. demand = .1 + 12.20 T2: Noncoin.6 kW Noncoincident max. using the Table of Diversity Factors in Table 2.3 #16 2135 17.5 Max. 66.40 Based upon the 15minute maximum kW diversiﬁed demand on each transformer and an assumed power factor of 0.30 T3: Noncoin.= 30.9 #8 1698 13.9 = 66.3 TRANSFORMER T3 Customer kWh kW #12 2098 17. Determine for each transformer the 15minute noncoincident maximum kW demand and.6 Max.9 #5 1456 11.7 Max.0 #13 1856 15. demand = .5 kW Noncoincident max.3 #17 1985 16.9.The Nature of Loads 23 The results of this calculation for the remainder of the customers is summarized below by transformer.9 + 11.0 1. div.1 + 16.1 + 17.= . max.4 + 16.0 = 117.2.3 #11 1765 14.3 kW Diversity factor for 5 2.8 + 14.4 + 13.7 + 18.3 + 17.3 + 14.1 #14 2058 16. 117.1 #7 1587 12. determine the 15minute maximum diversiﬁed kW demand. = 12.4 #3 1984 16.4 #2 1645 13. div.1 #18 2103 17.= .= .8 #9 1745 14.9 TRANSFORMER T2 Customer kWh kW #6 1235 10. .3 + 16. = 12. div.0 + 15.3 + 17.1 #4 1590 12. max.9 kW Diversity factor for 7 2.
6 . div.= 92. demand = .6 + 117. demand = 81.4 kW Max. With those selections.5 = 265.74) is used to compute the maximum diversiﬁed kW demand.9 49.5 kV. Noncoin.7 + 81. 37.= 72.9 kW .5 kW The maximum diversiﬁed kW demand is the computed by using the diversity factor for 18 customers.0 kW 199. demand = 48. Determine the 15minute noncoincident maximum kW demand and 15minute maximum diversiﬁed kW demand for each of the line segments.8 Max.6 kW 2. kVA T3 demand = . div. and 50 kVA. kVA T1 demand = .8 kW 2.4 . Segment N1 to N2: The maximum noncoincident kW demand is the sum of the maximum demands of all 18 customers. kVA T2 demand = . demand = . The diversity factor for 13 (2.86 Segment N2 to N3: This line segment “sees” 13 customers.6 + 117.3 Max. demand = 117.5 Max. 265. Noncoin. div.4 .74 Segment N3 to N4: This line segment sees the same noncoincident demand and diversiﬁed demand as that of transformer T3.= 39. only transformer T1 would experience a signiﬁcant maximum kVA demand greater than its rating (135%). 2.= 54.9 35.1 Max. Noncoin.24 Distribution System Modeling and Analysis the 15minute maximum kVA diversiﬁed demand on each transformer would be 30.5 = 199. The noncoincident maximum demand is the sum of customers number 6 through 18.= 33. max. respectively.9 The kVA ratings selected for the three transformers would be 25 kVA.0 Max. demand = 66.
at node N1 the maximum diversiﬁed demand ﬂowing down the line segment N1N2 is 92. There will be an inﬁnite amount of combinations of line ﬂow down N1N2 and through transformer T1 that will produce the maximum diversiﬁed demand of 72. The results of the program can also be used to allocate loads to transformers for feeder analysis purposes.3 kW. or 62. The process of developing such a table is generally not cost effective. the calculations for the maximum diversiﬁed demand in that segment was computed to be 72.6 kW. For example. the maximum diversiﬁed demand (allocated load) on each transformer on a feeder can be determined for each billing period. The explanation is that the maximum diversiﬁed demands for the line segments and transformers don’t necessarily occur at the same time. KCL would then predict that the maximum diversiﬁed demand ﬂowing down line segment N2N3 would be the difference of these. The transformer load management program relates the maximum diversiﬁed demand of a distribution transformer to the total kWh supplied by the transformer during a speciﬁc month.3 Transformer Load Management A transformer load management program is used by utilities to determine the loading on distribution transformers based upon a knowledge of the kWh supplied by the transformer during a peak loading month. 2. and the maximum diversiﬁed demand ﬂowing through transformer T1 is 30. This type of load survey meters the maximum demand on the transformer in addition to the total energy in kWh of all of the customers connected to the transformer. a curve similar to that shown in Figure 2. The usual relationship is the equation of a straight line.8 kW. 2. and the constants of the straightline equation can be computed.4 Metered Feeder Maximum Demand The major disadvantage of allocating load using the diversity factors is that most utilities will not have a table of diversity factors. the difference between the actual demand on line segment N1N2 and the demand of transformer T1 will be 72. Such an equation is determined from a load survey.10 can be developed. The program is primarily used to determine when a distribution transformer needs to be changed out due to a projected overloading condition.1.4.The Nature of Loads 25 Example 2. at the time segment N2N3 is experiencing its maximum diversiﬁed demand. line segment N1N2 and transformer T1 are not at their maximum values.5 kW. The major disadvantage of . This method has an advantage because the utility will have in the billing database the kWh consumed by each customer every month. However. At the time that line segment N2N3 is experiencing its maximum diversiﬁed demand.6 kW on line N2N3.1. All that can be said is that.1 demonstrates that Kirchhoff’s current law (KCL) is not obeyed when the maximum diversiﬁed demands are used as the load ﬂowing through the line segments and through the transformers.6 kW. As long as the utility knows which customers are connected to each transformer by using the developed equation.4. With the information available from several sample transformers.
26 Distribution System Modeling and Analysis the transformer load management method is that a database is required that speciﬁes which transformers serve which customers.5 92. The kVA ratings of all distribution transformers is always known for a feeder. Example 2. With this data.1 is 92. at minimum. Metered demand AF = kVA total where Metered demand can be either kW or kVA. and kVA total = sum of the kVA ratings of all distribution transformers The allocated load per transformer is then determined by: Transformer demand = AF ⋅ kVAtransformer (2.12) The transformer demand will be either kW or kVA depending upon the metered quantity. Since the metered data at the substation will include losses.= 0. The metered readings can be allocated to each transformer based upon the transformer rating. give either the total threephase maximum diversiﬁed kW or kVA demand and/or the maximum current per phase during a month. An “allocation factor” (AF) can be determined based upon the metered threephase kW or kVA demand and the total connected distribution transformer kVA.8 AF = . kVA.8249 kW /kVA 112. the reactive power can also be allocated. If there is no metered information on the reactive power or power factor of the feeder.9 kW. Allocate this load according to the kVA ratings of the three transformers. Most feeders will have metering in the substation that will. Again.5 + 50 = 112. this database is not always available. kVA total = 25 + 37.5 . then the load can be allocated by phase where it will be necessary to know the phasing of each distribution transformer. an iterative process will have to be followed so that the allocated load plus losses will equal the metered readings. Modern substations will have microprocessorbased metering that will provide kW. When the maximum current per phase is metered. When the kW or kVA is metered by phase. and current per phase.2 Assume that the metered maximum diversiﬁed kW demand for the system of Example 2. the load allocated to each distribution transformer can be done by assuming nominal voltage at the substation and then computing the resulting kVA. The load allocation will now follow the same procedure as outlined above. kvar. a power factor will have to be assumed for each transformer load. Allocating load based upon the metered readings in the substation requires the least amount of data.13) (2. power factor.
With a knowledge of the allocated loads ﬂowing in the line segments and through the transformers and the impedances.2.62 kW kW 1 = 0. The application of the diversity factors was demonstrated in Example 2.5 What Method to Use? Four different methods have been presented for allocating load to distribution transformers: • • • • Application of diversity factors Load survey Transformer load management Metered feeder maximum demand Which method to use depends upon the purpose of the analysis. For these studies it is assumed that the allocated loads will be modeled as constant real power and reactive power. The problem is that using those methods will result in a much larger maximum diversiﬁed demand at the substation than actually exists.8249 ⋅ 50 = 41.4. If the purpose is to determine as closely as possible the maximum demand on a distribution transformer.1.93 kW kW 1 = 0.24 kW 27 2. The various voltage drops will be computed using the loads allocated by two of the methods in the following examples. then either the diversity factor or the transformer load management method can be used.8249 ⋅ 37. . When the total feeder is to be analyzed.1 Application of Diversity Factors The loads allocated to a line segment or a distribution transformer using diversity factors are a function of the total number of customers downstream from the line segment or distribution transformer. Four different methods of allocating loads have been presented. 2.8249 ⋅ 25 = 20. the voltage drops can be computed.5 = 30. Neither of these methods should be employed when the analysis of the total feeder is to be performed.4.4. the only method that gives good results is that of allocating load based upon the kVA ratings of the transformers.2 VoltageDrop Calculations Using Allocated Loads The voltage drops down line segments and through distribution transformers are of interest to the distribution engineer. The assumption is that the allocated loads will be constant real power and reactive power. 2.The Nature of Loads The allocated kW for each transformer becomes: T1: T2: T3: kW 1 = 0.1.
2 kVA ST3 = 48.0 kVA S23 = 72. The system of Example 2. 2400240 volts. Example 2.8 + j45.3 For the system of Example 2.5 + j17. In order to avoid an iterative solution. is shown in Figure 2.8/40% Z = 1.9 + j23.1 the maximum diversiﬁed kW demands were computed. 2400240 volts. The same system and allocated loads from Example 2.6 kW P34 = 48. 50 kVA.9 + j23. Assume that the power factor of the loads is 0.9 lagging power factor. 37.3 kW PT2 = 35.12.7 kVA ST2 = 35.9/45% Z = 2.3 + j14. 2400240 volts. including segment distances.7 kVA .0/50% From Example 2. The impedance of the lines are: z = 0. Example 2.5 kVA.28 N1 5000' Distribution System Modeling and Analysis N2 500' N3 750' N4 T1 T2 T3 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 FIGURE 2.3 demonstrates how the method of load allocation using diversity factors is applied.5 kW PT3 = 48.9 lagging.6 + j35. Z = 1.1.3 + j0.1.9 kW S12 = 92.8 kW P23 = 72. Using the 0.6 Ω/mile The ratings of the transformers are T1: T2: T3: 25 kVA. assume the voltage at N1 is 2400 volts and compute the secondary voltages on the three transformers using the diversity factors.9 kW PT1 = 30. the voltage at the source is assumed and the voltage drops calculated from that point to the last transformer.12 Singlephase lateral with distances.7 kVA ST1 = 30.2 kVA S34 = 48. the maximum diversiﬁed kW and kVA demands for the line segments and transformers are Segment N1N2: Segment N2N3: Segment N3N4: Transformer T1: Transformer T2: Transformer T3: P12 = 92.1 are used.
3 + j0.4 Ω kVA 25 Z T1 = ( 0.16/ – 26.4 ⋅ 1000 Z base = .= 0.= .0852 Ω 5280 Calculate the current ﬂowing in segment N1N2: kW + jk var ∗ 92. =  = 43.6 ) ⋅ .6 ) ⋅ .378/ – 0.4 = 3.0568 Ω 5280 750 Z 34 = ( 0.4 ⋅ 1000 Z base = .24 = 2321.4 ⋅ 1000 Z base = .0/ – 25.= 230.67 Ω kV ⋅ 1000 2.4 kV Calculate the secondary voltage referred to the high side: V T1 = V 2 – Z T2 ⋅ I T1 V T1 = 2378.4 V Calculate the current ﬂowing into T1: kW + jk var ∗ 30.3 + j0.4/ – 0.77 Ω 2 2 2 2 2 2 T2: T3: Compute the line impedances: N1N2: N2N3: N3N4: 5000 Z 12 = ( 0.06 Ω kV ⋅ 1000 2.84 A 2.3 + j0.= 153.6 = 2.24 A 2.2 Ω kVA 50 Z T3 = ( 0.= .6 Ω kVA 37.2841 + j0.= 0.= .019/45 ) ⋅ 153.5/ – 0.4/0 kV Calculate the voltage at N2: V 2 = V 1 – Z 12 ⋅ I 12 V 2 = 2400/0 – ( 0.018/40 ) ⋅ 230.06 + j2.0284 + j0.16/ – 26.4 – ( 3.3 + j14.= 0.84 = 2378.The Nature of Loads 29 Convert transformer impedances to ohms referred to the highvoltage side T1: kV ⋅ 1000 2.02/50 ) ⋅ 115.5 Z T2 = ( 0.18 + j2.2841 + j0.18 + j2. =  = 14.67 ) ⋅ 14.8 V .0426 + j0.= 115.0 ∗ I 12 = .4/ – 0.9 + j 45.2 = 1.7 ∗ I T1 = .48 + j1.6 ) ⋅ .0/ – 25.5682 Ω 5280 500 Z 23 = ( 0.5682 ) ⋅ 43.
4 – ( 0. =  = 33.9/ – 26.4 V Calculate the current ﬂowing into T2: kW + jk var ∗ 35.8 Vlow T1 = .27 = 2331/1/ – 0.27 2.9/ – 26.1/ – 0.30 Distribution System Modeling and Analysis Compute the secondary voltage by dividing by the turns ratio of 10: 2321.7/ – 0.5 + j17.9/ – 26.0/ – 0.24 = 2376.24 A 2.5 V .27 A 2.3767/ – 0.4 kV Calculate the voltage at N4: V 4 = V 3 – Z 34 ⋅ I 34 V 4 = 2376.5/ – 0.6 + j35.378/ – 0. = .4 kV Calculate the voltage at N3: V 3 = V 2 – Z 23 ⋅ I 23 V 3 = 2378.0568 ) ⋅ 33.8 V 10 Calculate the current ﬂowing in line section N2N3: kW + jk var ∗ 72. = .58/ – 26.2 ∗ I 23 = .0852 ) ⋅ 22.8 Vlow T2 = .7 ∗ I 34 = .58/ – 26.0 + j23.0426 + 0.7/ – 0.27 = 2375.4 kV Calculate the secondary voltage referred to the high side: V T2 = V 3 – Z T2 ⋅ I T2 V T2 = 2376. = 16.3767/ – 0.4/ – 0.4 – ( 2.0284 + j0. = 22.= 232.8 V 10 Calculate the current ﬂowing in line section N3N4: kW + jk var ∗ 49.2 ∗ I T2 = .= 233.8 V Compute the secondary voltage by dividing by the turns ratio of 10: 2331.4 – ( 0.06 ) ⋅ 16.9/ – 26.7/ – 0.15/ – 0.1/ – 0.06 + j2.
Example 2.91/ – 26.1.9 ) = 92. Assume the load power factor is 0.4 For the system of Example 2.9 lagging.5 + 50 .0 Vlow T3 = . This method was discussed in Section 2. the feeder can be allocated based upon the metered demand and the transformer kVA ratings.7/ – 1.= 232.0 V Compute the secondary voltage by dividing by the turns ratio of 10: 2326.2 Load Allocation Based upon Transformer Ratings When only the ratings of the distribution transformers are known.84 AF = .⋅ 100 = 3.3.0 V 10 31 Calculate the percent voltage drop to the secondary of transformer T3. allocating the loads based upon the transformer ratings.30 A Calculate the secondary voltage referred to the high side: V T3 = V 4 – Z T3 ⋅ I T3 V T3 = 2375.84 kVA 0.11 V drop = . Example 2.5 – ( 1.9/ – 26.0/ – 0. The impedances of the line segments and transformers are the same as in Example 2. compute the kVA demand at N1 from the metered demand: 92.⋅ 100 = .0789% V1 2400 2.77 ) ⋅ 22.27 = 2326.84 25 + 37.9 kW. Use the secondary voltage referred to the high side: V 1 – V T3 2400 – 2326.8 + j45.0 = 103.= 0.The Nature of Loads The current ﬂowing into T3 is the same as the current from N3 to N4: I T3 = 22.4.2.48 + j1.2/25.2/25. assume the voltage at N1 is 2400 volts and compute the secondary voltages on the three transformers.9 Calculate the allocation factor: 103.3.3.9 –1 S 12 = . Assume that the metered kW demand at N1 is 92.9/ – 1.9/ – 1.4 demonstrates this method./cos ( 0.9175/25.
0 kVA S T3 = AF ⋅ kVA T3 = ( 0.0 + j15.3 + j35 kVA S 34 = S T3 = 41.9175/25.32 Distribution System Modeling and Analysis Allocate the loads to each transformer: S T1 = AF ⋅ kVA T1 = ( 0.7/ – 0.84 ) ⋅ 50 = 41.3. the node and secondary transformer voltages are V 2 = 2378.3 + j20.4 V Vlow T2 = 233. This is evidenced by the fact that the diversity factor became almost constant as the number of customers .4 V Vlow T1 = 234.3 + j20.9 V The percent voltage drop for this case is V 1 – VT 3 2400 – 2334. but as the demand is monitored on line segments working back toward the substation.5 = 31.8 V drop = .4 V V 4 = 2375.6 + j10.0 kVA Calculate the line ﬂows: S 12 = S T1 + S T2 + S T3 = 92. When this procedure is followed.3/ – 0. the procedure for computing the transformer secondary voltages is exactly the same as in Example 2.⋅ 100 = .4/ – 0. the effect of the diversity between demands becomes very slight.84 ) ⋅ 37.5 Summary This chapter has demonstrated the nature of the loads on a distribution feeder.0 kVA S 23 = S T2 + S T3 = 72.⋅ 100 = 2.7/0.9175/25.5/ – 0. There is great diversity between individual customer demands.4 V V 3 = 2376.0/ – 0.7179% V1 2400 2.9 + j45.8 V Vlow T3 = 233.0 kVA S T2 = AF ⋅ kVA T2 = ( 0. For the diversity factors of Table 2. The effect of diversity for short laterals can be taken into account in determining the maximum ﬂow on the lateral.9175/25.0 kVA Using these values of line ﬂows and ﬂows into transformers.2. It was shown that the effect of diversity between customer demands must be taken into account when the demand on a distribution transformer is computed.84 ) ⋅ 25 = 20. it was shown that when the number of customers exceeds 70 the effect of diversity has pretty much disappeared.
1 Time Cust #1 kW 8. A 25kVA singlephase transformer serves the four customers.12 6.32 5.88 Cust #2 kW 4.12 1.04 9.44 3.32 3.12 0.81 2.72 8.62 7.16 4.92 1.68 6.88 4.12 2.4 show that the ﬁnal node and transformer voltages are approximately the same.68 0.56 6.64 Cust #3 kW 11. Examples 2.69 1.72 3.44 1.92 2. Problems Shown below are the 15minute kW demands for four customers between the hours of 17:00 and 21:00.08 1.04 7.76 2.12 3. It must be understood that the number 70 will apply only to the diversity factors of Table 2.24 1.08 4.96 3.12 10.12 2. There was very little difference between the voltages when the loads were allocated using the diversity factors and when the loads were allocated based upon the transformer kVA ratings. For each of the customers determine: (a) Maximum 15minute kW demand (b) Average 15minute kW demand (c) Total kWh usage in the time period (d) Load factor .84 1.52 4.The Nature of Loads 33 approached 70.68 5.72 10.96 4.08 5.04 6.52 2. If a utility is going to use diversity factors.04 7.04 2. 2.33 1.41 17:00 17:15 17:30 17:45 18:00 18:15 18:30 18:45 19:00 19:15 19:30 19:45 20:00 20:15 20:30 20:45 21:00 1.96 6.48 2.16 6.46 0.32 Cust #4 kW 1.96 2.72 2.84 6.28 1.16 7.56 8.48 7.2.88 3.44 8.56 1.3 and 2.48 4. that utility must perform a comprehensive load survey in order to develop the table of diversity factors that apply to that particular system.12 9.12 6.08 3.89 1.24 4.84 4.24 1.62 2.
167) .2 Two transformers each serving four customers are shown in Figure 2. #2 #3 #4 #5 #6 #7 #8 The following table gives the time interval and kVA demand of the four customer demands during the peak load period of the year.13 System for Problem 2. For each transformer determine the following: (a) 30minute maximum kVA demand (b) Noncoincident maximum kVA demand (c) Load factor (d) Diversity factor (e) Suggested transformer rating (50.2. For the 25kVA transformer determine: (a) Maximum 15minute diversiﬁed demand (b) Maximum 15minute noncoincident demand (c) Utilization factor (assume unity power factor) (d) Diversity factor (e) Load diversity 3. Assume a power factor of 0. Plot the load duration curve for the transformer 2. 100. Tap #1 FIGURE 2.34 Distribution System Modeling and Analysis 2. 75.9 lagging.13. Time 3:00–3:30 3:30–4:00 4:00–4:30 4:30–5:00 5:00–5:30 5:30–6:00 6:00–6:30 6:30–7:00 #1 10 20 5 0 15 15 5 10 #2 0 25 30 10 5 15 25 50 #3 10 15 30 20 5 10 25 15 #4 5 20 15 10 25 10 15 30 #5 15 25 10 13 30 5 10 15 #6 10 20 30 40 30 20 10 5 #7 50 30 10 25 15 30 30 10 #8 30 40 10 50 5 25 25 30 1.
3 N1 5000' N2 2500' N3 T1 T2 1 2 3 4 5 6 7 8 9 10 11 12 FIGURE 2. .m. Compute the secondary voltage for each transformer.3. and 9:00 p. Determine the average kW demand for each customer. 5. 8.The Nature of Loads (f) Utilization factor (g) Energy (kWh) during the 4hour period 2. Determine the maximum diversiﬁed 30minute kVA demand at the Tap 35 Two singlephase transformers serving 12 customers are shown in Figure 2. Determine the diversity factor of the load for each transformer.0 power factor) for each transformer.14. 2. 2. The 15minute kW demands for the 12 customers between the hours of 5:00 p. Determine the maximum noncoincident demand for each transformer. are given in the tables that follow. 9.306 + j0.020/50 37. Assume a load power factor of 0.95 lagging. 10. Determine the kWH consumed by each customer in this time period.14 Circuit for Problem 2. Determine the maximum diversiﬁed demand for each transformer. The voltage at node N1 is 2500/0 V. Determine the load factor for each customer. 6. Determine the utilization factor (assume 1. taking diversity into account.6272 Ω/mile. Determine the maximum diversiﬁed demand at Node N1. 7.5 kVA 2400240 V Determine the maximum kW demand for each customer. 3. Transformer ratings: T1: T2: 1. 25 kVA 2400240 V Z pu = 0.018/40 Z pu = 0. 4.m. The impedance of the lines are z = 0.
21 10.87 0. Allocate the metered demand to each transformer based upon the transformer kVA rating.38 0.29 2.21 7.68 05:00 05:15 05:30 05:45 06:00 06:15 06:30 06:45 07:00 07:15 07:30 07:45 08:00 08:15 08:30 08:45 09:00 On a different day.17 0.58 #8 kW 0.19 2.43 kW.68 9.75 5.89 2.72 12.19 0.10 0.47 11.39 0.81 5.81 2.29 4.13 2.24 1.40 0.83 9.99 8.18 1.24 0.53 0.10 3.00 0.89 2.95 #10 kW 0.23 #12 kW 2.25 10.16 0.25 3.38 13.55 9.75 2. Assume a power factor of 0.26 9.99 0.96 6.09 9.11 4.42 3.18 2.39 0.72 8. 2.44 0.08 2.86 0.63 1.62 0.35 2.39 13.34 8.24 0.72 3.60 11.22 9.79 0.98 7.99 8.88 1.69 0.91 1.98 2.22 4.80 0.65 2.58 7.22 2.80 7.41 8.52 3.36 TRANSFORMER #125 kVA Time #1 kW 2.48 2.45 10.67 1.51 0.20 2.63 1.80 0.5 kVA Time #6 kW 0.47 0.34 5.08 2.01 7.27 4.21 Distribution System Modeling and Analysis #3 kW 4.11 7.35 6.22 8.74 2.92 3.58 6.74 11.07 1.01 4.22 2.93 3.73 2.60 0.51 0.34 0.26 7.64 3.00 2.27 10.04 4.16 0.44 0.11 4.05 0.70 0.58 7.28 1.24 1.63 1.76 1.04 #2 kW 0.66 6.40 #4 kW 8.3 is 72.46 0.26 8.47 0.94 2.78 6.78 2.33 0.54 5.65 7. the metered 15minute kW demand at node N1 for the system of Problem 2.92 5.57 0.55 3.27 2.09 9.42 2.11 0.79 #7 kW 2.72 1.42 8.93 1.43 0.38 1.31 9.56 0.97 0.05 1.12 1.59 1.29 11.81 2.76 2.87 0.09 2.95 0.97 0.98 8.36 #9 kW 8.85 1.62 7.60 0.95 lagging.4 .12 0.70 05:00 05:15 05:30 05:45 06:00 06:15 06:30 06:45 07:00 07:15 07:30 07:45 08:00 08:15 08:30 08:45 09:00 TRANSFORMER #237.57 8.15 2.73 0.26 3.36 0.19 5.72 2.60 2.48 0.52 1.19 0.39 0.37 0.13 2.37 4.72 3.07 0.55 1.56 1.24 2.12 2.75 8.36 0.25 2.88 #5 kW 0.24 4.82 7.46 0.30 2.04 1.52 0.35 13.71 3.19 #11 kW 9.23 13.60 7.78 8.13 1.66 0.53 2. Assume the loads are constant current and compute the secondary voltage for each transformer.39 5.
6.035 perunit Use the diversity factors found in Table 2.The Nature of Loads 2. divide by 18 (number of customers).4. Z = 0. (5) Repeat Part (4) above.5 kW + j7. Now calculate the voltages at all of the nodes listed in Part (3) using the instantaneous loads.5 kVA. but assume the loads are “constant current. The four transformers are rated as: T1 and T2: 37. 2400240 V.5 37 A singlephase lateral serves four transformers as shown in Figure 2.03 perunit T3 and T4: 50 kVA. (3) If the voltage at node 1 is 2600/0 V. take the current ﬂowing from node 1 to node 2 from Part (4) above. 1 380' 2 470' 4 750' 6 820' 8 T1 3 5 T2 7 T3 9 T4 FIGURE 2. The impedance of the singlephase lateral is z = 0.3213 Ω/1000 ft.15 System for Problem 2.8.4421 + j0. determine the voltage at nodes 2.15. calculate all of the voltages. and assign that as the instantaneous constant current load for each customer. (4) Use the 15minute maximum diversiﬁed demands at the lateral tap (Section 12) from Part (2) above.01 + j0. Again. In calculating the voltages. .5. (2) The 15minute maximum diversiﬁed kW and kvar demands for each line section.3. Z = 0.2 and determine: (1) The 15minute maximum diversiﬁed kW and kvar demands on each transformer.5.015 + j0. Assume that each customer’s maximum demand is 15. take into account diversity using the answers from (1) and (2) above. Divide these maximum demands by 18 (number of customers) and assign that as the “instantaneous load” for each customer. and 9. 2400240 V.5 kvar.” To do this.7.
38 Distribution System Modeling and Analysis (6) Take the maximum diversiﬁed demand from node 1 to node 2 and allocate that to each of the four transformers based upon their kVA ratings. calculate all of the voltages. take the maximum diversiﬁed demand and divide by 175 (total kVA of the four transformers). . (7) Compute the percent differences in the voltages for Parts (4). (5). Now multiply each transformer kVA rating by that number to give how much of the total diversiﬁed demand is being served by each transformer. Again. To do this. and (6) at each of the nodes using the Part (3) answer as the base.
In Figure 3. With these assumptions.2 the phasor for the voltage drop through the line resistance (RI) is shown in phase with the current phasor. The voltage drop down the line is deﬁned as the difference between the magnitudes of the source and the load voltages. many times only a “ballpark” answer is needed.3 Approximate Methods of Analysis A distribution feeder provides service to unbalanced threephase.1) The phasor diagram for Equation 3. This combination leads to threephase line currents and line voltages being unbalanced. twophase. twophase.2) 39 .1. and singlephase loads over untransposed threephase. and all line segments will be threephase and perfectly transposed. a single linetoneutral equivalent circuit for the feeder will be used. All of the approximate methods of modeling and analysis will assume perfectly balanced threephase systems. It is the purpose of this chapter to develop some of the approximate methods and leave for later chapters the exact models and analysis. In order to analyze these conditions as precisely as possible. it will be necessary to model all three phases of the feeder accurately. 3.2. Kirchhoff’s voltage law applied to the circuit of Figure 3. and the phasor for the voltage drop through the reactance is shown leading the current phasor by 90 degrees. and singlephase line segments.1 is shown in Figure 3. V drop = V S – V L (3.1 Voltage Drop A linetoneutral equivalent circuit of a threephase line segment serving a balanced threephase load is shown in Figure 3. however. some approximate methods of modeling and analysis can be employed. The dashed lines represent the real and imaginary parts of the impedance (ZI) drop. When this is the case. It will be assumed that all loads are balanced threephase.1 gives: V S = V L + ( R + jX ) ⋅ I = V L + R ⋅ I + jX ⋅ I (3.
1 Linetoneutral equivalent circuit. jX + I VL Load  VS ZI Im(ZI) Real(ZI) 0 I FIGURE 3.8419 A . Equation 3.3) For the purposes of this chapter. the voltage drop between the source and load voltage is approximately equal to the real part of the impedance drop.3.2 Phasor diagram. VL RI jXI The angle between the source voltage and the load voltage (δ ) is very small.1 In Example 2.3 will be used as the deﬁnition of voltage drop. the impedance of the ﬁrst line segment is Z 12 = 0.2841 + j0.5682 Ω The current ﬂowing through the line segment is I 12 = 43.0093/ – 25. That is V drop ≅ Re ( Z ⋅ I ) (3. Because of that.40 Distribution System Modeling and Analysis R + VS FIGURE 3. Example 3.
it will be assumed that the line segment is transposed.6486 Error = .) (3.0093/ – 25.3. only the positive sequence impedance of the line segment needs to be determined.5682 ) ⋅ 43.27% 21.2841 + j0.4098/ – 0.5) GMR = conductor geometric mean radius (from tables) (ft.3 is given by: D eq z positive = r + j0.8419 ] = 21.5902 V Computing the voltage drop according to Equation 3.0 V The exact voltage at node N2 is computed to be V 2 = 2400/0.) .3 gives: V drop = Re [ ( 0. Ω/mile GMR where r = conductor resistance (from tables) Ω/mile D eq = 3 (3.5682 ) ⋅ 43. The equation for the positive sequence impedance for the conﬁguration shown in Figure 3.2 Line Impedance For the approximate modeling of a line segment.3.5902 – 21.12134 ⋅ ln . With this assumption.0 – ( 0. 3.2841 + j0.4) D ab ⋅ D bc ⋅ D ca (ft.4098 = 21.8419 = 2378.0093/ – 25.Approximate Methods of Analysis The voltage at node N1 is V 1 = 2400/0. A typical threephase line conﬁguration is shown in Figure 3.⋅ 100 = – 0.0000 – 2378.6486 V 21.5902 41 This example demonstrates the very small error in computing voltage drop when using the approximate equation given by Equation 3.4015 V The voltage drop between the nodes is then: V drop = 2400.
3 Threephase line conﬁguration.5 ft. = 0. The spacings between conductors are D ab = 2.0244 ft Compute the equivalent spacing: D eq = Using Equation 3..2863 z positive = 0.2 A threephase line segment has the conﬁguration as shown in Figure 3.5 ft.0 ft.5 ⋅ 4.2863 ft . Example 3.0244 3 2.6272 Ω/mile 0.4: 4.306 Ω/mile GMR = 0.306 + j0. D bc = 4.0 = 4.. D ca = 7.5 ⋅ 7.306 + j0. The conductors of the line are 336.12134 ⋅ ln .400 26/7 ACSR.42 Distribution System Modeling and Analysis Dca Dab a b Dbc c n FIGURE 3. Determine the positive sequence impedance of the line in ohms/mile: SOLUTION From the table of conductor data in Appendix A: r = 0.3.
SOLUTION The impedance of one mile of line was computed to be Z = 0. There will be two types of K factors: one for voltage drop and the other for voltage rise calculations.9 lagging and a nominal voltage of 12.3 “K” Factors A ﬁrst approximation for calculating the voltage drop along a line segment is given by Equation 3. compute the Kdrop factor assuming a load power factor of 0.= 7199.9 ) = 0. Example 3.306 + j0.3./ – cos ( PF ) = ./ – cos ( 0.5) The Kdrop factor is determined by computing the percent voltage drop down a line that is one mile long and serving a balanced threephase load of 1 kVA.47 kV (linetoline).3 For the line of Example 3.046299/ – 25.84 ] = 0.2.Approximate Methods of Analysis 43 3.1 The Kdrop Factor The Kdrop factor is deﬁned as: Percent voltage drop K drop = kVA · mile (3.306 + j0. The percent voltage drop is referenced to the nominal voltage of the line.47 The voltage drop is computed to be V drop = Re [ Z ⋅ I ] = Re [ ( 0. the power factor of the load must be assumed.6272 Ω The current taken by 1 kVA at 0.025408 V The nominal linetoneutral voltage is 12470 V LN = .3. Another approximation is made by employing a ‘‘K” factor. In order to calculate this factor.6272 ) ⋅ 0.9 lagging power factor is given by: 1 kVA 1 –1 –1 I = .046299/ – 25.6 V 3 .84 A 3 ⋅ kV LL 3 ⋅ 12. 3.
. Example 3.6 The Kdrop factor computed in Example 3.= 5667. The Kdrop factor can be used to quickly compute the approximate voltage drop down a line section.4 A threesegment feeder is shown in Figure 3.9 lagging is a good approximation for a feeder serving a predominately residential load.400 26/7 ACSR conductor with the conductor spacings deﬁned in Example 3.00035291% drop/kVAmile 7199.00035291 ⋅ 7500 ⋅ 1. Fortunately.5 The application of the Kdrop factor is not limited to computing the percent voltage drop down just one line segment.025408 K drop = . the total percent voltage drop from the source to the end of the last line segment is the sum of the percent drops in each line segment.5 miles from the substation? 3. and one or two standard distribution voltages.9 lagging. When line segments are in cascade.3.5 miles from the substation with a resulting voltage drop of 3. The Kdrop factor for the line segments is K drop = 0.⋅ 100 = 0. Example 3. most utilities will have a set of standard conductors. Using the Kdrop factor computed in Example 3.5 = 3. How much load can be served 1. spacings. a simple spreadsheet program can be written that will compute the Kdrop factors for the standard conﬁgurations. For example. as would be the usual case. and voltages.44 The Kdrop factor is then: Distribution System Modeling and Analysis 0. but it must be understood that in all cases the percent drop is in reference to the nominal linetoneutral voltage.97%.9702% This example demonstrates that a load of 7500 kVA can be served 1. That is. assume a load of 7500 kVA is to be served at a point 1.00035291 ⋅ 1. Because of this. This seems logical. The assumed power factor of 0.47 kV.2. a nominal voltage of 12. the percent voltage drop down the line segment is computed to be V drop = K drop ⋅ kVA ⋅ mile = 0.2 kVA 0.00035291 Determine the percent voltage drop from N0 to N3.4 will demonstrate this application.0% kVA load = . but rather the nominal linetoneutral voltage.4. standard conductor spacings. the percent voltage drop in a line segment is not referenced to the source end voltage. Suppose now that the utility has a maximum allowable voltage drop of 3.5 miles from the substation. Unique Kdrop factors can be determined for all standard conductors. and a load power factor of 0.0%.3 is for the 336.
00035291 ⋅ 1550 ⋅ 1.8205% The total kVA ﬂowing in segment N1 to N2 is kVA 12 = 750 + 500 = 1250 kVA The percent voltage drop from N1 to N2 is Vdrop 12 = 0.75 = 0.00035291 ⋅ 500 ⋅ 0. 750 kVA 500 kVA SOLUTION The total kVA ﬂowing in segment N0 to N1 is kVA 01 = 300 + 750 + 500 = 1550 kVA The percent voltage drop from N0 to N1 is Vdrop 01 = 0.5 mile N1 0.2396% The application of the Kdrop factor provides an easy way of computing the approximate percent voltage drop from a source to a load. .Approximate Methods of Analysis N0 1.5 mile 45 N2 N3 0.3308 + 0.0882% The total percent voltage drop from N0 to N3 is Vdrop total = 0.5 = 0.8205 + 0.4 Threesegment feeder.00035291 ⋅ 1250 ⋅ 0.0882 = 1.3308% The kVA ﬂowing in segment N2 to N3 is kVA 23 = 500 The percent voltage drop in the last line segment is Vdrop 23 = 0. It should be kept in mind that the assumption has been a perfectly balanced threephase load.75 mile 300 kVA FIGURE 3.5 = 0.
2 The Krise Factor The Krise factor is similar to the Kdrop factor except that now the load is a shunt capacitor.5 Voltage rise phasor diagram.46 I Distribution System Modeling and Analysis cap VS ZI jXI cap Real(ZI) RI cap Im(ZI) VL FIGURE 3.6272 Ω . The Krise factor is deﬁned exactly the same as for the Kdrop factor: Percent voltage rise K rise = kvar mile.3 to 2. Determine the rating of a threephase capacitor bank to limit the voltage drop in Example 3.6) In Equation 3. 2. The impedance of one mile of line was computed to be Z = 0. Even with these assumptions the results will always provide a ‘‘ballpark” result that can be used to verify the results of more sophisticated methods of computing voltage drop. When a leading current ﬂows through an inductive reactance there will be a voltage rise across the reactance rather than a voltage drop. an assumed load power factor. and transposed line segments.5. Calculate the Krise factor for the line of Example 3. (3. the voltage rise is deﬁned as V rise = Re ( ZI cap ) = X ⋅ I cap (3.3.7) Example 3.6 it is necessary to take the magnitude of the real part of ZI so that the voltage rise is a positive number.5.5 1.306 + j0.5%. This is illustrated by the phasor diagram of Figure 3. SOLUTION 1. Referring to Figure 3.3. 3.
When the loads are uniformly distributed.47 The voltage rise per kvar mile is computed to be V rise = Re [ Z ⋅ I cap ] = Re [ ( 0.9702%.306 + j0.9702 – 2.5 = 1.3 was computed to be 3. Figure 3./90 = 0./90 = .Approximate Methods of Analysis The current taken by a 1kvar threephase capacitor bank is given by: 1 kvar 1 I cap = .4702 kvar = .046299/90 ] = 0. twophase.6 V 3 The Krise factor is then 0.6 2.5 47 3.029037 V The nominal linetoneutral voltage is 12.4 Uniformly Distributed Loads Many times it can be assumed that loads are uniformly distributed along a line where the line can be a threephase.= .00040331 ⋅ 1.⋅ 100 = 0.6272 ) ⋅ 0.18 kvar K rise ⋅ mile 0.029037 K rise = .046299/90 A 3 ⋅ kV LL 3 ⋅ 12.6 shows a generalized line with n uniformly distributed loads. To limit the total voltage drop to 2.= 7199. it is not necessary to model each load in order to determine the total voltage drop from the source end to the last load. . This is certainly the case on singlephase laterals where the same rating transformers are spaced uniformly over the length of the lateral.= 2430. the required voltage rise due to a shunt capacitor bank is V rise = 3.470 V LN = .4702% The required rating of the shunt capacitor is V rise 1.5%. or singlephase feeder or lateral. The percent voltage drop in Example 3.00040331% rise/kvar mile 7199.
It is desired to determine the total voltage drop from the source node (S) to the last node n.6 shows n uniformly spaced loads dx miles apart.48 Distribution System Modeling and Analysis length IT S dx 1 dx 2 dx 3 dx 4 dx 5 n di FIGURE 3. Let l z dx di n IT = = = = = = length of the feeder r + jx = impedance of the line in Ω/mile length of each line section load currents at each node number of nodes and number of line sections total current into the feeder The load currents are given by: IT di = n The voltage drop in the ﬁrst line segment is given by: Vdrop 1 = Re { z ⋅ dx ⋅ ( n ⋅ di ) } The voltage drop in the second line segment is given by: Vdrop 2 = Re { z ⋅ dx ⋅ [ ( n – 1 ) ⋅ di ] } (3. The total current into the feeder is IT .10) (3. The loads are all equal and will be treated as constant current loads with a value of di.8) The total voltage drop from the source node to the last node is then given by: Vdrop total = Vdrop 1 + Vdrop 2 + … + Vdrop n Vdrop total = Re { z ⋅ dx ⋅ di ⋅ [ n + ( n – 1 ) + ( n – 2 ) + … + ( 1 ) ]} (3.4. di di di di di 3.1 Voltage Drop Figure 3.11) .6 Uniformly distributed loads.9) (3.
13) (3.Vdrop total = Re z ⋅ .14) (3.16 gives the general equation for computing the total voltage drop from the source to the last node n for a line of length l. 2 n 1 1 Vdrop total = Re .13 results in: l IT n ⋅ (n + 1) . 2 The incremental distance is l dx = n The incremental current is IT di = n Substituting Equations 3.⋅ . Equation 3.17) .Approximate Methods of Analysis Equation 3.12) (3.11 becomes: n ⋅ (n + 1) Vdrop total = Re z ⋅ dx ⋅ di ⋅ .15 into Equation 3.15) (3. In the limiting case where n goes to inﬁnity.16) Equation 3.14 and 3.⋅ Z ⋅ I T ⋅ 1 + . 2 n where Z=z⋅l 49 (3.11 can be reduced by recognizing the series expansion: n(n + 1) 1 + 2 + 3 + … + n = 2 Using the expansion.⋅ . 2 n n 1 n+1 Vdrop total = Re z ⋅ l ⋅ I T ⋅ . the ﬁnal equation becomes: 1 Vdrop total = Re .⋅ .⋅ Z ⋅ I T 2 (3.
8 Onehalf load lumped at the end. the result is 2 3 2 P loss = 3 ⋅ I T ⋅ R = .7.⋅ I T ⋅ R 4 2 (3. The ﬁrst is to recognize that the total line distributed load can be lumped at the midpoint of the lateral as shown in Figure 3.7 is used to compute the total threephase power loss down the line. The voltage drop is the total from the source to the end of the line.17 is to lump onehalf of the total line load at the end of the line (node n). 3.7 Load lumped at the midpoint.18) When the model of Figure 3.7 and 3. length IT n S I T /2 FIGURE 3.2 Power Loss Of equal importance in the analysis of a distribution feeder is the power loss. This model is shown in Figure 3.17. If the model of Figure 3. Z represents the total impedance from the source to the end of the line. Figures 3. In Equation 3. the result is 3 2 R 2 P loss = 3 ⋅ I T ⋅ .19) .8 give two different models that can be used to calculate the total voltage drop from the source to the end of a line with uniformly distributed loads.⋅ I T ⋅ R 2 2 (3.= .8 is used to compute the total threephase power loss. A second interpretation of Equation 3.50 Distribution System Modeling and Analysis length IT S length/2 n IT FIGURE 3.8. The equation can be interpreted in two ways.4.
23) (3. The question is. reference is made to Figure 3. ⋅ .20) (3.25) .25 gives the total threephase power loss for a discrete number of nodes and line segments. The total threephase power loss down the line will be the sum of the power losses in each short segment of the line.Approximate Methods of Analysis 51 It is obvious that the two models give different results for power loss.24: ( n + 1 ) ⋅ ( 2n + 1 ) 2 Ploss total = 3 ⋅ R ⋅ I T ⋅ 2 6⋅n 2⋅n +3⋅n+1 2 Ploss total = 3 ⋅ R ⋅ I T ⋅ 2 6⋅n 1 1 1 2 Ploss total = 3 ⋅ R ⋅ I T ⋅ .24) (3. ⋅  n n 6 Simplifying Equation 3. 3.6 and the deﬁnitions for the parameters in that ﬁgure. For a truly uniformly distributed load. and 3.+ 3 2 ⋅ n 6 ⋅ n2 Where R = r ⋅ l. For example. Equation 3. the number of 2 (3.22 gives: l I T 2 n ⋅ ( n + 1 ) ⋅ ( 2n + 1 ) Ploss total = 3 ⋅ r ⋅ .15. which one is correct? The answer is neither one.+ .22 is the sum of the squares of n numbers and is equal to: n ⋅ ( n + 1 ) ⋅ ( 2n + 1 ) 2 2 2 2 1 + 2 + 3 + … + n = 6 Substituting Equations 3.22) The series inside the brackets of Equation 3.14.23 into Equation 3. To derive the correct model for power loss.21) The total power loss over the length of the line is then given by: Ploss total = 3 ⋅ ( r ⋅ dx ) ⋅ di [ n + ( n – 1 ) + ( n – 2 ) + … + 1 ] 2 2 2 2 2 (3. the threephase power loss in the ﬁrst segment is Ploss 1 = 3 ⋅ ( r ⋅ dx ) ⋅ ( n ⋅ di ) The power loss in the second segment is given by: Ploss 2 = 3 ⋅ ( r ⋅ dx ) ⋅ [ ( n – 1 ) ⋅ di ] 2 2 (3. the total resistance per phase of the line segment.
26) A circuit model for Equation 3. What is needed is one model that will work for both voltage drop and power loss calculations.25.9.9 Power loss model. cI T nodes goes to inﬁnity. In Figure 3.⋅ R ⋅ I T 3 (3.4. Figure 3.1 can be used for the computation of the total voltage drop down the line. From a comparison of Figures 3. to Figure 3.8. l I T S kl (1k) l n IX FIGURE 3.10 a portion (Ix) of .7 and 3. the ﬁnal equation for computing the total threephase power loss down the line is given by: 1 2 Ploss total = 3 ⋅ .4.26 is given in Figure 3.4. The ﬁrst models of Section 3.3 The Exact Lumped Load Model In the previous sections lumped load models were developed.2 developed a model that will give the correct power loss of the line.9. When that limiting case is taken in Equation 3. It was shown that the same models cannot be used for the computation of the total power loss down the line. used for voltage drop calculations. 3. it is obvious that the same model cannot be used for both voltage drop and power loss calculations.52 Distribution System Modeling and Analysis length IT S length/3 n IT FIGURE 3.10 General exact lumped load model.10 shows the general conﬁguration of the exact model that will give correct results for voltage drop and power loss. Section 3. used for power loss calculations.
10 is given by: Ploss total = 3 ⋅ [ k ⋅ R ⋅ I T + ( 1 – k ) ⋅ R ⋅ ( c ⋅ I T ) ] 2 2 (3. and the remaining current (cIT) will be modeled at the end of the line.Approximate Methods of Analysis 53 the total line current (IT) will be modeled kl miles from the source end. The values of k and c need to be derived.31) (3.⋅ Z ⋅ I T = [ k ⋅ Z ⋅ I T + ( 1 – k ) ⋅ Z ⋅ c ⋅ I T ] 2 Simplify Equation 3.⋅ Z ⋅ I T 2 Set Equation 3.5 – c k = 1–c (3.29 shows that the terms inside the brackets on both sides of the equal sign need to be set equal.29) (3.32) (3.10 the total voltage drop down the line is given by: Vdrop total = Re [ k ⋅ Z ⋅ I T + ( 1 – k ) ⋅ Z ⋅ c ⋅ I T ] where Z = total line impedance in ohms k = factor of the total line length where the ﬁrst part of the load current is modeled c = factor of the total current to place at the end of the line such that IT = Ix + c ⋅ IT In Section 3.= [ k + ( 1 – k ) ⋅ c ] 2 Solve Equation 3. that is 1 . The total threephase power loss in Figure 3.31 for k: 0.30) The same procedure can be followed for the power loss model.27) Equation 3.17 equal to Equation 3.⋅ Z ⋅ I T = Re [ k ⋅ Z ⋅ I T + ( 1 – k ) ⋅ Z ⋅ c ⋅ I T ] 2 (3. In Figure 3.33) .4.27: 1 Vdrop total = Re .30 by dividing both side of the equation by ZIT : 1 .1 it was shown that the total voltage drop down the line is given by: 1 Vdrop total = Re .28) (3.
⋅ R ⋅ I T = [k ⋅ R ⋅ IT + (1 – k) ⋅ R ⋅ (c ⋅ IT ) ] 3 1 2 .34 and simplify: 1 2 2 2 .34) Equate the terms inside the brackets of Equations 3.⋅ ( 1 – c ) + c 3 1–c Solving Equation 3.36) (3.54 Distribution System Modeling and Analysis The model for the power loss of Figure 3.11 Exact lumped load model. FIGURE 3.37) (3.38 is that onethird of the load should be placed at the end of the line.= [ k + ( 1 – k ) ⋅ ( c ) ] 3 1 2 2 2 2 . .= [ k + c – k ⋅ c ] = [ k ⋅ ( 1 – c ) + c ] 3 Substitute Equation 3.37 and 3.38) (3.37 into Equation 3.5 – c 2 2 .= .35: 1 0.33 and 3.36 for c results in: 1 c = 3 Substitute Equation 3.9 gives the total threephase power loss as: 1 2 Ploss total = 3 ⋅ . Figure 3.35) The interpretation of Equations 3. and twothirds of the load placed onefourth of the way from the source end.11 gives the ﬁnal exact lumped load model.32 into Equation 3.⋅ R ⋅ I T 3 (3.32 and solve for k: 1 k = 4 (3.
Figure 3. It is desired to determine the total voltage drop and the total threephase power loss down the primary main from node n to node m.5 Lumping Loads in Geometric Conﬁgurations Many times feeder areas can be represented by geometric conﬁgurations such as rectangles. and trapezoids. The feeder area is assumed to have a constant load density with threephase laterals uniformly tapped off of the primary main.Approximate Methods of Analysis 55 3. .5di IT n x i m w .5 di l FIGURE 3.5. the following deﬁnitions will apply: D PF z l w kVLL = = = = = load density in 2 mile assumed lagging power factor line impedance in Ω/mile length of the area width of the area kVA = nominal linetoline voltage in kV It will also be assumed that the loads are modeled as constant current loads. For all of the geographical areas to be evaluated. triangles. The approximate calculations can aid in the determination of the maximum load that can be served in a speciﬁed area at a given voltage level and conductor size. approximate calculations can be made for computing the voltage drop and total power losses. By assuming a constant load density in the conﬁgurations.12 is a model for the rectangular area.1 The Rectangle A rectangular area of length l and width w is to be served by a primary main feeder. dx .12 Constant load density rectangular area.12 represents a rectangular area of constant load density being served by a threephase main running from node n to node m. Figure 3. 3.
43 gives the same result as that of Equation 3. with the load modeled at the centroid.– cos ( PF ) 3 ⋅ kV LL (3. the voltage drop computed to the load point will represent the total voltage drop from node n to node m.39) An incremental segment is located x miles from node n.40) (3. which was derived for loads uniformly distributed along a feeder. It must be understood that in Figure 3.= I T ⋅ 1 –  l l The voltage drop in the incremental segment is x dV = Re ( z ⋅ i ⋅ dx ) = Re z ⋅ I T ⋅ 1 –  ⋅ dx l The total voltage drop down the primary main feeder is Vdrop = (3.⋅ Z ⋅ I T 2 2 where Z=z⋅l (3. The bottom line is that the total load of a rectangular area can be modeled at the centroid of the rectangle as shown in Figure 3.43) Equation 3.41) (3.⋅ l = Re .13. The only difference is the manner in which the total current (IT) is determined. The incremental current serving the load in the incremental segment is given by: IT di = .13.A/mile l The current in the incremental segment is given by: IT x i = I T – x ⋅ di = I T – x ⋅ . .56 Distribution System Modeling and Analysis The total current entering the area is given by: D⋅l⋅w –1 / I T = .42) ∫0 dV l l x = Re z ⋅ I T ⋅ ∫ 1 –  ⋅ dx l 0 Evaluating the integral and simplifying: 1 1 Vdrop = Re z ⋅ I T ⋅ .17.
The power loss in the incremental length is x 2 2 dp = 3 ⋅ i ⋅ r ⋅ dx = 3 ⋅ I T ⋅ 1 –  ⋅ r ⋅ dx l x x 2 = 3 ⋅ r ⋅ I T ⋅ 1 – 2 ⋅ .14.Approximate Methods of Analysis 57 IT n 1 2 m l IT w l FIGURE 3.14 represents the total power loss from node n to node m.+  ⋅ dx l l2 The total threephase power loss down the primary main is P loss = l x x 2 ∫0 dp = 3 ⋅ r ⋅ I T ⋅ ∫0 1 – 2 ⋅ .26. A similar derivation can be done in order to determine the total threephase power loss down the feeder main. again. The only difference.⋅ r ⋅ l ⋅ I T = 3 ⋅ .+  ⋅ dx l l2 l 2 2 2 Evaluating the integral and simplifying: 1 1 2 2 P loss = 3 ⋅ . is the manner in which the total current IT is determined.44 gives the same result as that of Equation 3. The model for computing the total threephase power loss of the primary main feeder is shown in Figure 3. it must be understood that the power loss computed using the model of Figure 3.44) Equation 3.13 Rectangle voltage drop model.⋅ R ⋅ I T 3 3 where R=r⋅l (3. Once again. .
84 A 3 ⋅ kV LL 3 ⋅ 4. Example 3. The load density of the area is 2500 kVA/mile with a power factor of 0. the impedance of the line was computed to be z = 0.16 kV and 12.000 l = . the total threephase power loss.6272 ) ⋅ 1. The primary main feeder uses 336.6 –1 I T = . also.6 It is proposed to serve a rectangular area of length 10. and width of 2 6000 ft.16 kV.15.16 2 . the total area current is kVA 5380. From Example 3.9 lagging.3.400 26/7 ACSR on a pole conﬁgured as shown in Example 3.000 ft.9 ) = 746.= .1522 = 5380. Compute.47 kV.1522 miles The total load of the area is kV = D ⋅ A = 2500 ⋅ 2. The area to be served is shown in Figure 3.2./ – cos ( 0.306 + j0.14 Rectangle power loss model.8939 = 0.306 + j0.8939 miles 5280 and 6000 w = .6 kVA A The total impedance of the line segment is Z = z ⋅ l = ( 0.= 1.1879 Ω For a nominal voltage of 4.6272 Ω/mile The length and width of the area in miles are 10.2.5795 + j1. Figure 3.1364 miles 5280 The total area of the rectangular area is A = l ⋅ w = 2.= 1. The question at hand is what minimum standard nominal voltage level can be used to serve this area without exceeding a voltage drop of 3% down the primary main? The choices of nominal voltages are 4.7/ – 25.58 Distribution System Modeling and Analysis IT n 1 3 m w l IT l FIGURE 3.
5795 + j1.⋅ ( 0.⋅ 100% = . The total voltage drop down the primary main is 1 1 V drop = Re .000' 10.1879 ) ⋅ 249.16% V LN 2401.Approximate Methods of Analysis 59 I T 6.8 It is clear that the nominal voltage of 4.47 The total voltage drop down the primary main is 1 1 V drop = Re . For a nominal voltage of 12.84 2 2 = 388.0%.6 –1 I T = .⋅ ( 0.⋅ Z ⋅ I T = Re .⋅ 100% = 16.000' FIGURE 3.⋅ Z ⋅ I T = Re .5795 + j1.84 2 2 = 129./ – cos ( 0.9 ) = 249. the total area current is kVA 5380.= .5 V .8 V 3 The percent voltage drop is V drop 388.1879 ) ⋅ 746.1 V % = .15 Example 3.1/ – 25.1/ – 25.8 rectangular area.47 kV.7/ – 25.84 A 3 ⋅ kV LL 3 ⋅ 12.= 2401.1 V The nominal linetoneutral voltage is 4160 V LN = .16 kV will not meet the criteria of a voltage drop less than 3.
⋅ 100% = 1.2 The Triangle Figure 3.60 Distribution System Modeling and Analysis The nominal linetoneutral voltage is 12.5 V % = .80% V LN 7199.= 7199. the total threephase power loss down the primary main is 2 1 .16 represents a triangular area of constant load density being served by a threephase main running from node n to node m.5di IT n w1 m w .⋅ 0.= 3 ⋅ . For the 12.5di x dx l FIGURE 3. It is desired to determine the total voltage drop and the total threephase power loss down the primary main from node n to node m.0% voltage drop constraint.5795 ⋅ 249.1 ⋅ R ⋅ IT 3 = 3 ⋅ .6 The nominal voltage of 12.= 35.965 kW 1000 1000 P loss 1 3 2 3.47 kV is more than adequate to serve this load.5. .6 V 3 The percent voltage drop is V drop 129. It would be possible at this point to determine how much larger the area could be and still satisfy the 3.47 kV selection.⋅ 100% = .470 V LN = . .16 Constant load density triangular area.
⋅ .52) .A/mile 1 l⋅w Area .– cos ( PF ) A 3 ⋅ kV LL Let IT IT 2 ⋅ IT 2 di = .⋅ x ⋅ .50 into Equation 3. w w 1 = x ⋅ l (3.49) (3.45) (3.51) (3.= .46) (3.47) The current entering the incremental line segment is i = I T – A 1 ⋅ di where A1 = area of triangle up to the incremental line segment.Approximate Methods of Analysis The area of the triangle is 1 Area = . = .50) (3.i = I T – .⋅ x 2 l 2 l 2 Substituting Equations 3. By similar triangles.⋅ x ⋅ x ⋅ .= .A 1 = .48) The area of the small triangle up to the incremental line segment is 1 w 1 w 2 1 .48: 1 w 2 2 x .⋅ I T = I T ⋅ 1 –  2 2 l l ⋅ w l The voltage drop in the incremental line segment is given by: x dv = Re [ i ⋅ z ⋅ dx ] = Re z ⋅ I T ⋅ 1 –  ⋅ dx 2 l The total voltage drop from node n to node m is V drop = l x 2 ∫0 dv = Re z ⋅ I T ⋅ ∫0 1 –  ⋅ dx l l 2 2 2 (3.⋅ .⋅ l ⋅ w 2 The total current entering the area is given by: D ⋅ Area –1 / I T = .47 and 3.⋅ l ⋅ w 2 61 (3.⋅ x ⋅ w 1 = .
The model for the voltage drop calculation is shown in Figure 3.17.⋅ Z T ⋅ I T 3 3 where ZT = z ⋅ l (3.53 shows that the total voltage drop from the vertex to the base of the triangular area can be computed by modeling the total triangle load twothirds of the distance between the vertex and the base of the triangle.+  ⋅ dx 2 4 2 l l l The total threephase power loss from node n to node m becomes: P loss = l x x 2 2 4 ∫0 dp = 3 ⋅ r ⋅ I T ⋅ ∫0 1 – 2 ⋅ .51 into Equation 3.54: x x x 2 2 dp = 3 ⋅ r ⋅ I T ⋅ 1 –  ⋅ dx = 3 ⋅ r ⋅ I T ⋅ 1 – 2 ⋅ .⋅ l = Re .53) Equation 3.62 Distribution System Modeling and Analysis IT n IT m w 2 3 l l FIGURE 3.17 Triangle voltage drop model. Evaluating the integral and simplifying: 2 2 V drop = Re z ⋅ I T ⋅ . The power loss in the incremental line segment is dp = 3 ⋅ [ r ⋅ i ⋅ dx ] Substituting Equation 3.+  ⋅ dx l l l 2 4 2 2 2 4 2 (3. A similar derivation can be made for the power loss model.54) .
000 l = .Approximate Methods of Analysis 63 IT n m w IT 8 15 l l FIGURE 3.55) Equation 3. Use the Kdrop factor from the line of Example 3.⋅ 2.8509 ⋅ 1. Example 3.= 2.3 the Kdrop factor was computed to be K drop = 0.= 1. From Example 3.55 gives the total threephase power loss down the primary main from node n to node m.400 26/7 ACSR.6142 miles 2 and 6000 w = .18.9 lagging.3.2 and determine the percent voltage drop from node n to node m.8409 miles 5280 The area of the triangle is 1 2 Area = .47 kV.00035291% drop/kVAmile The length and width of the triangle in miles is 15. The conductor on the primary main is 336.1364 = 1. The load density of the area is 3500 kVA/mile at a power factor of 0.19 is to be served by a feeder of nominal 2 voltage 12.18 Triangle power loss model.1364 miles 5280 . and the conﬁguration of the pole is that of Example 3. The model for the power loss is given in Figure 3.⋅ R ⋅ I T 15 (3.7 The triangular area shown in Figure 3.2 in Figure 3. Evaluating the integral and simplifying: 8 2 P loss = 3 ⋅ .
5 ⋅ 2.0%. the percent drop to node m is 2 2 V drop = . Since this is assumed –1 .7761% 3 3 Suppose now that a shunt capacitor bank is to be installed somewhere along the primary main in order to limit the percent voltage drop to node m to 3. Threephase rating of the capacitor bank 2.84 kVA = 5084.5/ – 25.⋅ K drop ⋅ kVA ⋅ miles = .64 Distribution System Modeling and Analysis IT n m 6.5 kVA A The total complex power of the triangular area is S = kVA / – cos ( PF ) = 5649.⋅ 0. Location of the capacitor bank The total reactive power of the area was computed to be 2462.6 Using the Kdrop factor and lumping the total load at the twothirds point.19 Example 3. That means that a capacitor bank rated up to 2462. The total load of the triangular area is kV = 3500 ⋅ 1.9 triangular area.000' 15.00035291 ⋅ 5649.6 + j2462.6 can be used without causing the feeder to go into a leading power factor condition.6142 = 5649.000' FIGURE 3.6 kvar. Two decisions must be made: 1.8409 = 3.
47 The total resistance of the primary main is R = r ⋅ l = 0.7761 dist = .306 ⋅ 2.20. Use the Krise factor from Example 3.= 1.⋅ l = 1.7761 The distance from node n is determined by: V rise 0.0 = 3. it is assumed that the load density is constant throughout the trapezoid.00040331 ⋅ 1800 The total threephase power loss down the primary main before the shunt capacitor is added is computed by lumping the total triangular load at: 8 l Load = .5: K rise = 0.0691 miles K rise ⋅ kvar 0.8693 Ω The total threephase power loss down the primary main is 3 8 3 8 2 2 P loss = .5./ – cos ( PF ) = 261.0%.8693 ⋅ 261.8409 = 0.5 –1 I T = .6/ – 25.= .⋅ 0.⋅ . Depending upon the load curve during the day. this bank may or may not have to be switched.⋅ .84 A 3 ⋅ kV LL 3 ⋅ 12.5 and determine how far from node n the capacitor bank should be installed in order to limit the voltage drop to 3. a capacitor bank rated at 1800 kvar (threephase) will be used in order to prevent a leading power factor condition for a smaller load.5151 miles from node n 15 The total load current is kVA 5649. From Example 3.00040331% rise/kvarmile The needed voltage rise due to the capacitor is V rise = V drop – 3.0 = 0. As before.3 The Trapezoid The ﬁnal geometric conﬁguration to consider is the trapezoid.⋅ R ⋅ I T = .Approximate Methods of Analysis 65 to be the peak load.6 = 95.7761 – 3. The general model of the trapezoid is shown in Figure 3. .16 kW 1000 15 1000 15 3.= .
5 di n x wx i .⋅ ( w 2 + w 1 ) ⋅ l 2 The current that is delivered to the trapezoid abef is D ⋅ Area x I x = 3 ⋅ kV LL where Areax = area of the trapezoid abef.20 General trapezoid.56) . 1 Area T = .59) (3. d Figure 3.57) (3. 1 Area x = .20 represents a trapezoidal area of constant load density being served by a threephase primary running from node n to node m. It is desired to determine the total voltage drop and the total threephase power loss down the primary main from node n to node m.66 Distribution System Modeling and Analysis dx a IT c b . The total current entering the trapezoid is D ⋅ Area T I T = 3 ⋅ kV LL where AreaT = total area of the trapezoid.60) (3.56 for D: 3 ⋅ kV LL ⋅ I T D = Area T (3. It is necessary to determine the value of the current entering the incremental line segment as a function of the total current and the known dimensions of the trapezoid.58) (3.5 di e m w2 w1 f l FIGURE 3. The known dimensions will be the length l and the widths w1 and w2.⋅ ( w x + w 1 ) ⋅ x 2 Solving Equation 3.
21 will be used to establish the relationship between the unknown width and the known dimensions.62) (3. = .⋅ I T Area T 3 ⋅ kV Area T LL The current entering the incremental line segment is Area x i = I T – I x = I T ⋅ 1 – . Area T (3.64) . d Substitute Equation 3.61) The only problem at this point is that the area of the small trapezoid cannot be determined since the width wx is not known. Figure 3.⋅ y 2 l But 1 y 2 = . ⋅ .21: wx = w1 + 2 ⋅ yx From similar triangles: x y x = .⋅ ( w 2 – w 1 ) 2 (3. Referring to Figure 3.60 into Equation 3.Approximate Methods of Analysis 67 b a w1 f e wx x i yx y2 w2 l FIGURE 3.21 Trapezoid dimensions.63) (3.65) (3.58: 3 ⋅ kV LL ⋅ I T Area x Area x I x = .
59: x 1 x Area x = .57 and 3.71) .68) 1 2 x (3.⋅ ( w 1 + w 2 ) ⋅ l – 2 ⋅ w 1 ⋅ x + w 1 ⋅ .⋅ w 2 + w 1 ⋅ x l 2 l Substitute Equations 3.66 into Equation 3.65 into Equation 3.⋅ ( w 2 – w 1 ) l 2 Substitute Equation 3.68 Distribution System Modeling and Analysis Substitute Equation 3.70: 2 2 IT x x dv = Re z ⋅ .69 into Equation 3.64: x 1 .⋅ ( w 2 – w 1 ) = w 1 + .68 into Equation 3.⋅ w 2 ) ] ⋅ x l l i = I T ⋅ 1 – 1 .⋅ l – 2 ⋅ x +  ⋅ w 1 + l –  ⋅ w 2 ( w1 + w2 ) ⋅ l l l (3.62: ⋅ [ ( w 1 ⋅ ( 2 – x ) + .⋅ .⋅ ( w 2 – w 1 ) = w 1 ⋅ 1 –  + .⋅ w 2 l 2 l l l (3.20 is given in Equation 3.70) (3.66) (3.⋅ . The voltage drop in the incremental line segment is given by: dv = Re [ z ⋅ i ⋅ dx ] Substitute Equation 3.y x = .67) Substitute Equation 3.⋅ w 1 ⋅ 1 –  + .63: x 1 x x x .69 and will be used to compute the voltage drop and power loss in the incremental line segment.⋅ ( w 2 + w 1 ) ⋅ l 2 2 2 IT x x i = .67 into Equation 3.w x = w 1 + 2 ⋅ .69) The current entering the incremental line segment of Figure 3.– w 2 ⋅ ( w1 + w2 ) ⋅ l l l 2 2 IT x x i = .⋅ l – 2 ⋅ x +  ⋅ w 1 + l –  ⋅ w 2 ⋅ dx l l ( w1 + w2 ) ⋅ l (3.
72) Equation 3.⋅ Z ⋅ I T 3 (3.⋅ ∫ l – 2 ⋅ x +  ⋅ w 1 + l –  ⋅ w 2 ⋅ dx ∫0 ( w1 + w2 ) ⋅ l 0 l l l V drop = Evaluating the integral and simplifying results in: w1 + 2 ⋅ w2 V drop = Re Z ⋅ I T ⋅ .76) Equation 3. Let w1 = w2 = w (3. THE RECTANGLE For a rectangular area the two widths w1 and w2 will be equal. 3 ⋅ ( w + w ) V drop 1 = Re .76 is the same as was derived for the triangular area.⋅ Z ⋅ I T 2 3⋅w = Re Z ⋅ I T ⋅ 6⋅w (3.73 into Equation 3. 3 ⋅ ( w 1 + w 2 ) (3. THE TRIANGLE For a triangular area the width w1 will be zero. The total threephase power loss down the line segment can be developed by starting with the derived current in the incremental segment as given by .73) Substitute Equation 3.75) Substitute Equation 3.72: w+2⋅w V drop = Re Z ⋅ I T ⋅ .72 is very general and can be used to determine the models for the rectangular and triangular areas.74) Equation 3.74 is the same that was initially derived for the rectangular area.75 into Equation 3.Approximate Methods of Analysis 69 The total voltage drop down the primary from node n to node m is given by: 2 2 l z ⋅ IT x x dv = Re .72: 01 + 2 ⋅ w2 V drop = Re Z ⋅ I T ⋅  3 ⋅ ( 0 + w 2 ) 2 = Re . Let w1 = 0 (3.
77) (3.44 that was previously derived for the rectangular area.70 Distribution System Modeling and Analysis Equation 3. RECTANGLE For the rectangle. 2 2 15 ⋅ ( w + w ) 15 ⋅ ( 2 ) 1 2 P loss = 3 ⋅ . = 3 ⋅ R ⋅ I T ⋅ . the width w1 is zero. The threephase power loss in the incremental segment is dp = 3 ⋅ r ⋅ i dx The total threephase power loss down the line segment is then: P loss = 3 ⋅ r ⋅ ∫ i dx 0 l 2 2 (3.81 is the same as the Equation 3.80. the two widths w1 and w2 are equal.69.⋅ ∫ l – 2 ⋅ x +  ⋅ w 1 + l –  ⋅ w 2 dx 2 2 l l 0 ( w1 + w2 ) ⋅ l 2 (3. 2 15 ⋅ ( w 1 + w 2 ) where R=r·l 2 2 (3.78) Substitute Equation 3.81) Equation 3.69 into Equation 3. Let w = w1 = w2 Substitute into Equation 3. Let w1 = 0 .79: 2 2 8⋅w +9⋅w⋅w+3⋅w 8+9+3 2 2 P loss = 3 ⋅ R ⋅ I T ⋅ .80) The rectangular and triangular areas are special cases of Equation 3. TRIANGLE For the triangular area.79) Evaluating the integral and simplifying results in: 8 ⋅ w2 + 9 ⋅ w1 ⋅ w2 + 3 ⋅ w1 2 P loss = 3 ⋅ R ⋅ I T ⋅ .78 and simplify: 2 2 2 l r ⋅ IT x x P loss = 3 ⋅ .⋅ R ⋅ I T 3 (3.
PWS Publishing Co. References 1. 1. once inside the ballpark more precise values of voltage drop and power loss are needed. Many times only a ballpark value is needed.. The conductors are 250.⋅ R ⋅ I T 2 15 15 ⋅ ( 0 + w 2 ) 2 2 71 (3. and Sarma. More often than not. = 3 ⋅ .000 cm. AA. These techniques are very useful for making quick calculations that will be ballpark values. which was previously derived for the total power loss in a triangular area.55. The remainder of this text will be devoted to the more precise methods for analyzing a distribution feeder under balanced and unbalanced. Power System Analysis and Design. .88 lag.80: 8 ⋅ w2 + 9 ⋅ 0 ⋅ w2 + 3 ⋅ 0 8 2 2 P loss = 3 ⋅ R ⋅ I T ⋅ . Determine the Krise factor. M. Boston. 1994. Determine the Kdrop factor assuming a power factor of 0..4 kV. and for geometric areas with constant load densities.1 Shown in Figure 3.Approximate Methods of Analysis Substitute into Equation 3. 2nd edition.82 is the same as Equation 3. The nominal linetoline voltage of the feeder is 14.82) Equation 3. 3. steadystate. 2. 3.D. Determine the series impedance per mile of this line. J. CON Lay. This will be especially true when the unbalanced nature of a distribution feeder is taken into account. and shortcircuit conditions.6 Summary This chapter has been devoted to the development of some useful techniques for computing the voltage drop and power loss of line segments with uniformly distributed loads. Problems 3. Glover.22 is the pole conﬁguration of conductors for a threephase primary feeder.
2. Determine the percent voltage drop to node E4.50 miles 500 kVA E2 0. 150 kVA 100 kVA 300 kVA 425 kVA 500 kVA The phase conductors are 4/0 ACSR and are conﬁgured on an 8ft. Determine the series impedance of the line segment in Ω/mile.3 0 0.24. and D ca = 7.225 mile 5 6 0.2 A 4.00298639% drop/kVAmile The Krise = 0. 3. 3. threephase feeder is shown in Figure 3. 1.5 ft.23. Determine the rating of a threephase shunt capacitor bank to be placed at E3 to limit the voltage drop to E4 to 5.175 mile 2 0. Determine the total percent voltage drop to node 6. crossarm with phase spacings of: D ab = 2.3.9 miles 750 kVA E4 FIGURE 3. 3. .0 ft. 2. E1 0. Determine the Kdrop and Krise factors assuming a load power factor of 0.22 Problem 3.9 lagging.2.65 miles 1200 kVA E3 0.2 mile 3 4 0..1 conﬁguration.16 threephase primary feeder is shown in Figure 3. The Kdrop = 0.5 ft.15 mile A 4160V.125 mile 200 kVA FIGURE 3.125 mile 0..24 System for Problem 3.00334353% rise/kvarmile 1. 1 0.72 Distribution System Modeling and Analysis a b c n 2' 2' 2' 25' FIGURE 3.0%. D bc = 4.23 System for Problem 3.
while the threephase lateral conductors are 266.25 illustrates the new feeder. from the substation. . 3. 0. The % voltage drop to the last customer in the ﬁrst lateral (point A).. The total threephase power loss for the total area. Determine the percent voltage drop to the last load point and the total threephase power loss for the feeder shown in Figure 3. Use the exact lumped load model of Figure 3. and 7. 2. Determine: 3.5 ft. 2. Lump the total feeder load at the midpoint of the feeder and compute the percent voltage drop to the end of the feeder. Figure 3.0 ft.800 26/7 ACSR. The rectangular area in Figure 3. poles with 8ft. Flash has decided that he will use 336.500 26/7 ACSR.25 System for Problem 3.5 1.0%. 1. 73 Flash Thunderbolt. The primary main and the laterals are constructed so that the equivalent spacing (Deq) is 3.4 Total length of feeder = 5000 ft. Load: 10–500 kVA (threephase). Determine the threephase kvar rating of a shunt capacitor to be placed at node 4 to limit the total percent voltage drop to node 6 to 3. The nominal voltage of the area being served is 4. with the ﬁrst load 500 ft.5 ft.. junior engineer for Tortugas Power and Light.Approximate Methods of Analysis 4. 4. has been given an assignment to design a new 4.11 and compute the percent voltage drop to the end of the line.26 has a uniform load density of 2000 2 kVA/mile at 0.4.9 lagging power spaced every 500 ft. Voltage drop: not to exceed 5% from the sub to the last load.16kV.16 kV. crossarms. 3. threephase feeder that will have the following characteristics: 3.5 ft.25. and the total threephase power loss down the line.400 26/7 ACSR (Linnet) conductors constructed on 45ft. The threephase primary main conductors are 556. 500' 500' 500' 500' 500' 500' 500' 500' 500' 500' SUB FIGURE 3.9 lagging power factor. The % voltage drop to the last customer in the last lateral (point B). The spacings of the conductors on the crossarms are 2.
The big black line will be the threephase primary main consisting 3.74 Distribution System Modeling and Analysis B 2.500' A 12.47kV substation that is located in the center of the square.5 mile 1.27 is a rectangletriangle area that is being fed from 2 a source at point X. The two plans are shown in Figure 3. Two different plans are being considered for serving the area. Determine the percent drop to point Z. what now is the percent drop to point Z? A square area of 20. and 0. Shown in Figure 3.000' FIGURE 3. and the Krise factor for the primary main conductors is 0.6 1.6.000 ft.00028436% rise/kvarmile. 2. In addition to the uniformly distributed loads. The Kdrop factor for the primary main conductors is 0. PlanA proposes to break the area into four square areas and serve it as shown. on a side has a load density of 2000 kVA/ 2 mile . With the capacitor in place.7 .5.9 lagging power factor is to be served from a 12.5 mile X 1 mile Y Z 2000 kVA FIGURE 3. Determine the kVAr rating (to the nearest 300 kVAr/phase) for a capacitor bank to be placed at point Y in order to limit the voltage drop to Z to 3%. 3.00022626% drop/kVAmile.26 Rectangular area for Problem 3. 3. 1.500' Source 2.28.27 Rectangulartriangular area of Problem 3. Both areas have a load density of 6000 kVA/mile . there is a “spot load” at point Z that is 2000 kVA. with loads being uniformly distributed as denoted by the dashed laterals.
29 Areas for Problem 3.8 Shown in Figure 3.5 mile 0.75 mi S a e 0.5 mile 1.5 mile 0. Again. 0.A 75 Plan .5 mile d c S' b 0.7.B A FIGURE 3. Under an emergency condition the switch at b is closed so that the feeder normally serving the triangle area must now serve both areas. Determine the percent voltage drop to the “last customer” (points A and B) for the two plans. 3.28 Two plans for Problem 3. Both the main and laterals are constructed such that Deq = 4.29 are the areas normally served by two feeders. B of 336/400 26/7 ACSR conductors. PlanB proposes to serve the area with four triangularly shaped feeders.5 mi FIGURE 3. The same conductors and Deq will be used in this plan.8.3795 ft. The threephase laterals will be spaced every 500 ft. and the dotted lines will be the threephase laterals consisting of 4/0 ACSR conductors. and are shown as the dotted lines. and the laterals are spaced every 500 ft. Assume both .Approximate Methods of Analysis Plan . the primary main is shown in the dark black line.
The primary feeder voltage is 13.9 lagging power factor.5 MVA/square mile. Determine the threephase kVAr rating of a shunt capacitor bank placed at c in order to limit the voltage drop to point d to 3. and 0. Determine the Kdrop and Krise factors. Determine the voltage drop to e with the source at S’ and the capacitor at c. . Determine the voltage drop to e with the capacitor bank at c. 4.0%. Laterals are uniformly tapped off of the primary main from S to a. 5. and laterals are tapped off from c to d and from c to S’.76 Distribution System Modeling and Analysis areas have a uniform load density of 2.8 kV. 1. The primary main conductors are 2/0 ACSR and are placed on a pole such that Deq = 4. No loads are tapped off the feed from a to b to c. Determine the voltage drop to point d. 2.3795 ft. 3.
1 Series Impedance of Overhead Lines The inductive reactance (self and mutual) component of the impedance is a function of the total magnetic ﬁelds surrounding a conductor. Figure 4. It is further assumed that the sum of the currents will be zero. Wbt/m D i2 D i1 GMR i D in (4.) 77 (4.2) where Din = Distance between conductor i and conductor n (ft. The currents in all conductors are assumed to be ﬂowing out of the page. That is I1 + I2 + · · Ii + · · In = 0 The total ﬂux linking conductor i is given by: 1 1 1 1 –7 λ i = 2 ⋅ 10 ⋅ I 1 ⋅ ln .1) . The series impedance of a singlephase.1 shows conductors 1 through n with the magnetic ﬂux lines created by currents ﬂowing in each of the conductors.+ · · I i ⋅ ln .) GMRi = Geometric Mean Radius of conductor i (ft.+ I 2 ⋅ ln . or threephase distribution line consists of the resistance of the conductors and the self and mutual inductive reactances resulting from the magnetic ﬁelds surrounding the conductors. The resistance component for the conductors will typically come from a table of conductor data such as found in Appendix A.+ · · I n ⋅ ln . twophase (Vphase).4 Series Impedance of Overhead and Underground Lines The determination of the series impedance for overhead and underground lines is a critical step before the analysis of a distribution feeder can begin. 4.
H/m GMR i 3 (4. D bc .H/m Ii GMR i λ in 1 –7 L in = . By deﬁnition: Self inductance: λ ii 1 –7 L ii = .1 Transposed ThreePhase Lines Highvoltage transmission lines are usually assumed to be transposed (each phase occupies the same physical position on the structure for onethird of the length of the line).1 Magnetic ﬁelds.3) Mutual inductance: (4.78 Distribution System Modeling and Analysis O n O O 1 2 1 2 O 1 D i2 D i1 On n O O i i 2 D in FIGURE 4. Phase inductance: D eq –7 L i = 2 ⋅ 10 ⋅ ln .H/m In D in (4. and D ca = distance between phases . With these two assumptions it is possible to combine the “self” and “mutual” terms 1 into one “phase” inductance.= 2 ⋅ 10 ⋅ ln .6) D ab . In addition to the assumption of transposition.4) 4.1. (4. The inductance of conductor i consists of the self inductance of conductor i and the mutual inductance between conductor i and all of the other n − 1 conductors.= 2 ⋅ 10 ⋅ ln . it is assumed that the phases are equally loaded (balanced loading).5) where D eq = D ab ⋅ D bc ⋅ D ca ft.
12134 ⋅ ln .Ω/mile D ij (4.Ω/mile GMR i 1 z ij = j0.4 are used to compute the self and mutual inductive reactances of the conductors.3 and 4. and the length of the conductor will be assumed to be one mile. Figure 4.Ω/mile GMR i (4.Ω/mile GMR i (4. the phase inductive reactance is given by: Phase reactance: D eq x i = ω ⋅ L i = 0.8) Untransposed Distribution Lines Because distribution systems consist of singlephase.2 illustrates a line consisting of two conductors (i and j) carrying currents (Ii and Ij) with the remote ends of the conductors tied to ground. With those assumptions the self and mutual impedances are given by: 1 z ii = r i + j0. Equations 4. Carson’s approach was to represent a line with the conductors connected to a source at one end and grounded at the remote end.12134 ⋅ ln .2 D eq z i = r i + j ⋅ 0.Series Impedance of Overhead and Underground Lines 79 Assuming a frequency of 60 Hz.9) (4.12134 ⋅ ln .12) (4. In Figure 4.2.7) The series impedance per phase of a transposed threephase line consisting of one conductor per phase is given by: Series impedance: 4. The ac resistance of the conductors is taken directly from a table of conductor data (Appendix A). A ﬁctitious “dirt” conductor carrying current Id is used to represent the return path for the currents. it is necessary to retain the identity of the self and mutual impedance terms of the conductors and take into account the ground return path for the unbalanced currents. The inductive reactance will be assumed to be at a frequency of 60 Hz. V ig = z ii ⋅ I i + z ij ⋅ I j + z id ⋅ I d – ( z dd ⋅ I d + z di ⋅ I i + z dj ⋅ I j ) Collect terms in Equation 4. taking into 2 account the return path of current through ground.11) .10) In 1926 John Carson published a paper in which he developed a set of equations for computing the self and mutual impedances of lines. Kirchhoff’s voltage law (KVL) is used to write the equation for the voltage between conductor i and ground.1.11: V ig = ( z ii – z di ) ⋅ I i + ( z ij – z dj ) ⋅ I j + ( z id – z dd ) ⋅ I d (4. and untransposed threephase lines serving unbalanced loads. twophase.12134 ⋅ ln .
17.– ln  GMR i GMR d D id D di D id ⋅ D di 1 ˆ z ii = r i + r d + j0.12 and collect terms: V ig = ( z ii + z dd – z di – z id ) ⋅ I i + ( z ij + z dd – z dj – z id ) ⋅ I j Equation 4.15) (4.2 Two conductors with dirt return path.14) (4. Substituting Equations 4. Note that in these two equations the effect of the ground return path is being folded into what will now be referred to as the “primitive” self and mutual impedances of the line. the “over bar” impedances are given by Equations 4. The equivalent primitive circuit is shown in Figure 4.9 and 4.10 into Equations 4.– ln .13 into Equation 4.13) In Equations 4.+ ln .14 is of the general form: ˆ ˆ V ig = z ii ⋅ I i + z ij ⋅ I j where ˆ z ii = z ii + z dd – z di – z id ˆ z ij = z ij + z dd – z dj – z id (4.16) (4.3.80 z ii + Vig + Vjg Id Ii Ij z jj Distribution System Modeling and Analysis z ij ground z dd z jd z id FIGURE 4.16 and 4.16 and 4.12134 ⋅ ln . the primitive self impedance is given by: ˆ z ii = r i + jx ii + r d + jx dd – jx id – jx di 1 1 1 1 ˆ z ii = r i + r d + j0. From Kirchhoff’s current law: Ii + I j + Id = 0 I d = –I i – I j Substitute Equation 4.+ ln .17.18) .12134 ⋅ ln .10. GMR i GMR d (4.9 and 4.17) (4.
In similar manner. the Geometric Mean Radius of dirt (GMRd). the most accurate analysis should not make any assumptions regarding the spacing between conductors.+ ln . Ddi .3 Carson’s Equations Since a distribution feeder is inherently unbalanced. conductor sizes.19) 4. every conductor at a given distance above ground has an image conductor the same distance below ground. and transposition.18 and 4.1. Carson made use of conductor images. Carson assumes the earth is an inﬁnite. The equations can also be applied to underground cables. .3 Equivalent primitive circuit. This is where John Carson’s work bails us out. Ddj ).Series Impedance of Overhead and Underground Lines 81 z ii + V ig + V jg Ii Ij z jj z ij + V'jg + V'ig  ground FIGURE 4. the primitive mutual impedance can be expanded: ˆ z ij = jx ij + r d + jx dd – jx dj – jx id 1 1 1 1 ˆ z ij = r d + j0.12134 ⋅ ln . In Carson’s 1926 paper he developed a technique whereby the self and mutual impedances for an arbitrary number of overhead conductors can be determined. (4. and the distances from the conductors to dirt (Did .19 is that we do not know the values of the resistance of dirt (rd).12134 ln . uniform solid with a ﬂat uniform upper surface and a constant resistivity.– ln . and are therefore neglected. With the advent of the digital computer. that is.+ ln . The technique was not met with a lot of enthusiasm because of the tedious calculations that would have to be done on the slide rule and by hand. Any end effects introduced at the neutral grounding points are not large at power frequencies.– ln  D ij D dj D id GMR d D dj ⋅ D id 1 ˆ z ij = r d + j0. D ij GMR d The obvious problem in using Equations 4. Carson’s equations have become widely used. In his paper. Djd .
1609344 × 10 Ω/mile RDi = radius of conductor i in feet GMRi = Geometric Mean Radius of conductor i in feet f = system frequency in Hertz ρ = resistivity of earth in Ωmeters Dij = distance between conductors i and j in feet (see Figure 4.4.20 and 4.4.20) Mutual Impedance between Conductor i and j: S ij ˆ z ij = 4 ω P ij G + j 2 ω G ⋅ ln .82 Distribution System Modeling and Analysis FIGURE 4.21.+ 4 ω Q ii G Ω/mile RD i (4. the original Carson equations are given in Equations 4.4 Conductors and images. This is illustrated in Figure 4. Self Impedance of Conductor i: S ii ˆ z ii = r i + 4 ω P ii G + j X i + 2 ω G ⋅ ln .21) . Referring to Figure 4.4) (4.+ 4 ω Q ij G Ω/mile D ij where ˆ z ii = self impedance of conductor i in Ω/mile ˆ z ij = mutual impedance between conductors i and j in Ω/mile ri = resistance of conductor i in Ω/mile ω = 2π f = system angular frequency in radians per second −3 G = 0.
4 Modiﬁed Carson’s Equations Only two approximations are made in deriving the modiﬁed Carson’s equations. 16 k ij 8 3 2 1 1 2 Q ij = – 0.24) (4.20 RD i S ii ˆ z ii = r i + 4ω P ii G + j 2ω G ⋅ ln .565 × 10 ⋅ S ij ⋅ ρ 2 83 (4.+ 2ω G ⋅ ln .⋅ ln .22) into Equation 4. π P ij = 8 1 2 Q ij = – 0.30) .Ω/mile GMR i k ij 1 2 π P ij = .Series Impedance of Overhead and Underground Lines Sij = distance between conductor i and image j in feet (see Figure 4.0386 + .k ij cos ( θ ij ) 2 k ij 3 2 f –4 k ij = 8.26) (4.03860 + .28) (4.+ 2Q ij D ij (4.+ 2Q ii GMR i Simplify Equation 4.22) (4.4) θij = angle between a pair of lines drawn from conductor i to its own image and to the image of conductor j (see Figure 4.+ .27) (4.4) RD i X i = 2 ω G ⋅ ln .– .21: S ij ˆ z ij = 4 ω P ij G + j2 ω G ln .+ 4ω Q ii G GMR i RD i Combine terms and simplify: S ii ˆ z ii = r i + 4 ω P ii G + j2 ω G ln .25) 4.k ij cos ( θ ij ) + .6728 + ln .ln 2 k ij Substitute Xi (Equation 4.cos ( 2 θ ij ) ⋅ 0. These approximations involve the terms associated with Pij and Qij by using only the ﬁrst term of the variable Pij and the ﬁrst two terms of Qij.29) (4.23) (4.1.
+ 2Q ii GMR i S ij 2 ˆ z ij = π fG + j4 π fG ln .+ 7.31) (4.8393 – .03860 + .2Q ij = 7.z ii = r i + π fG + j4 π fG ln .ln  + .565 ⋅ 10 ⋅ S ij ⋅ .ln  2 f –4 8.6786 – ln S ij + .ln . D ij 2 f 1 ρ 1 ˆ .36 into Equation 4.6786 – ln S ij + .ln . ρ Expand: 1 1 1 1 2 ρ Q ij = – 0. GMR i 2 f 1 ρ 1 ˆ .37) Substitute Equation 4.Q ij = 3.+ 7.84 Distribution System Modeling and Analysis Substitute expressions for P (Equation 4.34 can be reduced to 1 1 ρ .ln .+ 7.ln 2 8. D ij 2 f 2 (4.36 into Equation 4.27): 1 2 Q ij = – 0.ln .27) and ω ( 2 ⋅ π ⋅ f ): S ii 2 ˆ z ii = r i + π fG + j4 π fG ln .ln S ij + .03860 + .35) (4.+ 2Q ij D ij (4. GMR i 2 f 2 (4.ln 2 4 f or 1 ρ .32 and simplify: S ij 1 ρ 2 ˆ .z ij = π fG + j4 π fG ln .25) into the approximate expression for Qij (Equation 4.32) Substitute the expression for kij (Equation 4.33) (4.38) .+ .36) Substitute Equation 4.34) (4.6786 + .ln .ln 2 f (4.z ij = π fG + j4 π fG ln .z ii = r i + π fG + 4 π fG ln .6786 + .+ 7.565 ⋅ 10 –4 2 S ij 2 f Equation 4.6786 – ln S ii + .31 and simplify: S ii 1 ρ 2 ˆ .
40) (4.09530 + j0.ln .42) Recall that Equations 4.41 and 4.+ 7.39) GMR i 2 f 1 ρ 1 ˆ .44) The modiﬁed Carson’s equations will be used to compute the primitive self and mutual impedances of overhead and underground lines.+ 7.93402 GMR d GMR d (4.93402 Ω/mile. An overhead fourwire grounded wye distribution line segment will result in a 4 × 4 matrix.42 are used to compute the elements of an ncond x ncond primitive impedance matrix. the modiﬁed Carson’s equations are 1 ˆ z ii = r i + 0.+ 7.09530 Ω/mile D dj ⋅ D id D id ⋅ D di ln .42 shows that the modiﬁed Carson’s equations have deﬁned the missing parameters.5 Primitive Impedance Matrix for Overhead Lines Equations 4. A comparison of Equations 4. D ij 2 f It is now assumed: f = Frequency = 60 Hertz ρ = Earth resistivity = 100 Ohmmeter Using these approximations and assumptions.18 and 4. 4.6786 + .19 to Equations 4.1.Series Impedance of Overhead and Underground Lines Substitute in the values of π and G: 85 1 1 ρ ˆ . and the various distances from conductors to dirt were not known. the GMR of dirt.12134 ln . For an underground grounded wye line segment consisting of three concentric neutral cables.00202237 ⋅ f ln .12134 ln .+ 7.00202237 ⋅ f ln .43) (4. A comparison of the two sets of equations shows that: r d = 0.00158836 ⋅ f + j0. the .z ii = r i + 0.00158836 ⋅ f + j0.= 7.09530 + j0.z ij = 0.18 and 4.= ln .41 and 4.ln . (4.41) (4.19 could not be used because the resistance of dirt.6786 + .93402 Ω/mile GMR i 1 ˆ z ij = 0. D ij (4.
Equation 4.6 Phase Impedance Matrix for Overhead Lines For most applications the primitive impedance matrix needs to be reduced to a 3 × 3 phase frame matrix consisting of the self and mutual equivalent impedances for the three phases. The primitive impedance matrix for a threephase line with m neutrals will be of the form ˆ z aa ˆ z ba ˆ [ z primitive ] = ˆ z ca ˆ z n1a ˆ z ab ˆ z bb ˆ z ca ˆ z n1b ˆ z ac ˆ z bc ˆ z cc ˆ z n1c   ˆ z an1 ˆ z bn1 ˆ z an2 ˆ z bn2 ˆ z cn2 ˆ z n1n2 ˆ z anm ˆ z bnm ˆ z cnm ˆ z n1nm (4. .45 becomes ˆ [ z ij ] ˆ [ z nj ] ˆ [ z in ] ˆ [ z nn ] ˆ [ z primitive ] = (4.46) 4. One standard method of reduction is the Kron reduc3 tion. The assumption is made that the line has a multigrounded neutral as z aa + V ag + V bg + V cg + V ng Ia Ib Ic z bb z cc z nn In z ab z bc z z ac z bn z an + V'cg + V'ng + V'ag + V'bg cn       FIGURE 4.1.86 Distribution System Modeling and Analysis resulting matrix will be 6 × 6.5 Fourwire grounded wye line segment.5 shows a fourwire grounded neutral line segment.ˆ  z n1n1 ˆ  z n2n1  ˆ z nmn1 ˆ ˆ z n2n2 z n2nm ˆ ˆ z nmn2 z nmnm In partitioned form.45) ˆ ˆ ˆ z n2a z n2b z n2c ˆ ˆ ˆ z nma z nmb z nmc ˆ  z cn1 . Figure 4.
50 for [In]: [ I n ] = – [ z nn ] ⋅ [ z nj ] ⋅ [ I abc ] Substitute Equation 4.Series Impedance of Overhead and Underground Lines 87 shown in Figure 4.51) (4.51 into Equation 4.48) Because the neutral is grounded. n Substituting those values into Equation 4.47) In partitioned form. the voltages Vng and V′ g are equal to zero.54) .53 is the ﬁnal form of the Kron reduction technique. The Kron reduction method applies Kirchhoff’s voltage law to the circuit. The ﬁnal phase impedance matrix becomes: z aa z ab z ac [ z abc ] = z ba z bb z bc Ω/mile z ca z cb z cc (4.49) (4.5.50) (4.52) (4.53) Equation 4.49: [ V abc ] = [ V′ abc ] + ( [ z ij ] – [ z in ] ⋅ [ z nn ] ⋅ [ z nj ] ) ⋅ [ I abc ] [ V abc ] = [ V′ abc ] + [ z abc ] ⋅ [ I abc ] where [ z abc ] = [ z ij ] – [ z in ] ⋅ [ z nn ] ⋅ [ z nj ] –1 –1 −1 (4.47 becomes [ V abc ] [ V ng ] [ V′ abc ] [ V′ ng ] [ z ij ] [ z in ] [ z nj ] [ z nn ] [ I abc ] [In] = + ⋅ (4. Equation 4.48 and expanding results in: [ V abc ] = [ V′ abc ] + [ z ij ] ⋅ [ I abc ] + [ z in ] ⋅ [ I n ] [ 0 ] = [ 0 ] + [ z nj ] ⋅ [ I abc ] + [ z nn ] ⋅ [ I n ] Solve Equation 4. V ag V bg V cg V ng = V′ ag V′ bg V′ cg V′ ng + ˆ z aa ˆ z ba ˆ z ca ˆ z na ˆ z ab ˆ z bb ˆ z cb ˆ z nb ˆ z ac ˆ z bc ˆ z cc ˆ z nc ˆ z an ˆ z bn ˆ z cn ˆ z nn ⋅ Ia Ib Ic In (4.
Kron reduction will reduce the matrices to 2 × 2 and a single element. The phase impedance matrix can be used to accurately determine the voltage drops on the feeder line segments once the currents have been determined. For twophase (Vphase) and singlephase lines in grounded wye systems. These matrices can be expanded to 3 × 3 phase frame matrices by the addition of rows and columns consisting of zero elements for the missing phases. the matrix will be symmetrical. and the offdiagonal terms will not be equal to each other. the diagonal terms of Equation 4. the phase impedance matrix would be z ab 0 z ac [ z abc ] = 0 0 0 Ω/mile z ca 0 z cc (4.6 shows the general threephase model of Node n + Ia Ib + Zaa Zbb Zcc Node m Vagn Vbg n  Zab Zbc Zca + + Vag m Vbg m Vcg m + Ic Vcg n  +   FIGURE 4.54 will not be equal to each other.6 Threephase line segment model.88 Distribution System Modeling and Analysis For a distribution line that is not transposed. Figure 4. However.56) The phase impedance matrix for a threewire delta line is determined by the application of Carson’s equations without the Kron reduction step. for example) have been made regarding the spacing between conductors. the effect of the mutual coupling between phases is accurately taken into account. Since no approximations (transposition. a Vphase line consisting of phases a and c. the modiﬁed Carson’s equations can be applied which will lead to initial 3 × 3 and 2 × 2 primitive impedance matrices. . The application of the modiﬁed Carson’s equations and the phase frame matrix leads to the most accurate model of a line segment. For example.55) The phase impedance matrix for a phase b singlephase line would be 0 0 0 [ z abc ] = 0 z bb 0 Ω/mile 0 0 0 (4.
1. The ﬁrst incorporates the application of the modiﬁed Carson’s equations and the Kron reduction to obtain the phase impedance matrix.7 Sequence Impedances (4.61) 2 (4.58) Many times the analysis of a feeder will use only the positive and zero sequence impedances for the line segments. In condensed form.59 becomes: [ VLG abc ] = [ A s ] ⋅ [ VLG 012 ] 1 where 1 1 (4. The deﬁnition of linetoground phase voltages as a function of the line2 toground sequence voltages is given by : V ag Vcg 1 1 2 s 1 2 s VLG 0 (4.62) . Equation 4.Series Impedance of Overhead and Underground Lines 89 a line segment.0/120. The voltage equation in matrix form for the line segment is V ag V bg V cg n V ag = V bg Vcg m z aa z ab z ac z ca z cb z cc Ia (4.59) VLG 2 V bg = 1 a a s ⋅ VLG 1 1 as a where as = 1. Bear in mind that for Vphase and singlephase lines some of the impedance values will be zero.57) Ic + z ba z bb z bc ⋅ I b where Zij = zij ⋅ length.57 can be written in condensed form as: [ VLG abc ] n = [ VLG abc ] m + [ Z abc ] ⋅ [ I abc ] 4.60) [ As ] = 1 a2 as s 1 as as The phase line currents are deﬁned in the same manner: [ I abc ] = [ A s ] ⋅ [ I 012 ] (4. Equation 4. There are two methods for obtaining these impedances.
66 represent the mutual coupling between sequences.58 can be transformed to the sequence domain by multiplying −1 both sides by [As] and substituting in the deﬁnition of the phase currents as given by Equation 4.62.66 is the deﬁning equation for converting phase impedances to sequence impedances. In the idealized state these offdiagonal terms would .66) VLG 0 = VLG 1 VLG 2 m I0 Z 00 Z 01 Z 02 + Z 10 Z 11 Z 12 ⋅ I 1 Z 20 Z 21 Z 22 I2 (4.64) Equation 4.65 in expanded form is given by: VLG 0 VLG 1 VLG 2 n –1 –1 –1 (4.66 the diagonal terms of the matrix are the sequence impedances of the line such that Z00 = zero sequence impedance. [ VLG 012 ] n = [ A s ] ⋅ [ VLG abc ] n [ VLG 012 ] n = [ A s ] ⋅ [ VLG abn ] m + [ A s ] ⋅ [ Z abc ] ⋅ [ A s ] ⋅ [ I 012 ] [ VLG 012 ] n = [ VLG 012 ] m + [ Z 012 ] ⋅ [ I 012 ] Z 00 Z 01 Z 02 –1 [ Z 012 ] = [ A s ] ⋅ [ Z abc ] ⋅ [ A s ] = Z 10 Z 11 Z 12 Z 20 Z 21 Z 22 Equation 4.⋅ 1 a s a 2 s 3 2 1 as as –1 (4. In Equation 4. Z11 = positive sequence impedance.65) where (4.67) Equation 4.90 Distribution System Modeling and Analysis Equation 4. and Z22 = negative sequence impedance.60 can be used to solve for the sequence linetoground voltages as a function of the phase linetoground voltages: [ VLG 012 ] = [ A s ] ⋅ [ VLG abc ] 1 1 1 1 = .63) where [ As ] –1 (4. The offdiagonal terms of Equation 4.
70) When Equation 4.66 is used with this phase impedance matrix the resulting sequence matrix is diagonal (offdiagonal terms are zero).72) A second method commonly used to determine the sequence impedances directly is to employ the concept of Geometric Mean Distances (GMD).⋅ ( z aa + z bb + z cc ) Ω/mile 3 1 z m = . For highvoltage transmission lines this will sometimes be the case. and the offdiagonal terms equal to the average of the offdiagonal terms of Equation 4.69) (4. If a line is assumed to be transposed. When the lines are transposed the mutual coupling between phases (offdiagonal terms) are equal and. the phase impedance matrix is modiﬁed so the three diagonal terms are equal and all of the offdiagonal terms are equal.54.Series Impedance of Overhead and Underground Lines 91 be zero.73) The GMD between phases and neutral is deﬁned as: D in = GMD in = 3 D an ⋅ D bn ⋅ D cn ft (4. as a result. The usual procedure is to set the three diagonal terms of the phase impedance matrix equal to the average of the diagonal terms of Equation 4. Since distribution lines are rarely if ever transposed. The sequence impedances can be determined directly as: z 00 = z s + 2 ⋅ z m Ω/mile z 11 = z 22 = z s – z m Ω/mile (4.68) (4. In order for this to happen it must be assumed that the line has been transposed.74) . consequently. the offdiagonal terms of the sequence impedance matrix become zero. When this is done the self and mutual impedances are deﬁned as: 1 z s = .71) (4. the offdiagonal terms of the sequence impedance matrix will not be zero. the mutual coupling between phases is not equal and.⋅ ( z ab + z bc + z ca ) Ω/mile 3 The phase impedance matrix is now deﬁned as: zs zm zm [ z abc ] = z m z s z m Ω/mile zm zm zs (4.54. The GMD between phases is deﬁned as: D ij = GMD ij = 3 D ab ⋅ D bc ⋅ D ca ft (4.
77) (4. positive.93402 Ω/mile GMR n 1 ˆ z ij = 0.93402 Ω/mile D in (4. Ω/mile GMR i 2 (4. and the neutral conductor is 4/0 6/1 ACSR.78) Equations 4.75 through 4.0953 + j0.53) and the sequence impedance transformation (Equation 4.0953 + j0. Application of the Kron reduction (Equation 4.80) Equation 4. where ncond is the number of conductors (phases plus neutrals) in the line segment. SOLUTION From the table of standard conductor data (Appendix A) it is found that 336.0244 ft Resistance = 0.400 26/7 ACSR (Linnet).93402 Ω/mile GMR i 1 ˆ z nn = r n + 0.66) leads to the following expressions for the zero.75) (4.41 and 4.1 An overhead threephase distribution line is constructed as shown in Figure 4.12134 ⋅ ln . Ω/mile z nn ˆ ˆ ˆ z 11 = z 22 = z ii – z ij z 11 = z 22 D ij = r i + j0.42 to determine the various self and mutual impedances of the line.12134 ⋅ ln . + 7.0953 + j0. and negative sequence impedances: ˆ z in ˆ ˆ z 00 = z ii + 2 ⋅ z ij – 3 ⋅ .0953 + j0. resulting in: 1 ˆ z ii = r i + 0.400 26/7 ACSR: GMR = 0.93402 Ω/mile D ij 1 ˆ z in = 0.306 Ω/mile .92 Distribution System Modeling and Analysis The GMDs as deﬁned above are used in Equations 4.12134 ⋅ ln .12134 ⋅ ln . Example 4.76) (4.80 is recognized as the standard equation for the calculation of the line impedances when a balanced threephase system and transposition are assumed. Determine the phase impedance matrix and the positive and zero sequence impedances of the line.12134 ⋅ ln  + 7.79) (4.7. + 7. + 7. The phase conductors are 336.78 will deﬁne a matrix of order ncond × ncond.
7.8515 j0. Dca = 7.5' c 93 n 25.5' a b 3.4013 + j1.5 ft.0953 + 0.0953 + j0.5 = 0.7 Threephase distribution line spacings. Dan = 5.8515 j1. 4.4013 + 0.Series Impedance of Overhead and Underground Lines 2.0953 + j0.272 ft.4013 + j1.5 ft. Dbc = 4.0953 + 0. the self impedance for phase a is 1 ˆ z aa = 0.7266 j0.7674 j1.12134 ⋅ ln .7524 j0.7865 0.93402 0. Resistance = 0.0953 + j0. Dcn = 5.0953 + 0.0 ft.4013 + ˆ [ z ] = 0.7674 0. 4/0 6/1 ACSR: GMR = 0.0953 + j0.0' 4.7524 0.0953 + 0.0953 + j0.4133 0.0 ft.7802 j0.0953 + 0.6569 ft.41).0' FIGURE 4.7865 Ω/mile j0.0953 + 0.0' 4.4133 Ω/mile Applying Equation 4.4133 j0.306 + j0.5920 Ω/mile From Figure 4.7266 0.0953 + j1.4133 j0.0953 + j0.0244 = 0.0953 + 0.7802 0. Applying the modiﬁed Carson’s equation for self impedance (Eq.+ 7.6873 + j0.93402 2.8515 Ω/mile Applying the equations for the other self and mutual impedance terms results in the primitive impedance matrix: 0.00814 ft. Dbn = 4.42 for the mutual impedance between phases a and b: 1 ˆ z ab = 0.5465 .0953 + 0. the following distances between conductors can be determined: Dab = 2.+ 7.12134 ⋅ ln .
1 term is the zero sequence impedance.7802 Ω/mile 0.3061 + j0.9373 0.66: [ z 012 ] = [ A s ] ⋅ [ z abc ] ⋅ [ A s ] 0.4013 + j1.6270 – 0.4013 + j1.1560 + j0.7802 0.0953 + j0.4615 + j1.0953 + j0.0953 + j0. Note that the offdiagonal terms are not zero.1535 + j0.0060 Ω/mile 0.5465 ] Ω/mile ˆ [ z nj ] = 0.4236 Ω/mile 0.8515 0.4236 0.7674 ˆ [ z nn ] = [ 0.0953 + j0.7266 0.3849 0.2 term is the positive sequence impedance.0953 + j0.1580 + j0.4013 + j1.1580 + j0.7865 0.0953 + j0.1535 + j0.4133 0.8515 ˆ [ z ij ] = 0.4666 + j1.0115 0. This implies that there is mutual coupling between sequences.4576 + j1.2 and 3.3 terms are equal.7735 + j1. the positive and negative sequence impedances are equal.53 results in the phase impedance matrix: ˆ ˆ ˆ ˆ [ z abc ] = [ z ij ] – [ z in ] ⋅ [ z nn ] ⋅ [ z nj ] 0. The 2.5017 0. the 2.1560 + j.0943 + j0.5017 0.0953 + j0.0780 0.0159 [ z 012 ] = −0.7674 Ω/mile The Kron reduction of Equation 4.4133 .0059 0. With the offdiagonal terms –1 –1 0.0256 + j0.0953 + j0.0321 + j0.0256 + j0.4133 0. This is a result of the nonsymmetrical spacing between phases.0115 – 0.94 Distribution System Modeling and Analysis The primitive impedance matrix in partitioned form is 0.7524 0.0482 0.3 term is the negative sequence impedance.7524 ˆ [ z in ] = 0.0953 + j0.0723 – j0.0953 + j0.6270 In the sequence impedance matrix the 1.0159 0.0651 The phase impedance matrix can be transformed into the sequence impedance matrix with the application of Equation 4.7266 j0.3849 [ z abc ] = 0. which demonstrates that for line segments.6873 + j1.7865 Ω/mile 0.0953 + j0. and the 3.0723 – j 0.3061 + j0.0321 + j0.
1558 + j0.4619 + j1. The circuit of Figure 4. It should also be noted that the modiﬁed zero.3061 + j0.4368).4368 0. and replacing each offdiagonal term with the average of the offdiagonal terms (0. However. .8 will result in a 7 × 7 primitive impedance matrix. and negative sequence impedances are exactly equal to the exact sequence impedances that were ﬁrst computed.1558 + j0.0638 Using this modiﬁed phase impedance matrix in the symmetrical component transformation equation results in the modiﬁed sequence impedance matrix 0.2 Series Impedance of Underground Lines Figure 4.3061 + j0.4619 + j1.0638).4619 + j1.4619 + j1.9373 0 0 Ω/mile 0 0. For underground circuits that do not have the additional neutral conductor.6270 [ z1 012 ] = Note that now the offdiagonal terms are all equal to zero. 4. The transposition can be simulated in Example 4.4368 Ω/mile 0.4368 0. The modiﬁed Carson’s equations can be applied to underground cables in much the same manner as for overhead lines.6270 0 0 0 0.1558 + j0.1558 + j0.7735 + j1. the three sequence networks representing the line will not be independent. In highvoltage transmission lines.4368 [ z1 abc ] = 0. positive.1 by replacing the diagonal terms of the phase impedance matrix with the average value of the diagonal terms (0.1558 + j0. the primitive impedance matrix will be 6 × 6. The original phase impedance matrix must be used if the correct effect of the mutual coupling between phases is to be modeled.1558 + j0. it is noted that the offdiagonal terms are small relative to the diagonal terms. This modiﬁed phase impedance matrix becomes: 0. meaning there is no mutual coupling between sequence networks.0638 0.4368 0.1558 + j0.4368 0.0638 0.8 shows the general conﬁguration of three underground cables (concentric neutral or tape shielded) with an additional neutral conductor. The results of this example should not be interpreted to mean that a threephase distribution line can be assumed to have been transposed. it is usually assumed that the lines are transposed and that the phase currents represent a balanced threephase set.Series Impedance of Overhead and Underground Lines 95 nonzero.
9 Concentric neutral cable. the resistance and GMR of the phase conductor and the equivalent neutral must be known. In order to apply Carson’s equations to this cable. 4.2. the following data needs to be extracted from a table of underground cables (Appendix B): dc = phase conductor diameter (inches) dod = nominal diameter over the concentric neutrals of the cable (inches) ds = diameter of a concentric neutral strand (inches) GMRc = geometric mean radius of the phase conductor (ft. The solid strands of concentric neutral are spiraled around the semiconducting screen with a uniform spacing between strands.1 Concentric Neutral Cable Figure 4.9 shows a simple detail of a concentric neutral cable. The insulation is then covered by a semiconducting insulation screen.) . Two popular types of underground cables are the concentric neutral cable and the tapeshielded cable. To apply the modiﬁed Carson’s equations.8 Threephase underground with additional neutral. The cable consists of a central phase conductor covered by a thin layer of nonmetallic semiconducting screen. to which is bonded the insulating material. Phase Conductor Insulation R d od dc Jacket Concentric Neutral Strand Insulation Screen ds FIGURE 4.96 Distribution System Modeling and Analysis D14 D13 D12 a b D23 c D34 n FIGURE 4. Some cables will also have an insulating jacket encircling the neutral strands.
83) (4.Series Impedance of Overhead and Underground Lines GMRs = geometric mean radius of a neutral strand (ft. The equivalent geometric mean radius of the concentric neutral is computed using the equation for the geometric mean radius of bundled conductors used in high1 voltage transmission lines.10 shows the relationship between the distance between centers of concentric neutral cables and the radius of a circle passing through the centers of the neutral strands.84) where Dnm = centertocenter distance between phase conductors. GMR cn = k GMR s ⋅ k ⋅ R k−1 ft (4.81) where R = radius of a circle passing through the center of the concentric neutral strands d od – d R = s ft 24 The equivalent resistance of the concentric neutral is rs r cn = . The geometric mean distance between a concentric neutral and an adjacent phase conductor is given by: D ij = k D nm – R ft k k (4.82) The various spacings between a concentric neutral and the phase conductors and other concentric neutrals are as follows: Concentric neutral to its own phase conductor Dij = R (Equation 4.82 above) Concentric neutral to an adjacent concentric neutral Dij = centertocenter distance of the phase conductors Concentric neutral to an adjacent phase conductor Figure 4.Ω/mile k (4. For cables buried in a trench the distance between cables will be much greater than the .) rc = resistance of the phase conductor (Ω/mile) rs = resistance of a solid neutral strand (Ω/mile) k = number of concentric neutral strands 97 The geometric mean radii of the phase conductor and a neutral strand are obtained from a standard table of conductor data (Appendix A).
10 Distances between concentric neutral cables. 250. 6" radius R.2 Three concentric neutral cables are buried in a trench with spacings as shown in Figure 4.11.000 CM stranded allaluminum with 13 strands of #14 annealed. The outside diameter of the cable over the neutral strands is 1.11 Threephase concentric neutral cable spacing. In applying the modiﬁed Carson’s equations. the numbering of conductors and neutrals is important. a threephase underground circuit with an additional neutral conductor must be numbered as: 1 2 3 4 5 6 7 = = = = = = = phase conductor #1 phase conductor #2 phase conductor #3 neutral of conductor #1 neutral of conductor #2 neutral of conductor #3 additional neutral conductor (if present) Example 4.98 Distribution System Modeling and Analysis R R Dnm FIGURE 4. coated copper wires (1/3 neutral). For cables in conduit that assumption is not valid. 6" FIGURE 4.84 is equal to Dnm. . The cables are 15 kV.29 inches (Appendix B). Determine the phase impedance matrix and the sequence impedance matrix. and therefore it may be assumed that Dij in Equation 4. For example.
567 inches Resistance = 0.82) is d od – d R = s = 0. and 6.5 ft D 31 = D 13 = D 64 = D 46 = 1.0511 13−1 = 0.0641 inches The radius of the circle passing through the center of the strands (Equation 4. Resistance = 14.00208 ⋅ 13 ⋅ 0. 2.0171 ft. The conductortoconductor and concentric neutraltoconcentric neutral spacings are D 12 = D 21 = D 45 = D 54 = 0.5 ft D 23 = D 32 = D 56 = D 65 = 0.1438 Ω/mile k 13 The phase conductors are numbered 1. the distances between concentric neutrals and adjacent phase conductors are . and 3.4100 Ω/mile #14 copper neutral strands: GMRs = 0.= .8722 Ω/mile Diameter (ds) = 0. The concentric neutrals are numbered 4. 5. Diameter = 0.0486 ft The equivalent resistance of the concentric neutral is rs 14.Series Impedance of Overhead and Underground Lines SOLUTION 99 The data for the phase conductor and neutral strands from a conductor data table (Appendix A) are 250.8722 r cn = .0 ft The spacings between conductors and their concentric neutrals are D 14 = D 25 = D 36 = R = 0.0511 ft 24 The equivalent GMR of the concentric neutral is computed by: GMR cn = k GMR s ⋅ k ⋅ R k−1 = 13 0.000 AA phase conductor: GMR = 0.0511 ft Since the radius R is much smaller than the spacings between cables.= 1.00208 ft.
3236 .9627 .3236 ˆ [ z in ] = .0953 + j.0953 + j1.0953 + 1.144 + j0.3236 .0953 + j.2393 + j1.+ 7.5053 + j1.5 Continuing the application of the modiﬁed Carson’s equations results in a 6 × 6 primitive impedance matrix.4564 Ω/mile The self impedance for the concentric neutral for Cable #1 is 1 ˆ z 44 = 0.+ 7.0953 + j1.0953 + 0.0953 + j1.9627 .12134 ⋅ ln .0953 + j1.0953 + j1.+ 7.0468 .0468 .93402 0.0468 .0953 + j1.5 The mutual impedance between Cable #1 and its concentric neutral is 1 ˆ z 14 = 0.41 + j0.+ 7.5053 + j1.93402 0.0468 Ω/mile .0.100 Distribution System Modeling and Analysis just the centertocenter distances between conductors: D 15 = D 51 = 0.12134 ⋅ ln .9627 .5 D 26 = D 62 = 0.0468 Ω/mile 0.0953 + j1.12134 ⋅ ln .0953 + j1.0468 Ω/mile 0. This matrix in partitioned (Equation 4.0953 + j.12134 ⋅ ln .0953 + j1.0486 = 1.93402 = 0.0468 .+ 7.0511 The mutual impedance between the concentric neutral of Cable #1 and the concentric neutral of Cable #2 is 1 ˆ z 45 = 0.4564 .0953 + j0.0953 + j1.12134 ⋅ ln .0171 = 0.9627 .0953 + j1.0953 + j0.0468 .0953 + j0.0953 + j.0468 .5053 + j1.93402 = 0.4564 .0953 + j1.0953 + j1. The self impedance for the cable in position 1 is 1 ˆ z 11 = 0.33) form is .93402 = 0.5 ft D 61 = D 16 = 1.4564 ˆ [ z ij ] = .0953 + j1.3236 Ω/mile 0.3296 Ω/mile The mutual impedance between Cable #1 and Cable #2 is 1 ˆ z 12 = 0.5053 + j1.0468 Ω/mile .
4041 0.0328 0.7891 + j0.0451 Ω/mile – 0.0081 – 0.Series Impedance of Overhead and Underground Lines ˆ ˆ [ z nj ] = [ z in ] ˆ [ z nn ] = T 101 1.2391 + j1.3191 + j0.7981 + j0.4151 4.4463 0.2849 – j0.0028 – j0. .0523 + j0.2849 – j0.66: [ z 012 ] = [ A s ] ⋅ [ z abc ] ⋅ [ A s ] 1.0065 0.0003 0.12 Tapeshielded cable.3296 .3296 .4105 + j0.0028 – j0. The shield is bare copper AL or CU Phase Conductor Insulation d od ds dc CU Tape Shield T Jacket FIGURE 4.4874 + j0.4151 – 0. The cable consists of a central phase conductor covered by a thin layer of nonmetallic semiconducting screen to which is bonded the insulating material.0953 + j1.3191 + j0.0468 .7981 + j0.4874 + j0.0143 0.0328 0.4463 –1 [ z abc ] = The sequence impedance matrix for the concentric neutral threephase line is determined using Equation 4.0953 + j1.2 TapeShielded Cables –1 Figure 4. The insulation is covered by a semiconducting insulation screen.0081 0.3296 Using the Kron reduction results in the phase impedance matrix: ˆ ˆ ˆ ˆ [ z abc ] = [ z ij ] – [ z in ] ⋅ [ z nn ] ⋅ [ z nj ] 0.4665 – 0.0056 + j0.0953 + j.0468 1.9627 .0953 + j1.2.0328 0.0143 0.2391 + j1.9627 .0056 + j0.0468 Ω/mile 1.3191 + j0.2391 + j1.0953 + j1.0953 + j.0468 .12 shows a simple detail of a tapeshielded cable.0264 + j0.0065 [ z 012 ] = – 0.3191 + j0.0328 Ω/mile 0.
The singlephase line is connected to phase b.ft.89) Example 4.87) Tape shield to an adjacent tape shield Dij = centertocenter distance of the phase conductors (ft. Assume p = 2.) (4. the modiﬁed Carson’s equations will be applied to calculate the self impedances of the phase conductor and the tape shield. The resistance and GMR of the phase conductor are found in a standard table of conductor data. 24 GMR shield (4. .88) Tape shield to an adjacent phase or neutral conductor Dij = centertocenter distance between phase conductors (ft. The GMR of the tape shield is the radius of a circle passing through the middle of the shield and is given by: T d s – 1000 = .85 must be expressed in Ωmeters at 50°C. as well as the mutual impedance between the phase conductor and the tape shield.86) The various spacings between a tape shield and the conductors and other tape shields are as follows: Tape shield to its own phase conductor Dij = GMRshield = radius to midpoint of the shield (ft. as shown in Figure 4. An insulating jacket encircles the tape shield.13.102 Distribution System Modeling and Analysis tape helically applied around the insulation screen.9385×10 .85) The resistivity (ρ) in Equation 4.) (4. Parameters of the taped shielded cable are dc = diameter of phase conductor (inches) ds = outside diameter of the tape shield (inches) dod = outside diameter over jacket (inches) T = thickness of copper tape shield in mils Once again.) (4. The outside diameter of the tape shield (ds) is given in inches and the thickness of the tape shield (T) is in mils. The resistance of the tape shield is given by: ρ 8 r shield = 7.Ω/mile ds · T (4. Determine the phase imped−8 ance matrix.3 A singlephase circuit consists of a 1/0 AA. 220mil insulation tape shielded cable and a 1/0 Cu neutral conductor.3715 × 10 Ωmeter.
Distance between cable and neutral = Dnm = 3 inches The resistance of the tape shield is computed according to Equation 4.88 inches Resistance = 0.= 0.0365 ft 24 24 –8 GMR shield The conductors are numbered such that: #1 = 1/0 AA conductor #2 = tape shield #3 = 1/0 copper ground . Cable Data: 1/0 AA Outside diameter of the tape shield = ds = 0.= .Series Impedance of Overhead and Underground Lines 103 3" FIGURE 4.97 Ω/mile GMRp = 0.9385×10 · .2785 Ω/mile ds · T 0. Tape shield thickness = T = 5 mils Neutral Data: 1/0 Copper.86: T 5 d s – 0.01113 ft. 7 strand Resistance = 0.= 7.9385×10 · .88 · 5 The GMR of the tape shield is computed according to Equation 4.3715×10 ρ 8 8 r shield = 7.85: 2.607 Ω/mile GMRn = 0.13 Singlephase tape shield with neutral.= 4.0111 ft.88 – 1000 1000 = .
1309 ] .5088 ] ˆ [ z in ] = [ 0.0653 + j1.1309 0.0953 + j1.1309 0.0111 = 1.0953 + j1.0953 + j1.= 0.97 + j0.0365 = 0.3739 + j1.+ 7.5088 Ω/mile The mutual impedance between Conductor #1 and the tape shield (Conductor #2) is 1 ˆ z 12 = 0.+ 7.0953 + j1.7023 + j1.0953 + j1.3645 0.1309 Ω/mile 0.0953 + j1.93402 0.12134 ⋅ ln . the primitive impedance matrix is ˆ [ z ij ] = [ 1.0953 + j1.1309 0.5085 In partitioned form.0953 + j0.3645 0.3645 0.1309 0.0953 + j1.2786 + j0.1309 ˆ [ z nn ] = 4.25 12 The self impedance of Conductor #1 is 1 ˆ z 11 = 0.12134 ⋅ ln .3739 + j1.0953 + j1.3645 Ω/mile Continuing on.0953 + 0.0953 + j1.5085 0.3739 + j1.0953 + j1.7023 + j1.0365 3 D 13 = .0953 + 4.104 Distribution System Modeling and Analysis The spacings used in the modiﬁed Carson’s equations are D 12 = GMR shield = 0.93402 0.0653 + j1.3645 4.3645 Ω/mile The self impedance of the tape shield (Conductor #2) is 1 ˆ z 22 = 0.0953 + j1.+ 7.0953 + j1.5088 ˆ [ z ] = 0.12134 ⋅ ln .93402 0.3645 0.0653 + j1.1309 0.3645 ˆ [ z nj ] = 0. the ﬁnal primitive impedance matrix is 1.0365 = 4.
John R. the six basic reference frames. Wave propagation in overhead wires with ground return.3219 + j0. 2nd ed. Since the singlephase line is on phase b. Problems 4.6743 Ω/mile.. Boston. Since voltage drop is a primary concern on a distribution line.Series Impedance of Overhead and Underground Lines 105 Applying Kron’s reduction method will result in a single impedance which represents the equivalent singlephase impedance of the tape shield cable and the neutral conductor: ˆ ˆ ˆ z 1p = [ z ij ] – [ z in ] ⋅ [ z nn ] ⋅ [ z nj ] z 1p = 1. are determined. When using the modiﬁed Carson’s equations there is no need to make any assumptions. Vol. 1952.14. then the phase impedance matrix for the line is 0 0 0 [ z abc ] = 0 1. Bell System Technical Journal. AIEE Trans.3 Summary This chapter has been devoted to presenting methods for computing the phase impedances and sequence impedances of overhead lines and underground cables. M. Carson. such as transposition of the lines. 1994. G.6743 0 Ω/mile 0 0 0 –1 4. 3.. 1926.. Tensorial analysis of integrated transmission systems.. and Sarma. . PWSKent Publishing. D.. the most accurate values of the phase impedances. J. Part I.3219 + j0.1 Determine the phase impedance matrix [Zabc] and the sequence impedance matrix [Z012] in Ω/mile for the threephase conﬁguration shown in Figure 4. New York. 2. the impedances used for the line must be as accurate as possible. References 1. Vol. 71. self and mutual. By assuming an untransposed line and including the actual phasing of the line. Kron. Carson’s equations have been modiﬁed in order to simplify the computation of the phase impedances. Power System Analysis and Design. 5. Glover.
5' a c Distribution System Modeling and Analysis 4.0' FIGURE 4.15 Twophase conﬁguration for Problem 4.2 Determine the phase impedance [Zabc] matrix in Ω/mile for the twophase conﬁguration in Figure 4.14 Threephase conﬁguration for Problem 4.16.500 26/7 ACSR Neutral Conductor: 4/0 ACSR 4.400 26/7 ACSR Neutral Conductor: 4/0 6/1 ACSR Determine the phase impedance [Zabc] matrix in Ω/mile for the singlephase conﬁguration shown in Figure 4.15.2.3 Phase and Neutral Conductors: 1/0 6/1 ACSR .5' b 3. 7.0' FIGURE 4. Phase Conductors: 556.0' n 25.1. Phase Conductors: 336. 4.106 2.0' 4.0' 4.0' a 3.0' c n 25.
6 .3.2.5' n 25. and 4. 4. The phase and neutral conductors are 250.16 Singlephase pole conﬁguration for Problem 4.4 Create the spacings and conﬁgurations of Problems 4. Compare the phase impedance matrices to those computed in the previous problems.5 4' a 2 ' b c n 2' 2' 25’ FIGURE 4.Series Impedance of Overhead and Underground Lines b 107 5. 4. Compute the positive. negative.17.5.0' FIGURE 4. 4. 4.17.1.17 Threephase pole conﬁguration for Problem 4. using the Geometric Mean Distance (GMD) method for the pole conﬁguration shown in Figure 4.0' 0. Determine the phase impedance matrix [ Z abc ] and sequence impedance matrix [Z012] in Ω/mile for the threephase pole conﬁguration in Figure 4.000 allaluminum.3 in the Radial Distribution Analysis Package (RDAP). and zero sequence impedances in Ω/1000 ft.
Create the spacings and conﬁgurations of Problems 4. assuming phasing of cab.11).9 4. and zero sequence impedances in Ω/1000 ft.. Three onethird neutral.108 4. 4. Assume the cable jacket has a thickness of 0.8 2' 2' Compute the positive. Compute the phase impedance matrix in Ω/mile for this cabled line.000 allaluminum. 4.13 Create the spacing and conﬁguration of Problem 4.9 and 4. A 4/0 copper conductor is used as the neutral.000 CM aluminum concentric cables with onethird neutrals are buried in a trench in a horizontal conﬁguration (see Figure 4.14 . 4.10 Three 250.2inch and the cables lie in a triangular conﬁguration inside the conduit.68 and 4. assuming phase c. assuming the cable is connected to phase b. 4.000 CM aluminum tapeshielded cable. The cable has a full neutral (see Appendix B). 2/0 aluminumjacketed.7.12 A singlephase underground line is composed of a 350. negative.11 4. The phase conductors are 350. 2' n 2' b a c 25' FIGURE 4. 4.7 Distribution System Modeling and Analysis Determine the [ Z abc ] and [Z012] matrices in Ω/mile for the threephase conﬁguration shown in Figure 4. Determine the [Zabc] and [Z012] matrices in Ω/1000 ft. Determine the impedance of the cable and the resulting phase impedance matrix in Ω/mile. for the line of Figure 4.10 in RDAP. A 4/0 aluminum concentric neutral cable is to be used for a singlephase lateral. Compare the values of the phase impedance matrices to those computed in the previous problems.18 using the average self and mutual impedances deﬁned in Equations 4. concentric neutral cables are installed in a 6inch conduit. Determine the phase impedance matrix in Ω/mile for this singlephase cable line.18. Compare results.18 Threephase pole conﬁguration for Problem 4. The cable and neutral are separated by 4 inches.69.000 allaluminum. and the neutral conductor is 250.13 in RDAP.
5
Shunt Admittance of Overhead and Underground Lines
The shunt admittance of a line consists of the conductance and the capacitive susceptance. The conductance is usually ignored because it is very small compared to the capacitive susceptance. The capacitance of a line is the result of the potential difference between conductors. A charged conductor creates an electric ﬁeld that emanates outward from the center. Lines of equipotential are created that are concentric to the charged conductor, as illustrated in Figure 5.1. In Figure 5.1 a difference of potential between two points (P1 and P2) is a result of the electric ﬁeld of the charged conductor. When the potential difference between the two points is known, then the capacitance between the two points can be computed. If there are other charged conductors nearby, the potential difference between the two points will be a function of
P1
+ +
RD
D2
P2
D1
+
+
+
FIGURE 5.1 Electric ﬁeld of a charged round conductor.
109
110
Distribution System Modeling and Analysis
the distance to the other conductors and the charge on each conductor. The principle of superposition is used to compute the total voltage drop between two points, and then the resulting capacitance between the points. The points can be points in space, the surface of two conductors, or the surface of a conductor and ground.
5.1
General VoltageDrop Equation
Figure 5.2 shows an array of N positively charged solid, round conductors. Each conductor has a unique uniform charge density of q C meter. The voltage drop between Conductor i and Conductor j as a result of all of the charged conductors is given by: D1 j D ij RD j D Nj 1 V ij =  q 1 ln  + … + q i ln  + … + q j ln  + … + q N ln  D 1i RD i D ij 2 πε D Ni Equation 5.1 can be written in general form as: 1 V ij = 2 πε where (5.1)
n=1
∑ qn ln D ni
N
D nj
(5.2)
ε ε0 εr qn Dni Dnj RDn
= = = = = = =
ε0εr = permittivity of the medium −12 permittivity of free space = 8.85 × 10 µF/meter relative permittivity of the medium charge density on Conductor n cb/meter distance between Conductor n and Conductor i (ft.) distance between Conductor n and Conductor j (ft.) radius of Conductor n
1 N
+
+
n
+ V + i + ij
FIGURE 5.2 Array of round conductors.
j
+ 
Shunt Admittance of Overhead and Underground Lines
111
5.2
Overhead Lines
The method of conductors and their images is employed in the calculation of the shunt capacitance of overhead lines. This is the same concept that was used in Chapter 4 in the general application of Carson’s equations. Figure 5.3 illustrates the conductors and their images, and will be used to develop a general voltagedrop equation for overhead lines. In Figure 5.3 it is assumed that: q′ = −qi i and q′ = − q j j (5.3)
Applying Equation 5.2 to Figure 5.3: S ij D ij RD S ii 1 V ii =  q i ln  + q i′ ln i + q j ln  + q ′ ln  j RD i S ii D ij 2 πε S ij (5.4)
Because of the assumptions of Equation 5.3, Equation 5.4 can be simpliﬁed: D ij S ij RD S ii 1 V ii =  q i ln  – q i ln i + q j ln  – q j ln  S ii 2 πε S ij RD i D ij S ij S ij S ii S ii 1 V ii =  q i ln  + q i ln  + q j ln  + q j ln  2 πε RD i RD i D ij D ij S ij S ii 1 V ii =  2 ⋅ q i ln  + 2 ⋅ q j ln  2 πε RD i D ij (5.5)
FIGURE 5.3 Conductors and images.
112 where
Distribution System Modeling and Analysis
Sii = distance from Conductor i to its image i’ (ft.) Sij = distance from Conductor i to the image of Conductor j (ft.) Dij = distance from Conductor i to Conductor j (ft.) RD i = radius of Conductor i in ft. Equation 5.5 gives the total voltage drop between Conductor i and its image. The voltage drop between Conductor i and ground will be onehalf of that given in Equation 5.5: S ii S ij 1 V ig =  q i ln  + q j ln  RD i 2 πε D ij Equation 5.6 can be written in general form as: ˆ ˆ V ig = P ii ⋅ q i + P ij ⋅ q j (5.7)
(5.6)
ˆ ˆ where P ii and P ij are the self and mutual potential coefﬁcients. For overhead lines the relative permittivity of air is assumed to be 1.0 so that:
εair = 1.0 × 8.85 × 10
−2
−12
F/meter
(5.8)
εair = 1.4240 × 10 µF/mile
Using the value of permittivity in µF/mile, the self and mutual potential coefﬁcients are deﬁned as: S ii ˆ P ii = 11.17689 ⋅ ln  mile/µF RD i S ij ˆ P ij = 11.17689 ⋅ ln  mile/µF D ij (5.9)
(5.10)
NOTE In applying Equations 5.9 and 5.10, the values of RD i , Sii, Sij, and Dij must all be in the same units. For overhead lines the distances between conductors are typically speciﬁed in feet, while the value of the conductor diameter from a table will typically be in inches. Care must be taken to assure that the radius in feet is used in applying the two equations. For an overhead line of ncond conductors, the primitive potential coefﬁˆ cient matrix [P primitive] can be constructed. The primitive potential coefﬁcient
Shunt Admittance of Overhead and Underground Lines
113
matrix will be an ncond x ncond matrix. For a fourwire grounded wye line, the primitive coefﬁcient matrix will be of the form: ˆ P aa ˆ P ba ˆ P ab ˆ P bb ˆ ˆ P ac • P an ˆ ˆ P bc • P bn ˆ ˆ P cc • P cn
ˆ [ P primitive ] = P ˆ ca P cb ˆ • • • • • ˆ na P nb P nc • P nn ˆ ˆ ˆ P
(5.11)
The dots (•) in Equation 5.11 are partitioning the matrix between the third and fourth rows and columns. In partitioned form, Equation 5.11 becomes: ˆ ˆ ˆ [ P primitive ] = [ P ij ] [ P in ] ˆ ˆ [ P nj ] [ P nn ]
(5.12)
Because the neutral conductor is grounded, the matrix can be reduced using the Kron reduction method to an nphase x nphase phase potential coefﬁcient matrix [ P abc ]: ˆ ˆ ˆ ˆ [ P abc ] = [ P ij ] – [ P in ] ⋅ [ P nn ] ⋅ [ P nj ]
–1
(5.13)
The inverse of the potential coefﬁcient matrix will give the nphase x nphase capacitance matrix [ C abc ]: [ C abc ] = [ P abc ]
–1
(5.14)
For a twophase line, the capacitance matrix of Equation 5.14 will be 2 × 2. A row and column of zeros must be inserted for the missing phase. For a singlephase line, Equation 5.14 will result in a single element. Again, rows and columns of zero must be inserted for the missing phase. In the case of the singlephase line, the only nonzero term will be that of the phase in use. Neglecting the shunt conductance, the phase shunt admittance matrix is given by: [ y abc ] = 0 + j ⋅ ω ⋅ [ C abc ] µS/mile where (5.15)
ω = 2 ⋅ π ⋅ f = 2 ⋅ π ⋅ 60 = 376.9911
114
Distribution System Modeling and Analysis
Example 5.1
Determine the shunt admittance matrix for the overhead line in Example 4.1. Assume that the neutral conductor is 25 ft. above ground. The diameters of the phase and neutral conductors from the conductor table (Appendix A) are Conductor: 336,400 26/7 ACSR: dc = 0.721 inches, RDc = 0.03004 ft., ds = 0.563 inches, RDs = 0.02346 ft.
4/0 6/1 ACSR:
For the conﬁguration, the distances between conductors and images in matrix form are: 58 58.0539 58.4209 54.1479 58.0539 58 58.1743 54.0208 ft [S] = 58.4209 58.1743 58 54.0833 54.1479 54.0208 54.0833 50 The selfprimitive potential coefﬁcient for phase a and the mutual primitive potential coefﬁcient between phases a and b are 58 ˆ P aa = 11.17689 ln  = 84.5600 mile/µF 0.03004 58.0539 ˆ P ab = 11.17689 ln  = 35.1522 mile/µF 2.5 Using Equations 5.9 and 5.10, the total primitive potential coefﬁcient matrix is computed to be 84.5600 35.1522 23.7147 25.2469 ˆ [ P primitive ] = 35.4522 84.5600 28.6058 28.359 mile/µF 23.7147 28.6058 84.5600 26.6131 25.2469 28.359 26.6131 85.6659 Since the fourth conductor (neutral) is grounded, the Kron reduction method is used to compute the phase potential coefﬁcient matrix. Because only one ˆ row and column need to be eliminated, the [ P 44 ] term is a single element, so the Kron reduction equation for this case can be modiﬁed to: ˆ ˆ P i4 ⋅ P 4 ˆ P ij = P ij – j ˆ P 44 where i = 1, 2, 3 and j = 1, 2, 3.
7034 – j1.0031 0.6659 P Following the Kron reduction.0019 0.359 P 3. Figure 5.2923 115 Invert [ P abc ] to determine the shunt capacitance matrix: 0.9911 ⋅ [ C abc ] = – j1.6712 – j1.= 28.8714 26.0019 – 0.7944 75.015 – 0.8362 – j0.169 j5. the phase potential coefﬁcient matrix is 77.6131 ⋅ 28.2 – .8714 19.6058 – .8362 j5.9774 – j1.0159 – 0.1194 [ P abc ] = 26.Shunt Admittance of Overhead and Underground Lines For example.4 the following deﬁnitions apply: Rb = radius of a circle passing through the centers of the neutral strands dc = diameter of the phase conductor ds = diameter of a neutral strand k = total number of neutral strands .4 85.4 illustrates a basic concentric neutral cable with center conductor (black) being the phase conductor and the concentric neutral strands (gray) displaced equally around a circle of radius Rb.169 µS/mile – j0.3 Concentric Neutral Cable Underground Lines Most underground distribution lines consist of one or more concentric neutral cables.0049 – 0.= 19.7957 ˆ 4.7957 mile/µF 76.0031 µf/mile – 0. the value of Pcb is computed to be ˆ ˆ 26.7034 [ y abc ] = j ⋅ 376.3911 5.2 ˆ P cb = P 3.4 ⋅ P 4.0143 [ C abc ] = [ P ] –1 Multiply [ C abc ] by the radian frequency to determine the ﬁnal threephase shunt admittance matrix: j5.7957 15.0019 = – 0.172 19.7944 15. Referring to Figure 5.
4 Basic concentric neutral cable. Since all of the neutral strands are at the same potential. it is only necessary to determine the potential difference between phase conductor p and Strand 1.ln .+ … + ln . ln  .– . RD c Rb Rb Rb Rb 2 πε (5.17) .k 2 πε RD c k Rb (5. Rb RD D 12 D 1i D k1 1 V p1 = .116 Distribution System Modeling and Analysis 1 k D 12 Rb dc 012 Rb ds 2 j 3 i FIGURE 5.2 will be applied. Because of the stranding.18) (5. the general voltage drop of Equation 5. In order to compute the capacitance between the phase conductor and ground.16) where d RD c = c 2 d RD s = s 2 It is assumed that each of the neutral strands carries the same charge such that: qp q 1 = q 2 = q i = q k = – k Equation 5. 2 πε Rb Rb Rb RD c k R b V p1 RD s ⋅ D 12 ⋅ D 1i ⋅ … ⋅ D 1k qp Rb 1 = .+ … + q i ln .– . q p ln . ln s + ln .+ … + ln .+ q 1 ln s + q 2 ln . 4 5 The concentric neutral strands are grounded so that they are all at the same potential.16 can be simpliﬁed: RD D 12 D 1i D 1k Rb qp 1 V p1 = .+ … + q k ln .q p ln . it is assumed that the electric ﬁeld created by the charge on the phase conductor will be conﬁned to the boundary of the concentric neutral strands.
– 1 ln  . = 2 ⋅ R b ⋅ sin .Shunt Admittance of Overhead and Underground Lines 117 The numerator of the second ln term in Equation 5.20) Equation 5.k 2 πε RD c k R b (5.18 becomes: qp k ⋅ RD s ⋅ R k−1 Rb b V p1 = .ln . Referring to Figure 5.18 needs to be expanded. k (5. Equation 5.– 1 ln  2 πε RD. the following relations apply: 2⋅π θ 12 = k 4⋅π θ 13 = 2 ⋅ θ 12 = k In general. k Rb c .4.21) The term inside the bracket in Equation 5.19) θ 12 π D 12 = 2 ⋅ R b ⋅ sin  = 2 ⋅ R b ⋅ sin . Using that identity. ⋅ 2sin .20 can be used to expand the numerator of the second ln term of Equation 5. 2 k The distance between Strand 1 and any other Strand i is given by: θ 1i (i – 1) ⋅ π D 1i = 2 ⋅ R b ⋅ sin .ln . 2 k θ 13 2π D 13 = 2 ⋅ R b ⋅ sin  = 2 ⋅ R b ⋅ sin . 2 k (5.. the angle between Strand 1 and any other Strand i is given by: (i – 1) ⋅ 2π θ 1i = ( i – 1 ) ⋅ θ 12 = k The distances between the various strands are given by: (5. The numerator represents the product of the radius and the distances between Strand i and all of the other strands. ⋅ … ⋅ 2 sin  ⋅ … ⋅ 2sin k k k ( k – 1 ) .21 is a trigonometric identity that 1 is merely equal to the number of strands k.18: RD s ⋅ D 12 ⋅ … ⋅ D 1i ⋅ … ⋅ D 1k = RD s ⋅ R b k−1 (i – 1)π 2π π 2 sin .22) qp Rb k ⋅ RD s V p1 = .
Various types of insulation materials are used and each will have a range of values for the relative permittivity.0 Equation 5.1 gives the range of values of relative 2 permittivity for four common insulation materials.5–2. Therefore. the equation for the shunt admittance of the concentric neutral cable is given by: 77.ln .= V p1 R b 1 k ⋅ RD s ln .µS/mile R b 1 k ⋅ RD s ln .– .5–3. Crosslinked polyethlyene is a very popular insulation material.6 2.– .22 gives the voltage drop from the phase conductor to Neutral Strand 1.Rb RD c k (5.3–6. .Rb RD c k where (5. The electric ﬁeld of a cable is conﬁned to the insulation material. underground spacings are given in inches.118 TABLE 5.5 2. If the minimum value of relative permittivity is assumed (2.ln .0 2. so the radii of the phase conductor (RDc) and the strand conductor (RDs) should be speciﬁed in inches.2 Determine the threephase shunt admittance matrix for the concentric neutral line of Example 4. Care must be taken that the units for the various radii are the same. the capacitance from phase to ground for a concentric neutral cable is given by: qp 2 πε C pg = . Equation 5.22 gives the voltage drop between the phase conductor and ground. Table 5.3).4–8. Typically.23) ε = ε0 εr = permittivity of the medium ε0 = permittivity of free space = 0.01420 µF/mile εr = relative permittivity of the medium.24) Example 5. Since the neutral strands are all grounded.3619 y ag = 0 + j .1 Distribution System Modeling and Analysis Typical Values of Relative Permittivity (εr) Material Polyvinyl Chloride (PVC) EthylenePropylene Rubber (EPR) Polyethylene (PE) CrossLinked Polyethlyene (XLPE) Range of Values of Relative Permittivity 3.2 in Chapter 4.
= 0.6132 1 .5.24 .2: Rb = R = 0.0641 RD s = . The tapeshielded conductor can be visualized as a concentric neutral cable where the number of strands k has become inﬁnite.6132 inch Diameter of the 250. the electric ﬁeld is conﬁned to the insulation so that the relative permittivities of Table 5.567 inch.03205 inch 2 Substitute into Equation 5.⋅ ln  RD c k Rb 77.2835 13 0. As with the concentric neutral cable.5569 119 [ y abc ] = 5.000 AA phase conductor = 0.567 RD c = .= j96. 0.4 TapeShielded Cable Underground Lines A tapeshielded cable is shown in Figure 5.5569 µS/mile 0.Shunt Admittance of Overhead and Underground Lines From Example 4. 0.– .0511 ft.6132 The phase admittance for this threephase underground line is j96.= 0.5569 0 0 0 j96. When k in Equation 5.⋅ ln 13 ⋅ 0.2835 inch 2 Diameter of the #14 CU concentric neutral strand = 0.24: 77.– .1 will apply. Referring to Figure 5.5. = 0.0641 inch. Rb is the radius of a circle passing through the center of the tape shield.03205 ln 0.3619 y ag = j .5569 0 0 µS/mile 0 j96.3619 y ag = j Rb 1 k ⋅ RD s ln .
1840 inch 2 2 Substitute into Equation 5.= 0. the outside diameter of the tape shield is 0. the equation for the shunt admittance of a tapeshielded conductor becomes: 77.= j89.3619 y bg = j .005 R b = .184 . Therefore.3 in Chapter 4. The radius of a circle passing through the center of the tape shield is 5 T = .= .5 Tapeshielded conductor. approaches inﬁnity.4375 inch 2 2 The diameter of the 1/0 AA phase conductor = 0.368 RD c = . the second term in the denominator approaches zero.88 – 0.= 0.3 Determine the shunt admittance of the singlephase tapeshielded cable of Example 4.3179 µS/mile Rb  0.3.88 inch.= j .= 0. From Example 4.µS/mile Rb ln RD c (5. The thickness of the tape shield (T) is 5 mils.005 1000 d s – T 0.25) Example 5.3619 77.120 Distribution System Modeling and Analysis R AL or CU Phase Conductor b Insulation CU Tape Shield Jacket FIGURE 5.3619 y ag = 0 + j .25: 77.4375 ln ln RD c 0.368 inch dp 0.= .
Assume that the 3 × 3 admittance matrix is given in S/mile. Then the threephase capacitance currents as a function of the linetoground voltages are given by: Icap a Icap c y aa y ab y ac y ca y cb y cc V ag (5. the sequence admittances will be exactly the same as the phase admittances.26) V cg (5.3179 0 µS/mile 0 0 121 5.5 Sequence Admittance The sequence admittances of a threephase line can be determined in much the same manner as the sequence impedances were determined in Chapter 4.29) For a threephase overhead line with unsymmetrical spacing. a threephase underground line with three identical cables will only have the diagonal terms since there is no mutual capacitance between phases. In fact. the offdiagonal terms will be nonzero. the sequence admittance matrix will be full. However. That is.28 the sequence admittance matrix is given by: y 00 y 01 y 02 –1 [ y 012 ] = [ A s ] ⋅ [ y abc ] ⋅ [ A s ] = y 10 y 11 y 12 y 20 y 21 y 22 –1 –1 (5.28) (5.Shunt Admittance of Overhead and Underground Lines The line is on phase b so that the phase admittance matrix becomes: 0 [ y abc ] = 0 0 0 0 j89.27) Icap b = y ba y bb y bc ⋅ V bg [ Icap abc ] = [ y abc ] ⋅ [ VLG abc ] Applying the symmetrical component transformations: [ Icap 012 ] = [ A s ] ⋅ [ Icap abc ] = [ A s ] ⋅ [ y abc ] ⋅ [ A s ] ⋅ [ VLG 012 ] From Equation 5. .
1997.3 Determine the phase admittance matrix in µS/mile for the singlephase line of Problem 4. Underground cables have a much higher shunt admittance per mile than overhead lines. Arnold.122 Distribution System Modeling and Analysis 5.2. Determine the phase admittance matrix and sequence admittance matrix in µS/mile for the threephase line of Problem 4. T. Boston.P. 2nd ed. PWSKent Publishing.3 using RDAP. 5. GA.. Distribution lines are typically so short that the shunt admittance can be ignored. 2nd edition. there will be cases where the shunt admittance of an underground cable should be included in the analysis process.5.6 Determine the phase admittance matrix in µS/mile for the singlephase concentric neutral cable of Problem 4.D. Carrollton. and Sarma.7 . 2..1. 5. 5.4 5.9. Southwire Company. 5. Again. Why make a simplifying assumption when it is not necessary? References 1. lightly loaded overhead lines where the shunt admittance should be included. 5.1 5. Power Cable Manual.5 Verify the results of Problems 5. However.. Determine the phase admittance matrix and sequence admittance matrix for the threephase concentric neutral line of Problem 4.10. Problems Determine the phase admittance matrix [ Y abc ] and sequence admittance matrix [Y012] in µS/mile for the threephase overhead line of Problem 4. the approach to take is to go ahead and model the shunt admittance for both overhead and underground lines. and Mercier. and 5.2 Determine the phase admittance matrix in µS/mile for the twophase line of Problem 4. 1995.1. there are cases of long. J.2. M.3.6 Summary Methods for computing the shunt capacitive admittance for overhead and underground lines have been presented in this chapter. Power System Analysis and Design. Glover. C. When the analysis is being done using a computer.D. 5.
5.9 and 5.10 5.Shunt Admittance of Overhead and Underground Lines 5.7 using RDAP.6 and 5. Determine the phase admittance for the threephase tapeshielded cable line of Problem 4. 5.11 Verify the results of Problems 5.9 Determine the phase admittance matrix in µS/mile for the singlephase tapeshielded cable line of Problem 4. .10 using RDAP.13.8 123 Verify the results of Problems 5.12.
.
Chapters 4 and 5 developed the method for the computation of the phase impedance and phase admittance matrices without assuming transposition of the lines. When a line segment is twophase (Vphase) or singlephase. or singlephase overhead or underground line is shown in Figure 6.1. The values of the impedances and admittances in Figure 6. 6.6 Distribution System Line Models The modeling of distribution overhead and underground line segments is a critical step in the analysis of a distribution feeder. Therefore. Rows and columns of zeros for the missing phases represent twophase and singlephase lines. some of the impedance and admittance values will be zero. For the line segment of Figure 6.1 Exact Line Segment Model The exact model of a threephase. That is. the phase impedance matrix derived in Chapter 4 has been multiplied by the length of the line segment. It is important to include the actual phasing of the line and the correct spacing between conductors.1) m 125 . the equations relating the input (Node n) voltages and currents to the output (Node m) voltages and currents are developed as follows: Kirchhoff’s current law applied at Node m: Iline a Iline b Iline c n Ia = Ib Y aa Y ab Y ac V ag 1 + . one set of equations can be developed to model all overhead and underground line segments. Recall in Chapters 4 and 5 that in all cases the phase impedance and phase admittance matrices were 3 × 3.· Y ba Y bb Y bc · V bg 2 Ic m Y ca Y cb Y cc V cg (6. The phase admittance matrix derived in Chapter 5 has also been multiplied by the length of the line segment.1 represent the total impedances and admittances for the line.1. twophase. Those matrices will be used in the models for overhead and underground line segments.
In condensed form Equation 6.1 becomes: 1 [ Iline abc ] n = [ I abc ] m + .5) (6.2) = n (6.126 Ia n Ib n + Iline a Iline b Iline c Distribution System Modeling and Analysis Ia m Z ab Z bc Z ca Ib m Ic m + 1 [Yabc] 2 + Node .n + Z aa Z bb Z cc Node .6) (6.4) where (6.3) m Z ca Z cb Z cc Iline c In condensed form Equation 6.2 into Equation 6.[ Y abc ] · [ VLG abc ] m 2 Kirchhoff’s voltage law applied to the model gives: Vag Vbg Vcg Vag Vbg Vcg Z aa Z ab Z ac Iline a + Z ba Z bb Z bc · Iline b m (6.[ Y abc ] · [ VLG abc ] m 2 Collecting terms: 1 [ VLG abc ] n = [ U ] + .1 Threephase line segment model.· [ Z abc ] · [ Y abc ] · [ VLG abc ] m + [ Z abc ] · [ I abc ] m 2 1 0 0 [U] = 0 1 0 0 0 1 (6.7) .m + Vag n Vag m Vbg m Vcg m  Vbg n Ic n Vcg n + 1 [Yabc] [ICabc] n 2  [ICabc] m FIGURE 6.4: 1 [ VLG abc ] n = [ VLG abc ] m + [ Z abc ] · [ I abc ] m + .3 becomes: [ VLG abc ] n = [ VLG abc ] m + [ Z abc ] · [ Iline abc ] m Substituting Equation 6.
2 into Equation 6.14: 1 [ I abc ] n = [ Y abc ] + .12) .· [ Y abc ] · [ VLG abc ] n 2 2 Substitute Equation 6.10) Iline a Y aa Y ab Y ac Vag 1 = Iline b + .13: 1 1 [ I abc ] n = [ I abc ] m + .· [ Z abc ] · [ Y abc ] 2 [ b ] = [ Z abc ] The input current to the line segment at Node n is Ia Ib Ic n 127 (6.· [ Y abc ] ⋅ [ Z abc ] · [ Y abc ] · [ VLG abc ] m 4 1 + [ U ] + .9) (6.15) (6.[ Y abc ] · [ VLG abc ] m + .13) (6.· [ Y abc ] 2 2 1 × [ U ] + .· [ Z abc ] · [ Y abc ] · [ VLG abc ] m + [ Z abc ] · [ I abc ] m 2 Collecting terms in Equation 6.· [ Y abc ] · [ Z abc ] [ I abc ] m 2 (6.6 into Equation 6.6 is of the general form: [ VLG abc ] n = [ a ] · [ VLG abc ] m + [ b ] · [ I abc ] m where 1 [ a ] = [ U ] + .14) (6.· Y ba Y bb Y bc · Vbg 2 Vcg Iline c Y ca Y cb Y cc (6.· [ Y abc ] · [ VLG abc ] n 2 Substitute Equation 6.11) n In condensed form Equation 6.12: 1 1 [ I abc ] n = [ I abc ] m + .11 becomes: 1 [ I abc ] n = [ Iline abc ] m + .8) (6.Distribution System Line Models Equation 6.[ Y abc ] · [ VLG abc ] m + .
24) Sometimes it is necessary to compute the voltages at Node m as a function of the voltages at Node n and the currents entering Node m.8 and 6.· [ Z abc ] · [ Y abc ] 2 Equations 6.16 can be put into partitioned matrix form: [ VLG abc ] n [ I abc ] n [ VLG abc ] m = [a] [b] · [c] [d] [ I abc ] m (6.20) The inverse of the abcd matrix is simple because the determinant is [a] · [d] – [b] · [c] = [U] Using the relationship of Equation 6.19) Equation 6.18) (6.19 is very similar to the equation used in transmission line analy1 sis when the ABCD parameters have been deﬁned.17) (6.23) (6.128 Equation 6.19 can be turned around to solve for the voltages and currents at Node m in terms of the voltages and currents at Node n: [ VLG abc ] m [ I abc ] m = [a] [b] [c] [d] –1 · [ VLG abc ] n [ I abc ] n (6.21) Since the matrix [a] is equal to the matrix [d].16) (6. This is true in the iterative technique that is developed in Chapter 10.21.” Equation 6. Equation 6.22 in expanded form becomes: [ VLG abc ] m = [ a ] · [ VLG abc ] n – [ b ] · [ I abc ] n [ I abc ] m = – [ c ] · [ VLG abc ] n + [ d ] · [ I abc ] n (6. . In the case here. Equation 6.15 is of the form: Distribution System Modeling and Analysis [ I abc ] n = [ c ] · [ VLG abc ] m + [ d ] · [ I abc ] m where 1 [ c ] = [ Y abc ] + . the abcd parameters are 3 × 3 matrices rather than single variables and will be referred to as the “generalized line matrices.22) (6.· [ Y abc ] ⋅ [ Z abc ] · [ Y abc ] 4 1 [ d ] = [ U ] + .20 becomes: [ VLG abc ] m [ I abc ] m = [ d ] – [ b ] · [ VLG abc ] n –[ c ] [ a ] [ I abc ] n (6.
0482 0.4615 + j1.4236 0. A common method of describing the degree of unbalance is to use the National Electrical Manufacturers Association 2 (NEMA) deﬁnition of voltage unbalance as given in Equation 6.9774 – j1.25) (6.7034 [ y abc ] = j · 376.1. As a result.1 A balanced threephase load of 6000 kVA.000ft.1560 + j.3849 0.9 lagging power factor is being served at Node m of a 10.27) (6.25 is of the form: [ VLG abc ] m = [ A ] · [ VLG abc ] n – [ B ] · [ I abc ] m where [A] = [a] –1 –1 129 (6.8362 – j0.3911 . 0.26) (6. there will be different values of voltage drop on each of the three phases.0780 0. [c].3849 [ z abc ] = 0.5017 0.4666 + j1.1580 + j0.8 for the Bus m voltages gives: [ VLG abc ] m = [ a ] · { [ VLG abc ] n – [ b ] · [ I abc ] m } Equation 6.6712 – j1. threephase line segment.1535 + j0. determine the linetoground voltages and line currents at the source end (Node n) of the line segment. [b].7034 – j1.0651 j5.1 and 5.169 µ S/mile – j0. the voltages on a distribution feeder become unbalanced even when the loads are balanced.5017 0.29) Example 6.1580 + j0. Determine the generalized line constant matrices [a]. and [d].47 kV. Using the generalized matrices.28) [B] = [a] · [b] –1 Because the mutual coupling between phases on the line segments are not equal.4576 + j1.4236 Ω/mile 0.8362 j5.Distribution System Line Models Solving Equation 6.169 j5. SOLUTION The phase impedance matrix and the shunt admittance matrix for the line segment as computed in Examples 4.1560 + j0. The conﬁguration and conductors of the line segment are those of Example 4.9911 · [ C abc ] = – j1.· 100% V average (6.1535 + j0.1 are 0. 12.29 : Maximum deviation from average V unbalance = .
If more signiﬁcant ﬁgures are displayed.7409 – j3.2907 + j0.8837 + j1.0 0 0 0 1.9502 0.0 0 0 0 1.0 0 0 0 1.99999117 + j0.1 = 0.9852 0.2140 j10.8023 Ω 0.0 Because the elements of the phase admittance matrix are so small.2955 + j0.9852 0.2907 + j0.0417 0. 6.2992 + j0. line segment.1 term is c 1.0 0 0 0 1.2140 – j1.0000127144 .0000044134 + j0. the (1.4777 j11.8837 + j1. 6.130 Distribution System Modeling and Analysis For the 10. the total phase impedance matrix and shunt admittance matrix are 0.9502 0.2955 + j0. and 6.1) element of these matrices is a 1.000ft. The generalized matrices computed according to Equations 6.· [ Z abc ] · [ Y abc ] = 2 1.8667 + j2.7290 0.4777 – j1.2992 + j0.2955 + j0.7290 0.7290 [ b ] = [ Z abc ] = 0.8741 + j2.7290 [ Z abc ] = 0. if more signiﬁcant ﬁgures are displayed.2955 + j0.17.9502 0.0172 0 0 0 [c] = 0 0 0 0 0 0 [d] = 1. the [a] and [d] matrices appear to be the unity matrix. the 1.10.2907 + j0. Again.2907 + j0.9.0417 0.0172 j10.2104 µS It should be noted that the elements of the phase admittance matrix are very small.8023 0.8023 0.1 = – 0. the elements of the [c] matrix appear to be zero.2992 + j0.9502 0.8741 + j2.8023 0.18 are 1 [ a ] = [ U ] + .00000395 Also.3208 – j2.3322 – j2.0 0.2992 + j0.3322 [ Y abc ] = – j3.8667 + j2.
the load current matrix is 277. The degree of voltage unbalance is of concern since.57 [ VLG abc ] n = [ a ] · [ VLG abc ] m + [ b ] · [ I abc ] m = 7451. the .70/1. Using the NEMA deﬁnition for voltage unbalance (Equation 6.84 A 277.47 m For a 0.16 The linetoground voltages at Node n are computed to be 7538. the phase admittance matrix can be neglected.25/ – 118.84 [ I abc ] m = 277. the operating characteristics of a threephase induction motor are very sensitive to voltage unbalance. The magnitude of the linetoground voltage at the load is 12470 VLG = .79 3 · 12.56/0 = 7199.79/ – 25.= 277.93 It is important to note that the voltages at Node n are unbalanced even though the voltages and currents at the load (Node m) are perfectly balanced.29).Distribution System Line Models 131 The point here is that for all practical purposes. This is a result of the unequal mutual coupling between phases. the linetoground voltage matrix at the load is V ag V bg V cg m 7199.79/ – 145.56/120 The magnitude of the load currents is I 6000 = . for example.= 7199.11/121.30 V 7485.56/ – 120 V 7199.79/94.9 lagging power factor.56 3 Selecting the phaseatoground voltage as reference.
30 = 1.83 [ I abc ] n = [ c ] · [ VLG abc ] m + [ d ] · [ I abc ] m = 277.73/94. the perunit voltages at Bus n are 7538.56 7485.56). it does give an indication of how the unequal mutual coupling can generate an unbalance.11/121.70 – 7491.93 V ag V bg V cg n perunit By converting the voltages to perunit it is easy to see that the voltage drop by phase is 4.6275% 7491.0350/ – 118. and 3. The line currents at Node n are computed to be 277.71% for Phase a.70 + 7451.0%.69 = 47.93 1. 3.1 that the shunt admittance of a line is so small that it can be neglected.577 1.57 1 = .7451.71/ – 25.30 7199.70 Although this may not seem like a large unbalance. Figure 6.2 shows the modiﬁed line segment model with the shunt admittance neglected.= .01 V unbalance = .· 100% = 0.25 + 7485. A 6.82 277. Selecting rated linetoground voltage as base (7199.132 voltage unbalance is Distribution System Modeling and Analysis V ag n + V bg n + V cg n 7538.73/ – 148.= 7491.25/ – 118.11 V average = .2 The Modiﬁed Line Model It was demonstrated in Example 6.69 3 3 Vdeviation max = 7538.70/1.17 Comparing the computed line currents at Node n to the balanced load currents at Node m.50% for Phase b.01 47.0471/1.97% for Phase c. a very slight difference is noted that is another result of the unbalanced voltages at Node n and the shunt admittance of the line segment. . It is important to know that NEMA standards require that induction motors be derated when the voltage unbalance exceeds 1.0397/121.
33) (6.Distribution System Line Models Ia n Ib n + Ic n + Iline a Iline b Iline c Ia m Z ab Z bc Z ca Ib m Ic m + + 133 Node . However.2 Modiﬁed line segment model. it is possible to use “equivalent” linetoneutral voltages so that the equations derived to this point will still apply. the generalized matrices become: [a] = [U] [ b ] = [ Z abc ] [c] = [0] [d] = [U] [A] = [U] [ B ] = [ Z abc ] (6.n + Z aa Z bb Z cc Node . the voltage drops down the line must be in terms of the linetoline voltages and line currents.31) (6.34) (6.35) If the line is a threewire delta.37) .m + Vag n Vbg n Vag m Vbg m Vcg m  Vcg n  FIGURE 6.30) (6. When the shunt admittance is neglected. Writing the voltage drop equations in terms of linetoline voltages for the line in Figure 6.2 results in: V ab V bc V ca n V ab = V bc V ca m vdrop b vdrop a + vdrop b – vdrop c vdrop c vdrop a (6.36) where Iline a vdrop a Z aa Z ab Z ac vdrop b = Z ba Z bb Z bc · Iline b vdrop c Z ca Z cb Z cc Iline c (6.32) (6.
36 for Phase ab: Vab n = Vab m + vdrop a – vdrop b but Vab n = Van n – Vbn n Vab m = Van m – Vbn m (6.39 into Equation 6.54/95. .56/0 [ VLG abc ] n = 7199.40) Equation 6. Example 6.39) Substitute Equations 6.56/ – 120 V 7199. Determine also the voltage unbalance and the complex powers of the load.56/ – 145.134 Distribution System Modeling and Analysis Expanding Equation 6.38) (6. This is very important since it makes the development of general analyses techniques the same for fourwire wye and threewire delta systems.1 will be used to supply an unbalanced load at Node m.97/ – 24. The balanced linetoground voltages are 7199. Assume that the voltages at the source end (Node n) are balanced threephase at 12.8 A 305.5 = 277.40 can be broken into two parts in terms of equivalent linetoneutral voltages: Van n = Van m + vdrop a Vbn n = Vbn m + vdrop b (6.2 The line of Example 6.41) The conclusion here is that it is possible to work with equivalent linetoneutral voltages in a threewire delta line.38: Van n – Vbn n = Van m – Vbn m + vdrop a – vdrop b (6.2 Determine the linetoground and linetoline voltages at the load end (Node m) using the modiﬁed line model.47 kV linetoline.56/120 The unbalanced currents measured at the source end are given by: Ia Ib Ic n 249.
49 28.71 – 6916.9501 0.2 A The linetoground voltages at the load end are 6942.2993 + j0.19 –1 0 1 6887.7290 [ B ] = [ Z abc ] = 0. [ I abc ] m is equal to [ I abc ] n .71/117.= 6916.31 11903.05 V V ab V bc V ca V .54/95.43/28.53/ – 1.35/ – 121.53/ – 1.23/148.2907 + j0.20 The linetoline voltages at the load can be computed by: 6942.0172 Ω Since this is the approximate model.8838 + j1.43 1 –1 0 = 0 1 – 1 · 6918.35 + 6887.71 V average = .47 12008.8024 0.49 V unbalance = .47 [ VLG abc ] m = [ A ] · [ VLG abc ] n – [ B ] · [ I abc ] m = 6918.31 For this condition the average load voltage is 6942.97/ – 24.5 = 277.4119% 6916.2954 + j0.9852 0.20 = 28.2993 + j0.71/117.56/ – 145.· 100 = 0.2954 + j0.62/ – 92.9501 0. Therefore: Ia Ib Ic m 249.53 + 6918.55 = 12024.8024 0.8 305.8666 + j2.20 V 3 The maximum deviation from the average is on Phase c so that: Vdeviation max = 6887.Distribution System Line Models SOLUTION 135 The [A] and [B] matrices for the modiﬁed line model are 1 0 0 [A] = [U] = 0 1 0 0 0 1 0.55 6887.35/ – 121.8740 + j2.0417 0.2907 + j0.7290 0.
11 V cg · I c 6.3 The Approximate Line Segment Model Many times the only data available for a line segment will be the positive and zero sequence impedances.26/24.03 V ag · I a 1 = .· V bg · I b = 1920. Applying the approximate impedance matrix. the voltage at Node n is computed to be V ag V bg V cg n V ag = V bg V cg m 1 + .· 3 ( 2 · Z+ + Z0 ) ( Z0 – Z+ ) ( Z0 – Z+ ) ( Z0 – Z+ ) ( 2 · Z+ + Z0 ) ( Z0 – Z+ ) ( Z0 – Z+ ) ( Z0 – Z+ ) ( 2 · Z+ + Z0 ) Ia · Ib Ic m (6.42) [ Z seq ] = The reverse impedance transformation results in the following approximate phase impedance matrix: [ Z approx ] = [ A s ] · [ Z seq ] · [ A s ] ( 2 · Z+ + Z0 ) ( Z0 – Z+ ) ( Z0 – Z+ ) ( Z0 – Z+ ) ( 2 · Z+ + Z0 ) ( Z0 – Z+ ) –1 (6.136 Distribution System Modeling and Analysis The complex powers of the load are Sa Sb Sc 1735.42/23. This is the same result that is achieved if the line is assumed to be transposed. the sequence impedance matrix is given by: Z0 0 0 0 Z+ 0 0 0 Z+ (6.· 3 Notice that the approximate impedance matrix is characterized by the three diagonal terms being equal and all mutual terms being equal.43) ( Z0 – Z+ ) ( Z0 – Z+ ) ( 2 · Z+ + Z0 ) (6.45) .47/22.44) 1 [ Z approx ] = . The approximate line model can be developed by applying the reverse impedance transformation from symmetrical component theory. Using the known positive and zero sequence impedances.25 kVA 1000 2104.
51) Figure 6. Equation 6.50) (6.49) The same process can be followed in expanding Equation 6.46) (6.· ( I a + I b + I c ) 3 (6.· ( I a + I b + I c ) 3 ( Z0 – Z+ ) Vcg n = Vcg m + Z + · I c + .Distribution System Line Models In condensed form.45 for Phases b and c.47) Equation 6. Figure 6.45 for the Phase a voltage at Node n results in: 1 Vag n = Vag m + .46 is of the form: [ VLG abc ] n = [ a ] [ VLG abc ] m + [ b ] · [ I abc ] m where [a] = unity matrix [b] = [Zapprox] 137 (6.45 becomes: [ VLG abc ] n = [ VLG abc ] m + [ Z approx ] · [ I abc ] m Note that Equation 6.3 illustrates the approximate line segment model.· ( I a + I b + I c ) 3 (6.48 by adding and subtracting the term ( Z 0 – Z + ) I a and then combining terms and simplifying: 1 ( 2Z + + Z 0 )I a + ( Z 0 – Z + )I b + ( Z 0 – Z + )I c Vag n = Vag m + . The ﬁnal results are ( Z0 – Z+ ) Vbg n = Vbg m + Z + · I b + .45 can be expanded and an equivalent circuit for the approximate line segment model can be developed.{ ( 2Z + + Z 0 )I a + ( Z 0 – Z + )I b + ( Z 0 + Z + )I c } 3 (6.48) Modify Equation 6. Solving Equation 6.{ ( 3Z + )I a + ( Z 0 – Z + ) ( I a + I b + I c ) } 3 ( Z0 – Z+ ) Vag n = Vag m + Z + · I a + . 3 + ( Z 0 – Z + )I a – ( Z 0 – Z + )I a 1 Vag n = Vag m + .3 is a simple equivalent circuit for the line segment since no mutual coupling has .
138
Distribution System Modeling and Analysis
FIGURE 6.3 Approximate line segment model.
to be modeled. It must be understood, however, that the equivalent circuit can only be used when transposition of the line segment has been assumed.
Example 6.3
The line segment of Example 4.1 is to be analyzed assuming that the line has been transposed. In Example 4.1 the positive and zero sequence impedances were computed to be z + = 0.3061 + j0.6270 z 0 = 0.7735 + j1.9373 Ω/mile
Assume that the load at Node m is the same as in Example 6.1, that is kVA = 6000, kVLL = 12.47, Power factor = 0.8 lagging
Determine the voltages and currents at the source end (Node n) for this loading condition.
SOLUTION
The sequence impedance matrix is 0.7735 + j1.9373 0 0 Ω/mile 0 0.3061 + j0.6270 0 0 0 0.3061 + j0.6270
[ z seq ] =
Performing the reverse impedance transformation results in the approximate phase impedance matrix: [ z approx ] = [ A s ] · [ z seq ] · [ A s ]
–1
0.4619 + j1.0638 0.1558 + j0.4368 0.1558 + j0.4368 Ω/mile [ Z approx ] = 0.1558 + j0.4368 0.4619 + j1.0638 0.1558 + j0.4368 0.1558 + j0.4368 0.1558 + j0.4368 0.4619 + j1.0638
Distribution System Line Models
139
For the 10,000ft. line, the phase impedance matrix and the [b] matrix are 10,000 [ b ] = [ Z approx ] = [ z approx ] · 5280 0.8748 + j2.0147 0.2951 + j0.8272 0.2951 + j0.8272 [ b ] = 0.2951 + j0.8272 0.8748 + j2.0147 0.2951 + j0.8272 Ω 0.2951 + j0.8272 0.2951 + j0.8272 0.8748 + j2.0147 Note in the approximate phase impedance matrix that the three diagonal terms are equal and all of the mutual terms are equal. Again, this is an indication of the transposition assumption. From Example 6.1 the voltages and currents at Node m are 7199.56/0 [ VLG abc ] m = 7199.56/ – 120 V 7199.56/120 277.79/ – 25.84 [ I abc ] m = 277.79/ – 145.84 A 277.79/94.16 Using Equation 6.47: 7491.65/ – 1.73 [ VLG abc ] n = [ a ] · [ VLG abc ] m + [ b ] · [ I abc ] m = 7491.65/ – 118.27 V 7491.65/ – 121.73 Note that the computed voltages are balanced. In Example 6.1 it was shown that when the line is modeled accurately, there is a voltage unbalance of 0.6275%. It should also be noted that the average value of the voltages at Node n in Example 6.1 was 7491.69 volts. The V ag at Node n can also be computed using Equation 6.48: ( Z0 – Z+ ) Vag n = Vag m + Z + · I a +  · ( I a + I b + I c ) 3 Since the currents are balanced, this equation reduces to: Vag n = Vag m + Z + · I a = 7199.56/0 + ( 0.5797 + j1.1875 ) · 277.79/ – 25.84 = 7491.65/−1.73 V
140
Distribution System Modeling and Analysis
It can be noted that when the loads are balanced and transposition has been assumed, the threephase line can be analyzed as a simple singlephase equivalent, as was done in the calculation above.
Example 6.4
Use the balanced voltages and unbalanced currents at Node n in Example 6.2 and the approximate line model to compute the voltages and currents at Node m.
SOLUTION
From Example 6.2 the voltages and currents at Node n are given as: 7199.56/0 [ VLG abc ] n = 7199.56/ – 120 V 7199.56/120 Ia Ib Ic
n
249.97/ – 24.5 = 277.56/ – 145.8 A 305.54/95.2
The [A] and [B] matrices for the approximate line model are where [A] = Unity matrix [B] = [Zapprox] The voltages at Node m are determined by: 6993.12/ – 1.93 [ VLG abc ] m = [ A ] · [ VLG abc ] n – [ B ] · [ I abc ] n = 6881.16/ – 121.61 V 6880.24/117.50 The voltage unbalance for this case is computed by: 6993.12 + 6881.16 + 6880.24 V average =  = 6918.17 3 Vdeviation max = 6993.12 – 6918.17 = 74.94 74.94 V unbalance =  · 100 = 1.0833% 6918.17 Note that the approximate model has led to a higher voltage unbalance than the exact model.
Distribution System Line Models
141
6.4
Summary
This chapter has developed the exact, modiﬁed, and approximate line segment models. The exact model uses no approximations; that is, the phase impedance matrix, assuming no transposition is used, as well as the shunt admittance matrix. The modiﬁed model ignores the shunt admittance. The approximate line model ignores the shunt admittance and assumes that the positive and zero sequence impedances of the line are the known parameters. This is paramount to assuming the line is transposed. For the three line models, generalized matrix equations have been developed. The equations utilize the generalized matrices [a], [b], [c], [d], [A], and [B]. The example problems demonstrate that, because the shunt admittance is very small, the generalized matrices can be computed neglecting the shunt admittance with very little if any error. In most cases the shunt admittance can be neglected; however, there are situations where the shunt admittances should not be neglected. This is particularly true for long, rural, lightly loaded lines, and for many underground lines.
References
1. Glover, J.D. and Sarma, M., Power System Analysis and Design, 2nd ed., PWSKent Publishing, Boston, 1995. 2. ANSI/NEMA Standard Publication No. MG11978, National Electrical Manufacturers Association, Washington, D.C.
Problems
6.1 A twomilelong, threephase line uses the conﬁguration of Problem 4.1. The phase impedance matrix and shunt admittance matrix for the conﬁguration are
0.3375 + j1.0478 0.1560 + j0.5017 0.1535 + j0.3849 [ z abc ] = 0.1560 + j0.5017 0.3465 + j1.0179 0.1580 + j0.4236 Ω/mile 0.1535 + j0.3849 0.1580 + j0.4236 0.3414 + j1.0348 j5.9540 – j2.0030 – j0.7471 [ y abc ] = – j2.0030 j6.3962 – j1.2641 µ S/mile – j0.7471 – j1.2641 j5.6322
142
Distribution System Modeling and Analysis
The line is serving a balanced threephase load of 10,000 kVA, with balanced voltages of 13.2 kV linetoline and a power factor of 0.85 lagging. (1) Determine the generalized matrices. (2) For the given load, compute the linetoline and linetoneutral voltages at the source end of the line. (3) Compute the voltage unbalance at the source end. (4) Compute the source end complex power per phase. (5) Compute the power loss by phase over the line. (Hint: Power loss is deﬁned as power in minus power out.)
6.2
The positive and zero sequence impedances for the line of Problem 6.1 z+ = 0.186 + j0.5968 Ω/mile z0 = 0.6534 + j1.907 Ω/mile
are
Repeat Problem 6.1 using the approximate line model. The line of Problem 6.1 serves an unbalanced, grounded wye, connected constant impedance load of:
6.3
Z ag = 15/30 Ω,
Z bg = 17/36.87 Ω,
Z cg = 20/25.84 Ω
The line is connected to a balanced threephase 13.2kV source. (1) (2) (3) (4) (5) Determine the load currents. Determine the load linetoground voltages. Determine the complex power of the load by phase. Determine the source complex power by phase. Determine the power loss by phase and the total threephase power loss.
6.4 6.5
Repeat Problem 6.3, but change the impedance on Phase b to 50/36.87 Ω
The twophase line of Problem 4.2 has the following phase impedance matrix: 0.4576 + j1.0780 0 0.1535 + j0.3849 0 0 0 0.1535 + j0.3849 0 0.4615 + j1.0651
[ z abc ] =
Ω/mile
The line is two miles long and serves a twophase load such that: S ag = 2000 kVA at 0.9 lagging power factor and voltage of 7620/0 V S cg = 1500 kVA at 0.95 lagging power factor and voltage of 7620/120 V
Distribution System Line Models Neglect the shunt admittance and determine the following:
143
(1) The source linetoground voltages using the generalized matrices. (Hint: Even though Phase b is physically not present, assume that it is with a value of 7620/ – 120 V and is serving a 0 kVA load.) (2) The complex power by phase at the source. (3) The power loss by phase on the line. The singlephase line of Problem 4.3 has the following phase impedance matrix:
6.6
0 0 0 [ z abc ] = 0 1.3292 + j1.3475 0 Ω/mile 0 0 0 The line is one mile long and is serving a singlephase load of 2000 kVA, 0.95 lagging power factor at a voltage of 7500/ – 120 V. Determine the source voltage and power loss on the line. (Hint: As in the previous problem, even though Phases a and c are not physically present, assume they are, and, along with Phase b, make up a balanced threephase set of voltages.)
6.7 The threephase concentric neutral cable conﬁguration of Problem 4.10 is two miles long and serves a balanced threephase load of 10,000 kVA, 13.2 kV, 0.85 lagging power factor. The phase impedance and shunt admittance matrices for the cable line are
0.8040 + j0.4381 0.3176 + j0.0276 0.2824 – j0.0184 [ z abc ] = 0.3176 + j0.0276 0.7939 + j0.3966 0.3176 + j0.0276 0.2824 – j0.0184 0.2824 – j0.0184 0.8040 + j0.4381 [ y abc ] = j117.52 0 0 0 j117.52 0 0 0 j117.52
Ω/mile
µ S/mile
(1) Determine the generalized matrices. (2) For the given load, compute the linetoline and linetoneutral voltages at the source end of the line. (3) Compute the voltage unbalance at the source end. (4) Compute the source end complex power per phase. (5) Compute the power loss by phase over the line. (Hint: Power loss is deﬁned as power in minus power out.)
39 Ω/mile Determine the source voltage and the power loss for the loading condition.84 Ω The line is connected to a balanced threephase 13. Determine the load linetoground voltages.0 kV and 0.8 Distribution System Modeling and Analysis The line of Problem 6.2 kV source. Determine the complex power of the load by phase.87 Ω.7 serves an unbalanced grounded wye connected constant impedance load of: Z ag = 15/30 Ω. Determine the power loss by phase and the total threephase power loss. Z bg = 50/36. at 8.12 is two miles long and serves a singlephase load of 3000 kVA. The tapeshielded cable singlephase line of Problem 4. (1) (2) (3) (4) (5) Determine the load currents.144 6. Z cg = 20/25.5717 0 0 [ y abc ] = 0 0 0 0 0 µ S/mile 0 j140. .9 0 0 0 [ z abc ] = 0 0 0 0 0 0.9 lagging power factor. The phase impedance and shunt admittances for the line are 6.5287 + j0. Determine the source complex power by phase.
Each nominal system voltage pertains to a portion of the system bounded by transformers or utilization equipment.11995 for “Electric Power Systems and Equipment Voltage Ratings (60 Hertz)” 1 provides the following deﬁnitions for system voltage terms: • System Voltage: the root mean square (rms) phasor voltage of a portion of an alternatingcurrent electric system. and to which certain operating characteristics of the system are related. load tap changing transformers (LTC). Each system voltage pertains to a portion of the system that is bounded by transformers or utilization equipment. • Maximum System Voltage: the highest system voltage that occurs under normal operating conditions.1 Standard Voltage Ratings The American National Standards Institute (ANSI) standard ANSI C84. there must be some means of regulating the voltage so that every customer’s voltage remains within an acceptable level.7 Regulation of Voltages The regulation of voltages is an important function on a distribution feeder. • Service Voltage: the voltage at the point where the electrical system of the supplier and the electrical system of the user are connected. As the loads on the feeders vary. and shunt capacitors. • Nominal System Voltage: the voltage by which a portion of the system is designated. 7. 145 . and the highest system voltage for which equipment and other components are designed for satisfactory continuous operation without derating of any kind. Common methods of regulating the voltage are the application of steptype voltage regulators. • Utilization Voltage: the voltage at the line terminals of utilization equipment.
corrective measures shall be undertaken within a reasonable time to improve voltages to meet Range A.” Voltage unbalance is deﬁned as: Max.1) The task for the distribution engineer is to design and operate the distribution system so that under normal steadystate conditions the voltages at the . For a normal threewire 120/240 volt service to a user. The occurrence of voltages outside of these limits should be infrequent. the ANSI standard recommends that the “electric supply systems should be designed and operated to limit the maximum voltage unbalance to 3% when measured at the electricutility revenue meter under a noload condition.146 Distribution System Modeling and Analysis • Nominal Utilization Voltage: the voltage rating of certain utilization equipment used on the system. • Range B: Voltages above and below Range A. An oversimpliﬁcation of the voltage ranges is • Range A: Electric supply systems shall be so designated and operated such that most service voltages will be within the limits speciﬁed for Range A. The ANSI standard speciﬁes two voltage ranges. deviation from average voltage Voltage unbalance = . the Range A and Range B voltages are • Range A – Nominal utilization voltage = 115 V – Maximum utilization and service voltage = 126 V – Minimum service voltage = 114 V – Minimum utilization voltage = 110 V • Range B – Nominal utilization voltage = 115 V – Maximum utilization and service voltage = 127 V – Minimum service voltage = 110 V – Minimum utilization voltage = 107 V These ANSI standards give the distribution engineer a range of normal steadystate voltages (Range A) and a range of emergency steadystate voltages (Range B) that must be supplied to all users. In addition to the acceptable voltage magnitude ranges. When these voltages occur.⋅ 100% Average voltage (7.
The regulators and their controls allow the voltage output to vary as the load varies. Singlephase regulators can be connected in wye. Stepvoltage regulators can be singlephase or threephase. or open delta. Before proceeding to the autotransformer. 7.1 is modiﬁed by referring the primary impedance (Z1) to the secondary side as shown in Figure 7. The standards for these markings are such that at no load.1. Referring to Figure 7. A common device used to maintain system voltages is the stepvoltage regulator. The voltage change is obtained by changing the number of turns (tap changes) of the series winding of the autotransformer. Under a steadystate load condition. A stepvoltage regulator is basically an autotransformer with a load tap changing mechanism on the series winding.1 the highvoltage transformer terminals are denoted by H1 and H2. In Figure 7. the currents I1 and I2 will be in phase. the voltage between H1 and H2 will be in phase with the voltage between X1 and X2. the exact equivalent circuit of Figure 7.2. Without introducing a signiﬁcant error.1 Twowinding transformer exact equivalent circuit. and the lowvoltage terminals are denoted by X1 and X2 . and that the voltage unbalance will not exceed 3%. the total leakage Z1 H1 + N 1: N 2 I ex Z2 + IS I1 + I2 + X1 VS H2  Ym E1  E2  VL  X2 FIGURE 7. a review of twowinding transformer theory and the development of generalized constants will be presented. . in addition to operating as singlephase devices. An autotransformer can be visualized as a twowinding transformer with a solid connection between a terminal of the primary side of the transformer to a terminal on the secondary.Regulation of Voltages 147 meters of all users will lie within Range A.2.2 TwoWinding Transformer Theory The exact equivalent circuit for a twowinding transformer is shown in Figure 7. delta.
4) (7.2 Twowinding transformer approximate equivalent circuit.8) (7.3) In order to better understand the model for the stepregulator. Referring to Figure 7. impedance of the transformer is given by: Zt = n t ⋅ Z 1 + Z 2 N2 n t = N1 2 (7.7) (7.6 can be written as: VS = a ⋅ VL + b ⋅ I2 where 1 a = nt Z b = t nt (7.5) (7. a model for the twowinding transformer will ﬁrst be developed.2) where (7. Equation 7. the equations for the ideal transformer become: N2 E 2 = .⋅ I 2 = n t ⋅ I 2 N1 Applying KVL in the secondary circuit: E2 = V L + Zt ⋅ I 2 Z 1 1 V S = E 1 = .⋅ V L + t ⋅ I 2 nt nt nt In general form.9) .2.⋅ E 1 = n t ⋅ E 1 N1 N2 I 1 = .⋅ E 2 = .6) (7.148 Distribution System Modeling and Analysis N 1: N 2 + I ex H1 Z t + + IS VS H2  I1 Ym E1  I2 + X1 E2  VL  X2 FIGURE 7.
18) (7. The only difference at this point is that only a singlephase twowinding transformer is being modeled.7 for the load voltage yields: 1 b V L = .5 into Equation 7.⋅ V S – . These two equations are of the same form as Equations 6.⋅ V L + t + n t ⋅ I 2 nt nt In general form.11 can be written as: IS = c ⋅ VL + d ⋅ I2 where Ym c = nt Ym ⋅ Z d = t + n t nt 149 (7.10) (7. and d will be expanded to 3 × 3 matrices for all possible threephase regulator connections. particularly in an iterative process. b.⋅ V L + Y m ⋅ t ⋅ I 2 + n t ⋅ I 2 nt nt Ym Ym ⋅ Z I S = .15) . Solving Equation 7.10: Z 1 I S = Y m ⋅ .13) (7.17) (7.11) (7.7 and 7.16 that were derived in Chapter 6 for the threephase line models.15 results in: VL = A ⋅ VS − B ⋅ I2 where A = nt B = Zt (7.16) (7. Later in this chapter the terms a.⋅ I 2 a a Substituting Equations 7.9 into Equation 7. c. Sometimes. Equation 7.12) (7.12 are used to compute the input voltage and current to a twowinding transformer when the load voltage and current are known.6 and 7.8 and 6. the output voltage needs to be computed knowing the input voltage and the load current.Regulation of Voltages The input current to the twowinding transformer is given by: IS = Ym ⋅ VS + I1 Substitute Equations 7.14) Equations 7.8 and 7.
= 0.1 A singlephase transformer is rated 75 kVA.1 N1 V rated 1 2400 The equivalent transformer impedance referred to the lowvoltage side: Zt = Z2 + n t ⋅ Z1 = 0.150 Distribution System Modeling and Analysis Again.= 10 nt 0.0115 Ω (lowvoltage winding impedance) Ym = 1.0001 nt A = nt = 0.0235 2 −4 −4 . The transformer “turns ratio” is N2 V rated 2 240 n t = .16 is of the same form as Equation 6. b. c.0085 nt Ym ⋅ Z d = t + n t = 0.92 × 10 − j8. 2400240 volts.= 0.0122 + j0.1 Zt b = .612 + j1.0235 The generalized constants are 1 1 a = .52 × 10 S (referred to the highvoltage winding) Determine the generalized a.1222 + j0.0122 + j0.1002 – j0. Example 7.2 Ω (highvoltage winding impedance) Z2 = 0.= 0. The transformer has the following impedances and shunt admittance: Z1 = 0. Later in this chapter the expressions for A and B will be expanded to 3 × 3 matrices for all possible threephase transformer connections.= . d constants and the A and B constants.0019 – j0. Equation 7.1 Ym c = .26.= .0061 + j0.1 B = Zt = 0.= .235 0.
013 Y base Example 7. and d parameters computed above: VS = a ⋅ VL + b ⋅ I2 = 2466.0235 ) ⋅ ( 312. determine the load voltage.= 0. Determine the source voltage and current using the generalized constants.0122 + j0. the perunit impedance of the transformer is computed by: 2 V rated 2 240 Z base2 = .0/0 V For future reference.= 0.000 Zt 0.= .9/1.0655 perunit 0.75 A Using the computed source voltage and the load current.15 ) – ( 0. V L = A ⋅ V S – B ⋅ I S = ( 0.5 perunit 0.0115 Z pu = .9 ) = 312.0345/62.Regulation of Voltages 151 Assume that the transformer is operated at rated load (75 kVA) and rated voltage (240 V) with a power factor of 0.9 lagging.52 · Y pu = ./ – cos ( 0. V L = 240/0 75 ⋅ 1000 –1 I 2 = .1 ) ⋅ ( 2466.0148 – j0. c.92 ⋅ 10 – j 8.= 0.5/−25.= .013 S 2 kV 1 ⋅ 1000 –4 –4 Ym 1.0122 + j 0.768 Z base2 2 The perunit shunt admittance is computed by: kVA Y base1 = . b.= .768 Ω kVA ⋅ 1000 75.67/ – 28.84 ) V L = 240.1 demonstrates that the generalized constants provide a quick method for analyzing the operating characteristics of a twowinding transformer.= 0.84 240 Applying the values of the a.9/1. .15 V IS = c ⋅ VL + d ⋅ I2 = 32.5/ – 25.
Connecting the highvoltage terminal H1 to the lowvoltage terminal X2 as shown in Figure 7.21) N2 + E2 X H1 + VS H 2 FIGURE 7. VS is the source voltage and VL is the load voltage.19) Since the source voltage VS is equal to E1.5: E1 + nt ⋅ E1 = (1 + nt) ⋅ E1 = VL + Zt ⋅ I2 (7. The lowvoltage winding of the twowinding transformer will be referred to as the “series” winding of the autotransformer. Applying KVL in the secondary circuit: E1 + E2 = VL + Zt ⋅ I2 Using the ideal transformer relationship of Equation 7. Generalized constants similar to those of the twowinding transformer can be developed for the autotransformer.⋅ I L 1 + nt 1 + nt (7.3.⋅ V L + . and I2 is equal to IL .3 Stepup autotransformer.3 can create a “stepup” autotransformer.4 and 7.5 still apply. 2 I ex Zt I2 + X1 N1 I1 + VL IS Ym E1  .152 Distribution System Modeling and Analysis 7.20 can be modiﬁed to: Zt 1 V S = . The ideal transformer equations of 7. The total equivalent transformer impedance is referred to the “series” winding. In Figure 7. The source is connected to terminals H1 and H2. Equation 7.3 The TwoWinding Autotransformer A twowinding transformer can be connected as an autotransformer. and the highvoltage winding will be referred to as the “shunt” winding. while the load is connected between the X1 terminal and the extension of H2.20) (7.
⋅ I 2 1 + nt 1 + nt Ym Ym ⋅ Z I S = .24) (7. Applying KVL in the secondary circuit: E1 − E2 = VL + Zt ⋅ I2 (7. 7.25: Zt 1 I S = ( 1 + n t ) ⋅ I 2 + Y m .⋅ V L + t + n t + 1 ⋅ I 2 1 + nt 1 + nt IS = c ⋅ VL + d ⋅ I2 Ym c = 1 + nt Ym ⋅ Z d = t + n t + 1 1 + nt 153 (7.22) where (7.24.29) .28) Equations 7.4.27) (7.28 deﬁne the generalized constants relating the source voltage and current as functions of the output voltage and current for the stepup autotransformer.. and 7.21 into Equation 7.27. 7. Generalized constants can be developed for the stepdown connection following the same procedure as that for the stepup connection.25) (7. The twowinding transformer can also be connected in the stepdown connection by reversing the connection between the shunt and series winding as shown in Figure 7. 1 + nt Applying KCL at input node H1: IS = I1 + I2 + I ex IS = (1 + nt ) ⋅ I2 + Ym ⋅ VS Substitute Equation 7.23.23) (7.26) where (7.Regulation of Voltages VS = a ⋅ VL + b ⋅ IL 1 a = 1 + nt Zt b = .⋅ V L + .
for the stepdown connection.33 and 7.30) Since the source voltage VS is equal to E1 .23 and 7.⋅ V L + .30 can be modiﬁed to: Zt 1 V S = . This will also be the case for the c and d constants. the c and d constants are deﬁned by: .24 for the stepup connection. Equation 7.33) (7.34) It is observed at this point that the only difference between the a and b constants of Equations 7.31) (7. and I2 is equal to IL .5: E1 − nt ⋅ E1 = (1 − nt) ⋅ E1 = VL + Zt ⋅ I2 (7.4 Stepdown autotransformer.44 for the stepdown connection.32) where (7.154 Distribution System Modeling and Analysis I2 E2 + X H1 + VS H FIGURE 7. and Equations 7. is the sign in front of the turns ratio (nt ).⋅ I L 1 – nt 1 – nt = a ⋅ VL + b ⋅ IL 1 a = 1 – nt Zt b = 1 – nt (7. Therefore. + X2 1 IS Iex N1 I1 + VL Ym E1  Using the ideal transformer relationship of Equation 7.
In general.39) (7.37 through 7. it is sometimes necessary to relate the output voltage as a function of the source voltage and the output current.36) The only difference between the deﬁnitions of the generalized constants is the sign of the turns ratio nt. As with the twowinding transformer.41) (7.40.= 1 ± n t a b B = . Solving Equation 7. the generalized constants can be deﬁned by: 1 a = 1 ± nt Zt b = 1 ± nt Ym c = 1 ± nt Ym ⋅ Z d = t 1 ± n t 1 ± nt (7.35) (7.43) (7.Regulation of Voltages Ym c = 1 – nt Ym ⋅ Z d = t + 1 – n t 1 – nt 155 (7.40) In Equations 7. They are of exactly the same form as was derived for . the sign in the equations will be positive for the stepup connection.⋅ I 2 a a VL = A ⋅ VS – B ⋅ I2 where 1 A = .= Zt a (7. then.32 for the output voltage: 1 b V L = . and negative for the stepdown connection.37) (7.⋅ V S – .42) (7.38) (7.44) The generalized equations for the stepup and stepdown autotransformers have been developed.
⋅ E 1 ⋅ I 1 nt but therefore ( 1 ± nt ) kVA auto = . They are of exactly the same form as was derived for the twowinding transformer and for the line segment in Chapter 6. Deﬁne the rated kVA and rated voltages of the twowinding transformer and autotransformer as: kVA xfm kV auto A Vrated 1 Vrated 2 Vauto S Vauto L = = = = = = kVA rating of the twowinding transformer kVA rating of the autotransformer E1 = rated source voltage of the twowinding transformer E2 = rated load voltage of the twowinding transformer rated source voltage of the autotransformer rated load voltage of the autotransformer For the following derivation.45) (7. For the singlephase autotransformer the generalized constants are single values. neglect the voltage drop through the series winding impedance: Vauto L = E1 ± E2 = (1 ± nt) ⋅ E1 The rated output kVA is then: kVA auto = V auto L ⋅ I 2 = ( 1 ± n t ) ⋅ E 1 ⋅ I 2 but therefore ( 1 ± nt ) kVA auto = .3.46) (7.48) E 1 ⋅ I 1 = kVA xfm I1 I 2 = nt (7. but will be expanded later to 3 × 3 matrices for threephase autotransformers.1 Autotransformer Ratings The kVA rating of the autotransformer is the product of the rated input voltage VS times the rated input current IS.⋅ kVA xfm nt (7. 7. or the rated load voltage VL times the rated load current IL.47) .156 Distribution System Modeling and Analysis The generalized equations for the stepup and stepdown autotransformers have been developed.
For the stepup connection.1 ( 1.92 – j8.= 0.1 The voltage ratings are V auto S = V rated 1 = 2400 V V auto L = V rated 1 + V rated 2 = 2400 + 240 = 2640 V Therefore.⋅ 75 = 825 kVA 0.0235 ) d = .1 = 1.= ( 1.0235 b = .= 0. Example 7. 24002640 V. The rated kVA of the autotransformer using Equation 7.48 is given by: 1 + 0.5/ – 25.92 – j8.000 –1 I 2 = . In general.84 A V auto L 2640 Determine the generalized constants: 1 a = .52 ) ⋅ 10 ⋅ ( 0.1 ( 1.1 kVA auto = . Suppose now that the autotransformer is supplying rated kVA at rated voltage with a power factor of 0. so the kVA rating of the autotransformer will be considerably greater than the kVA rating of the twowinding transformer.1002 – j0. the autotransformer would be rated as 825 kVA. the turns ratio nt will be a relatively small value.0214 1 + 0.000005 1 + 0.0122 + j0.1 0. the sign of nt will be positive while the stepdown will use the negative sign.52 ) ⋅ 10 –4 c = .48 gives the kVA rating of a twowinding transformer when connected as an autotransformer.1 –4 –4 .1.7364 – j7.0111 + j0.= .2 The twowinding transformer of Example 7. From Example 7.Regulation of Voltages 157 Equation 7. Determine the source voltage and current: V L = V auto L = 2640/0 V kVA auto ⋅ 1000 825.+ 1 + 0.9091 1 + 0. Determine the kVA and voltage ratings of the autotransformer.9 ) = 312./ – cos ( 0.7455 ) ⋅ 10 1 + 0.0122 + j0.1 the turns ratio was determined to be nt = 0.1 is connected as a stepup autotransformer.9 lagging.
The base impedance of the twowinding transformer referred to the lowvoltage winding (series winding of the autotransformer) is V rated 2 Zbase xfm = kVA xfm ⋅ 1000 2 (7.84 = 2640.3.51) .06/ – 26.00/0 V 7.0/0. V rated 2 = Rated load voltage of the twowinding transformer.11 A When the load side voltage is determined knowing the source voltage and load current.1 B = Z t = 0. the A and B parameters are needed: A = 1 + n t = 1.2 PerUnit Impedance The perunit impedance of the autotransformer based upon the autotransformer kVA and kV ratings can be developed as a function of the perunit impedance of the twowinding transformer based upon the twowinding transformer ratings.04/0.5/ – 25.107 – B ⋅ 312.5/ – 25. that is V rated 2 V nominal = V rated 1 = nt (7.5/ – 25.84 = 345.0235 The load voltage is then: V L = A ⋅ 2406. Let Zpu xfm = Perunit impedance of the twowinding transformer based upon the twowinding kVA and kV ratings.49) The actual impedance of the transformer referred to the lowvoltage (series) winding is V auto series Zt actual = Zt pu ⋅ Zbase xfm = Zt pu ⋅ kVA xfm ⋅ 1000 2 (7.0111 + j0.1 V I S = c ⋅ 2640/0 + d ⋅ 312.84 = 2406.50) Assume that the rated source voltage of the autotransformer is the nominal voltage of the system.158 Distribution System Modeling and Analysis Applying the generalized constants: V S = a ⋅ 2640/0 + b ⋅ 312.
the value of nt is 0.53 into Equation 7.55) Equation 7.⋅ kVA xfm ⋅ 1000 nt 2 nominal 2 2 (7.48 and 7. Recall that the shunt admittance is represented on the source side of the twowinding transformer.= n t ⋅ ( 1 ± n t ) ⋅ Zt pu 2 V rating 2 n t ⋅ ( 1 ± n t ) ⋅ kVA xfm ⋅ 1000 2 (7.54) Zauto pu (7. The point is that the perunit impedance of the autotransformer is very small compared to that of the twowinding transformer.1 ) ⋅ Zt pu = 0.55 gives the relationship between the perunit impedance of the autotransformer and the perunit impedance of the twowinding transformer. When the autotransformer is connected to boost the voltage 10%.1 ⋅ ( 1 + 0.50 and 7.51 into Equation 7.54: V rating 2 kVA xfm ⋅ 1000 = Zt pu ⋅ .52: 2 Vrated  2 nt V = .11 ⋅ Zt pu (7.= kVA auto ⋅ 1000 1 ± nt .1.56) The perunit shunt admittance of the autotransformer can be developed as a function of the perunit shunt admittance of the twowinding transformer.52) Zbase auto (7. . and Equation 7.53) V rating 2 Zbase auto = n t ⋅ ( 1 ± n t ) ⋅ kVA xfm ⋅ 1000 The perunit impedance of the autotransformer based upon the rating of the autotransformer is Zt actual Zauto pu = Zbase auto Substitute Equations 7.Regulation of Voltages 159 The base impedance for the autotransformer referenced to the nominal system voltage is V nominal Zbase auto = kVA auto ⋅ 1000 Substitute Equations 7.57 becomes: Zauto pu = 0.
Yautopu = perunit admittance of the autotransformer based upon the autotransformer ratings.1 It has been shown that the perunit impedance and admittance values based upon the autotransformer kVA rating and nominal voltage are approximately onetenth that of the values for the twowinding transformer. 2 2 (7.59: kVA xfm ⋅ 1000 V rated 1 Yauto pu = Yt pu ⋅ .⋅ Yt pu kVA auto ( 1 ± nt ) ( 1 ± nt ) .57) The actual shunt admittance referred to the source side of the twowinding transformer is kVA xfm ⋅ 1000 Yt source = Yt pu ⋅ Ybase source = Yt pu ⋅ 2 V rating 1 The perunit shunt admittance for the autotransformer is given by: Yt source V rated 1 Yauto pu = .59) .60 becomes: 0.= .0909 ⋅ Yt pu 1 + 0. The base admittance of the twowinding transformer referenced to the source side is given by: kVA xfm ⋅ 1000 Ybase source = 2 V rating 1 (7.= Yt pu ⋅ . Equation 7.⋅ 2 kVA auto ⋅ 1000 V rated 1 (7.⋅ Yt pu = 0.58) (7.1 Ya pu = .1.= Yt source ⋅ Ybase auto kVA auto ⋅ 1000 Substitute Equation 7.58 into 7.160 Let Distribution System Modeling and Analysis Ytpu = Ympu = perunit admittance of the twowinding transformer based upon the transformer ratings. For an autotransformer in the raise connection with nt = 0.60) kVA xfm kVA xfm nt Yauto pu = Yt pu ⋅ .60 shows that the perunit admittance based upon the autotransformer ratings is much smaller than the perunit admittance of the twowinding transformer.⋅ kVA xfm nt Equation 7.
001341 – j 0. These equivalent circuits included the series impedance and shunt admittance.= 0. and when the autotransformer is to be a component of a system.014746 – j0.013 2 2400 1. very little error will be made by neglecting both the series impedance and shunt admittance of the equivalent circuit. Determine the perunit shunt admittance based upon the twowinding transformer ratings: 75 ⋅ 1000 Ybase source = . If a detailed analysis of the autotransformer is desired.52 ⋅ 10 Yt pu = .1432 0.005949 0.1432 2 2400 1.52 ⋅ 10 Yauto pu = . Determine the perunit admittance based upon the autotransformer kVA rating and a nominal voltage of 2400 V.065434 In this section the equivalent circuit of an autotransformer has been developed for the “raise” and “lower” connections. it has been shown that these values are very small. –4 –4 –4 –4 −4 −4 .92 ⋅ 10 – j 8. and determine the ratio of the perunit admittance of the autotransformer to the perunit admittance of the twowinding transformer: 825 ⋅ 1000 Ybase auto = .014746 – j0.2 the kVA rating of the twowinding transformer connected as an autotransformer was computed to be 825 kVA.92 ⋅ 10 – j 8.065434 0.52 ⋅ 10 S 1.2 is Yt = Ym = 1.013 2. In Example 7.= 0.Regulation of Voltages 161 Example 7. and the voltage ratings 2400−2640 V.= 0.= 0.= 0.3 The shunt admittance referred to the source side of the twowinding transformer of Example 7.0909 0.001341 – j0.92 ⋅ 10 − j8. However. the series impedance and shunt admittance should be included.005949 Ratio = .
162 Distribution System Modeling and Analysis 7. This amounts to a 5/8% change per step. on a 120V base. The more common Type B connection is shown in Figure 7. The stepvoltage regulator control circuit requires the following settings: 1. 2. Bandwidth: the allowed variance of the load center voltage from the set voltage level. .6 controls tap changing on a stepvoltage regulator. The block diagram circuit shown in Figure 7. Step regulators can be connected in a Type A or Type B connection according to the ANSI/IEEE C57. The voltage change is obtained by changing the taps of the series winding of the autotransformer. The load center may be the output terminal of the regulator or a remote node on the feeder. usually in 32 steps. For example.151986 stan2 dard.75V change per step. Standard step regulators contain a reversing switch enabling a ±10% regulator range. The voltage held at the load center will be ± onehalf the bandwidth. if the voltage level is set to 122 V Preventive Autotransformer Reversing Switch Series Winding Source Control CT Shunt Winding Control VT V Load FIGURE 7.5.5 Type B stepvoltage regulator. Voltage Level: the desired voltage (on a 120V base) to be held at the load center. The position of the tap is determined by a control circuit (line drop compensator).4 StepVoltage Regulators A stepvoltage regulator consists of an autotransformer and a load tap changing mechanism. or 0.
4. The kVA rating of the stepvoltage regulator is determined in the same manner as that of the previously discussed autotransformer. and the bandwidth is set to 2 V. The settings consist of R and X settings in volts corresponding to the equivalent impedance between the regulator and the load center. In general. Line Drop Compensator: set to compensate for the voltage drop (line drop) between the regulator and the load center. 7. 7.1 Type A StepVoltage Regulator The detailed equivalent circuit and abbreviated equivalent circuit of a Type A stepvoltage regulator in the raise position is shown in Figure 7. . the regulator will change taps until the load center voltage lies between 121 and 123 V. that if it is desired to include the impedance and admittance.4.6 Stepvoltage regulator control circuit. The required rating of a stepvoltage regulator is based upon the kVA transformed. however.1 SinglePhase StepVoltage Regulators Because the series impedance and shunt admittance values of stepvoltage regulators are so small. 4. this will be 10% of the line rating since rated current ﬂows through the series winding which represents the ±10% voltage change.1. they will be neglected in the following equivalent circuits. Time Delay: length of time that a raise or lower operation is called for before the actual execution of the command. they can be incorporated into the following equivalent circuits in the same way they were originally modeled in the autotransformer equivalent circuit. 3. This setting may be zero if the regulator output terminals are the load center. not the kVA rating of the line. This prevents taps changing during a transient or short time change in current. It should be pointed out.Regulation of Voltages 163 FIGURE 7.7.
The primary circuit of the system is connected. When the Type A connection is in the lower position.4. As shown in Figure 7. Since this is the more common connection.7. The effect of this reversal is to reverse the direction of the currents in the series and shunt windings.2 Type B StepVoltage Regulator The more common connection for stepvoltage regulators is the Type B. the deﬁning voltage and current equations for the voltage regulator will be developed only for the Type B connection.1. The series winding is connected to the shunt winding. In this connection the core excitation varies because the shunt winding is connected directly across the primary circuit.9. which is connected directly to the regulated circuit.7 Type A stepvoltage regulator in the raise position. The deﬁning voltage and current equations for the regulator in the raise position are as follows: + SL  + S IS S IS . In a Type B regulator the core excitation is constant because the shunt winding is connected across the regulated circuit. The series winding is connected to the shunt winding and. The detailed and abbreviated equivalent circuits of a Type B stepvoltage regulator in the raise position is shown in Figure 7. the reversing switch is connected to the L terminal. to the series winding of the regulator in the Type B connection. via taps. Figure 7. 7.164 + Distribution System Modeling and Analysis L E 2 R I2 N2 L + VL IL + N 1 SL + L IL VL E 1 I1 V S  VS  FIGURE 7. via taps. to the regulated circuit.8 shows the equivalent circuit and abbreviated circuit of the Type A regulator in the lower position. in turn. the primary circuit of the system is connected directly to the shunt winding of the Type A regulator.
Regulation of Voltages + 165 L E 2 R  N2 L + VL I2 IL + N 1 I1 E 1 SL IS N2 I2 + R E 2 L + IS IL IL L + S N1 + E 1 I1 VL SL SL + L IL VL V S  VS  FIGURE 7.9 Type B stepvoltage regulator in the raise position. + SL L + V L   + S IS S IS .8 Type A stepvoltage regulator in the lower position. S + IS V S V S  FIGURE 7.
⋅ I 2 = .⋅ I S N1 N1 N2 I L = 1 –  ⋅ I S N 1 IL = aR ⋅ IS (7.⋅ V L N1 N1 N2 V S = 1 –  ⋅ V L N 1 VS = aR ⋅ VL N2 a R = 1 – N1 N1 ⋅ I1 = N2 ⋅ I2 IL = IS − I1 I2 = IS N2 N2 I 1 = . The Type B stepvoltage connection in the lower position is shown in Figure 7.67) Equations 7.71) (7.74) .10.⋅ V L N1 N1 N2 V S = 1 +  ⋅ V L N 1 VS = aR ⋅ VL N1 ⋅ I1 = N2 ⋅ I2 IL = IS + I1 I2 = IS N2 N2 I 1 = .166 VOLTAGE EQUATIONS Distribution System Modeling and Analysis CURRENT EQUATIONS E1 E2 .⋅ E 1 = . The deﬁning voltage and current equations for the Type B stepvoltage regulator in the lower position are as follows: VOLTAGE EQUATIONS CURRENT EQUATIONS E1 E2 .66 and 7.62) (7. connection note that the direction of the currents through the series and shunt windings change. As in the Type A.70) (7.64) (7.⋅ E 1 = .⋅ I 2 = .68) (7.61) (7.65) (7.⋅ I S N1 N1 N2 I L = 1 +  ⋅ I S N 1 IL = aR ⋅ IS N2 a R = 1 + N1 (7. but the voltage polarity of the two windings remain the same.= N1 N2 VS = E1 − E2 VL = E1 N2 N2 E 2 = .67 are the necessary deﬁning equations for modeling a regulator in the raise position.= N1 N2 VS = E1 + E2 VL = E1 N2 N2 E 2 = .66) (7.63) (7.72) (7.69) (7.73) (7.
75) In Equation 7.4.67 and 7. 7.00625 ⋅ Tap + (7. the only difference between the voltage and current equations for the Type B regulator in the raise and lower positions is the sign of the turns ratio (N2/N1). It can now be shown that the generalized abcd constants can also be applied to the stepvoltage regulator.74 give the value of the effective regulator ratio as a function of the ratio of the number of turns on the series winding (N2) to the number of turns on the shunt winding (N1). the relationship between the source voltage and current to the load voltage and current are . the minus sign applies for the raise position and the plus sign for the lower position.10 Type B stepvoltage regulator in the lower position.75. Equations 7. the particular tap position will be known. however. the effective regulator ratio can be given by: a R = 1 − 0.67 and 7.00625 perunit.Regulation of Voltages 167 I I I I I I I FIGURE 7. Each tap changes the voltage by 5/8% or 0. In the ﬁnal analysis.1. Therefore. Equations 7. For both the Type A and Type B regulators. The actual turns ratio of the windings is not known.74 can be modiﬁed to give the effective regulator ratio as a function of the tap position.3 Generalized Constants In previous chapters and sections of this text generalized abcd constants have been developed for various devices.
which requires the voltage transformer in Figure 7.⋅ I L aR (7.4 The Line Drop Compensator The changing of taps on a regulator is controlled by the line drop compensator.75 and the sign convention is given in Table 7. Figure 7. The setting that is most critical is that of R′ and X′ calibrated in volts. The consistent set of base values is determined by .⋅ V L aR V S = aR ⋅ V L I S = aR ⋅ I L 1 I S = .1.168 TABLE 7.1. 7.78a) Type B: b = 0 c = 0 (7. The compensator input voltage is typically 120 volts.11 to reduce the rated voltage to 120 volts. The basic requirement is to force the perunit line impedance to be equal to the perunit compensator impedance.4. it is essential that a consistent set of base values be developed wherein the perunit voltage and currents in the line and in the compensator are equal.78b where aR is given by Equation 7. The compensator is an analog circuit that is a scale model of the line circuit.1 Distribution System Modeling and Analysis Sign Convention Table for Equation 7. These values must represent the equivalent impedance from the regulator to the load center. the generalized constants for a singlephase stepvoltage regulator become: Type A: 1 a = aR a = aR b = 0 c = 0 d = aR 1 d = aR (7. In order to cause this to happen.76) Type B: (7. For a regulator connected linetoground the rated voltage is the nominal linetoneutral voltage. while for a regulator connected linetoline the rated voltage is the linetoline voltage.75 Type A Raise Lower + − Type B − + of the form: Type A: 1 V S = .11 shows a simpliﬁed sketch of the compensator circuit and how it is connected to the distribution line through a potential transformer and a current transformer. The current transformer turns ratio is speciﬁed as CTp:CTs. The purpose of the line drop compensator is to model the voltage drop of the distribution line from the regulator to the load center. where the primary rating (CTp) will typically be the rated current of the feeder.77) Therefore.
Regulation of Voltages MVA rating kV hi .11 Line drop compensator circuit.79) .kV lo I line CTp:CTs I comp X' R line + jX line Load Center 169 R' 1:1 + V drop Npt:1 + V reg + VR Voltage Relay FIGURE 7. With the table of base values developed. and then computing the base voltage and current in the compensator by dividing the system base values by the voltage transformer ratio and current transformer ratio. the compensator R and X settings in ohms can be computed by ﬁrst computing the perunit line impedance: Rline Ω + jXline Ω R pu + jX pu = Zbase line R pu + jX pu CT P = ( Rline Ω + jXline Ω ) ⋅ V LN (7.2 Table of Base Values Base Voltage Current Impedance Line Circuit VLN CTP Compensator Circuit V LN N PT CTS V LN Zbaseline = CT P V LN Zbasecomp = N PT ⋅ CT S selecting a base voltage and current for the line circuit. the base system voltage is selected as the rated linetoneutral voltage (VLN) and the base system current is selected as the rating of the primary winding of the current transformer (CTP). respectively. Table 7. For regulators connected linetoground. TABLE 7.2 gives a table of base values and employs these rules for a regulator connected linetoground.
⋅ V LN N PT ⋅ CT S CT P RcompΩ + jXcompΩ = (RlineΩ + jXlineΩ) ⋅ N PT ⋅ CT S Ω (7. The compensator impedance in ohms is computed by multiplying the perunit impedance by the base compensator impedance: RcompΩ + jXcompΩ = ( R pu + jX pu ) ⋅ Zbase comp CT P V LN RcompΩ + jXcompΩ = (RlineΩ + jXlineΩ) ⋅ .81. the required value for the compensator settings in volts is determined by using Equation 7.= 2401.170 Distribution System Modeling and Analysis The perunit impedance of Equation 7.8 3 .⋅ CT S N PT ⋅ CT S CT P R′ + jX′ = (RlineΩ + jXlineΩ) ⋅ .4. The rated linetoground voltage of the substation transformer is 4160 V S = .11.16 grounded wye.4 Refer to Figure 7.V N PT Knowing the equivalent impedance in ohms from the regulator to the load center. and the equivalent line impedance from the regulator to the load center is 0.80 gives the value of the compensator R and X settings in ohms. (1) Determine the voltage transformer and current transformer ratings for the compensator circuit. (7. This is demonstrated in Example 7. The substation transformer is rated 5000 kVA.79 must be the same in the line and in the compensator.81) Example 7.9 Ω.3 + j0. The compensator R and X settings in volts are determined by multiplying the compensator R and X in ohms times the rated secondary current in amps (CTS) of the current transformer: R′ + jX′ = (RcompΩ + jXcompΩ) ⋅ CTS CT P R′ + jX′ = (RlineΩ + jXlineΩ) ⋅ .80) Equation 7. 115 delta − 4.
16 kV and 0.16 The primary rating of the CT is selected as 700 A.= 2.Regulation of Voltages 171 In order to provide approximately 120 V to the compensator.9 ) ⋅ .3 + j0.= 140 CT S 5 (2) Determine the R and X settings of the compensator in ohms and volts.1 + j6.9 power factor lag.5 R ohms + jX ohms = .= 10. Applying Equation 7.5 + j31.5 The substation transformer in Example 7.5 + j31. the voltage transformer ratio is 2400 N PT = . the CT ratio is CT P 700 CT = .5 V Voltage Level = 120 V (desired voltage to be held at the load center) Bandwidth = 2 V .4 is supplying 2500 kVA at 4.= .3 Ω 5 Understand that the R and X settings on the compensator control board are calibrated in volts.= 693. Example 7. and if the compensator current is reduced to 5 A.5 V 20 The R and X settings in ohms are determined by dividing the settings in volts by the rated secondary current of the current transformer: 10. The regulator has been set so that: R′ + jX′ = 10.9 3 ⋅ 4.78 to determine the settings in volts: 700 R′ + jX′ = ( 0.= 20 120 The rated current of the substation transformer is 5000 I rated = .5 + j31.
172 Distribution System Modeling and Analysis Determine the tap position of the regulator that will hold the load center voltage at the desired voltage level and within the bandwidth. the voltage regulator will have to change taps in the raise position to bring the load center voltage up to the required level. one step change on the regulator changes the voltage 0.9188 .24 Tap = .= 2.9 ) = 346.72 V The voltage across the voltage relay is V R = V reg – V drop = 120. Since this is well below the minimum voltage level of 119.458/45.97/ – 25.3 ) ⋅ 2.02 0. Recall that on a 120V base.09/ 0 V 20 The voltage drop in the compensator circuit is equal to the compensator current times the compensator R and X values in ohms: V drop = ( 2.4783/–25.= 120. the effective regulator ratio assuming a Type B regulator is aR = 1 − 0.24/ – 6.75 This shows that the ﬁnal tap position of the regulator will be “raise 13.09/0 – 16.84 = 16.84 A 140 The input voltage to the compensator is 2401./cos ( 0.1 + j6.19 V The voltage across the voltage relay represents the voltage at the load center.72 = 109.458/45.97/ – 25.75 V. This means that the tap on the regulator needs to be set so that the voltage at the load center lies between 119 and 121 V.” With the tap set at +13.4783/ – 25.84 A 3 ⋅ 4. The number of required tap changes can then be approximated by 119 – 109. The ﬁrst step is to calculate the actual line current: 2500 –1 I line = .00625 ⋅ 13 = 0.= 13.16 The current in the compensator is then: 346.84 I comp = .8/0 V reg = .
2/0 V a 0. the student should go back and.0884 0.15 V N pt 20 Therefore.= 120. That value will be computed to be exactly the same as the load center voltage on the 120V base.3 + j0.9 ) ⋅ 318. As an exercise.84 I L = .= .77/ – 25.97/ – 25.5. using the output voltage and current of the regulator on the +13 tap.= 1.6/ – 5. . The actual linetoground voltage and line current at the load side terminals of the regulator are V S 2401. calculate the voltage across the voltage relay in the compensator circuit.77/ – 25.8/ – 5.15 V On a 120V base.Regulation of Voltages 173 The generalized constants for modeling the regulator for this operating condition are a = a R = 0.2/0 – ( 0.9188 b = 0 c = 0 1 d = .9188 Example 7.15 V LC VLC 120 = .= 2614.84 = 2412.8/0 V L = .9188 I S 346.= 318.= .= . the load center voltage is 2412.84 A d 1.16 kV is measured at the substation transformer lowvoltage terminals. the +13 tap on the regulator has provided the desired voltage at the load center.8/ – 5.0884 The actual linetoground voltage at the load center is V LC = V L – Z line ⋅ I L = 2614.6 Using the results of Examples 7. calculate the actual voltage at the load center assuming the 2500 kVA at 4. It is important to understand that the value of equivalent line impedance is not the actual impedance of the line between the regulator and the load center.
Now the equivalent line impedance can be computed as: V regulator output – V load center Rline Ω + jXline Ω = .4. Two regulators connected in open delta 5. therefore. For this case it is up to the engineer to determine which phase current and voltage will be sampled by the compensator circuit.82) In Equation 7. Two regulators connected in “openwye” (sometimes referred to as “V” phase) 3. Three regulators connected in closed delta A threephase regulator has the connections between the singlephase windings internal to the regulator housing. Singlephase 2. Threephase regulators will only be connected in a threephase wye or closed delta. When three singlephase regulators are connected together. 7. as a result. Typical connections for singlephase stepregulators are 1. The compensator control circuit has been developed. . only one compensator circuit is required.2 ThreePhase StepVoltage Regulators Three singlephase stepvoltage regulators can be connected externally to form a threephase regulator. As a result. the load center is located down the primary main feeder after several laterals have been tapped. each regulator has its own compensator circuit and. The only way to determine the equivalent line impedance value is to run a powerﬂow program of the feeder without the regulator operating.82 the voltages must be speciﬁed in system volts and the current in system amperes. the taps on each regulator are changed separately. and it has been demonstrated how this circuit controls the tap changing of the regulator.174 Distribution System Modeling and Analysis Typically. This section has developed the model and generalized constants for Type A and Type B singlephase stepvoltage regulators. The next section will discuss the various threephase steptype voltage regulators. The threephase regulator is gang operated so that the taps on all windings change the same and. the current measured by the CT of the regulator is not the current that ﬂows all the way from the regulator to the load center.Ω I line (7. Three regulators connected in grounded wye 4. From the output of the program the voltages at the regulator output and the load center are known.
and c. Regardless of whether the regulator is raising or lowering the voltage. When the regulator is in the lower position a reversing switch will have reconnected the series winding so that the polarity on the series winding is now at the output terminal.84) .1 WyeConnected Regulators Three Type B singlephase regulators connected in wye are shown in Figure 7. b.4. In Figure 7.Regulation of Voltages B Ia A + VAn + V an Ib b 175 a Ic c C IC FIGURE 7.12 the polarities of the windings are shown in the raise position.83 is of the form: [ VLN ABC ] = [ a ] ⋅ [ VLN abc ] + [ b ] ⋅ [ I abc ] (7. B.12 Wyeconnected type B regulators. the phasing on the source side of the regulator will use capital letters A. 7.2. the following equations apply: VOLTAGE EQUATIONS V An V Bn = V Cn aR a 0 0 0 aR b 0 0 0 aR c V an ⋅ V bn V cn (7.83) where aR a . The load side phasing will use lowercase letters a. Equation 7. and C. In the regulator models to be developed in the next sections.12. and aR c represent the effective turns ratios for the three singlephase regulators. aR b .
176
CURRENT EQUATIONS
Distribution System Modeling and Analysis
IA IB IC =
1 aR a 0 0
0 1 aR b 0
0 0 1 aR c Ia = Ib Ic (7.85)
or
[ I ABC ] = [ c ] ⋅ [ VLG abc ] + [ d ] [ I abc ]
(7.86)
Equations 7.84 and 7.86 are of the same form as the generalized equations that were developed for the threephase line segment of Chapter 6. For a threephase wyeconnected stepvoltage regulator, neglecting the series impedance and shunt admittance, the generalized matrices are deﬁned as: aR a [a] = 0 0 0 aR b 0 0 0 aR c (7.87)
0 0 0 [b] = 0 0 0 0 0 0 0 0 0 [c] = 0 0 0 0 0 0 1 aR a [d] = 0 0 0 1 aR b 0 0 0 1 aR c
(7.88)
(7.89)
(7.90)
In Equations 7.87 and 7.90, the effective turns ratio for each regulator must satisfy: 0.9 ≤ aR abc ≤ 1.1 in 32 steps of 0.625%/step (0.75 V/step on 120V base) The effective turn ratios (aR a, aR b, and aR c ) can take on different values when three singlephase regulators are connected in wye. It is also possible to have a threephase regulator connected in wye where the voltage and current are
Regulation of Voltages
177
sampled on only one phase, and then all three phases are changed by the same number of taps.
Example 7.7
An unbalanced threephase load is served at the end of a 10,000ft., 12.47kV distribution line segment. The phase generalized matrices for the line segment were computed in Example 6.1 as 1 0 0 [a] = 0 1 0 0 0 1 0.8667 + j2.0417 0.2955 + j0.9502 0.2907 + j0.7290 [ b ] = 0.2955 + j0.9502 0.8837 + j1.9852 0.2992 + j0.8023 0.2907 + j0.7290 0.2992 + j0.8023 0.8741 + j2.0172 For this line the A and B matrices are deﬁned as: [A] = [a]
–1
[ B ] = [ a ] ⋅ [ b ] = [ Z abc ] The linetoneutral voltages at the substation are balanced threephase: 7200/0 [ VLN ABC ] = 7200/ – 120 7200/120 The line currents at the substation for the unbalanced loading are 258/ – 20 [ I abc ] = [ I ABC ] = 288/ – 147 324/86 For the measured substation voltages and currents, the linetoneutral voltages at the load are computed as: 6965.1/ – 2.1 [ Vload abc ] = [ A ] ⋅ [ VLN ABC ] – [ B ] ⋅ [ I abc ] = 6943.1/ – 121.2 6776.7/117.8 The load voltages on a 120V base are determined by dividing by the potential transformer ratio that transforms rated linetoneutral voltage V A V
–1
178 down to 120 V:
Distribution System Modeling and Analysis
7200 N pt =  = 60 120 The load voltages on a 120V base are 116.1/ – 2.1 1 [ V 120 ] =  ⋅ [ Vload abc ] = 115.7/ – 121.2 60 112.9/117.8
V
Three singlephase Type B stepvoltage regulators are connected in wye and installed in the substation. The regulators are to be set so that each linetoneutral load voltage on a 120V base will lie between 119 and 121 volts. The current transformers of the regulators are rated: CT P 600 CT =  =  = 120 5 CT S The equivalent line impedance for each phase can be determined by applying Equation 7.79: 7200/0 – 6965.7/ – 2.1 Zline a =  = 0.5346 + j1.2385 Ω 258/ – 20 7200/ – 12 0 – 6943.1/ – 121.2 Zline b =  = 0.5628 + j0.8723 Ω 288/ – 147 7200/120 – 6776.7/117.8 Zline c =  = 0.6386 + j1.418 Ω 324/86 Even though the three regulators will change taps independently, it is the usual practice to set the R and X settings of the three regulators the same. The average value of the three line impedances above can be used for this purpose: Zlineaverage = 0.5787 + j1.1763 The compensator R and X settings are computed according to Equation 7.78: CT P 600 R′ + jX′ = (RlineΩ + jXlineΩ) ⋅  = (0.5787 + j1.1763) ⋅ 60 N PT R′ + jX′ = 5.787 + j11.763 V The compensator controls are not calibrated to that many signiﬁcant ﬁgures, so the values set are R′ + jX′ = 6 + j12 V
Regulation of Voltages
179
The compensator control will be set such that the voltage level = 120 V with a bandwidth of 2 V. For the same unbalanced loading, and with the threephase wyeconnected regulators in service, the approximate tap settings are 119 – Vload a 119 – 116.1 Tap a =  =  = 3.8736 0.75 0.75 119 – Vload b 119 – 115.7 Tap b =  =  = 4.3746 0.75 0.75 119 – Vload c 119 – 112.9 Tap a =  =  = 8.0723 0.75 0.75 Since the taps must be integers, the actual tap settings will be Tap a = +4 Tap b = +5 Tap c = +9 The effective turns ratio for the three regulators and the resulting generalized matrices are determined by applying Equation 7.75 for each phase: 1 – 0.00625 ⋅ 4 0 0 0 1 – 0.00625 ⋅ 5 0 0 0 1 – 0.00625 ⋅ 8 0.975 0 0 0 0.9688 0 0 0 0.9500 1.0256 0 0 0 1.0323 0 0 0 1.0526
[a] =
=
[d] = [a]
–1
=
The metered substation voltages and currents are the inputs to the three voltage regulators. The output voltages and currents of the regulators are 7384.6/0 [ Vreg abc ] = [ a ] ⋅ [ VLN ABC ] = 7432.3/ – 120 V 7578/120
–1
180
Distribution System Modeling and Analysis
The output currents of the regulators are 251.6/ – 20 [ Ireg abc ] = [ d ] ⋅ [ I sub ] = 279.0/ – 147 307.8/86 With the regulators adjusted, the load voltages can be computed to be 7150.7/ – 2.0 [ Vload abc ] = [ A ] ⋅ [ Vreg abc ] – [ B ] ⋅ [ Ireg abc ] = 7185.5/121.2 7179.1/118.1 On a 120volt base the load voltages are 7150.7/ – 2.0 119.2/ – 2.0 1 [ V 120 ] =  ⋅ 7185.5/ – 121.2 = 119.8/ – 121.2 60 7179.1/118.2 119.7/118.1 V
–1
A
V
With the given regulator taps, the load voltages all now lie between the desired voltage limits of 119 and 121 V per the voltage level and bandwidth settings of the compensator circuit. Example 7.7 is a long example intended to demonstrate how the engineer can determine the correct compensator R and X settings knowing the substation and load voltages and the currents leaving the substation. Generally, it will be necessary to run a powerﬂow study to determine these values. The example demonstrates that with the regulator tap settings, the load voltages lie within the desired limits. The regulator has automatically set the taps for this load condition and, as the load changes, the taps will continue to change in order to hold the load voltages within the desired limits. 7.4.2.2 Closed DeltaConnected Regulators Three singlephase Type B regulators can be connected in a closed delta as shown in Figure 7.13. In the ﬁgure, the regulators are shown in the raise position. The closed delta connection is typically used in threewire delta feeders. Note that the voltage transformers for this connection are monitoring the load side linetoline voltages, and the current transformers are not monitoring the load side line currents.
Regulation of Voltages
A IA
181
SL
S
I ca
L
I
I a' A
I
L
a
a
I c' C
S
IC I bc
SL
I ab IB
S
SL
I B
C IB
L
I b'
Ib Ic
b c
FIGURE 7.13 Closed deltaconnected regulators.
The relationships between the source side and currents and voltages are needed. Equations 7.64 through 7.67 deﬁne the relationships between the series and shunt winding voltages, and currents for a stepvoltage regulator. These must be satisﬁed no matter how the regulators are connected. Kirchhoff’s voltage law is ﬁrst applied around a closed loop, starting with the linetoline voltage between phases A and B on the source side. Refer to Figure 7.13. VAB + VBb + Vba + VaA = 0 N2 V Bb = –  ⋅ V bc N1 N2 V aA =  ⋅ V ab N1 Vba = −Vab (7.91)
but
(7.92)
(7.93)
(7.94)
Substitute Equations 7.92, 7.93, and 7.94 into Equation 7.91 and simplify: N2 N2 V AB = 1 –  ⋅ V ab +  ⋅ V bc = a R ab ⋅ V ab + ( 1 – a R bc ) ⋅ V bc N1 N1
(7.95)
⋅ I A N1 N2 I ca = .⋅ I C N1 Substitute Equations 7.97 into Equation 7.102) Equation 7.103) .182 Distribution System Modeling and Analysis The same procedure can be followed to determine the relationships between the other linetoline voltages.96) Equation 7.96 and 7.99) (7.96 is of the generalized form: [ VLL ABC ] = [ a ] ⋅ [ VLL abc ] + [ b ] ⋅ [ I abc ] (7.101) (7.95 and simplify: N2 N2 I a = 1 –  ⋅ I A + I C = a R ab ⋅ I A + ( 1 – a R ca ) ⋅ I C N1 N 1 (7. The resulting threephase equation is Ia Ib Ic a R ab = 1 – a R ab 0 0 a R bc 1 – a R bc 1 – a R ca 0 a R ca IA ⋅ IB IC (7.100) The same procedure can be followed at the other two load side terminals.97) The relationship between source and load line currents starts with applying KCL at the load side terminal a: Ia = I′ + Ica = IA − Iab + Ica a but N2 I ab = . The ﬁnal threephase equation is V AB V BC V CA = a R ab 0 1 – a R ab 1 – a R bc a R bc 0 0 1 – a R ca a R ca V ab ⋅ V bc V ca (7.98) (7.102 is of the general form: [ I abc ] = [ d ] ⋅ [ I ABC ] –1 (7.
Regulation of Voltages where a R ab = 1 – a R ab 0 0 a R bc 1 – a R bc 1 – a R ca 0 a R ca 183 [d] –1 With the exception of the case when all three regulators are in the neutral −1 position (aR = 1). Note that the potential transformers monitor the linetoline voltages and the current transformers monitor the line currents. The inverse of the resulting matrix gives the matrix [d] so that the generalized form of the current equations can still be applied. As a result. Note that in both the voltage and current equations. increasing the tap in one regulator will affect the tap position of the second regulator.3 Open DeltaConnected Regulators Two Type B singlephase regulators can be connected in the open delta connection.4. Therefore. . Two additional open connections can be made by connecting the singlephase regulators between phases BC and AC. the elements of the inverse matrix can be determined. [ I ABC ] = [ c ] ⋅ [ VLL ABC ] + [ d ] ⋅ [ I abc ] (7. and also between phases CA and BA.2. and then the relationships of the other two possible connections will merely be stated. in most cases the bandwidth setting for the closed delta connection will have to be wider than that for wyeconnected regulators. Shown in Figure 7. However. the matrices [b] and [c] are zero as long as the series impedance and shunt admittance of each regulator are neglected. all of the elements of [d] will be nonzero. Once again. the basic voltage and current relations of the individual regulators are used to determine the relationships between the source side and load side voltages and currents.104) As with the wyeconnected regulators. The open delta connection is typically applied to threewire delta feeders. In general. 7. when the tap positions of each regulator are known. The connection shown in Figure 7.14 is an open delta connection where two singlephase regulators have been connected between phases AB and CB.14 will be used to derive the relationships. a change of the tap position in one regulator will affect voltages and currents in two phases. The closed delta connection can be difﬁcult to apply. [d] does not give a simple expression for each of the elements.
the drop across the series winding is N2 V AL = – .109) (7. (7.14 Open delta connection. the voltage equation is N2 V BC = 1 –  ⋅ V bc = a R cb ⋅ V bc N 1 Kirchhoff’s voltage law must be satisﬁed so that: VCA = −(VAB + VBC) = −aR ab ⋅ Vab − aR cb ⋅ Vbc (7.184 A +  Distribution System Modeling and Analysis IA S V CA L Ia + V a  V AB C I ab + S SL SL ab V ca V BC B + IC L I cb IB Ib b + V bc Ic c + FIGURE 7.⋅ V ab N1 Substituting Equation 7.106) Following the same procedure for the regulator connected across VBC.107) (7. The voltage drop VAB across the ﬁrst regulator consists of the drop across the series winding plus the drop across the shunt winding: VAB = VAL + Vab where VAL = drop across the series winding.105) Paying attention to the polarity marks on the series and shunt windings.106 into Equation 7.108) .105 yields: N2 V AB = 1 –  ⋅ V ab = a R ab ⋅ V ab N 1 (7.
110) – a R ab – a R cb 0 Equation 7. and 7.109 can be put into matrix form: V AB V BC V CA = a R ab 0 0 a R cb 0 V ab ⋅ V bc 0 V ca 185 (7.114) (7.75. [bLL] is zero. as long as the series impedance and shunt admittance of the regulators are neglected.0 a R ab a R cb . Equation 7. the necessary equation is 1 a R ab 0 0 1 a R cb 0 V AB 0 ⋅ V BC V CA (7. Up to this point the relationships between the voltages have been in terms of linetoneutral voltages. 7.111) – a R ab – a R cb 0 The effective turns ratio of each regulator is given by Equation 7.111 gives the linetoline voltages on the source side as functions of the linetoline voltages on the load side of the open delta using the generalized matrices.0 a R ab a R cb [ VLL abc ] = [ A LL ] ⋅ [ VLL ABC ] 1 a R ab where [ A LL ] = 0 0 1 a R cb 0 0 (7. When the load side linetoline voltages are needed as functions of the source side linetoline voltages.– .113) V ab V bc = V ca 1 1 – .115) 1 1 – .107.110 in generalized form is [ VLL ABC ] = [ a LL ] ⋅ [ VLL abc ] + [ b LL ] ⋅ [ I abc ] a R ab where [ a LL ] = 0 0 a R cb 0 0 (7. Again.Regulation of Voltages Equations 7.108. In Chapter 10 it will be shown how to convert this equation using equivalent linetoneutral voltages.– .112) (7.
116) but Therefore.118) In a similar manner.⋅ I a a R ab (7.119) (7.A a N 1 1 I A = .117) therefore (7.121) In generalized form Equation 7.0 – .⋅ I b a R cb a R ab Ic 1 0 0 a R cb 1 a R ab 0 0 (7.116 becomes: 1 – N 2 I = I .122) . Equation 7. then: 1 1 I B = – ( I A + I C ) = – . the current equations become: (7. the current equation for the second regulator is given by: 1 I C = .186 Distribution System Modeling and Analysis The current equations are derived by applying KCL at the L node of each regulator: IA = Ia + I ab N2 I ab = .120) IA IB IC Ia 1 1 = – .⋅ I A N1 (7.⋅ I c a R ab a R cb In matrix form.121 becomes: [ I ABC ] = [ c LL ] ⋅ [ VLL ABC ] + [ d LL ] ⋅ [ I abc ] (7.⋅ I a – .⋅ I c a R cb Because this is a threewire delta line.
124) [ I abc ] = [ D LL ] ⋅ [ I ABC ] a R ab where 0 0 0 0 [ D LL ] = – a R ab 0 – a R cb a R cb (7.0 – a R cb a R ab 0 0 1 a R cb When the series impedances and shunt admittances are neglected. it is possible that the third voltage may not be within the deﬁned limits. The third linetoline voltage will be dictated by the other two (Kirchhoff’s voltage law).123) 1 1 [ d LL ] = – . an equivalent impedance between the regulators and the load center must be computed. Therefore. the equivalent impedance is computed by taking the appropriate linetoline voltage drop and dividing by the sampled line current. care must be taken to recognize that in the open delta connection.15.128) . However.126) The determination of the R and X compensator settings for the open delta follows the same procedure as that of the wyeconnected regulators.15.Regulation of Voltages 187 1 a R ab where 0 0 (7. The open deltaconnected regulators will maintain only two of the linetoline voltages at the load center within deﬁned limits. With reference to Figure 7. Since each regulator is sampling linetoline voltages and a line current. the equivalent impedances are computed as: VR ab – VL ab Zeq a = Ia VR cb – VL cb Zeq c = Ic (7.125) (7. the constant matrix [cLL] will be zero. The load side line currents as functions of the source line currents are given by: Ia Ib Ic a R ab 0 0 IA = – a R ab 0 – a R cb ⋅ I B 0 0 a R cb IC (7. For the open delta connection shown in Figure 7.127) (7. the voltages applied to the compensator are linetoline and the currents are line currents.
470/0 VR bc = 12.4 VL bc = 12.911/ – 1.859/117.117/ – 122.470/ – 120 Ib = 264. For the open delta connection.3 VL ca = 11. as shown in Figure 7.0/70.3 Ia = 308.1 VR ca = 12. Example 7. the potential transformer will transform the system linetoline rated voltage down to 120 V.2/ – 58.188 A + Distribution System Modeling and Analysis  IA S V CA V AB C L Ia a + va + I ab + S SL SL VR ab Ib b  VL ab  Load Center  IC L I cb vb  V BC  VR cb Ic + c vc VL cb + B + IB FIGURE 7.8 A powerﬂow study has been run on a system prior to the installation of an open delta regulator bank at Node R.78. The results of the powerﬂow study are Node R: VR ab = 12. The units of these impedances will be in system ohms. The load center is at Node L. Example 7.2/ – 176.8 demonstrates how the compensator R and X settings are determined knowing the linetoline voltages at the regulator and at the load center.15 Open delta connected to a load center.8. They must be converted to compensator volts by applying Equation 7.16 Circuit for Example 7.470/120 Ic = 297.3 R L FIGURE 7.0 Node L: VL ab = 11.16. .
the load center voltages are V120 ab V120 bc V120 ca 11.= .6/ – 1. The ﬁrst regulator monitors the voltage Vab and the line current Ia.911/ – 1. As computed above.= 7.7002 V N PT 103. which are the negatives of the given bc voltages: 12.117/ – 57.= 1.4 VR ab – VL ab Zeq a = .5 V Two singlephase Type B regulators are to be installed in an open delta connection.1908 + j6.7 VR cb – VL cb Zeq c = .3925 Ω Ic 297.0483 Ω 308.= ( 0.4 114.3 114. it is necessary to use the cb voltages.0 Ia The second regulator monitors the voltage V cb and the line current Ic.2/ – 58.8012 + j9.1665 + j2.911/ – 1. The equivalent line impedance for this regulator is 12.470 N pt = .4 1 = .92 120 500 CT = .92 CT P 500 R′ cb + jX′ cb = Z c ⋅ . In the computation of the equivalent line impedance.= 100 5 On a 120V base.3 103.= ( 0.= .470/60 – 12.= 0. The compensator R and X settings must ﬁrst be determined using the results of the powerﬂow study. the potential transformer ratio and current transformer ratios are selected to be 12.6/ – 122.0483 ) ⋅ .3 Unlike the wyeconnected regulators.8555 V N PT 103.= 0.470/0 – 11. the load center voltages are not within the desired limits of 120 ± 1 V.3 = 116.4945 + j1.Regulation of Voltages 189 For this connection.3/117.0/70.92 11.1495 + j1. The voltage level will be set at 120 V with a bandwidth of 2 V. the compensator settings for the two regulators will be different.3925 ) ⋅ .⋅ 12. The settings calibrated in volts are CT P 500 R′ ab + jX′ ab = Z a ⋅ .117/ – 122.15.1665 + j2.= 103.859/117. The regulators are to be connected between phases AB and BC as shown in Figure 7.92 .
6 Tap cb = .= 2.2 + j6.= 0.98 CT secondary 5 R′ cb + jX′ cb 7.6/ – 1. The number of tap changes necessary to bring the load center voltage into the lower limit of the bandwidth for each regulator will be 119 – 114.2 + j 6.7 With regulators installed and in the neutral position.3 A CT 100 The compensator impedances in ohms are determined by dividing the settings in volts by the secondary rating of the current transformer: R′ ab + jX′ ab 0.75 119 – 116.2/ – 58.75 .8 + j9.470/60 Vcomp cb = .97/70.190 The compensator settings will be Distribution System Modeling and Analysis R′ ab + X′ ab = 0.9 R′ cb + jX′ cb = 7.= .= 3.47 ≈ 6 0.4 V Vrelay cb = Vcomp cb – ( R cb + jX cb ) ⋅ Icomp c = 116.= .0/70.082/ – 58.= 5.3 Ic Icomp c = .= 120/60 V 103. the currents and voltages in the compensator circuits are VR ab 12. the control circuit will send “raise” commands to change the taps on both regulators.6/57.470/0 Vcomp ab = .= 120/0 V N pt 103.8 + j 9.= 3.7 V Since the voltages are below the lower limit of 119.= .34 CT secondary 5 The voltages across the voltage relays in the two compensator circuits are Vrelay ab = Vcomp ab – ( R ab + jX ab ) ⋅ Icomp a = 114.7 R cb + jX cb = .9 R ab + jX ab = .= .20 ≈ 4 0.16 + j1.= .92 N pt 308.= 1.= .44 + j1.0 Ia Icomp a = . and with the same loading.92 – VR bc 12.6 Tap ab = .0 A CT 100 297.
the regulator ratios are a R ab = 1.9625 0 0 [ D LL ] = – 0.975 0.– .6/ – 58 [ I abc ] = [ D LL ] ⋅ [ I ABC ] = 255.0 – 0.9625 0 – 0.975 0 0 0.790/ – 120 V 12. The procedure is the same as was done initially to determine the load center voltages.112) and [DLL] (Equation 7.0 = 0 1.6/70.0 – 0.00625 ⋅ Tap ab = 0. the matrices [ALL] (Equation 7.9625 0.3 There are two ways to test if the voltages at the load center are within the limits. must be deﬁned: 1 0. With the taps adjusted.956/0 [ Vreg abc ] = [ A LL ] ⋅ [ VLL ABC ] = 12. The ﬁrst method is to compute the relay voltages in the compensator circuits.3 A 289.0 0.039 0 0 1 .975 In order to determine the load side regulator voltages and currents.975 – 1.874/120 The output currents from the regulators are 296.123).7/ – 175. a check can be made to determine if the voltages at the load center are now within the limits.00625 ⋅ Tap cb = 0.975 The output voltages from the regulators are 12.9625 a R cb = 1.039 – 1. First the voltages and currents in the compensator .Regulation of Voltages 191 With the taps set at 6 and 4.0256 0 0.9625 0 0 0 [ A LL ] = 1.0256 0 1 1 – .
6762 0.1804 + j1.420/ – 1.= .7604 + j2.1804 + j1.= 2.4773 Ω 0.956/0 VR ab Vcomp ab = .2762 [ Z abc ] = 0. The phase impedance matrix for the line between the regulator and the load center is 0.15.= 123.5/ – 1.6/ – 1.6125 0.2 V VL ca = Vreg ca – v c + v a = 12. the voltage drops per phase are 450.3 V Vrelay cb = Vcomp cb – ( R cb + jX cb ) ⋅ Icomp a = 119.8/142.= .896/70.7/0 V N pt 103.6/ – 58.= .1804 + j1.966/ – 58.447/ – 122.273/ – 118.1804 + j1.3 A CT 100 The voltages across the voltage relays are computed to be Vrelay ab = Vcomp ab – ( R ab + jX ab ) ⋅ Icomp a = 119.8 V Since both voltages are within the bandwidth.92 N pt 296.6762 0.8 The load center linetoline voltages are VL ab = Vreg ab – v a + v b = 12.1 V .7604 + j2.5 [ v abc ] = [ Z abc ] ⋅ [ I abc ] = 309.1/60 V 103.6125 0.8/57.0 A CT 100 289.6 V 402. no further tap changing will be necessary.790/60 Vcomp cb = .92 – VR bc 12.3 Ic Icomp c = .7604 + j2.= 2.0 Ia Icomp a = .1804 + j1.4/ – 106.3 V VL bc = Vreg bc – v b + v c = 12. The actual voltages at the load center can be computed using the output voltages and currents from the regulator and then computing the voltage drop to the load center.4773 0.6762 With reference to Figure 7.= 124.= .192 circuits are computed: Distribution System Modeling and Analysis 12.2762 0.6/70.1804 + j1.
only the terms in the rows and columns associated with the missing phases will be zero. There are two other possible open delta connections using phases BC and AC. The only way to bring that voltage up is to set a higher voltage level on the two regulators. and to determine the desired voltage level and bandwidth. then the [a] and [d] matrices will be of the same form as that of the wyeconnected regulators.2 V V120 ca = 118.1 V The desired voltages are being held on the two phases that have the voltage regulators. The open delta regulator connection using phases AB and CB has been presented. The generalized matrices developed in this chapter are of exactly the same form as those developed for the threephase line segments. and then how it will adjust taps so that the voltages at a remote load center node will be held within the set limits. In the next chapter. The derivations in this chapter were limited to threephase connections. and then CA and BA.5 Summary It has been shown that all possible connections for Type B stepvoltage regulators can be modeled using the generalized matrices. .Regulation of Voltages 193 Dividing the load center linetoline voltages by the potential transformer ratio gives the voltages on the 120V base as: V120 ab = 119. In actual practice.5/ – 1.1/118.3 V V120 bc = 119. the generalized matrices for all threephase transformers will be developed. Generalized matrices for these additional two connections can be developed using the procedures presented in this section. 7.8/ – 122. the rows and columns associated with the missing phases would be set to zero in the matrices developed for the open delta connection. Again. This cannot be helped since that voltage is being dictated by the other two linetoline voltages. If a singlephase regulator is connected linetoneutral or two regulators connected in open wye. The third linetoline voltage is below the limit. This example is very long but has been included to demonstrate how the compensator circuit is set. the only responsibilities of the engineer will be to correctly determine the R and X settings of the compensator circuit. The same can be said for a singlephase regulator connected linetoline.
37 × 10 S (referred to the highvoltage winding) (1) Determine the a.151986. 7.0078 Ω (lowvoltage winding impedance) Ym = 2. American National Standard for Electric Power – Systems and Equipment Voltage Ratings (60 Hertz). 0. ANSI C84. Problems 7. Determine the primary voltage. (2) The transformer is serving an 80kW. and Test Code for StepVoltage and InductionVoltage Regulators. Terminology. A.65 + j0. 0. ANSI/IEEE C57.85 lagging power factor load at 230 V. complex power. Including the impedance and shunt admittance.95 lagging power factor at a voltage of 2000 V. c. 2400240 V.0052 + j0.56 × 10 − j11. (3) Determine the perunit transformer impedance and shunt admittance based upon the transformer ratings.2 (1) Draw the connection diagram. National Electrical Manufacturers Association. current. IEEE Standard Requirements. current. and percent voltage drop. 1988. −4 −4 The singlephase transformer of Problem 7. c. (3) Determine the a. determine the input voltage. Rosslyn.194 Distribution System Modeling and Analysis References 1. How do these values compare to the perunit values of Problem 7. 1996.95 Ω (highvoltage winding impedance) Z2 = 0. Institute of Electrical and Electronic Engineers. The impedances and admittance of the transformer are Z1 = 0. (2) Determine the autotransformer kVA rating. and percent voltage drop. complex power. (5) Determine the perunit impedance and shunt admittance based upon the autotransformer rating.1 A singlephase transformer is rated 100 kVA. b. b. 2. d constants and the A and B constants. and B generalized constants. (4) The autotransformer is serving a load of 800 kVA.1? .11995.1 is to be connected as a stepdown autotransformer to transform the voltage from 2400 V to 2160 V. including the series impedance and shunt admittance. New York. VA. d.
3 195 A Type B stepvoltage regulator is installed to regulate the voltage on a 7200V singlephase lateral. (1) Determine the appropriate PT and CT ratios. The regulators have 32 – 5/8% taps (16 raise and 16 lower). (4) What would be the regulator tap settings for a load of 24 MVA. (2) Determine the equivalent line impedance between the regulator and the load center. (3) Assuming that the voltage level on the regulator has been set at 120 V with a bandwidth of 2 V. (3) The substation is serving a balanced threephase load of 16 MVA. 230 kV delta13.000 ft. Assume the voltage level is set at 121 V and a bandwidth of 2 V.Regulation of Voltages 7.8 kV and the regulators are set in the neutral position.264 + j0. The substation transformer is rated 24 MVA.4 Refer to Figure 7. The regulator taps are set on the +10 position when the voltage and current on the source side of the regulator are Vsource = 7200 V and Isource = 375 at a power factor of 0. 0.9 lagging power factor with the output voltages of the substation transformer balanced threephase 13.9 lagging power factor when the output linetoline voltages of the substation are balanced 13.8 kV wye.11.866 lagging power factor.47kV substation. Determine the ﬁnal tap position for each regulator (they will be the same).58 Ω/mile The distance to the load center node is 10. The equivalent line impedance between the regulators and the load center node is Zline = 0.5 . what tap will the regulator move to? 7. (1) Determine the voltage at the load center.8 kV? (5) What would be the load center voltages for the load of Part 4 above? Three Type B stepvoltage regulators are connected in wye and located on the secondary bus of a 12. (2) Determine the R′ and X′ settings in ohms and volts for the compensator circuit. Three singlephase Type B regulators are connected in wye. The potential transformer and current transformer ratios connected to the compensator circuit are Potential transformer: 7200120 V Current transformer: 500:5 A The R and X settings in the compensator circuit are: R = 5 V and X = 10 V. 0. The feeder is serving an 7.
05 The currents at the substation are 362.6/ – 1.5/117.4 [ I abc ] = 409.6 R′ = 3.2 Determine the ﬁnal tap settings for each regulator.5/98.49 [ VLC abc ] = 6825.0/ – 155.4/ – 154.9/ – 122.1 A 331.90 V 6990.3 V The voltages and currents at the substation bus are 7200/0 [ Vsub abc ] = 7200/ – 120 V 7200/120 320. The impedance compensator settings for the three step regulators of Problem 7.8/ – 27. (1) Determine the equivalent line impedance per phase between the regulator and the load center.5 have been set as: 7.0/98. A powerﬂow study has been run and the voltages at the substation and the load center node are 7200/0 [ Vsub abc ] = 7200/ – 120 V 7200/120 6890.7 A 329.0 V X′ = 9. Specify these values in volts and in ohms. The voltage level of the regulators is set at 121 V and the bandwidth at 2 V. (2) The compensators on each regulator are to be set with the same R and X values.196 Distribution System Modeling and Analysis unbalanced load.6/ – 27. .9 The regulator potential transformer ratio is 7200120 and the current transformer ratio is 500:5.3 [ I abc ] = 395.
4/ – 156. unbalanced constant impedance load is located at the load center node.3849 0. The load center node for the regulators described in Problem 7.1580 + j0.5017 0.3465 + j1.3414 + j1. ZLb = 22 + j12 Ω.3375 + j1.1535 + j0.1560 + j0.1580 + j0.1535 + j0. The regulators are set on neutral. The phase impedance matrix of the line segment is 7.5.0348 Ω/mile A wyeconnected.5 [ I abc ] = 213. ZLc = 18 + j10 Ω The voltages at the substation are balanced threephase of 7200 V linetoneutral.0179 0.8/98.8 0.3849 0.1/ – 28. There are no lateral taps between the substation and the load center.5017 0.Regulation of Voltages 7.4236 0. the taps on the regulators have been automatically set by the compensator circuit to: Tapa = +8 Tapb = +11 Tapc = +6 The load reduces so that the voltages and currents at the substation bus are 7200/0 [ Vsub abc ] = 7200/ – 120 V 7200/120 177. (1) Determine the linetoneutral voltages at the load center.3 Determine the new ﬁnal tap settings for each regulator. The load impedances are ZLa