strategic and robust deployment of. synchronized phasor measurement units with. restricted channel...
DESCRIPTION
sepTRANSCRIPT
Strategic and Robust Deployment of
Synchronized Phasor Measurement Units with
Restricted Channel Capacity
A Thesis Presented
by
Mert Korkalı
to
The Department of Electrical and Computer Engineering
in partial fulfillment of the requirements
for the degree of
Master of Science
in
Electrical and Computer Engineering
Northeastern University
Boston, Massachusetts
December 2010
c© copyright by Mert Korkalı 2010
All Rights Reserved
NORTHEASTERN UNIVERSITY
Graduate School of Engineering
Thesis Title: Strategic and Robust Deployment of Synchronized Phasor
Measurement Units with Restricted Channel Capacity
Author: Mert Korkalı
Department: Electrical and Computer Engineering
Approved for Thesis Requirements of the Master of Science Degree
Thesis Adviser: Professor Ali Abur Date
Thesis Reader: Professor Hanoch Lev-Ari Date
Thesis Reader: Professor Aleksandar M. Stankovic Date
Department Chair: Professor Ali Abur Date
Graduate School Notified of Acceptance:
Director of the Graduate School: Sara Wadia-Fascetti Date
Northeastern University
Abstract
Department of Electrical and Computer Engineering
Master of Science in Electrical and Computer Engineering
by Mert Korkalı
iv
Synchronized phasor measurements are changing the way power systems are moni-
tored and operated. Their efficient incorporation into various applications which are
executed in energy management control centers requires strategic placement of these
devices. Earlier studies which consider placement of synchronized phasor measure-
ment units (PMUs) to be used for state estimation assume that these devices will
have unlimited channel capacities to record as many phase voltages and currents as
needed. What differentiates this study from those already reported in the literature
is the fact that it accounts for the number of available channels for the chosen type
of PMU since all existing PMUs come with a limited number of channels and their
costs vary accordingly. This is shown to be a critical factor in strategic placement
of these devices. In this study, a revised formulation of the placement problem and
its associated solution algorithm will be presented. Examples will be used to illus-
trate the impact of having limited number of channels on the location and number of
required PMUs to make the system observable. Developed methods will take into ac-
count existing injection measurements, in particular the virtual measurements such
as zero-injections that are available at no cost at electrically passive buses.
Moreover, despite the advances in related technologies, it is almost impossible
to guarantee occasional device or communication failure that will lead to loss of
data to be received from a given PMU. This work is also aimed to illustrate how the
measurement design can be made reliable against such events while maintaining the
cost of PMU installations at a minimum by using strategically placed PMUs with
the proper number of channels.
Furthermore, it is also demonstrated that depending upon the topology of the
network, there will be an upper limit on the number of channels for the PMUs beyond
which installation costs will not be reduced any further. Accordingly, numerical
results of applying the developed optimization method to power systems with varying
sizes and topologies will be presented to illustrate the typical numbers of PMUs and
their channel capacities that are required for optimal performance. The results of
this work will be more useful as the number of PMU installations increases to levels
that will make the system fully observable based solely on PMUs with different
number of channel capacities.
Acknowledgments
I am deeply indebted to my mentor and research advisor, Professor Ali Abur, for an
incomparably rewarding educational and personal experience. I have been indescrib-
ably enlightened and inspired by his patient teaching and vast technical expertise.
His constant support, gentle guidance, and warm encouragement gave a positive
impetus to the successful completion of my thesis. His inspiring ability to treat
problems from a new perspective integrated with many hours of constructive discus-
sions were the raison d’etres of the progressive improvements in this thesis. Indeed,
being a research assistant to him will definitely fortify my competence to stay in the
forefront of my current research area.
In the meantime, this is an opportunity to thank some of the people who have
shaped my academic personality prior to my arrival to Northeastern. Special thanks
go to my undergraduate advisor, Professor Bulent Bilir, for his invaluable patience
and incessant encouragement throughout my studies at Bahcesehir University and
Professor H. Fatih Ugurdag for his irreplaceable endeavor that undoubtedly paved
the way for my being a graduate student in the United States. I am more than grate-
ful for experiencing a mentor–younger friend relationship as well as an instructor–
student relationship with them.
I would like to express my heartfelt gratitude to Professor Hanoch Lev-Ari and
Professor Aleksandar M. Stankovic not only for serving as my thesis committee
members, but also giving me an inspiration and immense knowledge of their areas
of expertise during my graduate studies.
My deepest gratitude and love are reserved for my parents, who made me who I
am and whose love embraces me everywhere regardless of the wide distance between
us.
v
Contents
Abstract iii
Acknowledgments v
List of Figures viii
List of Tables ix
1 Introduction 1
1.1 Motivations for the Study . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Contributions of the Thesis . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Thesis Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2 Optimal PMU Placement for State Estimation 5
2.1 Historical Overview of Phasor Measurement Units . . . . . . . . . . . 5
2.2 Applications of Synchronized Phasor Measurements in Power Systems 7
2.3 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3 Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 14
3.1 Statement of the PMU Placement Problem . . . . . . . . . . . . . . . 14
3.2 Proposed Formulation for the PMU Placement Problem . . . . . . . . 16
3.3 Modeling of Zero-Injection Buses . . . . . . . . . . . . . . . . . . . . 21
3.3.1 Case 1—Arbitrary Selection of Neighbor Buses . . . . . . . . . 21
3.3.2 Case 2—Selection of Neighbor Buses which Have MinimumNumber of Neighbors . . . . . . . . . . . . . . . . . . . . . . . 21
3.3.3 Case 3—Selection of Neighbor Buses which Have MaximumNumber of Neighbors . . . . . . . . . . . . . . . . . . . . . . . 22
3.3.4 Case 4—Modeling of Zero-Injection Buses as Linear Constraints 23
3.4 Effect of Network Sparsity on PMU Placement . . . . . . . . . . . . . 25
3.5 Optimal Placement Accounting for Single PMU Loss . . . . . . . . . 28
4 Simulation Results 31
4.1 Conventional PMU Placement with Fixed Channel Capacity . . . . . 31
vi
Contents vii
4.2 Impact of Network Sparsity on Strategic Placement of PMUs . . . . . 32
4.3 Reliable Measurement Design Against Loss of PMUs . . . . . . . . . 34
4.4 Illustration of Unified PMU Placement Schemes . . . . . . . . . . . . 35
5 Concluding Remarks and Further Study 49
5.1 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
5.2 Further Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
A Functions and Scripts Used in the PMU Placement Algorithm 51
A.1 Read Network Parameters and Build the Single-Hop Connectivity Ma-trix A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
A.2 Find the Required Number of PMUs for Complete Network Observ-ability (Ignoring Zero-Injection Measurements) . . . . . . . . . . . . . 54
A.3 Find the Required Number of PMUs for Complete Network Observ-ability (Considering Zero-Injection Measurements) . . . . . . . . . . . 57
B IEEE Test Systems Data Used in the PMU Placement Algorithm 64
B.1 IEEE 14-Bus Test System Data . . . . . . . . . . . . . . . . . . . . . 65
B.2 IEEE 30-Bus Test System Data . . . . . . . . . . . . . . . . . . . . . 68
B.3 IEEE 57-Bus Test System Data . . . . . . . . . . . . . . . . . . . . . 72
B.4 IEEE 118-Bus Test System Data . . . . . . . . . . . . . . . . . . . . 79
Bibliography 94
List of Figures
3.1 Phasor measurements provided by PMU. . . . . . . . . . . . . . . . . 17
3.2 7-bus system for illustration. . . . . . . . . . . . . . . . . . . . . . . . 19
3.3 PMU placement for IEEE 14-bus system when the channel limit is 3. 20
3.4 Network diagram and measurement configuration for the IEEE 14-bussystem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.5 Configuration of 3 one-channel PMUs in 7-bus system. . . . . . . . . 25
3.6 Network diagram for the 7-bus system with two-hop neighborhoodtopology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.7 Reliable placement against single PMU loss for the IEEE 14-bus sys-tem when the channel limit is 3. . . . . . . . . . . . . . . . . . . . . . 30
3.8 Zero-injections and optimally placed 22 PMUs for the IEEE 57-bussystem assuming no channel limits. . . . . . . . . . . . . . . . . . . . 30
4.1 Optimally placed 7 PMUs for the IEEE 14-bus system when the chan-nel limit is 1 (ignoring the zero-injections). . . . . . . . . . . . . . . . 37
4.2 Optimally placed 15 PMUs for the IEEE 30-bus system when thechannel limit is 1 (ignoring the zero-injections). . . . . . . . . . . . . 38
4.3 Optimally placed 19 PMUs for the IEEE 57-bus system when thechannel limit is 2 (ignoring the zero-injections). . . . . . . . . . . . . 39
4.4 Optimally placed 41 PMUs for the IEEE 118-bus system when thechannel limit is 2 (ignoring the zero-injections). . . . . . . . . . . . . 40
4.5 Zero-injection and optimally placed 3 PMUs for the IEEE 14-bussystem when the channel limit is 4. . . . . . . . . . . . . . . . . . . . 41
4.6 Zero-injections and optimally placed 7 PMUs for the IEEE 30-bussystem when the channel limit is 4. . . . . . . . . . . . . . . . . . . . 42
4.7 Zero-injections and optimally placed 14 PMUs for the IEEE 57-bussystem when the channel limit is 3. . . . . . . . . . . . . . . . . . . . 43
4.8 Zero-injections and optimally placed 29 PMUs for the IEEE 118-bussystem when the channel limit is 5. . . . . . . . . . . . . . . . . . . . 44
4.9 Reliable placement against single PMU loss for the IEEE 14-bus sys-tem when the channel limit is 3 (considering the zero-injection). . . . 45
4.10 Reliable placement against single PMU loss for the IEEE 30-bus sys-tem when the channel limit is 3 (considering the zero-injections). . . . 46
4.11 Reliable placement against single PMU loss for the IEEE 57-bus sys-tem when the channel limit is 3 (considering the zero-injections). . . . 47
4.12 Reliable placement against single PMU loss for the IEEE 118-bussystem when the channel limit is 3 (considering the zero-injections). . 48
viii
List of Tables
4.1 Conventional PMU Placement without Zero-Injections for Miscella-neous Power Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.2 PMU Placement with Zero-Injections for IEEE Test Systems . . . . . 33
4.3 Conventional PMU Placement with 2-Hop Connectivity for Five IEEETest Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.4 Reliable Placement Against Loss of PMUs without Zero-Injections forMiscellaneous Power Systems . . . . . . . . . . . . . . . . . . . . . . 35
4.5 Reliable Placement Against Loss of PMUs with Zero-Injections forIEEE Test Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
B.1 System Information of Studied IEEE Test Systems . . . . . . . . . . 64
ix
To my beloved parents,
Selma and Hasan Korkalı,
with love and gratitude
x
“A good companion shortens the longest road.”
—Turkish Proverb
xi
Chapter 1
Introduction
Power systems have long been monitored based on measurements provided by remote
terminal units (RTU). These measurements typically include branch power flows, bus
power injections, and magnitude of bus voltages. A critical quantity of interest is the
phase difference between a given pair of bus voltage phasors in the system. Until
recently, this quantity was not easily measurable. Direct measurement of phase
angles of voltage and/or current phasors is now possible by phasor measurement
units (PMUs) thanks to the availability of the Global Positioning System (GPS)
that facilitates time synchronization of measured signals at geographically remote
locations. However, even when there are no phase angle measurements, phase angle
associated with each bus voltage phasor can be estimated along with its magnitude
provided that there are sufficient number of power flow and bus injections measured
with negligible time skew. This is accomplished by the help of a power system state
estimator.
Having synchronized phasor measurements for bus voltages and branch currents
in a given power system has a significant effect on the application functions in control
centers. In particular, the state estimation application will be affected in a rather
fundamental manner in that the problem formulation can transform from nonlinear
to linear if sufficiently large number of such phasors can be measured. Hence, there
is interest on the part of state estimator users with respect to the requirements of
such a transformation, namely the cost associated with installing the right number
1
Chapter 1. Introduction 2
and type of PMUs at strategic locations in order to drastically improve their state
estimators.
1.1 Motivations for the Study
Synchronized phasor measurement units are rapidly populating power systems as
their benefits become more and more evident for various power system applications.
One such function is the state estimation [1]. In fact, state estimation provides the
much needed real-time database for several application functions which facilitate
power system control and efficient operation. Given the limited resources, it is
practical to place these devices strategically in order to minimize the cost/benefit
ratio.
Optimal placement methods to account for contingencies, loss of measurements
as well as existing conventional measurements and zero-injections have already been
presented in previous publications. A common assumption in all of these studies
is that each PMU can measure unlimited number of voltages and currents. This
assumption allows the PMU placement problem to be formulated in a straightforward
manner since placement of a PMU at a given bus automatically ensures availability
of phasors at all of its neighboring buses. This is true as long as the network
connectivity and associated branch parameters are perfectly known, which is also
commonly assumed by all state estimators. Available PMUs have limited number of
channels which are used to sample voltage or current signals. These sampled signals
are processed to generate a positive-sequence phasor voltage and current from the
three-phase voltages and currents. Hence, channel capacities of PMUs may play an
important role in their strategic placement for maximum coverage.
1.2 Contributions of the Thesis
The main contribution of this thesis is to recognize the effect of channel capacity
of a given type of PMU on their optimal placement for network observability and
Chapter 1. Introduction 3
to develop an optimal solution to the PMU placement problem given a specified
number of available channels for the candidate PMUs. As in earlier studies, problem
formulation accounts for bus injections which may be known due to measurements
or due to their zero values at passive buses.
In formulating the problem it is realized that for a given number of channel ca-
pacity there will be a finite combination of possible assignments of incident branches
to a given PMU placed at a bus. Hence, the choices will increase with the number of
incident branches for a given bus, but will remain bounded irrespective of the overall
system size, thanks to the sparse interconnection of power system buses.
1.3 Thesis Outline
This thesis comprises five chapters. It is organized as follows. In the current chapter,
the motivations for the research problem and our contributions to PMU placement
problem are discussed.
In the succeeding chapter (Chapter 2), we first present the general background
information about synchronized phasor measurement units and review the relevant
literature to the existing PMU placement strategies flourished in the field of power
systems state estimation.
Chapter 3 delineates the method of determining optimal number and locations
of PMUs, so that the system state of an entire power system will be observable. In
doing so, the technique to be introduced is a numerical procedure where the problem
is formulated as an integer linear programming (ILP) problem. In addition, different
cases are suggested for the modeling of zero-injection buses.
Chapter 4 is devoted to the simulation results of our optimization model includ-
ing various conditions where zero-injections are considered and ignored, the sparsity
of the studied networks is reduced fictitiously; and the reliability is maintained under
a single PMU failure. Several case studies are conducted to evaluate the algorithm’s
performance and effectiveness.
Chapter 1. Introduction 4
Finally, Chapter 5 concludes this thesis with the discussion on the benefits of
the proposed formulation for optimal PMU placement along with its usability in
the existing power systems. Also, we express our ideas about what can be done for
further study.
Chapter 2
Optimal PMU Placement for State
Estimation
This chapter is mainly devoted to the history of the evolution of synchronized phasor
measurement units used for the purpose of state estimation and provides an overview
of the applications and miscellaneous techniques that have been introduced so far in
the power systems literature regarding PMU placement.
2.1 Historical Overview of Phasor Measurement
Units
Synchronized phasor measurements are considered as a promising measurement tool
for electric power systems. They supply positive-sequence voltage and current mea-
surements synchronized to within a microsecond thanks to the availability of Global
Positioning System (GPS) and the sampled data processing algorithms designed
for computer relaying applications. Apart from the positive-sequence voltage and
current measurements, these systems are able to quantify both local frequency and
frequency rate-of-change. Moreover, they can be altered according to the several
needs of users in order to extract the data relating to zero- and negative-sequence
quantities, harmonics, as well as individual phase voltages and currents. Currently,
5
Chapter 2. Optimal PMU Placement for State Estimation 6
there are 24 commercial PMU manufacturers. In this respect, standards established
by the IEEE Power System Relaying Committee have enabled the data sharing capa-
bility among distant units from different manufacturers. Considering that blackouts
are occurring overwhelmingly on the existing power networks, widespread allocation
of PMUs has gained tremendous interest. In particular, positive-sequence measure-
ments provide the accessibility to the power system state at any instant. Various
applications of synchronized phasor measurements have been presented in the liter-
ature, and more applications will certainly be developed in coming years [2].
The modern era of phasor measurement technology has its origin in research
conducted on computer relaying of transmission lines. Early study on transmis-
sion line relaying with relays based on microprocessors unveiled that the available
computer power in 1970s was not substantial enough to manage the computations
required to execute all the line relaying operations.
Positive-sequence voltages of a network constitute the state vector of a power
system, and it is of fundamental importance in all of power system analyses. The
first paper to identify the importance of positive-sequence voltage and current pha-
sor measurements, and some of the uses of these measurements, was published in
1983 [3], and this work can be viewed as the starting point of modern synchronized
phasor measurement technology. The GPS was being entirely installed around these
years. Later on, it became evident that this system provided the most effective way of
synchronizing power system measurements over long distances. The first prototypes
of the existing PMUs using GPS were developed at Virginia Tech in 1980s. These
prototype PMUs built at Virginia Tech were placed at certain substations of the
Bonneville Power Administration (BPA), the New York Power Authority (NYPA),
and the American Electric Power Service Corporation (AEPSC) [2]. For the mo-
ment, a number of commercial manufacturers offer PMUs and placement of PMUs
on the modern power systems is being carried out profoundly in many countries.
Along with the development of PMUs, there is substantial amount of continuing
research on the applications of the measurements provided by the PMUs. In this
respect, the recent advancement in synchronized phasor measurements is reaching
Chapter 2. Optimal PMU Placement for State Estimation 7
towards improvements in its maturity, and a majority of modern power systems
across the globe are continually placing wide-area measurement systems made up of
the phasor measurement units.
2.2 Applications of Synchronized Phasor Measure-
ments in Power Systems
Advances and applications in research studies of synchronized phasor measurement
units have been presented by several recent papers. Reference [4] introduces a pos-
sible approach towards formulating a standard that would facilitate interoperability
of PMUs under transient conditions. In [5], the authors propose a wide-area network
of phasor measurement units as a means for monitoring and control of voltage sta-
bility. A new technique for wide-area protection utilizing PMU is suggested in [6].
In particular, the proposed protection scheme is dependent upon the comparison
of the positive-sequence voltage magnitudes for certain areas along with the differ-
ences of positive-sequence current phase angles for each line between two areas in
the power system. The authors of [7] show that the realization of optimal measure-
ment designs can be achieved in order to determine the types and locations of few
extra measurements that will considerably improve the capability of topology error
processing. To accomplish this task, emerging PMUs are suggested for use in addi-
tion to the conventional power flow and injection measurements. In order to help
prevent a large-scale blackout, the authors of reference [8] present an online voltage
security assessment scheme making use of synchronized phasor measurements as well
as decision trees which are periodically updated.
With the latest progress in smart grid technology, the use of phasor measure-
ment units has definitely drawn substantial interest in order to render power system
reliability within transmission and distribution infrastructure. Hence, the utilization
of wide-area monitoring systems (WAMS) using synchrophasor measurements has
gained momentum achieving improved system monitoring, control, and protection.
Chapter 2. Optimal PMU Placement for State Estimation 8
In one of recent studies [9], the authors discuss the detailed architecture and the re-
cent implementations and applications of a wide-area frequency monitoring network
(FNET). In [10], it is shown that single line outages can be detected by using pha-
sor angle measurements data provided by PMUs even if there is extremely limited
coverage. In a two-paper set [11, 12], the authors present a PMU-based technique
for fault detection/location as well as multifunction transmission line protection for
both arcing and permanent faults by processing the synchronized voltage and current
phasors. In order to avoid reclosure on a permanent fault, arcing fault discrimination
technique is proposed via processing the synchronized harmonic voltage and current
phasors.
In [13], energy function analysis has been adapted using phasor data in order
to monitor the dynamic security of power transfer paths. Utilizing the phasor data
provided by a system of well-located PMUs, the transfer paths and their associated
parameters are identified and the transfer path reactances and equivalent inertias
are estimated by using the power-angle curves and the oscillation frequencies. A
wide-area identification of long-term voltage instability from the bus voltage pha-
sors provided by synchronized phasor measurements is devised in [14, 15]. A gener-
alized fault section selector, as well as fault locator, is proposed by Liu et al. [16] for
multiterminal transmission lines based on synchronized phasor measurement units.
In [17], the authors investigate the feasibility of estimating the rotor angle of syn-
chronous generators from the measurements of field voltage of the generator and
terminal voltage measurements acquired from PMUs.
2.3 Related Work
In order to estimate the system state, power system state estimator makes use of the
set of available measurements. Given a set of measurements and their corresponding
locations, the network observability analysis will determine if a unique estimate
can be found for the system state. This analysis are carried out offline during
the initial phase of a state estimator installation in order to check the sufficiency
of the existing measurement configuration. If the system is not found observable,
Chapter 2. Optimal PMU Placement for State Estimation 9
then additional meters may have to be installed at certain locations. Observability
analysis is executed online prior to performing the state estimator. It ensures that a
state estimate can be obtained using the set of measurements at the last measurement
scan. Telecommunication errors, telemetry failures, or changes in topology may at
times result in the cases where the state of the whole system cannot be estimated.
Network observability test allows detection of such cases right before the execution
of the state estimator. Observability of a given network is determined by the type
and location of the available measurements as well as by the topology of the network.
Therefore, the analysis of network observability exploits the graph theory since it has
connection with networks, their respective equations and solutions. Also, the system
is said to be topologically observable if the meters are placed such that there exists at
least one spanning measurement tree of full rank [18]. On the other hand, installing
a PMU at every bus in a wide-area interconnected network is neither reasonable
nor prudent. For that reason, the optimal PMU placement problem deals with
determining the minimum number of PMUs to achieve full network observability.
Intrinsically, the optimal PMU placement problem is shown to be NP-complete
with a solution space having 2N possible combinations for an N -bus electric net-
work [19]. In this respect, it is regarded as a combinatorial optimization problem
and a considerable amount of work has been done by several researchers, accordingly.
These approaches are broadly classified into two main categories: the metaheuristic
techniques and conventional deterministic optimization methods. As formulations
based on metaheuristics (e.g., simulated annealing (SA), genetic algorithm (GA),
Tabu search (TS), etc.) do not involve derivative of cost functions and the variables
of the meter placement problem are discrete; they have been extensively used in
dealing with discrete variables when solving the optimal PMU placement problem.
The utilization and development of PMUs are first introduced in [20] and [21].
An algorithm for finding the minimum number of PMUs required for power system
state estimation is developed in [22] and [23] in which the simulated annealing (SA)
method and the graph theory are utilized in formulating and solving the problem.
Nuqui and Phadke [24], [25] utilize a simulated annealing (SA) technique in their
graph-theoretic approach to determine the optimal PMU locations. In their work, a
Chapter 2. Optimal PMU Placement for State Estimation 10
novel concept of “depth-of-unobservability” is presented and how this has an effect
on the PMU placement is also shown. An optimal placement method founded on
nondominated sorting genetic algorithm (NSGA) is proposed by Milosevic and Be-
govic [26]. The algorithm unites the graph theory and a simple GA to estimate each
optimal solution of the objective function. The best tradeoff between the competing
objectives is then searched by customized NSGA. This method is limited by the size
of the problem as it requires more complex computations. Another GA-based proce-
dure for the placement problem is presented by Marın et al. [19]. In this letter, the
relationship between the number of current phasors that must be measured on each
PMU and the required number of PMUs is also sought by the authors. Cho, Shin,
and Hyun [27] propose three approaches aiming at alleviating computational burden
of the optimal placement problem. First, SA method is modified in setting the ini-
tial temperature and cooling procedure. Second, direct combination (DC) method
is suggested using a simple yet effective heuristic rule to identify the most effective
sets in the observability sense. At the end, TS method is utilized to diminish the
searching spaces effectively. A novel technique established upon TS and augmented
incidence matrix is introduced by Peng, Sun, and Wang [28]. Aminifar et al. [29]
investigate the applicability of immunity genetic algorithm (IGA) for minimal PMU
placement problem. Chakrabarti et al. [30] propose a methodology based on binary
particle swarm optimization (BPSO). In this study, the objectives of the optimization
problem lie at the intersection of minimization of the required number of PMUs and
maximization of the measurement redundancy. Analogously, Hajian et al. [31] use a
modified BPSO algorithm as an optimization tool for obtaining the minimal num-
ber of PMUs for complete system observability. Sadu, Kumar, and Kavasseri [32]
solve the placement problem by particle swarm optimization (PSO) algorithm, and
the idea of introducing randomness in selecting the buses for the PMU placement
is suggested by the authors. Chakrabarti and Kyriakides [33] propose binary search
algorithm as a technique for solving the problem.
In addition to the metaheuristic methods, several conventional deterministic
techniques are applied to the optimal PMU placement problem [34–46]. In [34]
and [35], the algorithm for optimal placement of PMUs is developed using integer
Chapter 2. Optimal PMU Placement for State Estimation 11
programming (IP) established upon the network observability and installation costs
of PMUs. Gou [36] makes a simplification in the placement algorithm by using
ILP and considering both the presence and absence of the conventional flow and
injection measurements. In his another simultaneously published work [37], the
author extends the formerly developed model and generalizes the ILP formulation
to satisfy various needs by integrating redundant PMU placement, full observability
and incomplete observability cases. Dua et al. [42] propose another formulation using
ILP. Integer quadratic programming (IQP) model is proposed as a solution method
in [40] and [41].
Among the published techniques, a certain number of those take into account
the power system contingencies broadly associated with the line outages and/or
measurement losses [33, 38–44, 47–51]. The integration of such contingencies in the
placement problem would certainly contribute to the reliable measurement designs.
A sequential meter addition/elimination process based on the measurement sensi-
tivities has been presented by Park et al. [47]. Abur and Magnago [38] propose an
LP-based method in which a number of additional measurements are then systemat-
ically added to ensure full observability under the loss of any single network branch.
The same authors propose a numerical algorithm based on the measurement Jaco-
bian and sparse triangular factorization to optimally upgrade the measurements and
yield a configuration which can remain robust against loss of single measurement
and single branch outage without sacrificing network observability [48]. Xu, Yoon,
and Abur [39] address a binary integer programming method taking into account
the loss of a single PMU in order to lessen the vulnerability of state estimation to
PMU breakdowns. The identical efforts to obtain a reliable measurement system
based on numerical observability are made by Rakpenthai et al. [49]. The authors
utilize the minimum condition number of the normalized measurement matrix as a
criterion. Then, the sequential addition and elimination methods are employed to
determine the essential measurements and to identify the redundancy measurements
under the contingency, respectively. Later work by Chakrabarti and Kyriakides [33]
propose a strategy utilizing a binary search algorithm to find the minimum number
of PMUs for full topological observability under normal operating conditions, as well
Chapter 2. Optimal PMU Placement for State Estimation 12
as single branch outages. In the paper, the search process is said to be exhaustive;
as a result, they aspire to overcome the restrictions of the conventional optimization
methods such as the integer programming and the uncertainties of the evolutionary
programming techniques such as the genetic algorithm. In their another collabora-
tive works [40], [41], they propose an IQP approach to minimize the total number
of PMUs required to maintain the complete observability of the system for normal
operating conditions and under the outage of a single PMU or a transmission line.
Also, they aim to provide the maximization of the measurement redundancy at all
system buses. Dua et al. [42] devise a procedure for optimal multistage scheduling
of PMU placement phased over multiple time horizons. Furthermore, they suggest
zero-injection constraints be modeled as linear constraints in an ILP framework. The
two indices, Bus Observability Index (BOI) and System Observability Redundancy
Index (SORI), are utilized to rank the multiple solutions obtained via minimum
PMU placement problem. In their generic PMU placement formulation, the authors
offer some modifications to deal with the issues of PMU loss and communication
line outage. Likewise, Abbasy and Ismail [43] study the impact of single PMU loss
or multiple PMU losses on the decision strategy of the PMU placement problem.
In [50] and [51], the authors come up with the so-called branch PMUs which are
designed to monitor a single branch by measuring the associated current and termi-
nal voltage phasors. Further, they also address the robustness of the measurement
design by considering not only the cases of PMU loss or failure, but contingencies
stemming from line or transformer outages. More recently, Aminifar et al. [44] of-
fer a practical ILP-based model taking account of several contingency conditions
involving communication constraints, loss of measurements, and line outages.
Chen and Abur [45], [46] propose an IP-based solution that leads to the smallest
number of strategically located PMUs eliminating the measurement criticality in the
system. In these papers, it is shown that the bad data detection and identification
capability of a system can be enhanced greatly with few additional PMUs.
A fault location scheme for transmission networks using PMUs is developed and
the idea of fault location observability is presented by Lien et al. [52]. A method
for placing minimum number of PMUs to locate any fault in a power system is
Chapter 2. Optimal PMU Placement for State Estimation 13
proposed by Pokharel and Brahma [53]. The method is formulated on the basis
of ILP structure which is introduced in [34]. Mahmoodianfard et al. [54] utilize
a scheme based on decision trees to find an optimum PMU placement for voltage
security assessment. In the work by Zhou et al. [55], a virtual data preprocessing
technique and a matrix reduction algorithm are introduced to show the effectiveness
in reducing the computational effort for determining the optimal placement set.
The performance validation for the proposed algorithm is proven by applying the
method of Lagrangian relaxation to calculate the lower bound of the minimal number
of PMUs.
Chapter 3
Strategic Placement of Phasor
Measurement Units with Optimal
Number of Channels
3.1 Statement of the PMU Placement Problem
Power systems are assumed to operate in pseudo-steady-state due to the slow dy-
namics of system loads and generation. Hence, measurements of various quantities
such as power flows, voltages, and currents at various substations are used to ap-
proximately determine the operating conditions of this pseudo-steady-state of the
system. These measurements typically have time skew which may be in the order
of minutes and therefore an estimate of the system state obtained based on these
measurements will only be an approximation. Time skew among collected mea-
surements is eliminated when the measurements are replaced by those provided by
synchronized phasor measurement units (PMU). These devices are quickly becoming
the preferred metering choice at bulk power transmission substations. They provide
magnitude and phase angle measurements of bus voltages and branch currents for
the positive-sequence components of three-phase signals. These measurements are
time-stamped and phase angles are defined with respect to a common reference
14
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 15
determined by the global positioning satellite system. PMU measurements can im-
prove the performance and capabilities of various network applications due to their
unique features. One of the applications which will benefit most from availability
of PMU measurements is the state estimator. In fact, if the entire network can be
made fully observable by just using PMU measurements, then the state estimation
problem will become linear and will be solved directly without requiring the use of
iterative methods. If these devices can be strategically placed at proper substations
in sufficient numbers, then such a transformation will be possible, making a very
significant improvement in the performance of existing state estimators, essentially
eliminating the issues of divergence due to numerical problems.
All of the aforementioned studies in Section 2.3 consider the PMU placement
problem in such a way that these devices are assumed to have unlimited number of
channels. Hence, placing them at a given bus ensures that in addition to the phasor
voltage at that bus, phasor voltages of all its immediate neighbors will be available
due to the monitored phasor currents along all the branches incident to that bus.
In practice, every PMU comes with a channel limit and therefore a more realistic
placement of the PMUs should take into account their varying channel capacities.
The authors of [56] attempt to introduce the limit on the number of measurements
each PMU can make and a modified PageRank placement algorithm is utilized in the
importance modeling of the network nodes. However, the optimal numbers required
to make the studied test systems observable are not provided explicitly. Rather, the
effect of the number of measurements on the nodes to fully observe the network is
given as a criterion for the usefulness of the method.
In this study, the PMU placement problem is revisited with the aim of relaxing
the abovementioned assumption based on the fact that the bus voltage phasor and
all current phasors along branches connected to that bus are available. The problem
is reformulated where the number of channels for each PMU can be changed and
the problem can be solved repeatedly to find the optimal locations. Furthermore,
the formulation takes into account any existing injection measurements, in partic-
ular those virtual measurements provided by the zero-injections of passive buses.
Afterwards, the formulation is extended to account for loss of PMUs so that the
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 16
final PMU measurement design remains robust against loss of a PMU due to device
or communication link failures. Another contribution of this study is to recognize
the effect of channel capacity of a given type of PMU on their optimal placement
for network observability and to develop an optimal solution to the PMU placement
problem given a specified number of available channels for the candidate PMUs.
The approach taken in this study is one of exhaustive search among all possible
combinations at a given bus for a given limit on the number of available channels.
In formulating the problem it is realized that for a given number of channel capacity
there will be a finite combination of possible assignments of incident branches to
a given PMU placed at a bus. Hence, the choices will increase with the number
of incident branches for a given bus, but will remain bounded irrespective of the
overall system size, thanks to the sparse interconnection of power system buses.
The developed measurement placement procedure will be outlined and illustrated
by examples in the succeeding sections. Precisely, the main goal is to allow optimal
placement of PMUs which may have limited number of channels.
3.2 Proposed Formulation for the PMU Place-
ment Problem
Formulation of optimal PMU placement problem for the case of varying channels will
be briefly reviewed first. Subsequently, the revision of this formulation to account
for a PMU loss will be described in Section 3.5.
Consider a PMU which has L channels and installed at bus k as shown in
Figure 3.1. Also assume that bus k is connected to Nk number of buses. Note that the
actual number of channels may be three times more since the phasor measurements
are usually the positive-sequence components derived from sampled waveforms of
three-phase signals. So, it is understood that the number of channels refers to
the number of positive-sequence phasor measurements that can be produced by the
considered PMU.
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 17
PMU ......
Bus 1
Bus 2
Bus Nk
V phasor
I phasors
Bus k
Figure 3.1: Phasor measurements provided by PMU.
If the number of channels, L, is larger than the number of neighbors Nk, then a
single PMU placed at the bus will provide phasor voltages at all its neighbor buses.
Otherwise, there will be rk combinations of possible channel assignments to branches
incident at bus k:
rk =
NkCL if L < Nk,
1 if Nk ≤ L.
where the number of possible combinations of L out of Nk branches is defined as:
NkCL =Nk!
(Nk − L)!L!(3.1)
Since a PMU is able to measure both the voltage phasor of the bus at which it is
installed and the current phasors of all the lines in the neighborhood of this bus,
the placement of PMUs becomes a problem with an objective of finding a minimal
set of PMUs such that a bus must be observed at least once by the solution set of
the PMUs. This leads us to define the binary connectivity matrix H consisting of
all possible combinations at a given bus for a given limit on the number of available
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 18
channels such that each bus k will have rk rows, each row containing (L+1) nonzeros
for the bus itself and its neighbor buses. However, when Nk < L, i.e., the number of
branches incident to bus k is less than the channel limit of the PMUs, the associated
row needs to be kept unchanged. The channel limit constraints can thus be imposed
so that a PMU placed at a bus will observe its neighboring buses by selecting the
appropriate combination(s) of L.
Description of the procedure can be illustrated using a 7-bus system example
shown in Figure 3.2. Assuming a channel limit of 2 for the PMUs, the number of
rows for each bus is found to be r1 = r5 = r6 = r7 = 1, r2 = 6, and r3 = r4 = 3. In
fact, consider the buses connected to bus 2, which are buses 1, 3, 6, and 7; therefore,
the 2-combinations of this set will result in the pairs 1− 3, 1− 6, 1− 7, 3− 6, 3− 7,
and 6−7. In this sense, each row associated with bus 2 in the matrix H will include
a 1 corresponding to bus 2 and its neighbor pairs. Accordingly, let H be defined as
H =
1 1 0 0 0 0 0
1 1 1 0 0 0 0
1 1 0 0 0 1 0
1 1 0 0 0 0 1
0 1 1 0 0 1 0
0 1 1 0 0 0 1
0 1 0 0 0 1 1
0 1 1 1 0 0 0
0 1 1 0 0 1 0
0 0 1 1 0 1 0
0 0 1 1 1 0 0
0 0 1 1 0 0 1
0 0 0 1 1 0 1
0 0 0 1 1 0 0
0 1 1 0 0 1 0
0 1 0 1 0 0 1
Bus 1
Bus 2
Bus 3
Bus 4
Bus 5
Bus 6
Bus 7
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 19
Using this approach, the relevant PMU placement problem can be formulated as
follows:
7
6
1 2 3 4 5
Figure 3.2: 7-bus system for illustration.
Minimizen∑i=1
wixi
Subject to HTX ≥ U
X = [x1 x2 · · · xn]T
xi ∈ {0, 1}
(3.2)
where
n =N∑j=1
rj, U =
1
1...
1
N×1
, and w = [1 1 · · · 1]1×n.
Here, w represents the vector of installation cost of the PMUs, elements of which
are assumed to be uniform for simplicity, and xi are binary variables for the PMU
placement, and N denotes the number of buses in the system. In the above matrix
inequality, X represents a binary vector of all possible PMU channel assignments.
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 20
Correspondingly, the nonzero entries in X will point to the rows of associated buses,
voltage angles of which can be observed by these PMU measurements.
The solution of the aforecited PMU placement problem where the specified
channel limit for PMUs is 2, that is,
X =[
0 0 1 1 0 0 0 0 0 0 1 0 0 0 0 0]T
yields a total of 3 PMUs, two of which are to be located at bus 2, and one at bus
4, enabling the entire network observability. Indeed, one PMU installed at bus 2
measures the voltage phasor at bus 2 as well as the current phasors for branches
2− 1 and 2− 6; whereas, the other PMU at bus 2 measures branch current phasors
2 − 1 and 2 − 7. Furthermore, a PMU located at bus 4 will measure the voltage
phasor at bus 4 and current phasor on two of the incident branches 4− 3 and 4− 5.
Similarly, the location of PMUs for the IEEE 14-bus system with the channel
limit of 3 is illustrated in Figure 3.3. The optimal solution for this case points out
that 4 PMUs are required to achieve full network observability. A PMU located at
bus 2 will observe buses 1, 2, 3, and 5; the one at bus 6 will observe buses 6, 11, 12,
and 13; and so on.
Figure 3.3: PMU placement for IEEE 14-bus system when the channel limit is3.
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 21
3.3 Modeling of Zero-Injection Buses
The procedure of forming the constraint equations is straightforward for a system
without conventional measurements and/or zero-injections. However, when the zero-
injection pseudomeasurements are considered, the method of topology transforma-
tion introduced in [35] should be utilized in reforming the constraint equations.
Consider the IEEE 14-bus system shown in Figure 3.4, where the dot next to
bus 7 indicates that bus 7 is a zero-injection bus. It is obvious that if the phasor
voltages at any three out of the four buses 4, 7, 8, and 9 are known, then the fourth
one can be calculated using the power balance equations at bus 7 where the net
injection is known to be zero. Topology transformation method is then used to
merge the bus having the zero-injection measurement, with the buses connected to
that bus. In the selection of the candidate buses, three different strategies are first
considered and comparatively evaluated:
3.3.1 Case 1—Arbitrary Selection of Neighbor Buses
In this case, the bus which has the injection measurement is merged with one of its
arbitrarily chosen neighbors.
3.3.2 Case 2—Selection of Neighbor Buses which Have Min-
imum Number of Neighbors
In this case, the bus which has the injection measurement is merged with its neighbor
having the least number of neighbors.
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 22
3.3.3 Case 3—Selection of Neighbor Buses which Have Max-
imum Number of Neighbors
In this case, the bus which has the injection measurement is merged with its neighbor
having the most number of neighbors.
Figure 3.4: Network diagram and measurement configuration for the IEEE 14-bus system.
For Case 1, the appropriate algorithm is developed so that any of the neighbor buses
connected to bus 7, say bus 9, is chosen randomly. Since bus 8 has only one neighbor,
it is selected in the second case; whereas, bus 4 is to be selected in Case 3 since it has
five neighbors connected to it. The network will be updated after the merger of zero-
injection bus and the selected neighbor bus into a new single bus. In this context,
it is realized that the number of required PMUs can be reduced with the growing
system size, or almost equivalently, the increased availability of zero-injection buses.
Apart from these suggested cases involving intuition, we have developed a more
systematic methodology to incorporate the zero-injection buses into the problem
formulation by using linear constraints as shown in the following case.
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 23
3.3.4 Case 4—Modeling of Zero-Injection Buses as Linear
Constraints
In this case, zero-injection buses, which provide “free” measurements to the system,
are incorporated into the optimization formulation as done in [42]. Particularly, zero-
injections can be used to reduce the number of required PMUs by selectively allowing
some buses to be unreachable by the PMU measurements, as long as these buses
belong to a certain set of buses. This set is defined as the union of all zero-injection
buses and their immediate neighbors.
Let us define a set Ni as a set of buses including zero-injection bus i and all
its neighbors. Assuming “`” zero-injection buses to be present in the system, the
following set can be defined:
Φ =⋃i∈I
Ni = Ni1 ∪Ni2 ∪ · · · ∪ Ni`,
where I = {i1, i2, · · · , i`} designates the set of zero-injection buses.
Thus, the inequality constraints in (3.2) can be reestablished based on the above
considerations as follows: HT C
0 D
X
b
≥ R
c
(3.3)
where
Cjk =
−1 if j ∈ Ni and k ∈ {Φj},
0 otherwise.
Djk =
1 if j ∈ {Ii}i∈I and k ∈ {Φm}m∈Ni,
0 otherwise.
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 24
and X , R, b, and c are vectors of dimension n, N , |Φ|, and |I|, respectively, with
|{•}| denoting the cardinality of a set; and the matrices H, C, 0, and D in (3.3) are
of sizes n × N , N × |Φ|, |I| × n, and |I| × |Φ|, respectively. Also, ci are equal to
(|Ni| − 1) such that i ∈ I.
In building the matrices C and D, the set {Φε}ε∈E is defined such that the
elements of set Φ are indexed or labeled by means of set E . For the sake of con-
venience, the first row associated with the first partitioned matrix on the left-hand
side of (3.3) splits up into two parts in such a way that the zero-injection buses and
their neighbors are heaped together on the top of the new matrix. In this way, the
elements of the matrix D are clustered in the order of union set Φ. Moreover, the
elements of vector R on the right-hand side of (3.3) take on a “0” for the variables
related to zero-injection buses, and a “1” for those of the remaining buses.
For the sake of illustration, consider 7-bus system shown in Figure 3.5 where
single-channel PMUs are used and the dots designate the zero-injection buses present
in the system. Then, we can build the sets NBus 4 = {3, 4, 5, 7}, NBus 6 = {2, 3, 6},
and Φ = NBus 4 ∪ NBus 6 = {2, 3, 4, 5, 6, 7} based on the definitions above. In this
context, Ineq. (3.3) will take form of Ineq. (3.4) as shown in the following:
[HT
1
]6×16
−1
−1
−1
−1
−1
−1
[HT
2
]1×16 01×6
02×161 1 1 1
1 1 1
9×22
X
u2
u3
u4
u5
u6
u7
22×1
≥
0
0
0
0
0
0
1
3
2
9×1
(3.4)
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 25
The inequalities b3 + b4 + b5 + b7 ≥ 3 and b2 + b3 + b6 ≥ 2 with bi ∈ {0, 1} ensure that
there is at most one unobservable bus in these sets provided that their observability
is realized via the use of zero-injection buses in the corresponding sets. Indeed,
b6 = b7 = 0, meaning that bus 6 is not reached by a PMU by taking advantage of its
being zero-injection bus and bus 7 is observed via zero-injection bus 4. Figure 3.5
also illustrates the installed 3 PMUs along with the associated branches through
which the corresponding buses are observed.
7
6
1 2 3 4 5
PMU PMU PMU
Figure 3.5: Configuration of 3 one-channel PMUs in 7-bus system.
3.4 Effect of Network Sparsity on PMU Place-
ment
Buses in typical power networks are known to be sparsely connected. However, spar-
sity of systems may vary significantly depending on the geographic and operational
requirements. In order to study the effect of sparsity on the PMU placement, sys-
tems with increasingly dense bus interconnections are defined. This is accomplished
by systematically adding connections between second, third, etc. neighbors.
First, the binary adjacency matrix A is defined to describe the topology of the
network in which the ij-th entry is 1 if there is a connection between bus i and bus
j, and zero otherwise. All diagonal entries will also be 1 by default. The matrix A
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 26
for the simple network of Figure 3.2 will be given by:
A =
1 1 0 0 0 0 0
1 1 1 0 0 1 1
0 1 1 1 0 1 0
0 0 1 1 1 0 1
0 0 0 1 1 0 0
0 1 1 0 0 1 0
0 1 0 1 0 0 1
.
This matrix will be referred to as single-hop connectivity matrix since it represents
the connectivity of the buses with their immediate neighbors. Those buses that can
be reached from a given bus by a single hop over an existing branch will contain
nonzero entries. Similarly, matrices representing systems where buses have direct
connections to those buses that can be reached in two, three or more hops in the
original network can be easily generated by multiplying the single-hop connectivity
matrix by itself as many times as the number of hops [57]:
A(m+1) = A(m) ×A. (3.5)
Let A(1) = A. Then, A(m+1) is defined as follows:
A(m+1)(i, j) =
m + 1 if A(m)(i, j) = 0 and
A(m+1)(i, j) > 0;
A(m)(i, j) if A(m)(i, j) > 0.
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 27
Hence, the 2-hop connectivity matrix A(2) will be given by:
A(2) =
1 1 2 0 0 2 2
1 1 1 2 0 1 1
2 1 1 1 2 1 2
0 2 1 1 1 2 1
0 0 2 1 1 0 2
2 1 1 2 0 1 2
2 1 2 1 2 2 1
Bus 1
Bus 2
Bus 3
Bus 4
Bus 5
Bus 6
Bus 7
Bus 1 2 3 4 5 6 7.
which shows all the bus pairs that can reach each other within two hops. From the
1-hop connectivity matrix, we can obtain all (m+1)-hop connectivity matrices, thus
all possible (m + 1)-hop routes.
The binary version of the multi-hop connectivity matrix B(m+1) can be defined
by simply replacing nonzeros in the matrix A(m+1) by 1’s, as given below:
B(2) =
1 1 1 0 0 1 1
1 1 1 1 0 1 1
1 1 1 1 1 1 1
0 1 1 1 1 1 1
0 0 1 1 1 0 1
1 1 1 1 0 1 1
1 1 1 1 1 1 1
Bus 1
Bus 2
Bus 3
Bus 4
Bus 5
Bus 6
Bus 7
Bus 1 2 3 4 5 6 7.
The effect of reduced sparsity on the number and location of required PMUs
can then be studied by using this matrix instead of the original connectivity matrix.
The corresponding network connectivity is shown in Figure 3.6.
Again, from the newly formed network topology, we can easily build the matrix
H consisting of all potential set of combinations incident to each and every bus.
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 28
7
6
1 2 3 4 5
Figure 3.6: Network diagram for the 7-bus system with two-hop neighborhoodtopology.
When the PMU channel limit is assumed to be 2, the number of rows related to
each bus in the new matrix H becomes r1 = 6, r2 = r4 = r6 = 10, r3 = r7 = 15,
and r5 = 3. In this case, the solution of the PMU placement problem still yields 3
PMUs as the optimal number; however, the locations for these PMUs will now be
at buses 1, 4, and 7.
3.5 Optimal Placement Accounting for Single PMU
Loss
The initial studies consider a simplified model to represent the “reach” of individual
PMUs. Each PMU is assumed to provide the voltage phasor at the bus it is connected
and the current phasors at all of its neighbors as well. This assumption is relaxed in
[58] where the effects of channel capacity of a given type of PMU on their optimal
placement for network observability are taken into account.
In this section, the formulation is extended to account for loss of PMUs so that
the final PMU measurement design remains robust against loss of a PMU due to
device or communication link failures.
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 29
It is worth mentioning that PMUs are prone to failures like any other measuring
device even though they are highly reliable. Therefore, it is necessary to guard
against such unexpected failures of PMUs. In [34] and [39], the primary set of
PMUs is backed up by a secondary set which is determined based on the same
optimization formulation. In this study, the formulation of (3.2) is modified as done
in [41–43] to ensure that each bus will be observed by at least two PMUs. This
ascertains that a PMU loss will not lead to loss of observability. In the integer linear
programming framework, this can be easily achieved by multiplying U by 2, viz.,
U = [2 2 · · · 2]T1×N .
In this regard, the solution of the aforecited PMU placement problem where the
specified channel limit for PMUs is 2, will be given as:
X =[
0 0 1 1 0 0 0 0 0 1 1 0 1 0 0 0]T
which yields a total of 5 PMUs, two at buses 2 and 4 each and the remaining one
at bus 3, enabling the entire network to be observable even when the measurements
from any one of the PMUs are lost.
Similarly, the location of PMUs for the IEEE 14-bus system with the channel
limit of 3 is illustrated in Figure 3.7. The optimal solution for this case requires
placement of 9 PMUs to achieve full network observability under loss of a single
PMU.
Figure 3.8 illustrates the results of solving the above problem of PMU placement
for the IEEE 57-bus system assuming no channel limits for PMUs, accounting for
loss of a single PMU and making use of zero-injection measurements based on the
fourth case. The solution validates that each and every bus in the network is reached
at least twice either by PMUs or zero-injection measurements located at the bus or
its neighbors. Zero-injection buses are designated by dots next to the bus names in
Figure 3.8.
Chapter 3. Strategic Placement of Phasor Measurement Units with OptimalNumber of Channels 30
Figure 3.7: Reliable placement against single PMU loss for the IEEE 14-bussystem when the channel limit is 3.
Figure 3.8: Zero-injections and optimally placed 22 PMUs for the IEEE 57-bussystem assuming no channel limits.
Chapter 4
Simulation Results
4.1 Conventional PMU Placement with Fixed Chan-
nel Capacity
Simulations of the proposed method are carried out on various power systems. The
binary integer programming problem is solved using the TOMLAB /CPLEX Solver
Package [59]. The simulation results for the optimum number of PMUs with respect
to channel limits for the cases where zero-injections are ignored and considered, are
presented in Tables 4.1 and 4.2, respectively. As shown in Table 4.1, simulations are
carried out using five IEEE test systems as well as one larger-size system with 4520
buses. Also, Table 4.2 illustrates the results of simulations performed on the IEEE
14-, 30-, 57-, and 118-bus systems. For the case without zero-injection measurements,
the upper channel limit of the PMUs is determined by the maximum number of
branches incident to a certain bus in the corresponding test system. For those cases
where zero-injection measurements are considered, these limits are the number of the
branches incident to the fictitious bus, which is created by merging one or several
actual buses. Four ways to account for zero-injections are considered and compared
via simulations. The third case appears to have an advantage over the others in
particular when using multichannel PMUs. Second case may, however, be a better
choice when single-channel PMUs are to be used.
31
Chapter 4. Simulation Results 32
As evident from Table 4.2, zero-injections help reduce the number of required
PMUs. In Table 4.1, νmin/N is the ratio of the minimum number of required PMUs
to the number of buses in the system.
Table 4.1: Conventional PMU Placement without Zero-Injections for Miscella-neous Power Systems
Channel System under studyLimit
for theIEEE IEEE IEEE IEEE IEEE
4520-BusPMUs14-Bus 30-Bus 57-Bus 118-Bus 300-Bus
1 7 15 29 61 167 25432 5 11 19 41 105 16393 4 10 17 33 91 14904 4 10 17 32 89 14545 4 10 17 32 88 14366 10 17 32 88 14307 10 32 88 14298 32 87 14279 32 87 142210 87 142211 87 142212 142213 142214 142215 142216 1422
νmin/N 0.2857 0.3333 0.2982 0.2712 0.2900 0.3146
4.2 Impact of Network Sparsity on Strategic Place-
ment of PMUs
Considering the densely connected topologies, revised topologies with 2-hop connec-
tivity are obtained for five IEEE test systems. PMU placement problem is then
solved using these revised systems and the results are shown in Table 4.3. Among
the five power systems studied, it is observed that this ratio ranges from 27% to 33%
Chapter 4. Simulation Results 33
in Table 4.1, and from 10% to 21% in Table 4.3. In a similar vein, one can clearly
observe how the loss of sparsity leads to strategic placement of smaller number of
PMUs having larger number of channels.
Table 4.2: PMU Placement with Zero-Injections for IEEE Test Systems
ChannelIEEE Test Number of Limit Number of PMUsSystem Zero-Inj.’s for
the PMUs CASE 1 CASE 2 CASE 3 CASE 4
14-Bus 1
1 7 7 7 72 5 5 5 53 4 4 5 44 3 3 4 35 3 3 4 3
30-Bus 6
1 13 14 14 142 9 8 9 93 7 7 8 84 7 7 7 75 7 7 7 76 7 7 7 77 6 78 69 610 611 612 6
57-Bus 15
1 22 21 23 232 14 14 14 163 12 12 12 144 12 11 11 145 12 11 11 146 12 11 11 147 118 11
118-Bus 10
1 57 56 57 572 39 39 39 383 30 31 31 324 28 30 28 315 28 30 28 296 28 30 28 297 28 30 28 298 28 30 28 299 28 30 28 29
Chapter 4. Simulation Results 34
Table 4.3: Conventional PMU Placement with 2-Hop Connectivity for Five IEEETest Systems
IEEE Test SystemChannel Limitfor the PMUs
14-Bus 30-Bus 57-Bus 118-Bus 300-Bus
1 7 15 29 59 1502 5 10 19 40 1013 4 8 15 30 774 3 6 12 24 685 3 5 10 21 596 3 5 97 4 8
...
...17 3 ·20 823 4127 13
νmin/N 0.2143 0.1000 0.1404 0.1102 0.1367
4.3 Reliable Measurement Design Against Loss of
PMUs
PMU placement problem as formulated in Section 3.5 is solved for power systems of
different sizes. The solutions for the optimum number of PMUs for different channel
limits for the cases where zero-injections are ignored and considered, are presented
in Tables 4.4 and 4.5. Table 4.4 presents solutions that are obtained for five IEEE
test systems as well as for a 4520-bus utility system when zero-injections are ignored.
When the zero-injections are taken into account, results change significantly as shown
in Table 4.5 for four IEEE test systems. Once again, for the cases with and without
zero-injection measurements, the upper channel limit of the PMUs is determined
by the maximum number of incident branches to a bus in the corresponding test
system. Among the six power systems studied, it is observed that this ratio ranges
from 58% to 69%. As evident from Table 4.5, zero-injections help reduce the number
of PMUs required for complete network observability while maintaining robustness
against single PMU failure.
Chapter 4. Simulation Results 35
Table 4.4: Reliable Placement Against Loss of PMUs without Zero-Injectionsfor Miscellaneous Power Systems
Channel System under studyLimit
for theIEEE IEEE IEEE IEEE IEEE
4520-BusPMUs14-Bus 30-Bus 57-Bus 118-Bus 300-Bus
1 14 30 57 121 332 50802 10 22 38 82 219 34393 9 20 34 68 189 31604 9 20 33 68 190 31255 9 21 33 68 190 31356 21 33 68 193 31517 21 68 193 31628 68 195 32059 68 195 322410 195 323211 202 323512 326113 326514 326515 326516 3283
νmin/N 0.6429 0.6667 0.5789 0.5763 0.6300 0.6914
4.4 Illustration of Unified PMU Placement Schemes
In order to provide a more comprehensive picture of our overall methodology and
draw attention to the effect of channel limits on PMU placement accounting for
various combinations of abovementioned criteria, we have illustrated a number of
PMU placement strategies as shown in Figures 4.1–4.12. In cases where the zero-
injection measurements are considered, we have bounded our simulations merely by
the fourth case since it allows for a more systematic treatment of zero-injections.
Chapter 4. Simulation Results 36
Table 4.5: Reliable Placement Against Loss of PMUs with Zero-Injections forIEEE Test Systems
ChannelIEEE Test Number of Limit Number of PMUsSystem Zero-Inj.’s for
the PMUs CASE 1 CASE 2 CASE 3 CASE 4
14-Bus 1
1 13 13 13 122 9 9 9 83 7 7 8 74 8 7 8 75 8 7 7 7
30-Bus 6
1 26 28 28 212 18 16 19 143 15 14 15 134 14 14 14 135 15 15 14 136 16 15 14 137 13 138 139 1310 1311 1312 17
57-Bus 15
1 44 42 46 342 28 28 29 233 24 24 24 224 24 23 24 225 24 23 24 226 24 23 23 227 238 23
118-Bus 10
1 111 110 112 1032 75 76 77 693 62 63 62 584 61 63 61 585 61 63 60 586 62 63 63 587 62 63 63 598 62 63 63 599 62 63 63 59
Chapter 4. Simulation Results 37
Figure 4.1: Optimally placed 7 PMUs for the IEEE 14-bus system when thechannel limit is 1 (ignoring the zero-injections).
Chapter 4. Simulation Results 38
Figure 4.2: Optimally placed 15 PMUs for the IEEE 30-bus system when thechannel limit is 1 (ignoring the zero-injections).
Chapter 4. Simulation Results 39
Figure 4.3: Optimally placed 19 PMUs for the IEEE 57-bus system when thechannel limit is 2 (ignoring the zero-injections).
Chapter 4. Simulation Results 40
Figure4.4:
Op
tim
ally
pla
ced
41P
MU
sfo
rth
eIE
EE
118-
bu
ssy
stem
wh
enth
ech
an
nel
lim
itis
2(i
gn
ori
ng
the
zero
-in
ject
ion
s).
Chapter 4. Simulation Results 41
Figure 4.5: Zero-injection and optimally placed 3 PMUs for the IEEE 14-bussystem when the channel limit is 4.
Chapter 4. Simulation Results 42
Figure 4.6: Zero-injections and optimally placed 7 PMUs for the IEEE 30-bussystem when the channel limit is 4.
Chapter 4. Simulation Results 43
Figure 4.7: Zero-injections and optimally placed 14 PMUs for the IEEE 57-bussystem when the channel limit is 3.
Chapter 4. Simulation Results 44
Figure4.8:
Zer
o-i
nje
ctio
ns
and
opti
mal
lyp
lace
d29
PM
Us
for
the
IEE
E118
-bu
ssy
stem
wh
enth
ech
an
nel
lim
itis
5.
Chapter 4. Simulation Results 45
Figure 4.9: Reliable placement against single PMU loss for the IEEE 14-bussystem when the channel limit is 3 (considering the zero-injection).
Chapter 4. Simulation Results 46
Figure 4.10: Reliable placement against single PMU loss for the IEEE 30-bussystem when the channel limit is 3 (considering the zero-injections).
Chapter 4. Simulation Results 47
Figure 4.11: Reliable placement against single PMU loss for the IEEE 57-bussystem when the channel limit is 3 (considering the zero-injections).
Chapter 4. Simulation Results 48
Figure
4.12:
Rel
iab
lep
lace
men
taga
inst
sin
gle
PM
Ulo
ssfo
rth
eIE
EE
118-
bu
ssy
stem
wh
enth
ech
an
nel
lim
itis
3(c
on
sid
erin
gth
eze
ro-i
nje
ctio
ns)
.
Chapter 5
Concluding Remarks and Further
Study
5.1 Concluding Remarks
This thesis presents a new problem formulation and its associated solution based on
mixed integer linear programming method for obtaining the best locations of syn-
chronized phasor measurement units. The main contribution of the new formulation
is the way it accounts for the available number of PMU channels. Furthermore, zero-
injection measurements are incorporated into the problem formulation in order to
further minimize the required number of PMUs. Applying the developed technique
to different size systems, it is observed that PMUs having more than 4 channels
(positive-sequence) may not reduce the overall installation cost for medium-size sys-
tems. Moreover, it is observed that the channel limits which reduce the overall
installation cost will be larger for larger-size and/or more densely connected sys-
tems. In order to demonstrate the effect of sparsity on the required channel limits,
certain test systems are artificially modified by increasing connectivity in a system-
atic manner. The results indicate that densely connected systems will allow efficient
utilization of PMUs with large number of channels.
49
Chapter 5. Concluding Remarks and Further Study 50
This study also extends the results of conventional PMU placement to the case
where the solution is expected to be robust against failure of any single PMU. Any
existing injection measurements in the system, in particular those virtual ones at
passive buses with no generation or load, are also accounted for in the modified
optimization formulation. In this case, results of simulations on different type and
size test systems imply that using PMUs with large number of channels does not
minimize the investment in the measurement system. In most cases, having more
than 4 channels (positive sequence) does not reduce the required PMU count. Fur-
thermore, by strategic placement of PMUs, a very reliable metering design can be
achieved by placing PMUs at less than 70% of the buses in the system. This number
may be reduced significantly by taking advantage of zero-injection buses.
Ultimately, these results may be useful for the system planners as well as PMU
manufacturers when they make decisions on the next set of PMUs to be purchased
and installed or to be designed and marketed, respectively.
5.2 Further Study
We have studied and solved the problem of using PMUs with limited input capabil-
ities to achieve complete observability of the network. In other words, it is intended
to monitor at most a fixed number of currents from a bus. As a further study, the
PMU placement problem can be reinvestigated by taking into account the fact that
each PMU may have variations in channel capacity for a particular placement strat-
egy. Additionally, the costs for the proposed placement strategy and the prospective
placement procedures may be comparatively evaluated in order to determine the
best option. Undoubtedly, novel methodologies can also be implemented for model-
ing of zero-injection buses and reliable PMU placement to investigate the feasibility
of further reducing the number of PMUs required for entire network observability.
Appendix A
Functions and Scripts Used in the
PMU Placement Algorithm
A.1 Read Network Parameters and Build the Single-
Hop Connectivity Matrix A
function [y NoBran branch external_bus internal_bus] =
readAndBuildA(PowerFlowInputData,NoBus)
% read parameters of the network
fnet = fopen(PowerFlowInputData,’r’);
line = fgetl(fnet);
line = fgetl(fnet);
iter1 = 0;
iter2 = 0;
iter3 = 0;
while 1
51
Chapter A. Appendix A 52
line = fgetl(fnet);
iter1 = iter1 + 1;
if line(1:4) == ’BRAN’
break;
end
bus(iter1,1) = str2num(line(1:4));
end
bus(end,:) = []; % this entry corresponds to -9999
for ii = 1 : NoBus
% specify external bus numbers
external_bus(ii,1) = bus(ii,1);
% internal bus numbers
internal_bus(external_bus(ii),1) = it;
end
while 1
line = fgetl(fnet);
if line(1:4) == ’-999’
break;
end
iter2 = iter2 + 1;
cir(iter2,1) = str2num(line(17));
if cir(iter2,1) == 0
iter3 = iter3 + 1;
From_Bus(iter3,1) = str2num(line(1:4));
To_Bus(iter3,1) = str2num(line(6:9));
end
end
NoBran = iter3;
Chapter A. Appendix A 53
From_Bus = internal_bus(From_Bus);
To_Bus = internal_bus(To_Bus);
branch = [From_Bus To_Bus];
% create the bus admittance matrix, Y, with jX = j1.0.
ys = 1; [rowBran colBran] = size(From_Bus);
for iter = 1 : rowBran
Yi(ys) = From_Bus(iter,1);
Yj(ys) = To_Bus(iter,1);
Yv(ys) = 1; ys = ys + 1;
Yi(ys) = To_Bus(iter,1);
Yj(ys) = From_Bus(iter,1);
Yv(ys) = 1; ys = ys + 1;
Yi(ys) = From_Bus(iter,1);
Yj(ys) = From_Bus(iter,1);
Yv(ys) = 1; ys = ys + 1;
Yi(ys) = To_Bus(iter,1);
Yj(ys) = To_Bus(iter,1);
Yv(ys) = 1; ys = ys + 1;
end
y = sparse(Yi,Yj,Yv,NoBus,NoBus);
A = spones(y) + zeros(NoBus,NoBus);
disp(’Matrix A is found to be as follows: ’);
ST = fclose(fnet);
Chapter A. Appendix A 54
A.2 Find the Required Number of PMUs for Com-
plete Network Observability (Ignoring Zero-
Injection Measurements)
clear;
clc;
A = readAndBuildA(’pfinput14.dat’,14);
% A = readAndBuildA(’pfinput30.dat’,30);
% A = readAndBuildA(’pfinput57.dat’,57);
% A = readAndBuildA(’pfinput118.dat’,118);
%% reduce the network sparsity by adding 2nd, 3rd, etc. neighbors
% B2 = spones(A ^ 2) + zeros(size(A,1))
% B3 = spones(A ^ 3) + zeros(size(A,1))
[m,n] = size(A);
L = [];
for j = 1 : n
for i = 1 : j
if A(i,j) == 1
if j ~= i
L(i,j) = 1;
L(j,i) = 1;
end
end
end
end L;
Chapter A. Appendix A 55
ChannelLimit = input(’Choose a channel limit for the PMUs: ’);
H = [];
for k = 1 : n
if sum(L(:,k)) < ChannelLimit
V = find(L(:,k));
T = sparse(1,n);
T(1,k) = 1;
T(1,V) = 1;
H = sparse([H;T]);
else
V = nchoosek(find(L(:,k)),ChannelLimit);
[a,b] = size(V);
T = sparse(a,n);
for i = 1 : a
for j = 1 : b
T(:,k) = 1;
T(i,V(i,j)) = 1;
end
end
H = sparse([H;T]);
end
fprintf(’%d\n’,k)
end
H;
tic
f = ones(size(H,1),1);
Hnew = -H’;
b = -ones(n,1);
Chapter A. Appendix A 56
N = length(f);
x_L = zeros(N,1);
x_U = ones(N,1);
IntVars = ones(N,1);
PriLev = 1;
cpxControl.EPGAP = 0.1/100;
cpxControl.TILIM = 60*5;
[x, slack, v, rc, f_k, ninf, sinf, Inform, basis, lpiter, ...
glnodes, confstat, iconfstat, sa, cpxControl, presolve] = ...
cplex(f, Hnew, x_L, x_U, -inf*ones(n,1), b, ...
cpxControl, [], PriLev, [], IntVars);
disp(’The optimum number of PMUs is: ’)
fprintf(’%d\n’,sum(x))
toc
Chapter A. Appendix A 57
A.3 Find the Required Number of PMUs for Com-
plete Network Observability (Considering Zero-
Injection Measurements)
clear;
clc;
A = readAndBuildA(’pfinput14.dat’,14);
% A = readAndBuildA(’pfinput30.dat’,30);
% A = readAndBuildA(’pfinput57.dat’,57);
% A = readAndBuildA(’pfinput118.dat’,118);
[m,n] = size(A);
L = [];
for j = 1 : n
for i = 1 : j
if A(i,j) == 1
if j ~= i
L(i,j) = 1;
L(j,i) = 1;
end
end
end
end
L;
ChannelLimit = input(’Choose a channel limit for the PMUs: ’);
Chapter A. Appendix A 58
H = [];
for k = 1 : n
if sum(L(:,k)) < ChannelLimit
V = find(L(:,k));
T = zeros(1,n);
T(1,k) = 1;
T(1,V) = 1;
H = [H;T];
else
V = nchoosek(find(L(:,k)),ChannelLimit);
[a,b] = size(V);
T = zeros(a,n);
for i = 1 : a
for j = 1 : b
T(:,k) = 1;
T(i,V(i,j)) = 1;
end
end
H = [H; T];
end
fprintf(’%d\n’,k)
end
H;
tic
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%% for the IEEE 14-bus test system
C = zeros(size(H,2),4);
C(4,1) = 1; C(7,2) = 1;
Chapter A. Appendix A 59
C(8,3) = 1; C(9,4) = 1;
D = zeros(1,4);
D(1,1:4) = -1;
R = -ones(n,1);
% use R = -2 * ones(n,1) for reliable PMU placement, instead
R(4) = 0; R(7) = 0; R(8) = 0; R(9) = 0;
c = -3;
f = [ones(size(H,1),1);zeros(size(D,2),1)];
Hnew = [-H’,C;[zeros(size(D,1),size(H,1)),D]];
RHS = [R; c];
X = bintprog(f,Hnew,RHS);
X = X(1:size(H,1));
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%% for the IEEE 30-bus test system
% C = zeros(size(H,2),17);
%
% C(2,1) = 1; C(4,2) = 1; C(6,3) = 1; C(7,4) = 1;
% C(8,5) = 1; C(9,6) = 1; C(10,7) = 1; C(11,8) = 1;
% C(21,9) = 1; C(22,10) = 1; C(24,11) = 1; C(25,12) = 1;
% C(26,13) = 1; C(27,14) = 1; C(28,15) = 1; C(29,16) = 1;
Chapter A. Appendix A 60
% C(30,17) = 1;
%
% D = zeros(6,17);
% D(1,1:7) = -1; D(1,15) = -1; D(2,3) = -1;
% D(2,6:8) = -1; D(3,7) = -1; D(3,9:11) = -1;
% D(4,11:14) = -1; D(5,12) = -1; D(5,14:17) = -1;
% D(6,9) = -1; D(6,11) = -1; D(6,14:15) = -1;
%
% R = -ones(n,1);
% R(2) = 0; R(4) = 0; R(6) = 0; R(7) = 0;
% R(8) = 0; R(9) = 0; R(10) = 0; R(11) = 0;
% R(21) = 0; R(22) = 0; R(24) = 0; R(25) = 0;
% R(26) = 0; R(27) = 0; R(28) = 0; R(29) = 0;
% R(30) = 0;
%
% c = [-7 -3 -3 -3 -4 -3]’;
%
% f = [ones(size(H,1),1);zeros(size(D,2),1)];
% Hnew = [-H’,C;[zeros(size(D,1),size(H,1)),D]];
% RHS = [R; c];
% X = bintprog(f,Hnew,RHS);
% X = X(1:size(H,1));
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%% for the IEEE 57-bus test system
% C = zeros(size(H,2),39);
% C(3,1) = 1; C(4,2) = 1; C(5,3) = 1;
Chapter A. Appendix A 61
% C(6,4) = 1; C(7,5) = 1; C(8,6) = 1;
% C(9,7) = 1; C(11,8) = 1; C(13,9) = 1;
% C(14,10) = 1; C(15,11) = 1; C(18,12) = 1;
% C(20,13) = 1; C(21,14) = 1; C(22,15) = 1;
% C(23,16) = 1; C(24,17) = 1; C(25,18) = 1;
% C(26,19) = 1; C(27,20) = 1; C(29,21) = 1;
% C(32,22) = 1; C(34,23) = 1; C(35,24) = 1;
% C(36,25) = 1; C(37,26) = 1; C(38,27) = 1;
% C(39,28) = 1; C(40,29) = 1; C(41,30) = 1;
% C(43,31) = 1; C(44,32) = 1; C(45,33) = 1;
% C(46,34) = 1; C(47,35) = 1; C(48,36) = 1;
% C(49,37) = 1; C(56,38) = 1; C(57,39) = 1;
%
% D = zeros(15,39);
% D(1,1:4) = -1; D(1,12) = -1; D(2,4:6) = -1;
% D(2,21) = -1; D(3,7:9) = -1; D(3,30:31) = -1;
% D(4,13:15) = -1; D(5,14:16) = -1; D(5,27) = -1;
% D(6,16:19) = -1; D(7,17) = -1; D(7,19:20) = -1;
% D(8,22:24) = -1; D(9,24:26) = -1; D(9,29) = -1;
% D(10,25:28) = -1; D(11,26) = -1; D(11,28) = -1;
% D(11,39) = -1; D(12,25) = -1; D(12,29) = -1;
% D(12,38) = -1; D(13,11) = -1; D(13,32:33) = -1;
% D(14,10) = -1; D(14,34:35) = -1; D(15,27) = -1;
% D(15,35:37) = -1;
%
% R = -ones(n,1);
% R(3:9) = 0; R(11) = 0; R(13:15) = 0;
% R(18) = 0; R(20:27) = 0; R(29) = 0;
% R(32) = 0; R(34:41) = 0; R(43:49) = 0;
% R(56:57) = 0;
%
%
Chapter A. Appendix A 62
% c = [-4 -3 -4 -2 -3 -3 -2 -2 -3 -3 -2 -2 -2 -2 -3]’;
%
% f = [ones(size(H,1),1);zeros(size(D,2),1)];
% Hnew = [-H’,C;[zeros(size(D,1),size(H,1)),D]];
% RHS = [R; c];
% X = bintprog(f,Hnew,RHS);
% X = X(1:size(H,1));
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%% for the IEEE 118-bus test system
% C = zeros(size(H,2),32);
% C(3,1) = 1; C(4,2) = 1; C(5,3) = 1;
% C(6,4) = 1; C(8,5) = 1; C(9,6) = 1;
% C(10,7) = 1; C(11,8) = 1; C(17,9) = 1;
% C(26,10) = 1; C(30,11) = 1; C(33,12) = 1;
% C(34,13) = 1; C(35,14) = 1; C(37,15) = 1;
% C(38,16) = 1; C(39,17) = 1; C(40,18) = 1;
% C(59,19) = 1; C(61,20) = 1; C(63,21) = 1;
% C(64,22) = 1; C(65,23) = 1; C(68,24) = 1;
% C(69,25) = 1; C(70,26) = 1; C(71,27) = 1;
% C(72,28) = 1; C(73,29) = 1; C(80,30) = 1;
% C(81,31) = 1; C(116,32) = 1;
%
% D = zeros(10,32);
% D(1,1:5) = -1; D(1,8) = -1 ; D(2,5:7) = -1;
% D(3,5) = -1; D(3,9:11) = -1; D(3,16) = -1;
% D(4,12:18) = -1; D(5,11) = -1; D(5,15:16) = -1;
Chapter A. Appendix A 63
% D(5,23) = -1; D(6,19) = -1; D(6,21:22) = -1;
% D(7,20:23) = -1; D(8,23:25) = -1; D(8,31:32) = -1;
% D(9,26:29) = -1; D(10,24) = -1; D(10,30:31) = -1;
%
%
% R = -ones(n,1);
% R(3:6) = 0; R(8:11) = 0; R(17) = 0;
% R(26) = 0; R(30) = 0; R(33:35) = 0;
% R(37:40) = 0; R(59) = 0; R(61) = 0;
% R(63:65) = 0; R(68:73) = 0; R(80:81) = 0;
% R(116) = 0;
%
% c = [-5 -2 -4 -6 -3 -2 -3 -4 -3 -2]’;
% f = [ones(size(H,1),1);zeros(size(D,2),1)];
% Hnew = [-H’,C;[zeros(size(D,1),size(H,1)),D]];
% RHS = [R; c];
% X = bintprog(f,Hnew,RHS);
% X = X(1:size(H,1));
disp(’The optimum number of PMUs is: ’)
fprintf(’%d\n’,sum(X))
toc
Appendix B
IEEE Test Systems Data Used in
the PMU Placement Algorithm
This appendix section contains the data regarding the IEEE 14-, 30-, 57-, and 118-
bus test systems [60], which are utilized in our simulations. The system information
of these IEEE test systems is shown in Table B.1 given below:
Table B.1: System Information of Studied IEEE Test Systems
Test Number of Number of Zero-InjectionSystem Branches Zero-Injections Bus(es)
IEEE20 1 7
14-BusIEEE
41 6 6, 9, 22, 25, 27, 2830-BusIEEE
78 154, 7, 11, 21, 22, 24, 26, 34,
57-Bus 36, 37, 39, 40, 45, 46, 48IEEE
179 105, 9, 30, 37, 38, 63, 64, 68,
118-Bus 71, 81
64
B.1
IEE
E14-B
us
Test
Syst
em
Data
08/19/93UWARCHIVE
100.0
1962WIEEE14BusTestCase
BUSDATAFOLLOWS
14ITEMS
1Bus1
HV
11
31.060
0.0
0.0
0.0
232.4
-16.9
0.0
1.060
0.0
0.0
0.0
0.0
0
2Bus2
HV
11
21.045
-4.98
21.7
12.7
40.0
42.4
0.0
1.045
50.0
-40.0
0.0
0.0
0
3Bus3
HV
11
21.010-12.72
94.2
19.0
0.0
23.4
0.0
1.010
40.0
0.0
0.0
0.0
0
4Bus4
HV
11
01.019-10.33
47.8
-3.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
5Bus5
HV
11
01.020
-8.78
7.6
1.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
6Bus6
LV
11
21.070-14.22
11.2
7.5
0.0
12.2
0.0
1.070
24.0
-6.0
0.0
0.0
0
7Bus7
ZV
11
01.062-13.37
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
8Bus8
TV
11
21.090-13.36
0.0
0.0
0.0
17.4
0.0
1.090
24.0
-6.0
0.0
0.0
0
9Bus9
LV
11
01.056-14.94
29.5
16.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.19
0
10Bus10
LV
11
01.051-15.10
9.0
5.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
11Bus11
LV
11
01.057-14.79
3.5
1.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
12Bus12
LV
11
01.055-15.07
6.1
1.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
13Bus13
LV
11
01.050-15.16
13.5
5.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
14Bus14
LV
11
01.036-16.04
14.9
5.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
-999BRANCHDATAFOLLOWS
20ITEMS
12
1110
0.01938
0.05917
0.0528
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
15
1110
0.05403
0.22304
0.0492
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
23
1110
0.04699
0.19797
0.0438
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
24
1110
0.05811
0.17632
0.0340
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
25
1110
0.05695
0.17388
0.0346
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
34
1110
0.06701
0.17103
0.0128
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
45
1110
0.01335
0.04211
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
47
1110
0.0
0.20912
0.0
00
000
0.978
0.00.0
0.0
0.0
0.0
0.0
49
1110
0.0
0.55618
0.0
00
000
0.969
0.00.0
0.0
0.0
0.0
0.0
56
1110
0.0
0.25202
0.0
00
000
0.932
0.00.0
0.0
0.0
0.0
0.0
611
1110
0.09498
0.19890
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
612
1110
0.12291
0.25581
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
613
1110
0.06615
0.13027
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
78
1110
0.0
0.17615
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
79
1110
0.0
0.11001
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
910
1110
0.03181
0.08450
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
914
1110
0.12711
0.27038
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
10
11
1110
0.08205
0.19207
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
12
13
1110
0.22092
0.19988
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
13
14
1110
0.17093
0.34802
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
-999LOSSZONESFOLLOWS
1ITEMS
1IEEE14BUS
-99INTERCHANGEDATAFOLLOWS
1ITEMS
12Bus2
HV
0.0
999.99
IEEE14
IEEE14BusTestCase
-9TIELINESFOLLOWS
0ITEMS-999
ENDOFDATA
B.2
IEE
E30-B
us
Test
Syst
em
Data
08/20/93UWARCHIVE
100.0
1961WIEEE30BusTestCase
BUSDATAFOLLOWS
30ITEMS
1GlenLyn132
11
31.060
0.0
0.0
0.0
260.2
-16.1
132.0
1.060
0.0
0.0
0.0
0.0
0
2Claytor
132
11
21.043
-5.48
21.7
12.7
40.0
50.0
132.0
1.045
50.0
-40.0
0.0
0.0
0
3Kumis
132
11
01.021
-7.96
2.4
1.2
0.0
0.0
132.0
0.0
0.0
0.0
0.0
0.0
0
4Hancock
132
11
01.012
-9.62
7.6
1.6
0.0
0.0
132.0
0.0
0.0
0.0
0.0
0.0
0
5Fieldale132
11
21.010-14.37
94.2
19.0
0.0
37.0
132.0
1.010
40.0
-40.0
0.0
0.0
0
6Roanoke
132
11
01.010-11.34
0.0
0.0
0.0
0.0
132.0
0.0
0.0
0.0
0.0
0.0
0
7Blaine
132
11
01.002-13.12
22.8
10.9
0.0
0.0
132.0
0.0
0.0
0.0
0.0
0.0
0
8Reusens
132
11
21.010-12.10
30.0
30.0
0.0
37.3
132.0
1.010
40.0
-10.0
0.0
0.0
0
9Roanoke
1.0
11
01.051-14.38
0.0
0.0
0.0
0.0
1.0
0.0
0.0
0.0
0.0
0.0
0
10Roanoke
33
11
01.045-15.97
5.8
2.0
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.19
0
11Roanoke
11
11
21.082-14.39
0.0
0.0
0.0
16.2
11.0
1.082
24.0
-6.0
0.0
0.0
0
12Hancock
33
11
01.057-15.24
11.2
7.5
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
13Hancock
11
11
21.071-15.24
0.0
0.0
0.0
10.6
11.0
1.071
24.0
-6.0
0.0
0.0
0
14Bus14
33
11
01.042-16.13
6.2
1.6
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
15Bus15
33
11
01.038-16.22
8.2
2.5
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
16Bus16
33
11
01.045-15.83
3.5
1.8
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
17Bus17
33
11
01.040-16.14
9.0
5.8
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
18Bus18
33
11
01.028-16.82
3.2
0.9
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
19Bus19
33
11
01.026-17.00
9.5
3.4
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
20Bus20
33
11
01.030-16.80
2.2
0.7
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
21Bus21
33
11
01.033-16.42
17.5
11.2
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
22Bus22
33
11
01.033-16.41
0.0
0.0
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
23Bus23
33
11
01.027-16.61
3.2
1.6
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
24Bus24
33
11
01.021-16.78
8.7
6.7
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.043
0
25Bus25
33
11
01.017-16.35
0.0
0.0
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
26Bus26
33
11
01.000-16.77
3.5
2.3
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
27Cloverdle33
11
01.023-15.82
0.0
0.0
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
28Cloverdle132
11
01.007-11.97
0.0
0.0
0.0
0.0
132.0
0.0
0.0
0.0
0.0
0.0
0
29Bus29
33
11
01.003-17.06
2.4
0.9
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
30Bus30
33
11
00.992-17.94
10.6
1.9
0.0
0.0
33.0
0.0
0.0
0.0
0.0
0.0
0
-999BRANCHDATAFOLLOWS
41ITEMS
12
1110
0.0192
0.0575
0.0528
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
13
1110
0.0452
0.1652
0.0408
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
24
1110
0.0570
0.1737
0.0368
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
34
1110
0.0132
0.0379
0.0084
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
25
1110
0.0472
0.1983
0.0418
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
26
1110
0.0581
0.1763
0.0374
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
46
1110
0.0119
0.0414
0.0090
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
57
1110
0.0460
0.1160
0.0204
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
67
1110
0.0267
0.0820
0.0170
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
68
1110
0.0120
0.0420
0.0090
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
69
1110
0.0
0.2080
0.0
00
000
0.978
0.00.0
0.0
0.0
0.0
0.0
610
1110
0.0
0.5560
0.0
00
000
0.969
0.00.0
0.0
0.0
0.0
0.0
911
1110
0.0
0.2080
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
910
1110
0.0
0.1100
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
412
1110
0.0
0.2560
0.0
00
000
0.932
0.00.0
0.0
0.0
0.0
0.0
12
13
1110
0.0
0.1400
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
12
14
1110
0.1231
0.2559
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
12
15
1110
0.0662
0.1304
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
12
16
1110
0.0945
0.1987
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
14
15
1110
0.2210
0.1997
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
16
17
1110
0.0524
0.1923
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
15
18
1110
0.1073
0.2185
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
18
19
1110
0.0639
0.1292
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
19
20
1110
0.0340
0.0680
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
10
20
1110
0.0936
0.2090
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
10
17
1110
0.0324
0.0845
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
10
21
1110
0.0348
0.0749
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
10
22
1110
0.0727
0.1499
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
21
22
1110
0.0116
0.0236
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
15
23
1110
0.1000
0.2020
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
22
24
1110
0.1150
0.1790
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
23
24
1110
0.1320
0.2700
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
24
25
1110
0.1885
0.3292
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
25
26
1110
0.2544
0.3800
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
25
27
1110
0.1093
0.2087
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
28
27
1110
0.0
0.3960
0.0
00
000
0.968
0.00.0
0.0
0.0
0.0
0.0
27
29
1110
0.2198
0.4153
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
27
30
1110
0.3202
0.6027
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
29
30
1110
0.2399
0.4533
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
828
1110
0.0636
0.2000
0.0428
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
628
1110
0.0169
0.0599
0.0130
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
-999LOSSZONESFOLLOWS
1ITEMS
1IEEE30BUS
-99INTERCHANGEDATAFOLLOWS
1ITEMS-9
12Claytor
132
0.0
999.99
IEEE30
IEEE30BusTestCase
TIELINESFOLLOWS
0ITEMS-999
ENDOFDATA
B.3
IEE
E57-B
us
Test
Syst
em
Data
08/25/93UWARCHIVE
100.0
1961WIEEE57BusTestCase
BUSDATAFOLLOWS
57ITEMS
1Kanawha
V1
11
31.040
0.0
55.0
17.0
128.9
-16.1
0.0
1.040
0.0
0.0
0.0
0.0
0
2Turner
V1
11
21.010
-1.18
3.0
88.0
0.0
-0.8
0.0
1.010
50.0
-17.0
0.0
0.0
0
3Logan
V1
11
20.985
-5.97
41.0
21.0
40.0
-1.0
0.0
0.985
60.0
-10.0
0.0
0.0
0
4Sprigg
V1
11
00.981
-7.32
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
5Bus5
V1
11
00.976
-8.52
13.0
4.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
6BeaverCkV1
11
20.980
-8.65
75.0
2.0
0.0
0.8
0.0
0.980
25.0
-8.0
0.0
0.0
0
7Bus7
V1
11
00.984
-7.58
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
8ClinchRvV1
11
21.005
-4.45
150.0
22.0
450.0
62.1
0.0
1.005
200.0
-140.0
0.0
0.0
0
9SaltvilleV1
11
20.980
-9.56
121.0
26.0
0.0
2.2
0.0
0.980
9.0
-3.0
0.0
0.0
0
10Bus10
V1
11
00.986-11.43
5.0
2.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
11Tazewell
V1
11
00.974-10.17
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
12GlenLyn
V1
11
21.015-10.46
377.0
24.0
310.0
128.5
0.0
1.015
155.0
-150.0
0.0
0.0
0
13Bus13
V1
11
00.979
-9.79
18.0
2.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
14Bus14
V1
11
00.970
-9.33
10.5
5.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
15Bus15
V1
11
00.988
-7.18
22.0
5.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
16Bus16
V1
11
01.013
-8.85
43.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
17Bus17
V1
11
01.017
-5.39
42.0
8.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
18Sprigg
V2
11
01.001-11.71
27.2
9.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.10
0
19Bus19
V2
11
00.970-13.20
3.3
0.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
20Bus20
V2
11
00.964-13.41
2.3
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
21Bus21
V3
11
01.008-12.89
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
22Bus22
V3
11
01.010-12.84
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
23Bus23
V3
11
01.008-12.91
6.3
2.1
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
24Bus24
V3
11
00.999-13.25
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
25Bus25
V4
11
00.982-18.13
6.3
3.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.059
0
26Bus26
V5
11
00.959-12.95
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
27Bus27
V5
11
00.982-11.48
9.3
0.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
28Bus28
V5
11
00.997-10.45
4.6
2.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
29Bus29
V5
11
01.010
-9.75
17.0
2.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
30Bus30
V4
11
00.962-18.68
3.6
1.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
31Bus31
V4
11
00.936-19.34
5.8
2.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
32Bus32
V4
11
00.949-18.46
1.6
0.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
33Bus33
V4
11
00.947-18.50
3.8
1.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
34Bus34
V3
11
00.959-14.10
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
35Bus35
V3
11
00.966-13.86
6.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
36Bus36
V3
11
00.976-13.59
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
37Bus37
V3
11
00.985-13.41
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
38Bus38
V3
11
01.013-12.71
14.0
7.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
39Bus39
V3
11
00.983-13.46
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
40Bus40
V3
11
00.973-13.62
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
41Tazewell
V6
11
00.996-14.05
6.3
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
42Bus42
V6
11
00.966-15.50
7.1
4.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
43Tazewell
V7
11
01.010-11.33
2.0
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
44Bus44
V3
11
01.017-11.86
12.0
1.8
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
45Bus45
V3
11
01.036
-9.25
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
46Bus46
V3
11
01.050-11.89
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
47Bus47
V3
11
01.033-12.49
29.7
11.6
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
48Bus48
V3
11
01.027-12.59
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
49Bus49
V3
11
01.036-12.92
18.0
8.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
50Bus50
V3
11
01.023-13.39
21.0
10.5
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
51Bus51
V3
11
01.052-12.52
18.0
5.3
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
52Bus52
V5
11
00.980-11.47
4.9
2.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
53Bus53
V5
11
00.971-12.23
20.0
10.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.063
0
54Bus54
V5
11
00.996-11.69
4.1
1.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
55SaltvilleV5
11
01.031-10.78
6.8
3.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
56Bus56
V6
11
00.968-16.04
7.6
2.2
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
57Bus57
V6
11
00.965-16.56
6.7
2.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
-999BRANCHDATAFOLLOWS
80ITEMS
12
1110
0.0083
0.0280
0.1290
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
23
1110
0.0298
0.0850
0.0818
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
34
1110
0.0112
0.0366
0.0380
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
45
1110
0.0625
0.1320
0.0258
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
46
1110
0.0430
0.1480
0.0348
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
67
1110
0.0200
0.1020
0.0276
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
68
1110
0.0339
0.1730
0.0470
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
89
1110
0.0099
0.0505
0.0548
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
910
1110
0.0369
0.1679
0.0440
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
911
1110
0.0258
0.0848
0.0218
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
912
1110
0.0648
0.2950
0.0772
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
913
1110
0.0481
0.1580
0.0406
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
13
14
1110
0.0132
0.0434
0.0110
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
13
15
1110
0.0269
0.0869
0.0230
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
115
1110
0.0178
0.0910
0.0988
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
116
1110
0.0454
0.2060
0.0546
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
117
1110
0.0238
0.1080
0.0286
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
315
1110
0.0162
0.0530
0.0544
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
418
1110
0.0
0.5550
0.0
00
000
0.970
0.00.0
0.0
0.0
0.0
0.0
418
1110
0.0
0.4300
0.0
00
000
0.978
0.00.0
0.0
0.0
0.0
0.0
56
1110
0.0302
0.0641
0.0124
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
78
1110
0.0139
0.0712
0.0194
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
10
12
1110
0.0277
0.1262
0.0328
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
11
13
1110
0.0223
0.0732
0.0188
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
12
13
1110
0.0178
0.0580
0.0604
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
12
16
1110
0.0180
0.0813
0.0216
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
12
17
1110
0.0397
0.1790
0.0476
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
14
15
1110
0.0171
0.0547
0.0148
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
18
19
1110
0.4610
0.6850
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
19
20
1110
0.2830
0.4340
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
21
20
1110
0.0
0.7767
0.0
00
000
1.043
0.00.0
0.0
0.0
0.0
0.0
21
22
1110
0.0736
0.1170
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
22
23
1110
0.0099
0.0152
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
23
24
1110
0.1660
0.2560
0.0084
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
24
25
1110
0.0
1.1820
0.0
00
000
1.000
0.00.0
0.0
0.0
0.0
0.0
24
25
1110
0.0
1.2300
0.0
00
000
1.000
0.00.0
0.0
0.0
0.0
0.0
24
26
1110
0.0
0.0473
0.0
00
000
1.043
0.00.0
0.0
0.0
0.0
0.0
26
27
1110
0.1650
0.2540
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
27
28
1110
0.0618
0.0954
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
28
29
1110
0.0418
0.0587
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
729
1110
0.0
0.0648
0.0
00
000
0.967
0.00.0
0.0
0.0
0.0
0.0
25
30
1110
0.1350
0.2020
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
30
31
1110
0.3260
0.4970
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
31
32
1110
0.5070
0.7550
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
32
33
1110
0.0392
0.0360
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
34
32
1110
0.0
0.9530
0.0
00
000
0.975
0.00.0
0.0
0.0
0.0
0.0
34
35
1110
0.0520
0.0780
0.0032
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
35
36
1110
0.0430
0.0537
0.0016
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
36
37
1110
0.0290
0.0366
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
37
38
1110
0.0651
0.1009
0.0020
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
37
39
1110
0.0239
0.0379
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
36
40
1110
0.0300
0.0466
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
22
38
1110
0.0192
0.0295
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
11
41
1110
0.0
0.7490
0.0
00
000
0.955
0.00.0
0.0
0.0
0.0
0.0
41
42
1110
0.2070
0.3520
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
41
43
1110
0.0
0.4120
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
38
44
1110
0.0289
0.0585
0.0020
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
15
45
1110
0.0
0.1042
0.0
00
000
0.955
0.00.0
0.0
0.0
0.0
0.0
14
46
1110
0.0
0.0735
0.0
00
000
0.900
0.00.0
0.0
0.0
0.0
0.0
46
47
1110
0.0230
0.0680
0.0032
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
47
48
1110
0.0182
0.0233
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
48
49
1110
0.0834
0.1290
0.0048
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
49
50
1110
0.0801
0.1280
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
50
51
1110
0.1386
0.2200
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
10
51
1110
0.0
0.0712
0.0
00
000
0.930
0.00.0
0.0
0.0
0.0
0.0
13
49
1110
0.0
0.1910
0.0
00
000
0.895
0.00.0
0.0
0.0
0.0
0.0
29
52
1110
0.1442
0.1870
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
52
53
1110
0.0762
0.0984
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
53
54
1110
0.1878
0.2320
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
54
55
1110
0.1732
0.2265
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
11
43
1110
0.0
0.1530
0.0
00
000
0.958
0.00.0
0.0
0.0
0.0
0.0
44
45
1110
0.0624
0.1242
0.0040
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
40
56
1110
0.0
1.1950
0.0
00
000
0.958
0.00.0
0.0
0.0
0.0
0.0
56
41
1110
0.5530
0.5490
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
56
42
1110
0.2125
0.3540
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
39
57
1110
0.0
1.3550
0.0
00
000
0.980
0.00.0
0.0
0.0
0.0
0.0
57
56
1110
0.1740
0.2600
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
38
49
1110
0.1150
0.1770
0.0030
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
38
48
1110
0.0312
0.0482
0.0
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
955
1110
0.0
0.1205
0.0
00
000
0.940
0.00.0
0.0
0.0
0.0
0.0
-999LOSSZONESFOLLOWS
1ITEMS
1IEEE57BUS
-99INTERCHANGEDATAFOLLOWS
1ITEMS-9
18ClinchRvV1
0.0
999.99
IEEE57
IEEE57BusTestCase
TIELINESFOLLOWS
0ITEMS-999
ENDOFDATA
B.4
IEE
E118-B
us
Test
Syst
em
Data
08/25/93UWARCHIVE
100.0
1961WIEEE118BusTestCase
BUSDATAFOLLOWS
57ITEMS
1Riversde
V2
11
20.955
10.67
51.0
27.0
0.0
0.0
0.0
0.955
15.0
-5.0
0.0
0.0
0
2Pokagon
V2
11
00.971
11.22
20.0
9.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
3HickryCk
V2
11
00.968
11.56
39.0
10.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
4NwCarlsl
V2
11
20.998
15.28
30.0
12.0
-9.0
0.0
0.0
0.998
300.0
-300.0
0.0
0.0
0
5Olive
V2
11
01.002
15.73
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
-0.40
0
6Kankakee
V2
11
20.990
13.00
52.0
22.0
0.0
0.0
0.0
0.990
50.0
-13.0
0.0
0.0
0
7JacksnRd
V2
11
00.989
12.56
19.0
2.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
8Olive
V1
11
21.015
20.77
0.0
0.0
-28.0
0.0
0.0
1.015
300.0
-300.0
0.0
0.0
0
9Bequine
V1
11
01.043
28.02
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
10Breed
V1
11
21.050
35.61
0.0
0.0
450.0
0.0
0.0
1.050
200.0
-147.0
0.0
0.0
0
11SouthBnd
V2
11
00.985
12.72
70.0
23.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
12TwinBrch
V2
11
20.990
12.20
47.0
10.0
85.0
0.0
0.0
0.990
120.0
-35.0
0.0
0.0
0
13Concord
V2
11
00.968
11.35
34.0
16.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
14GoshenJt
V2
11
00.984
11.50
14.0
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
15FtWayne
V2
11
20.970
11.23
90.0
30.0
0.0
0.0
0.0
0.970
30.0
-10.0
0.0
0.0
0
16N.E.
V2
11
00.984
11.91
25.0
10.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
17Sorenson
V2
11
00.995
13.74
11.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
18McKinley
V2
11
20.973
11.53
60.0
34.0
0.0
0.0
0.0
0.973
50.0
-16.0
0.0
0.0
0
19Lincoln
V2
11
20.963
11.05
45.0
25.0
0.0
0.0
0.0
0.962
24.0
-8.0
0.0
0.0
0
20Adams
V2
11
00.958
11.93
18.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
21Jay
V2
11
00.959
13.52
14.0
8.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
22Randolph
V2
11
00.970
16.08
10.0
5.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
23CollCrnr
V2
11
01.000
21.00
7.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
24Trenton
V2
11
20.992
20.89
0.0
0.0
-13.0
0.0
0.0
0.992
300.0
-300.0
0.0
0.0
0
25TannrsCk
V2
11
21.050
27.93
0.0
0.0
220.0
0.0
0.0
1.050
140.0
-47.0
0.0
0.0
0
26TannrsCk
V1
11
21.015
29.71
0.0
0.0
314.0
0.0
0.0
1.015
1000.0-1000.0
0.0
0.0
0
27Madison
V2
11
20.968
15.35
62.0
13.0
-9.0
0.0
0.0
0.968
300.0
-300.0
0.0
0.0
0
28Mullin
V2
11
00.962
13.62
17.0
7.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
29Grant
V2
11
00.963
12.63
24.0
4.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
30Sorenson
V1
11
00.968
18.79
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
31DeerCrk
V2
11
20.967
12.75
43.0
27.0
7.0
0.0
0.0
0.967
300.0
-300.0
0.0
0.0
0
32Delaware
V2
11
20.964
14.80
59.0
23.0
0.0
0.0
0.0
0.963
42.0
-14.0
0.0
0.0
0
33Haviland
V2
11
00.972
10.63
23.0
9.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
34Rockhill
V2
11
20.986
11.30
59.0
26.0
0.0
0.0
0.0
0.984
24.0
-8.0
0.0
0.14
0
35WestLima
V2
11
00.981
10.87
33.0
9.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
36Sterling
V2
11
20.980
10.87
31.0
17.0
0.0
0.0
0.0
0.980
24.0
-8.0
0.0
0.0
0
37EastLima
V2
11
00.992
11.77
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
-0.25
0
38EastLima
V1
11
00.962
16.91
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
39NwLibrty
V2
11
00.970
8.41
27.0
11.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
40WestEnd
V2
11
20.970
7.35
20.0
23.0
-46.0
0.0
0.0
0.970
300.0
-300.0
0.0
0.0
0
41S.Tiffin
V2
11
00.967
6.92
37.0
10.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
42Howard
V2
11
20.985
8.53
37.0
23.0
-59.0
0.0
0.0
0.985
300.0
-300.0
0.0
0.0
0
43S.Kenton
V2
11
00.978
11.28
18.0
7.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
44WMVernon
V2
11
00.985
13.82
16.0
8.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.10
0
45N.Newark
V2
11
00.987
15.67
53.0
22.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.10
0
46W.Lancst
V2
11
21.005
18.49
28.0
10.0
19.0
0.0
0.0
1.005
100.0
-100.0
0.0
0.10
0
47Crooksvl
V2
11
01.017
20.73
34.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
48Zanesvll
V2
11
01.021
19.93
20.0
11.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.15
0
49Philo
V2
11
21.025
20.94
87.0
30.0
204.0
0.0
0.0
1.025
210.0
-85.0
0.0
0.0
0
50WCambrdg
V2
11
01.001
18.90
17.0
4.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
51Newcmrst
V2
11
00.967
16.28
17.0
8.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
52SCoshoct
V2
11
00.957
15.32
18.0
5.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
53Wooster
V2
11
00.946
14.35
23.0
11.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
54Torrey
V2
11
20.955
15.26
113.0
32.0
48.0
0.0
0.0
0.955
300.0
-300.0
0.0
0.0
0
55Wagenhls
V2
11
20.952
14.97
63.0
22.0
0.0
0.0
0.0
0.952
23.0
-8.0
0.0
0.0
0
56Sunnysde
V2
11
20.954
15.16
84.0
18.0
0.0
0.0
0.0
0.954
15.0
-8.0
0.0
0.0
0
57WNwPhil1
V2
11
00.971
16.36
12.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
58WNwPhil2
V2
11
00.959
15.51
12.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
59Tidd
V2
11
20.985
19.37
277.0
113.0
155.0
0.0
0.0
0.985
180.0
-60.0
0.0
0.0
0
60SWKammer
V2
11
00.993
23.15
78.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
61W.Kammer
V2
11
20.995
24.04
0.0
0.0
160.0
0.0
0.0
0.995
300.0
-100.0
0.0
0.0
0
62Natrium
V2
11
20.998
23.43
77.0
14.0
0.0
0.0
0.0
0.998
20.0
-20.0
0.0
0.0
0
63Tidd
V1
11
00.969
22.75
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
64Kammer
V1
11
00.984
24.52
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
65Muskngum
V1
11
21.005
27.65
0.0
0.0
391.0
0.0
0.0
1.005
200.0
-67.0
0.0
0.0
0
66Muskngum
V2
11
21.050
27.48
39.0
18.0
392.0
0.0
0.0
1.050
200.0
-67.0
0.0
0.0
0
67Summerfl
V2
11
01.020
24.84
28.0
7.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
68Sporn
V1
11
01.003
27.55
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
69Sporn
V2
11
31.035
30.00
0.0
0.0
516.4
0.0
0.0
1.035
300.0
-300.0
0.0
0.0
0
70Portsmth
V2
11
20.984
22.58
66.0
20.0
0.0
0.0
0.0
0.984
32.0
-10.0
0.0
0.0
0
71NPortsmt
V2
11
00.987
22.15
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
72Hillsbro
V2
11
20.980
20.98
0.0
0.0
-12.0
0.0
0.0
0.980
100.0
-100.0
0.0
0.0
0
73Sargents
V2
11
20.991
21.94
0.0
0.0
-6.0
0.0
0.0
0.991
100.0
-100.0
0.0
0.0
0
74Bellefnt
V2
11
20.958
21.64
68.0
27.0
0.0
0.0
0.0
0.958
9.0
-6.0
0.0
0.12
0
75SthPoint
V2
11
00.967
22.91
47.0
11.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
76Darrah
V2
11
20.943
21.77
68.0
36.0
0.0
0.0
0.0
0.943
23.0
-8.0
0.0
0.0
0
77Turner
V2
11
21.006
26.72
61.0
28.0
0.0
0.0
0.0
1.006
70.0
-20.0
0.0
0.0
0
78Chemical
V2
11
01.003
26.42
71.0
26.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
79CapitlHl
V2
11
01.009
26.72
39.0
32.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.20
0
80CabinCrk
V2
11
21.040
28.96
130.0
26.0
477.0
0.0
0.0
1.040
280.0
-165.0
0.0
0.0
0
81Kanawha
V1
11
00.997
28.10
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
82Logan
V2
11
00.989
27.24
54.0
27.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.20
0
83Sprigg
V2
11
00.985
28.42
20.0
10.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.10
0
84BetsyLne
V2
11
00.980
30.95
11.0
7.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
85BeaverCk
V2
11
20.985
32.51
24.0
15.0
0.0
0.0
0.0
0.985
23.0
-8.0
0.0
0.0
0
86Hazard
V2
11
00.987
31.14
21.0
10.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
87Pinevlle
V3
11
21.015
31.40
0.0
0.0
4.0
0.0
0.0
1.015
1000.0
-100.0
0.0
0.0
0
88Fremont
V2
11
00.987
35.64
48.0
10.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
89ClinchRv
V2
11
21.005
39.69
0.0
0.0
607.0
0.0
0.0
1.005
300.0
-210.0
0.0
0.0
0
90Holston
V2
11
20.985
33.29
78.0
42.0
-85.0
0.0
0.0
0.985
300.0
-300.0
0.0
0.0
0
91HolstonT
V2
11
20.980
33.31
0.0
0.0
-10.0
0.0
0.0
0.980
100.0
-100.0
0.0
0.0
0
92Saltvlle
V2
11
20.993
33.80
65.0
10.0
0.0
0.0
0.0
0.990
9.0
-3.0
0.0
0.0
0
93Tazewell
V2
11
00.987
30.79
12.0
7.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
94Switchbk
V2
11
00.991
28.64
30.0
16.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
95Caldwell
V2
11
00.981
27.67
42.0
31.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
96Baileysv
V2
11
00.993
27.51
38.0
15.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
97Sundial
V2
11
01.011
27.88
15.0
9.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
98Bradley
V2
11
01.024
27.40
34.0
8.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
99Hinton
V2
11
21.010
27.04
0.0
0.0
-42.0
0.0
0.0
1.010
100.0
-100.0
0.0
0.0
0
100GlenLyn
V2
11
21.017
28.03
37.0
18.0
252.0
0.0
0.0
1.017
155.0
-50.0
0.0
0.0
0
101Wythe
V2
11
00.993
29.61
22.0
15.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
102Smythe
V2
11
00.991
32.30
5.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
103Claytor
V2
11
21.001
24.44
23.0
16.0
40.0
0.0
0.0
1.01
40.0
-15.0
0.0
0.0
0
104Hancock
V2
11
20.971
21.69
38.0
25.0
0.0
0.0
0.0
0.971
23.0
-8.0
0.0
0.0
0
105Roanoke
V2
11
20.965
20.57
31.0
26.0
0.0
0.0
0.0
0.965
23.0
-8.0
0.0
0.20
0
106Cloverdl
V2
11
00.962
20.32
43.0
16.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
107Reusens
V2
11
20.952
17.53
28.0
12.0
-22.0
0.0
0.0
0.952
200.0
-200.0
0.0
0.06
0
108Blaine
V2
11
00.967
19.38
2.0
1.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
109Franklin
V2
11
00.967
18.93
8.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
110Fieldale
V2
11
20.973
18.09
39.0
30.0
0.0
0.0
0.0
0.973
23.0
-8.0
0.0
0.06
0
111DanRiver
V2
11
20.980
19.74
0.0
0.0
36.0
0.0
0.0
0.980
1000.0
-100.0
0.0
0.0
0
112Danville
V2
11
20.975
14.99
25.0
13.0
-43.0
0.0
0.0
0.975
1000.0
-100.0
0.0
0.0
0
113DeerCrk
V2
11
20.993
13.74
0.0
0.0
-6.0
0.0
0.0
0.993
200.0
-100.0
0.0
0.0
0
114WMedford
V2
11
00.960
14.46
8.0
3.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
115Medford
V2
11
00.960
14.46
22.0
7.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
116KygerCrk
V2
11
21.005
27.12
0.0
0.0
-184.0
0.0
0.0
1.005
1000.0-1000.0
0.0
0.0
0
117Corey
V2
11
00.974
10.67
20.0
8.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
118WHuntngd
V2
11
00.949
21.92
33.0
15.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0
-999BRANCHDATAFOLLOWS
80ITEMS
12
1110
0.03030
0.09990
0.02540
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
13
1110
0.01290
0.04240
0.01082
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
45
1110
0.00176
0.00798
0.00210
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
35
1110
0.02410
0.10800
0.02840
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
56
1110
0.01190
0.05400
0.01426
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
67
1110
0.00459
0.02080
0.00550
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
89
1110
0.00244
0.03050
1.16200
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
85
1111
0.00000
0.02670
0.0
00
000
0.985
0.00.0
0.0
0.0
0.0
0.0
910
1110
0.00258
0.03220
1.23000
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
411
1110
0.02090
0.06880
0.01748
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
511
1110
0.02030
0.06820
0.01738
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
11
12
1110
0.00595
0.01960
0.00502
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
212
1110
0.01870
0.06160
0.01572
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
312
1110
0.04840
0.16000
0.04060
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
712
1110
0.00862
0.03400
0.00874
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
11
13
1110
0.02225
0.07310
0.01876
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
12
14
1110
0.02150
0.07070
0.01816
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
13
15
1110
0.07440
0.24440
0.06268
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
14
15
1110
0.05950
0.19500
0.05020
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
12
16
1110
0.02120
0.08340
0.02140
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
15
17
1110
0.01320
0.04370
0.04440
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
16
17
1110
0.04540
0.18010
0.04660
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
17
18
1110
0.01230
0.05050
0.01298
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
18
19
1110
0.01119
0.04930
0.01142
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
19
20
1110
0.02520
0.11700
0.02980
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
15
19
1110
0.01200
0.03940
0.01010
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
20
21
1110
0.01830
0.08490
0.02160
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
21
22
1110
0.02090
0.09700
0.02460
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
22
23
1110
0.03420
0.15900
0.04040
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
23
24
1110
0.01350
0.04920
0.04980
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
23
25
1110
0.01560
0.08000
0.08640
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
26
25
1111
0.00000
0.03820
0.0
00
000
0.960
0.00.0
0.0
0.0
0.0
0.0
25
27
1110
0.03180
0.16300
0.17640
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
27
28
1110
0.01913
0.08550
0.02160
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
28
29
1110
0.02370
0.09430
0.02380
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
30
17
1111
0.00000
0.03880
0.0
00
000
0.960
0.00.0
0.0
0.0
0.0
0.0
830
1110
0.00431
0.05040
0.51400
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
26
30
1110
0.00799
0.08600
0.90800
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
17
31
1110
0.04740
0.15630
0.03990
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
29
31
1110
0.01080
0.03310
0.00830
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
23
32
1110
0.03170
0.11530
0.11730
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
31
32
1110
0.02980
0.09850
0.02510
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
27
32
1110
0.02290
0.07550
0.01926
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
15
33
1110
0.03800
0.12440
0.03194
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
19
34
1110
0.07520
0.24700
0.06320
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
35
36
1110
0.00224
0.01020
0.00268
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
35
37
1110
0.01100
0.04970
0.01318
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
33
37
1110
0.04150
0.14200
0.03660
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
34
36
1110
0.00871
0.02680
0.00568
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
34
37
1110
0.00256
0.00940
0.00984
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
38
37
1111
0.00000
0.03750
0.0
00
000
0.935
0.00.0
0.0
0.0
0.0
0.0
37
39
1110
0.03210
0.10600
0.02700
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
37
40
1110
0.05930
0.16800
0.04200
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
30
38
1110
0.00464
0.05400
0.42200
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
39
40
1110
0.01840
0.06050
0.01552
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
40
41
1110
0.01450
0.04870
0.01222
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
40
42
1110
0.05550
0.18300
0.04660
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
41
42
1110
0.04100
0.13500
0.03440
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
43
44
1110
0.06080
0.24540
0.06068
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
34
43
1110
0.04130
0.16810
0.04226
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
44
45
1110
0.02240
0.09010
0.02240
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
45
46
1110
0.04000
0.13560
0.03320
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
46
47
1110
0.03800
0.12700
0.03160
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
46
48
1110
0.06010
0.18900
0.04720
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
47
49
1110
0.01910
0.06250
0.01604
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
42
49
1110
0.07150
0.32300
0.08600
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
42
49
1110
0.07150
0.32300
0.08600
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
45
49
1110
0.06840
0.18600
0.04440
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
48
49
1110
0.01790
0.05050
0.01258
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
49
50
1110
0.02670
0.07520
0.01874
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
49
51
1110
0.04860
0.13700
0.03420
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
51
52
1110
0.02030
0.05880
0.01396
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
52
53
1110
0.04050
0.16350
0.04058
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
53
54
1110
0.02630
0.12200
0.03100
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
49
54
1110
0.07300
0.28900
0.07380
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
49
54
1110
0.08690
0.29100
0.07300
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
54
55
1110
0.01690
0.07070
0.02020
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
54
56
1110
0.00275
0.00955
0.00732
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
55
56
1110
0.00488
0.01510
0.00374
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
56
57
1110
0.03430
0.09660
0.02420
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
50
57
1110
0.04740
0.13400
0.03320
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
56
58
1110
0.03430
0.09660
0.02420
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
51
58
1110
0.02550
0.07190
0.01788
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
54
59
1110
0.05030
0.22930
0.05980
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
56
59
1110
0.08250
0.25100
0.05690
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
56
59
1110
0.08030
0.23900
0.05360
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
55
59
1110
0.04739
0.21580
0.05646
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
59
60
1110
0.03170
0.14500
0.03760
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
59
61
1110
0.03280
0.15000
0.03880
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
60
61
1110
0.00264
0.01350
0.01456
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
60
62
1110
0.01230
0.05610
0.01468
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
61
62
1110
0.00824
0.03760
0.00980
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
63
59
1111
0.00000
0.03860
0.0
00
000
0.960
0.00.0
0.0
0.0
0.0
0.0
63
64
1110
0.00172
0.02000
0.21600
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
64
61
1111
0.00000
0.02680
0.0
00
000
0.985
0.00.0
0.0
0.0
0.0
0.0
38
65
1110
0.00901
0.09860
1.04600
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
64
65
1110
0.00269
0.03020
0.38000
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
49
66
1110
0.01800
0.09190
0.02480
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
49
66
1110
0.01800
0.09190
0.02480
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
62
66
1110
0.04820
0.21800
0.05780
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
62
67
1110
0.02580
0.11700
0.03100
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
65
66
1111
0.00000
0.03700
0.0
00
000
0.935
0.00.0
0.0
0.0
0.0
0.0
66
67
1110
0.02240
0.10150
0.02682
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
65
68
1110
0.00138
0.01600
0.63800
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
47
69
1110
0.08440
0.27780
0.07092
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
49
69
1110
0.09850
0.32400
0.08280
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
68
69
1111
0.00000
0.03700
0.0
00
000
0.935
0.00.0
0.0
0.0
0.0
0.0
69
70
1110
0.03000
0.12700
0.12200
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
24
70
1110
0.00221
0.41150
0.10198
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
70
71
1110
0.00882
0.03550
0.00878
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
24
72
1110
0.04880
0.19600
0.04880
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
71
72
1110
0.04460
0.18000
0.04444
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
71
73
1110
0.00866
0.04540
0.01178
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
70
74
1110
0.04010
0.13230
0.03368
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
70
75
1110
0.04280
0.14100
0.03600
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
69
75
1110
0.04050
0.12200
0.12400
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
74
75
1110
0.01230
0.04060
0.01034
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
76
77
1110
0.04440
0.14800
0.03680
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
69
77
1110
0.03090
0.10100
0.10380
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
75
77
1110
0.06010
0.19990
0.04978
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
77
78
1110
0.00376
0.01240
0.01264
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
78
79
1110
0.00546
0.02440
0.00648
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
77
80
1110
0.01700
0.04850
0.04720
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
77
80
1110
0.02940
0.10500
0.02280
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
79
80
1110
0.01560
0.07040
0.01870
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
68
81
1110
0.00175
0.02020
0.80800
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
81
80
1111
0.00000
0.03700
0.0
00
000
0.935
0.00.0
0.0
0.0
0.0
0.0
77
82
1110
0.02980
0.08530
0.08174
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
82
83
1110
0.01120
0.03665
0.03796
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
83
84
1110
0.06250
0.13200
0.02580
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
83
85
1110
0.04300
0.14800
0.03480
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
84
85
1110
0.03020
0.06410
0.01234
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
85
86
1110
0.03500
0.12300
0.02760
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
86
87
1110
0.02828
0.20740
0.04450
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
85
88
1110
0.02000
0.10200
0.02760
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
85
89
1110
0.02390
0.17300
0.04700
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
88
89
1110
0.01390
0.07120
0.01934
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
89
90
1110
0.05180
0.18800
0.05280
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
89
90
1110
0.02380
0.09970
0.10600
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
90
91
1110
0.02540
0.08360
0.02140
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
89
92
1110
0.00990
0.05050
0.05480
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
89
92
1110
0.03930
0.15810
0.04140
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
91
92
1110
0.03870
0.12720
0.03268
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
92
93
1110
0.02580
0.08480
0.02180
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
92
94
1110
0.04810
0.15800
0.04060
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
93
94
1110
0.02230
0.07320
0.01876
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
94
95
1110
0.01320
0.04340
0.01110
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
80
96
1110
0.03560
0.18200
0.04940
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
82
96
1110
0.01620
0.05300
0.05440
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
94
96
1110
0.02690
0.08690
0.02300
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
80
97
1110
0.01830
0.09340
0.02540
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
80
98
1110
0.02380
0.10800
0.02860
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
80
99
1110
0.04540
0.20600
0.05460
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
92
100
1110
0.06480
0.29500
0.04720
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
94
100
1110
0.01780
0.05800
0.06040
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
95
96
1110
0.01710
0.05470
0.01474
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
96
97
1110
0.01730
0.08850
0.02400
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
98
100
1110
0.03970
0.17900
0.04760
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
99
100
1110
0.01800
0.08130
0.02160
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
100
101
1110
0.02770
0.12620
0.03280
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
92
102
1110
0.01230
0.05590
0.01464
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
101
102
1110
0.02460
0.11200
0.02940
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
100
103
1110
0.01600
0.05250
0.05360
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
100
104
1110
0.04510
0.20400
0.05410
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
103
104
1110
0.04660
0.15840
0.04070
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
103
105
1110
0.05350
0.16250
0.04080
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
100
106
1110
0.06050
0.22900
0.06200
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
104
105
1110
0.00994
0.03780
0.00986
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
105
106
1110
0.01400
0.05470
0.01434
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
105
107
1110
0.05300
0.18300
0.04720
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
105
108
1110
0.02610
0.07030
0.01844
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
106
107
1110
0.05300
0.18300
0.04720
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
108
109
1110
0.01050
0.02880
0.00760
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
103
110
1110
0.03906
0.18130
0.04610
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
109
110
1110
0.02780
0.07620
0.02020
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
110
111
1110
0.02200
0.07550
0.02000
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
110
112
1110
0.02470
0.06400
0.06200
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
17
113
1110
0.00913
0.03010
0.00768
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
32
113
1110
0.06150
0.20300
0.05180
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
32
114
1110
0.01350
0.06120
0.01628
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
27
115
1110
0.01640
0.07410
0.01972
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
114
115
1110
0.00230
0.01040
0.00276
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
68
116
1110
0.00034
0.00405
0.16400
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
12
117
1110
0.03290
0.14000
0.03580
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
75
118
1110
0.01450
0.04810
0.01198
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
76
118
1110
0.01640
0.05440
0.01356
00
000
0.0
0.00.0
0.0
0.0
0.0
0.0
-999LOSSZONESFOLLOWS
1ITEMS
1IEEE118BUS
-99INTERCHANGEDATAFOLLOWS
1ITEMS
180CabinCrk
V2
0.0
999.99
IEEE118IEEE118BusTestCase
-9TIELINESFOLLOWS
0ITEMS-999
ENDOFDATA
Bibliography
[1] A. Abur and A. G. Exposito, Power System State Estimation: Theory and
Implementation. New York: Marcel Dekker, 2004.
[2] A. G. Phadke and J. S. Thorp, Synchronized Phasor Measurements and Their
Applications. New York: Springer, 2008.
[3] A. G. Phadke, J. S. Thorp, and M. G. Adamiak, “A new measurement tech-
nique for tracking voltage phasors, local system frequency, and rate of change
of frequency,” IEEE Transactions on Power Apparatus and Systems, vol. 102,
no. 5, pp. 1025–1038, May 1983.
[4] A. G. Phadke and B. Kasztenny, “Synchronized phasor and frequency mea-
surement under transient conditions,” IEEE Transactions on Power Delivery,
2009.
[5] B. Milosevic and M. Begovic, “Voltage-stability protection and control using
a wide-area network of phasor measurements,” IEEE Transactions on Power
Systems, vol. 18, no. 1, pp. 121–127, Feb. 2003.
[6] M. M. Eissa, M. E. Masoud, and M. M. M. Elanwar, “A novel back up wide area
protection technique for power transmission grids using phasor measurement
unit,” IEEE Transactions on Power Delivery, vol. 25, no. 1, pp. 270–278, Jan.
2010.
[7] J. Chen and A. Abur, “Enhanced topology error processing via optimal mea-
surement design,” IEEE Transactions on Power Systems, vol. 23, no. 3, pp.
845–852, Aug. 2008.
94
Bibliography 95
[8] R. Diao, K. Sun, V. Vittal, R. J. O’Keefe, M. R. Richardson, N. Bhatt, D. Strad-
ford, and S. K. Sarawgi, “Decision tree-based online voltage security assessment
using PMU measurements,” IEEE Transactions on Power Systems, vol. 24,
no. 2, pp. 832–839, May 2009.
[9] Y. Zhang, P. Markham, T. Xia, L. Chen, Y. Ye, Z. Wu, Z. Yuan, L. Wang,
J. Bank, J. Burgett, R. W. Conners, and Y. Liu, “Wide-area frequency moni-
toring network (FNET) architecture and applications,” IEEE Transactions on
Smart Grid, vol. 1, no. 2, pp. 159–167, Sep. 2010.
[10] J. E. Tate and T. J. Overbye, “Line outage detection using phasor angle mea-
surements,” IEEE Transactions on Power Systems, vol. 23, no. 4, pp. 1644–
1652, Nov. 2008.
[11] Y.-H. Lin, C.-W. Liu, and C.-S. Chen, “A new PMU-based fault detec-
tion/location technique for transmission lines with consideration of arcing fault
discrimination-part I: theory and algorithms,” IEEE Transactions on Power
Delivery, vol. 19, no. 4, pp. 1587–1593, Oct. 2004.
[12] Y.-H. Lin, C.-W. Liu, and C.-S. Chen, “A new PMU-based fault detec-
tion/location technique for transmission lines with consideration of arcing fault
discrimination-part II: performance evaluation,” IEEE Transactions on Power
Delivery, vol. 19, no. 4, pp. 1594–1601, Oct. 2004.
[13] J. H. Chow, A. Chakrabortty, M. Arcak, B. Bhargava, and A. Salazar, “Synchro-
nized phasor data based energy function analysis of dominant power transfer
paths in large power systems,” IEEE Transactions on Power Systems, vol. 22,
no. 2, pp. 727–734, May 2007.
[14] M. Glavic and T. Van Cutsem, “Wide-area detection of voltage instability from
synchronized phasor measurements. part I: Principle,” IEEE Transactions on
Power Systems, vol. 24, no. 3, pp. 1408–1416, Aug. 2009.
[15] M. Glavic and T. Van Cutsem, “Wide-area detection of voltage instability from
synchronized phasor measurements. part II: Simulation results,” IEEE Trans-
actions on Power Systems, vol. 24, no. 3, pp. 1417–1425, Aug. 2009.
Bibliography 96
[16] C.-W. Liu, K.-P. Lien, C.-S. Chen, and J.-A. Jiang, “A universal fault location
technique for N-terminal (N = 3) transmission lines,” IEEE Transactions on
Power Delivery, vol. 23, no. 3, pp. 1366–1373, Jul. 2008.
[17] P. Tripathy, S. C. Srivastava, and S. N. Singh, “A divide-by-difference-filter
based algorithm for estimation of generator rotor angle utilizing synchropha-
sor measurements,” IEEE Transactions on Instrumentation and Measurement,
vol. 59, no. 6, pp. 1562–1570, Jun. 2010.
[18] K. A. Clements and W. A. Wollenberg, “An algorithm for observability determi-
nation in power system state estimation,” in IEEE Power Engineering Society
Summer Meeting, 1975.
[19] F. J. Marın, F. Garcıa-Lagos, G. Joya, and F. Sandoval, “Genetic algorithms
for optimal placement of phasor measurement units in electrical networks,”
Electronics Letters, vol. 39, no. 19, pp. 1403–1405, Sep. 2003.
[20] A. G. Phadke, J. S. Thorp, and K. J. Karimi, “State estimation with phasor
measurements,” IEEE Transactions on Power Systems, vol. 1, no. 1, pp. 233–
238, Feb. 1986.
[21] A. G. Phadke, “Synchronized phasor measurements in power systems,” IEEE
Computer Applications in Power, vol. 6, no. 2, pp. 10–15, Apr. 1993.
[22] L. Mili, T. Baldwin, and R. Adapa, “Phasor measurement placement for voltage
stability analysis of power systems,” in Proceedings of the 29th IEEE Conference
on Decision and Control, vol. 6, Dec. 5–7, 1990, pp. 3033–3038.
[23] T. L. Baldwin, L. Mili, M. B. Boisen, and R. Adapa, “Power system observ-
ability with minimal phasor measurement placement,” IEEE Transactions on
Power Systems, vol. 8, no. 2, pp. 707–715, May 1993.
[24] R. F. Nuqui and A. G. Phadke, “Phasor measurement unit placement based on
incomplete observability,” in IEEE 2002 Power Engineering Society Summer
Meeting, vol. 2, Jul. 25, 2002, pp. 888–893.
Bibliography 97
[25] R. F. Nuqui and A. G. Phadke, “Phasor measurement unit placement techniques
for complete and incomplete observability,” IEEE Transactions on Power De-
livery, vol. 20, no. 4, pp. 2381–2388, Oct. 2005.
[26] B. Milosevic and M. Begovic, “Nondominated sorting genetic algorithm for op-
timal phasor measurement placement,” IEEE Transactions on Power Systems,
vol. 18, no. 1, pp. 69–75, Feb. 2003.
[27] K.-S. Cho, J.-R. Shin, and S. H. Hyun, “Optimal placement of phasor measure-
ment units with GPS receiver,” in IEEE Power Engineering Society Winter
Meeting, vol. 1, Jan.-Feb. 2001, pp. 258–262.
[28] J. Peng, Y. Sun, and H.-F. Wang, “Optimal PMU placement for full network
observability using Tabu search algorithm,” International Journal of Electrical
Power & Energy Systems, pp. 223–231, May 2006.
[29] F. Aminifar, C. Lucas, A. Khodaei, and M. Fotuhi-Firuzabad, “Optimal place-
ment of phasor measurement units using immunity genetic algorithm,” IEEE
Transactions on Power Delivery, vol. 24, no. 3, pp. 1014–1020, Jul. 2009.
[30] S. Chakrabarti, G. K. Venayagamoorthy, and E. Kyriakides, “PMU placement
for power system observability using binary particle swarm optimization,” in
Australasian Universities Power Engineering Conference (AUPEC), Dec. 2008.
[31] M. Hajian, A. M. Ranjbar, T. Amraee, and A. R. Shirani, “Optimal placement
of phasor measurement units: particle swarm optimization approach,” in In-
ternational Conference on Intelligent Systems Applications to Power Systems
(ISAP), Nov. 5–8, 2007.
[32] A. Sadu, R. Kumar, and R. G. Kavasseri, “Optimal placement of phasor mea-
surement units using particle swarm optimization,” in 2009 World Congress
on Nature & Biologically Inspired Computing (NaBIC), Dec. 9–11, 2009, pp.
1708–1713.
[33] S. Chakrabarti and E. Kyriakides, “Optimal placement of phasor measurement
units for power system observability,” IEEE Transactions on Power Systems,
vol. 23, no. 3, pp. 1433–1440, Aug. 2008.
Bibliography 98
[34] B. Xu and A. Abur, “Observability analysis and measurement placement for
systems with PMUs,” in Proceedings of the IEEE PES 2004 Power Systems
Conference and Exposition, vol. 2, Oct. 10–13, 2004, pp. 943–946.
[35] B. Xu, “Optimal monitoring and visualization of steady-state power system
operation,” Ph.D. dissertation, Texas A&M University, May 2006.
[36] B. Gou, “Optimal placement of PMUs by integer linear programming,” IEEE
Transactions on Power Systems, vol. 23, no. 3, pp. 1525–1526, Aug. 2008.
[37] B. Gou, “Generalized integer linear programming formulation for optimal PMU
placement,” IEEE Transactions on Power Systems, vol. 23, no. 3, pp. 1099–
1104, Aug. 2008.
[38] A. Abur and F. H. Magnago, “Optimal meter placement for maintaining observ-
ability during single branch outages,” IEEE Transactions on Power Systems,
vol. 14, no. 4, pp. 1273–1278, Nov. 1999.
[39] B. Xu, Y. J. Yoon, and A. Abur, “Optimal placement and utilization of phasor
measurements for state estimation,” in Power System Computation Conference,
Liege, Belgium, Aug. 2005.
[40] S. Chakrabarti, D. Eliades, E. Kyriakides, and M. Albu, “Measurement un-
certainty considerations in optimal sensor deployment for state estimation,” in
IEEE International Symposium on Intelligent Signal Processing (WISP), Oct.
2007.
[41] S. Chakrabarti, E. Kyriakides, and D. G. Eliades, “Placement of synchronized
measurements for power system observability,” IEEE Transactions on Power
Delivery, vol. 24, no. 1, pp. 12–19, Jan. 2009.
[42] D. Dua, S. Dambhare, R. K. Gajbhiye, and S. A. Soman, “Optimal multistage
scheduling of PMU placement: an ILP approach,” IEEE Transactions on Power
Delivery, vol. 23, no. 4, pp. 1812–1820, Oct. 2008.
Bibliography 99
[43] N. H. Abbasy and H. M. Ismail, “A unified approach for the optimal PMU loca-
tion for power system state estimation,” IEEE Transactions on Power Systems,
vol. 24, no. 2, pp. 806–813, May 2009.
[44] F. Aminifar, A. Khodaei, M. Fotuhi-Firuzabad, and M. Shahidehpour,
“Contingency-constrained PMU placement in power networks,” IEEE Trans-
actions on Power Systems, vol. 25, no. 1, pp. 516–523, Feb. 2010.
[45] J. Chen and A. Abur, “Improved bad data processing via strategic placement
of PMUs,” in IEEE 2005 Power Engineering Society General Meeting, vol. 1,
Jun. 12–16, 2005, pp. 509–513.
[46] J. Chen and A. Abur, “Placement of PMUs to enable bad data detection in
state estimation,” IEEE Transactions on Power Systems, vol. 21, no. 4, pp.
1608–1615, Nov. 2006.
[47] Y. M. Park, Y. H. Moon, J. B. Choo, and T. W. Kwon, “Design of reliable mea-
surement system for state estimation,” IEEE Transactions on Power Systems,
vol. 3, no. 3, pp. 830–836, Aug. 1988.
[48] F. H. Magnago and A. Abur, “A unified approach to robust meter place-
ment against loss of measurements and branch outages,” IEEE Transactions
on Power Systems, vol. 15, no. 3, pp. 945–949, Aug. 2000.
[49] C. Rakpenthai, S. Premrudeepreechacharn, S. Uatrongjit, and N. R. Watson,
“An optimal PMU placement method against measurement loss and branch
outage,” IEEE Transactions on Power Delivery, vol. 22, no. 1, pp. 101–107,
Jan. 2007.
[50] R. Emami and A. Abur, “Reliable placement of synchronized phasor measure-
ments on network branches,” in IEEE PES 2009 Power Systems Conference
and Exposition (PSCE), Mar. 15–18, 2009.
[51] R. Emami and A. Abur, “Robust measurement design by placing synchronized
phasor measurements on network branches,” IEEE Transactions on Power Sys-
tems, vol. 25, no. 1, pp. 38–43, Feb. 2010.
Bibliography 100
[52] K.-P. Lien, C.-W. Liu, C.-S. Yu, and J.-A. Jiang, “Transmission network fault
location observability with minimal PMU placement,” IEEE Transactions on
Power Delivery, vol. 21, no. 3, pp. 1128–1136, Jul. 2006.
[53] S. P. Pokharel and S. Brahma, “Optimal PMU placement for fault location in a
power system,” in 2009 North American Power Symposium (NAPS), Oct. 4–6,
2009.
[54] F. Mahmoodianfard, M. Mohammadi, G. B. Gharehpetian, and H. A. Abyaneh,
“Optimal PMU placement for voltage security assessment using decision tree,”
in 2009 IEEE PowerTech, Jun. 28–Jul. 2, 2009.
[55] M. Zhou, V. A. Centeno, A. G. Phadke, Y. Hu, D. Novosel, and H. A. R. Vol-
skis, “A preprocessing method for effective PMU placement studies,” in Third
International Conference on Electric Utility Deregulation and Restructuring and
Power Technologies (DRPT), Apr. 2008, pp. 2862–2867.
[56] M. Hurtgen, P. Praks, J.-C. Maun, and P. Zajac, “Measurement placement
algorithms for state estimation when the number of phasor measurements by
each PMU is limited,” in 43rd International Universities Power Engineering
Conference (UPEC), Sep. 2008.
[57] J. Zhang and W. K. G. Seah, “Topology-based capacity analysis for ad hoc
networks with end-to-end delay constraints,” in Proceedings of the IEEE 6th
Circuits and Systems Symposium on Emerging Technologies: Frontiers of Mo-
bile and Wireless Communication, vol. 2, May 31–Jun. 2, 2004, pp. 541–544.
[58] M. Korkalı and A. Abur, “Placement of PMUs with channel limits,” in Pro-
ceedings of the IEEE PES 2009 General Meeting, Jul. 26–30, 2009.
[59] K. Holmstrom, A. O. Goran, and M. M. Edvall, User’s Guide for TOMLAB
/CPLEX v11.2, Nov. 2008.
[60] R. D. Christie, “Power systems test case archive,” Aug. 1999. [Online]. Avail-
able: http://www.ee.washington.edu/research/pstca.