Ain Shams Engineering Journal (2014) 5, 103111

Ain Shams University

Ain Shams Engineering Journal

www.elsevier.com/locate/asej
www.sciencedirect.com

ELECTRICAL ENGINEERING

Voltage stability assessment of a power

system incorporating FACTS controllers using
unique network equivalent
Nagendra Palukuru *, Sunita Halder nee Dey, Tanaya Datta, Subrata Paul
Department of Electrical Engineering, Jadavpur University, Kolkata 700032, India
Received 28 January 2013; revised 14 May 2013; accepted 2 July 2013
Available online 16 August 2013

KEYWORDS
Equivalent two-bus p-network model;
Global voltage stability indicator;
OPF;
SVC;
TCSC

Abstract Voltage instability is considered as a main threat to stability, security, and reliability in
the modern power systems. Prediction of voltage stability limit of multi-bus system through its
two-bus equivalent model is a hot topic of the research in the eld of power system operation
and control. This paper presents a novel method to assess voltage stability status using a unique
two-bus p-network equivalent derived with OPF solution of the actual system at different operating
conditions. As the FACTS controllers are now an ineluctable part of power system, this paper considers an SVC and a TCSC in OPF formulation to assess voltage stable states of any interconnected
power system in terms of its reduced two-bus integrated p-equivalent system. Simulation results for
a practical power system establish that the proposed methodology is highly promising to assess
voltage stability in a better way as compared to existing series equivalent model.
2013 Production and hosting by Elsevier B.V. on behalf of Ain Shams University.

1. Introduction
Present-day power systems are highly complex and widespread
and operating much closer to their breakdown limits due to
economical, environmental, political, and technical factors.
This scenario makes the power systems more vulnerable to
stability and security problems. Voltage stability has been
* Corresponding author. Tel.: +91 9804353148.
E-mail address: [email protected] (N. Palukuru).
Peer review under responsibility of Ain Shams University.

Production and hosting by Elsevier

recognized as one of these concerns and is referred to the

ability of a power system to maintain steady acceptable
voltages at all buses in the system under normal operating conditions and/or after being subjected to a disturbance [1,2]. A
system is said to have entered a state of voltage instability
when a disturbance, increase in load demand, or a change in
system condition causes a progressive and uncontrollable drop
in voltage which can occur because of the inability of the network to meet the increasing demand for reactive power [3,4].
Voltage instability is the cause of system collapse, in which
the system voltage decays to a level from which it is unable
to recover. Several large-scale power system blackouts in the
recent past all over the globe have been the consequences of
instabilities characterized by voltage collapse phenomena.
Hence, a proper analysis of voltage stability is essential for
successful operation and planning of the power system.

2090-4479 2013 Production and hosting by Elsevier B.V. on behalf of Ain Shams University.
http://dx.doi.org/10.1016/j.asej.2013.07.002

104
Furthermore, a better voltage stable system contributes to
reliability and reduction of system losses [5].
Over the last few decades, voltage stability is a very crucial
and hot point for both utilities and researchers. Several
techniques have been proposed in the literature for predicting
the voltage instability and collapse of the power system. Voltage stability being problem of power system under steady state
operation, load ow study has long been used to nd voltage
stability and security indicators [6,7]. Continuation power ow
technique enables the researchers to identify critical point of
voltage stability along with maximum system loadability. Load
ow Jacobian and bifurcation analysis have been proved to be
the effective tools to identify voltage collapse point [8,9]. Many
researchers also used the conventional PV or QV curves,
generated from the repetitive load ow solutions with successively increased load, for the assessment of voltage stability
of the critical bus in a power system [1]. A PQ plane of stability is also used as a tool to assess the voltage stability limit of a
power system [10].
Efforts also have been done to develop the concept of deriving single line two-bus equivalent network of any multi-bus
power system to get a quick overview on the system voltage
stability in a global mode [11]. Here, the actual system is reduced into an equivalent two-bus system and then the global
voltage stability indices for indicating the state of the actual
system are easily computed by using the parameters of the
equivalent model. This concept is very attractive due to its simplicity and less computational effort and the occurrence of
voltage collapse on the basis of the single line equivalent can
be studied easily and it is not necessary to consider every line
or bus of the system separately. In [12,13], the authors have
proposed a method to reduce the given power system into its
single line equivalent model on the basis of the equations of
load ow solution and total power loss of the actual system
for the assessment of voltage stability in the global scenario.
Signicant research has also been devoted for developing different voltage stability indicators to study the occurrence of
a voltage collapse, using the concept of reduced two-bus equivalent [1418]. This single line equivalent system is simply a
power line having series equivalent impedance which is obtained by lumping series impedances and shunt admittances
of transmission lines altogether available from the results of
load ow study performed on the actual system. Accurate
assessment of voltage stability is possible if the power system
is faithfully represented by an equivalent two-bus system. As
the series impedances and shunt admittances are lumped, the
series model indicates voltage collapse at higher voltage level
and species no appreciable change with increase in system
load. Also, in case of compensated power systems, the equivalent series impedance of series model results in capacitive line
impedance which apparently indicates over series capacitive
compensation of the actual system. But it does not represent
the actual scenario. So, it is hard to draw any sharp inference
regarding voltage stability from the series equivalent model.
On the other hand, the advances in exible ac transmission
system (FACTS) controllers have led their applications in
improving the overall performance of power networks [19].
Several studies analyzing the application of FACTS controllers
for voltage and angle stability have been reported in [20].
Among the well known FACTS controllers, SVC, TCSC,
and STATCOM are the most widely used controllers for
effective improvement of voltage stability and so the overall

N. Palukuru et al.
stability of the power system. To analyze the effect of these
controllers, steady state models have been developed over
the decades. Load ow analysis using such models would provide data necessary to calculate the voltage stability indicators
in order to evaluate the response of the system at any particular operating point.
On the basis of the previously mentioned literature, this paper proposes a unique p-network two-bus equivalent model for
a multi-bus power system for more accurate assessment of
voltage stability in global scenario. The proposed model is obtained by an innovative methodology considering the effect of
shunt branch admittances and series impedances of transmission lines separately available from the optimal power ow
solution of the multi-bus power system. In addition, this paper
considers the incorporation of steady state models of SVC and
TCSC controllers into the optimal power ow program to
investigate their impact on power system performance in terms
of voltage stability assessed in the equivalent two-bus domain
by digital computer simulation. Newtons method of OPF is
adopted here to calculate the system state variables for different operating conditions of the multi-bus power network considering economic criteria [21,22]. The simulation also includes
the identication of the weakest load bus and determination of
the global voltage stable states of the system following the
derived unique p-network two-bus equivalent system. Simulation results obtained with proposed approach are compared
with the results of well-established series equivalent system.
Improvement in voltage stability margins using the FACTS
controllers is also compared for the test system considered.
The proposed concept has been tested in a wide range of power
networks of varying sizes. In this paper, a real life power
system (203-bus Indian Eastern Grid) has been considered as
the test system to illustrate the utility of the proposed method.
2. Modeling of FACTS controllers
The FACTS technology has provided the power system greater
control of power, secure loading of transmission lines, greater
ability to transfer power, prevention of cascading outages and
damping of power system oscillations [19]. Among the important FACTS controllers, SVC and TCSC are most suitable for
the voltage control. This paper considers the steady state models of SVC and TCSC controllers in OPF formulation which
are discussed in the following section.
2.1. Static var compensator
Static var compensators have been extensively used in power
system applications to provide the controlled reactive power
and voltage stability improvement. The SVC ring angle model has been used here for optimal power ow analysis [22,23]. It
is made up of the parallel combination of a thyristor controlled
reactor (TCR) and a xed capacitor. The SVC is connected to
the transmission network via a step-down transformer as
shown in Fig. 1.
The SVC is considered as a continuous, shunt variable susceptance, which is adjusted in order to achieve a specied voltage magnitude while satisfying constraint conditions. Suitable
control of the equivalent reactance is brought about by varying
the current through the TCR by controlling the gate ring
instant of thyristors and thus the equivalent susceptance Bt_svc

Voltage stability assessment of a power system incorporating FACTS controllers

105

bus-type. However, if a goes out of limits, it is xed at the violated limit and the bus becomes a PQ type bus with xed susceptance connected to it.
The Lagrangian function can be optimized using the following set of equations given in the matrix form:
2

Figure 1 SVC connected to the transmission network via a stepdown transformer.

is a function of the ring angle a. The SVC effective reactance

Xsvc is determined by the parallel combination of XC and Xtcr
and is given by
Xsvc XC
p

XC XL
2p  a sin2a  XL

The partial derivatives required to calculate load ow Jacobian

with respect to the SVC (connected at mth bus) ring angle a
are,
@Pm @Pt svc
@Gt svc

V2m
@a
@a
@a
@Qm @Qt svc
2 @Bt svc

Vm
@a
@a
@a
where net active power injected at node m is given by
Pm active power injected by lines connected to the node
Pt svc
and net reactive power injected at node m is given by
Qm reactive power injected by lines connected to the node
Qt svc :
Here, Gt svc jBt svc 1=Rt jXt Xsvc
The Lagrangian function including the a iteration model of
SVC in OPF [22,23] is given below, where Qm (SVC is connected to the mth bus of the network) is a function of the thyristor ring angle a as well as bus voltage magnitudes |V| and
phase angles d (transformer resistance Rt and hence Gt_svc is assumed to be negligible).
NG
N
X
X
LPg ; jVj; d
Fc Pgi
kpi Pi jVj; d  Pgi
i1

 Ploadi 

i1
N
X

kqi Qi jVj; d  Qgi

@2 L
@Pgi @kpk

@ L
@di @dk

@ L
@di @kpk

@ L
@di @kqk

@ L
@di @jVk j

@2 L
@kpi @dk

@2 L
@kpi @jVk j

@ L
@jVi j@kqk

@2 L
@jVi j@jVk j

@2 L
@kqi @jVk j

@2 L
@a@kqk

@ L
@jVi j@dk

@ L
@jVi j@kpk

@2 L
@kqi @dk

2 @L 3
 @Pg
72 DPg 3
i
7
6
7
i
0 76
6  @L 7
7
76 Ddi 7 6 @di 7
7
7 6  @L 7
7
0 76
Dkpi 7 6
@kpi 7
76
76
76
7
7 6
6
6  @L 7
Dk
qi 7
0 7
6
@jVi j 7
76
7 6
7
6
7
5
4
@L
DjV
j
i
6  @k 7
@2 L 7
qi 5
4
@kqi @a 7
5 Da
 @L
@2 L
@a
0

 Qloadi  kqm Qm jVj; d; a  Qgm

2

where Pgi ; Qgi are the real and reactive power generations at ith
generator bus; NG, the total number of generators; Fci , the cost
of generation at ith generator; where Pi,Qi are the active and
reactive power injections at ith bus respectively; kpi and kqi
are Lagrangian multipliers for active power and reactive power
balance at the ith bus, respectively.
If a is within limits (90 6 a 6 180), the specied voltage
magnitude at the mth bus is attained and it remains a PV

@a2

At the end of the ith iteration, the variable ring angle a is updated like other state variables as, ai+1 = ai + Dai. Gt_svc
andBt_svc are calculated for the new value of ring angle a, and
hence, the admittance matrix of the system is modied incorporating the change in diagonal term Ymm of the admittance matrix. However, if the new angle a violates any of the limits,
then it is xed at the corresponding limit and a no longer serves
as a state variable, instead the voltage magnitude at bus m which
was a specied variable now becomes a state variable.
2.2. Thyristor controlled series compensator
TCSC is one of the most popular FACTS controllers, which allows rapid and continuous modulation of the transmission line
impedance. Active power ow along compensated transmission
line can be maintained at a specied value under a range of operating conditions [19,24]. Fig. 2 is a schematic representation of
TCSC which consists of a series capacitor in parallel with a thyristor controlled reactor (TCR). TCSC modies the line reactance in order to control the power ow through the line.
The equivalent reactance of the combination of xed capacitor and thyristor controlled reactor is a function of the ring
angle a of the TCR in TCSC and can be represented by the following equation [24],
XTCSC XC C1 f2p  a sin2p  ag  C2
 cos2 p  afx tanxp  a  tanp  ag

4X2LC
XL p

LC
where C1 XC X
; C2
; x xx0 ;
p
XC XL
1
p

x0 LC ; XLC XC XL ; x 2pf and 90 6 a 6 180

Thus, for a line between any two buses n and m with TCSC
in a system with N buses,
Znm RLINE jXLINE XTCSC
So, Ynn, Ynm,Ymn, and Ymm of the admittance matrix include 1/
Znm, of the line nm. We can write,
1
1
RLINE
Xeq

j
Znm RLINE jXLINE XTCSC
DT
DT

iNG1

 Qloadm 

@2 L
@Pg @Pg

6 i k
6 0
6
6
6 @2 L
6 @k @P
6 pi gk
6
6 0
6
6
6 0
6
4
0

where Xeq XLINE XTCSC and DT R2LINE X2eq

bus n

bus m
XC
RLINE

ILOOP
XLINE
XL
XTCSC

Figure 2

TCSC connected between bus n and bus m.

106

N. Palukuru et al.

The inclusion of ring angle model of TCSC in optimal

power ow formulation requires an additional constraint to
maintain the power ow through the controlled line at the
specied value. Active and reactive power injections at both
the ends of the controlled line now becomes a function of thyristor ring angle a as well as bus voltage magnitudes |V| and
angles d. So, the Lagrangian function can be modied as
NG
N
X
X
LPg ; jVj; d
Fc Pgi
kpi Pi jVj; d  Pgi
i1

i1
N
X

 Ploadi 

kqi Qi jVj; d  Qgi

iNG1

 Qloadi  lPnm jVj; d; a  Pnmsp 

The equations necessary for solving optimal power ow

including the TCSC at any particular line as per Newtons
method can be written in matrix form as [24,25]
2

@2 L

6 @Pgi @Pgk
6
6 0
6
6 2
6 @L
6 @kpi @Pgk
6
6
6 0
6
6
6 0
6
6
6 0
6
4
0

@2 L
@Pgi @kpk

@2 L
@di @dk

@2 L
@di @kpk

@2 L
@di @kqk

@2 L
@di @jVk j

@2 L
@di @l

@2 L
@kpi @dk

@2 L
@kpi @jVk j

@ L
@jVi j@dk

@ L
@jVi j@kpk

@ L
@jVi j@kqk

@2 L
@jVi j@jVk j

@2 L
@jVi j@l

@2 L
@kqi @dk

@2 L
@kqi @jVk j

@2 L
@l@dk

@2 L
@l@jVk j

@2 L
@a@jVk j

@2 L
@a@l

@ L
@a@dk

@ L
@a@kpk

@ L
@a@kqk

2 @L 3
 @Pg
72
3
i
7 DPgi
6
7
7
@L 7
76 Dd 7 6
6  @di 7
6
7
7
i
7
@ 2 L 76
7 6
6  @L 7
@kpi @a 76 Dkp 7
6 @kpi 7
i 7
76
6
7
2
6
7
7
@ L
@L 7
 @jV
6 Dkqi 7 6
@jVi j@a 7
ij 7
7 6
76
6
7 6  @L 7
76
7
@ 2 L 76 DjVi j 7
@kqi @a 76
7 6 @kqi 7
7
74 Dl 5 6
@L 7
6
2
@ L 7
4  @l 5
@l@a 7
Da
5
2
 @L
0

@2 L
@di @a

@ L
@a2

@a

At the end of the ith iteration, the variable ring angle a and
Lagrangian multiplier l is updated like other state variables
as, ai+1 = ai + Dai and li+1 = li + Dli. XTCSC is calculated
for the new value of ring angle a and hence the admittance
matrix of the system is modied, incorporating the change in
diagonal term Ymm, Ynn and off-diagonal term Ynm and Ymn
of the admittance matrix. If a is within limits
(90 6 a 6 180), the specied active power ow is attained.
However, if a goes out of limits, it is xed at the violated limit
and active power ow through the line is uncontrolled and
determined by xed reactance of the TCSC.
The total power loss of the entire multi-bus power network
being the algebraic sum of all line ows in the system, the
power balance equation for multi-bus power system is given by
Sg Sload Sl
also,
Sl Sse Ssh
where Sl is total complex power loss; Sse and Ssh are total complex series and shunt losses; Sg and Sload are total complex
source and load powers, respectively.
3. Evaluation of two-bus p-equivalent model and formulation of
global voltage stability indicator
Let us consider a two-bus system where sending end bus is assumed as a generator bus and receiving end bus a load bus
along with the series and shunt admittances representing the
equivalent of the entire multi-bus network as shown in
Fig. 3. The behavior of the proposed two-bus equivalent system will be the same as that of multi-bus network and it should
reect the common properties of original system and make
possible the evaluation of voltage stability [1218]. Therefore,

Ise

Is m
Vs
Sg
=Pg+jQg

Figure 3

Ishs

Ir

n
Ishr

Zse_eq

Y
Ysh_eq = sh
2

Ysh_eq

Vr
Sload
=Pr+jQr

Equivalent two-bus p-network system.

power balance equation for the two-bus equivalent network

can be written as
!!
Sg Pg jQg Vs Is  Sse Ssh Sload
! ! !
! !
! !
where Sse Vs  Vr Ise  and Ssh Vs Ishs  Vr Ishr 
Applying Kirchoffs current law at node m (Fig. 3), it can
be written that
!
!
Sg
Vs
!
Ise !  Ssh !
7
!
Vs
jVs j2 jVr j2
Similarly at node n,
!
!
Vr
Sload
!
Ise Ssh !
! !
Vr
jVs j2 jVr j2

where Vs, Vr andIs, Ir are the sending and receiving end voltages and currents; Ise is the current through series equivalent
impedance; Ishs, Ishr are the shunt branch currents at sending
and receiving end sides, respectively.
Substitution of Eq. (7) into Eq. (8) yields the following
! !
! !
! !
! !
Sg jVs j2 Vr Sg jVr j2 Vr  Ssh jVs j2 Vr  Ssh jVr j2 Vs
! !
! !
 Sload jVs j2 Vs  Sload jVr j2 Vs
0

Eq. (9) is dened here as the Global State Equation (GSE)

of the proposed two-bus p-network equivalent for any multibus power system and this can be separated into real and imaginary parts for solving receiving end voltage |Vr| and angle d by
assuming sending end voltage Vs 1\0 and Vr jVr j\d.
Then, the parameters of equivalent two-bus power network
can easily be determined as follows:
! !
Vr
Equivalent series impedance, Z se eq V s!
;
I se
! !
I shr I shs .
and equivalent shunt admittance, Y sh eq !
!
Vr
Vs
Thus, the equivalent two-bus p-network model is developed
at any particular network and load conguration where total
interconnected system has been replaced by a single line twobus system with same generation, load, and loss. The variables
of this equivalent network can be employed to assess the
behavior of the actual system as a whole, i.e., in global mode
at any particular operating condition.
Once the global two-bus p-network equivalent for the multi-bus power system is obtained, the global voltage stability
indicator could be formulated in a straightforward manner
from the parameters of the global network as follows:
The ABCD parameters of the two-bus p-equivalent system
are given by

Voltage stability assessment of a power system incorporating FACTS controllers

YZ
AD1
; B Z; C
2

 
YZ
Y 1
assuming Zse
4
Also assuming
A jAj\a; B jBj\b;
!
j Vr j\d; d < h

eq

Z and Ysh

!
!
Vs j Vs j\h;

eq

Y
2

ance (Zse_eq), equivalent shunt admittance (Ysh_eq), and

other parameters for the two-bus p-equivalent and hence
the global voltage stability margin.
Step 5. Increase the load of weakest bus by small steps at a
constant power factor and go to step 3.
Step 6. Stop.

!
Vr

5. Simulation results and discussion

Sending end voltage being constant (1\0 p.u.), the active

and reactive power at receiving end are given by
!
!
j Vr j
jAjj Vr j2
Pr
cosb d 
cosb  a
jBj
jBj
!
! 2
j Vr j
jAjj Vr j
sinb d 
sinb  a
Qr
jBj
jBj
Jacobian matrix of above power ow equation is given by
" !
#
!
1 j Vr j sinb d cosb d  2jAjj Vr j cosb  a
J
!
jBj j !
Vr j cosb d sinb d  2jAjj Vr j sinb  a
Then, at critical point of voltage stability, the determinant of
Jacobian matrix [2527], D[J] = 0
)

!
jVr j Vcr

1
2jAj cosd a

107

10

where Vcr is the critical value of receiving voltage at voltage

stability limit. Lower value ofVcr indicates that the system will
have better voltage prole along with higher load catering
capability and therefore better voltage stability.
To maintain global voltage stability, D[J] > 0; i.e., global
voltage stability margin can be dened as GVSM = D[J]. It
indicates how far the present operating condition is from the
global system voltage collapse, i.e., GVSM points on the global
voltage security status of the present operating condition.
In real time operation of power systems, the power control
centers would have the information of the various system state
measurements. Such information is similar to the result summary provided by the load ow or optimal power ow study.
So, the equivalent system can easily be evaluated and employed to assess the behavior of the system as a whole, i.e.,
in global mode without computation of Jacobian or Hessian
matrix. Therefore, the representation of any multi-bus power
system in an equivalent domain enables the fast assessment
of voltage stability and so useful for the practical on-line monitoring of power systems.
4. Algorithm
The proposed algorithm for the system simulation is given
below:
Step 1. Solve optimal power ow for base case load and
determine the weakest bus of the given multi-bus system.
Step 2. Make necessary modications in the bus admittance
matrix and Hessian matrix for incorporating SVC or TCSC.
Step 3. Solve optimal power ow problem to obtain the system states. Go to step 6 if OPF iterative process does not
converge.
Step 4. Calculate total generation, load, and transmission
line loss of the system. Calculate equivalent series imped-

The proposed algorithm has been tested on the West Bengal

State Electricity Board (WBSEB) 203-bus Indian Eastern Grid
system, which has a base load of 8887.48 MVA with 24 generators, 35 three-winding transformers, 37 two-winding transformers and 108 load buses which are interconnected by 267
transmission lines. The single line diagram of the test system
is given in Appendix A (Fig. 12). A computer software program has been developed in the MATLAB (Version 7.0) environment to perform the optimal power ow analysis including
the steady state models of SVC and TCSC. The reactive power
sensitivity analysis (dQ/dV criterion) [25,26] is used here to
diagnose the weakest bus of the system and the load bus
having minimum reactive power sensitivity is assumed to be
the weakest bus. This analysis reveals that bus number 172
as the weakest bus of the system. Proposed methodology is
applicable to any load bus of the power system; the weakest
bus is chosen here to present the simulation as it has highest
sensitivity toward voltage stability.
First, the optimal power ow is successively solved for
uniformly increasing load conditions (at an increment of 20%
of base value keeping the load power factor constant) at the
weakest bus until the OPF algorithm fails to converge. The
OPF problem is then similarly solved for the application of
SVC at the weakest bus or TCSC at critical line connected to
the weakest bus. It should be noted that only one FACTS controller at a time has been considered here for the simulation
purpose. For each case and each load set, the parameters of
two-bus p-equivalent model have been calculated and have
been used to assess the voltage stability of the actual system
in global scenario by calculating the global voltage stability
margin, global receiving end voltage, and global critical voltage. The parameters adopted for SVC and TCSC are given in
Appendix B. The results are quite encouraging and highly
promising as the proposed model can assess the voltage stability of any power system at any operating point, more accurately
compared to currently available two-bus series equivalent
model [1218]. The proles shown below validate this fact.
Fig. 4 exhibits the prole of global voltage stability margin
(GVSM) for WBSEB grid system indicating that the system
gradually moves toward voltage instability with increase in
load. Improved proles have been obtained with the application of FACTS devices. It is clear from the gure that with
the incorporation of SVC at weakest bus (no. 172) of the system, the GVSM has been improved with better loading catering capability. It is also observed here that with the application
of TCSC in a critical line (line no. 21 which is connected between buses 156 and 152), more power ow than original is
possible at each load set. TCSC enables the system to transmit
more power maintaining global voltage stability.
It is noted here that the parameters of the p-equivalent system vary with variation of system operating condition. At the

108

N. Palukuru et al.
Normal system
with SVC at bus 172

Normal system
with SVC at bus 172

with TCSC at 156-152

0.5652

7.2

Global critical voltage, p.u.

Global voltage stability margin (10 )

with TCSC at line 156-152

7.1

6.9

6.8

6.7
88.8

0.5649

0.5646

0.5643

0.564

0.5637
88.8
88.92

89.04

89.16

89.28

88.92

89.4

89.04

89.16

89.28

89.4

System operating load, p.u.

System operating load, p.u.

Figure 4 Prole of global voltage stability margin with the

variation of system operating load.

maximum loading point, the value of GVSM becomes zero

indicating that the system moving toward voltage instability.
A set of pre-calculated values of the GVSM corresponding
to different operating condition may be useful for the real time
operation where only the total line loss, total generation, and
total load of the entire system will be sufcient for calculating
the present indicator value from the measured system data
which if compared with the already pre-calculated GVSM data
may reveal whether system is at the verge of voltage collapse or
not, almost instantly. Thus, this approach may be benecial
due to its simplicity associated with high speed of decision
making.
To make a direct comparison between the well-established
series equivalent model and the proposed model, the proles
of global critical voltage (Vcr) for both models with variation
in operating load at the weakest load bus of the test system
are shown in Figs. 5 and 6, respectively. As the series impedance and shunt admittances of transmission lines are lumped
within the series impedance for the series equivalent two-bus
model, the proles for this model indicate voltage collapse at
higher voltage level and specify no appreciable change in Vcr
with increase in system load at the weakest bus. But in the case

Figure 6 Prole of global critical voltage with the variation of

system operating load (p-equivalent model).

of p-equivalent model, it is observed that Vcr is very much susceptible to change in system operating load at the weakest bus.
Here, the global critical voltage increases with increase in load
indicating more threatening operating condition since voltage
collapse occurs even at higher voltage magnitude. So, it is possible to take account of system status change in a better way in
the proposed model. It is also clear from the gure that with
the application of SVC or TCSC, the prole goes downward
which ensures the more stable system. Further, gures suggest
that the incorporation of the SVC enables the system to be stable even at much higher loading. But the TCSC provides best
voltage stability margin at all operating condition. Therefore,
selection of FACTS devices depends upon the system operating criteria.
Figs. 7 and 8 exhibit the variation in that global receiving
end voltage (Vg) for the existing series two-bus equivalent
and the proposed p-equivalent two-bus models respectively
for the load variation at the weakest bus of the test system.
The global receiving end voltage gradually decreases with
enhancement of load indicating that the system approaches local voltage collapse at the equivalent receiving end. The receiving end voltage for series equivalent has higher level indicating
better voltage stability limit, whereas higher value of the
critical voltage opposes this fact indicating voltage collapse
Normal system
with SVC at bus 172

with TCSC at line 156-152

Global receiving end voltage, p.u.

0.982

0.98

0.978

0.976

0.974

0.972
88.8

88.92

89.04

89.16

89.28

89.4

System operating load, p.u.

Figure 5 Prole of global critical voltage with the variation of

system operating load (series equivalent model).

Figure 7 Prole of global receiving end voltage with the

variation of system operating load (series equivalent model).

Voltage stability assessment of a power system incorporating FACTS controllers

Normal system
with SVC at bus 172

Normal system
with SVC at bus 172

with TCSC at line 15

0.843

Reactive power loss, p.u.

Global receiving voltage, p.u.

with TCSC at line 156-

1.6

0.845

0.841

0.839

0.837

0.835
88.8

109

88.92

89.04

89.16

89.28

1.2

0.8

0.4

0
88.8

System operating load, p.u.

88.92

89.04

89.16

89.28

System operating load, p.u.

Figure 8 Prole of global receiving end voltage with the

variation of system operating load (p-equivalent model).

at lower system loading. So, it is hard to draw any sharp inference regarding voltage stability from series equivalent model.
Actually in case of series equivalent, the critical voltage only
depends on power factor of load, the effect of system parameters are not accounted for in this model. It is clear from the gures that with inclusion of SVC, there is a sharp improvement
in voltage stability along with higher loading capability. Also,
the application of TCSC indicates improved voltage prole
though its actual signicance lies in its capability of handling
increased power ow and hence increased stability even under
stressed condition.
The bus voltage magnitude in Fig. 9 reveals that the voltage
corresponding to the weakest bus is gradually decreasing and
thereby approaches voltage instability for increase in system
loading. A atter voltage prole is possible where SVC is connected at the weakest bus of the system with better load handling capacity. The bus voltage starts drooping when SVC
reaches its ring angle limit. Fig. 9 also suggests that an
improvement in voltage prole with the incorporation of
TCSC though its actual signicance lies in its capability of
handling increased power ow and hence increased stability
of the system even under stressed condition. It is observed here
that the voltage prole is signicantly improved with the incorporation of SVC compared to TCSC.
Normal system
with SVC at bus 172

Figure 10 Variation of reactive power loss with the variation of

system operating load.

Figs. 10 and 11 exhibit the proles of reactive power loss

and active power loss, respectively, for the variation of system
operating load indicating that the losses are reduced with the
application of FACTS devices. Moreover, it is observed from
the gures that with the application of SVC, the losses are reduced in a better way compared to TCSC under every operating condition.
The proposed network equivalent could be evaluated if the
total generation, load, active and reactive losses, and voltages
are available. Employing the variables of this network equivalent, the global voltage stability margin can be calculated directly without any computation of the Hessian matrix. In the
real time operation of power systems, the power control centers
would be fed, through the communication links, the various
system state measurements. Such information is similar to the
result summary provided by the OPF or load ow study. So,
the equivalent two-bus model for the actual system can easily
be evaluated and employed to assess the behavior of the system
as a whole, i.e., in global mode. Hence, the proposed method
enables the fast assessment of voltage stability and is useful
for the practical on-line monitoring of power systems. It should
be made clear here that the OPF has been used in this paper
only as a tool to get the information of total generation, active
Normal system
with SVC at bus 172

with TCSC at line 156-152

with TCSC at line 156-152

1.935

Active power loss, p.u.

Weak bus voltage, p.u.

1.05

0.95

0.85

0.75

1.917

1.908

1.899

0.65

0.55
88.8

1.926

88.92

89.04

89.16

89.28

89.4

System operating load, p.u.

Figure 9 Prole of weak bus voltage with the variation of system

operating load.

1.89
88.8

88.92

89.04

89.16

89.28

System operating load, p.u.

Figure 11 Variation of active power loss with the variation of

system operating load.

110

N. Palukuru et al.

and reactive losses, and voltages of the actual system. In practice, the GVSM may be calculated using the information available at the power control centers and not through any OPF or
load ow runs. Therefore, the representation of any multi-bus
power system in an equivalent domain is very useful to assess
the overall voltage stability status of the system and to enable
the system operators or engineers to take quick action to avoid
any incidents of voltage collapse.

of any multi-bus power network in an equivalent domain is

very useful to assess overall voltage stability status of the system. Incorporation of steady state models of SVC and TCSC
in OPF analysis helps to examine their inuence in the equivalent mode and hence in the global voltage stability which in
due course helps the system operator to take quick decision
to select the suitable FACTS device depending upon particular
voltage stability related problem.

6. Conclusion
Appendix A.
A novel methodology is proposed in this paper to assess the
voltage stability of a multi-bus power system using a unique
two-bus p-network equivalent model obtained with optimal
power ow solution of the actual system at different operating
conditions. In addition, steady state models of the SVC and
TCSC have been developed and integrated into OPF program
to investigate their effect on voltage stability and security of
any multi-bus power system in terms of its reduced two-bus
integrated p-equivalent system. The equivalent network
parameters are able to sense any type of change in system status in better way as compared to the well-established two-bus
series equivalent methodology and thus extremely helpful in
assessment of global voltage stable states of the entire network
following any disturbance in the load structure of the network.
Simulation results clearly demonstrate that the representation

Figure 12

The single
. line diagram of the WBSEB grid system is given in
Fig. 12
Appendix B.
The SVC parameters adopted are the following: (1) Transformer reactance Xt = 0.334 p.u. (2) Transformer resistance
Rt = 0 p.u. (3) Inductor reactance for the TCR,
XL = 0.8741 p.u.
and
(4)
Capacitive
reactance,
XC = 3.2484 p.u. The maximum capacitive susceptance obtained is BSVC_max = 0.3431 p.u., i.e., 34.31 MVar is the maximum reactive power that the SVC can inject at 1.00 p.u.
terminal voltage. Resonance for the values adopted for the
SVC model occurs at about 128. Thus, an initial value of
140 has been adopted for the ring angle a.

Typical Indian Eastern Grid (WBSEB 203-bus) system.

Voltage stability assessment of a power system incorporating FACTS controllers

The TCSC parameters adopted for line no. 21, i.e., line
156152 of WBSEB system with line resistance 0.0034 p.u.
and line reactance 0.0128 p.u. are (1) Capacitive reactance,
XC = 0.005 p.u.
and
(2)
Inductive
reactance
XL = 0.00135 p.u. for which resonance occurs at 133.2346.
References
[1] Kundur P. Power system stability and control. New York:
McGraw-Hill; 1994.
[2] Hamouda A, Zehar K. Stability-index based method for optimal
Var planning in distribution feeders. Energy Convers Manage
2011;52(5):207280.
[3] Van Custem T. A method to compute reactive power margins with
respect to voltage collapse. IEEE Trans Power Syst 1991;6(1):14556.
[4] Raou H, Kalantar M. Reactive power rescheduling with
generator ranking for voltage stability improvement. Energy
Conserv Manage 2009;50(4):112935.
[5] De Mala, Goswami SK. Voltage support cost allocation. Energy
Conserv Manage 2011;52(2):118491.
[6] Sauer PW, Pai MA. Power system steady state stability and the
load ow Jacobian. IEEE Trans Power Syst 1990;5(4):137483.
[7] Salama MM, Saied EM, Abdel-Maksoud SM. Steady state
voltage instability assessment in a power system. Energy Conserv
Manage 1999;40(1):5969.
[8] Ajjarapu V, Christy C. The continuation power ow: a tool for
steady state voltage stability assessment. IEEE Trans Power Syst
1992;7(1):41623.
[9] Canizares CA. On bifurcations, voltage collapse and load
modeling. IEEE Trans Power Syst 1995;10(1):51222.
[10] Haque MH. Novel method of assessing voltage stability of a
power system using stability boundary in a PQ plane. Electric
Power Syst Res 2003;64(1):3540.
[11] Nagendra P, Datta T, Halder S, Paul S. Power system voltage
stability assessment using network equivalents a review. J Appl
Sci 2010;10(18):214753.
[12] Jasmon GB, Lee LHCC. Distribution network reduction for
voltage stability analysis and load ow calculation. Int J Electr
Power Energy Syst 1991;13(1):913.
[13] Jasmon GB, Lee LHCC. New contingency ranking technique incorporating a voltage stability criterion. IEE Proc C 1993;140(2):8790.
[14] Gubina F, Strmcnik B. A simple approach to voltage stability assessment in radial networks. IEEE Trans Power Syst 1997;12(3):11218.
[15] Moghavvemi M, Faruque MO. Power system security and voltage
collapse: a line outage based indicator for prediction. Int J Electr
Power Energy Syst 1999;21(6):45561.
[16] Dey S, Chanda CK, Chakrabarti A. Development of a concept of
global voltage security indicator (VSI) and role of SVC on it in longitudinal power supply (LPS). Electr Power Syst Res 2004;68(1):19.
[17] Sahoo NC, Prasad K. A fuzzy genetic approach for network
reconguration to enhance voltage stability in radial distribution
systems. Energy Conserv Manage 2006;47(18):3288306.
[18] Wang Yang, Li Wenyuan, Lu Jiping. A new node voltage stability
index based on local voltage phasors. Electr Power Syst Res
2009;79(1):26571.
[19] Hingorani NG, Gyugyi L. Understanding FACTS: concepts and
technology of exible AC transmission system. Wiley-IEEE Press;
1999.
[20] Abido MA. Power system stability enhancement using FACTS
controllers: a review. Arab J Sci Eng 2009;34(1B):15372.
[21] Sun David I, Ashley B, Brewer B, Hughes A, Tinney WF. Optimal
power ow by Newton approach. IEEE Trans Power Appl Syst
1984;PAS-103(10):286480.
[22] Nagendra P, Halder nee Dey S, Paul S. OPF based voltage stability
assessment of a multi-bus power system using network equivalencing technique, in: Proceedings of 3rd international conference on
power systems (ICPS-2009), IIT Kharagpur, India; 2009.

111

[23] Ambriz-Perez H, Acha E, Fuerte-Esquivel CR. Advanced SVC

models for NewtonRaphson load ow and Newton optimal
power ow studies. IEEE Trans Power Syst 2000;15(1):12936.
[24] Ambriz-Perez H, Acha E, Fuerte-Esquivel CR. TCSC-ring angle
model for optimal power ow solutions using Newtons method.
Int J Electr Power Energy Syst 2006;28(2):7785.
[25] Nagendra P, Halder nee Dey S, Paul S, Datta T. OPF based
global voltage stability assessment of a practical power system
incorporating TCSC controller. J Electr Syst 2010;7(1):10110.
[26] Chakrabarti A, Kothari DP, Mukhopadhyay AK, De A. An
introduction to reactive power control and voltage stability in
power transmission systems. India: PHI Ltd.; 2010.
[27] Arya LD, Choube SC, Shrivastava M. Technique for voltage
stability assessment using newly developed line voltage stability
index. Energy Conserv Manage 2008;49(2):26775.

Palukuru Nagendra obtained his B.Tech. in

Electrical and Electronics Engineering from
JNTU, Ananthapur, M.Tech in Energetics
(Power Systems) from NIT, Calicut and Ph.D
from Jadavpur University, Kolkata, India in
1995, 2004 and 2011 respectively. His research
interest includes Voltage Stability, OPF,
FACTS and ANN applications to Power
System.

Sunita Halder nee Dey received B.E. from

Jalpaiguri Govt. College, M.E. and Ph.D
from BESU, Kolkata, India in Electrical
Engineering in 1998, 2001 and 2006 respectively. She is with Electrical Engineering
Department at Jadavpur University, Kolkata.
She has published several research papers in
international, national journals and conferences. Her research interest includes Power
System Operation and Control, OPF, Voltage
Stability, FACTS applications.

Tanaya Datta obtained her B.E. from Jalpaiguri Govt. College in Electrical Engg, and M.E.
in Power Systems from Jadavpur Univerisity,
Kolkata in 2008 and 2010 respectively. She is
currently working toward the Ph.D degree at
the Department of Electrical Engineering, IIT
Kharagpur, INDIA. Her research interest
includes Voltage Stability, Optimal Power
Flow, and FACTS Applications.

Subrata Paul is Professor with the Department of Electrical Engineering at Jadavpur

University, Kolkata (West Bengal), India. His
work has been primarily concerned with
power system analysis, transient stability and
FACTS applications.