HARAMAYA UNIVERSITY

HARAMAYA INSTITUTE OF TECHNOLOGY

POST GRADUATE PROGRAM DIRECCTORATE
SCHOOL OF ELECTRICAL AND COMPUTER ENGINEERING
POWER ENGINEERING

GRADUATE SEMINAR

ADAPTIVE NEURO-FUZZY BASED POWER SYSTEM STABILIZERS

FOR DAMPING OSCILLATIONS IN MULTI-MACHINE POWER
SYSTEMS

Submitted by: Kenan yeshitela

Id. No :( PGP/888/15)

Submitted to: Dr.Ramesh Kumar

February 16, 2024

Haramaya, Ethiopia
 i

ACCRONYMS

PSS POWER SYSTEM STABILIZER

CPSS CONVENTIONAL POWER SYSTEM STABILIZER
FLPSS FUZZY LOGIC BASED POWER SYSTEM STABILIZER
ANFPSS ADAPTIVE NEURO FUZZY POWER SYSTEM STABILIZER
ANFIS ADAPTIVE NEURO FUZZY INFERENCE SYSTEM
 ii

Table of Contents
ACCRONYMS .................................................................................................................................................. i
LIST OF FIGURE ............................................................................................................................................. iii
ABSTRACT..................................................................................................................................................... iv
1. INTRODUCTION ......................................................................................................................................... 1
1.1 Background of Study ........................................................................................................................... 1
1.1.1 Classification of stability ....................................................................................................... 2
1.2 Statement of problem................................................................................................................... 2
1.3 Scope the study ............................................................................................................................. 4
1.4 Significant of the study ....................................................................................................................... 4
1.5 Objectives of the Study ....................................................................................................................... 4
2. LITERATURE REVIEW ................................................................................................................................. 6
3. METHODS ................................................................................................................................................ 11
4. CONCLUSION ........................................................................................................................................... 14
REFERENCE .................................................................................................................................................. 15
 iii

LIST OF FIGURE

Figure 1. Structure of Sugeno type ANFIS for PSS ................................................................................... 12

Figure 2 Procedure of the research. ............................................................................................................ 13
 iv

ABSTRACT

Low-frequency disturbances in power systems can lead to synchronization issues and system
failure. These disturbances, known as "low-frequency oscillations," have operational limits that
affect bulk power transmission in power networks and can cause the system to fail. For this
particular problem, power system stabilizers (PSSs) are utilized to produce supplementary control
signals for the excitation system. The purpose is to dampen the low-frequency power system
oscillations and provide additional damping for the synchronous generators. The supplementary
power system stabilizer must be capable of providing appropriate stabilization signals across a
wide range of operating conditions, as well as disturbances. However, a conventional power
system stabilizer (CPSS) offers positive damping torque in phase with the speed signal to
counteract the effect of the system's negative damping torque. This is because the gains of the
controller are determined for specific operating conditions. The conventional Power System
Stabilizer, which implements lead-lag compensation, suffers from poor performance under
different loading conditions. The constantly changing nature of the power system makes the design
of CPSS a challenging task. Thus, designing a stabilizer that can deliver good performance in all
operating points of electric power systems becomes quite difficult. This seminar aims to address
the drawback of conventional power system stabilizers (CPSSs) by proposing the design and
modeling of Adaptive Neuro-Fuzzy power system stabilizers (ANFPSSs). The new approach
effectively reduces oscillations caused by various disturbances on generators. The focus of this
thesis is centered around the design of a fuzzy logic-based PSS (FLPSS) and a self-learning
adaptive neural network-based power system stabilizer (ANFPSS). These aim to enhance dynamic
stability and provide additional signals, thereby extending power stability limits. The suggested
controller utilizes synchronous machine speed variation and acceleration as input signals.

The proposed technique for power system stabilization is simple, fast, effective, and affordable.
The simulation findings indicate that the ANFIS controller-based stabilizer out performs
conventional power system stabilizers in terms of inter-area power oscillation damping at low
frequencies. The ANFIS controller-based power system stabilizer provides 73% less damping time
than CPSS during three-phase fault circumstances. With these advancements, power systems can
become more resilient and reliable in the face of disturbances.
 1

1. INTRODUCTION
1.1 Background of Study

Since the 1920s, experts in electrical engineering have raised concerns about the stability of power
systems. As power systems become more advanced, their ability to recover from errors becomes
increasingly vital. Modern power systems are susceptible to significant disturbances that can
spread across interconnected networks without adequate safeguards. The primary issue arises from
the growing demand for power. In countries like Ethiopia, where a significant portion of the
population resides in villages and suburbs, providing power to these areas takes precedence. As
transmission lines become longer and carry greater power, the likelihood of power fluctuations
and faults increases. This situation can potentially lead to a complete failure of power grids [1].
The number of power transactions is increasing each day in the restructured power systems. These
systems are expected to operate at various operating points and closer to their operational limits.
One of the constraints that limit the transmission of bulk power through the network is known as
"low-frequency oscillations." In such cases, power system controls play a crucial role. These
controls can either contribute positively or negatively to damping the oscillations. Specifically, the
regulation of generator voltage, and in particular, high gain in this regulation, can lead to poor or
negative damping of the oscillation. To address this issue, Power System Stabilizers (PSS) have
been implemented to dampen the oscillations. For reliable power generation and transmission, it
is essential to maintain stability in the power system. Most modern power generators are equipped
with PSS to control slowly oscillating instability. These stabilizers produce control signals to
decrease low-frequency power system oscillations. Various methods, such as fuzzy logic, neural
networks, and PID controllers, have been proposed to overcome the limitations of common power
system stabilizers. This thesis focuses on the application of a Neuro-fuzzy controller to improve
system control based on simulations. It is found that the Neuro-fuzzy controller provides better
results compared to the traditional Fuzzy Logic Controller (FLC). FLC is chosen as a controller
for this thesis due to its advantages, such as control simplicity, low cost, and the ability to design
without an exact mathematical model of the process. The combination of neural networks and
fuzzy control in the Neuro-fuzzy controller enhances its performance [2].
 2

1.1.1 Classification of stability

Power system stability refers to the characteristic of a power system that allows it to maintain its
operational equilibrium under normal conditions and to restore a satisfactory state of equilibrium
after experiencing a disturbance [1]. There are three categories of power system stability.

Dynamic or Small Signal Stability

Dynamic stability refers to the response of the synchronous machine when faced with small,
oscillating perturbations. If these oscillations have a small amplitude, the system can be considered
stable. However, if the oscillations grow over time, the system may lose its stability. Typically,
heavy power flow in transmission lines or the interaction of the controller with the system
frequency can lead to small signal instabilities [3].

Steady State Stability

Steady-state stability analysis involves investigating a power system and its generators under
strictly steady-state conditions. The objective is to determine the maximum generator load that can
be transmitted without any loss of synchronism in any of the generators.

Transient Stability

Transient stability refers to the response of a synchronous machine to significant disturbances,

such as the application and clearing of faults, sudden load changes, or modifications in
transmission lines or generators. The accompanying figure provides an overview of the stability
issues. This thesis focuses on the aspect of small disturbance stability, which is a component of
rotor angle stability [1].

1.2 Statement of problem

The need to operate power systems close to their capacity limits has arisen due to the increasing
magnitude and complexity of interconnected power systems caused by competitive energy
markets, economy, and population development. However, this can lead to stability issues and
poor dynamic behaviors such as power oscillations. These oscillations have detrimental effects,
 3

including a reduction in the lifetime of system components, expensive operations of electrical

grids, and in the worst-case scenario, the risk of partial system collapse. Additionally, the action
of the excitation control system weakens the damping provided by the field and damper windings
in synchronous generators. This is because additional currents induced by voltage regulation
appear in the rotor circuits, opposing the currents induced by rotor speed deviations.

Currently, the installation of automatic voltage regulators (AVRs) in generators is common

practice in power systems. Many generators are equipped with high-gain and fast-acting AVRs to
enhance large-scale stability and ensure the generator remains in synchronism with the power
system during significant transient fault conditions. However, the high gain in generator voltage
regulation can result in poor or negative damping of the oscillations. This study addresses the use
of power system stabilizers (PSS) in synchronous generators to counteract the effect of high AVR
gain.

Power system stabilizers (PSS) should be capable of providing suitable stabilization signals across
a wide range of operating conditions and disturbances. However, traditional PSSs provide positive
damping torque in phase with the speed signal to counteract the effect of the system's negative
damping torque. Since the gains of this controller are determined for a specific operating condition,
they may not be valid for a wide range of operating conditions. This leads to high starting
overshoot, sensitivity to controller gains, and sluggish response to sudden disturbances and loading
conditions within a short time. Designing a conventional PSS for constantly changing power
systems is a challenging task. The nonlinear nature of power systems, along with random
disturbances like load changes and time-varying operating conditions, makes precise real-time
modeling of large power systems difficult. This study attempts to overcome these challenges by
using a hybrid of fuzzy logic and neural network control, which is a powerful tool for controlling
systems that are not precisely characterized.

The proposed Adaptive Neuro-Fuzzy Power System Stabilizer (ANFPSS) addresses the
limitations of conventional power system stabilizers. The new design effectively reduces
oscillations caused by various disturbances in generators. The ANFPSS dampens oscillations in
the excitation system, ensuring reliable and sustainable grid operations despite small and large
disturbances.
 4

1.3 Scope the study

The Adaptive Neuro-fuzzy power system stabilizer is restricted to examining a Single Machine
Infinite Bus (SMIB) model and conducting simulation trials in multi-machine systems.
Furthermore, the evaluation of the PSS effectiveness and the implementation of alternative control
structures are presented as a theoretical assessment of the findings backed by simulations.

1.4 Significant of the study

In my master thesis, I will develop a methodology for the adaptive neuro fuzzy power system that
is utilized in synchronous generators. I will also explore the use of a control structure that is
different from the one used in the generators for damping electro motion oscillations

 Enhance power flow and improve system reliability.

 Decrease the number of machines needed for both peak load operation and as spinning
reserves for sudden load changes.
 Offer a cost-effective power source for consumers.
 Electrical devices are susceptible to failure when faced with power stability issues.
 Improve current and future power generation and transmission systems.

1.5 Objectives of the Study

1.5.1 General objective

The main aim of my thesis is to create and analyze Adaptive Neuro-Fuzzy based power system
stabilizers to dampen oscillations using the MATLAB package with the control system toolbox.
The goal is to enhance the small signal stability of power systems.

1.5.2 Specific Objectives

The specific goals include:
 Analyzing power system stability, excitation systems, automatic voltage regulators
for synchronous generators, and power system stabilizers.
 5

 Reviewing various approaches to power system stabilizer design for potential

adoption in this thesis.
 Developing the Neuro-fuzzy Power System Stabilizer to swiftly stabilize the
system during transmission line faults and validating its effectiveness.
 Decreasing settling time, rise time, peak amplitude, and overshoot of the system.
 Operating the power system over a wide range.
 Enhancing the stability and reliability of synchronous generators.
 Using simulation to validate the Adaptive Neuro-fuzzy based power system
stabilizer and comparing its performance with conventional stabilizers.
 6

2. LITERATURE REVIEW

In this section, we will examine previous works related to the damping of low-frequency
oscillations in power systems to enhance stability. The reviewed works are categorized into
recently published journal papers, M.Sc. theses, and Ph.D. theses. While this work draws on a
variety of related materials, only selected works from each category will be reviewed due to space
constraints.

Yagami and Tamura (2009) [4] present a method for enhancing power system stability by
combining a fault current limiter and a thyristor-controlled braking resistor. The fault current
limiter is used to restrict fault currents, improve power system stability, and dampen turbine shaft
torsional oscillations. On the other hand, the thyristor-controlled braking resistor aims to quickly
manage generator disturbances. The effectiveness of both devices is demonstrated using a Three-
lines-to-ground (3LG) fault in a two-machine infinite bus system. Simulation results show
improved power system stability and damping of turbine shaft torsional oscillations within
acceptable temperature rise limits.
Li Zhengguo et al (2007) [5] present the concept of a switched controller to analyze a single
machine infinite bus power system when a symmetrical 3-phase short circuit fault occurs in one
of the transmission lines. Typically, a linear controller is unable to deliver satisfactory transient
performance for such a power system with a significant fault. The proposed switched controller
addresses both temporary and permanent faults. The effectiveness and efficiency of this approach
are demonstrated through simulation results. It is important to note that this method offers a
solution to the stabilization issue during fault occurrences. A future extension of this approach
involves exploring ways to achieve optimal post-fault performance.
Soon Kiat Yee and Milano Vic (2008) [6] proposed a Fuzzy Logic Controller (FLC) for the
decentralized stabilization of multi-machine power systems. The authors introduced a unique,
predominantly analytical technique for designing a robust Multi-Input-Single-Output (MISO) FLC
to improve the damping and stability of an electrical power system without affecting voltage
regulation. The suggested decentralized FLC employs a systematic analytical approach based on
a performance index to eliminate the need for prior knowledge about the system. The FLC tracks
speed deviations to zero to stabilize the generator's power output, while simultaneously controlling
and stabilizing the generator's terminal voltage.
 7

A. Dysko, W.E. Leithead, and J. O’Reilly (2010) [7] have outlined a systematic coordinated
design process for power system stabilizers (PSSs) and automatic voltage regulators (AVRs) in a
strongly interconnected system. The proposed coordinated PSS/AVR design process is established
within a frequency domain framework.

P. De Mello (1969) [8] explored the stability of synchronous machines under minor disturbances
by investigating a scenario of a single machine connected to an infinite bus through an external
reactance. The design of PSS for a single machine linked to an infinite bus was detailed using fast
output sampling feedback.

J.H. Chow and J.J. Sanchez-Gasca (1989) [9] studied the four pole-placement approach for
designing power system stabilizers, focusing on the frequency characteristics of these controllers.
However, their designs were not globally optimized.

The fundamental PSS design concept in this paper is rooted in the stabilizer proposed by De Mello
and Concordia (1969) [8]. Nonetheless, the phase characteristics were derived using a multi-
machine Eigen value program instead of a single machine model. This study emphasized
enhancing overall system stability, addressing simultaneous damping of inter-area and local
modes, and evaluating PSS performance under various system conditions. The authors asserted
that the frequency response method utilized to compensate for the delay between excitation input
and electrical torque was quite robust.
Hiyama et al (1994) introduced a PID (Proportional + Integral + Derivative) type fuzzy logic
power system stabilizer [10]. The integration of the speed deviation was utilized as one input, and
the phase plane origin was adjusted leftward or rightward based on the integral sign. Both
simulations and experiments were conducted to show case the effectiveness of this modification.
Hiyama's heuristic-based approach demonstrated some success in his series of research studies.
However, in his work, the fuzzy PSS parameters were not globally optimized as he argued that
these parameters are relatively insensitive to external conditions.
Son and Park (2000)[6] applied the Linear Quadratic Gaussian technique to design a TCSC
damping controller for a 3-machine 9-bus system. They utilized the optimal Hankel norm
 8

approximation technique to obtain a reduced-order power system model and designed a controller
based on this model. They also discussed using the Loop Transfer Recovery technique to maintain
the robustness of the designed controller. However, due to matrix size limitations, these methods
were not extended to higher-order systems.

Del Rosso et al. (2003)[11] proposed a hierarchical control strategy for enhancing dynamic and
steady state stability, including strategies to mitigate adverse interactions among TCSC
hierarchical controls. They analyzed and compared different locally measurable input signals
qualitatively using the equal area criterion. The focus was on comparing active power and line
current as input signals, without exploring the potential use of bus voltage and bus frequency as
input signals for the damping controller.
M. F. Othman, M. Mahfouf, and D.A. Linkens[12] described the design process for a fuzzy
logic-based power system stabilizer (FLPSS) and an Adaptive Neuro-fuzzy Inference System
(ANFIS), investigating their robustness for a single-machine power system. They selected speed
deviation and its derivative as input signals for the FLPSS.

Vani, M.U, Raju, G.S., and Prasad, K.R.L.[13] Outlined a systematic design methodology for
an Adaptive Neuro-Fuzzy Inference System and optimization-based automatic voltage regulator
and power system stabilizer.

Chun-Jung Chen[14] introduced an adaptive power system stabilizer (PSS) comprising a

recurrent neural network controller (RNNC) and a compensator to dampen power system
oscillations. The RNNC provides an adaptive control signal to the exciter or governor to damp
most oscillations.

Sumina D. (2008)[15] demonstrated the use of a neural network-based excitation control on a

single machine infinite bus. The proposed feedforward neural network integrates a voltage
regulator and a power system stabilizer.

Barton (2004) presented a robust artificially intelligent Adaptive Neuro-Fuzzy Inference System
(ANFIS) based PSS design for damping electromechanical oscillation modes and enhancing power
 9

system synchronous stability. The actual power system was divided into subsystems, each
containing one machine. Each subsystem was associated with a local ANFPSS, with input signals
including speed, power angle, and real power output. Nonlinear simulations illustrated the
robustness of the ANFPSS.
Hsu and Chen (1991)[14] utilized a neural network to adjust the parameters of a conventional PI
(Proportional + Integral) type PSS. Abdel-Magid et al. (2000)[16] aimed to identify a single set
of PSS parameters that could stabilize the power system across a wide range of operating
conditions simultaneously.
All the papers mentioned above, proposed by different researchers, are effective for specific
operating conditions but may not be suitable for a wide range of operating conditions due to issues
like high initial overshoot, sensitivity to controller gains, and slow response during sudden
disturbances and heavy loading conditions. This thesis explores the implementation of Neuro-
fuzzy techniques, which combine Artificial Neural Networks (ANN) and Fuzzy Inference Systems
(FIS) to address real-world problems. A Neuro-fuzzy system is essentially a fuzzy system trained
using a learning algorithm derived from neural network theory. The learning capability is
advantageous from the FIS perspective, while the formation of linguistic rule base is beneficial
from the ANN perspective. Various methods exist to integrate ANN and FIS, with the choice often
depending on the specific application.
The proposed Neuro-fuzzy controller processes speed deviation and acceleration error signals
through the fuzzy inference system to produce an output suitable for damping oscillations.
Simulation results demonstrate the superior performance of the ANFIS system compared to
traditional approaches. Multiple model adaptive controllers have been developed for Single
Machine Infinite Bus (SMIB) systems and Multi-machine systems as intelligent control solutions.
The performance evaluation of these controllers is carried out using Fuzzy Logic control and
Neural Network control methods.
In the field of Power System Stabilizers (PSS), new trends have emerged, leading to a surge in
research papers. Among these, Kothari et al. [17] developed a variable structure power system
stabilizer with desired eigenvalues in the sliding mode. Hariri and Malik [18] combined fuzzy
control with the learning capabilities of neural networks to create a PSS, which could potentially
trap the system in local minima. Abido and Abdel Magid [16] utilized an evolutionary
programming algorithm to determine the optimal values for a classical lead-lag PSS. Rashidi et
 10

al. [17] proposed adapting the gain of the discontinuous component of the control signal in a
sliding mode controller using a fuzzy inference system augmented by linear state feedback applied
to a sliding surface with an integral term. Elshafei et al. [19] suggested power system stabilization
using fuzzy logic and direct adaptive techniques. Hossein-Zadeh and Kalam [20] developed an
indirect adaptive indirect fuzzy approach. Elshafei et al. [19] expanded the direct adaptive fuzzy
method to include stabilization of multi-machine power systems.
 11

3. METHODS

The research procedure is illustrated in Fig. 2. MATLAB software package is selected as the
simulation environment for this study. It serves as the primary engineering tool for modeling and
simulating multi-machine power systems, as well as for interacting with the user and relevant
simulation programs. MATLAB was chosen for its robust set of programming tools, signal
processing capabilities, numerical functions, and user-friendly interface. Within this custom
simulation environment, evaluation procedures can be easily carried out. The Fuzzy Logic
Toolbox of MATLAB was utilized to develop the ANFIS model with 4 inputs and a single output.

Adaptive Neuro-Fuzzy PSS

The design process of the Adaptive Neuro-Fuzzy (ANFIS) for PSS go through the following steps:

1. Generating suitable training data: To apply the ANFIS technique for power system stability
using Power System Stabilizers (PSS), it is crucial to precisely determine the input parameter
limits. These parameters are typically collected from recording devices sparsely positioned at the
sending end in a power system network. Since practical fault data for transmission lines is limited,
generating training/testing data through simulations becomes necessary. A computer program has
been developed to create training data for various faults in a typical transmission system.

2. Selecting an appropriate ANFIS structure for a specific application: Different ANFIS structures
are tailored for PSS to expand stability boundaries by adjusting generator excitation to provide
damping to synchronous machine rotor oscillations relative to each other. The structure of a
Sugeno-type ANFIS for PSS is depicted in Figure 1.
 12

Figure 1. Structure of Sugeno type ANFIS for PSS

3. ANFIS Training: Different network configurations are trained to identify a suitable network
with satisfactory performance levels. The ANFIS models are trained to detect the presence of
faults, classify faults, and achieve system stability.

4. Evaluation of Trained ANFIS: The trained ANFIS models are evaluated using test patterns until
their performance meets the desired criteria. Once the network is trained, the ANFIS models
should provide accurate outputs for unseen data. When the output of test patterns and the network's
error fall within an acceptable range, the fuzzy system is adjusted optimally. This adjustment
ensures that the membership functions and fuzzy rules are well tuned. All these steps are conducted
offline, and once the structure and parameters of the ANFIS are adjusted, it can be utilized online
as a Power System Stabilizer (PSS).
 13

Figure 2 Procedure of the research.

 14

4. CONCLUSION

The real-time response of the non-linear system considered in this study indicates that the ANFIS-
based controller delivers superior performance compared to the conventional controller. The
ANFIS-based Power System Stabilizer (ANFLPSS) demonstrates better control performance in
terms of settling time and damping effect than the traditional power system stabilizer. In
conclusion, this thesis suggests that the Adaptive Neuro-Fuzzy Inference System (ANFIS)
controller can be implemented for small signal stability in power systems to achieve quicker
control action. The ANFIS-based tuning of the PSS is specifically developed for a Single Machine
Infinite Bus (SMIB) system, aiming to address local mode oscillations.
 15

REFERENCE

[1] Power-System-Stability-and-Control-by-Prabha-Kundur. 1994.

[2] A. B. Muljono, I. M. Ginarsa, I. M. A. Nrartha, and A. Dharma, “Coordination of adaptive
neuro fuzzy inference system (ANFIS) and type-2 fuzzy logic system-power system
stabilizer (T2FLS-PSS) to improve a large-scale power system stability,” Int. J. Electr.
Comput. Eng., vol. 8, no. 1, pp. 76–86, 2018, doi: 10.11591/ijece.v8i1.pp76-86.
[3] . S. S., “Efficient Power Oscillation Damping Control By Adaptive Neuro Fuzzy
Controller Based Power System Stabilizer,” Int. J. Res. Eng. Technol., vol. 05, no. 09, pp.
101–105, 2016, doi: 10.15623/ijret.2016.0509015.
[4] Energy Conversion Congress and Exposition, 2009. ECCE, IEEE. IEEE, 2009.
[5] S. Xie, X. Wang, C. Qu, X. Wang, and J. Guo, “Impacts of different wind speed
simulation methods on conditional reliability indices,” Int. Trans. Electr. energy Syst., vol.
20, no. August 2008, pp. 1–6, 2013, doi: 10.1002/etep.
[6] S. K. Yee and J. V. Milanović, “Fuzzy logic controller for decentralized stabilization of
multimachine power systems,” IEEE Trans. Fuzzy Syst., vol. 16, no. 4, pp. 971–981, 2008,
doi: 10.1109/TFUZZ.2008.917296.
[7] A. Dyśko, W. E. Leithead, and J. O’Reilly, “Enhanced power system stability by
coordinated PSS design,” IEEE Trans. Power Syst., vol. 25, no. 1, pp. 413–422, Feb.
2010, doi: 10.1109/TPWRS.2009.2036704.
[8] F. P. Demello and C. Concordia, “Concepts of Synchronous Machine Stability as Affected
by Excitation Control,” IEEE Trans. Power Appar. Syst., vol. PAS-88, no. 4, pp. 316–329,
1969, doi: 10.1109/TPAS.1969.292452.
[9] A. Jamal and R. Syahputra, “Design of Power System Stabilizer Based on Adaptive
Neuro-Fuzzy Method,” Int. Semin. Appl. Technol. Sci. Arts, pp. 14–21, 2011.
[10] P. Ghareeb, A. G. Asim, and M. Ahmed, “" The Design of a Fuzzy Logic Based Power
System Stabilizer Applied to Two Electric Power Generation Units " Researchers :
Abstract :,” pp. 41–68, 2020.
[11] A. D. Del Rosso, C. A. Cañizares, and V. M. Doña, “A Study of TCSC Controller Design
for Power System Stability Improvement,” IEEE Trans. Power Syst., vol. 18, no. 4, pp.
1487–1496, 2003, doi: 10.1109/TPWRS.2003.818703.
[12] M. F. Othman, M. Mahfouf, and D. A. Linkens, “Designing Power System Stabilizer for
Multimachine Power System Using Neuro-Fuzzy Algorithm,” J. Teknol., vol. 35, no. D,
pp. 55–63, 2012, doi: 10.11113/jt.v35.610.
[13] M. Uma Vani, G. S. Raju, and K. R. L. Prasad, “Robust supplementary controllers for
AVR and PSS,” Proc. INDICON 2009 - An IEEE India Counc. Conf., pp. 9–12, 2009, doi:
10.1109/INDCON.2009.5409474.
[14] C. J. Chen and T. C. Chen, “Design of a power system stabilizer using a new recurrent
 16

network,” Int. J. Innov. Comput. Inf. Control, vol. 3, no. 4, pp. 907–918, 2007.
[15] D. Sumina, N. Bulic, and G. Erceg, “Simulation model of neural network based
synchronous generator excitation control,” 2008 13th Int. Power Electron. Motion Control
Conf. EPE-PEMC 2008, no. 4, pp. 556–560, 2008, doi:
10.1109/EPEPEMC.2008.4635324.
[16] M. A. Abido and Y. L. Abdel-Magid, “Optimal design of power system stabilizers using
evolutionary programming,” IEEE Trans. Energy Convers., vol. 17, no. 4, pp. 429–436,
2002, doi: 10.1109/TEC.2002.805179.
[17] F. Rashidi, M. Rashidi, and H. Amiri, “An Adaptive Fuzzy Sliding Mode Control for
Power System Stabilizer,” IECON Proc. (Industrial Electron. Conf., vol. 1, pp. 626–630,
2003, doi: 10.1109/IECON.2003.1280053.
[18] P. Shamsollahi and O. P. Malik, “An adaptive power system stabilizer using on-line
trained neural networks,” IEEE Trans. Energy Convers., vol. 12, no. 4, pp. 382–387,
1997, doi: 10.1109/60.638951.
[19] A. L. Elshafei, K. A. El-Metwally, and A. A. Shaltout, “A variable-structure adaptive
fuzzy-logic stabilizer for single and multi-machine power systems,” Control Eng. Pract.,
vol. 13, no. 4, pp. 413–423, 2005, doi: 10.1016/j.conengprac.2004.03.017.
[20] N. Hossein-Zadeh and A. Kalam, “An indirect adaptive fuzzy-logic power system
stabiliser,” Int. J. Electr. Power Energy Syst., vol. 24, no. 10, pp. 837–842, 2002, doi:
10.1016/S0142-0615(01)00093-X.