Al-Mansour Journal Issue (18) 2012 (18) ‫ﻣﺠﻠﺔ اﻟﻤﻨﺼﻮر اﻟﻌﺪد‬

The Effect of Power System Stabilizers (PSS) on

Synchronous Generator Damping and
Synchronizing Torques


Asst. Lecturer Fadhil M. Kalaf

ABSTRACT

The synchronizing and damping torque coefficients in the range of the

electromechanical oscillation frequencies provide a measure of the
generators' contribution towards power system stability under small
deviations, and may be taken as stability indices even for multi machine
systems, although their interpretation is rather complex in this case. This
paper presents the effect of Power System Stabilizer (PSS) on the
synchronizing and damping torque coefficients for a synchronous generator
connected to an infinite busbar through impedance; it explicitly includes the
effects of its Automatic Voltage Regulator (AVR) and of the additional
stabilizer signal derived from the rotor speed. .
Results obtained were compared with the conventional (AVR) using
MATLAB 7 programming language.

Keywords: Power System Stabilizer, AVR, synchronizing torque, damping

torque


Technical College
101
Asst. Lecturer Fadhil M. Kalaf

INTRODUCTION

A traditional function of the excitation system is to regulate generator

voltage and thereby help control system voltage. The power system stabilizer
(PSS) is a supplementary controller, which is often applied as part of the
excitation control system. Grid codes and regulatory agencies are
increasingly specifying PSS controls for new generation and retrofit on
existing units. The basic function of the PSS is to apply a signal to the
excitation system, creating electrical torques that damp out power
oscillations.
Although the primary function of the PSS is to supply positive damping
torque contributions, experience indicates that it can impact the generator-
power system transient performance under certain conditions.
Dynamic instability is related to the high loading of modern electrical power
systems, to the design of lower-cost synchronous generators and especially
to the use of high-gain and quick-acting excitation systems. To improve the
damping of electromechanical oscillations when necessary, such systems
should be designed to allow proper processing of stabilization signals,
usually derived from the machine rotor-speed or electrical power. [1]
J.H.Chow and J.J.Sanchez-Gasca [2] examine the four pole-placement for
the design of the power system stabilizer with the emphasis on the frequency
characteristics of these controllers.
M. A. Abido and Y. L. Abdel-Magid [3] proposed a robust design of multi-
machine Power System Stabilizers (PSS) using Tabu Search (TS)
optimization technique. The eigenvalues analysis and the nonlinear
simulation results show the effectiveness of the proposed PSS to damp out
the local as well as the interarea modes and enhance greatly the system
stability over a wide range of loading conditions and system configurations.
K. Hongesombut, and Y. Mitani [4] focuses on the use of advanced
techniques in genetic algorithm for solving power system stabilization
control problems. These advanced techniques provide enhanced versatility
for solving problems in power system stabilization control.
K.T.Law et.al[5] exploited the internal model (IMC) technique to analyze
the inherent conflict between voltage regulation and damping improvement
for synchronous generators.

102
Al-Mansour Journal Issue (18) 2012 (18) ‫ﻣﺠﻠﺔ اﻟﻤﻨﺼﻮر اﻟﻌﺪد‬

Payman Shamsollahi [6] described an adaptive power system stabilizer using

on-line trained neural networks which is tested on a single-machine infinite-
bus power system model for a variety of disturbances.
C.X.Mao [7] described an adaptive optimal control algorithm for two real
time control applications for synchronous generator and power system
stabilizer.

2. PLANT MODEL
To investigate the effectiveness of using the power system stabilizer
(PSS) with the automatic voltage regulator (AVR), the plant model
which is the exciter-generator is shown in Figure 1 below. A
mathematical model for the exciter-generator plant was derived
depending on the Swing Equation viewpoint. The equation of central
importance in power system stability analysis are the rotational inertia
equation describing the effect of unbalance between the electromagnetic
torque and the mechanical torque of the individual machines.
The normalized swing equation then can be expressed as two first order
differential equations, then
d ( r ) 1
 (Tm  Te  K D  r )
dt 2H
(1)
d
 o  r
dt
(2)

The plant model will be introduced in the form of state space vector
differential equation x  Ax  Bu . [8, 9, 10, 11]

103
 Asst. Lecturer Fadhil M. Kalaf

Vref
K4
RegulatorExciter  Pm
+ +
1 K 3 E ' +  Pe - 1  r  o 
Σ KA K2 Σ
K E  sTE E Σ 1  sT3 Σ 2 Hs  K s
- fd + D

K1
Et K6

+
+
Σ K5

Figure 1 Exciter generator

The Vref and  P m will be taken as inputs to the plant then,

 KD  K1  K2 
 0  0 
  r   2 H 2H 2H   r   1
  o 0 0 0   0  P 
d         2 H
  K3K 4 1 K 3   0 m

dt  E '   0   E '   0  V 
   T3 T3 T3     0 KA   ref 

E fd   0  KAK 5  KAK 6


 K E  fd   0
E
 TE 
 TE TE TE 
(3)

Then from equation (3)

104
Al-Mansour Journal Issue (18) 20012 (18) ‫ﻣﺠﻠﺔ اﻟﻤﻨﺼﻮر اﻟﻌﺪد‬

 KD  K1  K2 
 2H 0 
2H 2H
  0 0 0 
 o 
A   0  K3 K 4 1 K3  and
 T3 T3 T3 
  KAK 5  KAK 6  KE 
 0 
 TE TE TE 
 1 0 
 2H 0 
B   0 0 
 0 KA 
 0 T 
 E 

3. PERFORMANCE EVALUATION of DESIGNED

CONVENTIONAL AVR

For the conventional AVR the output in the state-space can be

taken arbitrary, for the model of equation (3), the terminal voltage
Et is taken to be an output and can be given as:

Et  K 5   K 6 E '
(4)
or in a matrix form
  r 
  
Et  0 K 5 K6 0  
 E ' 
 
E fd 
(5)

105
 Asst. Lecturer Fadhil M. Kalaf

The synchronizing torque and the damping torque coefficients

can be calculated from the equations

 n  K S o rad/s
2H
(6)
And the damping ratio is
1 KD

2 2 H n
1 KD

2 K S 2 Ho
(7)
By changing the value of KA from 1 to 90 (the gain 90 is the
maximum that after this value the instability will be reached). Then for these
changes evaluating the values of the KS, KD (as given in equations 5 and 6),
and the time domain specifications is evaluated are shown in Table 1.
Figures 2 and 3 displays the plots of the output terminal voltage Et versus t
for number of KA values

Table 1 Conventional AVR

Settling Peak
KA ωn ωd ζ KS KD
time Amplitude
1 0.37 0.36 0.265 0.7668 1.395 37.2 1.43
20 1.71 1.71 0.0542 0.8185 1.2975 42.5 1.86
40 2.46 2.46 0.0336 0.8766 1.1572 47.5 2.01
50 2.77 2.77 0.0273 0.9067 1.0587 51.1 2.11
60 3.08 3.07 0.0219 0.9404 0.9443 57.3 2.19
80 3.67 3.67 0.0107 1.0143 0.5498 96.8 2.26
90 3.99 3.99 0.0018 1.0598 0.1033 504 2.31

(Note: if Vref is taken to be zero p.u, then the output will be the rate of
change of the terminal voltage Et , else if Vref is taken to be 1p.u then the
output will be the terminal voltage Et , which is the case in this paper).
106
Al-Mansour Journal Issue (18) 2012 (18) ‫ﻣﺠﻠﺔ اﻟﻤﻨﺼﻮر اﻟﻌﺪد‬

AVR Characteristic Peak Amplitude

when KA=1
1.5

Settling Time
Terminal Voltage
1

0.5

0
0 10 20 30 40 50 60

Time (sec)

AVR Characteristic when KA=20

2

1.8

1.6

1.4
Terminal Voltage

1.2

0.8

0.6

0.4

0.2

0
0 10 20 30 40 50 60

Time (sec)

Figure 2 Terminal voltage of the generator with KA = 1 and KA = 20

107
Asst. Lecturer Fadhil M. Kalaf

AVR Characteristic when KA=40

Terminal Voltage
1.5

0.5

0
0 10 20 30 40 50 60

Time (sec)

AVR Characteristic when KA=60

2
Terminal Voltage

1.5

0.5

0
0 10 20 30 40 50 60

Time (sec)

Figure 3 Terminal voltage of the generator with KA = 40 and KA =

60
108
 Al-Mansour Journal Issue (18) 2012 (18) ‫ﻣﺠﻠﺔ اﻟﻤﻨﺼﻮر اﻟﻌﺪد‬

4. PERFORMANCE EVALUATION of THE DESIGNED

CONVENTIONAL AVR AND PSS

The basic function of the power system stabilizer (PSS) is to

add damping to the generator rotor oscillations by controlling its
excitation using auxiliary stabilizing signal(s). To provide damping,
the stabilizer must produce a component of electrical torque in phase
with the rotor speed deviations.
The basis for a PSS may be illustrated with the aid of the block
diagram shown in Figure 4. This is an extension of the block diagram
of Figure 1 and includes the effect of PSS.
Since the purpose of a PSS is to introduce a damping torque
component, a logical signal to use for controlling generator excitation
is the speed deviation.

PSS

K4
v
T m
+ +
1 K 3 E ' T e + 1  o 
Vref
+
Σ KA +Σ K2 + Σ Σ
- E  sTEE  sT3 - Hs  K D r s
fd +

E
K1
K6
+
Σ K5
+
Figure 4 Block diagram representation with
109
 Asst. Lecturer Fadhil M. Kalaf

The PSS transfer function, should have appropriate phase lead

compensation circuits to compensate for the phase lag between the
exciter input and the electrical torque.
If the phase-lead network provides more compensation than
the phase lag between Te and vs , the PSS introduces, in addition to
a damping component of a torque a negative synchronizing torque
component. Conversely, under compensation a positive synchronizing
torque is introduced.
The PSS representation in Figure 5 consists of three blocks: a
phase compensation block, a signal washout block, and a gain block.

Phase
Washout Compensation
Gain
sTw v2 1  sT1
 r K STAB vs
1  sTw 1  sT2

System state matrix including PSS is shown below

Figure 5 Block diagram showing the PSS

110
 Al-Mansour Journal Issue (18) 20012 (18) ‫ﻣﺠﻠﺔ اﻟﻤﻨﺼﻮر اﻟﻌﺪد‬

  KD  K1  K2 
 0 0 0
2H 2H 2H
  0 0 0 0 0   
 r   o  r
     K K  1 K3
   0 3 4
0 0   
T3 T3 T3 
d  E'    KAK5  KAK6  KE  E' 
  0 E 
dtEfd   0 0
 TE TE TE  fd 
 v2    K K  K K  K  v2 
2KSTAB 1
   D STAB 1 STAB
0 0  
 vs   2H 2H 2H Tw  vs 
T1KDKSTAB T1K1KSTAB T1K2KSTAB 1 T1 1
 2HT 0 
 2 2HT2 2HT2 T2 T2Tw T2 
 0 
 0 
 
 0 
 
  KA  Vref
T 
 E
 0 
 0 
 
(8)

The terminal voltage Et is taken to be an output and can be given as

in equation (4)
or in state-space representation
  r 
  
 
 E ' 
Et  0 K 5 K 6 0 0 0  
E fd 
 v 2 
 
 v s 
(9)
By changing the value of KA from 1 to 90. Then for these
changes evaluating the values of the KS, KD and the time domain
111
Asst. Lecturer Fadhil M. Kalaf

specifications, as given in Table 2. Figures 6 and 7 displays the plots

of the output terminal voltage Et versus t for number of KA values in
the presence of PSS. From these figures, it can be shown how the
damping torque coefficient (KD) increases and with respect to the
damping torque coefficient evaluated in Table 1.

Table 2 Conventional AVR and PSS

Settling Peak
KA ωn ωd ζ KS KD
time Amplitude
1 0.37 0.36 0.2650 0.7668 1.395 37.2 1.43
20 1.71 1.71 0.0563 0.8185 1.3478 40.7 1.86
40 2.46 2.45 0.0371 0.8766 1.2777 43.6 2
50 2.77 2.77 0.0316 0.9067 1.2254 44.3 2.1
60 3.08 3.07 0.0271 0.9404 1.1686 46.1 2.18
80 3.67 3.67 0.0186 1.0143 0.9557 55.7 2.29
90 4.00 4.00 0.0127 1.0613 0.7112 73.8 2.3

112
Al-Mansour Journal Issue (18) 20012 (18) ‫ﻣﺠﻠﺔ اﻟﻤﻨﺼﻮر اﻟﻌﺪد‬

AVR+PSS Characteristic when KA=1

1.5

Terminal Voltage 1

0.5

0
0 10 20 30 40 50 60

Time (sec)

AVR+PSS Characteristic when KA=20

2

1.8

1.6

1.4
Terminal Voltage

1.2

0.8

0.6

0.4

0.2

0
0 10 20 30 40 50 60

Time (sec)

Figure 6 Terminal voltage of the generator with KA = 1 and KA = 20

113
Asst. Lecturer Fadhil M. Kalaf

AVR+PSS Characteristic when KA=40

Terminal Voltage
1.5

0.5

0
0 10 20 30 40 50 60

Time (sec)

AVR+PSS Characteristic when KA=60

2
Terminal Voltage

1.5

0.5

0
0 10 20 30 40 50 60

Time (sec)

Figure 7 Terminal voltage of the generator with KA = 40 and KA = 60

114
Al-Mansour Journal Issue (18) 2012 (18) ‫ﻣﺠﻠﺔ اﻟﻤﻨﺼﻮر اﻟﻌﺪد‬

5. CONCLUSIONS
In this paper at first the conventional AVR was adopted to the
synchronous generator whose data are given in Appendix A. The
synchronizing torque coefficient and the damping torque coefficient
are given in Table 1 and the time domain responses are displayed in
Figures 2 and 3. Then the conventional AVR with the power system
stabilizer (PSS) was adopted to the same synchronous generator. The
synchronizing torque coefficient and the damping torque coefficient
are given in Table 2 and the time domain responses are displayed in
Figures 6 and 7.
From the evaluated Tables and the time domain plots it is
shown that the damping is low, and if the gain of the regulator is
increased then the damping will be less. In order to solve this problem
and to increase damping, the Power System Stabilizer (PSS) was
used. Then KD was increased after using the PSS from 0.1033 to
0.7112 for a gain KA = 90, also the settling time was enhanced and
decreased from 504 sec to 73.8 sec and the same thing is true for the
other gains. It is clear that the synchronizing torque does not so
effected by the PSS as shown in Tables 1 and 2.

115
Asst. Lecturer Fadhil M. Kalaf

REFERENCES
1. MAURY, F.: Summary of the answers to the questionnaire on the
dynamic performances of future power system, CIGRE, Electra, pp. 87-100,
1975

2. J.H. Chow and J.J. Sanchez-Gasca: Pole-Placement Designs Of

Power System Stabilizers, IEEE Transactions on Power Systems, Vol. 4, No.
1, February 1989

3. M. A. Abido and Y. L. Abdel-Magid: Robust design of electrical

power-based stabilizers using Tabu search, Power Engineering Society
Summer Meeting, Vol. 3, pp. 1573-1578, 2001

4. K. Hongesombut, and Y. Mitani: Implementation of Advanced

Genetic Algorithm to Modern Power System Stabilization Control, Power
Systems Conference and Exposition, pp. 1050-1055 vol.2, 2004

5. K.T. Law et. al.: Robust Co-ordinated AVR-PSS Design, IEEE

Transactions on Power Systems, Vol. 9, No. 3, August 1994

6. Payman S hamsollahi: Neural Adaptive Power System Stabilizer,

Thesis at University of CALGARY, CALGARY, ALBERTA, July, 1997

7. C. X. Mao et. al.: Studies of Real-Time Adaptive Optimal Excitation

Controller and Adaptive Optimal Power System Stabilizer, IEEE
Transactions on Energy Conversion, Vol. 7, No. 3, September 1992.

8. P. Kundur, Power System Stability and Control, McGraw-Hill,

1994.

9. P. M. Anderson and A. A. Fouad, Power System Control and

Stability; A John Wiley & Sons, Inc., Publications 2003.

10. Haddi Saadat, Power System Analysis, McGraw-Hill Inc., 1999.

11. Computer Representation of Excitation Systems; IEEE Committee

Report; IEEE Trans. on PAS-87, No.6, June 1968.

116
Al-Mansour Journal Issue (18) 2012 (18) ‫ﻣﺠﻠﺔ اﻟﻤﻨﺼﻮر اﻟﻌﺪد‬

List of symbols

o  Rated speed in elect.rad/s

 rotor Speed
 r  Speed deviation in pu
  Rotor angle deviation in elect.rad
KA = Regulator gain
TE = Time constant of the exciter
 Rotor angle with respect to infinite bus
H Inertia constant in MW.s/MVA
Tm = Mechanical torque in N.M
Te = Air-gap torque or electromagnetic torque
KD = Damping torque coefficient
K1 to K6 = Constants

APPENDIX (A)
The power system data are as follows:
K A = (1 to 90), K D = 0, K 1 = 0.7643, K 2 = 0.8649, K 3 = 0.323, K 4 =
1.4187, K 5 = -0.1463, K 6 = 0.4168, K E = -0.02, H= 3.5 MW.s/MVA,
TE = 0.56, T3 = 2.465 s.

117
‫‪Asst. Lecturer Fadhil M. Kalaf‬‬

‫‪   ‬ا‪‬رة ‪  ‬ا‪ ‬و‪ ‬ا‪‬ا‪‬‬

‫‪ ‬ا‪‬ا‪‬‬
‫‪‬م‪ .‬م‪ .‬ﻓﺎﺿﻞ ﻣﺤﺴﻦ ﺧﻠﻒ‬
‫اﻟﺨﻼﺻﺔ‬
‫أن ﻣﻌﺎﻣﻼتِ ﻋﺰوم اﻟﻤﺰاﻣﻨﺔ واﻟﺘﺨﻤﯿﺪ ﻓﻲ ﻣﺪى ﺗﺮدداتِ ﺗﺬﺑﺬبِ ﺧﺎﺻﯿﺔ‬
‫اﻟﻜﮭﺮوﻣﯿﻜﺎﻧﯿﻚ ﯾُﻌﻄﯿﺎنِ ﻣﻘﯿﺎس ﻟﻤﺴﺎھﻤﺔِ اﻟﻤﻮﻟّﺪاتَ اﻟﻤﺘﺰاﻣﻨﺔ ﻓﻲ إﺳﺘﻘﺮارِﯾﺔ اﻟﻤﻨﻈﻮﻣﺔِ‬
‫اﻟﻜﮭﺮﺑﺎﺋﯿِﺔ ﻓﻲ ﺣﺎﻟﺔ اﻹﻧﺤﺮاﻓﺎتِ اﻟﺼﻐﯿﺮةِ‪ ،‬و َﻗﺪْ ُﯾﺆْﺧَﺬانِ ﻛﻔﮭﺎرس إﺳﺘﻘﺮارِ ﺣﺘﻰ ﻷﻧﻈﻤﺔِ‬
‫اﻟﻤﺎﻛﻨﺎتِ اﻟﻤﺘﻌﺪدةِ‪ ،‬ﺑﺎﻟﺮﻏﻢ ﻣﻦ ﺗﻌﻘﯿﺪ اﻟﻤﻮﺿﻮع ﻓﻲ ھﺬه اﻟﺤﺎﻟﺔِ‪ .‬ﯾُﻘﺪّمُ ھﺬا اﻟﺒﺤﺚ ﺗﺄﺛﯿﺮَ ﻣﺜﺒّﺖِ‬
‫اﻟﻨﻈﺎمِ اﻟﻜﮭﺮﺑﺎﺋﻲِ )‪ (PSS‬ﻋﻠﻰ ﻣﻌﺎﻣﻼتِ ﻋﺰوم اﻟﻤﺰاﻣﻨﺔ واﻟﺘﺨﻤﯿﺪ ﻟﻤﻮﻟّﺪ ﻣﺘﺰاﻣﻦ أوﺻﻞَ‬
‫إﻟﻰ ﻗﻀﯿﺐ ﻻﻧﮭﺎﺋﻲ ﺧﻼل ﻣﻌﺎوﻗﺔ ﻛﮭﺮﺑﺎﺋﯿﺔِ‪ .‬وﻣﻦ ﺿ ّﻤﻦُ اﻟﺘﺄﺛﯿﺮاتَ ﺗﺄﺛﯿﺮ ﻣﻨﻈّﻢِ اﻟﻔﻮﻟﻄﯿﺔ‬
‫اﻵﻟﻲِ )‪(AVR‬وﺗﺄﺛﯿﺮ إﺷﺎرةِ اﻟﻤﺜﺒّﺖِ اﻹﺿﺎﻓﯿﺔِ اﻟﻤﺄﺧﻮذة ِﻣﻦْ ﺳﺮﻋﺔِ اﻟﺪوّار ‪.‬‬

‫أن اﻟﻨَﺘﺎﺋِﺞ اﻟﻤﻜﺘﺴﺒﺔ ﻗُﻮرﻧﺖْ ﺑﺈﺳﺘﻌﻤﺎل ﻣﻨﻈّﻢِ اﻟﻔﻮﻟﻄﯿﺔِ اﻵﻟﻲِ اﻟﺘﻘﻠﯿﺪيِ )‪(AVR‬‬
‫وﺑﺄﺳﺘﺨﺪام اﻟُﻠﻐﺔَ اﻟﺒﺮﻣﺠﯿﺔ ‪.MATLAB 7‬‬

‫‪ ‬اﻟﻜﻠﯿﺔ اﻟﺘﻘﻨﯿﺔ‬
‫‪118‬‬