HARAMAYA UNIVERSITY

POSTGRADUATE PROGRAM DIRECTORATE

Adaptive Neuro-Fuzzy-Based Power System Stabilizers for Low-Frequency

Damping Oscillations in Multi-Machine Power Systems

M.Sc. Thesis Proposal

Kenan Yeshitela

College: Haramaya Institute of Technology

School/Department: School of Electrical and Computer Engineering
Program: M.Sc. In Electrical Power Engineering
Major Advisor: Dr. Mohammad Firoz Alam Khan

March 2024

Haramaya University, Haramaya

 ACRONYMS AND ABBREVIATIONS

PSS POWER SYSTEM STABILIZER

CPSS CONVENTIONAL POWER SYSTEM STABILIZER

FLPSS FUZZY LOGIC POWER SYSTEM STABILIZER

ANFPSS ADAPTIVE NEURO-FUZZY POWER SYSTEM STABILIZER

ANFIS ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM

SMIB SINGLE MACHINE INFINITE BUS

AC ALTERNATE CURRENT
ANN ARTIFICIAL NEURAL NETWORK
AVR AUTOMATIC VOLTAGE REGULATOR
DC DIRECT CURRENT

i
 Table of Contents
ACRONYMS AND ABBREVIATIONS I

LIST OF TABLES III

LIST OF FIGURES IV

1. INTRODUCTION 1

1.1. BACKGROUND OF STUDY 1

1.2. STATEMENT OF PROBLEM 4
1.3. SCOPE OF THE STUDY 5
1.4. SIGNIFICANT OF STUDY 5
1.5. OBJECTIVES OF THE STUDY 6
1.5.1. General Objective 6
1.5.2. Specific Objectives 6

2. LITERATURE REVIEW 7
2.1. REVIEW OF RELATED WORK 7
2.2. NEURO-FUZZY CONTROLLER 11

3. MATERIALS AND METHODS 12

3.1. METHODS 12
3.2. MATERIALS 15
3.2.1. Power System Stabilizer Model 15

4. WORK PLAN 19
5. RESEARCH BUDGET BREAKDOWN 20
5.1. PERSONAL COST 20
5.2. TRANSPORT EXPENSE 20
5.3. STATIONARY EXPENSE 21
5.4. BUDGET SUMMARY 21

6. REFERENCE 22
APPROVAL SHEET 24

ii
 LIST OF TABLES
Table Page
4.1. Work Plan 19
5.1. Per Diem 20
5.2 Supervision Fee 20
5.3 Transport Expense 20
5.4 Stationary Expense 21
5.5. Budget Summary 21

iii
 LIST OF FIGURES

Figure Page

1.1 Classification of Power System Stability 3

3.1. Structure of Sugeno type ANFIS for PSS 13
3.2 Procedure of the research. 14
3.3. Haffron- Phillips model of excitation system with PSS 16
3.4. General Structure of PSS 17

iv
 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 power demand. In countries like Ethiopia, where most of the population
resides in villages and suburbs, power supply to these places takes priority. As transmission
lines become large and carry greater power, the likelihood of power fluctuations and faults
increases. This situation can potentially lead to a complete failure of power grids (“Power-
System-Stability-and-Control-by-Prabha-Kundur,” 1994).
The quantity of power transactions is growing in restructured power networks. These systems
are projected to run at different operating points, drawing closer to their operational limitations.
The One that Limits bulk power transmission over the network is referred to as "low-frequency
oscillations." These modifications may have a good or positive effect on dampening
oscillations. The power system must be stable to provide and transmit dependable electricity.
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 power system stabilizers. This thesis focuses on 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 (Muljono et al., 2018).
 2

1.1.1. Classification of Power System Stability

Power system stability refers to the characteristics 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 (“Power-System-Stability-and-Control-by-
Prabha-Kundur,” 1994). 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 stable.
However, if the oscillations grow over time, the system loses its stability. Typically, heavy
power flow in transmission lines or the interconnection of the controller with the system
frequency can lead to small signal instabilities (Shubi Sharda, 2016).

Steady-state stability

Steady-state stability analysis refers to examining a power system and its generators under
stable operating conditions to establish the maximum load that can be transmitted without
causing any loss of synchronism in the generators.

Transient stability

Transient stability refers to the response of the 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 small disturbance stability, a component of rotor angle stability
(“Power-System-Stability-and-Control-by-Prabha-Kundur,” 1994).
 3

Power system
stablity

Angle Frequency Voltage

Stability Stability Stability

Mid Long
Term Term
Stability Stability
Small Large Small
Transient Signal disturban-
Stability disturbance
Stability ce

Non Oscillatory
Oscillatory instability

Local Torsional
Inter Area Control
Plant Modes
Modes Modes
Modes

Figure 1.1 Classification of Power System Stability

 4

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,
including a reduction in the lifetime of system components, Electrical grid operations are
expensive, and the worst-case scenario is a partial system breakdown. Furthermore, the actions
of the excitation control system diminish the damping. Provided by the field and damper
windings in synchronous generators because additional currents induced by voltage regulation
appear in the rotor circuits, opposing the currents induced by rotor speed deviations. Nowadays,
the development of automatic voltage regulators (AVRs) in generators is widespread in power
systems. Many generators are equipped with high-gain and fast-acting AVRs to enhance large-
scale stability and ensure the generator remains synchronized 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 function of
power system stabilizers (PSSs) 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 aligned with the speed signal to counteract 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. It leads to
high starting overshot, sensitivity to controller gains, and sluggish response to sudden
disturbances and loading conditions within a brief time. Designing a conventional PSS for
constantly changing power systems is a challenging task. The non-linear nature of power
systems, 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 using a fuzzy logic and neural network control hybrid.
 5

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

1.3. Scope of the Study

The Adaptive Neuro-fuzzy power system stabilizer is restricted to examining an SMIB model
and conducting simulation trials in multi-machine systems. Furthermore, the evaluation of PSS
efficacy and the installation of various control structures are provided as a theoretical
assessment of the findings, supported by simulations.

1.4. Significant of Study

This thesis will develop a methodology for the adaptive neuro-fuzzy power system
implemented in synchronous generators and explore the function of a control structure different
from the one used in the generators for damping electro-motion oscillations.

i. Enhance power flow and improve system reliability.

ii. Decrease the number of machines needed for peak load operation and spinning reserves
for sudden load changes.
iii. Offer a cost-effective power source for consumers.
iv. Electrical devices are susceptible to failure when faced with power stability issues.
v. Improve current and future power generation and transmission systems.
 6

1.5. Objectives of the Study

1.5.1. General objective
The main aim of this 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
and to improve the small signal stability of the power system.

1.5.2. Specific Objectives

The specific goals include:
i. Analyzing Power system stability, Excitation systems, Automatic Voltage Regulators
for synchronous generators, and Power system stabilizers.
ii. Developing the Neuro-fuzzy Power System Stabilizer to swiftly stabilize the system
during transmission line faults and validate its effectiveness.
iii. Decreasing settling time, rise time, peak amplitude, and overshot of the system.
iv. Operating the power system over a wide range.
v. Enhancing the stability and reliability of synchronous generators.
vi. Utilize simulation to validate the Adaptive Neuro-Fuzzy power system stabilizer and
compare its performance with traditional stabilizers.
 7

2. LITERATURE REVIEW
2.1. Review of Related Work
This section will examine previous works related to the damping of low-frequency oscillations
in power systems to enhance stability. The reviewed are categorized into recently published
journal papers, M.Sc., and Ph.D. thesis. While this work draws on related materials, only
selected works from each category will be reviewed due to space constraints.

Yagami and Tamura (2009) (Energy Conversion Congress and Exposition, 2009. ECCE,
IEEE., 2009) Present a method for improving power system stability by combining a fault
current limiter with a thyristor-controlled brake resistor. The fault current limiter is intended to
reduce fault currents, enhance power system stability, and prevent turbine shaft torsional
oscillations. On the other hand, the thyristor-controlled braking resistor aims to 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) (Xie et al., 2013) Present the concept of a switched controller to
analyze an SMIB power system when a symmetrical 3-phase short circuit fault occurs in one
of the transmission lines. Typically, a linear controller does not deliver satisfactory transient
performance for such a power system with a significant error. 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) (Yee & Milanović, 2008) The authors suggest a Fuzzy
Logic Controller (FLC) for the decentralized stabilization of multi-machine power systems.
They have developed an analytical approach for developing a resilient Multi-Input-Single-
Output (MISO) FLC targeted at improving the damping and stability of an electrical power
system while preserving voltage control.
 8

The proposed decentralized FLC employs a systematic analytical method based on a

performance indicator, removing the need for previous knowledge. This FLC tracks speed
variations to zero to stabilize the generator's power output while regulating and stabilizing the
terminal voltage.

Dysko, W.E. Leithead, and J. O'Reilly (2010) (Dyśko et al., 2010) Highly interconnected
systems established a coordinated design method for power system stabilizers (PSSs) and
automated voltage regulators (AVRs). The proposed coordinated PSS/AVR design method is
structured on a frequency domain framework.

P. De Mello (1969) (Demello & Concordia, 1969) The stability of synchronous machines under
small shocks was examined by analyzing a situation in which a generator is connected to an
infinite bus via an external reactance. Power System Stabilizers (PSSs) for a single generator
connected to a bus were designed to provide quick output sampling feedback.

J.H. Chow and J.J. Sanchez-Gasca (1989) (Jamal & Syahputra, 2011) 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.

De Mello and Concordia (1969) (Demello & Concordia, 1969). Nonetheless, the phase
characteristics were derived using a multi-machine Eigenvalue 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 claimed the frequency response approach was reliable
in correcting the delay between excitation input and electrical torque.

Hiyama (1994) introduced a PID-type FLPSS (Ghareeb et al., 2020). 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
display the effectiveness of this modification.
 9

Hiyama's heuristic-based approach demonstrated some success in his series of research studies.
The fuzzy PSS parameters were not globally optimized. He argued that these parameters are
relatively insensitive to external conditions.

Son and Park (2000) (Yee & Milanović, 2008) 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 approximation technique to obtain a reduced-order power system
model and analyzed a controller based on His model. They also discussed using the Loop
Transfer Recovery technique to maintain the robustness of the designed controller. Due to
matrix size limitations, these methods were not extended to higher-order systems.

Del Rosso et al. (2003) (Del Rosso et al., 2003) suggested a hierarchical control system for
improving dynamic and steady-state stability, including strategies for mitigating negative
interactions among TCSC hierarchical controls. They analyzed different locally measurable
input signals qualitatively using the equal area criterion. The focus was on relating 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. Lankans (Othman et al., 2012) outlined the design
methodology for an FLPSS and an ANFIS, examining their effectiveness for a single-machine
power system. They chose speed deviation and its derivative as input signals for the FLPSS.

Vani, M.U., Raju, G.S., and Prasad, K.R.L. (Uma Vani et al., 2009) They outlined a
structured design approach for an ANFIS and optimization-based automatic voltage regulator
and PSS.

Chun-Jung Chen (Chen & Chen, 2007) presented an adaptive power system stabilizer
incorporating a recurrent neural network controller (RNNC) and a compensator to mitigate
power system oscillations. The RNNC delivers an adaptive control signal to the exciter or
governor to dampen most oscillations.
 10

Sumina D. (2008) (Sumina et al., 2008) highlighted the application of a neural network-based
excitation control on an SMIB. The proposed feedforward neural network combines a voltage
regulator and a power system stabilizer.

Barton (2004). He presented a robust artificially intelligent ANFIS-based PSS design for
damping electromechanical oscillation modes and enhancing power 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 power output. Nonlinear simulations illustrated the robustness of the ANFPSS.

Hsu and Chen (1991) (Chen & Chen, 2007) utilized a neural network to adjust the parameters
of a conventional PI (Proportional + Integral) type PSS.

Abdel-Magid et al. (2000) (Abido & Abdel-Magid, 2002) 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
high initial overshot, 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 a fuzzy system trained
using a learning algorithm derived from neural network theory. The learning capability is
advantageous from the FIS perspective, while a linguistic rule base is beneficial from the ANN
perspective. Various methods exist to integrate ANN and FIS, often depending on the specific
application.
 11

2.2. Neuro-Fuzzy Controller

The suggested Neuro-fuzzy controller analyses speed deviation and acceleration error data
using a fuzzy inference approach to generate an output suited for dampening oscillations.
Simulation findings indicate that the ANFIS system outperforms previous methodologies.
Intelligent control solutions have been created for Single Machine Infinite Bus (SMIB) and
Multi-machine systems using multiple model adaptive controllers.
The performance of these controllers is assessed using Fuzzy Logic Control and Neural
Network control techniques.
In the field of Power System Stabilizers, new trends have emerged, leading to a surge in
research papers. Among these, Kothari et al. (Rashidi et al., 2003) developed a variable
structure power system stabilizer with desired eigenvalues in the sliding mode. Hariri and
Malik (Shamsollahi & Malik, 1997) 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 (Abido & Abdel-Magid, 2002) utilized an evolutionary programming
algorithm to determine the optimal values for a classical lead-lag PSS. Rashidi et al. (Rashidi
et al., 2003) 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. (Elshafei et al., 2005) suggested
power system stabilization using fuzzy logic and direct adaptive techniques. Hossein-Zadeh
and Kalam (Hossein-Zadeh & Kalam, 2002) developed an indirect adaptive fuzzy approach.
Elshafei et al. (Elshafei et al., 2005) expanded the direct adaptive fuzzy method to include the
stabilization of multi-machine power systems.
 12

3. MATERIALS AND METHODS

3.1. 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 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 two inputs and a single
output.

Adaptive Neuro-Fuzzy PSS

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

1. Generating suitable training data: Applying the ANFIS technique for power system stability
using Power System Stabilizers is crucial precisely to 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 2.
 13

Figure 3.1. 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 constructed, the ANFIS
models should provide accurate outputs for unseen data. The fuzzy system is optimally adjusted
when the result of test patterns and the network's error falls within an acceptable range. 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).
 14

Start

Literature Study

Create the Multi Machine Power System Model

Develop Robust PSS

Perform the CPSS Analysis

Perform the ANFPSS Analysis

Compare the Performance of CPSS with ANFPSS

Conclusion

Finish

Figure 2.2 Procedure of the research.

 15

3.2. Materials
3.2.1. Power System Stabilizer Model
The primary role of a power system stabilizer is to dampen generator rotor oscillations by
managing their excitation with auxiliary stabilizing signal(s). The stabilizer must provide
electrical torque in phase with rotor speed variations to provide damping. It is generally
understood that fast-acting exciters with high gain AVR can cause rhythmic instability in power
systems. This form of instability is distinguished by low-frequency oscillations (0.2 to 3.0 Hz)
that can persist (and even expand in size) for no apparent cause. This form of instability might
jeopardize system security and impede power transfer.

The major factors that contribute to the instability are:

1. Loading of the generator or tie lines

2. Power transfer capability of transmission lines
3. Power factor of the generator (leading power factor operation is more problematic than
lagging power factor operation)

A cost-efficient and satisfactory solution to the problem of oscillatory instability is to provide

damping for generator rotor oscillations. This is conveniently done by providing Power System
Stabilizers (PSS) supplementary controllers in the excitation systems. The objective of
designing PSS is to provide additional damping torque without affecting the synchronizing
torque at critical oscillation frequencies needed for PSS will be felt in situations when power is
transmitted over long distances with weak AC ties. Even when PSS may not be required under
normal operating conditions, they allow satisfactory operation under unusual or abnormal
conditions that may be encountered at times. Thus, PSS has become a standard option with
modern static exciters power engineers use these effectively. Retrofitting of existing excitation
systems with PSS may also be required to improve system stability. Since the purpose of a PSS
is to introduce a damping torque component, a logical signal to use for controlling generator
excitation is the speed deviation ∆ωr. If the exciter transfer function and the generator transfer
function between ∆Efd and ∆Te were pure gains, direct feedback of ∆ωr would result in a
damping torque component. However, in practice, the generator and the exciter exhibit
frequency-dependent gain and phase characteristics.
 16

Therefore, the PSS transfer function G(s) should have appropriate phase compensation circuits
to compensate for the phase lag between the exciter input and the electrical torque. In the ideal
case, with the phase characteristic of G(s) being an exact inverse of the exciter and generator
phase characteristics to be compensated, the PSS would result in a pure damping torque at all
oscillating frequencies.

Figure 3.3. Haffron- Phillips model of excitation system with PSS

 17

3.2.1.1. General Structure of Power System Stabilizer

Figure 5 depicts a block schematic of the fundamental construction of the power system
stabilizer. It contains a gain block, a washout circuit, a dynamic compensator, and a limiter.

Figure 3.4. General Structure of PSS

The following sections describe the functions of each PSS component.

1. PSS Gain
Stabilizing gain, KPSS, affects the amount of damping introduced by PSS. Ideally, the
PSS gain is chosen to provide the maximum damping of the oscillatory modes. However,
for practical reasons, a high gain may not always be the ideal solution, resulting in undue
amplification of the stabilizer input signal. In general, the gain amount is chosen. That
provides adequate dampening of system modes while maintaining stability constraints.
2. Washout Circuit
The washout circuit is designed to reduce steady-state bias in the PSS output, which
modifies the generator terminal voltage. The PSS is designed to respond only to
transitory fluctuations in the input signal, such as rotor speed, and not to DC offsets in
the signal. The washout circuit functions as a high-pass filter that must pass all
frequencies of interest. If only local modes are of interest, the time constant Tw can be
selected between 1 and 2. However, if inter-area modes are to be dampened, Tw must
be set between 10 and 20. Tw = 10 is required to improve the damping of the inter-area
modes. When Tw is increased from 1.5 to 10, initial swing steadiness improves
noticeably. The increased Tw value enhanced the overall terminal voltage
responsiveness under system islanding situations.
 18

3. Lead-Lag Compensator
The lead-lag compensator block provides an appropriate phase lead for the phase lag
between the exciter input and the generator's electrical torque. In practice, the dynamic
compensator consists of numerous lead-lag compensators, depending on the phase
adjustment required.
4. PSS Output Limits
The stabilizer output voltage is regulated to maximum and minimum values to reduce
generator VT volatility under transient circumstances. Maximum output limitations
guarantee that stabilizers contribute to their full potential, although generator terminal
voltage can fluctuate significantly. The primary goal of setting PSS output limits is to
allow the stabilizer to exert maximum force while keeping the terminal voltage within
desirable limits. The most typical maximum limit value is between 0.1 and 0.2 p.u.,
whereas the minimum limit is between -0.05 and -0.1 p.u.
5. Input of PSS
Many signals, such as rotor speed deviation, frequency deviation, change in load angle,
change in electrical power, etc., can be used as input signals to PSS. However, from a
practical point of view, the following three types of input signals are most frequently
utilized as input to power system stabilizers:
i. Rotor Speed Deviation (∆ω)
ii. Frequency Deviation (∆f)
iii. Electrical Power Deviation (∆P)

Though the speed deviation signal is commonly utilized as an input to PSS, it is intrinsically
sensitive to torsional oscillations in the 8 to 20 Hz frequency band, which might result in
negative damping for the torsional mode. As a result, it is also recommended to apply a torsional
filter, often a low pass filter, to prevent PSS interaction with the torsional mode of oscillations.
 19

4. WORK PLAN
The proposed research will be broken down into subtasks for accomplishing the study within
the given time. Therefore, the following table shows us the subcomponents of the study
concerning their starting and finishing times.

Table 4.1. Work Plan

Name of Year 2023/2024

Task Month

December January February March April May June

Title √
Introduction √ √

Literature √ √ √ √ √ √ √
Review
Methodology √ √
Formulation
Data √ √ √
Collection
and Analysis
Simulation √ √ √
Design
Result √ √
analysis and
discussion
Draft report √ √
Preparation
Final report √
 20

5. RESEARCH BUDGET BREAKDOWN

5.1. Personal Cost
Table 5.1. Per Diem

Activities No. of ETB/day No. of Total

Person Days Cost
(Birr)
1 Researcher 1 250.00 30 7,500.00
2 Advisor 1 300.00 10 3,000.00
3 Data 18 150.00 4 10,800.00
Collectors
Sub-total 21,300.00

Table 5.2 Supervision Fee

No. Description Total price

1. Supervision 3,000.00
Sub-total 3,000.00

5.2. Transport Expense

Table 5.3 Transport Expense

No. Traveler Types of From To Cost (ETB)

Transportation

1. Researcher Minibus Haramaya Dire Dawa 70.00

2. Researcher Minibus Dire Haramaya 70.00

Dawa

Subtotal 140.00
 21

5.3. Stationary Expense

Table 2.4 Stationary Expenses

No. Items unit Quantity Unit price(birr) Total price(birr)

1 Notebook Pcs 1 100.00 100.00

2. Pen Pcs 2 20.00 40.00

3. Paper Packet 1 420.00 420.00

Subtotal 560.00

5.4. Budget Summary

In conducting this research, a number of the costs will be invested. The detailed list of items
and cost estimations are described in the below table.

Table 5.5. Budget Summary

No. Items Cost (ETB)

1. Personal cost 21,300.00

2. Transportation expense 140.00
3. Stationary expense 560.00
4. Supervisor fee 3,000.00
Total 25,000.00
Budget Source: Haramaya University
 22

6. REFERENCE

Abido, M. A., & Abdel-Magid, Y. L. (2002). Optimal design of power system stabilizers
using evolutionary programming. IEEE Transactions on Energy Conversion, 17(4), 429–436.
https://doi.org/10.1109/TEC.2002.805179
Chen, C. J., & Chen, T. C. (2007). Design of a power system stabilizer using a new recurrent
network. International Journal of Innovative Computing, Information and Control, 3(4), 907–
918.
Del Rosso, A. D., Cañizares, C. A., & Doña, V. M. (2003). A Study of TCSC Controller
Design for Power System Stability Improvement. IEEE Transactions on Power Systems,
18(4), 1487–1496. https://doi.org/10.1109/TPWRS.2003.818703
Demello, F. P., & Concordia, C. (1969). Concepts of Synchronous Machine Stability as
Affected by Excitation Control. IEEE Transactions on Power Apparatus and Systems, PAS-
88(4), 316–329. https://doi.org/10.1109/TPAS.1969.292452
Dyśko, A., Leithead, W. E., & O’Reilly, J. (2010). Enhanced power system stability by
coordinated PSS design. IEEE Transactions on Power Systems, 25(1), 413–422.
https://doi.org/10.1109/TPWRS.2009.2036704
Elshafei, A. L., El-Metwally, K. A., & Shaltout, A. A. (2005). A variable-structure adaptive
fuzzy-logic stabilizer for single and multi-machine power systems. Control Engineering
Practice, 13(4), 413–423. https://doi.org/10.1016/j.conengprac.2004.03.017
Energy Conversion Congress and Exposition, 2009. ECCE, IEEE. (2009). IEEE.
Ghareeb, P., Asim, A. G., & Ahmed, M. (2020). " The Design of a Fuzzy Logic Based Power
System Stabilizer Applied to Two Electric Power Generation Units " Researchers : Abstract :
41–68.
Hossein-Zadeh, N., & Kalam, A. (2002). An indirect adaptive fuzzy-logic power system
stabilizer. International Journal of Electrical Power and Energy Systems, 24(10), 837–842.
https://doi.org/10.1016/S0142-0615(01)00093-X
Jamal, A., & Syahputra, R. (2011). Design of Power System Stabilizer Based on Adaptive
Neuro-Fuzzy Method. International Seminar on Applied Technology, Science and Arts
(APTECS), 14–21.
Muljono, A. B., Ginarsa, I. M., Nrartha, I. M. A., & Dharma, A. (2018). Coordination of
adaptive neuro-fuzzy inference system (ANFIS) and type-2 fuzzy logic system-power system
stabilizer (T2FLS-PSS) to improve large-scale power system stability. International Journal of
Electrical and Computer Engineering, 8(1), 76–86. https://doi.org/10.11591/ijece.v8i1.pp76-
86
Othman, M. F., Mahfouf, M., & Linkens, D. A. (2012). Designing Power System Stabilizer for
Multimachine Power System Using Neuro-Fuzzy Algorithm. Jurnal Teknologi, 35(D), 55–63.
https://doi.org/10.11113/jt.v35.610
Power-System-Stability-and-Control-by-Prabha-Kundur. (1994). In P.Kundur, Power System
 23

Stability and Control. Mc-Graw Hill, 1994.

Rashidi, F., Rashidi, M., & Amiri, H. (2003). An Adaptive Fuzzy Sliding Mode Control for
Power System Stabilizer. IECON Proceedings (Industrial Electronics Conference), 1, 626–630.
https://doi.org/10.1109/IECON.2003.1280053
Shamsollahi, P., & Malik, O. P. (1997). An adaptive power system stabilizer using online
trained neural networks. IEEE Transactions on Energy Conversion, 12(4), 382–387.
https://doi.org/10.1109/60.638951
Shubi Sharda. (2016). Efficient Power Oscillation Damping Control By Adaptive Neuro-Fuzzy
Controller Based Power System Stabilizer. International Journal of Research in Engineering
and Technology, 05(09), 101–105. https://doi.org/10.15623/ijret.2016.0509015
Sumina, D., Bulic, N., & Erceg, G. (2008). Simulation model of neural network-based
synchronous generator excitation control. 2008 13th International Power Electronics and
Motion Control Conference, EPE-PEMC 2008, 4, 556–560.
https://doi.org/10.1109/EPEPEMC.2008.4635324
Uma Vani, M., Raju, G. S., & Prasad, K. R. L. (2009). Robust supplementary controllers for
AVR and PSS. Proceedings of INDICON 2009 - An IEEE India Council Conference, 9–12.
https://doi.org/10.1109/INDCON.2009.5409474
Xie, S., Wang, X., Qu, C., Wang, X., & Guo, J. (2013). Impacts of different wind speed
simulation methods on conditional reliability indices. International Transactions on Electrical
Energy Systems, 20(August 2008), 1–6. https://doi.org/10.1002/etep
Yee, S. K., & Milanović, J. V. (2008). Fuzzy logic controller for decentralized stabilization of
multimachine power systems. IEEE Transactions on Fuzzy Systems, 16(4), 971–981.
https://doi.org/10.1109/TFUZZ.2008.917296
 24

APPROVAL SHEET

HARAMAYA UNIVERSITY

POSTGRADUATE PROGRAM DIRECTORATE

Adaptive Neuro-Fuzzy-Based Power System Stabilizers for Low-Frequency

Damping Oscillations in Multi-Machine Power Systems

Submitted by:
Kenan Yeshitela Mokuria
____________________ ________________ ______________
Name of Student Signature Date

Approved by:

1. Prof./Dr./Mr._____________________ ____________ ____________

Major Advisor Signature Date

2. Prof./Dr./Mr._______________________ ____________ ____________

Chairperson, DGC/SGC Signature Date

3. Prof./Dr./Mr.__________________________ ____________ ____________

Research Thematic Area Leader Signature Date

4. Prof./Dr./Mr.__________________________ ____________ ____________

Associate Director for PGRP, HiT Signature Date

5. Prof./Dr./Mr.___________________________ ____________ ____________

Director, Postgraduate Programs, HU Signature Date