Dasu et al.

Protection and Control of

Protection and Control of Modern Power Systems (2021) 6:35
https://doi.org/10.1186/s41601-021-00215-w Modern Power Systems

ORIGINAL RESEARCH Open Access

Small signal stability enhancement of a large

scale power system using a bio‑inspired whale
optimization algorithm
Butti Dasu1* , Sivakumar Mangipudi1 and Srinivasarao Rayapudi2

Abstract
A whale optimization algorithm (WOA)-based power system stabilizer (PSS) design methodology on modified single
machine infinite bus (MSMIB) and multi-machine systems to enhance the small-signal stability (SSS) of the power
system is presented. The PSS design methodology is implemented using an eigenvalue (EV)-based objective func-
tion. The performance of the WOA is tested with several CEC14 and CEC17 test functions to investigate its potential
in optimizing the complex mathematical equations. The New England 10-generator 39-bus system and the MSMIB
system operating at various loading conditions are considered as the test systems to examine the proposed method.
Extensive simulation results are obtained which validate the effectiveness of the proposed WOA method when com-
pared with other algorithms.
Keywords: Small-signal stability, Power system stabilizer, Stability, Modified single machine infinite bus system, Multi-
machine system, Whale optimization algorithm, Eigenvalues

1 Introduction classical PSS design techniques on various scenarios have

It is well-known that a power system is a practical exam- been developed after the realization of the role PSS role
ple of a highly non-linear interconnected system formed in the power system. Such techniques include the pole
by a number of transmission lines, switchgear equipment, placement technique [9, 10], sliding mode approach
various generators with diversified configurations, and a [11–14], linear quadratic regulator [15, 16], H − ∞ tech-
variety of loads. Increasing electrical energy demand has nique [17, 18], and fuzzy logic [19, 20]. The design of a
led almost all power grids to operate at their stability lim- PSS using the above-mentioned conventional methods
its. Under such highly stressed conditions, the nonlinear for single machine and multi-machine interconnected
behavior of the system is more pronounced by small dis- systems of highly varying loading conditions is tedi-
turbances. The Power System Stabilizer (PSS) has been ous, requires extensive expertise, and consumes much
developed as a supplementary controller to produce the computational time. Many mathematical calculations
necessary damping torque to suppress electromechani- and several parameters are required to design the PSS
cal oscillations in the range of 0.1–1.5 Hz originated from parameters for such a highly non-linear, time-varying,
small disturbances [1]. and complex large scale power system. In addition, the
The role of the PSS in coordination with the excitation design of a PSS using conventional techniques for the
system in a generator is illustrated in [2–8], while many multi-machine power system, one that always operates
at variable loading conditions is a complex process. In
recent decades, heuristic search algorithms have evolved
*Correspondence: dasu.geceeee@gmail.com and have proven their significance in solving complex
1
Department of EEE, Gudlavalleru Engineering College, Gudlavlleru, problems, optimizing design parameters, and minimizing
Krishna (Dt.), Andhra Pradesh 521301, India
Full list of author information is available at the end of the article

© The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.
 Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 2 of 17

the cost of fuel. PSS design using heuristic algorithms can infinite bus voltage. The details of this model are illustrated
reduce the above- mentioned drawbacks. in [44]. The well-known IEEE New England 10-generator
Extensive literature is available on the methods of PSS 39-bus MM test system is considered as the second test
design for the single machine and multi-machine power case to design the PSS parameters using the WOA.
systems including-Tabu search [21, 22], and algorithms The remainder of the paper is as follows: the test sys-
such as genetic [23, 24], particle swarm optimization tems considered are described in Sect. 2, while the struc-
(PSO) [25–29], differential evolution (DE) [30, 31], hybrid ture of the PSS and formulation of the objective function
differential evolution [32, 33], chaotic swarm optimization for the PSS design are provided in Sect. 3. PSS design
[34], fire-fly [35], harmony search [36], bacterial foraging using the WOA is described in Sect. 4, and in Sect. 5, the
search [37], cuckoo search [38], bat [39, 40], gravitational performance analysis of the WOA on CEC 14 and CEC
search [41], and grey wolf [42] algorithms. The above PSS 17 test functions is presented. Simulation results are pre-
design techniques, implemented on single and multi- sented in Sect. 6 and finally, Sect. 7 draws the conclusion.
machine systems have proven their efficacy in improving
the small-signal stability of the power system. However, 2 Systems under consideration
some drawbacks and limitations have also been observed 2.1 Case 1: MSMIB system
including lack of fast convergence, the need for large num- For a stability investigation, power system compo-
bers of design parameters, and not maintaining a good nents, for example, synchronous generators, exciters,
balance between the exploitation and exploration stages. and transmission networks are required to structure
To address these issues, a Whale Optimization Algorithm proper dynamic models. A unique model of the power
(WOA) is implemented in this paper on single machine system, one which consolidates all the elements of the
and large scale multi-machine systems to design the PSS system, is represented by the SMIB system, which is
parameters. Recently, WOA [43, 44] has been developed modelled as the K-constant model or HP model. This
based on the hunting strategy of the humpback whale. model has become very familiar to those studying SSS
The advantage of the WOA is the maintenance of a good issues in a power system. Numerous controllers have
balance between exploitation and exploration stages to also been developed for this model to investigate the
identify the best global solution, while very few numbers small-signal stability behavior [45]. A, modified ver-
of control parameters are required to. Compared to other sion of this model is developed which gives similar
heuristic search algorithms the number of steps involved performance to the Heffron Phillip’s (HP) model and
in the evolving process and the rate of the convergence in is called the Modified HP model (MHP). In this study,
the WOA are also improved. To validate the performance the PSS designed on the MHP model is considered as
of the WOA, it is tested on various CEC 14 and CEC 17 MPSS and on the multi-machine system is considered
test functions before designing the PSS parameters. Two as PSS. It is created by taking the generator side trans-
test cases working with various loading conditions under former secondary bus voltage as the reference rather
several disturbance conditions are considered in the design than considering an infinite bus voltage. One line dia-
of the PSS parameters using WOA. The first test case is the gram for the MSMIB system is shown in Fig. 1 and the
recently developed MSMIB system which is the modified block diagram of the MHP model is represented in
version of the SMIB system. This takes the transformer Fig. 2. Table 1 depicts the loading conditions consid-
secondary bus voltage of the generator side, instead of an ered for case 1.

The MSMIB (Multi-Machine Single-Infinite Bus) system is indeed a modified version of the SMIB (Single-Machine Infinite
Bus) system.

Fig. 1 One line diagram of the MSMIB system

CEC (Congress on Evolutionary Computation) benchmark functions are a set of standardized test functions used to assess the
performance of optimization algorithms. These functions are designed to provide a challenging environment for algorithms to
solve, with various characteristics such as multimodality, separability, and dimensionality.
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 3 of 17

Fig. 2 Block diagram of the MHP model

Table 1 Loading conditions considered for case 1 conditions is a crucial task for the operation engineers.
Loading Xe Pt Qt
PSS has developed as the supplementary controller to
condition the excitation system to provide the required damping
to mitigate the oscillations originatig from the distur-
1 0.3-Higher loading 1 0.2
bances. Typically, the PSS is formulated by three impor-
2 0.4-Nominal loading 0.8 0.41
tant blocks namely, the phase compensation, washout,
3 0.8-Weak loading 1 0.5
and gain blocks. The damping performance of the system
4 Lead.P.F loading 1 − 0.5
is affected by the contribution of each block. The phase
compensation block acts as a lead-lag compensator, the
wash out block acts as the high pass filter, and the gain
block provides necessary gain value to damp the oscilla-
2.2 Case 2: The New England 10‑generator 39‑bus system tions. The transfer function of the PSS is represented as
This test system [46] is the second case study for the sTWi

(1 + sT1i )(1 + sT2i )

design of PSS. Optimal tuning of PSS parameters for VS = KPSSi �ω1 (s)
1 + sTWi (1 + sT3i )(1 + sT4i )
the test case is carried out using the proposed tech-
(1)
niques. Various disturbances are created on the system
to investigate the performance of the proposed tech- here Vs is the output from the PSS, Kpssi is the gain value
nique. The block diagram of the New England 10-gen- of the PSS, subscript ’i’ represents the ith machine, Twi is
erator 39-bus system is shown in Fig. 3. the time constant of the washout block, T1i, T2i, T3i, and
T4i are the phase compensation blocks’ time constants
3 PSS structure and Δωi is the speed deviation of the ith machine. It has
It is well-known that a complex interconnected power been shown in the literature that the performance of the
system is always subjected to various types of dis- PSS for any test system can be significantly affected by its
turbances. Maintaining system stability under such parameters. Thus the proposed algorithm is developed
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 4 of 17

Fig. 3 Block diagram of the New England 10-generator 39-bus system

to properly design the parameters using the mentioned function. Hence the EVs will move to the left region of
objective function for all the case studies. the imaginary axis as shown in Fig. 4a. The damping
ratios of the lightly damped oscillating modes of the sys-
tem will be improved, if the second objective function is
3.1 Formulation of the EV‑based objective function for PSS
considered alone. This is represented as
design
The single objective function for the design of the PSS is Np
 
formed by combining two independent objective func- min Jb = (ζ0 − ζi )2 (3)
tions to place the lightly damped and unstable EVs of the j=1 ζi ≥ζ0
system into the required s-plane regions. Figure 4 repre-
sents eigenvalue regions of the EV-based objective func- where ζi is the damping ratio of the ith EV of the sys-
tion. The first objective function is represented as tem, ζ0 is a constant value of the expected damping ratio
and is considered to be greater than 0.3. The EVs will be
NP 
 pushed into the wedge shape sector of Fig. 4b, when the
min Ja = (σ0 ≥ σi )2 (2) PSS parameters are optimized based on this objective
j=1 σi ≥σ0 function. The advantage of this objective function is that
the damping ratio of the low damped EVs of the system
where Np is the total number of loading conditions con-
can be improved. Finally, the desired objective function
sidered for the optimization process, σi is the real part of
is constructed by combining the two objective functions
the ith eigenvalue of the system and σ0 is a constant value
into one by assigning a weighting factor-C, used to offset
of the expected damping factor and is chosen as − 1.5.
the weights of Ja and Jb as the value of c is chosen as 10
The real parts of poorly damped EVs of the system will be
[22].
improved when the PSS is designed using the objective
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 5 of 17

Fig. 4 Eigenvalue regions of the EV-based objective function

Table 2 Ranges of control parameters to be evolved

Loading condition-1 Loading condition-2 Loading condition-3 Loading condition-4

10 ≤ Kpss ≤ 100 10 ≤ Kpss ≤ 100 10 ≤ Kpss ≤ 100 10 ≤ Kpss ≤ 100

0.001 ≤ T1 ≤ 1 0.001 ≤ T1 ≤ 1 0.001 ≤ T1 ≤ 1 0.001 ≤ T1 ≤ 1
0.001 ≤ T2 ≤ 1 0.001 ≤ T2 ≤ 1 0.001 ≤ T2 ≤ 1 0.001 ≤ T2 ≤ 1
0.001 ≤ T3 ≤ 1 0.001 ≤ T3 ≤ 1 0.001 ≤ T3 ≤ 1 0.001 ≤ T3 ≤ 1

global best solution and the exploitation stage is required

JI = Ja + cJb (4)
to get the local best solution. Achieving a good balance
between the two stages is the most challenging task for an
Np Np
 
2
  optimization algorithm because of the stochastic nature of
JI = (σ0 − σi ) + C (ζ0 − ζi )2 (5)
the algorithm. If an algorithm can succeeded in obtaining
j=1 σi ≥σ0 j=1 ζi ≥ζ0
a good balance between the two, the optimal solutions can
Thus, all the poorly damped eigenvalues will be shifted in be found. As the WOA has been more successful in main-
the marked regions as shown in Fig. 4c, when JI is used. taining a good balance between the exploitation and the
exploration stages than the other algorithms [43], it is used
4 PSS design using WOA here for the design of the parameters of the PSS for the test
The WOA was developed by Seyedali Mirjalili in 2016, cases.
by observing the hunting strategy of Humpback whales.
Humpback whales, when attacking targets, follow a spe- 4.1 Steps involved in the WOA algorithm
cial type of hunting technique which is known as the In the WOA, the steps of the hunting strategy of the whales
bubble-net feeding method. In this technique, the whales are listed as:
create two types of mechanisms to reach the target, i.e.,:
a shrinking encircle mechanism and a spiral mechanism. 4.1.1 Step 1: Initialization
The WOA has been developed based on these two hunting To start the optimization process, PSS parameters are
mechanisms and it has several advantages over other meth- selected as control variables and their ranges with mini-
ods. The biggest advantage of the WOA is that only one mum and maximum values are listed in Table 2. The num-
design parameter (a) is required for the optimization pro- ber of iterations taken as 100, and the population size is
cess [43]. This is fewer than other PSO and DE algorithms. selected to be 40. The initial solutions are randomly gener-
The convergence rate of the WOA is superior and the num- ated by using the expression given as:
ber of steps required in the WOA is also fewer than PSO  
and DE. After initialization of the design parameters, PSO Xj0i = Xjmin + rand. Xjmin − Xjmax (6)
requires an update of the position and velocity for each
particle. For DE algorithm, crossover and mutation steps where ’X’ is the control variable, and Xjmin and Xjmax are
are required for the optimization process. For any optimi- its minimum and maximum values, respectively. j = 1,
zation algorithm, exploration and exploitation are the two 2,…, N, where N is the number of control variables, and
important stages upon which the convergence of the opti- i = 1, 2, 3, …, NP, where NP is the population size, rand
mal solution depends. The exploration stage is to get the €[0, 1] is a random number that changes between 0 and 1.
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 6 of 17

All the tuneable parameters are selected with their ranges where A  is a random value that decreases from 2 to 0 dur-
and randomly generated using (16). ing the optimization process and r is an arbitrary number
that varies between (0and,1). In every iteration process,
4.1.2 Step 2: Determination of EV‑based objective function the values of a, A, R are updated for each hunting agent.
To place the EVs in pre-determined regions, the EV-based The present hunting agents update their locations by (7),
objective function described in Sect. 3 is used. After ini- but if the value of A is less than 1, the following are used
tial solutions of the PSS are randomly generated, the Sim- instead:
ulink models of the test cases in the MATLAB program are
called and then the EVs of the test systems are determined.
 
S� = R
� �
· Prand − P�  (11)

The EV-based objective function for the given population
size s then found, and the EV-based objective function is
determined for both test cases working with various oper- � + 1) = P� ∗ (t) − A
P(t � · S� (12)
ating conditions.

4.1.3 Step 3: Updating hunting agent path using shrinking

encircling mechanism
After determining the objective function, updating the
hunting agent process using the WOA is imitated. All the
PSS parameters are treated as hunting agents in the opti-
mization process. When the location of the target is iden-
tified, the hunting agent updates its posture by using this
mechanism to reach the prey. Since the best solution is not
obvious, WOA assumes that the current solution is the
optimum solution and updates its position using:
 
S� = R
� � � 
· P ∗ (t)P(t) (7)

� + 1) = P� ∗ (t) − A
P(t � · S� (8)

where S is the distance between the whale and target, R 


and, A are the coefficient vectors, t is the current itera-
tion, P* corresponds to the optimum solution attained up
to now, and P stands for the position vector.; || represents
the absolute value, and ’. is the multiplication of elements
to elements. The vectors A  , and R
 are characterized as:

� = 2�a · �r − a�
A (9)

� = 2 · �r
R (10)

Table 3 Design parameters of optimization algorithms

#Algorithm# Parameters

PSO Number of generations = 100

Population size = 50
C1= 2, C2 = 2
Weighing factor, w = (0.9–04)
DE Number of generations = 100
Population size = 50
Mutation constant, F = 0.5
Cross over constant, C.R = 0.8
WOA Number of generations = 100
Population size = 50
Constant, a = varies between 2 and 0 Fig. 5 Flowchart for WOA to design the PSS parameters
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 7 of 17

where �ω is a position of the whale selected randomly where δ is an arbitrary number that varies between 0 to 1.
from the present population. The flowchart for the WOA to design the PSS
parameters is depicted in Fig. 5, which describes vari-
ous steps involved in the WOA. It starts with the ini-
4.1.4 Step 4: Updation of hunting agent path using a spiral tialization of the PSS parameters with their minimum
mechanism and maximum values, and then leads to the genera-
In this step, the hunting agents follow in a spiral- tion of initial solutions, determination of the objective
shaped path to reach the target. To simulate the spi- function, optimization of the PSS parameters accord-
ral-shaped track between the whale and target, a spiral ing to the mechanisms mentioned above, and stopping
equation is formulated. All the hunting agents update criteria of the algorithm when the number of genera-
their position based on the formulated equations as: tions is completed. Table 3 shows the design param-
eters of the optimization algorithms.
� + 1) = S� ′ ebl · cos(2 I) + P� ∗ (t) − A
� · S� (13)

P(t
5 Performance analysis of WOA
where Many complex test functions are tested with the pro-
  posed WOA to prove its performance in finding the
S� = P� ∗ (t) − P(t)
�  (14)

global solution, and the test results for CEC 14 and CEC
17 functions are displayed in Tables 4 and 5 respec-
Here ’l’ is a stochastic limit that diverges between 0 and 1. tively. CEC14 test functions,e.g., High Conditioned
The shrinking encircle itinerary and spiral itinerary Elliptic, Discuss, Rosenbrocks, Weierstrass, Griewanks,
are merged by giving 50% probability to each of them Rastrigins’s, Katsuura, Happy Cat, and Griewanks plus
to update the positions of the hunting agents to reach Rosenbrocks Functions are with the proposed WOA.
the prey as: The test results are also compared with PSO and, DE
algorithms. As shown in Table 4, the results show that
P� ∗ (t) − A � · S�

� + 1) = if δ∠0.5
P(t
S� ′ ebl · cos(2 I) + P� ∗ (t) − A

� · S� if δ ≥ 0.5
the WOA is succeeded in finding the global best fit-
(15) ness values for all test functions better than the other
algorithms.

Table 5 Test results of CEC 17 test functions

S. nos. Name of the function Order of the WOA DE PSO
function Fitness value

1 Bent Cigar function f1(x) 4.0897 e−88 3.3893 e−04 1.9313 e−15
2 Sum of differential power function f2(x) 99.515675945535875 99.5157 99.5157
3 Zakharov function f3(x) 9.5876 e−80 4.1649 e−09 2.7229 e−06
4 Rosenbrocks function f4(x) 1.1866 e−06 3.0034 e−03 1.6720 e−02
5 Rastrigins’s function f5(x) 1.3548 e−80 5.6960 e−05 0.001.2 e−03
6 Expanded Schaffers function f6(x) 0.0097 0.0065 0.0532
7 Lunaeek bi-Rastrigin function f7(x) 2.7252 e−07 0.0373 0.1205
8 Levy function F8(x) 70.0737 e+00 70.3698e+00 78.0292e+00
9 Modified Schwefel’s function F9(x) 4.1183e+06 4.1479e+06 4.1482e+06
10 High conditioned elliptic function f10(x) 3.5101 e−80 3.5632 e−04 1.6684 e−80
11 Discuss function F11(x) 1.7573 e−73 8.6146 e−13 3.1554 e−22
12 Ackley’s function F12(x) 1.6556e+002 1.6587e+002 8.5076e+00
13 Weierstrass function F13(x) 9.441375732713 e−03 9.441375732812 e−03 9.441375733129 e−03
14 Griewanks function F14(x) 5.6654 e−08 0.3131 e−04 1.0852 e−07
15 Katsuura function f15(x) 1.7937 e−10 4.4305 e−10 1.7377e−10
16 HappyCat function f16(x) 1.7126 e−04 0.9555 1.6875 e−08
17 HGBat function f17(x) 2.7428 e−06 2.5256 e−03 8.2076 e−04
18 Expanded Griewanks plus Rosen- f18(x) 63.6412 4.5131 7.9816
brocks function
19 Schffer’s function F19(x) 0.1606 0.0090 0.4613
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 8 of 17

Table 4 Test results of CEC 14 test functions

S. nos. Name of the function Order of the WOA DE PSO
function Fitness value

1 High Conditioned Elliptic function f1(x) 3.5101e−80 3.5632 e−04 1.6684 e−80
2 Bent Cigar function f2(x) 4.0897e−88 3.3893e−04 1.9313e−15
3 Discuss function f3(x) 1.7573 e−73 8.6146 e−13 3.1554 e−22
4 Rosenbrocks function f4(x) 1.1866 e−06 3.0034 e−03 1.6720 e−02
5 Ackley’s function f5(x) 4.4409 e−14 4.4409 e−14 4.4409 e−14
6 Weierstrass function f6(x) 9.441375732713 e−03 9.441375732812 e−03 9.441375733129 e−03
7 Griewanks function f7(x) 4.6928 e−08 7.0573 e−07 4.2514 e−06
8 Rastrigins’s function f8(x) 1.3548 e−80 5.6960 e−05 0.001.2 e−03
9 Modified Schwefel’s function f9(x) 0.0073 0.0038 0.0083
10 Katsuura function f10(x) 1.7937 e−10 4.4305 e−10 1.7377 e−10
11 HappyCat function f11(x) 1.7126 e−04 0.9555 1.6875 e−08
12 HGBat function f12(x) 2.7428 e−06 2.5256 e−03 8.2076 e−04
13 Expanded Griewanks plus Rosen- f13(x) 63.6412 4.5131 7.9816
brocks function
14 Expanded Schaffer’s function f14(x) 0.009.7 e−03 0.006.5 e−03 5.32 e−02

Fig. 6 Convergence plots of CEC14 and CEC 17 functions

Similarly, the CEC17 test functions are such as Ras- 6 Simulation results and discussions
trigins’s, Lunaeek bi-Rastrigin, Ackley’s Function, Grie- 6.1 Case 1
wanks, HG Bat, Griewanks plus Rosenbrocks and The WOA and DE algorithms are run several times con-
Schffer’s functions are tested with the proposed WOA to sidering the new objective function. Evolved PSS param-
derive the best fitness value. The test results prove that eters for four loading conditions are listed in Table 6. The
the WOA can derive the best i.e. better than the other following disturbances are considered to test the efficacy
algorithms. The convergence plots of CEC14 and CEC and vigor of the proposed WOA-MPSS.
17 functions are shown in Fig. 6. From these results, it
can be seen that WOA performed better in deriving the i. 10% step change at Vref
global best fitness values than the other algorithms. ii. 10% step change at Tm
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 9 of 17

Table 6 Evolved Parameters of PSS for the four loading conditions

Loading conditions #Algorithm# PSS parameters
Kpss T1 T2 T3 T4

1 WOA 42.33 0.1053 0.0247 0.0481 0.0192

DE 39.2799 0.1004 0.02330 0.0499 0.0210
PSO 40.9652 0.1001 0.0257 0.0120 0.0251
MPSS 13.00 0.0952 0.0217 0.0952 0.0217
2 WOA 43.125 0.129 0.031 0.0601 0.0301
DE 39.2597 0.113 0.028 0.0505 0.029
PSO 40.6325 0.1011 0.0284 0.0101 0.0284
MPSS 13.00 0.0952 0.0217 0.0952 0.0217
3 WOA 39.789 0.0124 0.0119 0.0348 0.0147
DE 39.2799 0.1004 0.0233 0.0499 0.0210
PSO 25.025 0.1000 0.0232 0.1110 0.1240
MPSS 13.00 0.0952 0.0217 0.0952 0.0217
4 WOA 41.0215 0.1124 0.0214 0.0487 0.0412
DE 39.2597 0.1130 0.0280 0.0505 0.0340
PSO 40.0257 0.251 0.0284 0.3000 0.0284
MPSS 13.00 0.0952 0.0217 0.0952 0.0217

Fig. 7 System response plots for the disturbance of 10% step change at ­Vref

The simulation plots are shown in Fig. 7 for distur- plots obtained using the ISE objective function by PSO
bance of 10% step change at Vref and Fig. 8 for the distur- and PSO-based PID are compared with the WOA and
bance of 10% step change at Tm for the four operating DE based PSS which are tuned with the EV-based objec-
conditions. The plots represent system responses for tive function. The reason behind using the EV-based
the conventional design of PSS (MPSS), PSO-based objective function is that it can relocate the unstable and
MPSS (PSO-MPSS), PSO-PID-based MPSS (PSO-PID- lightly damped poles into the desired locations. From
MPSS), and WOA-based MPSS (WOA-MPSS). The the results, it can be seen that that the intensity of the
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 10 of 17

oscillations has been minimized with the WOA-MPSS. complex conjugate poles with low damping ratios are
The time taken for the oscillations to settle down is also considered to be shifted into the stable regions. Because
reduced with WOA-MPSS when compared to other sta- of this reason, the EV-based objective function is con-
bilizers in all cases. Hence it is concluded that the pro- structed in such a way that only lightly damped oscillat-
posed WOA-MPSS tuned with the EV-based objective ing modes are shifted into the desired locations.
function shows superior performance over the MPSS, For example, for loading condition 1, with MPSS, the
PSO-MPSS, DE-MPSS, and PSO-PID-MPSS for all the pair of complex conjugate poles which causes the oscil-
loading conditions under the typical disturbances. latory behavior are located at − 1.34 ± 6.39i with the
To further, demonstrate the performance of the pro- damping ratio (ς0) of 0.20. It is shifted to − 2.03 ± 6.44i
posed PSS design technique, EV analysis (EVA) is car- with a damping ratio of 0.30 when MPSS is optimized
ried out for all the loading conditions and is displayed in with PSO. It means that the real value (σ) of the com-
Table 7 for four loading conditions with NO-PSS, MPSS, plex conjugate pole is shifted from − 1.34 to − 2.03. Then
PSO-MPSS, DE-MPSS, PSO-PID-MPSS, and WOA- it is further shifted to − 6.16 ± 10.1i with a damping
MPSS. It is seen from the EVs that some of the poles lie ratio of 0.52 when PSS is tuned with PSO- PID-MPSS.
on the right-hand side of the s-plane when the PSS is not Finally, with WOA-PSS, it is shifted to − 2.06 ± 1.57i and
established in the system and hence the system becomes the damping ratio is improved to 0.79. Thus the damp-
unstable. It is known that, for any system, the complex ing ratio has been improved from 0.20 to 0.79 with the
conjugate or imaginary poles with low damping ratio are proposed WOA-MPSS. Similarly, for loading condition
responsible for the oscillatory behavior and the oscilla- 2, the oscillation causing eigenvalue pair are located at
tory behavior is pronounced when they are located near − 1.24 ± 6.27i with the damping ratio of 0.19. When the
the imaginary axis. On the other hand, the real poles do PSO-MPSS is established in the system the damping ratio
not cause any oscillatory behavior in the system as the is improved to 0.23 and, with DE-MPSS it is improved to
damping ratio of the real root is unity. In the present 0.26. After that with PSO-PID-MPSS, the damping ratio
work, the real roots of the system with the proposed and of the oscillatory causing eigenvalue is improved to 0.43,
other methods for all the loading conditions are already and finally to 0.80 with the proposed WOA-MPSS For
located in the desired and stable regions. However, the the loading condition 3, the damping ratio is improved
damping ratio of the oscillation causing eigenvalues is from 0.15 to 0.70 with the proposed WOA-MPSS. A sim-
less than unity and there is a need to improve the damp- ilar improvement has also been observed with the fourth
ing ratio of these eigenvalues to improve the damping operating condition. Hence, from all these results it is
performance of the system. Therefore, in this work, much seen that when the PSS parameters are optimized with
attention is paid to these complex conjugate poles with the proposed WOA, the damping ratios of the oscilla-
low damping ratios which causes oscillations in the sys- tion causing eigenvalues are improved and the damping
tem. Therefore, the oscillation causing eigenvalues or performances of the system with the proposed optimized

Fig. 8 System response plots for the disturbance of 10% step change at ΔTm
Table 7 Eigenvalue analysis of four loading conditions
Loading condition 1 Loading condition 2 Loading condition 3 Loading condition 4

NO-PSS 0.91 ± 8.55i − 0.10 8.60 1.84 ± 9.59i − 0.188 9.77 1.75 ± 8.90i − 0.19 9.07 2.88 ± 12.8i − 0.220 13.1
− 26.1 ± 13.3 0.89 29.3 − 27.1 ± 15.1i 0.87 31.0 22.0 1.00 22.0 − 28.1 ± 24.3i 0.757 37.2
MPSS − 0.51 1.00 0.51 − 0.515 1.00 0.51 − 0.51 1.00 0.51 − 0.59 1.00 0.52
− 1.34 ± 6.39i 0.20 6.53 − 1.24 ± 6.27i 0.19 6.39 − 1.03 ± 6.73i 0.15 6.81 − 1.71 ± 6.07i 0.27 6.30
− 15.6 1.00 15.6 − 16.9 1.00 16.9 − 17.8 1.00 1.78 − 8.74 1.00 8.74
− 44.3 1.00 44.3 − 56.2 ± 15.6i 0.96 58.4 − 49.0 1.00 49.1 − 29.8 1.00 29.8
1.00 76.6 1.00 66.6 1.00 116
Dasu et al. Protection and Control of Modern Power Systems

− 76.6 − 66.6 − 116

PSO-MPSS − 0.5 1.00 0.52 − 0.51 1.00 0.51 − 0.51 1.00 0.51 − 0.55 1.00 0.55
− 2.03 ± 6.44i 0.30 6.75 − 1.58 ± 6.55i 0.23 6.73 − 1.63 ± 6.79, 0.23 6.99 − 3.66 ± 5.71i 0.54 6.78
− 17.7 1.00 17.7 − 19.5 1.00 19.5 − 20.1 1.00 20.1 − 8.11 1.00 8.11
− 45.3 ± 4.50i 0.99 45.5 − 45.5 ± 12.5i 0.96 47.2 − 42.3 ± 7.18i 0.98 42.9 − 28.2 1.00 28.2
(2021) 6:35

PSO-PID-MPSS − 1.70e−15 1.00 1.7e−15 − 2.1e−16 1.002 1e−16 − 1.00e−15 1.00 1.0e−15 − 2.1e−15 1.00 2.1e−15
− 0.26 1.00 0.26 -0.37 1.00 0.37 − 0.39 1.00 0.39 − 0.02 1.00 0.02
− 6.16 ± 10.1i 0.52 11.8 − 3.53 ± 7.42i 0.43 8.22 − 3.86 ± 8.10i 0.43 8.97 − 3.73 ± 10.0i 0.34 10.7
− 13.6 1.00 13.6 − 21.3 1.00 21.3 − 19.1 1.00 19.1 − 11.6 1.00 11.6
− 109 1.00 109 − 106 1.00 106 − 107 1.00 107 − 158 1.00 158
DE-MPSS 0.51 1.00 0.51 48.1 1.00 48.1 − 0.44 1.00 0.44 − 3.32 ± 5.42i 0.52 6.35
− 1.89 ± 6.46 0.28 6.73 − 1.70 ± 6.061i 0.26 7.24 − 1.60 ± 6.84i 0.22 7.03 − 8.84 1.00 8.84
− 17.6 1.00 17.6 − 20.2 1.00 20.2 − 19.9 1.00 19.9 − 27.7 1.00 27.7
43.9 1.00 43.9 − 40.5 1.00 40.5 − 38.7 1.00 38.7 − 67.8 1.00 67.8
− 61.9 1.00 61.9 − 41.4 1.00 41.4 − 46.5 46.5 1.00
Proposed WOA-MPSS − 0.10 1.00 0.10 − 0.10 1.00 0.10 − 0.10 1.00 0.10 − 0.10 1.00 0.10
− 2.06 ± 1.57i 0.79 2.59 − 1.85 ± 1.41i 0.80 2.33 − 1.98 ± 2.26i 0.70 3.00 − 1.91 ± 1.36 0.81 2.34
− 7.25 1.00 7.25 − 14.0 ± 7.85i 0.90 0.16 − 8.26 1.00 8.26 − 14.6 ± 1.36i 0.98 15.00
− 35.10 ± 10.00i 0.96 36.50 − 117.00 1.00 117 − 27.10 1.00 27.1 − 128.00 1.00 128.0
Page 11 of 17
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 12 of 17

Table 8 Evolved parameters of PSS for case 2 using WOA incorporated (without proper tuning), more oscillations
#Gen Kpss T1 T2 T3 T4
are observed in the system. This is because of the lack
of sufficient damping torque. This is not desirable from
G2 51.234 0.345 0.0136 0.412 0.014 the stability point of view. On the other hand, when PSS
G3 29.126 0.674 0.0275 0.623 0.039 parameters are designed with the WOA and placed at
G4 42.654 0.728 0.0501 0.698 0.048 respective generators, the oscillations are reduced. The
G5 50.143 0.126 0.0621 0.294 0.026 intensity of the oscillations is considerably lower with
G6 47.865 0.701 0.0152 0.964 0.065 WOA-PSS compared to PSO-PSS and DE-PSS, and the
G7 2.879 0.379 0.0282 0.478 0.122 simulation results show that WOA-PSS provides better
G8 27.125 0.852 0.0169 0.757 0.019 dynamic performance characteristics over DE-PSS and
G9 5.654 0.248 0.0512 0.265 0.298 PSO-PSS.
G10 19.597 1.023 0.0389 1.21 0.031 The system behavior under the second disturbance
of line outage between 14 and 15 is studied. Extensive
simulation results on the speed deviations are obtained
and presented in Fig. 12. The simulation results show
parameters of the WOA are better enhanced for all the
that WOA-PSS provides better damping performance
loading conditions than other methods.
than the other two methods. For the third scenario, the
speed deviation under the disturbance of a 25% increase
6.2 Case 2 in loads at buses 16 and 21, a 25% increase in the gen-
To test the efficacy of the suggested WOA-PSS, a second eration of G7, and the line outage between 21 and 22 are
test case is considered. The optimal design of PSS is car- shown in Fig. 13. From these responses, it is noted that
ried out by considering the EV-based objective function the oscillations are reduced and settled in quicker with
using the WOA, DE, and PSO algorithms. All the gen- the WOA-PSS compared to DE-PSS and PSO-PSS at all
erators except generator one, are equipped with PSS and the generators shown. Hence, WOA-PSS can provide
the parameters are optimized using the WOA. In total better damping performance than the other two stabiliz-
45 parameters named ­KPssi, ­T1i, ­T2i, ­T3i and T4i are opti- ers. Table 9 depicts the time response specifications of
mized, as listed in Table 8. To test the effectiveness and generators under scenario 2. As seen, the settling time
robustness of the proposed WOA-PSS various distur- is decreased from 9.8300 to 8.9371 s with WOA-PSS for
bances are created at various locations of the test system. generator two. In the case of generator three, the settling
These conditions are extremely harsh from a stability time is reduced from 8.4502 to 8.4913 s, where as for
point of view with the following cases studied to investi- generator five, oscillations are settled in 7.9892 s with the
gate the efficacy of the WOA-PSS: WOA-PSS. At generator six the settling time is 7.0971 s
when PSO-PSS is used and is reduced to 6.1351 s with
Scenario 1 line outage between 21 and 22. DE-PSS and to 5.9735 s with WOA-PSS. Similarly, the
Scenario 2 line outage between 14 and 15. time taken for settling down the oscillations at generator
Scenario 3 25% increase in loads at buses 16 and 21, seven with PSO-PSS is 7.1417 and is reduced to 6.7459 s
25% increase in the generation of G7, and line outage with WOA-PSS. A similar decrement is also observed
between 21 and 22. in the remaining generators. The peak overshoot is also
decreased with WOA-PSS when compared to PSO-PSS
Figure 9 represents Scenario 1 of case 2. To create the and DE-PSS. Hence it is shown that the WOA-based sta-
disturbance, the line between buses 21 and 22 is open- bilizers give better transient responses than the other DE
circuited and the stability behavior of the generators with and PSO based stabilizers for all the generators under
the proposed WOA-PSS and other PSSs are studied. this disturbance with significant improvements in peak
After that, another disturbance of line outage between overshoot and settling time.
buses 14 and 15 (scenario 2) is created and the results The strength of the WOA-PSS is tested with the third
are shown in Fig. 10. Similarly, the third disturbance con- disturbance condition. and Fig. 13 depicts the speed
dition of the line outage between 21 and 22 is created deviation plots under the disturbance of a 25% increase
under scenario 3 to effectiveness of the proposed PSS in loads at buses 16 and 21, a 25% increase in the gen-
design technique, and Fig. 11 shows the corresponding eration of G7, and the line outage between 21 and 22.
speed deviation of the system. It can be observed from The peak overshoots of the oscillations at generator two,
the results that the system becomes unstable when PSSs seven, eight, and nine are reduced with the WOA-PSS
are not connected to the generators. When PSSs are compared to other stabilizers designed with the DE and
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 13 of 17

Fig. 9 Scenario 1 of case 2

Table 9 Time response specifications of generators under PSO algorithms. The oscillations are also settled quicker
scenario 2 with the WOA-PSS. Thus, it is concluded that WOA-PSS
#Gen Time specifications PSO DE WOA can provide better damping performance than DE-PSS
and PSO-PSS.
G2 Settling time (s) 9.8300 8.7539 8.3971 EVA is carried out for all the scenarios to test the effec-
%Overshoot 218.9909 180.5292 146.1780
tiveness of the proposed design technique. The EVA
G3 Settling time (sec) 8.4502 8.6446 8.4913
%Overshoot 88.6049 65.3729 52.1090 of the system under the three disturbance conditions is
G4 Settling time (s) 7.8388 8.2694 8.0895 depicted in Table 10 and it can be seen that the lightly
%Overshoot 145.0600 115.3542 93.0143 damped EVs are shifted more towards the left-hand side
G5 Settling time (s) 7.6893 8.1954 7.9892 of the s-plane with the proposed WOA-based design
%Overshoot 221.1063 152.8303 103.0659 technique at all the disturbance conditions than the
G6 Settling time (s) 7.0971 6.1351 5.9735 other methods. This reveals that the damping factor of
%Overshoot 660.1628 670.0134 667.3342
the lightly damped EVs is better improved and system
G7 Settling time (s) 7.1417 6.9144 6.7549
%Overshoot 474.768 470.9769 465.1441 enhanced with the WOA-PSS than with PSO-PSS and
G8 Settling time (s) 8.0548 8.3397 8.259 DE-PSS under all the disturbance conditions. From all
%Overshoot 85.6159 66.6644 52.7958 the simulation results it can be concluded that the SSS of
G9 Settling time (s) 8.3247 8.1519 8.5480 the system is improved with the proposed WOA-based
%Overshoot 83.1522 53.5511 44.7755 PSS design technique, and better than with the other
G10 Settling time (s) 7.9526 7.8140 7.6259 design approaches.
%Overshoot 162.2931 149.6141 141.9001
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 14 of 17

Fig. 10 Scenario 2 of case 2

Fig. 11 Speed deviation plots under the disturbance of line outage between 21 and 22
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 15 of 17

Fig. 12 Speed deviation plots under the disturbance of line outage between 14 and 15

Fig. 13 Speed deviation plots under the disturbance of 25% increase in loads at buses 16 and 21, 25% increase in generation of G7, and the line
outage between 21 and 22

7 Conclusion requires less time for tuning, and reduces computational

A PSS design technique on the MSMIB and the large complexity compared to PSO-PSS, and DE-PSS. Results
scale New England 10-generator 39-bus multi-machine reveal that the proposed WOA-based PSS can improve
systems using a WOA is proposed. The PSS design the dynamic performance of the MSMIB and the large
approach has been carried out by considering an EV- scale IEEE New England 10-generator 39-bus systems
based objective function. The performance of the pro- operating with various loading conditions at several dis-
posed WOA has been tested on standard CEC14 and turbance conditions, and do it more effectively than the
CEC17 test functions. The results have shown that the other methods.
proposed WOA-PSS design technique is very effective,
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 16 of 17

Table 10 Eigenvalue analysis of three scenarios

PSO DE WOA

Scenario 1 − .366 ± 0.547i,0.557 − 0.385 ± 0.563i,0.565 − 0.401 ± 0.574i,0.572

− 0.410 ± 0.588i,0.572 − 0.436 ± 0.603i,0.586 − 0.457 ± 0.613i,0.598
− 0.408 ± 0.613i,0.554 − 0.437 ± 0.626i,0.571 − 0.459 ± 0.635i,0.586
− 0.563 ± 0.568i,0.670 − 0.739 ± 0.630i,0.585 − 0.563 ± 0.578i,0.698
− 0.616 ± 0.673i, 0.675 − 0.750 ± 0.653i,0.754 − 0.669 ± 0.663i,0.711
− 0.701 ± 0.640i,0.743 − 0.827 ± 0.670i,0.778 − 1.01 ± 0.623i,0.851
− 0.731 ± 0.663i, 0.741 − 18.2 ± 11.3i,0.850 − 2.29 ± 1.77i,0.792
Scenario 2 − 0.362 ± 0.547i, 0.551 − 0.108 ± 0.122i,0.663 − 0.108 ± 0.122i, 0.663
− 0.407 ± 0.613i, 0.553 − 0.380 ± 0.563i,0.560 − 0.396 ± 0.575i,0.567
− 0.492 ± 0.601i,0.633 − 0.434 ± 0.627i,0.574 − 0.458 ± 0.636i,0.585
− 0.514 ± 0.568i, 0.6714 − 0.520 ± 0.612i,0.648 − 0.544 ± 0.618i,0.661
− 0.673 ± 0.665, 0.711 − 0.542 ± 0.575i,0.686 − 0.564 ± 0.578i,0.698
− 0.686 ± 0.645i, 0.728 − 0.696 ± 0.659i,0.726 − 1.01 ± 0.619i,0.854
− 0.981 ± 0.0051i, 1.000 − 0.719 ± 0.631i,0.752 − 1.13 ± 1.81i,0.538
− 1.01 ± 0.0546i, 0.999 − 0.772 ± 0.649i,0.765 − 1.32 ± 3.30i,0.371
− 1.02 ± 0.0474i, 0.999 − 0.872 ± 0.657i,0.799 − 2.05 ± 2.87i,0.581
− 0.362 ± 0.547i, 0.551 − 0.108 ± 0.122i,0.663 − 0.108 ± 0.122i, 0.663
Scenario 3 − 0.674 ± 0.664i, 0.712 − 0.696 ± 0.659i,0.726 − 0.714 ± 0.653i,0.738
− 0.981 ± 0.0563i, 1.000 − 0.870 ± 0.657i,0.798 − 0.985 ± 0.0173i,1.000
− 0.746 ± 0.666i, 0.746 − 0.960 ± 0.664i,0.822 − 0.790 ± 0.635i,0.779
− 0.835 ± 0.692i, 0.770 − 1.83 ± 0.113i,0.998 − 0.887 ± 0.635i,0.814
− 0.895 ± 0.682i, 0.895 − 1.12 ± 1.83i,0.522 − 1.01 ± 0.619i,0.853
− 1.69 ± 0.150i, 0.996 − 2.36 ± 1.66i,0.818 − 2.06 ± 0.135i,0.998

Abbreviations Funding
WOA: Whale optimization algorithm; PSS: Power system stabilizer; MPSS: Modi- This work is carried out without support of any funding agency.
fied power system stabilizer; MSMIB: Modified single machine infinite bus;
MM: Multi-machine; SSS: Small signal stability; EV: Eigenvalue; PSO: Particle Availability of data and materials
swarm optimization; DE: Differential evolution; HP: Heffron Phillips; MHP: Please contact author for data and material request.
Modified Heffron Phillips; EVA: Eigenvalue analysis.
Declaration
Acknowledgements
Not applicable. Competing interests
The authors declare that they have no known competing financial interests
Authors’ contributions or personal relationships that could have appeared to influence the work
BD designed the study and formulated the objective function. BD and MS reported in this paper.
performed the simulations on test systems. MS and RS as supervisors helped
in pursing the work with constructive suggestions and edited the manuscript. Author details
1
All authors read and approved the final manuscript. Department of EEE, Gudlavalleru Engineering College, Gudlavlleru, Krishna
(Dt.), Andhra Pradesh 521301, India. 2 Department of EEE, University College
Authors’ information of Engineering, JNTUK, Kakinada, East Godavari (Dt.), Andhra Pradesh 533001,
B. Dasu: The author completed his M. Tech from JNTU Kakinada and pursuing India.
his PhD in JNTU Kakinada. His research interests include Control application to
Power Systems and Evolutionary Algorithms. Received: 15 July 2020 Accepted: 2 November 2021
Mangipudi Siva Kumar: The author completed his M.E and PhD from
Andhra University. He is presently working as Professor in EEE Department,
Gudlavalleru Engineering College, Gudlavalleru. He has contributed more than
40 technical papers in various referred journals and conference. He is a life
member of ISTE, member of IEEE and IAEng and Fellow of Institution of Engi- References
neers. His research interests include model order reduction, interval system 1. Demello, F. P., & Concordia, C. (1969). Concepts of synchronous machine
analysis, design of PI/PID controllers for Interval systems, sliding mode control stability as affected by excitation control. IEEE Transactions on Power
and Soft computing Techniques. Apparatus and Systems PAS, 88(4), 316–329.
R. Srinivasa Rao: The author completed his M.E from IISC Bangalore and PhD 2. Schleif, F., Hunkins, H., Martin, G., & Hattan, E. (1968). Excitation control
from JNTU Hyderabad. He is currently working as professor in EEE Department to improve powerline stability. IEEE Transactions on Power Apparatus and
of JNTU Kakinada. His research interests include Optimization Algorithms, Systems PAS, 87(6), 1426–1434.
State estimation, Modeling and control of Induction Generators. He has pub-
lished papers in IEEE transactions, Elsevier Publications.
Dasu et al. Protection and Control of Modern Power Systems (2021) 6:35 Page 17 of 17

3. Larsen, E. V., & Swann, D. A. (1981). Applying power system stabilizers multi-source power system considering high renewable energy penetra-
parts I, II and III: Practical considerations. IEEE Transactions on Power Appa- tion. Protection and Control of Modern Power Systems, 3(39), 1–15.
ratus and Systems PAS, 100(6), 3034–3046. 26. El-Zonkoly, M., Khalil, A. A., & Ahmied, N. M. (2009). Optimal tuning of
4. Kundur, P., Klein, M., Rogers, G. J., & Zywno, M. S. (1989). Application of lead-lag and fuzzy logic power system stabilizers using PSO. Expert
power system stabilizers for enhancement of overall system stability. IEEE Systems with Applications, 36(2), 2097–2106.
Transactions on Power Systems, 4(2), 614–626. 27. Butti, D., Mangipudi, S. K., & Rayapudi, S. R. (2019). Design of robust modi-
5. Padiyar, K. R. (1996). Power system dynamics stability and control. Interline fied power system stabilizer for dynamic stability improvement using
Publishing. Particle Swarm Optimization technique. Ain Shams Engineering Journal,
6. Gibbard, M. J. (1988). Co-ordinated design of multimachine power system 10(4), 769–783.
stabilizers based on damping torque concepts. IEE Proceedings C Genera- 28. Keumarsi, V., Simab, M., & Shahgholian, G. (2014). An integrated approach
tion, Transmission and Distribution, 135(4), 276. for optimal placement and tuning of power system stabilizer in multi-
7. Roger, G. (2000). Power system oscillations. Dordrecht: Kluwer. machine systems. International Journal of Electrical Power & Energy Systems,
8. Dysko, A., Leithead, W. E., & O’Reilly, J. (2010). Enhanced power system 63, 132–137.
stability by coordinated PSS design. IEEE Transmission Power Systems, 29. Eslami, M., Shareef, H., Mohamed, A., & Khajehzadeh, M. (2012). An
25(1), 1200. efficient PSO technique with chaotic sequence for optimal tuning and
9. Chow, J. H., & Sanchez-Gasca, J. J. (1989). Pole-placement designs of placement of PSS in power systems. International Journal of Electrical
power system stabilizers. IEEE Transmission Power Systems, 4(1), 271–277. Power & Energy Systems, 43(1), 1467–1478.
10. Kashki, M., Abido, M. A., & Abdel-Magid, Y. L. (2010). Pole placement 30. Wang, Z., Chung, C. Y., Wong, K. P., & Tse, C. T. (2008). Robust power system
approach for robust optimum design of PSS and TCSC-based stabiliz- stabiliser design under multi-operating conditions using differential
ers using reinforcement learning automata. Electrical Engineering, 91(7), evolution. IET Generation, Transmission and Distribution, 2(5), 690.
383–394. 31. Panda, S. (2009). Differential evolutionary algorithm for TCSC-based con-
11. Gomes, S., Guimarães, C. H. C., Martins, N., & Taranto, G. N. (2018). Damped troller design. Simulation Modelling Practice and Theory, 17(10), 1618–1634.
Nyquist Plot for a pole placement design of power system stabilizers. 32. Wang, S.-K., Chiou, J.-P., & Liu, C.-W. (2009). Parameters tuning of power
Electric Power Systems Research, 158, 158–169. system stabilizers using improved ant direction hybrid differential evolu-
12. Kothari, M. L., Nanda, J., & Bhattacharya, K. (1993). Design of variable tion. International Journal of Electrical Power & Energy Systems, 31(1), 34–42.
structure power system stabilizers with desired Eigenvalues in the sliding 33. Wang, S. K. (2016). Coordinated parameter design of power system
mode. IEE Proceedings C Generation, Transmission and Distribution, 140(4), stabilizers and static synchronous compensator using gradual hybrid
263–268. differential evaluation. International Journal of Electrical Power & Energy
13. Bandal, V., & Bandyopadhyay, B. (2007). Robust decentralised output Systems, 81, 165–174.
feedback sliding mode control technique-based power system stabiliser 34. Chatterjee, A., Ghoshal, S. P., & Mukherjee, V. (2011). Chaotic ant swarm
(PSS) for multimachine power system. IET Control Theory and Applications, optimization for fuzzy-based tuning of power system stabilizer. Interna-
1(5), 1512–1522. tional Journal of Electrical Power & Energy Systems, 33(3), 657–672.
14. Dash, P. K., Patnaik, R. K., & Mishra, S. P. (2018). Adaptive fractional integral 35. Sambariya, D. K., Prasad, R., & Birla, D. (2015). Design and performance
terminal sliding mode power control of UPFC in DFIG wind farm pen- analysis of PID based controller for SMIB power system using Firefly
etrated multimachine power system. Protection and Control of Modern algorithm. In 2015 2nd international conference on recent advances in
Power Systems, 3(8), 1–14. engineering & computational sciences (RAECS) (pp. 1–8).
15. Aldeen, M. (1995). Multimachine power system stabiliser design based 36. Sambariya, D. K., & Prasad, R. (2015). Optimal tuning of fuzzy logic power
on new LQR approach. IEE Proceedings C Generation, Transmission and system stabilizer using harmony search algorithm. International Journal of
Distribution, 142(5), 494. Fuzzy Systems, 17(3), 457–470.
16. Ko, H. S., Lee, K. Y., & Kim, H. C. (2004). An intelligent based LQR controller 37. Abd-Elazim, S. M., & Ali, E. S. (2015). A hybrid PSO and bacterial foraging
design to power system stabilization. Electric Power Systems Research, for power system stability enhancement. Complexity, 21(2), 245–255.
71(1), 1–9. 38. Chitara, D., Niazi, K. R., Swarnkar, A., & Gupta, N. (2018). "Cuckoo search
17. Yang, T. (1997). Applying H − ∞ optimisation method to power system optimization algorithm for designing of a multi-machine power system
stabiliser design, part 1: single-machine infinite-bus systems. International stabilizer. IEEE Transactions on Industry Applications, 54(4), 3056–3065.
Journal of Electrical Power & Energy Systems, 19(1), 29–35. 39. Ali, E. S. (2014). Optimization of power system stabilizers using BAT search
18. Hardiansyah, S. F., & Irisawa, J. (2006). A robust H∞ power system stabi- algorithm. International Journal of Electrical Power & Energy Systems, 61,
lizer design using reduced-order models. International Journal of Electrical 683–690.
Power & Energy Systems, 28(1), 21–28. 40. Chaib, L., Choucha, A., & Arif, S. (2017). Optimal design and tuning
19. Hoang, P., & Tomsovic, K. (1996). Design and analysis of an adaptive fuzzy of novel fractional order PID power system stabilizer using a new
power system stabilizer. IEEE Transactions on Energy Conversion, 11(2), metaheuristic Bat algorithm. Ain Shams Engineering Journal, 8(2), 113–125.
455–461. 41. Khadanga, R. K., & Satapathy, J. K. (2015). Time delay approach for PSS
20. Ghadimi, N. (2015). A new hybrid algorithm based on optimal fuzzy and SSSC based coordinated controller design using hybrid PSO–GSA
controller in multimachine power system. Complexity, 21(1), 78–93. algorithm. International Journal of Electrical Power & Energy Systems, 71,
21. Abido, M. A. (1999). A novel approach to conventional power system 262–273.
stabilizer design using tabu search. International Journal of Electrical Power 42. Shakarami, M. R., & Faraji Davoudkhani, I. (2016). Wide-area power system
& Energy Systems, 21(6), 443–454. stabilizer design based on Grey Wolf Optimization algorithm considering
22. Abido, M. A., & Abdel-Magid, Y. L. (2000). Robust design of multimachine the time delay. Electric Power Systems Research, 133, 149–159.
power system stabilisers using tabu search algorithm. IEE Proceedings C 43. Mirjalili, S., & Lewis, A. (2016). The whale optimization algorithm. Advances
Generation, Transmission and Distribution, 147(6), 387. in Engineering Software, 95, 51–67.
23. Abdel-Magid, Y. L., Abido, M. A., Al-Baiyat, S., & Mantawy, A. H. (1999). 44. Dasu, B., Sivakumar, M., & Srinivasarao, R. (2019). Interconnected multi-
Simultaneous stabilization of multimachine power systems via genetic machine power system stabilizer design using whale optimization
algorithms. IEEE Transactions on Power Systems, 14(4), 1428–1439. algorithm. Protection and Control of Modern Power Systems, 4(2), 1–11.
24. Abdel-Magid, Y. L., & Abido, M. A. (2003). Optimal multiobjective design 45. Gurrala, G., & Sen, I. (2008). A modified Heffron-Phillip’s model for the
of robust power system stabilizers using genetic algorithms. IEEE Transac- design of power system stabilizers. In Joint international conference on
tions on Power Systems, 18(3), 1125–1132. power system technology and IEEE power India (pp. 1–6).
25. Magdy, G., Shabib, G., Elbaset, A. A., et al. (2018). Optimized coordi- 46. Pai, M. A. (1989). Energy function analysis for power system stability. Boston,
nated control of LFC and SMES to enhance frequency stability of a real MA: Springer, US.