International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT) - 2016

Enhancement of transient stability on power system

with the use of power System stabilizers
1. Mihir R. Patel, 2. Nilay A. Patel,
Electrical Engineering Department, Electrical Engineering Department,
Charusat University, Charusat University,
Changa, India-388 421 Changa, India-388 421
1. [email protected] 2. [email protected]

Abstract-As we all know that most of the power systems are range 0.1to 0.7 Hz. For small scale stability the mathematical
interconnected through long distance tie lines, in order to model presented is a set of linear time invariant differential
transfer a large amount of electric power. There are some equations [1]. The concepts of synchronous machine stability
constraints which affect to build new transmission lines, which as affected by excitation control and the phenomenon of
are: limited availability of resources, strict environment and
stability of synchronous machines under small perturbations
shortage of funds. Because of these constraints it necessitates the
operation of transmission lines near or for short term on their in the case of single machine connected to an infinite bus
thermal loading limits. Day by day the demand in electric power through external reactance has been presented in [2]. The
rapidly increases and is expected to continue growing; while analysis also develops insights into effects of thyristor-type
growth of generation is restricted too, due to the same excitation systems and establishes understanding of the
constraints mentioned above, this necessitates the generators to stabilizing requirements for such systems [2]. These
operate near their maximum stability limits. As a result of all stabilizing requirements include the voltage regulator, gain
these factors, power systems oscillations occurs and if not parameters and the transfer function characteristics for a
damped out these oscillations, the magnitude of these oscillations machine speed derived signal superposed on the voltage
will go on increasing until loss of the system synchronism. So,
regulator reference for providing damping machine
for the improvement of power transfer capability within the safe
stability limit it is necessary to enhance the transient stability of oscillations [3].Kundur et al. presented a detailed analytical
the power system. The objective of this paper is to investigate work to determine the parameters of phase-lead PSSs so as to
the enhancement of transient stability on power system by enhance the steady state as well as transient stability of both
different types of power system stabilizers, which are: (I) Speed local and inter-area modes. These parameters included the
Input (II) Power Input (III) Frequency Input (IV) Del-P-Omega signal washout, stabilizer gain, and the stabilizer output
(V) Combination of frequency-electrical power input PSS and limits. They concluded that by proper tuning, the fixed-
(VI) Speed-electrical power input PSS. The implementation and parameter PSS can satisfy the requirements for a wide range
simulation carried out on the Single Machine Infinite Bus of system conditions and hence the need of adaptive PSS is of
system.
little incentive [4].This paper discusses the performance of
Index Terms-Transient Stability, Power System Stabilizer(PSS), various types of power system stabilizers for SMIB system
Single Machine Infinite Bus (SMIB). and shows which one is better among all. All the test system
and MATLAB software used in the paper will be freely
I. INTRODUCTION available by email to anyone who request.
In the power system disturbances occurs because of
variations in load, include electro mechanical oscillations of II. POWER SYSTEM SATBILIZER
electrical generators. These oscillations are also termed as Power system stabilizer (PSS) is a cost effective way of
power swings and for maintaining system stability, these improving the damping of electromechanical oscillations of
oscillations must be effectively damped out. rotors and in turn it improves the power transfer capability of
Electromechanical oscillations can be classified in two main transmission lines. The function of a power system stabilizer
categories (i) Local Plant Mode Oscillations and (ii) Inter- (PSS) is to add damping to the generator rotor oscillations.
area Oscillations. Local Plant Mode Oscillations are This is achieved by modulating the generator excitation so as
associated with units at a generating station swinging with to develop a component of electrical torque in phase with the
respect to the rest of the power system. The frequencies of rotor speed deviations. Such a way of producing damping
these oscillations are typically in the range 0.8Hz to 2.0 Hz. torque is the most cost-effective method of enhancing the
Inter-area Oscillations are associated with the swinging of small signal stability of power systems, in comparison to
many machines in one part of the system against machines in FACTS-based controllers [5].The necessary power
other parts. The frequencies of these oscillations are in the application is brought about in the normal process of torque

978-1-4673-9939-5/16/$31.00 ©2016 IEEE

Authorized licensed use limited to: New Mexico State University. Downloaded on November 01,2024 at 16:39:24 UTC from IEEE Xplore. Restrictions apply.
1914
development mechanism in the generator. Generally, PSS is If the degree of phase compensation required is small, a
installed to improve damping of local modes which is single first-order phase-compensation block may be used.
destabilized by the use of a high gain fast acting exciter. 3. Torsional Filter:
However, by judiciously placing power system stabilizers in a The torsional filter in the PSS is essentially a band reject or a
system, and with appropriate tuning of PSS parameters, it is low pass filter to attenuate the first torsional mode frequency.
possible to even improve the damping of inter area modes. The transfer function of the filter can be expressed as [6].
While designing a PSS to produce damping torque in a
desired frequency range, care must be taken to see that the ω a
PSS does not destabilize the other oscillatory modes, for FILT(s) = = (2)
s + 2ζω s + ω s +a s+a
example, torsional modes [6]. In the following, structure of a
PSS is briefly discussed. For a typical structure of a PSS see Torsional filter is necessitated by the adverse interaction of a
Fig. 1. It consists of a washout circuit, compensator, torsional slip-signal-based PSS with the torsional oscillations. This can
filter, gain and a limiter. The functions of each of the be lead to shaft damage, particularly at light generator loads
components of PSS with guidelines for the selection of when the inherent mechanical damping is small. Stabilizer
parameters are given below [6]. output can go into saturation (due to torsional frequency
components) making it ineffective. The criteria for designing
the torsional filter are:
3.1 The maximum possible change in damping of any
torsional mode is less than some fraction of the
inherent torsional damping.
3.2 The phase lag of the filter in the frequency range of
Fig. 1Basic structure of single input PSS the filter 1 to 3 Hz is minimized.
(Speed/slip input PSS)
4. Stabilizer Gain:
The amount of damping associated with the rotor oscillations
1. Washout Circuit:
depends on the stabilizer gain KS. The damping increases
Washout circuit is essentially a high-pass filter which
with an increase in stabilizer gain up to a certain value
removes dc offsets in the input signal and it eliminates the
beyond which further increase in gain results in a decrease in
steady state bias in the output of PSS which will modify the
the damping. To set the gain of the PSS, the following
field voltage. From the viewpoint of the washout function, the
criterions are generally employed:
value of Tw is not critical and may be anywhere in the range
4.1 Based on the gain for instability: The optimal PSS
of 1 to 20 seconds. For local mode oscillations in the range of
gain is chosen for the particular tuning condition as
0.8 to 2.0 Hz, a washout time constant of 1.5s is satisfactory.
the gain that results in the maximum damping of the
From the viewpoint of low-frequency inter-area oscillations,
critical (least damped) mode. The optimal gain (KS)
a washout time constant of 10 seconds or higher is desirable,
is related to the value of the gainK ∗ at results in
since lower-time constants result in significant phase lead at
instability. For example, for speed input
low frequencies. Unless this is compensated for elsewhere, it ∗ ∗
will reduce the synchronizing torque component at inter-area stabilizers,Ks = may be used [9]. In [10],Ks =
frequencies. has been used. These studies are carried out using
2. Lead-Lag Compensator: root locus method.
To damp rotor oscillations, a PSS must produce a component 4.2 Damping factor of the critical mode: Here, the gain
of electrical torque in phase with rotor speed deviation. This is selected such that damping factor for the mode is
requires a phase-lead circuits to be used to compensate for the above some typical value say 0.05 [8].
lag between the PSS output point (at the exciter) and the 4.3 High frequency gain: The high frequency gain of
resulting electrical torque developed. The amount of phase PSS is given byKs . This should not be too high
lag to be compensated depends on the generator parameters,
the type of exciters used and the system conditions. Though as it would lead to noise amplification decreasing
the degree of phase compensation should be designed so that the effectiveness of a PSS.
the PSS contributes to damping over a wide range of 5. Stabilizer Limits:
frequencies covering both inter-area and local modes of In order to restrict the level of generator terminal voltage
oscillation, a phase characteristic acceptable for different fluctuations during transient conditions and to prevent the
system conditions is selected. Generally, slight under PSS acting to counter the action of AVR, limits are imposed
compensation is preferable to overcompensation so that the on the PSS output.
PSS does not contribute to the negative synchronizing torque The positive output limits of the stabilizer are set at a
component. General transfer function of G (s) is given by relatively large value in the range of 0.1 to 0.2 pu. This
allows a high level of contribution from the PSS during large
(1 + sT ) (1 + sT ) swings. With such a high value of stabilizer output limit, it is
G (s) = (1) essential to have a means of limiting the generator terminal
(1 + sT ) (1 + sT )

Authorized licensed use limited to: New Mexico State University. Downloaded on November 01,2024 at 16:39:24 UTC from IEEE Xplore. Restrictions apply.
1915
voltage to its maximum allowable value, typically in the 1.12 x The phasor ∆ ( ) leads the
to 1.15 pu range [1]. ∆ ( )phasor by 90º (having accounted
The negative limit of PSS output is of importance during the negative sign in(V ). This implies that
the back swing of the rotor (after initial acceleration is over). using electrical power signal is equivalent
Negative side limit are raised to prevent the PSS from to using slip signal with 90º phase lead. In
reducing the generators terminal voltage excessively other words, to get the phase angle of the
following a fault. The AVR action is required to maintain the compensated GEPS, , it is required to
voltage (and thus prevent loss of synchronism) after the simply add 90º to the angle of ( )
angular separation has increased. Typically, -0.02 to -0.05 pu From the above observations, a power-input PSS is
is used for the negative limit. This allows sufficient control implemented along with a measurement delay
range while providing satisfactory transient response [4]. transfer function and a washout-circuit as shown in
III. TYPES OF PSS Fig. 1.
2. Dual input power system stabilizer: In this kind of
As per IEEE standards 421.5 - 1992 [7], the following are the
PSS design, a combination of signals such as speed
two main categories of PSS:
and electrical power output are used.
1. Single input power system stabilizer: It is known
2.1 Delta-P-Omega Signal PSS:
that in order to modify a mode of oscillation by
A dual input PSS where speed and electrical power
feedback, the chosen input must excite the mode and
deviations are used as input, is analyzed. This PSS is
it must be visible in the chosen output [8]. Thus, for
referred to as Delta-P-Omega PSS. The objective of
this kind of PSS design, commonly used input
this PSS is to derive an equivalent speed
signals are shaft speed, terminal bus frequency and
signal∆ω so that it does not contain torsional
electrical power output.
1.1 Slip/Speed Input Signal PSS: modes. The principle of this type of stabilizer is
The block schematic of a speed input type PSS is illustrated below [1, 10]:
shown in Fig.1.It consists of a washout circuit,
1 1
compensator, torsional filter, gain and a limiter. The ∆ = ∆ − ∆ (4)
functions of each of the block have been explained 2 2
same as above.
1.2 Frequency Input Signal PSS: Note that torsional components are inherently
Frequency input type PSS use terminal frequency as attenuated in ∆Te signal. Now the problem is to
stabilizer input signal. Even though a frequency measure the integral of ∆Tm free of torsional modes.
input type PSS is less sensitive to torsional In many applications, the ∆Tm component is
oscillations, its structure is assumed to be the same neglected. This is satisfactory, except when
as that of a speed input type PSS, as shown in Fig. 1. changing load on the unit and other system
1.3 Power-signal Input PSS: conditions when the mechanical power changes.
In a single-input power system stabilizer, electrical Under such conditions, a spurious stabilizer output is
power output of the machine is normally used as the produced if ∆Te alone is used as the stabilizing
input signal as it provides high degree of attenuation signal. This in turn results in transient oscillations in
to torsional modes unlike slip-signal [11]. An voltage and reactive power. A way to measure
electrical power-input based PSS can be realized by integral of ∆Tmis presented below: Rewriting
observing the following relationship between∆S equation 4, we have,
and ∆T given by
1 1
∆Tm dt = ∆Sm + ∆Te dt (5)
1 2H 2H
∆ ( )=− ∆ ( ) (3)
2
The delta-P-omega stabilizer makes use of the above
Note that the above relationship is obtained from relationship to simulate a signal proportional to the
swing equation by neglecting the deviation in the integral of mechanical power change by adding
mechanical power input to the machine and signals proportional to shaft-speed change and
mechanical damping, and for sinusoidal variation of integral of electrical power change. This signal will
quantities. From equation 3 it can be seen that contain torsional oscillations unless a filter is used.
x To get the same effect as a slip-input PSS, Because mechanical power changes are relatively
with the∆T input, the output of the PSS slow even for fast-valve movements, the derived
block(V )is fed to the exciterV summing integral of the mechanical power signal can be
junction with a negative sign (as against a conditioned with a simple low-pass filter to remove
positive sign that has been used with the torsional frequencies. The output of the filter
slip-input PSS). provides ( ∫ ∆Tm dt)signal which is free from
torsional oscillation. Using this signal, the slip signal

Authorized licensed use limited to: New Mexico State University. Downloaded on November 01,2024 at 16:39:24 UTC from IEEE Xplore. Restrictions apply.
1916
 is synthesized which is completely free of torsional gain experience with the analytical techniques, using simple
frequency components. This obtained by realizing low-order systems, we will be in a better position to deal with
equation 4, having simulated ( ∫ ∆Tm dt)signal. large complex system. For the analysis it’s necessary to
develop the Linearized Model of each part of the system.
This permits the selection of a higher stabilizer gain
Likewise: Generator, Transmission line, Power system
that results in better damping of system oscillations,
Stabilizer (PSS).For the analysis purpose the system is
without causing the destabilization of exciter/swing
reduced using Thevenin’s equivalent of the transmission
modes unlike that is observed in a slip-signal-based
network external to the machine and the adjacent
PSS with torsional filter, FILT(s) [1]. The block
transmission. Because of the relative size of the system to
schematic of a Delta-P Omega type PSS is shown in
which the machine is supplying power, dynamics associated
Fig. 2. In the figure, note that an integrator is
with the machine will cause virtually no change in the voltage
approximated by a first order transfer function by
and frequency of Thevenin’s voltage EB. Such a voltage and
suitably choosing the time constant T7. This is done
constant frequency is referred to as an infinite bus. The single
to avoid offset problem in a pure integrator circuit.
line diagram of SMIB power system is shown in Fig. 5. The
system details are adopted from Example 6.6 in [6].The
simulation is carried out in MATLAB and the system details
are given in appendix.

Fig. 2 Delta-P-Omega Signal PSS

3. Frequency-electrical power input PSS:

Fig. 5 Line diagram of SMIB system
Which is basically frequency input PSS. Here
frequency and electrical power deviations are used V. SIMULATION RESULT AND
as input signal to the frequency input type PSS with DISCUSSIONS
two individual different gains as shown in fig. 3. 1. Without PSS action: Without PSS the oscillation is
goes on increasing and system will out of step.
Without PSS
2

1.5
Delta(pu)

Fig. 3 Frequency-electrical power input PSS 0.5

0 5 10 15
Time(sec)

4. Speed-electrical power input PSS:

Which is basically speed input type PSS. Here speed 2. With speed input PSS: The peak value of first swing
and electrical power deviations are used as input is 1.754pu and damp out after 9.935 sec.
signal to the speed input type PSS with two
individual different gains as shown in fig. 4.
Speed input PSS
1.8

1.7

1.6

1.5

1.4
Delta(pu)

1.3

1.2

1.1

0.9
0 5 10 15
Time(sec)

3. With Power input PSS: The peak value of first swing

Fig. 4 Speed-electrical power input PSS is 1.776pu and damp out time is more than 15 sec.

IV. SMIB TEST SYSTEM

The Single Machine Infinite Bus System (SMIB) is used for
analysis. This SMIB system analysis gives the simple
configurations to extremely useful in understanding basic
effects and concepts. After that we can able to develop an
appreciation for the physical aspects of the phenomena and

Authorized licensed use limited to: New Mexico State University. Downloaded on November 01,2024 at 16:39:24 UTC from IEEE Xplore. Restrictions apply.
1917
 Power input PSS
1.8

1.7
Frequency Input 1.754 More than 15
1.6

1.5 Del-P-Omega 1.754 10.954

Frequency and
1.4

1.752 5.195
Delta(pu)

1.3

1.2
electrical power
Speed and electrical
1.1

1
1.751 3.050
0.9
power
0.8
0 5 10 15
Time(sec)

4. With Frequency input PSS: The peak value of first

swing is 1.754pu and damp out time is more than 15 VI. CONCLUSION
sec. This paper shows the work of different types of PSS on SMIB
system with the applied two new types of PSS. The
comparison graph clearly signifies that the combination of
Frequency input PSS
2

speed and electrical power input PSS which yields the better
1.8

1.6

1.4 results among all types of PSS. Numerical results shown in

Delta(pu)

1.2
table I from this we can observe that, speed and electrical
1
power input PSS which is same as Del-P-Omega PSS (respect
to input signals) from both this PSS the proposed one is give
0.8

very better and superior results. Its rotor oscillation damp out
0 5 10 15
Time(sec)

5. With Del-P-Omega PSS: The peak value of first time is 3.050sec which is best among all. One more
swing is 1.754pu and damp out time is 10.954 sec. combination of PSS is used, Frequency and electrical power
Del-p-Omega PSS
input PSS, this PSS is basically frequency input PSS but by
combination with the signal of electrical power it gives better
1.8

1.7

1.6 performance compare to normal frequency input PSS. Here

1.5

1.4
the selection of gain value for both the new approach PSS are
based on trial and error based so, if we consider any
Delta(pu)

1.3

1.2

1.1
optimization methods for the selection of gain value it may be
1 give better results.
0.9
0 5 10 15

Time(sec)

6. Frequency and electrical power: The peak value of ACKNOWLEDGMENT

first swing is 1.752pu and damp out time is 5.195 I am heartily very thankful to Dr. K.N. Shubhanga (Associate
sec. Professor, Department of Electrical Engineering, NITK,
Surathkal) for guideline me and providing their helping arm
1.8
Proposed PSS__Frequency
to me during the work.
1.7

1.6

1.5
VIII. REFERENCES
1.4

[1].P.Kundur, “Power system stability and control” New York:

Delta(pu)

1.3

1.2

1.1
TataMcGraw-Hill, 1994.
1
[2]F.P.demello, C.Concordia, “Concepts Of Synchronous MachineStability
As Affected By Excitation Control,” IEEE Trans.On Powersystem and
0.9

0.8

apparatus, Vol-PAS-88, No.4, April 1969, pp. 316-329.

0 5 10 15
Time(sec)

[3].IEEE Committee Report: “Computer representation of

excitationsystems”, IEEE Trans., 1968, PAS-87, pp 1460-1464.
7. Speed and electrical power: The peak value of first
[4]. Kundur P., Klien, M., Rogers, G.J., and Zywno, M.S.: “Application of
swing is 1.751pu and damp out time is 3.050 sec Power System Stabilizer for the enhancement of overall systemstability”,
IEEE Trans., 1989, PWRS-4, pp. 614-626.
1.8
Proposed PSS
[5]Hingorani N.G and Gyugyi.L, “Understanding FACTS,” IEEE press,
1.7
NewYork, 2000.
[6]. K.R. Padiyar, Power System Dynamics - Stability and Control, BS
1.6

1.5

1.4 Publications, Hyderabad, India, 2002.

Delta(pu)

1.3
[7] IEEE Recommended Practice for Excitation Systems Model for Power
System Stability Studies, IEEE Standard 421.5-1992.
1.2

1.1

1 [8] Graham Rogers, Power System Oscillations, Kluwer Academic

0.9
0 5
Time(sec)
10 15 Publishers, London, 2000.
[9]E.V. Larsen and D.A. Swann,“Applying Powers System Stabilizers, Part
I; GeneralConcepts, Part II; Performance Objectives and Tuning Concepts,
TABLE I. SUMMARY OF PERFORMANCE DATA FOR ALL PSS
Part III; Practicalconsiderations ”, IEEE Trans on power Apparatus and
Systems Vol PAS-100, No.6, pp. 3017-3046, June, 1981.
Peak value of Delta Damp out Time [10] P. Kundur, D.C. Lee and H.M. Zein El-Din,“Power System Stabilizers
Types Of PSS
(pu) (sec.) for ThermalUnits: Analytical Techniques and On-site Validation,” IEEE
Speed Input 1.754 9.935 Trans.on Power Apparatusand Systems, Vol. PAS-100, pp. 81-95, January,
1981.
Power Input 1.776 More than 15 [11] P.W. Sauer and M.A. Pai, Power System Dynamics and Stability,
Prentice Hall,Upper Saddle River, New Jersy, 1998.

Authorized licensed use limited to: New Mexico State University. Downloaded on November 01,2024 at 16:39:24 UTC from IEEE Xplore. Restrictions apply.
1918
APPENDIX
The system details are as follows:

Generator: xd = 1.7572, x’d = 0.4245, T’do = 6.66 s, xq =

1.5845, x’q = 1.0400, T’qo = 0.44 s, H = 3.542, fo = 50 Hz, D
=0
Exciter: Single-time constant static exciter, KA = 400, TA
=0.025 s, Efdlimits = ± 6
Transformer: xt = 0.1364
Lines: R = 0.08593, xL = 0.81250 and Bc = 0.1184, xth =
0.13636
Loads: Constant impedance type.
A three phase fault is applied at bus 2 at time t = 1 s and is
cleared by tripping the line 1. Fault clearing time is 0.08 sec

Authorized licensed use limited to: New Mexico State University. Downloaded on November 01,2024 at 16:39:24 UTC from IEEE Xplore. Restrictions apply.
1919