Beruflich Dokumente
Kultur Dokumente
l / 0 7 1
TEKNOLOGI
UNIVERSITI MALAYSIA
OF THESIS
DECTARATION PROJECT
/ UNDERGRADUATE PAPERAND COPYRIGHT
Author'sfullnome LINGTINGYI
Title ANALYSISUTITIZING
VOTTAGESECURITY VOLTAGECOLTAPSE
INDICATOR
PROXIMITY IN POWER SYSTEM
Certifiedby:
D ote : tl MA Y2Ol1 Do t e : l2 MA Y2 o tI
Signature
LING TING YI
MAY 2011
DECLARATION
"It is hereby declared that all the materials in this thesis are the effort of my own work
and idea except for works that have been cited clearly in the references. The thesis has
not been accepted for any degree and is not concurrently subrnitted in candidature ofany
degree."
ACKNOWLEDGMENT
ABSTRACT
ABSTRAK
TABLE OF CONTENTS
1 INTRODUCTION 1
1.1 Introduction 1
1.2 Background of the Study 1
1.3 Problem Statement 2
1.4 Objectives of the Study 3
1.5 Scope of the Study 3
1.6 Significance of the Study 4
1.7 Thesis Organization 4
viii
2 LITERATURE REVIEW 6
2.1 Introduction 6
2.2 Power System Security 6
2.3 Voltage Security Assessment 8
2.4 Relative Voltage-Change Method (VC) 10
2.4 Continuation Power Flow (CPF) 11
2.4 Voltage Collapse Proximity Indicator (VCPI) 12
2.4 Summary 14
3 RESEARCH METHODOLOGY 15
3.1 Introduction 15
3.2 Research Procedure 15
3.2.1 Critical Loading Condition 16
3.2.2 Relative Voltage-Change Method 17
3.2.3 Continuation Power Flow 18
3.2.4 Voltage Collapse Proximity Indicator 19
3.3 Research Instruments 20
3.3.1 MATPOWER Software 21
3.4 Data Analysis 27
3.4.1 PV Curve 27
3.4.2 Microsoft Office Excel 2007 SP2 27
3.5 Summary 28
REFERENCES 52
APPENDICES A E 55-70
x
LIST OF TABLES
LIST OF FIGURES
LIST OF ABBREVIATIONS
LIST OF APPENDICES
CHAPTER 1
INTRODUCTION
1.1 Introduction
Thus the power system is expected to be more heavily loaded from day to day.
However, many environmental and economic constraints preventing the constructions of
new or upgrading power system. The power producing utilities such as Tenaga Nasional
Berhad (TNB) shows reluctance to expand their generation and transmission capabilities
due to social pressure, such as the public concerns about the effect of electric and
magnetic field around the housing area near to transmission lines.
All the constraints lead the current power system to operate closely to stability
limits, causing loss of control of the voltage levels in a power system. Usually the
voltage decay is gradual that makes system operators unaware that it is the symptom of
voltage collapse which will lead to complete blackout. Therefore, constant attention is
required to ensure the systems are operated above desired level of voltage stability
margin.
Voltage Stability Assessment (VSA) is the process to ensure the voltage security
in power system. There are two main scope of the assessment, which is static security
and dynamic security of system. One of the various steps of carrying out the assessment
is contingency screening and ranking of weak load buses. Numerous ways were
developed by researchers and industries to indicate the weakest bus in the power system.
This step is vital to make preventive maintenance before the voltage collapse happen.
ii. Static study is appropriate for bulk power system study, which involves
enormous number of buses and generators;
iii. Static voltage instability is most affected by reactive power imbalance.
i. The PQ-buses with zero loads are assumed to be of zero loads throughout the
analysis;
ii. The parameters of the PV-buses, i.e. bus voltage magnitude and injected power,
P are assumed to be constant throughout the analysis;
iii. The slack bus is capable to absorb the losses in the system.
There is limitation in the project. Only load with sufficient reactive power will be
considered as the proposed VCPI is considering reactive power solely.
Although there are various ways of contingency screening and ranking in the
research field, the findings of this study are important as part of Voltage Security
Assessment. Critical load bus identification is vital to ensure the priority of preventive
and correction action is given to the most critical load bus in the power system.
Chapter 4 presents the findings and results obtained from the project. The data
are analyzed and the three critical bus rankings are compared and discussed.
Chapter 5 discusses the conclusion of the project, and suggestions for further
extension on the current work.
6
CHAPTER 2
LITERATURE REVIEW
2.1 Introduction
In this chapter, literature review starts with screening the concept of power
system security, which is the big picture of the research. The research is part of the effort
to ensure power system to operate without interruption of supply to the consumers in the
same time it can withstand credible contingencies. Next the voltage stability analysis
procedure is examined and the scope steep down to static security assessment. Finally
two benchmark methods and the proposed method are reviewed from the original
authors, and the reviews of some other methods are presented as well.
condition for every bus in the system, the bus voltage magnitude increases as the
reactive power injection at the same bus is increased. A system is voltage unstable if, for
at least one bus in the system, the bus voltage magnitude (V) decreases as the reactive
power injection (Q) at the same bus is increased. [3]
There are some other important terms to understand for this research, which is
defined in [6]:
8
i. Voltage stability - the ability of a system to maintain voltage magnitude at all the
buses in the system after disturbance;
ii. Voltage collapse - a process by which voltage instability leads to a very low
voltage;
iii. Voltage security - the ability of a system not only to operate stably but also to
remain stable following credible contingencies or adverse system changes.
a. Where the voltage collapse occurs for any equipment outage or operating
change,
b. All contingencies and operating changes that cause voltage collapse in that
location (a specific sub region in the transmission, sub transmission, or
distribution network),
c. The cause of the voltage collapse in terms of
Figure 2.1 is the modified flowchart from [8] which represents the voltage
stability assessment. The VSA environment receives its input from a real time database.
Voltage stability assessment of the current operating point is necessary to enable the
system engineer know the voltage stability status of the system. The outcome of this
assessment determines the next line of action. If the result of the assessment is positive,
i.e. the system is secured at the present operating point, the next step would be to initiate
some credible contingencies, such as line outages and critical loading conditions, which
would be analyzed further. The large list of contingencies is screened and ranked with
respect to their margins to voltage collapse, using any fast and accurate ranking
algorithm available. Finally, the contingencies flagged as potentially harmful to the
systems stability are investigated further using tools like continuation power flow (CPF)
and consequently develop some control schemes to be executed in either a pre-
contingency or post-contingency mode. [9]
In this research, the focus is on the dotted box in the flowchart, which is
contingency selection, screening and ranking of potential harmful load bus in power
system.
System dynamics influencing voltage stability are usually slow. Therefore many
aspects of the problem can be effectively analyzed by static methods, which examine the
viability of the equilibrium point represented by a specified operating condition of the
power system. The static analysis techniques allow examination of a wide range of
system conditions and, if appropriately used, can provide much insight into the nature of
the problem and identify the key contributing factors. Dynamic analysis, on the other
hand, is useful for detailed study of specific voltage collapse situations, coordination of
protection and controls, and testing of remedial measures. Dynamic simulations also
examine whether and how the steady state equilibrium point will be reached. [3]
Introduced by [10] and adopted by [2], the first benchmark method is based on
the relative change in the bus voltages going from the initial operating point to the
voltage stability limit.
(2.1)
As mentioned previously, the weak or critical bus in the network is the most
(electrically) remote bus from the point of constant or controllable voltage. It is expected
that the critical bus would be the worst affected (voltage wise) because of a shortage of
11
local VARs or VARs transferred from a remote source. [10] It is anticipated that for a
specified operating regime, going from an initial operating point to the voltage stability
limit, the weakest bus would experience the largest voltage change (or drop), i.e., the
largest index VC, defined by eqn. 14. Therefore if bus k is the weakest bus,
VCk max{VCi }
iJ L
(2.2)
Based on the index VCi, the system buses may be arranged in order of weakness,
the weakest bus corresponding to that with the largest index.
Fig. 2.1 presents the PV curves of the power flow solution when generator limits
are neglected. Any attempt to increase PR (QR) beyond point A in the figure would result
in a system voltage collapse. The maximum loading points are depicted in the figure
12
with A and C. Different PV and QV curves can be computed based on the system
parameters chosen to do so. Each curve shows the maximum power that can be
transferred at a particular power factor. [9]
index L1 was adopted as a fast and accurate tool to trace the SNB point, regardless of
the type of load model. This takes care of the limitations of the index-L proposed by
Kessel [21] that is only suitable for constant power type of load. Fast curve fitting
method was proposed by Ejebe et al [8] to calculate the limit of the nose curve.
(2.3)
VCPI Qk max{VCPI i }
i L
(2.4)
14
Based on the index VCi, the system buses may be arranged in order of weakness,
the weakest bus corresponding to that with the largest index.
2.8 Summary
This chapter has presented the related knowledge of the research and numerous
of past works which were developed by other researchers. It starts from power system
security concept, and go deep to voltage stability assessment overview, until the
benchmark methods and proposed method as well as other methods of developing VCPI.
15
CHAPTER 3
RESEARCH METHODOLOGY
3.1 Introduction
This section discusses the methodology used to archive the objective of the
research. Three procedures of implementing the indices were explained, namely VC,
CPF and VCPI. Before that another additional procedure is the pre-requisite of acquiring
VC and VCPI ranking, which is obtaining the critical loading condition. The main
research software, MATPOWER and Microsoft Office Excel 2007 SP2 were discussed.
Finally data analysis was done utilizing PV curve.
Before the indices VC and VCPI could be calculated, critical loading condition
must be obtained to ensure the weakest load bus was exposed for identification. When
the power system is stressed to heavy load state, the voltage magnitude will drop and
reactive power shortage will appear. In this research, 10% margin is adopted as one of
the test systems exceeds 5% tolerance at basecase. Undervoltage will cause voltage
instability and malfunction of electrical equipments, inducing economic losses
especially for heavy industries.
To obtain the critical loading condition for a power system, firstly a load flow
analysis is ran at basecase, and the voltage magnitude on each load bus is recorded. Next,
the load of the power system is increased linearly at constant power factor, on a step of
10%. Constant power factor load increment is done by increasing both real and reactive
power at the same time with same step size. To obtain more precise of critical loading
condition, a smaller step size such as 5% or 1% can be adopted. The increment is
continued with the record of load bus voltage magnitudes until the magnitudes exceeded
the specified range of voltage margin, which is 0.9 p.u. to 1.1 p.u. in this research.
Therefore the last increment of the load before the voltage magnitude exceeded the limit
will be the critical loading condition of the power system. Figure 3.1 shows the
procedure of obtaining critical loading condition for this research in flowchart.
17
END
VC is calculated based on the relative voltage change during initial and critical
loading condition of the power system. The weakest bus will be experiencing largest
voltage drop; hence will produce the largest index. Firstly the critical loading condition
is obtained for the power system. Next, from the load flow results of basecase and
critical loading condition, the voltage magnitude for each load bus is tabulated. After
that, VC index is calculated according to the formula in Chapter 2. When all the indices
are obtained for each load bus, they are ranked from highest to lowest value, indicating
the weakest to strongest bus in the system. Figure 3.2 shows the procedure of obtaining
critical bus ranking utilizing VC for this research in flowchart.
18
END
CPF method is a graphical method, plotting the PV curve and acquires the data
from the voltage collapse point. Firstly, PV curve is plotted for each load bus using
MATPOWER, the primary research software which will be discussed in the latter
section. Next, values for the voltage and real power magnitude at the voltage collapse
point are tabulated. The ranking is done by examine the real power value from lowest to
highest, indicating the lowest power handling bus as the weakest bus. Higher the value
of real power at voltage collapse point, stronger the load bus. Figure 3.3 shows the
procedure of obtaining critical bus ranking utilizing CPF for this research in flowchart.
19
Figure 3.3: Flowchart for methodology of critical bus ranking utilizing CPF
VCPI is based on the reactive power compensation of the power system. When a
load bus is having small increment of reactive power, other generation buses will
compensate the load increment by generating more reactive power. The load bus that
needs more reactive power compensation from the generation will be indicated as the
weakest bus as it will cause reactive power shortage in the system more likely than other
buses with the same increment of reactive power loading. Therefore the weakest bus will
be taking largest reactive power compensation; hence will produce the largest index.
Critical loading condition is implemented to ensure the system was critically stressed
and this will amplify the effect of reactive power shortage, giving larger value of VCPI
index.
Firstly the critical loading condition is obtained for the power system, same as
the methodology for VC. Next, the reactive power for specific load bus is increased by a
20
small increment, in this research it is simplified by value one (1). The load flow is rerun
after the increment to monitor the additional reactive power generated by the generation
buses. The increment of reactive power of each generation bus is summed up as one of
the parameter for the calculation of VCPI. Then VCPI index is calculated according to
the formula mentioned in Chapter 2. When all the indices are obtained for each load bus,
they are ranked from highest to lowest value, indicating the weakest to strongest bus in
the system. Figure 3.4 shows the procedure of obtaining critical bus ranking utilizing
VCPI for this research in flowchart.
Figure 3.4: Flowchart for methodology of critical bus ranking utilizing VCPI
The main research software used in the research was MATPOWER. It is a third
party freeware MATLAB power system simulation package, including several M-files
for solving power flow and optimal power flow problems. The latest version for
MATPOWER is Version 4.0b4, 21-May-2010. Data analysis was done using
21
3.3.1 MATPOWER
The input data for the case to be simulated are specified in a set of data matrices
packaged as the fields of a MATLAB struct, referred to as a MATPOWER case struct
and conventionally denoted by the variable mpc. This struct is typically defined in a case
file, either a function M-file whose return value is the mpc struct or a MAT-file that
defines a variable named mpc when loaded. The main simulation routines, whose names
begin with run (e.g. runpf, runopf), accept either a file name or a MATPOWER case
struct as an input. Use loadcase to load the data from a case file into a struct if
modifications need to be made to the data before passing it to the simulation. [12]
22
loadcase is used to load the data from a case file into a struct if modifications
need to be made to the data before passing it to the simulation. To load the IEEE 14-bus
test system, defined in case14.m M-file into the mpc variable, the following function can
be entered:
The solver is invoked by calling one of the main simulation functions, such as
runpf, passing in a case file name or a case struct as the first argument [12]. To run a
Newton power flow with default options on the 14-bus system, the following function
can be entered at the MATLAB prompt:
Figure 3.5 to 3.7 shows the results of AC power flow results when command
runpf(case14) was entered in MATPOWER:
System summary, bus data, and branch data are displayed. The bus data includes
the voltage, angle and total generation and load at each bus. The branch data shows the
flows and losses in each branch. From the minimum and maximum voltage magnitude
printed in Figure 3.1, it is used in the research for determination of critical loading
condition while bus voltage magnitude in Figure 3.2 is used for VC computation.
On the other hand, real and reactive power demand can be modified to suit the
research need. To load the IEEE 30-bus test system data denoted from case30.m,
increase its real power demand at bus 2 to 30 MW, then run a Newton power flow with
default options, this could be accomplished as follows:
>> define_constants;
>> mpc = loadcase('case30');
>> mpc.bus(2, PD) = 30;
>> runpf(mpc);
The define constants in the first line is simply a convenience script that defines a
number of variables to serve as named column indices for the data matrices. In this
example, it allows us to access the real power demand column of the bus matrix using
the name PD without having to remember that it is the 3rd column [12]. Another
variable used in the research is reactive power demand, which is denoted as QD.
For realization of CPF, continuation power flow code contributed by Rui Bo and
implemented in MATPOWER is used. Implementation of continuous power flow solver
allows the plot of PV curve as well as the prediction-correction trajectory [12]. A
MATLAB M-file test_cpf as the test program for CPF is a PV curve plotter for IEEE 30-
bus test system with respect to load at bus 7. The program can be simply run by typing
test_cpf in the command window. The full code can be obtained in Appendix B.
25
To suit the research need, the program is modified in order to change the case
file to be analyzed. To analyze 14-bus system, line 34 is modified:
>>casename=(case14);
30-bus test system can be implemented by changing case14 into case30 that
represented 30-Bus data. Currently, continuous power flow with respect to demand
being provided to one bus only. So, only one graph for one bus can be drawn at a time.
The number of bus to be analyzed can be simply done by changing the next line:
>>loadvarloc=4
Figure 3.8 shows the PV curve for 14-bus system, with respect to bus 4. It is
significant to ensure which buses are critical in this project. PV curves were used to
determine system load handling capability. System performance can be shown for
various types of contingencies. In addition, the curves reflects how much load can be
served at minimum operating voltage level and the contingencies combination that lead
to system voltage collapse. The voltage and power limit for the specific bus can be
determined. For this research, CPF serves as a graphical method to obtain critical bus
ranking for the test power system.
26
>>tic
>>runpf('case14');
>>toc
To analyze the data, various ways are adopted including graphical method and
tables.
The data from the power flow results are transcribed and analyzed and tabulated
in tables using spreadsheet software. The results present through tables. Table is the best
way to show the ranking of a series of data. In this research, the main purpose is to
produce weak bus ranking in power system network, thus table is the most effective way.
Ranking is done by arranging the indices ascending. Comparison table is tabulated for
clearer judgement in terms of accuracy and deviation of results.
3.4.1 PV Curve
short time is very helpful in the research. The implementation can be referred from the
attached CD to the thesis.
3.5 Summary
This research proposes two benchmarks, which are VC and CPF, and one
simplified VCPI to examine the ranking of critical bus in power system. As mention
earlier in the introduction, the purpose of this study is to develop an indicator to perform
contingency screening and ranking as part of the voltage security assessment on standard
IEEE test system using MATLAB language, as well as compare the results obtained
from the proposed VCPI to two benchmark results. The research instruments that the
researchers are going to use are MATPOWER and Microsoft Office Excel 2007 SP2.
Then, researcher performs a data analysis base on the results in the form of table and PV
curve.
29
CHAPTER 4
4.1 Introduction
This section presents the results of critical bus ranking tested on IEEE 14-bus test
system and IEEE 30-bus test system. Two benchmark results has been adopted, which
are relative voltage-change method (VC) and continuation power flow (CPF). For CPF,
the results are tabulated using data obtained from PV curves plotted on each load bus in
test systems while for VC, the relative change of bus voltage magnitude between initial
state and critical state are recorded. The actual results are computed by proposed VCPI
utilizing the reactive power compensation for small increase on each load bus. Both
results are compared in terms of accuracy and computation time.
Results for the benchmark results, VC and CPF as well as the proposed method,
VCPI are presented, and analysis is done in terms of accuracy and computation time.
30
This project adopts IEEE 14-bus system which is part of American Electric
Power System at February 1962 as shown in Figure 4.1. This power system network
consists of 14 buses with five machines and 11 loads. There is no line limit for 14-bus
system, but it has low base voltages and an overabundance of voltage control capability.
Full data of the test system can be referred at Appendix A section.
Table 4.1: Determination of critical loading condition for IEEE 14-bus system
% of increment 0% 10% 20% 170% 180% 190%
Minimum 1.010 1.010 1.010 0.918 0.909 0.899
voltage p.u. @ p.u. @ p.u. @ p.u. @ p.u. @ p.u. @
magnitude bus 3 bus 3 bus 3 bus 14 bus 14 bus 14
Maximum 1.090 1.090 1.090 1.090 1.090 1.090
voltage p.u. @ p.u. @ p.u. @ p.u. @ p.u. @ p.u. @
magnitude bus 8 bus 8 bus 8 bus 8 bus 8 bus 8
Benchmark results consist of VC and CPF. The calculation of the indices and the
critical bus ranking are shown.
Table 4.2 shows the calculation of VC on IEEE 14-bus system during critical
loading condition. For load bus 1, 2, 3, 6, 8, the VC index appears as nil due to their
generation bus or P-V bus characteristics, which will maintain their voltage magnitude
in spite of load change. Bus 7 is not considered for the ranking as it is not a load bus,
containing no load data for active and reactive power.
32
Table 4.2: Calculation of VCi on heavy load state (IEEE 14-bus system)
Load bus
1 1.060 1.060 0
2 1.045 1.045 0
3 1.010 1.010 0
4 1.018 0.933 0.091104
5 1.020 0.934 0.092077
6 1.070 1.070 0
7* 1.062 0.984 0.079268
8 1.090 1.090 0
9 1.056 0.944 0.118644
10 1.051 0.943 0.114528
11 1.057 0.995 0.062312
12 1.055 1.019 0.035329
13 1.050 0.998 0.052104
14 1.036 0.909 0.139714
Table 4.3 shows the critical bus ranking using VC index. Eight rankings are
produced as there are 8 load buses out of 14 buses available for the use this research.
Load buses are ranked ascending from weak to strong from the calculated VC index
above. Bus 14 appears as the weakest bus according to the index; follow by bus 9, 10, 5,
4, 11, 13 and finally bus 12 as the strongest bus.
33
Table 4.3: Critical load bus ranking of IEEE 14-bus system using VCi
Critical bus Load bus
ranking (weak to
strong)
1 14
2 9
3 10
4 5
5 4
6 11
7 13
8 12
Table 4.4 shows the tabulated real power and voltage magnitude from CPF on
IEEE 14-bus system. Only pure load bus is considered for the plotting of CPF, therefore
there is no data for bus 1, 2, 3, 6, 7, and 8.
Table 4.4: Real power and voltage magnitude of P-Q bus at voltage collapse point from
PV curve (IEEE 14-bus system)
Load bus P (p.u.) V (p.u.)
4 7.266 0.6824
5 6.055 0.6249
9 2.536 0.5905
10 1.695 0.5870
11 1.871 0.5781
12 1.826 0.5705
13 2.690 0.5878
14 1.354 0.6008
Table 4.5 shows the critical bus ranking using CPF. Eight rankings are produced
from the tabulated data above and load buses are ranked ascending from weak to strong.
Bus 14 appears as the weakest bus according to the index; follow by bus 10, 12, 11, 9,
13, 5 and finally bus 4 as the strongest bus.
34
Table 4.5: Critical load bus ranking of IEEE 14-bus system using voltage collapse point
from PV curve
Critical bus Load bus
ranking (weak to
strong)
1 14
2 10
3 12
4 11
5 9
6 13
7 5
8 4
Table 4.6 shows the calculation of VCPI on IEEE 14-bus system during critical
loading condition. For load bus 1, 2, 3, 6, 7, and 8, there were no VCPI index appears
due to their generation bus or P-V bus characteristics. As shown in column 3, proposed
VCPI is simplified by stating small change in load reactive power to one (1), compared
to the original VCPI introduced by Chen [2].
35
Table 4.6: Calculation of VCPI on heavy load state (IEEE 14-bus system)
Load bus
1 - - -
2 - - -
3 - - -
4 1.31 1 1.31
5 1.41 1 1.41
6 - - -
7 - - -
8 - - -
9 1.56 1 1.56
10 1.54 1 1.54
11 1.29 1 1.29
12 1.13 1 1.13
13 1.21 1 1.21
14 1.59 1 1.59
Table 4.7 shows the critical bus ranking using VCPI index. Eight rankings are
produced from the tabulated data above and load buses are ranked ascending from weak
to strong. Load buses are ranked ascending from weak to strong from the calculated
VCPI index above. Bus 14 appears as the weakest bus according to the index; follow by
bus 9, 10, 5, 4, 11, 13 and finally bus 12 as the strongest bus.
Table 4.7: Critical load bus ranking of IEEE 14-bus system using VCPI
Critical bus Load bus
ranking
(weak to strong)
1 14
2 9
3 10
4 5
5 4
6 11
7 13
8 12
36
4.2.1.3 Discussion
Table 4.8: Comparison of 3 proximity measures for contingency ranking of IEEE 14-
bus system
Rank (weakest to Proximity measures
strongest) CPF VC VCPI
1 14 14 14
2 10 9 9
3 12 10 10
4 11 5 5
5 9 4 4
6 13 11 11
7 5 13 13
8 4 12 12
Table 4.8 shows the load buses of IEEE 14-bus test system ordered from the
weakest to strongest using continuous power flow (CPF) and relative voltage change
index (VC), compared to voltage collapse proximity indicator (VCPI). All three
indicator noted bus 14 as the weakest bus. For VC and VCPI, both of them produce the
same rank of weak load buses, it is evidenced by their same choice of strongest bus in
system, which is bus 12, followed by bus 13, bus 11, bus 4, bus 5, bus 10, bus 9 and
finally bus 14. Besides the weakest bus, CPF screens different results compared to
another two indices. Except ranking of bus 10 and bus 13 are similar to those shown by
others, the remaining rank of buses deviate much, as it can be seen that CPF ranked bus
4 as the strongest bus while others ranked it as the fifth of weakest.
While for IEEE 30-bus system, it consists of 30 buses, 6 generators and 20 loads.
The test system data can be viewed in Appendix A.
37
Table 4.9: Determination of critical loading condition for IEEE 30-bus system
% of
increment 0% 10% 20% 70% 80% 90%
Minimum 0.961 0.955 0.949 0.915 0.907 0.899
voltage p.u. @ p.u. @ p.u. @ p.u. @ p.u. @ p.u. @
magnitude bus 8 bus 8 bus 8 bus 8 bus 8 bus 8
Maximum 1.000 1.000 1.000 1.000 1.000 1.000
voltage p.u. @ p.u. @ p.u. @ p.u. @ p.u. @ p.u. @
magnitude bus 1 bus 1 bus 1 bus 1 bus 1 bus 1
Benchmark results consist of VC and CPF. The calculation of the indices and the
critical bus ranking are shown.
Table 4.10 shows the calculation of VC on IEEE 30-bus system during critical
loading condition. For load bus 1, 2, 13, 22, 23 and 27, the VC index appears as nil due
to their generation bus or P-V bus characteristics, which will maintain their voltage
magnitude in spite of load change. Bus 5, 6, 9, 11, 25 and 28 are not considered for the
ranking as it is not a load bus, containing no load data for active and reactive power.
39
Table 4.10: Calculation of VCi on heavy load state (IEEE 30-bus system)
Load bus
1 1 1 0
2 1 1 0
3 0.983 0.952 0.032563
4 0.98 0.946 0.035941
5* 0.982 0.956 0.027197
6* 0.973 0.932 0.043991
7 0.967 0.923 0.047671
8 0.961 0.907 0.059537
9* 0.981 0.957 0.025078
10 0.984 0.972 0.012346
11* 0.981 0.957 0.025078
12 0.985 0.97 0.015464
13 1 1 0
14 0.977 0.955 0.023037
15 0.98 0.961 0.019771
16 0.977 0.957 0.020899
17 0.977 0.957 0.020899
18 0.968 0.94 0.029787
19 0.965 0.935 0.032086
20 0.969 0.942 0.028662
21 0.993 0.988 0.005061
22 1 1 0
23 1 1 0
24 0.989 0.978 0.011247
25* 0.99 0.981 0.009174
26 0.972 0.948 0.025316
27 1 1 0
28* 0.975 0.932 0.046137
29 0.98 0.961 0.019771
30 0.968 0.939 0.030884
* Bus 5, 6, 9, 11, 25 and 28 were neglected from ranking as it is not a P-Q bus
Table 4.11 shows the critical bus ranking using VC index. Eighteen rankings are
produced from the tabulated data above and load buses are ranked ascending from weak
to strong. The top three weakest buses appear as bus 8, 7 and 4, while the top three
strongest buses are bus 21, 24 and 10.
40
Table 4.11: Critical load bus ranking of IEEE 30-bus system using VCi
Critical bus Load bus
ranking (weak to
strong)
1 8
2 7
3 4
4 3
5 19
6 30
7 18
8 20
9 26
10 14
11 16
12 17
13 15
14 29
15 12
16 10
17 24
18 21
Table 4.12 shows the tabulated real power and voltage magnitude from CPF on
IEEE 30-bus system. Only pure load bus is considered for the plotting of CPF, therefore
there is no data for bus 1, 2, 5, 6, 9, 11, 13, 22, 23, 25, 27 and 28.
41
Table 4.12: Real power and voltage magnitude of P-Q bus at voltage collapse point
from PV curve (IEEE 30-bus system)
Load bus P (p.u.) V (p.u.)
3 3.702 0.5237
4 5.676 0.5995
7 2.586 0.5166
8 2.254 0.4937
10 3.307 0.7693
12 2.850 0.6237
14 1.337 0.5350
15 2.544 0.5963
16 1.523 0.5495
17 2.048 0.5601
18 1.297 0.5229
19 1.277 0.5208
20 1.400 0.5366
21* 2.600 0.9388
24 1.930 0.5212
26 0.352 0.4964
29 0.754 0.5364
30 0.748 0.5463
Table 4.13 shows the critical bus ranking using CPF. Eighteen rankings are
produced from the tabulated data above and load buses are ranked ascending from weak
to strong. The top three weakest buses appear as bus 26, 30 and 29, while the top three
strongest buses are bus 4, 3 and 10. PV curve for bus 21 is not plotted accurately as it
does not showed a full swing curve as others, thus the lowest point of the curve is
adopted for the research.
42
Table 4.13: Critical load bus ranking of IEEE 30-bus system using voltage collapse
point from PV curve
Critical bus Load bus
ranking (weak to
strong)
1 26
2 30
3 29
4 19
5 18
6 14
7 20
8 16
9 24
10 17
11 8
12 15
13 7
14 21*
15 12
16 10
17 3
18 4
Table 4.14 shows the calculation of VCPI on IEEE 30-bus system during critical
loading condition. For load bus 1, 2, 5, 6, 9, 11, 13, 22, 23, 25, 27 and 28, there are no
VCPI index appeared due to their generation bus or P-V bus characteristics. As shown in
column 3, proposed VCPI is simplified by stating small change in load reactive power to
one (1), compared to the original VCPI introduced by Chen [2].
43
Table 4.14: Calculation of VCPI on heavy load state (IEEE 30-bus system)
Load bus
1 - - -
2 1.00 1 1.00
3 1.10 1 1.10
4 1.11 1 1.11
5 - - -
6 - - -
7 1.15 1 1.15
8 1.19 1 1.19
9 - - -
10 1.07 1 1.07
11 - - -
12 1.09 1 1.09
13 - - -
14 1.10 1 1.10
15 1.08 1 1.08
16 1.10 1 1.10
17 1.10 1 1.10
18 1.11 1 1.11
19 1.12 1 1.12
20 1.11 1 1.11
21 1.02 1 1.02
22 - 1 -
23 1.00 1 1.00
24 1.04 1 1.04
25 - 1 -
26 1.09 1 1.09
27 - 1 -
28 - 1 -
29 1.04 1 1.04
30 1.06 1 1.06
Table 4.15 shows the critical bus ranking using VCPI index. Twenty rankings are
produced from the tabulated data above and load buses are ranked ascending from weak
to strong. Load buses are ranked ascending from weak to strong from the calculated
VCPI index above. The top three weakest buses appear as bus 8, 7 and 19, while the top
three strongest buses are bus 23, 2 and 21.
44
Table 4.15: Critical load bus ranking of IEEE 30-bus system using VCPI
Critical bus Load bus
ranking (weak to
strong)
1 8
2 7
3 19
4 4
4 18
4 20
7 3
7 14
7 16
7 17
11 12
11 26
13 15
14 10
15 30
16 24
16 29
18 21
19 2
19 23
45
4.2.2.3 Discussion
Table 4.16: Comparison of 3 proximity measures for contingency ranking of IEEE 30-
bus system
Rank (weakest to Proximity measures
strongest) CPF VC VCPI
1 26 8 8
2 30 7 7
3 29 4 19
4 19 3 4
5 18 19 18
6 14 30 20
7 20 18 3
8 16 20 14
9 24 26 16
10 17 14 17
11 8 16 12
12 15 17 26
13 7 15 15
14 21* 29 10
15 12 12 30
16 10 10 24
17 3 24 29
18 4 21 21
19 - - 2
20 - - 23
Table 4.16 shows the load buses of IEEE 30-bus test system ordered from the
weakest to strongest using continuous power flow (CPF) and relative voltage change
index (VC), compared to voltage collapse proximity indicator (VCPI). In overall, VC
and VCPI produce similar ranking of weak load buses while CPF shows an irrelevant
ranking compared to others. VC and VCPI indicate bus 8 as the weakest bus while CPF
votes for bus 26. For the strongest bus, both of them choose bus 12 but CPF goes for bus
4. When zooming into the difference of results by both VC and VCPI indices, their
ranking of load buses deviate not more than three (3) position of rank. For CPF, the
ranking is better at the middle rank. It is proofed that rank 5 and 10 for CPF is exactly
same as VCPI and ranking for bus 19, 14, 20, 16, 15, 12 and 10 are similar to what VC
and VCPI have ranked.
46
In terms of computation time, CPF and VCPI require one complete load flow
solution per load bus. CPF method need to plot the PV curve per load bus and obtain the
voltage collapse point, while it is vital for VCPI to obtain small increase in generation
and load reactive power. For VC, it requires two complete power flow solutions to
obtain the load bus voltage magnitude for initial and critical state. Table 4.17 shows the
computation time per power flow and table shows the overall computation time of all
indices. Stopwatch time function in MATLAB utilizing tic and toc code is used in the
analysis.
47
Table 4.17: Comparison of 3 proximity measures in terms of computation time per load
flow for both test systems
Test system Run Computation time per load flow (second)
attempt CPF (on bus 4) VC VCPI
IEEE 14-bus 1 0.381236 0.018805 0.018805
2 0.384566 0.018985 0.018985
3 0.385031 0.018978 0.018978
4 0.391615 0.017533 0.017533
5 0.385207 0.018658 0.018658
Average 0.385531 0.018592 0.018592
From Table 4.18, it can be seen that CPF has the longest computation time
compared to another two indices, and VC appears to be the fastest indices to compute.
For IEEE 14-bus test system, VCPI is seven (7) times slower than VC while twenty (20)
times faster than CPF. For IEEE 30-bus test system, VCPI is seven (7) times slower than
VC while fifteen (15) times faster than CPF. The assumption made for the analysis is the
power flow is attempted on bus 4 solely for CPF, and the power flow process for VC
and VCPI is the same. All the computation time tabulated is based on pure CPU
calculation time, neglecting time delay by user interaction. In comparison, VC has the
shortest computation time compared to another two indices.
4.3 Summary
This chapter has discussed the comparison of three indices to indicate weak load
buses in the power system, tested on IEEE 14-bus test system and 30-bus test system.
Two benchmark methods are adopted which were VC and CPF. The proposed VCPI is
rated as satisfactory in terms of accuracy, but has longer computation time compared to
performance of VC. While for comparison with CPF, their ranking results deviate much
due to the assumption of critical loading condition on VCPI index calculation, and VCPI
is superior to CPF from the perspective of computation time.
49
CHAPTER 5
5.1 Introduction
5.2 Conclusions
This research has presented a comparative study and analysis of the performance
of some static voltage collapse indices. The objectives for the research are archived: to
develop an indicator to perform contingency screening and ranking part of Voltage
Security Assessment on standard IEEE test system using MATLAB language and to
compare the results obtained from the VCPI to two benchmark results, which are VC
and CPF. The software used to analysis primary data included Matpower and Matlab.
Data collected is then analyzed by spreadsheet software Microsoft Office Excel.
For the results, all the indices VC, CPF and VCPI point bus 14 as the weakest
bus in the system for IEEE 14-bus test system, and VC and VCPI produce the exact
50
ranking from each other. While for IEEE 30-bus test system, VC and VCPI produce
similar ranking as well, rate bus 8 as the weakest bus but ranking of CPF deviates much,
it rates bus 26 which is ranked 12 by VCPI. The researcher believes that the deviation is
due to the critical loading condition assumption for calculation of VC and VCPI. In
terms of computation time, VC is the best among three indices, followed by VCPI and
CPF.
The results of this study indicate that the proposed VCPI is rated as satisfactory
in terms of accuracy, but has longer computation time compared to performance of VC.
While for comparison with CPF, their ranking results deviate much due to the
assumption of critical loading condition, and VCPI is superior to CPF from the
perspective of computation time. Compared to the original VCPI by Chen [2], proposed
VCPI is simplified by stating the small increase of load reactive power, Qi to one (1).
5.3 Recommendations
Based on the findings and conclusion of the study, here are several
recommendations to be considered:
2. Complex power VCPI proposed VCPI is applicable for load with sufficient
reactive power, obviously it would produce inaccurate index for unity power
factor loads. Therefore a complex power index will be a better indicator. Further
development of the index could refer to [22].
3. Precision on critical loading condition to obtain more precise of critical loading
condition, a smaller step size such as 5% or 1% can be adopted instead of 10%
that was used in this research.
This study should be conducted with large buses test system such as 57-bus or
118-bus to increase the validity of the research. Researchers should do more reading on
the topic of voltage security assessment and previous methods of voltage collapse
proximity indicators and indices to identify weak load buses in power system.
Source of the information should not depend solely on internet articles. Journals
and newspaper archive should also be taken into consideration. Methods of analyzing
data collected should not be restricted on reactive power and voltage magnitude, but
from more complex method such as right singular vector by Chen [2].
REFERENCES
1. R, Saidura and H.H, Masjukia and S, Mekhilef and H.W, Ping and Rahim,
N.A and Jahirul, M.I (2008) Energy Audit And Consumption Study Of Malaysian
Industrial Sector. In: The 2nd Industrial Applications Of Energy Systems
Conference, April 1-2, 2008, Oman.
2. Y. L. Chen, C. W. Chang, and C. C. Liu, "Efficient methods for identifying weak
nodes in electrical powernetworks," IEE Proceedings-Generation, Transmission
and Distribution, vol. 142, pp. 317-322, 1995.
3. P. Kundur. (1994). Power System Stability and Control: McGraw Hill, Inc.
4. A. M. Chebbo , M. R. Irving and M. J. H. Sterling Voltage collapse proximity
indicator: Behavior and implications, Proc. Inst. Elect. Eng., C, vol. 139, pp.
241 1992.
5. D S Kirschen, Power System Security, IEE Power Engineering Journal, Vol.
16, No. 5, pp.241-248, October 2002.
6. Hadi Saadat (1999), Power System Analysis, Boston: McGraw-Hill.
7. R. A. Schlueter, "A voltage stability security assessment method", IEEE Trans.
Power Syst., vol. 13, no. 4, pp. 1423 - 1438, 1998.
8. G. C. Ejebe, G. D. Irisarri, S. Mokhtari, O. Obadina, P. Ristanovic, and J. Tong,
"Methods for contingency screening and ranking for voltage stability analysis of
power systems," 1995.
9. Akorede, M.F.; Hizam, H.; Aris, I.; Kadir, M.Z.A.; , "Contingency evaluation for
voltage security assessment of power systems," Research and Development
(SCOReD), 2009 IEEE Student Conference on , vol., no., pp.345-348, 16-18 Nov.
2009
53
10. Obadina, O.O.; Berg, G.J.; , "Identifying electrically weak and strong segments
of a power system from a voltage stability viewpoint," Generation, Transmission
and Distribution, IEE Proceedings C , vol.137, no.3, pp.205-212, May 1990
11. H. Song, S. Kim, B. Lee, S.H. Kwon, V. Ajjarapu, Determination of interface
flow margin using the modified continuation power flow in voltage stability
analysis, IEE Proc. Gen. Transm. Distrib. 148 (2) (2001) 128132.
12. R. D. Zimmerman, C. E. Murillo-Sanchez, and R. J. Thomas, Matpower's
Extensible Optimal Power Flow Architecture, Power and Energy Society
General Meeting, 2009 IEEE, pp. 1-7, July 26-30 2009.
13. Reis, C.; Andrade, A.; Maciel, F.P.; , "Voltage stability analysis of electrical
power system," Power Engineering, Energy and Electrical Drives, 2009.
POWERENG '09. International Conference on , vol., no., pp.244-248, 18-20
March 2009
14. A.C. Zambroni, Identifying a vanishing eigenvalue in voltage collapse analysis
with consideration of limits. IEE Proc. Gen. Transm. Distrib. 148 2 (2001), pp.
263267.
15. A.A. El-Keib, X. Ma, Application of artificial neural networks in voltage
stability assessment, IEEE Trans. Power Syst. 10 (4) (1995) 18901896.
16. A.C.Z. de Souza, C.A. Canizares and V.H. Quintana, New techniques to speed
up voltage collapse computations using tangent vectors. IEEE Trans. Power
Syst. 12 3 (1997), pp. 13801387
17. S. Greene, I. Dobson, and F. L. Alvarado, "Contingency ranking for voltage
collapse via sensitivities from a single nose curve," IEEE Transactions on Power
Systems, vol. 14, pp. 232-240, 1999.
18. N. Yorino, H. Q. Li, S. Harada, A. Ohta, and H. Sasaki, "A method of voltage
stability evaluation for branch and generator outage contingencies," IEEE
Transactions on Power Systems, vol. 19, pp. 252-259, 2004.
19. T. Van Cutsem, C. Moisse, and R. Mailhot, "Determination of secure operating
limits with respect to voltage collapse," IEEE Transactions on Power Systems,
vol. 14, pp. 327-335, 1999.
54
20. J. Hongjie, Y. Xiaodan, and Y. Yixin, "An improved voltage stability index and
its application," International Journal of Electrical Power and Energy Systems,
vol. 27, pp. 567-574, 2005.
21. P. Kessel and H. Glavitsch, "Estimating the voltage stability of a power system,"
IEEE Transactions on Power Delivery, vol. 1, pp. 346-354, 1986.
55
Appendix A1
IEEE 14-bus system MATLAB M-file
% MATPOWER
% $Id: case14.m,v 1.11 2010/03/10 18:08:15 ray Exp $
%% bus data
% bus_i type Pd Qd Gs Bs area Vm Va baseKV zone Vmax Vmin
mpc.bus = [
1 3 0 0 0 0 1 1.06 0 0 1 1.06 0.94;
2 2 21.7 12.7 0 0 1 1.045 -4.98 0 1 1.06 0.94;
3 2 94.2 19 0 0 1 1.01 -12.72 0 1 1.06 0.94;
4 1 47.8 -3.9 0 0 1 1.019 -10.33 0 1 1.06 0.94;
5 1 7.6 1.6 0 0 1 1.02 -8.78 0 1 1.06 0.94;
6 2 11.2 7.5 0 0 1 1.07 -14.22 0 1 1.06 0.94;
7 1 0 0 0 0 1 1.062 -13.37 0 1 1.06 0.94;
8 2 0 0 0 0 1 1.09 -13.36 0 1 1.06 0.94;
9 1 29.5 16.6 0 19 1 1.056 -14.94 0 1 1.06 0.94;
10 1 9 5.8 0 0 1 1.051 -15.1 0 1 1.06 0.94;
11 1 3.5 1.8 0 0 1 1.057 -14.79 0 1 1.06 0.94;
12 1 6.1 1.6 0 0 1 1.055 -15.07 0 1 1.06 0.94;
13 1 13.5 5.8 0 0 1 1.05 -15.16 0 1 1.06 0.94;
14 1 14.9 5 0 0 1 1.036 -16.04 0 1 1.06 0.94;
];
%% generator data
% bus Pg Qg Qmax Qmin Vg mBase status Pmax Pmin Pc1 Pc2 Qc1min
Qc1max Qc2min Qc2max ramp_agc ramp_10 ramp_30 ramp_q apf
56
mpc.gen = [
1 232.4 -16.9 10 0 1.06 100 1 332.4 0 0 0 0 0 0 0 0 0 0 0 0;
2 40 42.4 50 -40 1.045 100 1 140 0 0 0 0 0 0 0 0 0 0 0 0;
3 0 23.4 40 0 1.01 100 1 100 0 0 0 0 0 0 0 0 0 0 0 0;
6 0 12.2 24 -6 1.07 100 1 100 0 0 0 0 0 0 0 0 0 0 0 0;
8 0 17.4 24 -6 1.09 100 1 100 0 0 0 0 0 0 0 0 0 0 0 0;
];
%% branch data
% fbus tbus r x b rateA rateB rateC ratio angle status angmin angmax
mpc.branch = [
1 2 0.01938 0.05917 0.0528 9900 0 0 0 0 1 -360 360;
1 5 0.05403 0.22304 0.0492 9900 0 0 0 0 1 -360 360;
2 3 0.04699 0.19797 0.0438 9900 0 0 0 0 1 -360 360;
2 4 0.05811 0.17632 0.034 9900 0 0 0 0 1 -360 360;
2 5 0.05695 0.17388 0.0346 9900 0 0 0 0 1 -360 360;
3 4 0.06701 0.17103 0.0128 9900 0 0 0 0 1 -360 360;
4 5 0.01335 0.04211 0 9900 0 0 0 0 1 -360 360;
4 7 0 0.20912 0 9900 0 0 0.978 0 1 -360 360;
4 9 0 0.55618 0 9900 0 0 0.969 0 1 -360 360;
5 6 0 0.25202 0 9900 0 0 0.932 0 1 -360 360;
6 11 0.09498 0.1989 0 9900 0 0 0 0 1 -360 360;
6 12 0.12291 0.25581 0 9900 0 0 0 0 1 -360 360;
6 13 0.06615 0.13027 0 9900 0 0 0 0 1 -360 360;
7 8 0 0.17615 0 9900 0 0 0 0 1 -360 360;
7 9 0 0.11001 0 9900 0 0 0 0 1 -360 360;
9 10 0.03181 0.0845 0 9900 0 0 0 0 1 -360 360;
9 14 0.12711 0.27038 0 9900 0 0 0 0 1 -360 360;
10 11 0.08205 0.19207 0 9900 0 0 0 0 1 -360 360;
12 13 0.22092 0.19988 0 9900 0 0 0 0 1 -360 360;
13 14 0.17093 0.34802 0 9900 0 0 0 0 1 -360 360;
];
Appendix A2
IEEE 30-bus system MATLAB M-file
% MATPOWER
% $Id: case30.m,v 1.12 2010/03/10 18:08:13 ray Exp $
%% bus data
% bus_i type Pd Qd Gs Bs area Vm Va baseKV zone Vmax Vmin
mpc.bus = [
1 3 0 0 0 0 1 1 0 135 1 1.05 0.95;
2 2 21.7 12.7 0 0 1 1 0 135 1 1.1 0.95;
3 1 2.4 1.2 0 0 1 1 0 135 1 1.05 0.95;
4 1 7.6 1.6 0 0 1 1 0 135 1 1.05 0.95;
5 1 0 0 0 0.19 1 1 0 135 1 1.05 0.95;
6 1 0 0 0 0 1 1 0 135 1 1.05 0.95;
7 1 22.8 10.9 0 0 1 1 0 135 1 1.05 0.95;
8 1 30 30 0 0 1 1 0 135 1 1.05 0.95;
9 1 0 0 0 0 1 1 0 135 1 1.05 0.95;
10 1 5.8 2 0 0 3 1 0 135 1 1.05 0.95;
11 1 0 0 0 0 1 1 0 135 1 1.05 0.95;
12 1 11.2 7.5 0 0 2 1 0 135 1 1.05 0.95;
13 2 0 0 0 0 2 1 0 135 1 1.1 0.95;
14 1 6.2 1.6 0 0 2 1 0 135 1 1.05 0.95;
59
%% generator data
% bus Pg Qg Qmax Qmin Vg mBase status Pmax Pmin Pc1 Pc2 Qc1min
Qc1max Qc2min Qc2max ramp_agc ramp_10 ramp_30 ramp_q apf
mpc.gen = [
1 23.54 0 150 -20 1 100 1 80 0 0 0 0 0 0 0 0 0 0 0 0;
2 60.97 0 60 -20 1 100 1 80 0 0 0 0 0 0 0 0 0 0 0 0;
22 21.59 0 62.5 -15 1 100 1 50 0 0 0 0 0 0 0 0 0 0 0 0;
27 26.91 0 48.7 -15 1 100 1 55 0 0 0 0 0 0 0 0 0 0 0 0;
23 19.2 0 40 -10 1 100 1 30 0 0 0 0 0 0 0 0 0 0 0 0;
13 37 0 44.7 -15 1 100 1 40 0 0 0 0 0 0 0 0 0 0 0 0;
];
%% branch data
% fbus tbus r x b rateA rateB rateC ratio angle status angmin angmax
mpc.branch = [
1 2 0.02 0.06 0.03 130 130 130 0 0 1 -360 360;
1 3 0.05 0.19 0.02 130 130 130 0 0 1 -360 360;
2 4 0.06 0.17 0.02 65 65 65 0 0 1 -360 360;
3 4 0.01 0.04 0 130 130 130 0 0 1 -360 360;
2 5 0.05 0.2 0.02 130 130 130 0 0 1 -360 360;
2 6 0.06 0.18 0.02 65 65 65 0 0 1 -360 360;
4 6 0.01 0.04 0 90 90 90 0 0 1 -360 360;
5 7 0.05 0.12 0.01 70 70 70 0 0 1 -360 360;
6 7 0.03 0.08 0.01 130 130 130 0 0 1 -360 360;
6 8 0.01 0.04 0 32 32 32 0 0 1 -360 360;
6 9 0 0.21 0 65 65 65 0 0 1 -360 360;
6 10 0 0.56 0 32 32 32 0 0 1 -360 360;
9 11 0 0.21 0 65 65 65 0 0 1 -360 360;
9 10 0 0.11 0 65 65 65 0 0 1 -360 360;
60
2 0 0 3 0.025 3 0;
];
62
Appendix B
test_cpf MATLAB M-file
function test_cpf
%TEST_CPF Test continuation power flow (CPF).
% created by Rui Bo on 2007/11/12
% MATPOWER
% $Id: test_cpf.m,v 1.4 2010/04/26 19:45:26 ray Exp $
% by Rui Bo
% Copyright (c) 2009-2010 by Rui Bo
%
% This file is part of MATPOWER.
% See http://www.pserc.cornell.edu/matpower/ for more info.
%
% MATPOWER is free software: you can redistribute it and/or modify
% it under the terms of the GNU General Public License as published
% by the Free Software Foundation, either version 3 of the License,
% or (at your option) any later version.
%
% MATPOWER is distributed in the hope that it will be useful,
% but WITHOUT ANY WARRANTY; without even the implied warranty of
% MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
% GNU General Public License for more details.
%
% You should have received a copy of the GNU General Public License
% along with MATPOWER. If not, see <http://www.gnu.org/licenses/>.
%
% Additional permission under GNU GPL version 3 section 7
%
% If you modify MATPOWER, or any covered work, to interface with
% other modules (such as MATLAB code and MEX-files) available in a
% MATLAB(R) or comparable environment containing parts covered
% under other licensing terms, the licensors of MATPOWER grant
% you additional permission to convey the resulting work.
%% test cpf
fprintf('\n------------testing continuation power flow (CPF) solver\n');
loadvarloc = 7;%6;%7 % bus number at which load changes
sigmaForLambda = 0.2;%0.05; % stepsize for Lambda
sigmaForVoltage = 0.05;%0.025; % stepsize for voltage
[max_lambda, predicted_list, corrected_list, combined_list, success, et] = cpf(casename,
loadvarloc, sigmaForLambda, sigmaForVoltage);
fprintf('maximum lambda is %f\n\n', max_lambda);
63
%% draw PV curve
flag_combinedCurve = true;
busesToDraw = [];%[3:6];
drawPVcurves(casename, loadvarloc, corrected_list, combined_list,
flag_combinedCurve, busesToDraw);
64
Appendix C
PV curves for each load bus in IEEE 14-bus system
Figure C(a): PV curve of IEEE 14-bus system, bus 4 Figure C(b): PV curve of IEEE 14-bus system, bus 5
Figure C(c): PV curve of IEEE 14-bus system, bus 9 Figure C(d): PV curve of IEEE 14-bus system, bus 10
Figure C(e): PV curve of IEEE 14-bus system, bus 11 Figure C(f): PV curve of IEEE 14-bus system, bus 12
65
Figure C(g): PV curve of IEEE 14-bus system, bus 13 Figure C(h): PV curve of IEEE 14-bus system, bus 14
66
Appendix D
PV curves for each load bus in IEEE 30-bus system
Figure D(a): PV curve of IEEE 30-bus system, bus 3 Figure D(b): PV curve of IEEE 30-bus system, bus 4
Figure D(c): PV curve of IEEE 30-bus system, bus 7 Figure D(d): PV curve of IEEE 30-bus system, bus 8
Figure D(e): PV curve of IEEE 30-bus system, bus 10 Figure D(f): PV curve of IEEE 30-bus system, bus 12
67
Figure D(g): PV curve of IEEE 30-bus system, bus 14 Figure D(h): PV curve of IEEE 30-bus system, bus 15
Figure D(i): PV curve of IEEE 30-bus system, bus 16 Figure D(j): PV curve of IEEE 30-bus system, bus 17
Figure D(k): PV curve of IEEE 30-bus system, bus 18 Figure D(l): PV curve of IEEE 30-bus system, bus 19
68
Figure D(m): PV curve of IEEE 30-bus system, bus 20 Figure D(n): PV curve of IEEE 30-bus system, bus 21
Figure D(o): PV curve of IEEE 30-bus system, bus 24 Figure D(p): PV curve of IEEE 30-bus system, bus 26
Figure D(q): PV curve of IEEE 30-bus system, bus 29 Figure D(r): PV curve of IEEE 30-bus system, bus 30
69
Appendix E
MATLAB command for constant power factor load increment
(for 14-bus system, 180% increment)
define_constants;
mpc = loadcase('case14');
mpc.bus(2, PD) = 60.76 ;
mpc.bus(3, PD) = 263.76 ;
mpc.bus(4, PD) = 133.84 ;
mpc.bus(5, PD) = 21.28 ;
mpc.bus(6, PD) = 31.36 ;
mpc.bus(9, PD) = 82.6 ;
mpc.bus(10, PD) = 25.2 ;
mpc.bus(11, PD) = 9.8 ;
mpc.bus(12, PD) = 17.08 ;
mpc.bus(13, PD) = 37.8 ;
mpc.bus(14, PD) = 41.72 ;
mpc.bus(2, QD) = 35.56 ;
mpc.bus(3, QD) = 53.2 ;
mpc.bus(4, QD) = -10.92 ;
mpc.bus(5, QD) = 4.48 ;
mpc.bus(6, QD) = 21 ;
mpc.bus(9, QD) = 46.48 ;
mpc.bus(10, QD) = 16.24 ;
mpc.bus(11, QD) = 5.04 ;
mpc.bus(12, QD) = 4.48 ;
mpc.bus(13, QD) = 16.24 ;
mpc.bus(14, QD) = 14 ;
runpf(mpc);
70
Appendix F
MATLAB M-file for command computation time
tic
test_cpf;
runpf('case14');
runpf('case30');
toc