2018 Twentieth International Middle East Power Systems Conference (MEPCON), Cairo University, Egypt

A design of a Modified Power System Stabilizer

for Power System Transient Stability
Enhancement
Zakaria Ebrahim Khalil Abd El-Fath El-Said Eliwa Waheed Sabry
Teaching Assistant Associate Professor Professor
Military Technical College Military Technical College Science Valley Academy
Cairo, Egypt Cairo, Egypt Cairo, Egypt
z.khalil@mtc.edu.eg a.eliwa@mtc.edu.eg ws123@sva.edu.eg

Abstract— In this paper, an adaptive fractional order
proportional integral derivative (FO-PID) power system
stabilizer (PSS) is proposed to enhance the transient stability
of a synchronous generator (SG) connected to an infinite bus.
A single machine to infinite bus (SMIB) power system is
subjected to various operating conditions and disturbances.
To damp the oscillations resulting at these situations, the SG
is equipped with a PSS. An adaptive FO-PID controller is
suggested to play the role of the PSS. To interpret the
effectiveness of the proposed FO-PID-PSS controller, several
simulation results are presented for the SMIB power system,
and compare the system in two cases: when equipped with
and without the proposed controller. The novel proposed FO-
PID-PSS controller has proved its efficiency in system
transient stability enhancement for the SMIB power system.
Keywords— Transient Stability, Single Machine Infinite
Bus System, Power System Stabilizer, Adaptive Fractional
Order Proportional – Integral – Derivative Controller.
Abbreviations:
AVR … Automatic Voltage Regulator
FO … Fractional Order
GL … Grunwald – Letnikov
IB … Infinite Bus
IO … Integral Order
PID … Proportional – Integral – Derivative
PSS … Power System Stabilizer
RL … Riemann – Liouville
SG … Synchronous Generator
SMIB … Single Machine Infinite Bus System
T.L. … Transmission Line
I. INTRODUCTION
The recent current trends in power system design are
toward the application of a large size generating units to
feed higher expected loads. For each power system, the SG
is the most imperative component. The most effective
control mean of the SGs is through the excitation system,
which implements control and protective functions essential
978-1-5386-6654-8/18/$31.00 ©2018 IEEE
to the satisfactory performance of the power system. The
PSS is an important part of the excitation system which
provides an extra input signal to damp power system
oscillations [1].
Stability analysis for power system has been received a
great deal of attention. Transient stability enhancement is
one of the essential aspects in modern power system.
Therefore, to achieve Transient stability enhancement,
a robust, strong, adequate and fast action control systems
are required to provide the compensation with which the
reduction in stability margin is offset [2].
There are different types of controllers and various
techniques to enhance stability using conventional PSS;
e.g., conventional PID optimized by particle swam [3, 4],
adaptive neuro fuzzy [5] and H ∞ mixed sensitivity
technique [6]. For a wide range of operating condition, the
conventional PSS cannot perform a robust acceptable
response. Therefore, it is required to design a more robust
PSS to enhance stability against the wide range of operating
condition [7].
In this paper, a novel model of adaptive FO-PID
controller is proposed to act as a PSS. The proposed FO-
PID based PSS controller is analyzed on a developed
nonlinear mathematical model for the SMIB. The SMIB
system model is also the base to analyze and to tune
the PSS controller to enhance the transient stability of the
SG through the excitation control system. The influence of
the proposed PSS on the transient performance of the SG is
investigated when the system is subjected to different
disturbances and operating at different conditions as stated
in the results and discussions.
II. POWER SYSTEM MODEL
The power system under consideration is shown in “Fig.
1”. It consists of a SG connected to an IB through a power
transformer and two parallel T.Ls [1].
Fig. 1. The Single Machine Infinite Bus System (SMIB)
712
 The mathematical model of the power system shown in u(t) = k e(t) + T D e(t) + T D e(t) (10)
“Fig. 1” is based on the state-space formulation. In this
model, the state-space variables are chosen to be the It can be expected that PIλDμ controller (as in
currents. This can be expressed in the matrix form as: “equations (10)”), may enhance the systems control
performance due to more tuning knobs introduced [10].
[I]∙ = [A]. [I] + [B]. [U] (1) One of the most important advantages of the PIλDμ
Where controller is the better control of dynamical systems, which
are described by fractional order mathematical models.
[I] = [i i i i i ] (2) Another advantage lies in the fact that the PIλDμ controllers
are less sensitive to changes of parameters of a controlled
The torque equation can be expressed in the form of:
system [11]. This is due to the two extra degrees of freedom
. ω∙ = T − T − T (3) to better adjust the dynamical properties of a FO control
system. It was shown that the best FO-PID works better
Where than IO-PID. For actually implementation, we introduced a
T = D. ω (4) modified approximation method to realize the designed FO-
PID controller.
Also, the relation between δ and ω can be defined as:
In Ref. [12] shows more strategies for efficient
∆ω = δ∙ = ω − 1 (5) implementation of FOPID in industrial applications. The
The SG is equipped with an IEEE static type 1-S exciter FOPID controller generalizes the IOPID controller and
and by feeding-back the terminal voltage signal and extends it from point to plane as shown in “Fig. 2”. This
comparing this signal with a reference value, the excitation expansion appends more flexibility to controller design
system is related to be an AVR [8]. [13].
The equations of the excitation system only can be
written in the state-space form as following:
−1 K
E T T E
= .
V −K K .K 1 V
−( + )
T .T T.T T
K E (6)
. (V + )
T K
+
K .K E
. (V + )
T.T K
Combining “(1), (3), (5) and (6)”, the complete system Fig. 2. Comparison between FOPID and IOPID
model will be constructed. Using FO-PID controllers, we have significantly
reduced percentage overshoot and rise and settling times
III. FO CONTROLLER compared to integral PID controllers [13].
There are commonly two definitions used for which are
the RL definition and the GL definition. The RL definition IV. PROPOSED ADAPTIVE FO-PID-PSS
is given as: In this paper, a proposed adaptive FO-PID-PSS is
1 d f(τ) designed. The parameters of this controller are not constant
aD f(t) = dτ (7) but they are computed according to the variation of the
Γ(m − r) dt (t − τ) system operating conditions. “Fig. 3” shows the block
for m − 1 < r < m where Γ( ) is the Euler’s gamma diagram of the controlled process, AVR, and the proposed
function. The GL definition is given as: PSS. The figure shows the method on which the design is
depend [14].
[ ]
h For a wide range of system operating conditions (active
r (8)
aD f(t) = lim h (−1) f(t − mh) power (P) and reactive power (Q)), the obtained controller
h→ m parameters are stored in look-up table against the system
operating conditions. During on-line operation, the
Where [ ] is an integer. controller monitors the P and Q values of the system and
h
picks up the corresponding controller parameters at each
In general form, the transfer function of PIλDμ is given sampling instant. The system of equations is:
by [9]:
X ∙ = A. X + B. U
U(s) (11)
C(s) = = k +Ts +T s (9) Y = C. X
E(s)
Involving an integrator of order λ and a differentiator of Where
order μ, where λ and μ are positive real numbers; KP is the [X] = i i i i i ω δ E V (12)
proportional gain, KI is the integral gain and KD is the
derivative gain. Clearly, taking λ = 1 and μ = 1, we obtain a
classical PID controller. The time domain formula is:

713
 (a) Rotor speed – time curve

Fig. 3. Adaptive FO-PID Controllers

Taking the Laplace transform for system of “Eq. (11)”:

. ( )= . ( )+ . ( )
(13)
( )= . ( )
(b) Terminal voltage – time curve
This can be rewritten as: Fig. 4. Comparing between FOPID PSS controller at various values of

( )=( . − ) . . ( ) (14) As shown in “Fig. 4”, it is clear that at λ = 0.75, μ =

Hence, 0.25, which indicates with the red curve, has the best
transient stability from the others. The simulation results
( ) = .( . − ) . . ( ) (15) for all the power system variables at λ = 0.75 and =
The control signal will be: 0.25 are shown in “Fig. 5”.

( )= ( ). ( ) In all figures, the curve in black represents the system

when working at steady-state, the curve in blue represents
. (16) the system when working under fault and without the
( )= . + + . . ( ) proposed control strategy, and the curve in red represents
1+ .
the system when working under fault and with the proposed
By equating “(15) and (16)” we get control strategy.
1
( )=
. . − ) .
(
. (17)
= .[ +
1+ .
+ . ]
The gains KP, KI and KD may be computed by finding
the given values of the open loop system, prespecifying the
eigen values of the closed loop system, and substituting the
three eigen values of “(17)”, we can get three equations
when solved together, we get KP, KI and KD. The input
signal to the PSS may be expressed as
(a) Rotor speed – time curve
= .∆ (18)
Where the damping amount is determined by the gain .
V. RESULTS AND DISCUSSIONS
To test the validity of the proposed control strategy
explained in previous sections, a simple disturbance of 10%
sudden increase in the input mechanical torque of SG is
supposed after 0.5 seconds and the fault is cleared after 2
seconds. The simulation results at various values of
integrator order (λ) and differentiator order (μ) of FOPID
controller are shown in “Fig. 4”.
(b) Terminal voltage – time curve

714
 (c) Torque angle – time curve
(d) Exciter field voltage – time curve
(e) Stator d-axis current – time curve
(f) Stator q-axis current – time curve
(g) Field winding current – time curve
(h) D-axis current of damper winding – time curve
715
(i) Q-axis current of damper winding – time curve
Fig. 5. Power system response to 10% sudden increase in the input
mechanical torque
In all figures, by comparing both curves red and blue,
the max overshoot of the red one is less than the blue one at
arise time less, which indicates better transient stability.
Also, it is clear that the oscillations of red curve is less than
that of blue one at less magnitude and frequency which
indicates better dynamic stability. From both curves, the
steady state stability represented by settling time and steady
state error becomes better in red curve.
The SG response curves are shown in “Fig. 5”, show
that the proposed adaptive FO-PID PSS has high capability
in improving the performance of the SG in comparison with
that of conventional AVR. Also, it is clear that both of the
transient and dynamic stability was enhanced which
represents a new achievements of this type of controllers.
To prove the robustness of adaptive FO-PID-PSS over
conventional PID-PSS, more check for this contribution
can be verified by the comparison between the
effectiveness of both the adaptive FO-PID-PSS and the
conventional PID-PSS ( = = 1 ) at the same fault
conditions. The simulation results are shown in “Fig. 6”.
In “Fig. 6”, the curve in blue represents the system
when working under fault and with the proposed control
strategy and the controller is conventional PID PSS (λ =
μ = 1), and the curve in red represents the system when
working under the same fault conditions and with the
proposed control strategy and the controller is adaptive
FOPID PSS.
(a) Rotor speed – time curve
(b) Terminal voltage – time curve
 (c) Torque angle – time curve
(d) Exciter field voltage – time curve
Fig. 6. Comparison between adaptive FOPID PSS and the conventional
(e) Stator d-axis current – time curve
(f) Stator q-axis current – time curve
(g) Field winding current – time curve
716
(h) D-axis current of damper winding – time curve
(i)Q-axis current of damper winding – time curve
PID PSS
By comparing both curves red and blue, the max
overshoot of the red one is less than the blue one at arise
time less, which indicates better transient stability. Also, it
is clear that the oscillations of red curve is less than that of
blue one at less magnitude and frequency which indicates
better dynamic stability. From both curves, the steady state
stability represented by settling time and steady state error
becomes better in red curve.
For more check of the validity of the proposed control
strategy, a severe disturbance is considered and simulated
as a short circuit in one T.L. after 0.5 seconds at its
midpoint with a successful enclosure of the circuit breakers.
We assume that the short circuit remains for 0.08 second
and the breakers are reclosed after 0.16 second. The
simulation results are shown in “Fig. 7”.
(a) Rotor speed – time curve
(b) Terminal voltage – time curve
Fig. 7. Power system response to a short circuit in one T.L.
 VI. CONCLUSIONS
In this paper, the transient performance of a SG when
equipped with a continuous acting AVR and adaptive FO-
PID-PSS is described. The effect of rotor speed error
feedback stabilizing signal on the generator response is also
examined. The proposed adaptive FO-PID-PSS is proved to
be an efficient mean for improving the SG transient
stability when the generator operate under light and at
severe disturbance conditions.
VII. REFERENCES
[1] P. Kundur, “Power system stability and control”, EPRI Power System
Engineering Series, McGraw-Hill, 1994.
[2] W. Sabry, “Effects of zero-reflection controller on the mitigation of
electromechanical wave propagation in one-dimensional ring power
system ”, International Review of Automatic Control, IREACO, Vol. (5),
No. (4), July 2012.
[3] Sayed Mojtaba Shirvani, Hemmati Reza, Delafkar Hamideh, Boroujeni
Amin Safarnezhad, “ Optimal PID power system stabilizer tuning based
on particle swarm optimization ”, Indian J Sci Technol 2011.
[4] Gaing ZL, “ A particle swarm optimization approach for optimum design
of PID controller in AVR system ”,IEEE Trans Energy Convers 2004.
[5] Ramirez-Gonzalez Miguel, Malik OP, “ Power system stabilizer design
using an online adaptive neuro-fuzzy controller with adaptive input link
weights ”,IEEE Trans Energy Convers, 2008.
[6] Sil Amitava, Gangopadhyay TK, Paul S, Maitra AK., “ Design of robust
power system stabilizer using H∞ mixed sensitivity technique ”,Third
international Conference on Power Systems, Kharagpur, INDIA
Dec.27−29, 2009.
[7] K. Haseena, J. Jacob, and A. T. Mathew, “A Robust Fractional order PSS
optimized using Genetic Algorithm for electromechanical oscillation
damping of SMIB system”, ICCC, International Conference on 2015, pp.
285-290, 2015.
[8] P. M. Anderson and A. A. Fouad, “Power system control and stability”,
GALGOTIA Publisher, 1984.
[9] J. A. Tenreiro Machado, “Theory analysis and design of fractional-order
digital control systems”, Journal: Systems Analysis Modeling Simulation,
Vol. 27, No. (2-3) 1997.
[10] Y. Chen, "Ubiquitous fractional order controls," IFAC Proceedings
Volumes, vol. 39, pp. 481-492, 2006.
[11] I.Podlubny, "Fractional-order systems and PI/sup/spl
lambda//D/sup/spl mu//-controllers," IEEE Transactions on automatic
control, vol. 44, pp. 208-214, 1999.
[12] B. M. Vinagre, C. A. Monje, A. J. Calderón, and J. I. Suárez, “Fractional
PID controllers for industry application. A brief introduction”, Journal of
Vibration and Control, vol. 13, pp. 1419-1429, 2007.
[13] D. Maiti, S. Biswas, and A. Konar, “Design of a fractional order PID
controller using particle swarm optimization technique”, 2nd National
Conference on Recent Trends in Information Systems, 2008.
[14] Y. Y. Hsu and C.J. Wu, “Adaptive control of a synchronous machine using
the auto-searching method”, IEEE Transactions of Power systems, Vol.
(3), No. (4), 1988.

717