International Conference on Computing, Communication and Automation (ICCCA2017)

Fuzzy Tuned PID Controller for Power System

Stability

Zubair Iqbal Shoeb Hussain Mohammad Abid Bazaz

Dept. of Electrical Engineering Dept. of Electrical Engineering Dept. of Electrical Engineering
NIT, Srinagar NIT, Srinagar NIT, Srinagar
J&K INDIA J&K INDIA J&K INDIA
[email protected] [email protected] [email protected]

Abstract— This paper introduces a hybrid fuzzy based PID As of late, with momentous increment of demand of power,
controller to improve the performance of power system development of interconnected power frameworks and
stabilization. The inadequately damped electromechanical especially the deregulation of industry issues identified with
motions in power frameworks has been an issue to power low frequency oscillations have been generally revealed. To
engineers for quite a while. This restrains the ability of transfer maintain such huge interconnected power system within safe
of power in transmission lines and further incites worry in the and stable operating modes various strategies have been
mechanical shaft of machine. Owing to large scale presented by researchers and it has become one of the
interconnections and structured complexity of power systems, important research themes from last decade. The most savvy
some disturbances arise which result in poorly damped low- damping controller, Power system stabilizer (PSS) has been
frequency oscillations that don’t damp out with time and at some
broadly used to stifle the low frequency oscillations and
occasions system is shaken out of equilibrium. The dynamics of
this framework is additionally influenced by these oscillations at
upgrade the framework dynamic stability. In this way,
low frequency. numerous analysts are contemplating and showing different
The oscillations can be reduced and in turn the stability can PSS sorts on the premise of power framework stability since
be increased by proper design of controllers. The oscillations can the late 1980s.
be diminished and thus the solidness can be expanded by Demello proposed a determination technique for PSS
appropriate outline of controllers. The essential element of the parameters in light of the expanding guideline of braking
proposed approach is plan of a fuzzy based PID controller
torque in low frequency region through the stage compensator
(FPID) that uses a rearranged configuration. FPID controller,
plan of PSS. Be that as it may, on account of higher request
which exploits the properties of fuzzy PI and PD controllers, is
proposed. This approach doesn't require exact learning of plant's PSS, it can't be connected [1]. Hanson proposed an enhanced
characteristics for execution of the controller. The performance PSS strategy which could manage the increase steady of
when compared with the conventional controllers, the results of generator control base (i.e., governor and voltage controller,
fuzzy controller are quite encouraging and satisfactory. and so on), however his technique was not needy upon a
precise approach strategy but rather upon a basic
Keywords— PSS; Fuzzy logic; PID; AVR; FPID experimentation strategy [2]. Furthermore, Gooi and Fleming
et al. introduced an enhancement approach of PSS pick up
I. INTRODUCTION parameters [3, 4]. Yu presented a choice technique for PSS
parameters with the expansion of supplementary damping
The interconnection of large electric power systems has twisting by stage compensator [5]. However, the precise
resulted in spontaneous system oscillations at very low demonstrating of power framework is not obtained due to its
frequencies in the order of several cycles per minute. With nonlinearity including sudden load change and parameter
oscillations initializing, they proceed for some time and after irritations, for example, generator and other framework
that vanish, or keep on growing, bringing on poorer framework coefficients. In this way, it is hard to plan a phenomenal
coordination. With more experience collected from controller, which can be worked in the huge working reach.
interconnected electric power framework operation, control
framework designers are presently persuaded that the low Hence, in this paper a power framework stabilizer sort is
frequency oscillations are because of the absence of damping displayed. That is, this controller is consolidated with a
of the mechanical method of the interconnected framework, customary PID sort PSS and a fuzzy derivation identifier for
and the required extra damping can be given by supplementary on-line tunings of PID increases. Every parameter of PID
excitation control. controller is auto tuned by on-line fuzzy induction identifier

ISBN: 978-1-5090-6471-7/17/$31.00 ©2017 IEEE 1379

 International Conference on Computing, Communication and Automation (ICCCA2017)
under proposed fuzzy calculation for given dynamic and y( t ) = Cx( t ) (2)
receptive power conditions.
After the introduction in section I, section II describes the ⎡− D − K1 − K2 ⎤
plant model. Section III introduces to the conventional PSS. In ⎢M 0 ⎥
section IV, the design of hybrid fuzzy PID controller is studied. M M
⎢ 2πf 0 0 0 ⎥
Simulation results are presented in section V. Finally ⎢ ⎥
conclusions are drawn in section VI. A= ⎢ 0 − K4 1 −1 ⎥ (3)
⎢ Tdo' Tdo' K 3 Tdo' ⎥
II. PLANT MODEL ⎢ K K K K −1 ⎥
⎢ 0 − A 5 − A 6
A synchronous machine is appeared in Fig. 1, associated with ⎣ TA TA TA ⎥⎦
an infinite bus considered as the contextual analysis for this KA
paper. A solitary machine-infinite bus (SMIB) framework is B = [0 0 0 ]' (4)
considered in this paper for examination. This machine, which TA
is associated with an expansive framework through a C = [ 0 1 0 0] (5)
transmission line might be diminished to a SMIB framework
by the utilization of Thevenin's likeness transmission arrange X (t ) = [ Δδ Δω ΔEq ' ΔE fd ] (6)
outside to the machine.
The linearized one-machine infinite-bus display with III. POWER SYSTEM STABILIZER (PSS)
synchronous generator and exciter is appeared in Fig.1. In this The PSS essentially includes a settling sign that adjusts for the
figure, ω is the angular velocity [rad/sec], δ the rotor angle oscillations in voltage mistake of the excitation framework for
[rad], eq' the quadrate pivot voltage of generator, Efd the the dynamic/transient state operation, and to give a damping
electromotive drive, Δ the deviation amounts from given segment when it is on stage with the rotor speed deviation of
working focuses, and K1-K6 the linearized constants. the synchronous machine. Really PSS balances the AVR
By and large, the Heffron-Phillips model is taken as a fourth reference and thus realizes an adjustment in the control without
order linear model for the synchronous generator containing a influencing the essential controller prerequisite. An extra
third order synchronous machine and a first order AVR. This contribution to AVR is drawn from PSS as appeared in Fig 2.
model where direct estimate is helpful is reasonable for small
signal investigation. All things considered, all input and output
signals of this model have variations in little space around the
ostensible esteem delineated by delta 'Δ'. The linearization
method alongside the non-linear conditions are given in [5].
The piece graph of the direct model is appeared in Fig.1. There
are two noteworthy loops in Fig.1, the mechanical loop on top
and the electrical loop at the base. For little occasional Fig 2. Block diagram of PSS
oscillations, linearised conditions can be considered.
A. Conventional PSS (CPSS)
It comprises a wash out block, lead-lag compensator and
gain block. Wash out piece is utilized to decrease the
overreaction of the damping amid serious occasions. A lead lag
compensator is utilized for repaying the slack between the
electrical torque segment created by PSS and the speed
deviation [5]. The traditional PSS utilized as a part of this
paper is appeared in Fig.3.
2
sT W ⎛ 1 + sT 1 ⎞ (7)
V pss ( s ) = K pss
⎜⎜ ⎟⎟ Δ ω
1 + sT W ⎝ 1 + sT 2 ⎠
Fig. 1. Block diagram of Heffron-Phillips model
Where, V pss ( s ) is PSS output towards AVR
State equation of one machine infinite-bus model is as follows
The performance of CPSS gets degraded when operating
. point is changed. Also it is much sensitive to parameter
x ( t ) = Ax ( t ) + Bu ( t ) (1) variations [2, 3, 4].

1380
 International Conference on Computing, Communication and Automation (ICCCA2017)
TABLE I: FUZZY CONTROL RULES FOR KI

E
Ki
NB NS ZO PS PB
ΔE PB PS PB PB PB PB
Fig.3. Conventional PSS vs proposed controller (T=3.0; PS ZO PS PB PB PB
T1=0.685; T2=0.1; Kc=7.09)
ZO NS ZO PS PB PB
IV. CONSTRUCTION AND STRUCTURE OF FUZZY PID NS NB NS ZO PS PB
CONTROLLER
NB NB NB NS ZO PS
In the previous decade, Fuzzy Logic Controller (FLC) has
been created as controller for various applications in electrical
engineering including the outline of FLC as PSS. It has thus The used 5 linguistic variables used in fuzzy inference
incredibly enhanced the elements of power system [6]. In light algorithm are as follows: NB(negative big), NS(negative
of the unpredictability and multi-variable states of the power small), ZO(zero), PS(positive small), PB(positive big). Fig. 7
system, ordinary control strategies may not give palatable shows the initial membership functions of antecedent and
arrangements [5]. The robust control nature and high consequent variables.
unwavering quality of fuzzy rationale controllers makes it
exceedingly helpful in various control applications in power TABLE II: FUZZY CONTROL RULES FOR KD
system[7]. E
PSS is produced as a course of action of heuristic control Kd
NB NS ZO PS PB
rules, and the control variable is particularly gotten from the
lead base. PID controller additions are tuned on-line from the ΔE PB PS PB PB PB PB
information gave as the run base and fuzzy derivation, and a PS ZO PS PB PB PB
short time later, the standard PID controller makes the control
signal [9]. The fundamental block diagram of fuzzy PID ZO NS ZO PS PB PB
controller is appeared in Fig.4. NS NB NS ZO PS PB
NB NB NB NS ZO PS

TABLE III: FUZZY CONTROL RULES FOR KP

E
Kp
NB NS ZO PS PB
ΔE PB PS PB PB PB PB
PS ZO PS PB PB PB
Fig 4: Block diagram of Hybrid Fuzzy PID controller ZO NS ZO PS PB PB
The SIMULINK model of controller is appeared in Fig.5. NS NB NS ZO PS PB
The controller piece of fuzzy is framed by fuzzification of (ΔE)
where E is error (edge speed deviation) and ΔE is rate of NB NB NB NS ZO PS
progress of error, the interface system and defuzzification [8].
A control signal 'U', is therefore connected to the excitation
arrangement of every generator set [10]. The parameters, Kp,
Ki and Kd are dictated by an arrangement of fuzzy guidelines.
The parameters of PID controller at different working modes,
for example, stack aggravations and parameter's annoyances
can tune naturally and are dictated by fuzzy interface rules
appeared in tables I, II, III. The entire fuzzy model is appeared
in Fig.6. Considering the framework yield, the PID controller
brings about the accompanying control vector,
Fig.5: Simulink model of hybrid fuzzy PID controller
t
(8)
U = K pΔE + K i ∫ Δ Edt
0
+ K dΔE

1381
 International Conference on Computing, Communication and Automation (ICCCA2017)

Fig.6: complete Fuzzy system in SIMULINK

(b2)

(a1)

(b3)
Fig.7: Membership functions for input (a1,a2) and output variables
(b1, b3, b3)

V. SIMULATION RESULTS
The utilized power framework test system is given in Fig.
1. With the end goal of this paper, a routine PSS with a
compensator is picked with the proposed fuzzy PID controller
[5]. The framework parameters and starting qualities are
appeared in table IV.
(a2)
TABLE IV: SYSTEM PARAMETERS AND INITIAL VALUES

System parameters: Initial values

M =9.26 xd= 0.973 Peo=1.0
D=0 xd’= 0.19 Qeo=0.015
Td0·= 7.76 xq= 0.55 Vto=1.05
KA= 50 TA= 0.05
x =0.997 B = 0.262

The dynamic control exhibitions of fuzzy PID controller

(b1) and the customary PSS are contrasted and each other with
starting conditions indicated in Table IV. The Fig.8
demonstrates the dynamic qualities of rotor edge deviation in

1382
 International Conference on Computing, Communication and Automation (ICCCA2017)
the event of no PSS (Case1), with PSS (Case 2) in fig. 9 and of considerably stifled and the union rate additionally was
Hybrid fuzzy PID controller (Case 3) in fig. 10. surprisingly diminished.
From the outcomes, the exhibitions of the proposed fuzzy
PID controller are seen to be better than traditional ones in
connection to deviations and joining times.
At that point the linearized constants K1-K6 in Fig. 1 are
auto-tuned on-linearlly by the proposed fuzzy surmising
calculation as K1=0.5423, K2=1.2068, K3=0.6584,
K4=0.6982, K5=-0.0955, K6=0.8l59. The last estimations of
PID controller were obtained as Kp=32.5796, Ki=-0.1551and
Kd=3.2114.

Fig10: Dynamic chrecteristics of rotor angle deviation (Case 3,

Hybrid FPID)

Fig8: Dynamic chrecteristics of rotor angle deviation (Case

1,without CPSS)

Fig11:Comparison of Dynamic charecteristics of rotor angle

deviation (case 1,case 2,case 3)

Fig9: Dynamic chrecteristics of rotor angle deviation (Case 2,with

CPSS)

VI. CONCLUSION
Customary PSS brings about damping and frequency
oscillations. Consequently in this paper, a Hybrid Fuzzy PID
controller for the adjustment of SMIB power framework has
been proposed. The adjustment of power framework has been
brought out through PID Controller. The fuzzy controller is
executed to tune the parameters of the PID controller.
Fig12:Comparison of Dynamic chrecteristics of rotor velocity(w)
Simulation has been done in MATLAB. Reproduction deviation (case 1,case 2,case 3)
comes about demonstrate that the adjustment of power
framework utilizing FPID controller is conceivable and is
better than the routine PSS. The overshoot amount was

1383
 International Conference on Computing, Communication and Automation (ICCCA2017)
REFERENCES [6] M. A. Iskandar, “Stabilizing Control of Power System using
Fuzzy Control”, IEE J., Vol. 112-B, No. 12, pp. 1111-1120, 1992.
[1] F. P. Demello, C. Concordia, “Concepts of Synchronous Machine
Stability as Affected by Excitation Control”, IEEE Trans. on Power [7] A. Ishigame, H. Kaizu et al., “Power System Stabilizing Control
Apparatus and System, Vol. PAS-88, No. 4, pp. 189-202, 1969. Based on Simplification of Constructing Fuzzy Control Rules”,
IEEJ., Vol. 113-B, No. l, 1993.
[2] O. W. Hanson, C. J. Goodwin, P. L. Dandeno, “Influence of
Excitation and Speed Control Parameters in Stabilizing Intersystem [8] Witold Pedrycz, “Fuzzy Control and Fuzzy Systems, second
Oscillations”, IEEE Trans on Power Apparatus and System, Vol edition”, JOHN WILEY & SONS, 1993.
PAS-87, No. 5, pp.1306-1311, 1968. [9] H. H. Chung, D. I. Chung, S. M. Joo, “Stabilization of Power
System using Self Tuning Fuzzy Controller”, J. of Fuzzy Logic and
[3] H.B. Gooi, et al., “Coordinate Multi-Machine Stabilizer Settings Intell. Syst., Vol. 5, No. 2, pp. 58-69,1995.
Without Eigen value Drift”, IEEE Trans. On Power Apparatus and
System, Vol. PAS- 100, No. 8 [10] M.Hilal, Y. Errami, M.Benchagra and M. Maaroufi,”Fuzzy
Power control for doubly fed induction generator based wind farm”,
[4] R. J. Fleming, M. A. Mohan, K. Parvatisam, “Selection Journal of Theoretical and Applied Information Technology Vol. 43
Parameters of Stabilizers in Multi-machine Power System”, IEEE No.2, 2012, pp 321-330.
Trans. on Power Apparatus and System, Vol. PAS-100,No. 5, pp.
2329-2333, 1981.
[5] Yao-Nan Yu, “Electric Power System Dynamics”, pp.65-94,
Academic, Press 1983.

1384