The Effect of SVC-FACTS

Controller on Power System Oscillation

Damping Control
Ghazanfar Shahgholian1, Seyed Mohsen Mirbagheri1, Hosein Safaeipoor1, Mehdi Mahdavian2
1
Najafabad Branch, Islamic Azad University, Esfahan, Iran
2
Naein Branch, Islamic Azad University, Esfahan, Iran
[email protected]

Abstract — Static var compensator (SVC) is used to improve flow to select the best control parameters to minimize the total
the dynamic stability of the system using supplementary control. generation fuel cost and keep the power flows within the secu-
In this paper, dynamic behavior of the SMIB power system with rity limits in power systems incorporating FACTS is presented
SVC has been investigated. The gains of controller were selected
in [11].
based on a pole-placement technique. The IEEE type-ST1 is used
for excitation system and the SVC is modeled by a controllable FACTS controller helps to controlling both real and reactive
susceptance. Several simulations have been done to show the power, which in addition to this control could provide an
effect of the proposed approach to damping controller on power excellent tool for improving power system dynamics. This
system oscillation stability. paper is proposed the dynamic simulation of the SMIB power
system equipped with SVC based on mathematical modeling.
I. INTRODUCTION A control strategy namely power oscillation damping (POD)
Power system damping controllers such as power system of the SVC is also presented to improve damping of the low
stabilizer (PSS) and FACTS based stabilizers may be used to frequency oscillations. Eigenvalues and eigenvectors are used
damp dynamic oscillations and increase damping of the swing to examine the stability of the power system.
modes. PSS increases damping torque of a generator by affect-
ing the generator excitation control, while FACTS devices im- II. SYSTEM MATHEMATICAL MODEL
prove damping by modulating the equivalent power-angle cha- A. SVC Model
racteristic of the system [1]. The main circuit of a SVC is shown in Fig. 1. It consists of
Dynamic application of FACTS controllers includes dynam- four components: a fixed capacitor, bypass inductor, bidire-
ic stability, voltage stability enhancement and transient stabili- ctional thyristor valve and control system. The SVC can be
ty improvement. Among FACTS controllers, the shunt con- operated in voltage regulation mode and in var control mode.
trollers have shown feasibility in term of cost effectiveness in It is capable of injecting controllable reactive power to bus
a wide range of problem-solving from transmission to distrib- where it is connected.
ution levels. In principle, all shunt-type controllers inject addi-
tional current into the system at the point of common coupling.
The SVC is the most widely used FACTS device in power
system regulation. It usually installs in power transmission
systems and serves in various ways to improve the system
performance. SVC is an effective and economical means of
solving problems of transient stability, system reliability, static
voltage stability, dynamic stability and steady state stability in
long transmission lines [2, 3].
Many papers have been published in the field of damping
control for power system oscillation by FACTS controller [4-
7]. A systematic approach for designing of SVC based dampi- Fig. 1. The main circuit of SVC
ng controllers for damping of low frequency oscillations in a
SMIB power system is presented in [8]. In [9] discusses and TCR is a variable inductive reactor controlled by the firing
compares different control techniques for damping undesirable angle α. The equivalent reactance of the TCR, XV(α), at fund-
inter area oscillation in power systems by means of power amental frequency is given by [12]:
system stabilizers, SVC and STATCOM. The simultaneous
coordinated tuning of the FACTS POD controller and the con- π XL
X V (α ) = (1)
ventional PSS controllers in multi-machine power systems is 2 (π − α) + sin 2α
presented in [10], which the linearized system model and the
parameter-constrained nonlinear optimization algorithm is us- where XL is the fundamental frequency reactance of the react-
ed. A hybrid genetic algorithm method to solve optimal power or without thyristor control. For the range of 90 o to 180 o of α,
XV(α) start vary from actual reactance XL to infinity. The Fig. 4 show the critical angle of SVC response for compen-
conduction angle σ of the TCR depends on the α. The valves sation ratio variations, from 2 to 4. We can see the αr is
turn off automatically at the zero crossing of the ac current. between 129.3957o to 149.1471o.
A SVC can be modeled by a variable shunt susceptance. It
can be made to generate or absorb reactive power by means of
thyristor controlled elements. The relationship between σ and
stead state value of the susceptance of the SVC is given as
follows:

1 sin σ − σ
B E (σ) = + (2)
XC πXL

where XC is the reactance of the shunt capacitor and σ=2(π-α).

The limits of the SVC are given by the firing angle limits,
which are fixed by design.
The V-I characteristic is show in Fig. 2. It is specified by
three operating regions: SVC is in regulation range, SVC is fu- Fig. 4. Critical angle
lly capacitive and SVC is fully inductive [13]. The reference
voltage UR is the voltage at the terminal of the SVC when it is B. System Equation
neither absorbing nor generating any reactive power. The sus- The system model for stability analysis is usually that of a
ceptance of the capacitor (slope of OA) is BC and the suscepta- SMIB power system. The configuration of the SMIB with
nce of the reactor (slope of OBC) is BL. The slope of the SVC used in this study is shown in Fig. 5. The equivalent
control characteristics is XSL. A positive value of XSL reduces circuit is show in Fig. 6, where the transfer reactance between
the rating of a SVC.
terminal machine and the infinite bus, XT, is as follows:

X T = XS + X R − BE XSX R (3)

Fig. 2. Voltage–current characteristics of the SVC

The limits of the controller are given by the firing angle li- Fig. 5. SMIB power system with SVC
mits, which are fixed by design. Fig. 3 depicts the values of
the susceptance of the SVC for different values of firing angle
α, for compensation ratio λ=2, 3, 4. Notice that there is a valu-
e of angle critical (αr) that causes steady-state resonance. For
the SVC, resonance is not a problem, as it is shunt connected
to the system and, hence, at the resonant point the controller is
basically switched off the system without any notable effects. Fig. 6. The equivalent circuit

Using the variables defined in SMIB system the following

equations can be written:

(1 − X 2 B E ) U T − U B
IT = (4)
jXT

The power system dynamic model can be written as a set of

differential equations and a set algebraic equation. The non-
linear dynamics of the SMIB system installed with SVC is
given by following set of differential equations, where the
dynamics of the generator is expressed in terms of the second
order electromechanical swing equation and the internal volta-
Fig. 3. Equivalent susceptance ge equation, the exciter by three first order differential equati-
ons and the SVC is modeled as controllable susceptance by a X 2 (I do X 'd − E 'qo )
first order differential equation. Damping effects due to damp- + ΔB E (9)
X DO
er winding and rotor body eddy currents are neglected: 
 

K dc

d
δ = ωo ω Figure 7 shows the block diagram representation of the
dt
small signal performance of the SMIB installed with a SVC.
d 1
ω= (PM − PE − K D ω) GV(s) is the exciter transfer function, GF(s) is the field circuit
dt 2H transfer function and GM(s) is the dynamic machine transfer
(5)
d ' 1 function.
E q = ' [E F − E 'q + (X 'd − X d ) i d ]
dt Tdo
d 1 K
EF = − E F + A (U T − U R )
dt TA TA

where δ is rotor angle, E'q is the machine internal voltage, X’d

and Xq are the direct and quadrature reactance of the gener-
ator.
The active electrical power delivered by the generator is ex-
pressed as:

E 'q U B U 2B (X q − X 'd )
PE = sin δ − sin 2δ (6)
XD 2 XDXQ
Fig. 7. Linearised dynamic model of the SMIB power system with SVC
where XD and XQ are given by:
D. Controller Design
The effectiveness of the SVC in enhancing small-signal sta-
⎧⎪X D = X T + X 'd (1 − X R B E )
(7) bility depends on the location of the SVC, input control signal-
⎨
⎪⎩X Q = X T + X q (1 − X R B E ) s used, and controller design. The structure of the SVC contro-
ller with control loop is shown in Fig. 8.
The stability can be improved by operating the SVC in
inductive (BE<0) or capacitive (BE>0) mode. So the improve
the dynamic performance of the system is achieved by the
control of the susceptance of the compensator (BE), which
changes the electrical power output.

C. System Linear Model Fig. 8. Block diagram of the SVC controller

The linearized model of the power system including SVC
about an equilibrium operating point is used for small signal An SVC with firing control system can be represented by a
analysis and damping controller design. It can be obtained by first order model characterized by a gain KR and time constant
linearizing equations for id and iq around an operating point TR as shown in Fig. 9. The input signal of the controller is the
and then substituting in non-linear equation of the system. The speed deviation and the output signal is the susceptance offer-
d and q components of generator current of SMIB system inst- ed by the SVC. The control signal US is the added modulation
alled with SVC can be expressed as: signal of SVC for damping control and BCR is a reference
signal which is initial value of the SVC. The limits of the
U B cos δ o I qo X 2 X q controller are given by the firing angle limits, which are fixed
Δi q = Δδ + ΔB E (8) by design. Due to fast switching characteristics of thyristors,
X QO X QO
 
 
the time constants of an SVC device are very small. The susc-
K qd K qc eptance of the SVC could be described by:

U sin δ o 1 − X 2 B CO d K 1
Δi d = − B Δδ + ΔE 'q B C = R (B CR − U S ) − BC (10)
X DO X DO dt TR TR

 
  

K dd K de The structure of POD controller has a similar structure to

that of the power system stabilizer controller as shown in Fig.
10. The transfer function of the POD controller is given by:
 are: -0.0999, -2.0355, -5, -18.9530, -22.4732, -77.3181 and -
0.6640±j8.8558.
The system response for a step increase in mechanical
power input with normal loading show in Figs. 11-14.
The system response for a step increase in mechanical
power input with heavy loading show in Figs. 15-18.
Fig. 9. Block diagram of the SVC control model
TABLE I
TW s 1 + T1 s 1 + T3 s POWER SYSTEM PARAMETER
G P (s) = K P ( )( ) (11) Generator
1 + TW s 1 + T2 s 1 + T4 s JM=10, Xq=1.7, Xd=1.8, X'd=0.23, KD=0, T'do=8, f=60
Transmission line
where KP is positive constant gain and TW is the washout time XS=0.5, XR=0.5
Static var compensator
constant. T1 and T2 are the lead and lag time constant of leas- KR=1, TR=0.05
lag block, respectively. Normal loading
PEo=1, QEo=0.015, UTo=1.05
Heavy loading
PEo=1.1, QEo=0.4, UTo=1.05
Exciter
KA=200, TA=0.01

TABLE II
Fig. 10. POD controller SENSITIVITY CONSTANTS OF MODEL POWER SYSTEM WITH NORMAL
LOADING
The POD controller is modeled by state-space analytical Constant Value Constant value
equations: K1 1.5998 Kic 0.5851
K2 0.4481 Kid 0.1621
d K K K K K K
ΔU W = − P 1 Δδ − P D Δω − P 2 ΔE 'q K3 1.1603 Kie -0.0740
dt JM JM JM K4 -0.7035 Kdc -0.1926
1 K K5 1.4704 Kqd 0.9978
− ΔU W + P ΔPM (12)
TW JM K6 0.5641 Kdd -0.4481
Kde -0.0880 Kqc 1.3828
d K K T K K T TK K
ΔU P = − P 1 1 Δδ − P D 1 Δω − 1 P 2 ΔE 'q
dt J M T2 J M T2 J M T2 TABLE III
SENSITIVITY CONSTANTS OF MODEL POWER SYSTEM WITH HEAVY LOADING
1 T1 1 K T
+( − ) ΔU W − ΔU P + P 1 ΔPM (13) Constant Value Constant value
T2 T2 T W T2 J M T2 K1 2.8683 Kic 0.2718
K2 0.3893 Kid 0.1521
d K K TT K K TT K3 1.1603 Kie -0.0806
ΔU S = − P 1 1 3 Δ δ − P D 1 3 Δ ω
dt J M T2 T4 J M T2 T4 K4 -0.6111 Kdc -0.2292
T1T3 K P K 2 T 1 T1 K5 1.6243 Kqd 1.2698
− ΔE 'q + 3 ( − ) ΔU W K6 0.7039 Kdd -0.3893
J M T2 T4 T4 T2 T2 TW
Kde -0.0880 Kqc 1.2012
1 T 1 K TT
+ (1 − 3 )ΔU P − ΔU S + P 1 3 ΔPM (14)
T4 T2 T4 J M T2 T4

III. SIMULATION RESULTS

The SMIB system is implemented for the simulation studies
with and without the controller. Simulation results are present-
ed for a SMIB power system using data in Table I.
Table II summarizes the value of the sensitivity constants
with normal loading. In this case, the eiegnvalues of the
system are: -0.0997, -2.0744, -5, -12.6740, -21.0923, -83.0952
and -1.5861±j4.2618.
Table III summarizes the value of the sensitivity constants
with heavy loading. In this case, the eiegnvalues of the system Fig. 11. Characteristic of voltage-current of the generator terminal with
normal loading
 Fig. 12. Power-angle curve with normal loading Fig. 16. Variation in susceptance of SVC with heavy loading

IV. CONCLUSION
SVC is used to damp out power system oscillations. The
function of the SVC is either to inject reactive power to bus or
to absorb reactive power from the bus where it is connected.
The small signal stability can be improved by SVC.
Simulation results show that it can indeed improve the dam-
ping of the power systems. An eigenvalue-based objective
function to increase the system damping ratio was developed.

REFERENCES
[1] G. Shahgholian, J. Faiz, "Static synchronous compensator for improving
Fig. 13. Variation in susceptance of SVC with normal loading performance of power system: A review", Int. Rev. of Elec. Eng., Vol.4,
No.2, pp., Oct. 2010.
[2] A. Luo, Z. Shuai, W. Zhu, Z. J. Shen, "Combined system for harmonic
suppression and reactive power compensation", IEEE Tran. on Ind. Ele.,
Vol.56, No.2, pp.418-428, Feb. 2009.
[3] E. Zhijun, D.Z. Fang, K.W. Chan, S.Q. Yuan, "Hybrid simulation of
power systems with SVC dynamic phasor model", Elec. Pow. and Ene.
Sys., No.31, pp.175–180, 2009.
[4] X. Lei, E.N. Lerch, D. Povh, "Optimization and coordination of
damping controls for improving system dynamic performance", IEEE
Trans. Pow. Sys., Vol.16, pp.473-480, Aug. 2001.
[5] R.H. Adware, P.P. Jagtap, J.B. Helonde, "Power system oscillations
damping using UPFC damping controller", IEEE/ICETET, pp.340-344,
Nov. 2010.
[6] H.Wang, “A unified model for the analysis of FACTS devices in
damping power system oscillations – part III: unified power flow
controller”, IEEE Tran. On Pow. Del., Vol.15, No.3, pp.978-983, July
Fig. 14. Characteristic of voltage-current of the generator terminal with 2000.
heavy loading [7] P.K. Dash, S.Mishra, G.Panda, “Damping multimodal power system
oscillation using a hybrid fuzzy controller for series connected facts
devices”, IEEE Tran. On Pow. Sys., Vol.15, pp.1360-1366, 2000.
[8] V.K. Chandrakar, S.N. Dhurvey, S.C. Suke, "Performance comparison
of SVC with POD and PSS for damping of power system oscillations",
IEEE/ICETET, pp.247-252, Nov. 2010.
[9] N. Mithulananthan, C.A. Canizares, J. Reeve, G.J. Rogers, “Comparison
of PSS, SVC, and STATCOM controllers for damping power system
oscillations”, IEEE Tran. On Pow. Sys., Vol.18, No.2, pp.786-792, May
2003.
[10] L.J. Cai, I. Erlich, "Simultaneous coordinated tuning of PSS and FACTS
damping controllers in large power systems", IEEE Tran. on Pow. Sys.,
Vol.20, No.1, pp.294-300, Feb.2005.
[11] T.S. Chung, Y.Z. Li, "A hybrid GA approach for OPF with consideratio-
n of FACTS devices", IEEE Pow. Eng. Rev., pp.47-57, Feb. 2001.
[12] S. Biansoongnern, S. Chusanapiputt, S. Phoomvuthisarn, "Optimal SVC
and TCSC placement for minimization of transmission losses",
Fig. 15. Power-angle curve with heavy loading
IEEE/ICPST, 2006.
[13] N. Sabai, H. Nandar Maung, T. Win, "Voltage control and dynamic
Simulation results indicated that the SVC can significantly performance of power transmission system using static var compe-
improve system damping. Based on the time domain results, it nsator", Worl. Acad. of Sci., Eng. and Tec., pp.425-429, 2008.
is obvious that at normal loading is a better option as it damp
out oscillation faster than heavy loading.