Optimal Placement of Static Series Voltage Regulator (SSVR) in Distribution System for Voltage improvement Using Particle Swarm Optimization

M. Laxmidevi Ramanaiah.et.al. Int. Journal of Engineering Research and Application www.ijera.com
ISSN : 2248-9622, Vol. 7, Issue 3, ( Part -1) March 2017, pp.23-28
www.ijera.com DOI: 10.9790/9622- 0703012328 23 | P a g e
Optimal Placement of Static Series Voltage Regulator (SSVR) in
Distribution System for Voltage improvement Using Particle
Swarm Optimization
M. Laxmidevi Ramanaiah *, Dr. M. Damodar Reddy**, T. Chandra Sekhar***
*(Research Scholar, Department of E.E.E., S.V.University, Tirupati, Andhra Pradesh, India.
** (Professor, Department of E.E.E., S.V.University, Tirupati, Andhra Pradesh, India.
*** (M.Tech student, Department of E.E.E., S.V.University, Tirupati, Andhra Pradesh, India.
ABSTRACT
This paper presents optimal placement of Static Series Voltage Regulator (SSVR), for voltage profile
improvement and power loss reduction in radial distribution systems under steady state condition. SSVR consists
of a series compensator. The series compensator injects the series voltage in quadrature with the branch current
in such a way that the receiving end voltage is maintained at desired value (up to 1 p.u). The criteria for selection
of optimum location of SSVR are under voltage problem mitigation and loss reduction in the network under
steady sate condition. Particle Swam Optimization (PSO) technique is used to find the rating of the device. The
proposed model is tested using standard distribution system consisting of 33 nodes.
Keywords: Static Series Voltage Regulator (SSVR), Voltage Source Converter (VSC), Series Voltage (Vse),
Particle Swarm Optimization (PSO).
I. INTRODUCTION
The radial distribution system usually
suffers from high power loss, poor voltage profile at
the end buses, poor voltage stability and line
loadability. Optimal reactive power compensation
can improve all these problems. Therefore, there are
several reactive power compensation devices
reported in the literature. For instance, capacitor
placement [1-2], combined operation of on load tap
changer and capacitor banks [3], integration of
distributed generation (DG) [4].
Flexible AC Transmission System
(FACTS) devices were originally developed for
transmission system to alleviate the system from the
problem of congestion, FACTS used in distribution
systems are called as Distribution FACTS
(DFACTS) or Custom Power Devices (CPD). These
devices are used mainly for compensation of power
quality problems like voltage unbalance voltage sag
and flicker. Some of the DFACTS devices are
distribution static compensator (DSTATCOM) [5],
Dynamic Voltage Restorer (DVR) [6]. Comparison
of DVR and DSTATCOM are presented in [7] and
Static Series Voltage Regulator in [8].
Most of the DFACTS devices are modeled
and utilized for power quality problems in a small
two bus distribution system with a sensitive load and
source. But, the research on the effect of DFACTS
on reactive power compensation in a large
distribution system is very less in steady state
conditions.
In the proposed work, a SSVR [8] model is
used and its impact on voltage compensation is
studied in a large radial distribution system. In the
SSVR model, the voltage source converter injects
controllable series voltage (Vse) in quadrature with
the branch current. Thus, the load voltage magnitude
changes to desired value (up to 1 p.u.) during
healthy operating condition. Particle Swarm
optimization (PSO) method is used to find the rating
of SSVR.
II. RADIAL DISTRIBUTION SYSTEM
LOAD FLOW IN STEADY STATE
CONDITION
For any power system, load flow is the
basic and essential method for analysis, planning and
operation studies under steady-state condition. Most
of the distribution systems are fed at one point and
system has a radial structure as shown in Fig 1.
Several load flow methods have been developed for
radial distribution systems [9-10]. The load flow
method [10] based on BIBC and BCBV matrices
formulation is used in this paper. The proposed load
flow algorithm requires formation of Bus Injection
to Branch Current (BIBC) matrix with 1’s & 0’s as
elements and Branch Current to Bus Voltage
(BCBV) matrix with primitive impedances as
elements and Distribution Load Flow (DLF) matrix.
DLF matrix is the product of BCBV & BIBC
matrices.
RESEARCH ARTICLE OPEN ACCESS

Fig 1: Simple 6 bus radial distribution system
At bus i the complex load SLi is given by
 Li Li Li
S P j Q 
(1)
Where, i = 2 ….. N
SLi is the complex power at ith
bus;
PLi is the real power at ith
bus;
QLi is the reactive power at ith
bus
N=number of buses.
The load current at node i is given by
*
( )
L i L i
k
i
L
P jQ
I
V


(2)
Where Vi
k
is the bus voltage at kth
iteration for ith
bus; IL is load current at the kth
iteration for ith
bus.
The branch currents can be written as
     B L
I B IB C I
(3)
The relation between branch current and bus
voltages is given by
     B
V B C B V I 
(4)
The formation of BIBC and BCBV matrices is
presented in [11]. Substituting the [IB] matrix, the
resultant equation expressed as
       L
V BC BV BIBC I 
     L
V D L F I 
Node voltage is obtained as
1 1
[ ]
k k k
V V V
 
  
(5)
Thus, we obtain the voltage magnitude and phase
angle, which completes the load flow.
2.1 Power loss calculations
The power losses in a distribution system can be
written as
2
1
b
B m m
m
L o s s
P I R

  (6)
2
1
b
B m m
m
L o s s
IQ X

  (7)
Where m is number of branches, Rm is the resistance
and IBm is the magnitude of mth
branch current.
III. STATIC SERIES VOLTAGE
REGULATOR (SSVR) IN
DISTRIBUTION SYSTEM
SSVR consists of a series compensator as shown in
Fig.2.
Fig 2: Single line diagram of two buses of a
distribution system with SSVR
Here Vs is the sending end voltage; Vse is the
voltage injected by series compensator; VL is the
load bus voltage; IB is the branch current. From fig
1, if SSVR is placed in the branch B2, then the
receiving end bus 3 is maintained at 1p.u. From
Fig.3, the series compensator injects the series
voltage (Vse) in quadrature with the branch current
in such a way that the receiving end voltage (VL) is
maintained at desired value (up to 1 p.u).Since,
SSVR injects the series voltage in quadrature to
branch current it can provide only reactive power to
the system. The phasor diagram in Fig 4 shows the
effect of series injected voltage on the desired load
bus voltage.
Fig 3: Model of SSVR and phasor diagram of
reactive power exchange operation
Fig 4: Phasor diagram for voltage and currents with
series compensator

From the phasor diagram given in Fig.4, the new
load bus voltage is given as,
'
L L se
V V V     (8)
Where
'
L
V is magnitude of load bus voltage after injecting
series voltage.
δ is displacement angle between VL and
'
L
V
Vse is series injected voltage
 is the angle of series injected voltage with
respect to the angle of load voltage VL.
The series power is obtained as follows:
*
.( )S S V R se
B
S V I (9)
Where SSSVR
is the complex power rating of SSVR;
se
V is series voltage injected by SSVR; B
I is
branch current in which SSVR injects the series
voltage
IV. PARTICLE SWARM
OPTIMIZATION
Particle Swarm Optimization is swarm
intelligence based technique [2] that mimics the
social behavior of a flock of birds. It has stable
convergence characteristics. The behavior of birds to
search for food is modeled in this optimization. PSO
is found to be superior to evolutionary optimization
methods. The terms associated with PSO are particle
best and global best. The searching mechanism of
PSO involves velocity update and position update.
The new velocities are calculated using equation
(10)
1 ( , ) ( , ) 2 2
(1, ) ( ,
1
( , ) ( , ) 1
)
(p b est X )
(g b est X
*
)
(
)
*
t t
m
t t
m n m n n m n
t t
n m n
V ra n d c ra n dk w V c

 

 
(10)
Where k is a constriction factor.
c1, c2 are the learning constants.
rand1, rand2 are the random numbers in the range of
0 and 1
( , )
t
m n
V = velocity of mth
particle in nth
dimension ( , )
X
t
m n
= position of mth
particle in nth
dimension ( , )
pbest
t
m n
= personal best of mth
particle
in nth
dimension
(1, )
gbest
t
n
=global best of the particle in nth
dimension.
W is the inertia weight given by
m ax m ax m in
((w w ) * t/ T )w w   (11)
w is an adjustable parameter between wmax and wmin
t = current iteration number
T = maximum number of iterations
The positions of the particles are updated using
equation (12)
1
( , ) ( , )
1
( , )
X X
t t t
mn nm m n
V
 
  (12)
4.1. Algorithm to find optimum value of series
injected voltage using PSO technique
Step 1: Initialize the particles, maximum iterations,
and dimension, minimum and maximum reactive
power limits of SSVR, inertia weights and velocity
of the particles. Here wmax and wmin are taken as 0.9
and 0.4 respectively.c1=c2=2.05.
Step 2: Generate the particles and velocities
randomly within the limits. Here the particles are
reactive power rating of SSVR. Series voltage is
calculated from reactive power rating of SSVR and
branch current.
Step 3: Run the load flow by adding the series
voltage in the branch as given in the equation (8).
The voltage constraint imposed is the voltage at the
optimal location should not exceed the desired
value (1 p.u.). Discard the particles which violate the
voltage constraint. Find the fitness. Here the fitness
is minimization of real power losses.
Step 4: Obtain the pbest values for all the particles
from the fitness values and the best value among all
the pbest values (gbest) is identified.
Step 5: Iteration count is set to one.
Step 6: Calculate the velocity using equation (10) for
all the particles.
Step 7: Update the position of each particle as given
in equation (12).
Step 8: Calculate new fitness functions for the new
positions of all the particles. Update the pbest as
follows:
 
 
1
( , ) ( , )
1
( , )
1 1
( ,
( , )
( , ) ( , ) )
( )
( )
t t t
b est n b est nm n
m n m n
t
b est n
t t t
b est n
P if f f P
P
X
f fXi f PX


 
 

 

(13)
Update the gbest from the latest pbest.
Step 9: Increment the iteration count. If iteration
count does not reach its maximum, then go to step 6.
Step 10: Identify the gbest particle which gives the
SSVR rating and series voltage.
V. RESULTS OF 33-BUS TEST SYSTEM
The single line diagram of the 12.66 kV,
33-bus, 4-lateral radial distribution system and the
data of the system are obtained from [11]. The total
load of the system is considered as 3715 kW and
2300 kVAr. In 33 bus system 21 nodes out of 33
nodes of distribution system (63.63%) have under
voltage problem.
Table 1 illustrates results of 33 bus system
before and after installation of SSVR in branch 5 (6th
node). Table 2 gives voltage profiles. Table 3 shows
results for SSVR Location, Series injected voltage,

Reactive power rating, real and reactive power loss,
loss reduction and number of nodes with under
voltage problem for 33 bus distribution system.
SSVR placement in distribution system
changes all the downstream node voltages, whereas
upstream node voltages are slightly improved. When
SSVR is placed in branch 5 (6th
node), the number of
nodes with under voltage problem are zero. The
reduction in real and reactive power losses with
SSVR placed in branch 5 is 17.1230 kW and
11.7233 kVAr respectively. Thus, from the
viewpoint of loss reduction and under voltage
problem mitigation, branch 5 is the optimal location.
The rating of SSVR obtained for best location is
224.2600 kVAr. SSVR placement in the branches
from 18 to 24 has no significant improvement in
voltage and loss reduction.
Table 1: Results of 33 bus system
Table 2: Voltage profiles of 33 bus system without
and with SSVR
Table 3: SSVR Location, Series injected voltage, real and reactive power injected, real and reactive power loss,
loss reduction and number of nodes with under voltage problem for 33 bus distribution system
SSVR
Location
(branch)
Series injected
voltage (p.u)
Real
power
injected
(kW)
Reactive
power
injected
(kVAr)
Power loss Loss Reduction No. of
Nodes with
under
voltage
problem
Real
(kW)
Reactive
(kVAr)
Real
(kW)
Reactive
(kVAr)
1 0.0029 + 0.0048i 0 25.8606 201.2971 134.2188 1.3800 0.9222 19
2 0.0164 + 0.0272i 0 127.7100 195.1452 130.1111 7.5319 5.0299 14
3 0.0234 + 0.0355i 0 122.1400 193.5001 128.8989 9.1770 6.2421 10
4 0.0300 + 0.0464i 0 149.4800 191.1581 127.2746 11.5190 7.8664 3
5 0.0454 + 0.0737i 0 224.2600 185.5541 123.4177 17.1230 11.7233 0
6 0.0433 + 0.1321i 0 168.0000 191.5033 127.5493 11.1738 7.5917 8
7 0.0455 + 0.1501i 0 153.8000 192.0447 127.9254 10.6324 7.2156 9
8 0.0473 + 0.1769i 0 137.8100 192.7451 128.3799 9.9320 6.7611 10
9 0.0504 + 0.1904i 0 135.1200 192.7968 128.4073 9.8803 6.7337 11
10 0.0524 + 0.1853i 0 119.9300 193.7173 129.0231 8.9598 6.1179 12
11 0.0508 + 0.2016i 0 118.1900 193.8112 129.0839 8.8659 6.0571 13
12 0.0478 + 0.2411i 0 122.0100 193.5202 128.8801 9.1569 6.2609 14
13 0.0390 + 0.2833i 0 121.9500 193.5172 128.8754 9.1599 6.2656 15
14 0.0000 + 0.3194i 0 93.8061 195.4742 130.2074 7.2029 4.9336 16
15 0.0000 + 0.3627i 0 82.9298 196.2188 130.7063 6.4583 4.4347 17
16 0.0000 + 0.3790i 0 61.9845 197.7337 131.7331 4.9434 3.4079 18
Bus
no
Voltage in p.u
Bus
no
Voltage in p.u
Without
SSVR
With
SSVR
Without
SSVR
With
SSVR
1 1.0000 1 18 0.9131 0.9654
2 0.9970 0.9971 19 0.9965 0.9966
3 0.9829 0.9834 20 0.9929 0.9930
4 0.9755 0.9763 21 0.9922 0.9923
5 0.9681 0.9692 22 0.9916 0.9917
6 0.9497 1 23 0.9794 0.9799
7 0.9462 0.9967 24 0.9727 0.9732
8 0.9413 0.9921 25 0.9694 0.9699
9 0.9351 0.9862 26 0.9477 0.9982
10 0.9292 0.9807 27 0.9452 0.9957
11 0.9284 0.9799 28 0.9337 0.9849
12 0.9269 0.9784 29 0.9255 0.9771
13 0.9208 0.9727 30 0.9220 0.9738
14 0.9185 0.9705 31 0.9178 0.9698
15 0.9171 0.9692 32 0.9169 0.9690
16 0.9157 0.9679 33 0.9166 0.9687
17 0.9137 0.9660

17 0.0244 + 0.3526i 0 34.8108 199.8109 133.1544 2.8662 1.9866 19
18 0.0035 + 0.0079i 0 3.4196 202.6590 135.1281 0.0181 0.0129 21
19 0.0069 + 0.0162i 0 5.2032 202.6462 135.1183 0.0309 0.0227 21
20 0.0076 + 0.0179i 0 3.8252 202.6536 135.1233 0.0235 0.0177 21
21 0.0082 + 0.0194i 0 2.0767 202.6639 135.1309 0.0132 0.0101 21
22 0.0194 + 0.0451i 0 51.0598 201.3734 134.3725 1.3037 0.7685 21
23 0.0250 + 0.0617i 0 62.0383 201.0736 134.1913 1.6035 0.9497 21
24 0.0278 + 0.0705i 0 35.2720 201.7125 134.5614 0.9646 0.5796 21
25 0.0496 + 0.0540i 0 99.3743 193.7699 129.0734 8.9072 6.0676 12
26 0.0520 + 0.0545i 0 97.3326 193.7359 129.0308 8.9412 6.1102 13
27 0.0621 + 0.0632i 0 108.0000 192.5758 128.2116 10.1013 6.9294 14
28 0.0693 + 0.0674i 0 111.7200 192.1019 127.8751 10.5752 7.2659 15
29 0.0731 + 0.0645i 0 99.6253 193.0583 128.5330 9.6188 6.6080 16
30 0.0537 + 0.2198i 0 106.3300 192.4493 128.1172 10.2278 7.0238 17
31 0.0567 + 0.2144i 0 67.4496 196.0367 130.5669 6.6404 4.5741 18
32 0.0702 + 0.1538i 0 12.1930 201.4373 134.2829 1.2398 0.8581 20
VI. CONCLUSIONS
The effect of SSVR is considered by
injecting series voltage in quadrature with branch
current in such a way that the load bus voltage
magnitude is maintained at desired value. The
optimum location is the branch which has mitigated
the problem of under voltage in all the load buses
and which has minimum power loss. PSO has been
used to find the optimum rating of SSVR. The
results show the substantial improvement in voltage
profile and considerable reduction in power loss
after insertion of SSVR model in large radial
distribution networks.
REFERENCES
[1] M. Damodar Reddy and V.C. Veera Reddy, “A
two-stage methodology for capacitor placement
for the reconfigured network,” Indian Journal of
engineering and materials sciences.vol 17, pp.
105-112 April 2010,
[2] M. Damodar Reddy and V.C. Veera Reddy.
“Capacitor placement using fuzzy and particle
swarm optimization method for maximum
annual savings,” ARPN Journal of Engineering
and Applied Sciences. vol. 3, no. 3, January
2008.
[3] R H Liang, Y S Wang, “Fuzzy-based reactive
power and voltage control in a distribution
system”, IEEE Transactions on Power Delivery,
vol.18, no.2, pp. 610–618, 2003.
[4] V.Ramalingaiah, M. Damodar Reddy, “Optimal
placement of dg units using fuzzy and real
coded genetic algorithm,” Journal of
theoretical and applied information technology,
2005 – 2009
[5] S. Jazebi, S.H. Hosseinian, B. Vahidi,
“DSTATCOM allocation in distribution
networks considering reconfiguration using
differential evolution algorithm”, Energy
Conversion Management, vol.52, no.7 pp.
2777–2783, 2011.
[6] Ghosh A, Ledwich G, “Compensation of
distribution system voltage using DVR,” IEEE
Transactions on Power Delivery, vol.17, no.4,
pp.1030–6, 2002.
[7] Haque M H, “Compensation of distribution
system voltage sag by DVR and DSTATCOM,”
IEEE Porto power tech conference, vol. 1;
2001. pp. 223–8, 2001.
[8] Mehdi Hosseini, Heidar Ali Shayanfar,
Mahmoud Fotuhi-Firuzabad, “Modeling of
Static Series Voltage Regulator (SSVR) in
Distribution Systems for Voltage Improvement
and Loss Reduction,” Leonardo Electronic
Journal of Practices and Technologies, ISSN
1583-1078, no. 12, pp. 61-82. January-June
2008.
[9] Haque M H. “Efficient load flow method for
distribution systems with radial or meshed
configuration,” IEEE Proceedings – Generation
Transmission and Distribution, vol.143, no.1,
pp. 33–8, 1996.
[10] S. Ghosh and D. Das, “Method for load-flow
solution of radial distribution networks,” IEEE
Proceedings - Generation, Transmission and
Distribution, vol. 146, no. 6, pp. 641, 1999.
[11] Baran M. E. and Wu F.F, “Network
reconfiguration in distribution systems for loss
reduction and load balancing,” IEEE
Transactions on Power Systems, vol. 4, no. 3,
pp. 1401–1407, Aug. 1989.

Author Profile’s:
M. Laxmidevi Ramanaiah is a
Research Scholar in the Department
of EEE, S.V.University, Tirupati,
India. She received her M.Tech and
B.Tech Degree from JNTU
Anantapur, India in 2012 and 2009
respectively. Her research interests
include Power System Optimization and FACTS.
Prof. M. Damodar Reddy is
working as a Professor at
S.V.University College of
Engineering, Tirupati, India. His
areas of interests include Power
System Optimization and reactive
power compensation.
T. Chandra Sekhar is a M. Tech
student in S. V. University College
of Engineering, S.V.University, and
Tirupati, India. He received his
B.Tech Degree from RVR & JC
College of Engineering, Guntur,
India in the year 2013

Optimal Placement of Static Series Voltage Regulator (SSVR) in Distribution System for Voltage improvement Using Particle Swarm Optimization

More Related Content

What's hot (17)

Viewers also liked (9)

Similar to Optimal Placement of Static Series Voltage Regulator (SSVR) in Distribution System for Voltage improvement Using Particle Swarm Optimization (20)

Recently uploaded (20)

Optimal Placement of Static Series Voltage Regulator (SSVR) in Distribution System for Voltage improvement Using Particle Swarm Optimization