2015 5th Nirma University International Conference on Engineering (NUiCONE)

Optimal Placement of TCSC for Improvement of

Static Voltage Stability
Aesha Sheth Dr. C. D. Kotwal Shital Pujara
PG Student, EE Dept. Professor, EE Dept. Asst. Professor, EE Dept.
EE Dept. SVIT, Vasad. SVIT, Vasad. SVIT Vasad.

is a relatively slow process thus being primarily considered as a

Abstract—Voltage instability in power system is a major issue small signal phenomenon. So Voltage stability determination
in electrical power system operation. Voltage instability occurs have been developed on static analysis techniques based on the
mainly due to deficit of reactive power at load buses during some
power flow model since they are simple, fast and convenient to
contingency or under increased loading condition. Flexible AC
use [2].
Transmission System (FACTS) devices can improve the voltage
stability with reduction of losses and generation cost. Thyristor The major factors affecting voltage stability of a power
Controlled Series Compensator (TCSC) is a series FACTS system are its generators reactive power limit, voltage control
device, which can improve voltage stability with improved power limits, characteristics of connected loads, reactive power
transfer capability of the transmission line. Like other FACTS
compensation devices characteristics and their actions. Several
devices, TCSC is costly and it is required to find its optimal
location and size for improvement of voltage stability. Here
methods have been used for static voltage stability analysis
Particle Swarm Optimization (PSO) and PSO with time varying such as the P-V and Q-V curves, modal analysis, artificial
acceleration co-efficient (PSO-TVAC), used to locate a TCSC of neural networks etc. Line stability index (LSI) provides
an optimum rating for voltage stability improvement of power important information about the proximity of the system to
system. Line Stability Index (LSI) is considered here for voltage collapse and also used to identify the critical line of the
optimization purpose which provides the information about the system [3-5].
stability condition of a power system and its proximity to voltage
collapse. The analysis has been demonstrated on a standard Voltage instability [3] can be avoided by: (a) appropriate
IEEE 6-bus system with some modification in load data and load shedding on the consumer network; (b) on load tap
overloading condition using MATLAB. PSO-TVAC gives fast changers; (c) reactive compensation (series and/or shunt). An
convergence than PSO. effective solution to improve the voltage profile and voltage
stability of power system with reactive power compensation is
Keywords— Voltage Stability, FACTS devices, TCSC, Line Stability
Index, PSO, PSO-TVAC
to place an appropriate Flexible AC transmission system
(FACTS) controller [6] in the system. FACTS controllers are
I. INTRODUCTION solid state converters which use reliable high speed thyristor
based control elements and have the capability of controlling
In recent years, due to restructuring of the electricity and improving the various electrical parameters in transmission
industry, increasing load demands, some economic and network i.e. power transmission capacity, voltage profile,
environmental constraints power systems are being operated minimization of transmission losses, power system stability
closer to their stability limits, which may sometimes leads to etc. FACTS controllers are costly. To obtain maximum benefits
voltage collapse. Maintaining a stable and secure operation of a from these controllers, the choice of type of controller and its
power system is therefore a very important and challenging optimal location and size is also a challenging issue [7].
issue. Voltage stability is the ability of a power system to
Thyristor controlled series capacitor (TCSC) is a series
maintain acceptable voltages at all the load buses of the system
FACTS device. It is a series capacitor, connected with thyristor
under normal conditions and also following a disturbance [1].
controlled inductor in parallel so that variable reactance of the
Many analysis are carried out on voltage stability, they are capacitor can be achieved. As other FACTS devices, TCSC is
either static analysis or dynamic analysis. The dynamic also a costly device and hence it is important to place it at
analysis emphasizes on large disturbance or transient stability optimal location and to find its optimal size [10], so that
occurrence, while static analysis is considered as a small signal maximum benefits from it may be achieved. Sequential
phenomenon, load increasing and line outage. Voltage collapse quadratic programming, mixed integer programming and line

978-1-4799-9991-0/15/$31.00 ©2015 IEEE

stability index have been proposed for optimal location and To avoid overcompensation, the working range of TCSC is
sizing of TCSC for voltage stability enhancement. In this paper between -0.7 XL and 0.2 XL [9].
line stability index is used for optimally locate a TCSC in the
system. III. LINE STABILITY INDEX
The line stability index determines the critical line and
This paper gives comparison of Particle Swarm Optimization provides a measure to proximity to the voltage collapse point
(PSO) and a modified Particle swarm optimization technique of the system [10]. In an interconnected system, the value of
with time varying acceleration coefficient (PSO-TVAC) to line stability index that is closed to one indicates the line has
optimally locate a TCSC to enhance static voltage stability. reached its instability limit.
Optimal location and size of TCSC has been determined by
applying PSO and PSO-TVAC for different line outage In any power system, the voltage stability analysis is carried
contingencies one by one to minimize Line Stability Index. out for two purposes: (a) to determine how much any system is
Effectiveness of the proposed approach has been tested on close to its instability limit, and (b) which is the critical line or
IEEE 6 bus system with some modification in load data. weak bus in a power system.

II. MATHEMATICAL MODEL OF TCSC

Derivation of LSI:
Thyristor controlled series compensator is one of the most Moghavvemi [12], derived a voltage stability criterion based
important device of FACTS controllers' family. TCSC is in use on a power transmission concept in a single line. An
for many years to increase line power transfer and to enhance interconnected system is reduced to a single line network and
stability. TCSC can enhance the power system stability by then applied to assess the overall system stability. Utilizing the
effectively controlling the line power flows. Controlling the same concept but using it for each line of the network a
power flows in the system helps in reducing the flows in stability criterion is developed. Consider a single line of
heavily loaded lines, resulting in increased system loadability interconnected network shown in Fig. 2. The line is connected
and improved stability of the power system [8]. to the other lines forming a grid network. Any of the lines from
that network can be represented with the following parameters
The TCSC can serve as the capacitive or inductive
as shown in the Fig. 2.
compensation respectively by modifying the resistance of the
transmission line. In this study, the reactance of the
transmission line is adjusted by TCSC directly. The rated value
s r + jx r
of the TCSC is a function of the reactance of the transmission
line, where the TCSC is located.

, ,
i j
Fig. 2. Typical one line diagram of transmission line
Utilizing the concept of power flow in the line and analyzing
with ‘π’ model representation, the power flow at the receiving
and sending end can be expressed as
Fig. 1. Mathematical Model of TCSC

| || | | |
(1) (4)
(2) | | | || |
(5)
(3)
From these power equations one can separate real and
Where, XL = Reactance of transmission line. reactive power,
XTCSC = Reactance of TCSC. V V V
P cos θ δ δ cosθ (6)
Z Z
rTCSC = Coefficient which represents the
V V V
compensation of degree of TCSC. Q sin θ δ δ sinθ (7)
Z Z
 Putting, δ δ δ into equation 4 and solving it for V V wV C rand P S C rand G S (12)

S S V (13)
.
V V ZQ
V (8) Where,

Vik= Velocity of agent i at kth iteration

Now, for Z sinθ x, we have Vik+1= Velocity of agent i at (k +1)th iteration
W = the inertia weight
.
V V Q
V (9) C1 = C2 = individual and social acceleration constants
rand1= rand2= random numbers (0 to1)

To obtain real values of Vr in terms of Q, the equation must Sik = present position of agent i at kth iteration
have real roots. Thus the following conditions, which can be Sik+1= present position of agent i at (k+1)th iteration
used as a stability criterion, need to be satisfied: Pbesti = Particle best of agent i
V sin θ δ 4xQ 0 (10) Gbest = Global best of the set
Or
B. PSO-TVAC
Q
L 1.00 (11) PSO-TVAC is particle swarm optimization with time
V
varying acceleration co-efficient algorithm is the algorithm
worked on the principle same as that of PSO with the varying
L is termed the stability index of the line. The stability acceleration co-efficient in each of iteration in contrast with the
criterion is used to find the stability index for each line classical PSO where acceleration coefficients are kept constant.
connected between two bus bars in an interconnected network. PSO-TVAC enhance global search of particles in early parts of
Based on the stability indices of lines, voltage collapse can be optimization and allow the particles to converge towards global
accurately predicted. As long as the stability index L remains optimum solution at the end of the algorithm. To achieve so c1
less than 1, the system is stable and when this index exceeds and c2 are varied with time such that cognitive component is
the value 1, the whole system loses its stability and voltage large and social component is small initially and they are
collapse occurs. Thus the proposed method can be used in decreased and increased respectively. This condition allows the
voltage collapse prediction. particles to move around the search space in spite of moving
IV. PSO AND PSO-TVAC towards the Pbest initially and converge towards global optimal
in later parts.
A. Simple PSO
PSO employs a population of individuals known as particles. Following are the formulae of acceleration co-efficient
change in each of iteration.
These particles fly across the problem work field with some
specified initial velocities. The velocities of the particles are C1= C1i + (C1f – C1i)×(iter/max_iter) (14)
stochastically attuned considering, the past best location of the C2= C2i + (C2f – C2i)×(iter/max_iter) (15)
particles and their vicinity best position, in each iteration.
Here, C1i, C1f, C2i and C2f are initial and final values of
These positions are determined according to some predefined
cognitive and social acceleration co-efficient respectively. Due
fitness function and then, the movement of each particle as you
to such variation of acceleration co-efficient, PSO-TVAC
would expect emerges to an optimal or near-optimal solution.
converges faster than classical PSO [8].
Each particle follows the path of its coordinates in the
problem work space which are related with the best solution V. SIMULATION AND DISCUSSION
(fitness), it has attained until now. The fitness value is also Effectiveness of TCSC in the system is analyzed on a 6 bus
recorded and denoted as Pbest. When a particle takes all the test system (modification in standard IEEE 6-bus system) using
population as its geological neighbors, of which the best value power world simulator. This test system is consist of 6 buses,
is a global best and denotes as Gbest. three generators, three loads and eleven transmission line
connected as shown in Fig.3. Data of this system is given in
The particle revises its velocity and positions, with following
APPENDIX.
equation, after getting the two best values.
 2. Calculation of LSI of each line for base case and single
line contingency (12 cases).
3. Implementation of PSO algorithm to find optimal rating
of TCSC at different locations. The objective function of the
algorithm is to minimize (Σ LSI).
Table II shows different PSO parameters values used in
simulation.
TABLE II. PSO PARAMETERS

Parameters Value

Number of Particles 15
Inertia weight Linearly decreased
Individual acceleration co- 2.5
efficient(C1)
Fig. 3. 6-Bus Test System
Social acceleration co-efficient(C2) 4-C1
From the given data, the model of the test system is Number of Iterations(max ) 20
prepared using power world simulator and the power flow is rand1 0.3
checked. With normal operating condition, LSI values for all
rand2 0.2
eleven lines are computed. This normal condition is termed as
Velocity bounds [-2,2]
the base case here. Then after some line outage contingencies
are considered and the power flow is analyzed. Again the LSI Position bounds(Search Space) [-0.7,0.2] (range of
r)
values of each line for these cases are computed. The line
C1 initial and C1 final(for PSO- 0.2 and 2.5
outage providing highest value of LSI is considered as the TVAC) respectively
most severe line outage from the viewpoint of voltage C2 initial and C2 final(for PSO- 2.5 and 0.2
stability. TVAC) respectively
From the Table I, it can be seen that LSI of line no. 8 is
A. Classical PSO applied to 6-Bus Test System
maximum in most of the cases. So compensation should
By applying PSO for different cases, to improve voltage
provide at this line. The TCSC having capacitive reactance
stability, location and optimal rating of TCSC can be found.
0.0091p.u.is placed at line 8. The rating of TCSC is found
Results for all cases are shown in Table III.
using trial and error method such that voltages improve at
each bus. Table shows that after placing TCSC voltage For Base Case, Line 8 is the critical line, the line having
stability is improved i.e. maximum LSI is reduced. maximum LSI value. Similarly for different line outage
contingencies, critical lines are different.
TABLE I. LSI VALUES WITH AND WITHOUT TCSC From these results it is observed that, location of TCSC is
not same as the weakest line for each case. But for all the cases
No Case Critical Max LSI Max LSI
Line with TCSC after placement of TCSC summation of LSI is reduced after
1 Base Case 8 0.3274 0.32061 placement of TCSC. So voltage stability is improved of power
2 Outage of Line 1 8 0.3602 0.3534 system.
3 Outage of Line 2 1 0.3895 0.3892
4 Outage of Line 3 8 0.404 0.3940 B. PSO-TVAC applied to 6-Bus Test System
5 Outage of Line 4 8 0.3471 0.34 With the same parameters as of classical PSO and only keep
6 Outage of Line 5 2 0.4602 0.4591 updating C1 and C2 in each iteration, as the principle of PSO-
7 Outage of Line 6 8 0.3824 0.374
TVAC, same analysis is carried out to find optimal location
From this it is clear that LSI can be used to measure the and rating of TCSC. Application of PSO-TVAC gives the same
voltage stability of the system. results as of PSO (Table III.), both TCSC location and rating.
Analysis is carried out using MATLAB using the following Thus PSO and PSO-TVAC can be used as the substitutes of
steps: each other.

1. Load flow analysis with Newton- Rapshon Method.

 VI. CONCLUSIONS
TABLE III.TCSC PLACEMENT FOR VOLTAGE STABILITY
ENHANCEMENT PSO and PSO-TVAC algorithm locate a TCSC of optimal
rating such that static voltage stability of the system is
TCSC Sum of LSI improved, which is indicated by reduced sum of LSI. It is not
Line Critical TCSC Optimal
Outage Line Location Rating Without necessary that critical line of the system is optimum for
With
(pu) TCSC TCSC placement of TCSC. PSO-TVAC converges fast as compared
Base
to PSO with the same optimization, so one can apply PSO-
8 9 0.0699 2.0861 1.8811
Case TVAC in place of classical PSO. The test system taken is a 6
1 8 6 0.1859 1.8977 1.6675 bus system. Further the PSO-TVAC can be implemented for
2 1 3 0.06 2.2803 2.2249
higher bus number system and its superiority compared to
conventional PSO can be verified.
3 8 6 0.1902 2.2764 2.0737
4 8 9 0.0343 1.9943 1.8759 REFERENCES
5 2 2 0.1399 2.7508 1.9626 [1] P. Kundur, Power System Stability and Control, McGraw-Hill,
1993.
6 8 3 0.2097 2.1883 1.9453
[2] C. W. Taylor, Power System Voltage Stability, McGraw-Hill,
7 8 2 0.1399 2.0109 1.8264
1994.
8 6 3 0.2098 2.2765 1.9633 [3] C. Reis and F.P. M. Barbosa, "A Comparison of Voltage
9 6 7 0.1398 2.7140 2.2182 Stability Indices," Electro technical Conference IEEE
10 8 9 0.0698 2.0889 1.8936
MELECON 2006, Benalmadena (Malaga), Spain, pp. 1007-
1010,2006.
11 8 9 0.0699 2.0827 1.8490
[4] C. Reis and F.P. M. Barbosa, "Line indices for voltage stability
assessment," IEEE Power Tech Conference, Bucharest,
Romania, pp. 1-6, 2009.
C. Comparison of PSO and PSO-TVAC
[5] M. y. Suganyadevia and C.K. Babulal, "Estimating of
PSO and PSO-TVAC give the same optimization. But
loadability margin of a power system by comparing voltage
convergence of both algorithms is different. Analysis is stability indices," Int. Conf. on Control, Automation,
carried out for the comparison of PSO and PSO-TVAC. For Communication and Energy Conservation, I NCACEC
the base case and for different single line outage contingency 2009,Tamilnadu, India, pp. 1-4,2009.
convergence of PSO and PSO-TVAC is determined. [6] R. Yadav, S. Varshney and L. Srivastava, "Enhancement of
Fig. 4. shows the convergence comparison for PSO and voltage stability through optimal placement of TCSC," Int.
PSO-TVAC for different cases. Journal of Power System Operation and Energy Management,
vol. I , pp. 39- 47,2011
[7] S.N. Singh, Electrical Power Generation, Transmission and
Comparison PSO vs PSO-TVAC
Distribution, PHI Leaning Pvt. Ltd., 2 004.
[8] L. Shrivastava, S.Dixit, Ganga Agnihotri ,“Optimal Location
120 and Size of TCSC for Voltage Stability Enhancement using
No. of Iterations

100 PSO-TVAC” 2014 Power and Energy Systems: Towards

80 Sustainable Energy (PESTSE 2014)
60 [9] Tiwari P.K., Sood Y.R., “Optimal Location of FACTS Devices
40 in Power System Using Genetic Algorithm” Nature &
20 Biologically Inspired Computing, 2009. NaBIC 2009. World
0
Congress.
0 5 10 15
[10] A. Yazdanpanah- Goharrizi, R. Asghari, "A novel line stability
Line Outage index (NLSI) for voltage stability assessment of power
systems," Proceedings 7th WSEAS Int. Conf. on Power
PSO PSO-TVSC Systems, pp. 164- 167,China, September , 2007
[11] G. 1. Rashed, Y. Sun, and H.1. Shaheen, Optimal location and
Fig. 4. Convergence Comparison of PSO and PSO-TVAC parameter setting of TCSC for loss minimization based on
Differential Evolution and Genetic Algorithm", Physics
Procedia, vol. 33, pp. 1864-1878,2012
[12] M Moghavemmi, and F M Omar, “Technique for Contingency
Monitoring and Voltage Collapse Prediction”, IEEE
proceeding on Generation, Transmission and Distribution, Vol.
145, No. 6, pp634-640, 1998.

APPENDIX
6 BUS TEST SYSTEM DATA
(BASE MVA=100)

Table A Demand data of six bus system

Bus Type Real Reactive Base

Number Power Power KV (pu) (pu)
Demand Demand

(pu) (pu)
1 Slack 0 0 230 1.05 1.05
2 PV 0 0 230 1.05 1.05
3 PV 0 0 230 1.07 1.07
4 PQ 0.85 0.8 230 1.05 0.95
5 PQ 0.85 0.8 230 1.05 0.95
6 PQ 0.85 0.8 230 1.05 0.95

Table B Generator Data of six bus system

Bus Real Power Reactive Reactive P P

No. Generation Power power (pu) (pu)
P Generation Generation
(pu) Q Limit
(pu) Q Q
(pu) (pu)
1 0 0 1.0 -1.0 200 50
2 0.5 0 1.0 -1.0 150 37.
5
3 0.6 0 1.0 -1.0 180 45

Table C Branch Data of six bus system

Line Bus Line Line Line

Number Number resistance reactance charging
R(pu) X(pu) admittance
B(pu)
1 1-2 0.1 0.2 0.04
2 1-4 0.05 0.2 0.04
3 1-5 0.08 0.3 0.06
4 2-3 0.05 0.25 0.06
5 2-4 0.05 0.1 0.06
6 2-5 0.1 0.3 0.04
7 2-6 0.07 0.2 0.05
8 3-5 0.12 0.26 0.05
9 3-6 0.02 0.1 0.02
10 4-5 0.2 0.4 0.08
11 5-6 0.1 0.3 0.06