Content uploaded by K R S S Prasad
Author content
All content in this area was uploaded by K R S S Prasad on Apr 25, 2016
Content may be subject to copyright.
Page
175
Optimal Placement of SVC and UPFC in
Transmission Networks using SFLA
Publication History
Received: 24 August 2015
Accepted: 28 September 2015
Published: 25 October 2015
Citation
Prasad KRSS, Damodar Reddy M. Optimal Placement of SVC and UPFC in Transmission Networks using SFLA. Discovery,
2015, 45(210), 175-181
Discovery
ANALYSIS
The International Daily journal
ISSN 2278 – 5469 EISSN 2278 – 5450
© 2015 Discovery Publication. All Rights Reserved
Page
176
Optimal Placement of SVC and UPFC in
Transmission Networks using SFLA
K.R.S.S.Prasad M.Damodar Reddy
Electrical & Electronics Engineering Electrical & Electronics E
ngineering
S V University S V University
Tirupati - 517502, India Tirupati -517502, India
sivasai.rajarshi@gmail.com
Abstract –This paper proposes an application of
Shuffle frog leaping Algorithm (SFLA) based on
minimization of Active power loss and improvement
of voltage profile in transmission networks by
incorporating Static Var Compensator (SVC) and
Unified Power Flow Controller (UPFC) (Power
injection model) devices. The main objective of this
paper is to reduce Active power loss and improve the
voltage profile of transmission networks. Here the
Fuzzy approach is used for optimal locations of SVC
and active power loss Sensitivity factors are used to
find the optimal location’s of UPFC. SFLA is used to
find out the optimal SVC sizes and optimal control
parameter settings of UPFC with regard to the power
loss minimization. The proposed method is tested on
IEEE 14-bus and IEEE 30-bus test systems and the
results are discussed.
Keywords— Transmission network, optimal placement,
Fuzzy approach, SFLA.
I. INTRODUCTION
In the present day scenario, the growing
demand and tight restrictions on construction of
new lines has resulted in unscheduled power flows
and higher transmission losses. This has made the
transmission systems increasingly stressed, more
difficult to operate and vulnerable to security
threats. In order to improve the performance of
power system, power flow across it must be
controlled rather than generation rescheduling or
topology changes. The reliability of power systems
while handling large volumes of energy
transactions is maintained using several new
technologies. Flexible AC Transmission Systems
(FACTS) are one of these technologies and are
most reliable to implement. Rapid growth in the
power electronics technology made FACTS a
favourable and promising concept for power
system applications [1-8]. Application of FACTS
technology made the control of power flow along
transmission lines more flexible.
In this paper, two FACTS devices Static
Var Compensator (SVC) & Unified Power Flow
Controller (UPFC) are used. SVC is used for shunt
compensation. It is a shunt-connected static Var
generator or absorber which minimises power loss
and improves voltage profile. UPFC is a versatile
FACTS device, proposed for real time control and
dynamic compensation of AC transmission
systems. It can independently or simultaneously
control the Active power, reactive power and bus
voltage to which it is connected. In literature
different UPFC models are available [2,5,6]. In this
paper the power injection model of UPFC is put to
use and optimal control parameter settings of
UPFC are obtained with the help of SFLA. Fuzzy
& active power loss Sensitivity factors are used to
find the optimal location of SVC & UPFC
respectively in transmission network. SFLA is a
meta-heuristic search method inspired from a
memetic natural evolution of a group of frogs when
exchanging information in search for the location
that has the maximum amount of food available [9,
10,11].
II. MODELLING OF SVC & UPFC
A. Modelling of SVC
SVC is a shunt compensation device
which is mainly employed for the purpose of
voltage regulation. It is an important component for
voltage control [3,4], normally it is installed at the
receiving node of the transmission line. The basic
form of SVC is quite different from its working
point of view. In its basic form, it is a parallel
combination of thyristor controlled reactor with a
bank of capacitors. On the other hand in its
working point of view, it resonates a shunt
connected variable reactance, which is either,
generates (or) absorbs reactive power in order to
regulate the voltage magnitude.
Fig.1. Injection model of SVC
The above Fig.1 depicts SVC as a shunt
branch with a compensated reactive power QSVC,
set by available inductive and capacitive
Page
177
susceptance. In this paper the SVC model is
realized as an element which feeds a certain
amount of reactive power at selected bus [4].
B. Modelling of UPFC
UPFC comprises of two voltage source
converters having a DC link in common and are
connected back to back. The DC link is provided
by a DC storage capacitor as shown in Fig.2. The
real power demanded by converter 2 is either
supplied or absorbed by the converter 1 through a
common DC link. Converter 1 can also provide
independent shunt reactive compensation for the
line either by generating or absorbing controllable
reactive power. Converter 2 injects an AC voltage
with controllable magnitude and phase angle in
series with transmission line via a series
transformer [5]. This is the main function of UPFC,
through which converter 2 either supplies or
absorbs reactive power and active power is
exchanged on account of series injected voltage.
Fig.2. UPFC Schematic Diagram
Fig.3. Steady-state UPFC power injection model
Here in this study the power injection
model of UPFC has been used. The power injection
model of UPFC is quite simple and can easily be
incorporated in load flow studies [5, 6, 7, 8]. This
type of UPFC model is more useful for the study
state analysis. The above Fig.3 portrays the power
injection model of UPFC. The elements of the
equivalent power injections in Figure 3 are,
, = 0.02 − 1.02
−
+) (1)
, =
− +(2)
, =−(3)
, =
− +(4)
In a power system where UPFC is located between
nodes i and j, the admittance matrix is modified by
adding a reactance between the nodes. The
Jacobian matrix is modified by addition of
appropriate powers.
III. OPTIMAL LOCATIONS OF SVC & UPFC
A. Optimal Location of SVC using Fuzzy Approach
The optimal locations of SVC on load
buses are found out by using Fuzzy approach. Here
Fuzzy logic is developed by considering two
objective functions they are by reducing real power
losses and maintaining voltage profile within the
prescribed limits (0.9p.u-1.1p.u). The Fuzzy rules
are developed on the basis of two inputs Power loss
index (PLI) and nodal voltages (p.u). Power loss
index value for node can be finding by using
the following equation (5).
()=()− ()
()− ()(5)
Where LR (n) is known as the loss
reduction at node and the minimum and
maximum values of loss reduction are denoted by
LR(min) and LR(max) respectively. The fuzzy
rules are taken from [4].The suitability index for
the SVC placement is given by the output of fuzzy.
The appropriate locations of the SVC will be the
maximum values of the Fuzzy output.
Fig.4. Membership function plot for power loss index
Page
178
Fig.5. Membership function plot for p.u. nodal voltage
Fig.6. Membership function plot for SVC suitability
B. Optimal Location of UPFC using Active power
loss sensitivity indices:
In order to find the optimal location of
UPFC the most unstable line in the system should
be find out. In this regard the active power loss
sensitivity factors are used [8]. For UPFC placed
between buses i and j the active power loss
sensitivity factors with respect to the parameters of
the UPFC is given as
=
(6)
The active power loss sensitivity factors is
calculated with the help of the equation given
below which is deduced from the equation (6).
= 2
cos− + 2
sin(− ) (7)
IV. SHUFFLED FROG LEAPING ALGORITHM
(SFLA)
SFLA is an evolutionary computation
method, inspired from the frog prey behaviour.
This was proposed by Eusuff and Lansey. It is
based on swarm intelligence and is similar to
Memetic Algorithm, which is based on co-
operative search [9,10,11]. It has the combined
benefits of genetic based Mas and the social
behaviour based PSO algorithms. In this algorithm
a set of frogs are taken as the population that are
divided into number of subsets known as
memplexes. A local search is performed in each
memplex and the different memplexes are
considered as different cultures of Frogs. Within
each memplex the individual frogs contain ideas,
that can be influenced by the ideas of the other
frogs, and evolve through a process of memetic
evolution. After a defined number of steps, the
ideas among the memplexes are passed as a
shuffling process. The above two processes i.e. the
local search and the shuffling processes continue
until a defined convergence criterion is satisfied.
For S-dimensional problem (S variables), a frog i is
represented as Xi= (xi1, xi2,……,xip) and an initial
population of P frogs is created randomly.
Afterwards, according to their fitness values all the
frogs are sorted in a descending order. Then the
entire population is divided into m memplexes,
each containing n number of frogs (P=n×m). In
this process first frog goes to first memplex and the
second frog goes to the second memplex, frog m
goes to the memplex and frog m+1 goes to the
first memplex and so on. Each memplex consists of
one best value and one worst value identified as
and respectively. In each cycle of process the
frog with the worst fitness is improved with a
process similar to PSO. Accordingly, the position
of the frog with the worst fitness is adjusted as
follows. Change in frog position:
(D) = rand ( )×(X-X) (8)
New position X = current position X+D (9)
D ≥D≥ −D
Where rand ( ) is a random number between 0 and
1 and D is the maximum allowed change in a
frog’s position. The calculations in (8) and (9) are
repeated to replace the worst frog with a better
result but with respect to the global best frog (X
replaces). If no improvement is possible in this
case, the then new solution is randomly generated
to replace that frog. The calculations will continue
until the termination criteria is satisfied [11].
V. PROCEDURE OF SFLA
The SFLA-based approach for finding the
optimal sizing of SVC and optimal control
parameter setting of UPFC to minimize the total
active power loss and to improve the voltage
profile takes the following steps [10].
Step1: Create the initial population of P frogs
generated randomly. SFLA population
[X,X,.....,X]× where, P=n×m, m is number
of memplexes, n is the number of frogs in each
memplex.
Step2: After generating the population randomly,
the population is sorted in the descending order and
the frogs are divided into the memplexes such that
Page
179
P=n×m. The division is done with the first frog
going to the first memplex, second one going to the
second memplex and the m one going to the mth
memeplex, the m+1th frog goes back to the first
memplex. The following Fig.7 portrays memplex
partitioning process
Fig.7. Memplex Partitioning process
Step3: Within each constructed memplexes, the
frogs are affected by other frog’s ideas, hence they
experience heuristics evolution. Memetic evolution
improves the quality of the meme of an individual
and enhances the individual frog’s performance
towards the goal to achieve optimal solution.
Step4: For each memplex, the frogs with the wrost
fitness and best fitness are identified as X and X
respectively. Also the frog with the global best
fitness X is identified, and then the position of the
worst frog X for the memplex is adjusted with the
help of equations (8) and (9). The below Fig. 8
demonstrates the original frog leaping rule.
Fig.8 The original frog leaping rule
If the evolution process gives a better frog
(solution), it replaces the older frog, otherwise X is
replaced by X and the process is repeated. If no
improvement is possible in this case a random frog
is generated which replaces the old frog.
Step5: Check the convergence. If the convergence
criteria are satisfied stop, otherwise consider the
new population set as the initial population and
return to Step2. The best solution is found in the
search process is considered as the output results of
the algorithm. By observing the flow chart we can
easily understand that how this method is
implemented.
Flowchart:
Fig.9. SFLA flow chart
VI. IMPLEMENTATION OF SFLA TO
DETERMINE THE SIZE OF SVC AND
CONTROL PARAMETER SETTINGS OF UPFC
Algorithm:
1.) Initialize the population size (number of frogs),
maximum number of iterations, minimum and
maximum values of SVC size and control
parameter setting of UPFC to be installed in
transmission network.
2.) Choose the parameters that are to be optimized.
Here the parameters are reactive power that is to be
injected and r and values which are control
parameters of UPFC.
3.) Randomly generate the initial solutions for all
frogs within their operating limits and assume them
as initial SVC sizes, r and values.
4.) Run the load flow and obtain the voltage profile
and real and reactive power losses of the system.
5.) Assume the fitness function as the real power
loss as we need to find the optimal SVC size and
control parameter settings of UPFC that minimizes
the losses to a maximum extent.
6.) Sort the population of frogs in descending order
by iterating through all the values of fitness
function.
Page
180
7.) Divide the population of frogs into ‘m’ number
of memeplexes. In this process, first frog goes to 1st
memplex, mth frog goes to mth memplex and frog
m+1 go to the 1st memeplex and so on.
8.) Within each memeplex, determine the best and
worst frogs.
9.) In each memeplex, improve the worst frog
position using the equation (9).
10.) Check whether the newly obtained solution is
better than the old or not. If it is better, then replace
the old frog with new solution.
11.) Combine the evolved memplexes.
12.) Check for the termination criteria. If it is
satisfied, terminate the loop and plot the results.
Else, repeat the steps 6 to 11 until termination
criteria are satisfied.
13) The solution vector of frogs corresponding to
best fitness value gives the optimal SVC sizes and
optimal control parameter settings of UPFC in
optimal locations.
VII. RESULTS
The proposed method was tested on IEEE-
14 bus and IEEE-30 bus and the results of IEEE 14
bus are shown in the following tables.
Table.1 – Results for 14 bus system losses with and without
UPFC
Aspect
Losses
Without
UPFC
UPFC
Location.
SFLA
UPFC
Parameters
Losses
With
UPFC
Total
Active
Power
loss
13.3938
5-6
=168.80
°
r=0.0659
13.3138
Table.2 – Results for 14 bus system losses with and without
SVC
Aspect
Losses
Without
SVC
SVC
Location.
SFLA
SVC
Ratings
(MVAR)
Losses
With
SVC
Total
Active
Power
loss
13.3938
5,
14
25.4258
6.9893
13.2742
Table.3 – Results for 14 bus system losses with and without
UPFC & SVC
Table.4-Voltages of 14 bus system for all the three case studies
Bus
no: Without
FACTS UPFC in
line 5-6
SVC at
5 & 14 SVC at 14
& UPFC in
line 5
-
6
1 1.0600 1.0600 1.0600 1.0600
2 1.0450 1.0450 1.0450 1.0450
3 1.0100 1.0100 1.0100 1.0100
4 1.0183 1.0249 1.0253 1.0255
5 1.0200 1.0308 1.0312 1.0314
6 1.0700 1.0700 1.0700 1.0700
7 1.0608 1.0637 1.0659 1.0659
8 1.0900 1.0900 1.0900 1.0900
9 1.0541 1.0567 1.0612 1.0613
10 1.0495 1.0516 1.0554 1.0554
11 1.0561 1.0572 1.0591 1.0592
12 1.0550 1.0553 1.0570 1.0570
13 1.0501 1.0505 1.0537 1.0537
14 1.0343 1.0360 1.0502 1.0503
VIII. CONCLUSION
In this paper, a two stage methodology for
finding the optimal locations, sizes of SVC and the
optimal control parameter settings of UPFC for
active power loss minimization of standard tested
IEEE-14 bus system is presented. The following
conclusions are drawn based on the results depicted
in the paper:
By placing the UPFC in the optimal
branch location which is found out with the help of
Active power loss sensitivity factors the total real
power loss of the system is minimised and the
voltages are also improved simultaneously. The
loss reduction by placing two SVC’s in the optimal
locations is more compare to the first case. The loss
Aspect
Losses
Without
UPFC
UPFC
Location.
SVC
Location.
SFLA
UPFC
Parameters
&
SVC
Ratings
(MVAR)
Losses
With
UPFC
&
SVC
Total
Active
Power
loss
13.3938
5-6
14
5.3161
=165.503
°
r=0.0594
13.2734
Page
181
reduction is increased in the case three when both
SVC and UPFC are placed simultaneously in the
test system. The proposed SFLA algorithm is more
accurate and faster in giving the results.
Reference
[1] K. Venkata Ramana Reddy, M. Padma Lalitha,
Harsha vardan Reddy, ‘‘Enhancement of voltage stability
by using FACTS under normal and transient conditions”,
International Journal of Electrical and Data
communication, ISSN: 2320,vol.1,Issue.7,sep-2013.
[2] Dr.M.Damodara Reddy and P.Ramesh, ‘‘Loss reduction
through optimal placement of Unified power-flow
controller using Firefly Algorithm”, IJAREEIE, ISSN:
2320 3765, vol.2 Issue.10, October-2013.
[3] B.Venkateswara Rao, Dr.G.V.Nagesh Kumar, M.Ramya
Priya, and P.V.S.Sobhan, ‘‘Implementation of Static VAR
Compensator for Improvement of Power System
Stability”,IEEE-2009.
[4] Dr.M.Damodar reddy and K.Dhanunjaya Babu ‘‘optimal
placement of SVC using fuzzy and PSO algorithm”,
International journal of engineering research and
applications, ISN: 2248-9622. vol.3,Issue 1,January-
February 2013.
[5] A. Mete Vural and Mehmat Tumay, ‘‘Mathematical
modelling and analysis of a unified power flow controller:
A comparison of two approaches in power flow studies
and effects of UPFC location”, Electrical Power and
Energy Systems 29 (2007) 617-629.
[6] S.V.Ravi Kumar and S. Siva Nagaraju, ‘‘Loss
Minimization by incorporation of UPFC in Load Flow
Studies”,International Journal of Electrical and power
engineering 1 (3): 321-327,2007.
[7] Ch. Chengaiah, G.V.Marutheswar and
R.V.S.Satyanarayana, ‘‘Control Setting of Unified Power
Flow Controller Through Load Flow
Calculation”,ARPNJournal,vol.3,December,2008.
[8] Dr.shivasharanappa G C, Sunil Kumar A V, Roopa V,
Javid Akthar, ‘‘Transmission Loss Allocation and Loss
Minimization by Incorporating UPFC in LFA”,
International journal of engineering Resarch,
vol.1,Issue.1,pp-236-245, 2009.
[9] M.M.Eusuff and K.E. Lansey, ‘‘Optimization of water
distribution network design using the shuffled frog leaping
algorithm”. J.water Resources Planning &Management ,
vol.129(3),pp.210-225,2003.
[10] M.M.Eusuff and K.E.Lansey, F. Pasha: Shuffled frog
leaping algorithm: a memetic meta-heuristic for discrete
optimization”,EngineeringOptimization,2006,vol,38,No.2,
pp.129-154.
[11] R.Jahani, H.A.Shayanfar, N.M.Tabatabaei, J.Olamaei
‘‘Optimal Placement of UPFC in power system by a new
advanced heuristic method”, IJTPE Journal, ISSN 2077-
3528,December 2010, vol.2,No.4,pp.13-18.
