Particle Swarm Optimization Based Power System

Stabilizer For SMIB System

Pothula Jagadeesh Dr.M.Sai Veerraju

Assistant Professor, Department of EEE Professor, Department of EEE
SRKR Engineering College SRKR Engineering College
Bhimavaram, India. Bhimavaram, India.
[email protected] [email protected]

Abstract—Power system stabilizers (PSS) are used to generate disturbances such as sustained oscillations in the speed or
supplementary damping control signals for the excitation system periodic variations in the torque that is applied to the
in order to damp the low frequency oscillations (LFO) of the generator. These disturbances may result in voltage or
electric power system. The PSS is usually designed based on frequency fluctuation that may affect the other parts of the
classical control approaches but this conventional PSS (CPSS)
interconnected power system[8]. External factors, such as
has some problems. To overcome the drawbacks of CPSS,
numerous techniques have been proposed in literatures. In this lightning, can also cause disturbances to the power system. All
paper a PID type PSS is considered for damping electric power these disturbances are termed as faults. When a fault occurs, it
system oscillations. The parameters of this PID type PSS are causes the generators to lose synchronism. With these factors
tuned based on particle swarm optimization method. The in mind, the basic condition for a power system with stability
proposed PSS (PSO-PSS) is evaluated against the conventional is synchronism[8]. Besides this condition, there are other
power system stabilizer (CPSS) at a single machine infinite bus important conditions such as steady-state stability, transient
power system considering system parametric uncertainties. The stability, harmonics and disturbance, collapse of voltage and
simulation results clearly indicate the effectiveness and validity of the loss of reactive power.
the proposed method.
The stability of a system is defined as the tendency and
Keywords—power system stabilizer, particle swarm ability of the power system to develop restoring forces equal
optimization,damping, to or greater than the disturbing forces to maintain the state of
equilibrium[10]. There are many major blackouts caused by
I. INTRODUCTION instability of a power system which illustrates the importance
of this phenomenon. The stability has been acknowledged as
an important problem for secure system operation since the
The electrical energy is a primary prerequisite for
economic growth. The demand for electrical energy has greatly 1920‟s.
increased due to large-scale industrialization[1]. Modern power Damping of power system oscillation between
interconnected areas is very important for the system secure
system operates under much stressed conditions because of
growth in demand and deregulation of electric power operation. Power system stabilizer (PSS) is the most widely
system[4]. This leads to many problems associated with used device for resolving oscillatory stability problems, and to
operation and control of power systems. The economics of enhance the power system damping[2]. Traditionally, lead-lag
power generation has a major concern for the power utilities. structures have been used as power system stabilizers[5].
Therefore, the power utilities always need new technology to Many researches had been published explaining the ways of
tuning the parameters of the lead lag controller. The methods
solve its problems.
used for tuning range from pole placement, to the more recent
The complexity of power systems is continuously one using the heuristic optimization techniques such as
growing due to the increasing number of generation plants and Genetic Algorithms (GAs), Taboo Search Algorithm (TSA),
load demand[3]. Power systems are becoming heavily stressed Simulated Annealing (SA), Particle Swarm Optimization
due to the increased loading of the transmission lines and due (PSO), and Bacteria Foraging Algorithm (BFA).
to the difficulty of constructing new transmission systems as The PID controller is a well-established type of
well as the difficulty of building new generating plants near controller and has been in use for a long time. Tuning PID
the load centers. All of these problems lead to the voltage controllers are traditionally tuned using standard techniques
stability problem in the system. such as the root locus, and classical PID controllers which
An interconnected power system basically consists of tuned by “Ziegler-Nichols” methods. This paper produces a
several essential components[7]. They are namely the design method for the stability enhancement of a single
generating units, the transmission lines and the loads. During machine infinite bus power system using PID-PSS which its
the operation of the generators, there may be some parameters are tuned by Hybrid Particle Swarm- Bacteria

978-1-4673-6725-7/16/$31.00 ©2016 IEEE

Forging Optimization method[6]. The advantage of tuning the We note that the constants Ki (i = 1,..., 6) are uncertain and
parameters of the optimum controller is that the possibilities of depend upon the network parameters, the quiescent operating
including time domain specifications such as raise time, conditions and the infinite bus voltage.
maximum overshoot, damping ratio, and steady-state error.
III. PARTICAL SWARM OPTIMIZATION

II. SYNCHRONOUS MACHINE MODELLING Particle Swarm Optimization (PSO) algorithm was
The power system considered in this study is modeled as a originally introduced by Kennedy and Eberhart in 1995[12].
synchronous generator connected through a transmission line Many researches have been introduced to improve the
to infinite bus bar. A simplified model that describing the performance of the PSO with respect to the speed of
system dynamics used in this study is given by the following convergence and to make sure that the PSO will not get stuck
state space equations.[2] in a local minima[13]. The PSO has been developed through a
x˙ = Ax(t)+B1w(t)+B2u(t) (1) (1.a) simulation of simplified social models. The features of the
z(t)=C1x(t)+ D11w(t) + D12u(t) (2) method are as follows:
y(t)=C2x(t)+D21w(t) + D22u(t) (3) 1) The method is based on researches on the swarms such as
fish schooling and flocking.
Where u represents the PSS output added to the voltage set 2) It is based on a simple concept. Therefore, the computation
points ΔVref, ω is an external disturbance represented by the time is short and it requires few memories.
mechanical power ΔPm. The matrices A, B1, B2, the vector z, y According to the research results for bird flocking, birds
and the state vector x are defined by are finding food by flocking (not by each individual). It leaded
the assumption that information is owned jointly in
flocking[11]. According to observation of behavior of human
groups, behavior pattern on each individual is based on several
behavior patterns authorized by the groups such as customs
and the experiences by each individual (agent).
PSO is basically developed through simulation of bird
T
; T flocking in two-dimension space. The position of each
individual (agent) is represented by XY axis position and the
T,
y=∆Pe velocity is expressed by vx(the velocity of X axis) and vy(the
velocity of Y axis). Modification of the agent position is
T realized by the position and velocity information.
Where δ, ω, E'q, EFD, Pe and Vt are respectively the An optimization technique depends on the above concept.
torque angle, the angular velocity the internal machine Bird flocking optimizes a certain objective function[9]. Each
voltage, the excitation voltage, power output and generator agent knows its best value so far (pbest) and its XY position.
terminal voltage[14][15]. T'do is the open-circuit transient Moreover, each agent knows the best value so far in the group
time constant, Δ represents a small deviation around the (gbest) among pbests. Each agent tries to modify its position
operation point The operating conditions for the above using the following information:
systems are completely defined by the values of the real (P) 1 the current position (x,y),
and reactive (Q) powers at the generator terminals and the 2 the current velocities (vx, vy),
transmission line impedance Xe. A detailed block diagram of 3 the distance between the current position, and pbest and
the power system (open loop) is shown in Fig. 1. gbest.
This modification can be represented by the concept of
velocity. Velocity of each agent can be modified by the
following equation:

viK+1=wvik + C1rand1(pbesti-Sik)+C2rand2(gbest-sik) (4)

Using the above equation, a certain velocity, which

gradually gets close to pbest and gbest can be calculated. The
current position (searching point in the solution space) can be
modified by the following equation:
Sik+1=Sik + vik+1 (5)

Fig.1 block diagram of SMIB system

 Fig.2 Concept of modification of a searching point

A. PSO Algorithm

Step.1: Initialize an array of particles with random positions

and their associated velocities to satisfy the inequality
constraints.

Step.2: Check for the satisfaction of the quality constraints and

modify the solution if required.

Step.3 Evaluate the fitness function of each particle.

Step.4 Compare the current value of the fitness function with

the particles previous best value (pbest). If the current fitness
value is less, then assign the current coordinates (positions) to
pbest x.

Step.5 Determine the current global minimum fitness value

among the current positions.

Step.6 Compare the current global minimum with the previous

global minimum (gbest). If the current global minimum is
better than gbest, then assign the current global minimum to IV. PSO BASED PID CONTROLLER
gbest and assign the current coordinates (positions) to gbest x. In the proposed design approach, the PID control structure
shown in Fig.is used as the power system stabilizer as opposed
Step.7 Change the velocities according to equation. to the traditional lead-lag controller[16]. The PID parameters,
and are tuned using the PSO technique discussed in above.
Step.8 Move each particle to the new position according to
equation (5) and return to step 2.

Step.9 Repeat Step 2-8 until optimization is satisfied or the

maximum number of iterations is reached.

B) Flowchart for particle swarm optimization

Fig.3 PID based power system stabilizer
Where the speed deviation is the input to the controller, u is
the supplementary stabilizing signal, and the washout filter is
used to remove the controller effect at steady state
conditions[16]. The design starts with incorporating the PID
structure with the washout in the system matrix discussed in
equation (1) to form the augmented A matrix given below:
 Particle swarm 80 22.60 47.1979
optimization
Table.1 Parameters of PID controller

VII. RESULTS
The simulation results for the dynamic stability analysis
of considering SMIB system without using any power system
stabilizer that is without providing any damping to the system
(6) is shown in fig.4 and the response of conventional PID
controller with this SMIB system is shown in fig.5
V. OBJECTIVE FUNCTION
In this paper particle swarm optimization has been used for
the optimization of PID-PSS parameters. Just like any other
optimization problem, a cost or an objective function needs to
be formulated for the optimal PID-PSS design. The objective in
the optimal PID-PSS design is to maximize damping; in other
words minimize the overshoots and settling time in system
oscillations[6].
The Integral of Squared Error (ISE) is considered as the
cost function to be minimized. ISE accounts mainly for error at
the beginning of the response and to a lesser degree for the
steady state duration. The objective function is given by (7). Fig.4 Response without using PSS
2
dt (7)

Where ∆w is the speed deviation of the generator obtained

from time domain simulation. Therefore, the design problem
can be formulated as the following optimization problem.
Minimize J
Subject to
Zmin≤z≤zmax (8)
Where z is a vector, which consists of the parameters of the
PID-PSS. Where z is a vector, which consists of the parameters Fig.5. Response with conventional PID controller
of the PID-PSS. The proposed approach employs PSO to solve Here we have used PID controller based power system
this optimization problem and search for the optimal set of
stabilizer it’s parameters are tuned by using Particle Swarm
PID-PSS parameters.
Optimization algorithm, it is implemented in the above SMIB
system the response obtained is shown in fig.6 and the error
VI. IMPLEMENTATION OF PSO BASED PSS which is occur in the optimization is represented in a plot is
Here we used particle swarm optimization algorithm to shown in fig..7
tune the PID controller. It is much better and much accuracy
than conventionally obtained parameters of the PID controller
and also the Zigluer-Nickolas method of tuning of PID
controller parameters. By using Particle Swarm Optimization
algorithm we can easily know the boundaries of the control
parameters in a certain interval and also we can easily obtain
the exact values of the control parameters. We can observe the
result in Fig.4 the unstable response of the SMIB system.
That’s why we choose this PSO algorithm and got the control
parameters as shown in Table.1.
Fig.6 Response with PSO-PID based PSS
Kp Ki Kd The error signal value that is the fitness value if the each
particle can be obtained in this optimization that is the
Trial and error 1.05 5 2 magnitude of
Ziegler-Nicholos 2 3.22 2.8 Error = 4.1818e-0.06
It can be plotted as shown in fig.7,
 CONCLUSION
Here The dynamic stability analysis of considering
single machine connected to an infinite bus system is done in
three cases, they are without using any power system stabilizer
for given system, with using conventionally tuned PID
controller based Power system stabilizer and We proposed
Particle Swarm optimization based tuning of PID controller
based Power System Stabilizer for SMIB system.
Fig.7 Error plot This proposed power system stabilizer not only reduces
Comparison between simulation results: the settling time of the oscillations but also minimizing both
positive and negative overshoots very effectively can observe
The comparison between the responses of with using
in above results. Thus it can be concluded stability
PSO based PID controller power system stabilizer and without
enhancement is greatly achieved by this Particle Swarm
using any power system stabilizer is shown in fig.8
Optimization algorithm tuned PID controller based stabilizer
compared with conventional PID controller.

APPENDIX
System data:
The Parameters of the synchronous machine, excitation
system and conventional PSS are as follows.
[a] Synchronous machine constants:
xd= 2.64 pu, x’d= 0.28 pu
Fig.8 Combined response of with PSS and Without PSS xq= 1.32 pu, xq= 0.29 pu
The combined response representation of both RE= 0.004 pu, XE= 0.73 pu
conventional PID controller based PSS and particle swarm f= 60 Hz, H= 4.5 sec
optimization based PID controller PSS is shown in fig.9 [b] Excitation system constants:
KA= 100, TA= 0.05, TR= 0.015
EFMAX= 5.0, EFMIN= -5.0

PSO algorithm Parameters:

Population size =50
Initial weights Wmax=0.9
Final weights Wmin=0.4
Crossover rate Cr=0.65
Fig.9 Response with conventional PID and PSO-PID based Acceleration coefficient C1=C2=2
PSS
The combined response of PSO-PID based power system Maximum iteration number =50
stabilizer with different values of iterations can be shown in Minimum gain value=-10
fig.10
Maximum PID gain = 80
Maximum velocity of particles Vmax=5
Minimum velocity of particles Vmin= -5

REFERENCES
[1] Zwe-Lee Gaing, “A Particle Swarm Optimization Approach for
Optimum Design of PID Controller in AVR System”, IEEE
Transactions on Energy Conversion, Vol. 19, No.2, June 2004,
pp384- 391.
[2] F. P. Demello and C. Concordia , “Concepts of Synchronous
Machine Stability as Effected by Excitation Control”, IEEE
Fig.10 Response with different iteratios of PSO algorithm Transactions on Power Apparatus and Systems,Vol.PAS-88,
No.4,April 1969.
[3] T. K. Das, and G. K. Venayagamoorthy, “Optimal Design of
Power System Stabilizers Using a Small Population Based
PSO”IEEE Power Engineering Society General Meeting, 2006.
[4] E. V. Larsen, and D. A. Swann, “Applying Power System
Stabilizers, parts I, II & III,” IEEE Transactions on Power
Apparatus and Systems. Vol. PAS-100, No. 6 June 1981, pp. 3025-
3046.
[5] P. Kundur, Power System Stability and Control, McGraw-Hill
Press, 1994.
[6] Nelendran Pillay and Poobalan Govender, “A Particle Swarm
Optimization Approach for Model Independent Tuning of PID
Control Loops”, Conference Publications: AFRICON 2007,
Windhoek , 26-28 Sept. 2007, pp. 1 – 7
[7] Amit Shrivastava, Manisha Dubey, Yogendra Kumar
“optimization Techniques Based Power System Stabilizer's: An
Overview, International Journal on Emerging Technologies 4(1):
50-54(Dec-2013) ISSN No. (Print): 0975-8364ISSN No. (Online):
2249-3255.
[8] M. A. Abido, “An efficient heuristic optimization technique for
robust PSS design,” EPSR 58, 2001,pp 53-62.
[9] J.G. Ziegler and N.B. Nichols, “Optimum settings for automatic
controllers,” Trans.-ASME, Vol. 65, pp. 433-444, 1942.
[10] Freddy Prasetia Ridhuan , Mohd Fauzi Othman “Application Of
Artificial Immune System To Design Power Systems Stabilizer”
The 5th Student Conference on Research and Development
Malaysia,2007.
[11] Das, T.K.; Venayagamoorthy, G.K.; Aliyu, U.O. “Bio-Inspired
Algorithms for the Design of Multiple Optimal Power System
Stabilizers: SPPSO and BFA ” IEEE Transactions on Industry
Applications, Vol: 44 , Issue: 5 , 2008, pp:1445 – 1457.
[12] Sidhartha Panda, N. P. Padhy “Robust Power System Stabilizer
Design using Particle Swarm Optimization Technique” World
Academy of Science, Engineering and Technology 46, 2008.
[13] Z. Rafiee, S. Ganjefar, and A. F. Meyabadi“A New PSS Tuning
Technique Using ICA and PSO Methods with the Fourier
Transform” Electrical and Electronic Engineering; 1(1), 2011, pp:
17-23.
[14] H. Shayeghi, A. Ghasemi “Application Of PSO-TVAC To
Improve Low Frequency Oscillations” International Journal on
“Technical and Physical Problems of Engineering” (IJTPE) , Vol:
3, No: 4 Issue 9, 2011, pp 36-44.
[15] B.Selvabala, Dr.D.Devaraj “Co-Ordinated Tuning Of AVR-PSS
Using Differential Evolution Algorithm”IEEE conference, 2010.
[16] Åström K., and Hägglund T., “PID controllers: Theory, Design and
Tuning”, ISA, Research Triangle Park, NC, 1995.