Energies 14 05644

energies
Article
Power System Stabilizer Tuning Algorithm in a Multimachine
System Based on S-Domain and Time Domain System
Performance Measures
Predrag Marić 1, *, Ružica Kljajić 1 , Harold R. Chamorro 2 and Hrvoje Glavaš 1
1 Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, University of Osijek,
31000 Osijek, Croatia; ruzica.kljajic@ferit.hr (R.K.); hrvoje.glavas@ferit.hr (H.G.)
2 KTH Royal Institute of Technology, 100 44 Stockholm, Sweden; hr.chamo@ieee.org
* Correspondence: predrag.maric@ferit.hr
Abstract: One of the main characteristics of power systems is keeping voltages within given limits,
done by implementing fast automatic voltage regulators (AVR), which can raise generator voltage
(i.e., excitation voltage) in a short time to ceiling voltage limits while simultaneously affecting
the damping component of the synchronous generator electromagnetic torque. The efficient way
to increase damping in the power system is to implement a power system stabilizer (PSS) in the
excitation circuit of the synchronous generator. This paper proposes an enhanced algorithm for PSS
tuning in the multimachine system. The algorithm is based on the analysis of system participation
factors and the pole placement method while respecting the time domain behavior of the system
after being subdued with a small disturbance. The observed time-domain outputs, namely active
power, speed, and rotor angle of the synchronous generator, have been classified and validated with

proposed weight functions based on the minimal square deviation between the initial values in a
Citation: Marić, P.; Kljajić, R.;
steady-state and all sampled values during the transitional process. The system weight function
Chamorro, H.R.; Glavaš, H. Power
proposed in this algorithm comprises s-domain and time-domain indices and represents a novel
System Stabilizer Tuning Algorithm
approach for PSS tuning. The proposed algorithm performance is validated on IEEE 14-bus system
in a Multimachine System Based on
S-Domain and Time Domain System
with a detailed presentation of the results in a graphical and table form.
Performance Measures. Energies 2021,
14, 5644. https://doi.org/10.3390/ Keywords: multi-machine stability; s-domain analysis; participation analysis; PSS tuning; time-
en14185644 domain analysis
Academic Editor: Abu-Siada Ahmed
Received: 8 August 2021 1. Introduction

Accepted: 3 September 2021
Low-frequency oscillations, also known as electromechanical oscillations, were first
Published: 8 September 2021
noticed in the American West system. They can appear in large interconnected systems
within the frequency range between 0.01 to 3 Hz [1], mainly caused by an imbalance
Publisher’s Note: MDPI stays neutral
between energy demand and generation. Electromechanical oscillations represent char-
with regard to jurisdictional claims in
acteristic system responses (low-frequency modes) to a disturbance in the system and, if
published maps and institutional affil-
not properly damped, can lead to system instability [2]. PSSs are a practical solution for
iations.
electromechanical oscillations damping and their implementation as an additional block in
generator excitation systems is simple. PSS is the most efficient for local oscillatory modes
with frequencies 6–12 rads−1 and can also be used to improve transient stability mostly
combined with AVR [3,4] or flexible AC transmission system devices (FACTS) [5–7].
Copyright: © 2021 by the authors.
However, PSS tuning is a complex task. One of the most common methods for the PSS
Licensee MDPI, Basel, Switzerland.
tuning is a pole placement method where the dominant pair of the poles of the observed
This article is an open access article
power system is damped and shifted to the left side of the S-plane by selecting the proper
distributed under the terms and
PSS parameters. One of the challenges for PSS implementation in a multi-machine system
conditions of the Creative Commons
is to find an optimal PSS location. Participation factor analysis shows the impact of each
Attribution (CC BY) license (https://
generator on the system state variables and thus indicates the location where PSS should
creativecommons.org/licenses/by/
4.0/).
be implemented [8]. Participation factors can be found for both monotone and oscillatory
Energies 2021, 14, 5644. https://doi.org/10.3390/en14185644 https://www.mdpi.com/journal/energies

Energies 2021, 14, 5644 2 of 29
modes. In comparison with oscillatory modes, participation factors are very low for
monotone modes [9]. In papers [10,11], the participation analysis is applied on a large
multi-machine system to identify critical generators and modes in the system. Several
papers combine both the participation factor analysis in the multi-machine system and
advance methods for PSS tuning [12–14]. The author [15] presents global pole placement
problems as the composition of separate global and local pole-placement problems used to
solve local oscillatory modes, usually undamped. This solution was applicable for multiple
operating conditions, which led to a robust design technique [16]. In paper [17], extensive
research on system stability for various combinations of system parameters is performed. In
a three-part paper [18–20], PSS application and a complete tuning process are overviewed.
Over the years, application and tuning processes evolved into more sophisticated methods,
mostly heuristic methods [21–23], like genetic algorithms [24], fuzzy logic [25,26], or other
optimization techniques [27,28]. Paper [29] compares three different types of stabilizers,
conventional PSS (CPSS), fuzzy, and adaptive neuro-fuzzy inference system (ANFIS) based
PSS. All three types show appropriate responses to a system disturbance, giving a slight
advantage in performance to ANFIS. Artificial intelligence methods are mainly applied on
single machine infinite bus (SMIB) system models, as in [30–34]. Additionally, many novel
approaches for PSS and AVR tuning are applied on SMIB, for example, nonlinear robust
PSS-AVR controller [35]. However, PSS tuning in multi-machine systems is much more
complex in comparison to SMIB since one generator can deteriorate the performance of
other generators in the system.
In most papers, tuning of the PSS parameters is based on results in the S-domain [36–38],
and, afterwards, results were verified in the time domain by comparing the system response
to disturbances before and after PSS implementation. Due to the possible inertia decrease
with high penetration of converter-interfaced renewable energy sources, the modern power
system may become more vulnerable to disturbances that can initiate low-frequency oscilla-
tions. From the consumer side, the amplitude of active power low-frequency oscillations is
particularly unfavorable since they can lead to load shedding. The PSS parameters obtained
with the traditional pole-placement method show satisfactory results regarding s-domain
system performance measures; however, this paper points to the improvement of damping
of the low-frequency oscillations when the time-domain system performance measures are
included in the PSS tuning too. Hence, the paper proposes the algorithm for the PSS1A
tuning considering both the s-domain and time-domain system performance measures. The
synthesis of these measures is achieved through the proposed weight functions that output
comparable absolute values; the s-domain weight function is proposed as the damping ratio of
the damping oscillatory mode since the damping ratio takes values between 0 and 1; the time
domain weight functions are proposed as the distance between value 1 and the sum of the
square differences from the initial steady-state value and the sampled values of the observed
variable during the time-domain simulation of the disturbance in the system. The proposed
time-domain weight functions are related to active power, speed, and the rotor angle of the
generator where the PSS has been implemented according to the participation factor analysis
and represents the generator with the most significant contribution to the dominant system
mode. The output of the weight functions closer to value 1 indicates the higher function
weight. The sum of the s-domain and time-domain weight functions constitutes a system
performance weight function. The maximum of the system performance weight function
is obtained with the PSS parameters variations in an ascending and a descending order in
separate clusters. The PSS parameters that correspond to the maximum of the system weight
function are classified as optimal.
This paper consists of six sections. The Introduction provides a literature overview of
the PSS tuning techniques and briefly highlights the novelties proposed in the paper. The
second section gives a brief description of the PSS1A structure. The proposed algorithm
for PSS1A tuning, considering both the s-domain and time domain system performance
measures, is the scope of Section 3. In the fourth section, the application of the proposed al-
gorithm is examined on the performance of the IEEE 14-bus system model after initiation of
rithm for
s 2021, 14, x FOR PEER REVIEW PSS1A tuning, considering both the s-domain and time domain system 3 ofperfor-
30
mance measures, is the scope of Section 3. In the fourth section, the application of the
proposed algorithm is examined on the performance of the IEEE 14-bus system model
afterrithm
initiation of thetuning,
for PSS1A disturbances withboth
considering a different duration
the s-domain within
and time threesystem
domain case studies.
perfor- The
Energies 2021,Discussion
14, 5644 3 of 29
mance measures, is the scope of Section 3. In the fourth section, the application of theFinal
follows the Results, presented in figures and tables for all case studies.
remarks and algorithm
proposed possible benefits fromonthe
is examined theapplication
performanceofofthe
theproposed algorithm
IEEE 14-bus are given
system model
in the Conclusion.
after initiation of the disturbances with a different duration within three case studies. The
the disturbances with a different duration within three case studies. The Discussion follows
Discussion follows the Results,
the Results, presented
presented in figures
in figures and for
and tables tables for all
all case case studies.
studies. Final and possible
Final remarks
remarks and
2. Materials andpossible
benefitsbenefits
Methods from thefrom the application
application of the proposed
of the proposed algorithmalgorithm
are givenare given
in the Conclusion.
in the Conclusion.
The authors conducted an extensive literature review on the topic of PSS tuning, cov-
2. Materials and Methods
ering
2. both the fundamentals
Materials and Methods of tuning and modern optimization methods. Based on this
The authors conducted an extensive literature review on the topic of PSS tuning,
research, it was found that commonly used PSS tuning procedures based solely on the s-
The authors conducted
covering bothanthe
extensive literature
fundamentals of review
tuningon andthemodern
topic of optimization
PSS tuning, cov- methods. Based
domain
eringsystem
both theperformance
fundamentals measures
of tuning canmodern
and be improved by involving
optimization the time-domain
on this research, it was found that commonly usedmethods. Based
PSS tuning on this based solely
procedures
system performance
research, measures.
it wasonfound
the Tosystem
that commonly
s-domain addressused this
PSSdeficiency,
tuning
performance the can
authors
procedures
measures based
be proposed
solely on
improved byathe
method
s-
involving the time-
thatdomain
includes both domain
systemdomain
performancesystem performance
system measures
performance measures
canmeasures.
be improved to find
by involving
To address the optimal PSS
the time-domain
this deficiency, param-
the authors proposed a
eters. The results
system were measures.
performance
method verified on
To the
that includes IEEE
address
both 14–bus
this
domain test performance
deficiency,
system system consisting
the authors proposed
measuresofafive genera-
method
to find the optimal PSS
tors that
withincludes both
parameters domain
publicly
parameters. system
The performance
available.
results measures
All generators
were verified on areto find
the the
equipped
IEEE optimal
14–bus testPSS
with param-
standard
system IEEE of five
consisting
eters. The results were
generators verified
with on the
parameters IEEE
AVRs models with parameters available in the Appendix A. 14–bus
publicly test system
available. Allconsisting
generators of five
are genera-
equipped with standard
IEEE AVRs
tors with parameters models
publicly with parameters
available. available
All generators in the Appendix
are equipped A.
with standard IEEE
AVRs
3. PSS models with parameters available in the Appendix A.
Structure 3. PSS Structure
3.Power system stabilizer
PSS Structure provides
Power system an additional
stabilizer providesdamping torque
an additional in a phase
damping withingen-
torque a phase with
erator speed variation and it is simply implemented as a part of the synchronous genera-
Power system stabilizer provides an additional damping torque in a phase with gen- synchronous
generator speed variation and it is simply implemented as a part of the
tor excitation system
erator speed [18]—Figure
generator
variation excitation
and 1.system
it is simply [18]—Figure
implemented as a1.part of the synchronous genera-
tor excitation system [18]—Figure 1.
Figure 1. Synchronous
Figure generator
1. Synchronous Figure 1.excitation
generator excitation system.
Synchronous generator excitation system.
system.
TheThe main
main structure Theof
structure main thestructure
ofthe PSSpresented
PSS of the PSS
presented in presented
in Figure
Figure in known
2 2(also
(alsoFigure
known 2in(also
[39] known in [39] as PSS1A),
as PSS1A),
in [39] as PSS1A),
consists of the consists
gain, of the gain,
wash-out wash-out
filter, and filter, and
phase phase compensation
compensation blocks blocks (lead-lag
(lead-lag filters). filters). Some
consists of the gain, wash-out filter, and phase compensation blocks (lead-lag filters).
Some additional
additional blocks,suchblocks, such as
suchasashigh-frequency high-frequency
high-frequency torsional torsional filters or transducer
con-con- constant
time
Some additional blocks, torsionalfilters oror
filters transducer
transducer timetime
compensation, may be implemented [39]. Torsional filter block can be implemented by
stantstant compensation, may be implemented [39]. Torsional filter block can be implemented
compensation, may constants
defining be implementedA and A [39].
. If Torsional
the filter block can be implemented
by defining constants A1 and A2. If 1the torsional 2 filter is to be neglected, these constants constants are
torsional filter is to be neglected, these
by defining constants A and
1 1 Athis
2. Ifblock
the torsional filter is to betorsional
neglected, these constants
are set to be 1 and this block is used to compensatecompensate
set to be and is used to torsional modes, which modes, which
are the resultare the result of
are set to be 1 and this block is used to compensate torsional modes, which
of rotational movement. The first block is used to compensate transducer time constant constant and
rotational movement. The first block is used to compensate are
transducer the result
time
of rotational
and can bemovement.
can be by
neglected The
neglected firstby
setting block
timesetting istime
constant used to compensate
Tconstant
6 to 1.
T6 to 1. transducer time constant
and can be neglected by setting time constant T6 to 1.
Figure
Figure 2. Type
2. Type PSS1A
PSS1A single-input
single-input power
power system
system stabilizer.
stabilizer. Reproduced
Reproduced from
from [39],[39], IEEE: 2016.
Copyright 2016 IEEE.
Figure 2. Type PSS1A single-input power system stabilizer. Reproduced from [39], IEEE: 2016.
Gain KS definesGain
a circuit gain and
KS defines damping
a circuit gain torque. Gain increase
and damping torque. leads to damping
Gain increase leads to damping
increase until it reachesuntil
increase a maximum
it reachespoint after which
a maximum a further
point gain aincrease
after which further results in
gain increase results in
Gain KS defines a circuit gain and damping torque. Gain increase leads to damping
less damping torque [40]. Gain
less damping should
torque [40].beGain
tuned following
should the maximum
be tuned damping
following the maximum but damping but
increase until it reaches a maximum point after which a further gain increase results in
also all the other limitations need to be taken into consideration. The second block is the
less damping torque [40]. filter
wash-out Gain(defined
should bybeits
tuned following
time constant T5 ).the maximum
This damping
is a high-pass but defines
filter which
the operating margin for PSS. PSS should not respond to small and slow changes which
Energies 2021, 14, 5644 4 of 29
are mostly caused by a change of the system operating point. Wash-out filter eliminates
these disturbances and only reacts to transient disturbances [40]. The value of T5 must
be in a range in which it does not change observed frequencies but also does not lead to
unwanted voltage trips. The range for T5 is 1 to 20 s [41]; for local modes, it should be 1–2 s
and 10–20 s for interarea modes [42]. Input in an excitation system and electrical torque as
an output signal has a phase shift, which needs to be compensated. Phase compensation
is achieved with cascade lead-lag filters, which are presented with constants T1 –T4 . To
simplify the process, it is assumed to be T2 = T4 and T1 = T3 . T2 and T4 usually have fixed
values and the other two-time constants need to be tuned. PSS output voltage needs to be
limited to restrict terminal voltage fluctuations [43]. The output voltage is limited between
values V STmin and V STmax .
4. PSS Tuning Algorithm

The first step in PSS tuning in the proposed algorithm is the system modal analysis
in order to identify a dominant oscillatory mode since it determines the dynamic system
properties. Each oscillatory mode is the system eigenvalue represented as a complex
pair with the real part—the damping constant, and the imaginary part—the oscillation
(damping) frequency. The dominant system oscillatory mode is identified as the complex
pair closest to the imaginary axis on the system S-plane. The pole placement method used
in this algorithm aims to nullify the imaginary part of the dominant system mode and to
significantly shift it left from the initial location on the S-plane integrating the PSS into
the system [1]. The most efficient PSS application would be for the damping frequencies
range of the dominant system mode within 1.5–12 rads−1 , i.e., for the inter-area and local
oscillatory modes [44]. Hence, the system oscillatory modes with damping frequencies
outside this range will not be the subject for this paper’s analysis.
Once the appropriate dominant system mode is identified, the proposed algorithm
tends to find an optimal location for the PSS implementation in the multi-machine system.
Participation factors determine the contributions of each active component in the system
oscillatory properties [41]. Since the PSS affects the system damping, the participation of
the state variable speed for the dominant oscillatory mode is chosen as a criterion for the
PSS allocation, which implies that the excitation system of the synchronous machine with
the highest participation of the state variable speed in the dominant oscillatory mode is the
most appropriate for the PSS implementation.
The PSS tuning optimization is based on the optimal selection of the PSS parameters:
the washout filter time constant-T5 , gain-K, and lead-lag time constants-T2 = T4 and
T1 = T3 . According to the suggestions in literature [18–20,39,45], the lead-lag-T2 = T4
constants should be set to fixed value T2 = T4 = 0.02 s. The washout filter time constant
is set to T5 = 10 s according to the suggested values in [18–20,40]. The initial gain-Kinit is
determined according to the damping of the dominant oscillatory mode-αdom represented
as [45]:
K
− αdom = d (1)
4H
and should be at least equal to coefficient-Kd :
Kinit = Kd (2)
While coefficient-H represents an inertia time constant of the synchronous machine

with the highest participation in the dominant oscillatory mode.
As recommended in [18–20,39], the initial value for time constants T1 and T3 should
be T1 = T3 = 0.2 s.
The PSS tuning procedure proposed in this paper is the optimization problem to find
the PSS gain-K and the PSS time constant-T = T1 = T3 for the maximal left-shift and the
maximal damping ratio ζ dom of the dominant oscillatory mode in the S-plane considering
the time-domain behavior of the rotor angle, speed, and active power of the synchronous
Energies 2021, 14, 5644 5 of 29
machine with the highest participation in the dominant oscillatory mode after initiating a
(small) disturbance in the system.
The time-domain indices presented in this paper are related to the time-domain
measures of the system performance after initiating a small disturbance. The proposed
measures are based on the sum of the square differences between the initial values of the
rotor angle-δ (related to synchronizing torque); speed-ω (related to damping torque); active
power-P in steady-state-δinit , ωinit , Pinit ; and all of the simulated values δt , ωt , Pt during
the time domain simulation period tsim − tinit after a (small) disturbance has been subdued
at the time instant t = tinit :
t = tsim
δt − δinit 2

∑ δinit
(3)
t=t init
t = tsim 2
ωt − ωinit
∑ ωinit
(4)
t = tinit
t = tsim 2
Pt − Pinit
∑ Pinit
(5)
t = tinit
The smaller outputs of the sum of the square differences (3)–(5) correspond to better
time-domain measures of the system performance. Based on (3)–(5), the weight functions
Γδ ,Γω , and Γ P are introduced:
t = tsim 2
δt − δinit
Γδ = 1 − ∑ δinit
(6)
t = tinit
t = tsim 2
ωt − ωinit
Γω = 1 − ∑ ωinit
(7)
t = tinit
t = tsim 2
Pt − Pinit
ΓP = 1 − ∑ Pinit
(8)
t = tinit
Hence, the output of each weight function closer to 1 represents better time-domain
system performance. Since the damping ratio ζdom of the dominant system mode falls into
the interval (0, 1) and indicates better system dynamic properties when it tends to a value
of 1, as the s-domain system performance measure, the damping ratio is comparable with
the proposed weight functions (6)–(8). It follows that a new weight function based on the
damping ratio-Γ(ζ ) can be presented as:
Γζi = ζ dom (9)
Comprising both the time-domain and s-domain measures, the system performance
function can be represented as the weight function-Γ:
Γ= ∑ Γζ,δ,ω,P (10)
ζ,δ,ω,P
The optimal PSS time constant T is selected as Ti,j value obtained for the maximal
output of weight function Γ j (i ) which represents the system performance weight function-
Γ calculated in the iterative procedure for the j-th cluster composed of five PSS time
constants-Ti,j increasing the initial value-Tinit for the increment of 0.1 s in each subsequent
i-th iteration:
T = Ti,j f or maxΓ j (i ) (11)
i =4
Ti,j = Tinit + ∑ i ·0, 1 (12)
i =0
Energies 2021, 14, 5644 6 of 29
Until Γ j (4) > Γ j (3), a new cluster, composed of five Ti,j will be defined according
to (12), where the initial PSS time constant value will be the last value from the previous
cluster j−1:
Tinit = T4,j−1 (13)
After the PSS time constant T has been determined, the PSS gain K will be obtained in
iterative procedures with invariant T.
The PSS gain K is chosen as one of the values obtained in the following procedure:
If
maxΓi,um (i ) > maxΓi,ln (i ) (14)
K = Ku
else
K = Kl
where -Ku is the PSS gain obtained for the maximal output of weight function Γi,um (i),
which represents the system performance weight function-Γ calculated in the iterative
procedure for the u-th cluster composed of five PSS gain constants-Ki,um increasing the
initial PSS gain-Kinit for increment 1 in each subsequent i-th iteration:
Ku = Ki,um f or maxΓi,um (i ) (15)
i =4
Ki,um = Kinit + ∑ i ·1 (16)
i =0
Until Γ4,um (4) > Γ3,um (3), a new cluster composed of five PSS gain constants-Ki,um, ,
will be defined according to (16) where the initial PSS time constant value will be the last
value from the previous cluster u−1:
Kinit = K4,um−1 (17)
Kl is the PSS gain obtained for the maximal output of weight function Γi,ln (i) which
represents the system performance weight function-Γ calculated in the iterative procedure
for the l-th cluster composed of five PSS gain constants-Ki,ln decreasing the initial PSS
gain-Kinit for decrement 1 in each subsequent i-th iteration:
Kl = Ki,ln f or maxΓi,ln (i ) (18)
i =4
Ki,ln = Kinit − ∑ i ·1 (19)
i =0
Until Γ4,ln (4) > Γ3,ln (3), a new cluster, composed of five PSS gain constants -Ki,ln , will
be defined according to (19) where the initial PSS time constant value will be the last value
from the previous cluster l−1:
Kinit = K4,l −1 (20)
More detailed insight into the proposed algorithm is presented in Figure 3. The
proposed principle makes both the s-domain and time-domain measures of the system
performance comparable and can be considered beneficial for the PSS fine-tuning.
Energies 2021, 14, x FOR PEER REVIEW 7 of 30
Energies 2021, 14, 5644 7 of 29
MODAL ANALYSIS
OF THE SYSTEM
j=1 Tinit = T4j

N – EIGENVALUES Tinit = Tij j=j+1
m=1 m=m+1 n=1 n=n+1
λ = α ±ϳω
FOR i = 0 TO i = 4 FOR i = 0 TO i = 4 FOR i = 0 TO i = 4

Tij = Tinit + 0.1 i Ki,um = Kinit + i·
1 Ki,ln = Kinit - i·
1
NO 1.5 ≤ ϳω ≤
≤ 12 rad s-1 MODAL ANALYSIS MODAL ANALYSIS MODAL ANALYSIS
(PSS WITH Tij) (PSS WITH Ki,um) (PSS WITH Ki,ln)
YES FIND DOMINANT FIND DOMINANT FIND DOMINANT

OSCILLATORY MODE OSCILLATORY MODE OSCILLATORY MODE
FIND DOMINANT
OSCILLATORY MODE
λdom = αdom ±ϳωdom Γζi = ζ dom Γζi = ζ dom Γζi = ζ dom
EXECUTE PARTICIPATION
ANALYSIS TIME-DOMAIN TIME-DOMAIN TIME-DOMAIN
COMPARE PARTICIPATONS OF SIMULATION OF SIMULATION OF SIMULATION OF
STATE VARIABLES SMALL DISTURBANCE SMALL DISTURBANCE SMALL DISTURBANCE
GENERATOR SPEED
CALCULATE CALCULATE CALCULATE
MAX PARTICIPATION Γδ Γδ,um Γδ,ln
Γω Γω,um Γω,ln
PSS LOCATION Γp Γp,um Γp,ln
Γj(i)= Γδ + Γω+ Γp+Γζ Γj,um= Γδ,um + Γω,um+ Γp,um+Γζi Γj,ln= Γδ,ln + Γω,ln+ Γp,ln+Γζi
Kd
αdom = Kinit = Kd
4H
T5 = 10s
T1 = T3 = 0.2 s YES YES YES
Γj(4) > Γj(3) Γ4,um(4) > Γ2,um(3) Γ4,ln(4) > Γ2,ln(3)
T2 = T4 = 0.02 s
Tinit = T1 = T3
NO NO NO
COMPARE Γj(i) FOR Ku = Ki,um Ku = Ki,ln

max Γj(4) = FOR max Γi,um(i) FOR max Γi,ln(i)
T= Tij
IF
YES
K = Ku max Γj,um(i) >
max Γj,ln(i)
NO
K = Kl
END
Figure 3.
Figure 3. PSS
PSS tuning
tuning algorithm.
algorithm.
5. PSS Tuning Algorithm Impact on the System Performance

5. PSS Tuning Algorithm Impact on the System Performance
The proposed algorithm’s impact on the system performance has been analyzed on
The 14–bus
the IEEE proposed algorithm’s impact
multi-machine systemoninthe system performance
DIgSILENT PowerFactoryhas been analyzed
simulation on the
interface.
The single line diagram system model is presented in Figure 4. All the generators in The
IEEE 14–bus multi-machine system in DIgSILENT PowerFactory simulation interface. the
single line
model diagram system
are equipped model(parameters
with AVRs is presentedare
in Figure 4. the
given in All Appendix
the generators in the model
A). Busbars 2, 4,
are equipped
and with AVRs
5 are equipped (parameters
with static are given
compensation in the (parameters
devices Appendix A).areBusbars 2, the
given in 4, and 5 are
Appen-
equipped
dix A). with static compensation devices (parameters are given in the Appendix A).
Modal analysis of the system indicates, on 55 modes (eigenvalues), 22 modes are
oscillatory while 33 modes are monotone. Damping frequencies of five oscillatory modes
fall in the range of 8.6–13 rads−1 (1.3–2.1 Hz), which corresponds to local modes and will be
further analyzed in Table 1. Other oscillatory modes correspond to low-frequency modes
mainly caused by AVRs and are supposed to be well-damped after the PSS implementation.
rgies 2021, 14, x FOR PEER REVIEW 8 of
Energies 2021, 14, 5644 8 of 29
Figure 4. IEEE4. 14
Figure bus
IEEE 14 system model
bus system modelininDIgSILENT PowerFactory
DIgSILENT PowerFactory simulation
simulation interface.
interface.
Table 1. Local
Modal oscillatory
analysis modes
of the for theindicates,
system analyzed system without
on 55 the PSS
modes implementation.22 modes are
(eigenvalues),
cillatory Mode
whileNumber
33 modes are monotone.
α [s−1 ] Damping frequencies
±jω [rad s−1 ] of fiveΞ oscillatory mod
fall in the range
Mode 1of 8.6–13 rads
−1 (1.3–2.1 Hz), which
−0.6029112884 corresponds0.069933275444
8.6001286177 to local modes and w
be further analyzed in Table 1. Other oscillatory modes correspond to low-frequen
Mode 2 − 3.0888275889 9.1371914398 0.32024636109
Mode 3 −3.3668021306 10.953437301 0.30026554697
modes mainly caused by AVRs
Mode 4
and are supposed
−3.7886935644
to be well-damped
11.819709815
after the PSS imp
0.30524242934
mentation. Mode 5 −4.5222909467 12.584279377 0.33818650996
Table 1. Local oscillatorydiagram

The root-locus modes of
forthe
theanalyzed
analyzed system
system without the
is presented PSS implementation.
in Figure 5; monotone
modes are marked green, low-frequency oscillatory modes are marked red, while local
Mode Number
oscillatory α [s−1]
modes are marked blue. ±jω [rad s−1] Ξ
Mode 1
The damping closest−0.6029112884
to the horizontal axis on the8.6001286177
S-plane indicates Mode 0.069933275444
1 as the
dominant
Mode 2 oscillatory mode. According
−3.0888275889 to the algorithm proposed
9.1371914398 in the previous chapter,
0.32024636109
the PSS location will be determined considering the highest participation of the generator
Mode
speed state3variable in the −3.3668021306
dominant oscillatory mode. 10.953437301 0.30026554697
As can be seen in Table 2, the highest
Mode 4 the value of−3.7886935644
participation, 0.954, belongs to synchronous 11.819709815 0.30524242934
generator 1 (Gen1, highlighted in
Figure 4),
Mode 5 which implies that the excitation
−4.5222909467 system of this generator
12.584279377 is the most suitable for
0.33818650996
the PSS implementation. The participation of other synchronous generators (Gen 2–Gen 5
and External Grid) in the dominant system mode can almost be neglected.
The root-locus diagram of the analyzed system is presented in Figure 5; monoto
modes are marked green, low-frequency oscillatory modes are marked red, while lo
oscillatory modes are marked blue.
14, x FOR PEER REVIEW 9 of 30
Energies 2021, 14, 5644 9 of 29
Figure 5.Figure
Root locus diagram
5. Root of the of
locus diagram analyzed system
the analyzed without
system the the
without PSSPSS
implementation.
implementation.
The damping closest

Table to the horizontal
2. Participation axisspeed
of the generator on the
stateS-plane indicates
variable for Mode
the dominant 1 as the
oscillatory mode.
dominant oscillatory mode. According to the algorithm proposed in the previous chapter,
Generator Unit Participation in Mode 1
the PSS location will be determined considering the highest participation of the generator
Gen1
speed state variable in the dominant oscillatory mode. As can be seen in Table 0.954
2, the high-
Gen2 0.120
est participation, the value of 0.954, Gen3belongs to synchronous generator 10.005 (Gen1, high-
lighted in Figure 4), which impliesGen4 that the excitation system of this generator 0.003is the most
suitable for the PSS implementation. Gen5The participation of other synchronous 0.002generators
(Gen 2–Gen 5 and External Grid) in the dominant system mode can almost 0.046
External Grid
be neglected.
Table 2. Participation ofAfter the PSS location

the generator has been
speed state determined,
variable the tuning
for the dominant process begins
oscillatory mode. with washout
filter time constant setting to T5 = 10 s according to the proposed algorithm, i.e., the
Generator
suggestedUnitvalues in [18–20,39]. This valueParticipation
will not affectinthe
Mode 1
regulation loop regarding
the Gen
phase1 shift. According to (1) and (2), the initial PSS
0.954
gain is set to K = Kinit = 10. The
value for the time constants T1 and T3 is initially set
Gen2 to T1 = T3 = 0.2 s (recommended
0.120
in [18–20,39]).
Gen3 0.005
5.1. Gen4
PSS Lead-Lag Tuning 0.003
Gen5 0.002
According to (11) and (12) in the proposed algorithm, modal analyses and time-domain
External Grid
simulations are made in clusters containing five samples,0.046
changing T = T1 = T3 in each iteration
while maintaining the PSS gain as a constant value. Table 3 shows the modal analyses results
forlocation
After the PSS time constants
has been of cluster 1 − Ti,1the
determined, =T = 0.2–0.6
tuning s for begins
process the dominant oscillatory mode.
with washout
Increasing
filter time constant settingPSSto Ttime constants T shifts this pole further to the left side of the S-plane and at
5 = 10 s according to the proposed algorithm, i.e., the sug-
the same time increases the damping ratio. The maximum value of the damping ratio and
gested values in [18–20,39]. This value will not affect the regulation loop regarding the
the minimum value of the damping are fulfilled with the time constant(s) T = T3,1 = 0.5 s. A
phase shift. According
further to (1) and
increase of (2),
timethe initial PSS
constant(s) gainthe
pushes is set to 𝐾 =oscillatory
dominant 𝐾𝑖𝑛𝑖𝑡 = 10.pole
Theback
valuein the right
for the time constants T 1 and T3 is initially set to T1 = T3 = 0.2 s (recommended in [18–20,39].
part of the S-plane while at the same time the damping ratio decreases (Figure 6). Mode T00
represents the dominant pole without implemented PSS.
5.1. PSS Lead-Lag Tuning
According to (11) and (12) in the proposed algorithm, modal analyses and time-do-
main simulations are made in clusters containing five samples, changing T = T1 = T3 in each
iteration while maintaining the PSS gain as a constant value. Table 3 shows the modal
analyses results for time constants of cluster 1−𝑇𝑖,1 = T = 0.2 s–0.6 s for the dominant oscil-
latory mode. Increasing PSS time constants T shifts this pole further to the left side of the
damping ratio and the minimum value of the damping are fulfilled with the time con
stant(s) T = T3,1 = 0.5 s. A further increase of time constant(s) pushes the dominant oscilla
tory pole back in the right part of the S-plane while at the same time the damping rati
Energies 2021, 14, 5644 10 of 29
decreases (Figure 6). Mode T00 represents the dominant pole without implemented PSS.
Table 3. Dominant oscillatory pole for PSS parameters K = 10, T = Ti,1 = 0.2 s–0.6 s (cluster 1).
Table 3. Dominant oscillatory pole for PSS parameters K = 10, T = Ti,1 = 0.2–0.6 s (cluster 1).
I i Ti,1 [s]T [s] α [sα−1[s]−1 ] ±jω
±jω [rad
[rad s−1s] ]
−1
ξ ξ
i,1
0 0
0.2 0.2 −0.846729311
−0.846729311 8.6001286177
8.6001286177
0.094299372
0.094299372
1 1 0.3 0.3 −1.306793920
−1.306793920 9.1371914398
9.1371914398 0.141835333
0.141835333
2 2 0.4 0.4 −2.059476510
−2.059476510 10.953437301
10.953437301 0.214871539
0.214871539
3 0.5 −2.442664922 11.819709815 0.507928180
3 4 0.5 0.6 −2.442664922
−2.024097138 11.819709815
12.584279377 0.507928180
0.425562142
4 0.6 −2.024097138 12.584279377 0.425562142
Figure6.6.Dominant
Figure Dominant oscillatory
oscillatory polepole movement
movement in theinS-plane
the S-plane after
after the PSSthe PSS implementation
implementation with with
= 10 and the increasing time constants from T = 0.2 s to T = 0.6
K = 10 and the increasing time constants from T = 0.2 s to T = 0.6 s. s.
The
Theobserved
observed variables in the
variables in time-domain
the time-domainsimulations are rotor
simulations areangle,
rotorspeed,
angle,and
speed, an
active power of Gen 1 after initiating a (small) disturbance (3-phase short-circuit) on
active power of Gen 1 after initiating a (small) disturbance (3-phase short-circuit) on BU BUS
4 with the duration of 30 ms at the time instance tinit = 0.1 s from the simulation begin-
4 with the duration of 30 ms at the time instance tinit = 0.1 s from the simulation beginning
ning. Figure 7 shows the rotor angle response with different PSS time constant(s)-T = Ti,1
Figure 7 shows the rotor angle response with different PSS time constant(s)-T = Ti,1 setting
setting. As can be seen from the figure, increasing the PSS time constant(s) results in a
As cansettling
shorter be seen from
time andthe figure,overshoot
decreased increasingandthe PSS timeofconstant(s)
undershoot results
the rotor angle in a shorte
during
settling
the time and
transitional decreased
process. Similarovershoot
behavior isand
shown undershoot of the speed
by the generator rotor angle
(Figureduring
8) and the tran
sitional process. Similar
the active power (Figure 9). behavior is shown by the generator speed (Figure 8) and the ac
tive power (Figure 9).
FOR PEER REVIEW -1.175 11 of 30
Energies 2021, 14, 5644 11 of 29
-1.180
-1.185
-1.190
-1.175
[rad] angle [rad]
-1.195
-1.180
-1.200
-1.185
-1.205
-1.190
-1.210
Rotor angleRotor
-1.195
-1.215
-1.200
-1.220
-1.205
-1.225
-1.210
-1.230
-1.215 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
-1.220 Time [s]
-1.225 T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
-1.230
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Figure 7. Gen1 rotor angle response on a small disturbance with PSS parameters K = 10, and T = Ti,1
Time [s]
= 0.2 s–0.6 s.
T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
Visual observation of the mentioned figures implies the most favorable time-domain
Figure
Figure 7.
system Gen1rotor
Gen1 rotor
performance angle
angle response
response
measures on PSS
onwith adisturbance
a small small disturbance
gain-K with
= 10PSS with PSSconstant(s)-T
parameters
and time parameters
K = 10, and T =KT=
=i ,110, and
T=4,10.2–0.6
= 0.6Ts.=s.Ti,1
= 0.2 s–0.6 s.
1.0004
Visual observation of the mentioned figures implies the most favorable time-domain
system1.0003
performance measures with PSS gain-K = 10 and time constant(s)-T = T4,1 = 0.6 s.
[rads-1]
1.0002
1.0001
1.0004
[radsspeed
1.0000
1.0003
-1]
0.9999
1.0002
Generator
0.9998
1.0001
Generator speed
0.9997
1.0000
0.9996
0.9999
0.9995
0.9998
0.9994
0.9997
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
0.9996
Time [s]
0.9995
T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
0.9994
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Figure 8. Gen1
Figure speed
8. Gen1 speedresponse ona asmall
response on small disturbance
disturbance with
with PSS PSS parameters
parameters K T= =10,
K = 10, and Ti ,and T = Ti,
1 = 0.2–0.6 s. 1 = 0.2
Time [s]
s–0.6 s. Visual observation of the mentioned figures implies the most favorable time-domain
T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
system performance measures with PSS gain-K = 10 and time constant(s)-T = T4,1 = 0.6 s.
Table 4 shows results of the weight functions calculated following the proposed
Figure 8. Gen1 speedalgorithm
responsefor
onthe first cluster.
a small disturbance with PSS parameters K = 10, and T = Ti,1 = 0.2
s–0.6 s.
Energies 14, 5644
2021,2021,
Energies 14, x FOR PEER REVIEW 12 of 30 12 of 29
465
460
455
Active power [MW]

450
445
440
435
430
425
420
415
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time [s]
T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
Figure 9. Gen1 active power response on a small disturbance with PSS parameters K = 10, and T =
Figure 9. Gen1 active power response on a small disturbance with PSS parameters K = 10, and
Ti,1 = 0.2 s–0.6 s.
T = Ti ,1 = 0.2–0.6 s.
Table 4 shows results of the weight functions calculated following the proposed al-
Table 4. Weight
gorithm for the functions for K = 10, T = Ti,1 = 0.2–0.6 s (cluster 1).
first cluster.
i 4. Weight
Table Ti,1functions
[s] for KΓζi Γωis–0.6 s (cluster
= 10, T = Ti,1 = 0.2 Γδi1). ΓPi Γ1 (i)
i0 Ti,1 [s]0.2 Γζi 0.094299 Γωi0.999995596 Γ0.989720196

δi Γ0.936069374
Pi Γ3.02008453797
1(i)
1 0.3 0.141835 0.999996902 0.991708475 0.955296445 3.08883715433
0 0.2 0.094299 0.999995596 0.989720196 0.936069374 3.02008453797
2 0.4 0.214872 0.999997748 0.993060487 0.968084688 3.17601446330
1 0.3 0.141835 0.999996902 0.991708475 0.955296445 3.08883715433
3 0.5 0.507928 0.999998262 0.993814845 0.976163024 3.47790431049
24 0.4 0.6 0.214872 0.999997748
0.425562 0.993060487
0.999998447 0.968084688
0.993415907 3.17601446330
0.980405629 3.39938212468
3 0.5 0.507928 0.999998262 0.993814845 0.976163024 3.47790431049
4 0.6 0.425562 0.999998447 0.993415907 0.980405629 3.39938212468
Except for Γδi , the time-domain weight functions-Γωi and ΓPi increase with increasing
the PSS time
Except forconstant(s)-T
Γδi, the time-domaini,1 . However, the system
weight functions-Γ performance
ωi and weight
ΓPi increase with function-Γ1 (i)
increasing
obtained maximum
the PSS time for TΓi,11 (3),
constant(s)- which the
. However, corresponds to PSS parameters
system performance K = 10 and
weight function–Γ T = T3,1 =
1(i) ob-
tained
0.5 s. Since Γ1 (4) <for
maximum Γ1Γ(3),
1(3),awhich corresponds
new cluster to 2)
(cluster PSSofparameters
five samples K =of
10PSS
andtime
T = Tconstant(s)-T
3,1 = 0.5
i,2
s. Since
will not Γbe
1(4) < Γ1(3), T
created, a 3,1
newwillcluster (cluster for
be selected 2) of five
the samples
PSS of PSS time constant(s)-Ti,2
time constant(s)-T:
will not be created, T3,1 will be selected for the PSS time constant(s)-T:
max Γ
max (i)==Γ1Γ(3)
Γ11(i) →→
1 (3) T3,1T=3,1 T =s 0.5 s
T ==0.5
The weight
The weightfunctions
functionsfrom Table
from 1 are
Table illustrated
1 are in Figures
illustrated 10–14.10–14.
in Figures
Γζ vs PSS time constant(s)- T

0.6
0.5
0.4
0.3
0.2
0.1
0
NO PSS T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
Figure 10. Γζi for PSS parameters K = 10, T = Ti,1 = 0.2–0.6 s.

Γω vs PSS time constant(s)- T

0.9999990
0.9999985
0.9999980
0.9999975
0.9999970
0.3
0.1
0.2
0
Energies 2021, 14, 5644 0.1 NO PSS T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s 13 of 29
0
NO PSS T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
Γω vs
Figure 10. Γζi for PSS parameters
PSS time constant(s)- T
K = 10, T = Ti,1 = 0.2–0.6 s.
0.9999990
0.9999985 Γω vs PSS time constant(s)- T
0.9999990
0.9999980
0.9999985
0.9999975
0.9999980
0.9999970
0.9999975
0.9999965
0.9999970
0.9999960
0.9999965
0.9999955
0.9999960
0.9999950
0.9999955
0.9999945
0.9999950
0.9999945
0.9999940
0.9999940 T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
Figure11.
11.ΓΓωifor
forPSS
PSSparameters
parametersKK==10,
10,TT== TTi,, 1 =
= 0.2–0.6 s.
Figure ωi i 1 0.2–0.6 s.
Figure 11. Γωi for PSS parameters K = 10, T = Ti,1 = 0.2–0.6 s.
Γ vs PSS time consant(s)- T

δ PSS time consant(s)- T
0.995
0.995
Γδ vs
0.994
0.994
0.993
0.993
0.992
0.992
0.991
0.991
0.99
0.99
0.989
0.989
0.988
0.988
0.987
0.987 T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
Energies 2021, 14, x FOR PEER REVIEW T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.614s of 30
Figure 12. Γδi for PSS parameters K = 10, T = Ti,1 = 0.2–0.6 s.
Figure 12. Γδi for PSS parameters K = 10, T = Ti ,1 = 0.2–0.6 s.
Figure 12. Γδi for PSS parameters K = 10, T = Ti,1 = 0.2–0.6 s.
ΓP vs PSS time constant(s) -T

0.99
0.98
0.97
0.96
0.95
0.94
0.93
0.92
0.91
T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
Figure 13.
13. ΓΓPi for PSS parameters K = 10, T = Ti,1 = 0.2–0.6 s.
Figure Pi for PSS parameters K = 10, T = Ti ,1 = 0.2–0.6 s.
Γ1 vs PSS time constant(s) -T

3.600
3.500
3.400
3.300
0.93
0.92
0.91
T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
Energies 2021, 14, 5644 14 of 29
Figure 13. ΓPi for PSS parameters K = 10, T = Ti,1 = 0.2–0.6 s.
Γ1 vs PSS time constant(s) -T

3.600
3.500
3.400
3.300
3.200
3.100
3.000
2.900
2.800
2.700
T = 0.2 s T = 0.3 s T = 0.4 s T = 0.5 s T = 0.6 s
Figure 14.
14. ΓΓ1 for PSS parameters K = 10, T = Ti,1 = 0.2–0.6 s.
Figure 1 for PSS parameters K = 10, T = Ti ,1 = 0.2–0.6 s.
5.2. PSS
5.2. PSS Gain
GainTuning
Tuning
Performed time-domain
Performed time-domain simulations imply
simulations the insignificant
imply influence
the insignificant of the (small)
influence of the (small)
disturbance duration on the PSS time constant tuning. However, the
disturbance duration on the PSS time constant tuning. However, the PSS PSS gain tuning
gain will
tuning will
be studied for different durations of the same (small) disturbance (3-phase short-circuit)
be studied for different durations of the same (small) disturbance (3-phase short-circuit)
on BUS 4.
on BUS 4.
5.3. Case Study 1–Disturbance Duration 30 ms
5.3. Case Study 1—Disturbance Duration 30 ms
Following the proposed algorithm, the PSS gain is varied from the initial value K
Following
= 𝐾𝑖𝑛𝑖𝑡 = 10. Tablethe proposed
5 presents algorithm,
the dominant the PSS
oscillatory gain
pole is varied
s-domain from for
measures thethe clus- value K
initial
Kinit
=ter = 10. Table 5 presents the dominant oscillatory pole s-domain measures for the cluster
where the PSS initial value is decreased by 1 in each subsequent iteration (lower clus-
where
ter). the PSS initial value is decreased by 1 in each subsequent iteration (lower cluster).
Table5.5. Dominant
Table Dominant oscillatory
oscillatorypole
poleforfor
PSS parameters
PSS T=T
parameters T1 == T
T13 ==0.5
T3 s,= K0.5
i,ln =s,6–10,
Ki,ln n= =6–10,
1 (lower
n = 1 (lower
cluster 1).
cluster 1).
i Ki,l1 α [s−1] ±jω [rad s−1] ξ

i Ki,l1 α [s−1 ] ±jω [rad s−1 ] ξ
NO PSS −0.602911288 ±8.600128618 0.069933275
NO PSS −0.602911288 ±8.600128618 0.069933275
0 6 −2.044897210 ±9.123657619 0.218705293
1 7 −2.323979821 ±9.251852650 0.243622413
2 8 −2.615513529 ±9.361915009 0.269074414
3 9 −2.536071719 ±4.118411476 0.524347042
4 10 −2.442664922 ±4.142534779 0.50792818
Table 6 presents the dominant oscillatory pole s-domain measures for the cluster where
the PSS initial value is increased by 1 in each subsequent iteration (upper cluster).
Table 6. Dominant oscillatory pole for PSS parameters T = 0.5 s, Ki,um = 10–14, m = 1 (upper cluster 1).
i Ki,u1 α [s−1 ] ±jω [rad s−1 ] ξ

0 10 −2.442664922 ±4.142534779 0.50792818
1 11 −2.350630618 ±4.154693103 0.492426473
2 12 −2.262346344 ±4.156618791 0.478053895
3 13 −2.179202113 ±4.150363212 0.464877617
4 14 −2.101785372 ±4.137873414 0.452866896
Increasing the PSS gain in the lower cluster 1 resulted in the initial dominant system
oscillatory pole (system without the PSS) better damping ratio and movement to the left
side of the S-plane until a complex pair of the eigenvalues obtained for PSS gain K3,l1 = 9
Energies 2021, 14, 5644 15 of 29
became a new dominant oscillatory pole. However, the further increase of the PSS gain in
lower cluster 1 resulted in the dominant system oscillatory pole decreased damping ratio
and movement to right-side on the S-plane, which was continued in the first upper cluster 1
too (Figure 15).
Figure
Figure 15. Dominantoscillatory
15. Dominant oscillatory pole
pole before
before the the
PSS PSS implementation-K
implementation-K 0,0 and 0,0
and
after theafter the PSS imple-
PSS imple-
mentation with T = 0.5 s, PSS gain increase in the lower cluster 1-K
mentation with T = 0.5 s, PSS gain increase in the lower cluster 1- Ki,ln = 6–10 = 6–10 and the upper
i,lnand the upper cluster 1 cluster
1KKi,um =
i,um 10–14.
= 10–14.
FigureOn15.the
Dominant oscillatory
other hand, pole before
a continual the PSSof
increase implementation-K
the PSS gain in 0,0 and
bothafter thecluster
lower PSS imple-
1 and upper
mentation with T = 0.5 s, PSS gain increase in the lower cluster 1- Ki,ln = 6–10 and the upper cluster 1
cluster-1.185
1 resulted in better time-domain measures of the system performance (Figures 16–18).
Ki,um = 10–14.
-1.190
-1.195
-1.200
[rad] angle [rad]
-1.185
-1.205
-1.190
-1.210
-1.195
-1.215
Rotor angleRotor
-1.200
-1.220
-1.205
-1.225
-1.210
-1.230
-1.215
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
-1.220
Time [s]
-1.225
K=6 K=7 K=8 K=9 K = 10
-1.230
K0 = 11 1 2 3 K = 412 5 6 K7 = 13 8 9 10 K =11
14 12 13 14 15
Time [s] with PSS parameters T = 0.5 s, the PSS
Figure 16. Gen1 rotor angle response on a small disturbance
gain increaseK in
= 6the lower cluster
K =1-K
7 i,ln = 6–10 and Kthe
= 8upper cluster 1 KK=i,um9 = 10–14. K = 10
K = 11 K = 12 K = 13 K = 14
Figure 16. Gen1 rotor angle response on a small disturbance with PSS parameters T = 0.5 s, the PSS
gain increase in the lower cluster 1-Ki,ln = 6–10 and the upper cluster 1 Ki,um = 10–14.
gain increase in the lower cluster 1-Ki,ln = 6–10 and the upper cluster 1 Ki,um = 10–14.
Energies
Energies 2021,
2021, 14,14, x FOR PEER REVIEW
5644 17 of1630of 29
1.0003
1.0003
1.0002
1.0002
1.0001
1.0001
Generator speed [rads-1]

1.0000
1.0000
0.9999
0.9999
0.9998
0.9998
0.9997
0.9997
0.9996
0.9996
0.9995
0.9995
0 0 1 1 2 2 3 3 4 4 5 5 6 6 77 88 99 10
10 11
11 12
12 13
13 14
14 15
15
Time[s][s]
Time
K=6 K=7 K=8 K=9 K = 10
K=6 K=7 K=8 K=9 K = 10
K = 11 K = 12 K = 13 K = 14
K = 11 K = 12 K = 13 K = 14
Figure17.
17. Gen1
Gen1 speed
speed response on aasmall
smalldisturbance with PSS parameters T = 0.5 s, the PSSPSS
gaingain
Figure
Figure 17. Gen1
increase speed
in the
response
lowerresponse
cluster 1-Koni,lnon
a=small disturbance
6–10, disturbance
and the upper
with
with PSSPSS
cluster
parameters
1parameters T =T0.5
Ki,um = 10–14.
= 0.5 s, the
s, the PSS gain
increase
increase in the
in the lower
lower cluster
cluster 1-K1-K
i,ln =i,ln = 6–10,
6–10, andand the upper
the upper cluster
cluster 1 K=i,um
1 Ki,um = 10–14.
10–14.
465
465
460
Active power [MW]
460
455
Active power [MW]
455 450
450 445
445 440
440 435
435 430
430 425
425 420
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
420
0 1 2 3 4 5 6 7
Time8 [s] 9 10 11 12 13 14 15
K=6 K=7 K=8 K=9 K = 10
Time [s]
K = 6K = 11 K =K7= 12 K K= =8 13 KK==914 K = 10
K = 11 K = 12 K = 13 K = 14
Figure 18. Gen1 active power response on a small disturbance with PSS parameters T = 0,5 s, the
PSS gain increase in the lower cluster 1-Ki,ln = 6–10 and the upper cluster 1 Ki,um = 10–14.
Figure
Figure 18. 18. Gen1
Gen1 active
active power
power responseonona asmall
response smalldisturbance
disturbance with parameters TT==0,5
with PSS parameters 0,5s,s,the
thePSS
PSSgain
gainincrease
increase in
in the
the lower
lower cluster
cluster 1-K
1-K i,ln== 6–10
6–10 and
and the
the upper
upper cluster
cluster 11KK i,um =
Tables 7 and 8 give an overview of the proposed weight functions. The damping ratio
i,ln i,um = 10–14.
10–14.
of the dominant oscillatory mode reached the maximum in the lower cluster 1 for the PSS
gain
Tables
TablesK3,l17=and
7 and
9. Weight
8 give
8 give an an overview
overview
functions
of the
of the
representing
proposed
proposed weight
weight
the generator
functions.
functions.
speed-Γ ωi andThe The damping
damping
active
ratio
ratio
power-Γ Pi
ofincrease
of the the dominant
dominant theoscillatory
with oscillatoryPSS gain modemode reached
reached
increase. The the the
weight maximum
maximum function in the
inrepresenting
the lower lower cluster
cluster
the 1 for
1 for thethe
PSS
rotor-angle-Γ PSS
δi
gain
gain K K = 9. = 9.
Weight Weight functions
functions representing
representing the the generator
generator speed-Γ
speed-Γ
increases with the PSS gain increase before it reaches the maximum in upper cluster 1, for Pi
3,l1 3,l1 ωi and
ωi and active
active power-Γ
power-Γ Pi
increase
increase 11.with
K1,u1 =with thethe PSSPSSgaingain increase.
increase. TheThe weight
weight function
function representing
representing thethe rotor-angle-Γ
rotor-angle-Γ δi δi
increases
increasesSince with
withΓthe the
i,l1(i)PSS
PSS gain
gain the
reached increase
increase before
before
maximum it reaches
forit Γreaches
1,l1(1) and
the
theΓ4,l1 maximum
maximum
(4) < Γ3,l1(3), in upper
in according cluster
upper cluster 1,
it for
1, for
to (18),
K =
K1,u1follows:
= 11.
1,u1 11.
Since
Since Γi,l1Γ(i)i,l1reached
(i) reachedthe the maximum
maximum for for Γ(1)
𝐾𝑙 Γ=1,l1𝐾
(1) and
1,l1and Γ4,l1<(4)
Γ4,l1(4) Γ3,l1
Γ3,l1<(3), (3), according
according to it
to (18), (18),
it follows: 1,𝑙1 = 9
follows:
Further, Γi,u1(i) reached the maximum Kl = for K1,l1 =(0)
Γ0,u1 9 and Γ4,u1(4) < Γ3,u1(3), according to
(15), it follows: 𝐾𝑙 = 𝐾1,𝑙1 = 9
Further, Γi,u1 (i) reached the maximum for Γ0,u1 (0) and Γ4,u1 (4) < Γ3,u1 (3), according to
Further,
(15), it follows:Γi,u1(i) reached the maximum for Γ0,u1(0) and Γ4,u1(4) < Γ3,u1(3), according to
(15), it follows: Ku = K0,u1 = 10
𝑚𝑎𝑥𝛤𝑖,𝑢𝑚 (𝑖 ) < 𝑚𝑎𝑥𝛤𝑖,𝑙𝑛 (𝑖 )
Which implies that the PSS gain K is chosen as:
Energies 2021, 14, 5644 17 of 29
𝐾 = 𝐾𝑙 = 𝐾1,𝑙1 = 9
Further, according to (14):

Table 7. Weight functions for PSS parameters T = 0.5 s, Ki,ln = 6–10, n = 1 (lower cluster 1), t = 30 ms.
maxΓ (i ) < maxΓ (i )
i Ki,l1 Γζi,l1 Γωi,l1 Γi,um
δi,l1
i,ln
ΓPi,l1 Γi,l1(i)
0 10 0.507928 0.9999982622
Which implies that the PSS0.9938148445
gain K is chosen 0.9764722440
as: 3.8037042787
1 9 0.524347 0.9999981598 0.9937118105 0.9749639780 3.8180089833
K = Kl = K1,l1 = 9
2 8 0.269074 0.9999980102 0.9934398148 0.9729530840 3.5597830173
3 7 0.243622 0.9999978261
Table 7. Weight 0.9930707387
functions for PSS parameters T = 0.5 s,0.9705117330 3.5307066213
Ki,ln = 6–10, n = 1 (lower cluster 1), t = 30 ms.
4 6 0.218705
i K 0.9999976055
Γ 0.9926580155
Γ Γ 0.9675305460 Γ 3.5014016407 Γ (i)
i,l1 ζi,l1 ωi,l1 δi,l1 Pi,l1 i,l1
0 10 0.507928 0.9999982622 0.9938148445 0.9764722440 3.8037042787

Table 8. Weight functions1 for PSS
9 parameters
0.524347 T = 0.9999981598
0.5 s, Ki,um = 10–14, m = 1 (upper
0.9937118105 cluster 1); t =3.8180089833
0.9749639780 30 ms.
2 8 0.269074 0.9999980102 0.9934398148 0.9729530840 3.5597830173
i Ki,u1 Γζi,u1
3 7 Γ0.243622
ωi,u1 Γδi,u1 0.9930707387ΓPi,u1 0.9705117330Γi,u1(i)
0.9999978261 3.5307066213
4 6 0.218705 0.9999976055 0.9926580155 0.9675305460 3.5014016407
0 10 0.507928 0.9999982622 0.9938148445 0.976472244 3.80370427871
1 11 0.492426 0.9999983415 0.9938523457 0.977678002 3.78984782950
Table 8. Weight functions for PSS parameters T = 0.5 s, Ki,um = 10–14, m = 1 (upper cluster 1); t = 30 ms.
2 12 0.478054 0.9999984003 0.9938187092 0.978677412 3.77677422066
3 13 0.464878
i Γζi,u1
Ki,u10.9999984437 Γωi,u1
0.9937383989 Γδi,u1
0.979498216ΓPi,u1 3.76461208143 Γi,u1 (i)
4 14 0.452867
0 10 0.9999984808
0.507928 0.9936327709
0.9999982622 0.980223698
0.9938148445 3.75346307772
0.976472244 3.80370427871
1 11 0.492426 0.9999983415 0.9938523457 0.977678002 3.78984782950
2 12 0.478054 0.9999984003 0.9938187092 0.978677412 3.77677422066
It can be noticed 3that the
13 overall weight
0.464878 𝛤𝑖,𝑢𝑚 and 𝛤𝑖,𝑙𝑛0.979498216
functions0.9937383989
0.9999984437 are mainly3.76461208143
influ-
4 14 0.452867 0.9999984808 0.9936327709 0.980223698
enced by the s-domain system performance measures. However, the weight functions re- 3.75346307772
lated to time-domain system performance measures increase with the PSS gain increase,
It can be and
resulting in better over-shoots noticed that the overallofweight
under-shoots functions Γi,um
the simulated and Γi,lnafter
variable are mainly influenced
initiating
by the s-domain system performance measures. However, the weight functions related to
a (small) disturbance. High amplitudes of the active power oscillations can cause the loss
time-domain system performance measures increase with the PSS gain increase, resulting
of load and trigger the protection
in better devices
over-shoots in the system.
and under-shoots Decreased
of the simulatedamplitude
variable afterofinitiating
the activea (small)
power oscillations obtained
disturbance. with
Highthe PSS gain
amplitudes of increase supports
the active power the idea
oscillations canof tuning
cause PSSof load
the loss
and trigger
taking into consideration the protection
both devices in
the s-domain the system.
and Decreasedsystem
time-domain amplitude of the active power
performance
oscillations obtained with the PSS gain increase supports the idea of tuning PSS taking into
measures.
consideration both the s-domain and time-domain system performance measures.
The weight functions Thefrom
weightTables 7 and
functions 8 Tables
from are presented
7 and 8 arein Figuresin19–23.
presented FiguresFor theFor
19–23. sakethe sake
of better visibility, the results for both lower cluster 1 and upper cluster 1 are shown
of better visibility, the results for both lower cluster 1 and upper cluster 1 are shownon a on a
single diagram. single diagram.
Γζ vs PSS gain
0.600
0.500
0.400
0.300
0.200
0.100
0.000
NO PSS K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
Figure 19. ΓζFigure 19. ΓΓ
(Γζi,l1 and from
(Γζi,l1
ζ ζi,u1 Γζi,u1 from
and Table 7 and Table
Tables 8) 8)
7 and forforPSS
PSSparameters
parameters T T = 0.5
= 0.5 s, Ks,=K = 6–14.
6–14.
4,
1, xx14,
FOR PEER
x FOR PEERREVIEW
REVIEW 1919
of of
30 30
4, FOR PEER REVIEW
Energies 2021, 14, 5644 19 of 30 18 of 29
ΓΓωω vs PSS gain

gain
0.999999
0.999999 ω vs PSS gain
0.999999
0.999998
0.999998
0.999998
0.999998
0.999998
0.999998
0.999998
0.999998
0.999998
0.999998
0.999998
0.999998
0.999998
0.999998
0.999998
0.999997
0.999997
0.999997
0.999997
0.999997
0.999997
0.999997
0.999997
0.999997 K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
FigureFigure
20. Γω 20.
(Γωi,l1 and Γ from Table 7 and Table 8) for PSS parameters T = 0.5 s, K = 6–14.
Γω (Γωi,l1 and Γωi,u1 from Tables 7 and 8) for PSS parameters T = 0.5 s, K = 6–14.
ωi,u1
Figure 20. Γω (Γωi,l1 and Γωi,u1 from Table 7 and Table 8) for PSS parameters T = 0.5 s, K = 6–14.
Γ vs PSS gain
0.9940 ΓΓδδ vs PSS gain
0.9940
0.9940
0.9938 δ vs PSS gain
0.9938
0.9936
0.9938
0.9936
0.9934
0.9936
0.9934
0.9932
0.9934
0.9932
0.9930
0.9932
0.9930
0.9928
0.9930
0.9928
0.9926
0.9928
0.9926
0.9924
0.9926
0.9924
0.9922
0.9924
0.9920
0.9922
0.9922
K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
0.9920
0.9920
Figure 21. Γδ K=6
(Γδi,l1 and ΓK=7
K=6 K=7
K=8 7 andK=9
δi,u1 from Table
K=8 Table 8) K=10
K=9 K=10
K=11
for PSS parameters
K=11
K=12
T = 0.5 s, KK=13
K=12 = 6–14. K=14
K=13 K=14
Figure 21.Figure
Γδ (Γδi,l1
21.and
Γδ (ΓΓ δi,u1 from Table 7 and Table 8) for PSS parameters T = 0.5 s, K = 6–14.
Γδi,u1 parameters TT == 0.5
Figure 21. Γδ (Γδi,l1 and δi,u1 and
Γδi,l1 from Tablefrom Tables 7 and
7 and 8)8)for
forPSS
PSSparameters 0.5s,s,KK==6–14.
ΓP Table
vs PSS gain 6–14.
0.985
ΓΓP vs PSS gain
0.985 P vs PSS gain
0.985
0.980
0.980
0.975
0.980
0.975
0.970
0.975
0.970
0.965
0.970
0.965
0.960
0.965
K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
0.960
0.960
Figure 22. ΓP (ΓPi,l1 and ΓPi,u1 from Table 7 and Table 8) for PSS parameters T = 0.5 s, K = 6–14.
K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
Due
Figure 22. ΓPto protection
(ΓPi,l1 and ΓPi,u1activation
from Tabletime and
7 and circuit-breaker
Table opening T
8) for PSS parameters time,
= 0.5ins,the
K = real
6–14.power
Figure 22. ΓP (ΓPi,l1
Figure 22. Γ
and (ΓΓ from
Pi,u1 and Γ Table 7
from and Table
Tables 7 8)
and for
8) PSS
for parameters
PSS parameters
system, disturbance duration lasts longer than the one simulated in this case study,
P Pi,l1 Pi,u1 T = 0.5
0.5 s,
s, K
K == 6–14.hence
6–14.
theDue
proposed algorithm
to protection performance
activation time will
and be examined for opening
circuit-breaker longer disturbance
time, in thedurations.
real power
Due to protection activation time and circuit-breaker opening time, in the real power
system, disturbance duration lasts longer than the one simulated in this case study, hence
system, disturbance duration lasts longer than the one simulated in this case study, hence
the proposed algorithm performance will be examined for longer disturbance durations.
the proposed algorithm performance will be examined for longer disturbance durations.
5.4. Case Study 2-Disturbance Duration 150 ms
The
observations from the aforementioned case study will be supplemented with the 19 of 29
Energies 2021, 14, 5644
OR PEER REVIEW new time-domain system performance measures for the disturbance duration of 15020ms. of 30
The simulation results for the lower cluster 1 and the upper cluster 1 are presented in Figures
23–25.
5.4. Case-1.14
Study 2-Disturbance Duration 150 ms
-1.16
The observations from the aforementioned case study will be supplemented with the
-1.18
new time-domain system performance measures for the disturbance duration of 150 ms.
Rotor angle [rad]
The simulation
-1.20 results for the lower cluster 1 and the upper cluster 1 are presented in Figures
23–25. -1.22
-1.14-1.24
-1.26
-1.16
-1.28
-1.18
Rotor angle [rad]
-1.30
-1.20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
-1.22
Time [s]
K=6 K=7 K=8 K=9 K = 10
-1.24
K = 11 K = 12 K = 13 K 0 14
-1.26
Figure 23. Gen1 rotor angle response on a small disturbance with PSS parameters T = 0.5 s, the PSS gain increase in the
lower cluster
-1.281-Ki,ln = 6–10, and the upper cluster 1 Ki,um = 10–14.
gain increase in the lower cluster 1-Ki,ln = 6–10, and the upper cluster 1 Ki,um = 10–14.
-1.30 Due to protection activation time and circuit-breaker opening time, in the real power
1.0015
0 1 2system,
3 disturbance
4 5 duration
6 7lasts longer
8 9than 10 11 simulated
the one 12 13 14case15
in this study, hence
Time [s] will be examined for longer disturbance durations.
the proposed algorithm performance
K=6
1.0010 K=7 K=8 K=9 K = 10
K = 11 K = 12 K = 13 K 0 14
1.0005 5.4. Case Study 2—Disturbance Duration 150 ms

The observations from the aforementioned case study will be supplemented with the
1.0000
Figure 23. Gen1 rotor new time-domain
angle response on system performance
a small disturbancemeasures for the
with PSS disturbanceT duration
parameters of 150
= 0.5 s, the PSSms. The
gain increase simulation results for the lower cluster 1 and the upper cluster
0.9995in the lower cluster 1-Ki,ln = 6–10, and the upper cluster 1 Ki,um = 10–14.1 are presented in Figures 23–25.
0.9990
1.0015
0.9985
1.0010
0.9980
1.0005 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time [s]
1.0000 K=6 K=7 K=8 K=9 K = 10
0.9995 K = 11 K = 12 K = 13 K = 14
Figure
0.9990 24. Gen1 speed response on a small disturbance with PSS parameters T = 0.5 s, the PSS gain
increase in the lower cluster 1-Ki,ln = 6–10, and the upper cluster 1 Ki,um = 10–14.
0.9985
0.9980
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time [s]
K=6 K=7 K=8 K=9 K = 10
K = 11 K = 12 K = 13 K = 14
Figure
Figure 24. 24.
Gen1 Gen1
speedspeed response
response on a small on disturbance
a small disturbance with PSSTparameters
with PSS parameters T =gain
= 0.5 s, the PSS 0.5increase
s, the PSS gain
in the lower
increase
cluster 1-Ki,lnin the lower
= 6–10, cluster
and the upper 1-K i,ln =
cluster 1 6–10,
Ki,um =and the upper cluster 1 Ki,um = 10–14.
10–14.
x FOR PEER REVIEW 21 of 30
Energies 2021, 14, 5644 20 of 29
490
470
Active power [MW]
450
430
410
390
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time [s]
K=6 K=7 K=8 K=9 K = 10
K = 11 K = 12 K = 13 K = 14
Figure
Figure 25. active
25. Gen1 Gen1power
activeresponse
power onresponse on a smallwith
a small disturbance disturbance with T
PSS parameters PSS parameters
= 0.5 T =increase
s, the PSS gain 0.5 s, the
in the
lowerPSS gain
cluster 1-Kincrease
i,ln = in
6–10, the
and lower
the cluster
upper cluster1-K
1 K i,ln ==6–10,
i,um
and
10–14. the upper cluster 1 K i,um = 10–14.
Gen1 rotor
Gen1 rotor angle settling angle
time and settling time andundershoot
the maximal the maximal undershoot increase
increase with with the increase
the increase
of the PSS gain in both upper cluster 1 and lower cluster 1 in this study case. The active
of the PSS gain in both upper cluster 1 and lower cluster 1 in this study case. The active
power and the generator speed undershoot decrease with the PSS gain increase while
power and the generator speed
the settling undershoot
times decrease
remain almost with the
the same. PSS9 gain
Tables increase
and 10 give anwhile the of the
overview
settling times remain almost
proposed the same.
weight Tables 9 and 10 give an overview of the proposed
functions.
weight functions.
i Ki,l1 Γζi,l1 Γωi,l1 Γδi,l1 ΓPi,l1 Γi,l1 (i)
i Ki,l1 Γ0ζi,l1 10 Γωi,l1
0.507928 Γδi,l1 0.8671126304
0.9999717530 ΓPi,l1 0.6144343930Γi,l1(i)2.9894469569
1 9 0.524347 0.9999706671 0.8723687207 0.5977923551 2.9944787854
0 10 0.507928
2 8 0.9999717530
0.269074 0.8671126304
0.9999693749 0.6144343930
0.8766792552 2.9894469569
0.5791968020 2.7249198464
1 9 0.524347
3 7 0.9999706671
0.243622 0.8723687207
0.9999679219 0.5977923551
0.8801760142 2.9944787854
0.5594264612 2.6831928107
4 6 0.218705 0.9999662783 0.8826227186 0.5379485031 2.6392427924
2 8 0.269074 0.9999693749 0.8766792552 0.5791968020 2.7249198464
3 7 0.243622 0.9999679219 0.8801760142 0.5594264612 2.6831928107
4 6 0.218705
Table 10. 0.9999662783
Weight 0.8826227186
functions for PSS parameters T = 0.5379485031 2.6392427924
0.5 s, Ki,um = 10–14, m = 1 (upper cluster 1);
t = 150 ms.
Table 10. Weight functions
i for PSS parameters
Ki,u1 Γζi,u1 T = Γ0.5 s, Ki,um = 10–14,
Γδi,u1 m = 1 (upper cluster 1); t = 150
ΓPi,u1 Γi,u1 (i)
ωi,u1
ms.
0 10 0.507928 0.9999717530 0.8671126304 0.6144343930 2.9894469569
1 11 0.492426 0.9999726603 0.8609451785 0.6293255630 2.9826698748
i Ki,u1 Γζi,u1
2 12
Γωi,u1
0.478054
Γδi,u1 ΓPi,u1 0.6432479360Γi,u1(i)2.9752607011
0.9999734435 0.8539854267
0 10 0.507928
3 13 0.9999717530
0.464878 0.8671126304
0.9999741856 0.6144343930
0.8468810809 2.9894469569
0.6563139020 2.9680467857
4 14 0.452867 0.9999748504 0.8395454183 0.6686698430 2.9610570069
1 11 0.492426 0.9999726603 0.8609451785 0.6293255630 2.9826698748
2 12 0.478054 0.9999734435 0.8539854267 0.6432479360 2.9752607011
Speed weight function Γω as well as active power weight function ΓP increase with
3 13 0.464878 0.9999741856 0.8468810809 0.6563139020 2.9680467857
the PSS gain increase. As in the previous study case:
4 14 0.452867 0.9999748504
Γ (i) reached the maximum 0.8395454183
for Γ (1) and0.6686698430 2.9610570069
Γ (4) < Γ (3), and according to (18),
i,l1 1,l1 4,l1 3,l1
it follows:
Speed weight function Γω as well as active powerKl = K weight
1,l1 = 9 function ΓP increase with
the PSS gain increase. As in the previous study case:
Γi,l1(i) reached the maximum for Γ1,l1(1) and Γ4,l1(4) < Γ3,l1(3), and according to (18), it
follows:
𝐾𝑙 = 𝐾1,𝑙1 = 9
Γi,u1(i) reached the maximum for Γ0,u1(0) and Γ4,u1(4) < Γ3,u1(3), and according to (15), it
follows:
Energies 2021, 14, 5644 21 of 29
Γi,u1 (i) reached the maximum for Γ0,u1 (0) and Γ4,u1 (4) < Γ3,u1 (3), and according to (15),
it follows:
Ku = K0,u1 = 10
maxΓi,um (i ) < maxΓi,ln (i )
Which implies that the PSS gain 𝑚𝑎𝑥𝛤 is: (i) < 𝑚𝑎𝑥𝛤𝑖,𝑙𝑛 (i)
K𝑖,𝑢𝑚
𝑚𝑎𝑥𝛤𝑖,𝑢𝑚 (i) < 𝑚𝑎𝑥𝛤𝑖,𝑙𝑛 (i)
Which implies that the PSS gain K is:
Which implies that the PSS gainK K=is:Kl = K1,l1 = 9
K = 𝐾𝑙 = 𝐾1,𝑙1 = 9
The difference between the maximal K = 𝐾𝑙 outputs
= 𝐾1,𝑙1 =of9the Γi,um (i) and Γi,ln (i) functions is
The difference between the maximal outputs of the Γi,um (i) and Γi,ln(i) functions is
smallerThethan in the previous
difference case
betweencase study,
thestudy,
maximal implying
outputs that the disturbance duration affects is
the
smaller than in the previous implying thatofthe
thedisturbance
Γi,um(i) andduration
Γ i,ln(i) functions
affects the
weight
smaller functions
than in related
the to the
previous time-domain
case study, system
implying performance
that the measures
disturbance duration andaffects
emphasize
the
weight functions related to the time-domain system performance measures and empha-
their
weightsignificance
functions in the PSS
related tuning process.system
to the As in the previous measures
case study, the weight
size their significance in the PSStime-domain
tuning process. As inperformance
the previous case study,and theempha-
weight
functions from Tables
size their significance 9 and 10 are illustrated in Figures 26–29.
functions from Tablesin the PSS
9 and tuning
10 are process.
illustrated in As in the26–29.
Figures previous case study, the weight
functions from Tables 9 and 10 are illustrated in Figures 26–29.
0.600
0.600
0.500
0.500
0.400
0.400
0.300
0.300
0.200
0.200
0.100
0.100
0.000
0.000 NO PSS K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
NO PSS K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
Figure26.
26. ΓΓζ (Γζi,l1 and Γζi,u1 from Table 9 and Table 10) for PSS parameters T = 0.5 s, K = 6–14.
Figure ζ (Γζi,l1 and Γζi,u1 from Tables 9 and 10) for PSS parameters T = 0.5 s, K = 6–14.
Figure 26. Γζ (Γζi,l1 and Γζi,u1 from Table 9 and Table 10) for PSS parameters T = 0.5 s, K = 6–14.
Γω vs PSS gain
0.999976 Γω vs PSS gain
0.999976
0.999974
0.999974
0.999972
0.999972
0.999970
0.999970
0.999968
0.999968
0.999966
0.999966
0.999964
0.999964
0.999962
0.999962
0.999960
0.999960 K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
Figure27.
27. ΓΓωω (Γ and Γωi,u1 from Table 9 and Table 10) for PSS parameters T = 0.5 s, K = 6–14.
Figure ωi,l1 and Γωi,u1 from Tables 9 and 10) for PSS parameters T = 0.5 s, K = 6–14.
(Γωi,l1
s 2021, 14, x FOR PEER
Energies REVIEW
2021, 23 of 30
Energies 2021, 14,14, x FOR PEER REVIEW
5644 23 22
of of
3029
Γδ vs PSS gain
Γδ vs PSS gain
0.890 0.890 Γδ vs PSS gain
0.890
0.880 0.880
0.880
0.870 0.870
0.870
0.860 0.860
0.860
0.850 0.850
0.850
0.840 0.840
0.840
0.830 0.830
0.830
0.820 0.820
0.820
0.810 0.810
K=6 K=6
K=8 0.810K=7
K=7 K=9 K=8
K=10 K=9 K=11 K=10K=12 K=11 K=13 K=12 K=14 K=13 K=14
K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
Figure
Figure 28. Γδ (Γδi,l1 28.ΓΓδ (Γ
and Γ28.
Figure δi,u1 from
(Γδi,l1
Tableand
and9Γδi,u1
and from
Γδi,u1 Table
Table
from 10) 9 and
for
Tables Table
PSS 10)10)
forfor
parameters
9 and PSS
PSS = parameters
T parameters TT==0.5
0.5 s, K = 6–14. 0.5s,s,KK==6–14.
6–14.
Figure 28. Γδδ (Γδi,l1
δi,l1
and Γδi,u1 from Table 9 and Table 10) for PSS parameters T = 0.5 s, K = 6–14.
ΓP vs PSS gain
ΓP vs PSS gain
0.680
0.680 Γ vs PSS gain
P
0.680
0.660
0.660
0.660
0.640
0.640
0.640
0.620
0.620 0.620
0.600
0.600 0.600
0.580
0.580 0.580
0.560
0.560 0.560
0.540
0.540 0.540
0.520
0.520 0.520
0.500
0.500 0.500 K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
K=7 K=6 K=8
K=6 K=9
K=7 K=10 K=9
K=8 K=11 K=10K=12 K=11 K=13 K=12K=14 K=13 K=14
Figure 29. ΓP (ΓPi,l1 and Γ29.
Figure
Figure 29. Γfrom
Pi,u1 ΓP (Γ Table
(ΓPi,l1 and 9ΓPi,u1
and and Table
Γ from from10)
Table for
Tables PSS
9 and parameters
9 Table
and 10)for
10) T =parameters
forPSS
PSS 0.5 s, K = 6–14.
parameters TT== 0.5
0.5 s,
s, K
K== 6–14.
6–14.
P Pi,l1 Pi,u1
5.5. Case Study 3–Disturbance Duration-220 ms
5.5. Case Study 5.5.
5.5. Case
Case Study
3–Disturbance 3—Disturbance
StudyDuration-220
3–Disturbancems Duration-220msms
Duration-220
New time-domain system performance measures for the (small) disturbance dura-
New time-domain New
tion New
of time-domain
time-domain
system
220 system
system
ms areperformance
obtained performance
performance
measures
from measures
formeasures for
for
the presented
the simulations (small) the
the (small)disturbance
in (small)
disturbance
Figures disturbanceduration
dura-
30–32. dura-
tion of 220 ms oftion
are220 ofms
220are
obtained ms are obtained
obtained
from fromfrom the simulations
the simulations
the simulations presented presented
presented
in Figures in Figures
in30–32.
Figures 30–32.
30–32.
-1.10
-1.10 -1.10
-1.15
-1.15 -1.15
[rad]
-1.20
[rad]
Rotor angle [rad]
-1.20
angle
-1.20
angle
-1.25
Rotor
-1.25 -1.25
Rotor
-1.30
-1.30
-1.30
-1.35
-1.35 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
-1.35
0 1 2 3 4 5 6 Time 7 [s]8 9 10 11 12 13 14 15
0 1 2 3K = 6 4 5 6 K =7 8 9 TimeK10 8 11
= [s] 12 13 K = 9 14 15 K = 10
KK == 611 KKTime
== 712 [s] KK == 813 KK == 914 K = 10
K=6 K=7 K=8 K=9 K = 10
K = 11 K = 12 K = 13 K = 14
K = 11 Figure 30. Gen1 K = rotor
12 angle response K = 13 K = 14 with PSS parameters T = 0.5 s, the PSS
on a small disturbance
Figure 30. Gen1
gain increase rotor
in the angle
lower response
cluster 1-Ki,lnon a small
= 6–10, anddisturbance with PSS
the upper cluster 1 Ki,parameters
um = 10–14. T = 0.5 s, the PSS
Figure
Figure 30. Gen1 30. Gen1
rotor angle rotor angleonresponse
response
gain increase ainsmall on acluster
small 1-K
disturbance
the lower disturbance
with
i,ln = PSS
with PSSupper
6–10,parameters
and the parameters
T =cluster TK=i,um
0.5 s, 1the 0.5
PSS s,gain
the increase
= 10–14. PSS in the
lower clustergain increase
1-Ki,ln = 6–10,inand
thethe
lower cluster
upper 1-K1i,lnK=,6–10,
cluster i um = and the upper cluster 1 Ki,um = 10–14.
10–14.
Energies 2021, 14, 5644 23 of 29
14,
4, xxFOR
FORPEER
PEERREVIEW
REVIEW 24 of
24 of 30
30
1.0015
1.0015
1.0010
1.0010
[rads-1] ]
-1
speed[rads
1.0005
1.0005
1.0000
Generatorspeed
1.0000
0.9995
0.9995
Generator
0.9990
0.9990
0.9985
0.9985
0.9980
0.9980
0.9975
0.9975
00 11 22 33 44 55 66 77 88 99 10
10 11
11 12
12 13
13 14
14 15
15
Time[s]
Time [s]
KK==66 KK==77 KK==88 KK==99 KK==10
10
KK==11
11 KK==12
12 KK==13
13 KK==14
14
31. Gen1
Figure Figure 31.speed
Gen1response on a smallon
speedresponse
response disturbance with PSS parameters Tparameters
= 0.5 s, the PSS gain
0.5s,s,increase
PSSin the lower
Figure 31. Gen1 speed on aasmall
smalldisturbance
disturbance withPSS
with PSSparameters TT==0.5 thePSS
the gain
gain
cluster 1-K
increase
i,ln = 6–10, and the upper cluster 1 K , = 10–14.
in the lower cluster 1-Ki,ln = i6–10, and the upper cluster 1 Ki,um = 10–14.
um
increase in the lower cluster 1-Ki,ln = 6–10, and the upper cluster 1 Ki,um = 10–14.
530
530
510
510
490
490
[MW]
power[MW]
470
470
Activepower
450
450
430
430
Active
410
410
390
390
370
370
00 11 22 33 44 55 66 77 88 99 10
10 11
11 12
12 13
13 14
14 15
15
Time[s]
Time [s]
KK==66 KK==77 KK==88 KK==99 KK==10
10
KK==11
11 KK==12
12 KK==13
13 KK==14
14
Figure
32. Gen1
FigureFigure 32.
32. Gen1
active
Gen1 active
power
active poweron
response
power response on aa small
smallwith
a small disturbance
response on disturbance with PSS
PSS parameters
disturbance with PSS parameters
0.5 s, the PSSTT
T =parameters == 0.5
gain 0.5 s,s, the
increase thein the
PSS
lower clustergain
1-K increase
i,ln = 6–10, in
andthe
thelower
uppercluster
cluster 1-K
1 Ki ,
i,ln ==6–10, and
10–14. the upper cluster 1 K i,um = 10–14.
PSS gain increase in the lower cluster 1-Ki,ln = 6–10, and the upper cluster 1 Ki,um = 10–14.
um
Tables 11
Tables 11 and 12Tables
and 12 11 overview
give an
an and 12 give an overview
of the
the of the weight
weight functions.
functions. infunctions.
As in As incase,
the previous
previous the previous
case, give
the weightoverview
functionsof weight
related to generator speedAs the
and active case, with the
power increase
the weight
the weight functions
functions related
related to
to generator
generator speed
speed andand active
active power
power increase
increase with
with the
the PSS
PSS gain increase in both lower cluster 1 and upper cluster 1 while the rotorPSS
angle weight
gainincrease
gain increasein
in both
both lower
lower
function
cluster11and
cluster
decreases.
andupper
upper cluster
cluster11 while
while the
therotor
rotorangle
angle weight
weightfunc-
func-
tiondecreases.
tion decreases. The damping ratio of the dominant oscillatory mode reached the maximum in the
lower cluster 1 for the PSS gain K1,l1 = 9.
Table 11. Weight functions forTT==0.5
Table 11. Weight functions for 0.5s,s,KKi,ln
i,ln = 6–10, n = 1 (lower cluster 1); t = 220 ms.
= 6–10, n = 1 (lower cluster 1); t = 220 ms.
ii KKi,l1
i,l1
ΓΓζi,l1
ζi,l1
ΓΓωi,l1
ωi,l1
ΓΓδi,l1
δi,l1
ΓΓPi,l1
Pi,l1
ΓΓi,l1
i,l1(i)
(i)
00 10
10 0.507928
0.507928 0.9999430777
0.9999430777 0.7031884278
0.7031884278 0.211588861
0.211588861 2.42264854628
2.42264854628
11 99 0.524347
0.524347 0.9999411008
0.9999411008 0.7163057813
0.7163057813 0.183731701
0.183731701 2.42432562562
2.42432562562
22 88 0.269074
0.269074 0.9999389630
0.9999389630 0.7280097024
0.7280097024 0.154920355
0.154920355 2.15194343419
2.15194343419
33 77 0.243622
0.243622 0.9999366821
0.9999366821 0.7383721799
0.7383721799 0.124432751
0.124432751 2.10636402651
2.10636402651
Energies 2021, 14, 5644 24 of 29
Table 11. Weight functions for T = 0.5 s, Ki,ln = 6–10, n = 1 (lower cluster 1); t = 220 ms.
14, x FOR PEER REVIEW 25 of 30
i Ki,l1 Γζi,l1 Γωi,l1 Γδi,l1 ΓPi,l1 Γi,l1 (i)
0 10 0.507928 0.9999430777 0.7031884278 0.211588861 2.42264854628
1 9 0.524347 0.9999411008 0.7163057813 0.183731701 2.42432562562
Table 12. Weight functions
2 for8 T = 0.5, Ki,um = 10–14,
0.269074 m = 1 (upper
0.9999389630 cluster 1); t = 220
0.7280097024 ms.
0.154920355 2.15194343419
3 7 0.243622 0.9999366821 0.7383721799 0.124432751 2.10636402651
i Ki,u1 4Γζi,u1 6 Γωi,u1
0.218705 Γδi,u1 0.7469502268
0.9999341808 ΓPi,u1 0.091936998Γi,u1(i)2.05752669831
0 10 0.507928 0.9999430777 0.7031884278 0.211588861 2.42264854628
1 11 0.492426 0.9999449201 0.6889357381 0.238297313 2.41960444481
Table 12. Weight functions for T = 0.5, Ki,um = 10–14, m = 1 (upper cluster 1); t = 220 ms.
2 12 0.478054 0.9999466013 0.6733378546 0.263934882 2.41527323297
3 13 i K Γζi,u1
0.464878 i,u1 0.9999480314 Γωi,u1
0.6559384555 Γδi,u1
0.286974980 ΓPi,u12.40773908371 Γi,u1 (i)
4 14 0
0.452867 0.507928
10 0.9999493362 0.9999430777
0.6376144533 0.7031884278
0.308911693 0.211588861
2.39934237831 2.42264854628
1 11 0.492426 0.9999449201 0.6889357381 0.238297313 2.41960444481
2 12 0.478054 0.9999466013 0.6733378546 0.263934882 2.41527323297
The damping ratio 3 of13the dominant
0.464878 oscillatory mode
0.9999480314 reached
0.6559384555 the0.286974980
maximum in2.40773908371
the
lower cluster 1 for the4 PSS gain
14 K1,l1 = 9.
0.452867 0.9999493362 0.6376144533 0.308911693 2.39934237831
Γi,l1(i) reached the maximum for Γ1,l1(1) and Γ4,l1(4)< Γ3,l1(3) and, according to (18), it
follows: Γi,l1 (i ) reached the maximum for Γ1,l1 (1) and Γ4,l1 (4) < Γ3,l1 (3) and, according to
(18), it follows:
𝐾𝑙 = 𝐾1,𝑙1 = 9
Kl = K1,l1 = 9
Further, Γi,u1(i) reached
Further, the
Γi,u1maximum formaximum
(i) reached the Γ0,u1(0) and
for ΓΓ0,u1
4,u1(4) < Γ3,u1(3) and, according
(0) and Γ4,u1 (4) < Γ3,u1 (3) and, according
to (15), it follows:to (15), it follows:
K = K0,u1 = 10
𝐾𝑢 = 𝐾0,𝑢1 = u10
maxΓ
i,um (i ) < maxΓi,ln (i ).
𝑚𝑎𝑥𝛤𝑖,𝑢𝑚 (i) < 𝑚𝑎𝑥𝛤 𝑖,𝑙𝑛 (i)
Which
Which implies that theimplies thatKthe
PSS gain K is chosen as:
PSS gainas:
is chosen
K ==Kl9= K1,l1 = 9
K = 𝐾𝑙 = 𝐾1,𝑙1
Here, it can alsoHere,
be noticed thatbedifference
it can also noticed thatbetween
differenceweight
between functions Γ1,l1(1) →
weight functions Γ1,l1K(1) = K1,l1 = 9
1,l1 →
0,l1 =Γ10
9 and Γ0,u1(0) → Kand 0,u1 (0) → K0,l1 smaller
becomes as the
= 10 becomes fault as
smaller duration
the fault increase. The weight
duration increase. func- functions
The weight
from Tables 11 and 12 are presented
tions from Tables 11 and 12 are presented in Figures 33–37. in Figures 33–37.
0.600
0.500
0.400
0.300
0.200
0.100
0.000
NO PSS K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
Figure Figure 33.ζi,l1Γζand

33. Γζ (Γ Γζi,u1
(Γζi,l1 and Γζi,u1
from Table
from11 and 11
Tables Table
and 12) for
forPSS
PSSparameters
parametersT =T 0.5
= 0.5
s, Ks,=K6–14.
= 6–14.
Energies
Energies 2021,
2021, 14,14, x FOR PEER REVIEW
5644 26 of 25
30of 29
Γω vs PSS gain
0.99996
0.99995
0.99995
0.99995
0.99995
0.99995
0.99995
0.99994
0.99994
0.99994
0.99994
0.99994
0.99994
0.99993
0.99993
0.99993
0.99993
0.99993 K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
0.99993 K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
Figure 34.
34. Γ K=6 and ΓK=7 ωi,u1 from K=8 K=9 TableK=10 K=11 K=12 T = 0.5
K=13 K=14
Figure Γωω(Γ(Γωi,l1
ωi,l1 and Γ TableTables
from 11 and 11 and 12)
12) for
forPSS
PSS parameters
parameters T = s, s,
0.5 K =K6–14.
= 6–14.
Figure 34. Γω (Γωi,l1 and Γωi,u1ωi,u1
from Table 11 and Table 12) for PSS parameters T = 0.5 s, K = 6–14.
0.760 Γδ vs PSS gain
0.740
0.740
0.720
0.740
0.720
0.700
0.720
0.700
0.680
0.700
0.680
0.660
0.680
0.660
0.640
0.660
0.640
0.620
0.640
0.620
0.600
0.620
0.600
0.580
0.600
0.580 K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
0.580 K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
Figure 35. ΓK=6
δ (Γδi,l1 andK=7Γδi,u1 fromK=8
Table 11 K=9
and Table K=10 K=11
12) for PSS K=12 T = 0.5
parameters K=13 K=14
s, K = 6–14.
Figure 35.
Figure 35. Γ
Γδ (Γ(Γδi,l1 and Γδi,u1 from Table 11 and Table 12) for PSS parameters T = 0.5 s, K = 6–14.
and Γ from Tables 11 and 12) for PSS parameters T = 0.5 s, K = 6–14.
Figure 35. Γδδ (Γδi,l1
δi,l1
and Γδi,u1δi,u1
from Table 11 and Table 12) for PSS parameters T = 0.5 s, K = 6–14.
0.350 ΓP vs PSS gain
0.300
0.300
0.300
0.250
0.250
0.250
0.200
0.200
0.200
0.150
0.150
0.150
0.100
0.100
0.100
0.050
0.050
0.050
0.000
0.000 K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
0.000 K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
Figure 36. ΓK=6
P (ΓPi,l1 andK=7ΓPi,u1 fromK=8
Table 11 K=9
and TableK=10
12) for PSSK=11 K=12 T = 0.5
parameters K=13 K=14
s, K = 6–14.
Figure36.
Figure 36. ΓΓP (Γ
(ΓPi,l1 and ΓPi,u1
and Γ fromfromTable 11 and
Tables 11Table 12) for
and 12) for PSS
PSS parameters
parametersTT==0.50.5s,s,KK= =
6–14.
6–14.
P Pi,l1 Pi,u1
The difference between the outputs of the weight functions Γi,l1(i) and Γi,u1(i) becomes
Energies 2021, 14, 5644
insignificant with further PSS gain increase after the maximum has been reached for K26=of 29
9 (Figure 37).
2.500
Weight functions Γi.l1(i) and Γi,u1(i)
2.400
2.300
2.200
2.100
2.000
1.900
1.800
K=6 K=7 K=8 K=9 K=10 K=11 K=12 K=13 K=14
Figure 37. Weight functions: Γi,l1(i) –(i = 0–4 for K = 6–10), Γi,u1(i) (i = 0–4 for K = 10–14), T = 0.5 s.
Figure 37. Weight functions: Γi ,l1 (i)—(i = 0–4 for K = 6–10), Γi,u1 (i) (i = 0–4 for K = 10–14), T = 0.5 s.
Hence,
The the weight
difference between functions representing
the outputs the time-domain
of the weight functions Γi,l1system
(i) and performance
Γi,u1 (i) becomes
measures have a more significant influence on the PSS parameters
insignificant with further PSS gain increase after the maximum has been reached selection. The weight
for K = 9
function
(Figure 37). associated with the generator active power oscillations is of particular im-
portance since the active power oscillations can lead to the load loss. All the simulation
Hence, the weight functions representing the time-domain system performance mea-
results indicate the lower amplitude of active power oscillations if the PSS parameters are
sures have a more significant influence on the PSS parameters selection. The weight
selected according to a higher value of this function. Therefore, the impact of the proposed
function associated with the generator active power oscillations is of particular importance
weight function associated with the generator active power oscillations can be empha-
since the active power oscillations can lead to the load loss. All the simulation results
sized by multiplying it with the corrective factor. Even multiplied with value 1.1, the
indicate the lower amplitude of active power oscillations if the PSS parameters are selected
weight functions ΓPi,l1 and ΓPi,u1 will increase the PSS gain from K = 9 to K = 10. With the
according to a higher value of this function. Therefore, the impact of the proposed weight
further increase of this corrective factor, the reduction of the active power oscillations am-
function associated with the generator active power oscillations can be emphasized by mul-
plitudes will be more emphasized. The decision of the most appropriate corrective factor
tiplying it with the corrective factor. Even multiplied with value 1.1, the weight functions
value will be subjected to future works applying machine learning methods.
ΓPi,l1 and ΓPi ,u1 will increase the PSS gain from K = 9 to K = 10. With the further increase of
this corrective factor, the reduction of the active power oscillations amplitudes will be more
6. Conclusions
emphasized. The decision of the most appropriate corrective factor value will be subjected
In this paper, an algorithm for the PSS1A tuning comprising both the s-domain and
to future works applying machine learning methods.
the time domain system performance measures was proposed and tested on multi-ma-
6.chine IEEE 14-bus system model. The location of the PSS implementation is determined
Conclusions
by analyzing the participation of the generator speeds state variables in the dominant os-
In this
cillatory paper,
mode of theananalyzed
algorithm for the
system. ThePSS1A tuning
well-known comprising
phase both method
compensation the s-domainfor
and the time domain system performance measures was
the PSS tuning based on the s-domain system performance measures has only beenproposed and tested onup-multi-
machine IEEE 14-bus system model. The location of the PSS implementation
graded with the time-domain system performance measures incorporating the weight is determined
by analyzing
functions the participation
comparable of the generator
with the damping ratio of thespeeds state system
dominant variablesmodein the
as thedominant
pro-
oscillatory
posed weight function related with the s-domain system performance measures. Formethod
mode of the analyzed system. The well-known phase compensation dif-
for the system
ferent PSS tuning based on
disturbance the s-domain
durations, system
three case performance
studies measures
were examined. As thehasduration
only been
upgraded
of system with the time-domain
disturbances system
increases, the impactperformance measuressystem
of the time-domain incorporating
measuresthe on weight
PSS
functions comparable
tuning increases with
as well. the damping
Besides ratio of
the damping ofthe
thedominant system mode
electromechanical as the proposed
oscillations in the
weight
power function
system, the related with the s-domain
implementation of the PSS, system performance
according measures.
to the proposed For different
algorithm, can
system
reducedisturbance
the amplitudes durations, three case
of the active power studies were examined.
oscillations to which theAs modern
the duration
power of sys-
system
disturbances
tems with possibleincreases, thereduction
inertia impact ofare the time-domain
particularly system Hence,
vulnerable. measures the on
PSSPSS tuning
imple-
mentationasand
increases well.tuning,
Besides according to the of
the damping proposed algorithm, can oscillations
the electromechanical be beneficialinfor theboth
power
small-signal
system, and transient stability
the implementation of thein the according
PSS, power system. to the proposed algorithm, can reduce
the amplitudes of the active power oscillations to which the modern power systems with
possible inertia reduction are particularly vulnerable. Hence, the PSS implementation and
tuning, according to the proposed algorithm, can be beneficial for both small-signal and
transient stability in the power system.
Author Contributions: Conceptualization, P.M. and R.K.; methodology, R.K.; software, P.M.; val-
idation, P.M. and H.R.C.; formal analysis, H.R.C. and H.G.; investigation, R.K.; resources, R.K.;
Energies 2021, 14, 5644 27 of 29
data curation, H.G.; writing—original draft preparation, R.K.; writing—review and editing, P.M.;
visualization, H.G.; supervision, H.R.C.; project administration, P.M.; funding acquisition, P.M., R.K.
and H.G. All authors have read and agreed to the published version of the manuscript.
Funding: The APC was funded by Faculty of Electrical Engineering, Computer Science, and Infor-
mation Technology Osijek.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: IEEE 14-bus test system data are available online. AVR and PSS
standard models are available online; AVRs parameters are given in Appendix A. All other data
presented in this study are available on request.
Conflicts of Interest: The authors declare no conflict of interest.
Appendix A
AVR parameters (AVR, IEEE URST5B):
AVR1, AVR2, AVR3: Tr = 0.01 s, Tb1 = 10 s, Tc1 = 2 s, Tb2 = 1 s, Tc2 = 1 s, Kr = 200 p.u.,
T1 = 0.01 s, Kc = 0.1 p.u., Vmin = −3 p.u., Vmax = 3.4 p.u.
AVR4, AVR5: Tr = 0.01 s, Tb1 = 10 s, Tc1 = 2 s, Tb2 = 1 s, Tc2 = 1 s, Kr = 150 p.u., T1 = 0.01 s,
Kc = 0.1 p.u., Vmin = −3 p.u., Vmax = 3.4 p.u.
Static compensation devices parameters:
SC: SB = 100 MVA, UB = 13.8 kV, UN = 1.07 p.u., Qmax = 0.24 p.u., Qmin = −0.06 p.u.,
Vmax = 1.2 p.u., Vmin = 0.8 p.u.
SC1: SB = 100 MVA, UB = 69 kV, UN = 1.01 p.u., Qmax = 0.4 p.u., Qmin = 0.00 p.u.,
Vmax = 1.2 p.u., Vmin = 0.8 p.u.
SC2: SB = 100 MVA, UB = 18 kV, UN = 1.09 p.u., Qmax = 0.24 p.u., Qmin = −0.06 p.u.,
Vmax = 1.2 p.u., Vmin = 0.8 p.u.
References
1. Peres, W.; Coelho, F.C.; Costa, J.N. A pole placement approach for multi-band power system stabilizer tuning. Int. Trans. Electr.
Energy Syst. 2020, 30, e12548. [CrossRef]
2. Prasertwong, K.; Mithulananthan, N.; Thakur, D. Understanding low-frequency oscillation in power systems. Int. J. Electr. Eng.
Educ. 2010, 47, 248–262. [CrossRef]
3. Falehi, A.D.; Rostami, M.; Mehrjadi, H. Transient Stability Analysis of Power System by Coordinated PSS-AVR Design Based on
PSO Technique. Energy Power Eng. 2011, 3, 478–484. [CrossRef]
4. Dudgeon, G.J.; Leithead, W.E.; Dysko, A.; O’Reilly, J.; McDonald, J.R. The effective role of AVR and PSS in power systems:
Frequency response analysis. IEEE Trans. Power Syst. 2007, 22, 1986–1994. [CrossRef]
5. Benaissa, O.M.; Hadjeri, S.; Zidi, S.A. Impact of PSS and SVC on the power system transient stability. In Proceedings of the 8th
International Conference on Modelling, Identification and Control (ICMIC), Algiers, Algeria, 15–17 November 2016; pp. 303–307.
[CrossRef]
6. Dysko, A.; Leithead, W.E.; O’Reilly, J. Enhanced power system stability by coordinated PSS design. IEEE Trans. Power Syst. 2009,
25, 413–422. [CrossRef]
7. Sahare, S.B.; Bonde, U.G. Transient stability enhancement in multi-machine power system by using power system stabilizer (Pss)
and static var compensator (Svc). Int. J. Eng. Res. 2020, 9. [CrossRef]
8. Calderón-Guizar, J.G.; Ramírez-González, M.; Castellanos-Bustamante, R. Castellanos, identification of low frequency oscillation
modes in large transmission systems. Rev. Fac. Ing. 2017, 82, 31–39.
9. Pagola, F.L.; Perez-Arriaga, I.J.; Verghese, G.C. On sensitivities, residues and participations: Applications to oscillatory stability
analysis and control. IEEE Trans. Power Syst. 1989, 4, 278–285. [CrossRef]
10. Djalal, M.R.; Imran, A.; Robandi, I. Optimal placement and tuning power system stabilizer using participation factor and
imperialist competitive algorithm in 150 kV South of Sulawesi system. In Proceedings of the International Seminar on Intelligent
Technology and Its Applications (ISITIA), Surabaya, Indonesia, 20–22 May 2015; pp. 147–152. [CrossRef]
Energies 2021, 14, 5644 28 of 29
11. Benasla, M.; Allaoui, T.; Brahami, M.; Hadji, B. Frequency response analysis to select the best PSS locations in power systems. In
Proceedings of the 2016 8th International Conference on Modelling, Identification and Control (ICMIC), Algiers, Algeria, 15–17
November 2016; pp. 269–273. [CrossRef]
12. Flores, H.B.; Cepeda, J.C.; Gallardo, C.F. Optimum location and tuning of PSS devices considering multi-machine criteria and
a heuristic optimization algorithm. In Proceedings of the IEEE PES Innovative Smart Grid Technologies Conference—Latin
America (ISGT Latin America), Quito, Ecuador, 20–22 September 2017; pp. 1–6. [CrossRef]
13. Khormizi, A.B.; Nia, A.S. Damping of power system oscillations in multi-machine power systems using coordinate design of PSS
and TCSC. In Proceedings of the 10th International Conference on Environment and Electrical Engineering, Rome, Italy, 1–7 May
2011; pp. 1–4. [CrossRef]
14. Sharma, A.; Kumar, R. Optimum location of PSS and its parameters by using particle swarm optimization. Int. J. Adv. Res. Eng.
Technol. 2019, 10. [CrossRef]
15. Chow, J.H. A pole-placement design approach for systems with multiple operating conditions. In Proceedings of the 27th IEEE
Conference on Decision and Control, Austin, TX, USA, 7–9 December 1988; pp. 1272–1277. [CrossRef]
16. Othman, H.; Sanchez-Gasca, J.J.; Kale, M.A.; Chow, J.H. On the design of robust power system stabilizers. In Proceedings of the
28th IEEE Conference on Decision and Control, Tampa, FL, USA, 13–15 December 1989; pp. 1853–1857. [CrossRef]
17. Demello, F.P.; Concordia, C. Concepts of synchronous machine stability as affected by excitation control. IEEE Trans. Power Appar.
Syst. 1969, 88, 316–329. [CrossRef]
18. Larsen, E.V.; Swann, D.A. Applying power system stabilizers. Part I: General concepts. IEEE Power Eng. Rev. 1981, PER-1, 62–63.
[CrossRef]
19. Larsen, E.V.; Swann, D.A. Applying power system stabilizers part II: Performance objectives and tuning concepts. IEEE Trans.
Power Appar. Syst. 1981, PAS-100, 3025–3033. [CrossRef]
20. Larsen, E.V.; Swann, D.A. Applying power system stabilizers part III: Practical considerations. IEEE Trans. Power Appar. Syst.
1981, PAS-100, 3034–3046. [CrossRef]
21. Sebaa, K.; Boudour, M. Robust power system stabilizers design using multi-objective genetic algorithm. In Proceedings of the
IEEE Power Engineering Society General Meeting, Tampa, FL, USA, 24–28 June 2007; pp. 1–7. [CrossRef]
22. Spoljaric, T.; Pavic, I. AVR and PSS coordination strategy by using multi-objective ant lion optimizer. In Proceedings of the
2020 43rd International Convention on Information, Communication and Electronic Technology (MIPRO), Opatija, Croatia, 28
September–2 October 2020; pp. 928–933. [CrossRef]
23. Attia, A.-F. Online tuning based fuzzy logic controller for power system stabilizers. In Proceedings of the 2017 Nineteenth
International Middle East Power Systems Conference (MEPCON), Cairo, Egypt, 19–21 December 2017; pp. 952–957. [CrossRef]
24. Shafiullah, M.; Rana, M.J.; Alam, M.S.; Abido, M.A. Online tuning of power system stabilizer employing genetic programming
for stability enhancement. J. Electr. Syst. Inf. Technol. 2018, 5, 287–299. [CrossRef]
25. Bakolia, V. Design and analysis of fuzzy logic based power system stabilizer. Int. J. Eng. Res. 2020, 9. [CrossRef]
26. Douidi, B.; Mokrani, L.; Machmoum, M. A new cascade fuzzy power system stabilizer for multi-machine system stability
enhancement. J. Control. Autom. Electr. Syst. 2019, 30, 765–779. [CrossRef]
27. Verdejo, H.; Torres, R.; Pino, V.; Kliemann, W.; Becker, C.; Delpiano, J. Tuning of controllers in power systems using a heuristic-
stochastic approach. Energies 2019, 12, 2325. [CrossRef]
28. Kim, J.J.; Park, J.H. A novel structure of a power system stabilizer for microgrids. Energies 2021, 14, 905. [CrossRef]
29. Ahmed, M.; Roy, N.K. A comparative analysis on different types of power system stabilizers. In Proceedings of the 2016 5th
International Conference on Informatics, Electronics and Vision (ICIEV), Dhaka, Bangladesh, 13–14 May 2016; pp. 797–802.
[CrossRef]
30. Farah, A.; Guesmi, T.; Abdallah, H.H.; Ouali, A. Real-time stability enhancement based on neural fuzzy networks and genetic
algorithms. In Proceedings of the 10th International Multi-Conferences on Systems, Signals & Devices 2013 (SSD13), Hammamet,
Tunisia, 18–21 March 2013; pp. 1–6. [CrossRef]
31. Rafiee, Z.; Ganjefar, S.; Fattahi, A. A new PSS tuning technique using ICA and PSO methods with the fourier transform. Electr.
Electron. Eng. 2012, 1, 17–23. [CrossRef]
32. Li, X.; Wang, Z.; Xu, J.; Chen, B. Power system stabilizer parameters designing based on genetic simulated annealing algorithm. J.
Clean Energy Technol. 2015, 4, 178–182. [CrossRef]
33. Castrillon, N.J.; Sanchez, H.M.; Perez, J.A. A tool for optimal PSS tuning for the Colombian power system. In Proceedings of the
2014 IEEE PES T & D Conference and Exposition, Chicago, IL, USA, 14–17 April 2014; pp. 1–5. [CrossRef]
34. Alhaj, A.K.; Hassan, M.O. Design and performance evaluation of TCSC and PSS to mitigate low frequency oscillations. In
Proceedings of the 2018 International Conference on Computer, Control, Electrical, and Electronics Engineering (ICCCEEE),
Khartoum, Sudan, 12–14 August 2018; pp. 1–5. [CrossRef]
35. Verrelli, C.M.; Marino, R.; Tomei, P.; Damm, G. Nonlinear robust coordinated PSS-AVR control for a synchronous generator
connected to an Infinite Bus. IEEE Trans. Automat. Contr. 2021. [CrossRef]
36. Zea, A.A. Power System Stabilizers for The Synchronous Generator: Tuning and Performance Evaluation. Master’s Thesis,
Chalmers University of Technologyg, Gothenburg, Sweden, 2013.
Energies 2021, 14, 5644 29 of 29
37. PPande, W.; Chakrabarti, S.; Srivastava, S.C. Online updating of synchronous generator linearised model parameters and PSS
Tuning. In Proceedings of the 2018 20th National Power Systems Conference (NPSC), Tiruchirappalli, India, 14–16 December
2018; pp. 1–6. [CrossRef]
38. Lu, S.; Zhang, W.; Wang, T.; Cai, T.; Sui, X.; Li, H.; Zhu, T.; Gang, Y.; Yu, Y. Parameter tuning and simulation analysis of PSS
function in excitation system with suppression of low frequency Oscillation. In Proceedings of the 2019 IEEE 8th International
Conference on Advanced Power System Automation and Protection (APAP), Xi’an, China, 21–24 October 2019; pp. 474–479.
[CrossRef]
39. IEEE Power Engineering Society. IEEE Recommended practice for excitation system models for power system stability studies.
In IEEE Standard 421.5-2016 (Revision of IEEE Standard 421.5-2005—Redline); IEEE: New York, NY, USA, 2016. [CrossRef]
40. Dey, P.; Bhattacharya, A.; Das, P. Tuning of power system stabilizer for small signal stability improvement of interconnected
power system. Appl. Comput. Inform. 2017, 16, 3–28. [CrossRef]
41. Kundur, P. Power System Stability and Control; McGraw-Hilll: New York, NY, USA, 1993.
42. Tharani P., A.S.; Jyothsna, T.R. Design of PSS for small signal stability improvement. Int. J. Eng. Sci. Technol. 2010, 3, 2–7.
43. Kundur, P.; Klein, M.; Rogers, G.J.; Zywno, M.S. Application of power system stabilizers for enhancement of overall system
stability. IEEE Trans. Power Syst. 1989, 4, 614–626. [CrossRef]
44. JChow, H.; Sanchez-Gasca, J.J. Power System Modeling, Computation, and Control; Wiley-IEEE Press: New York, NY, USA, 2020.
45. Pourbeik, P.; Vowles, D.J.; Gibbard, M.J. Small-Signal Stability, Control and Dynamic Performance of Power Systems; University of
Adelaide Press: Adelaide, Australia, 2015.

Energies 14 05644

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Energies 14 05644

Uploaded by

Copyright:

Available Formats

energies

Academic Editor: Abu-Siada Ahmed

Received: 8 August 2021 1. Introduction

Energies 2021, 14, 5644. https://doi.org/10.3390/en14185644 https://www.mdpi.com/journal/energies

4. PSS Tuning Algorithm

While coefficient-H represents an inertia time constant of the synchronous machine

Γζi = ζ dom (9)

Ku = Ki,um f or maxΓi,um (i ) (15)

Kinit = K4,um−1 (17)

Kl = Ki,ln f or maxΓi,ln (i ) (18)

j=1 Tinit = T4j

FOR i = 0 TO i = 4 FOR i = 0 TO i = 4 FOR i = 0 TO i = 4

YES FIND DOMINANT FIND DOMINANT FIND DOMINANT

COMPARE Γj(i) FOR Ku = Ki,um Ku = Ki,ln

5. PSS Tuning Algorithm Impact on the System Performance

Table 1. Local oscillatorydiagram

The damping closest

Table 2. Participation ofAfter the PSS location

Active power [MW]

i0 Ti,1 [s]0.2 Γζi 0.094299 Γωi0.999995596 Γ0.989720196

Γζ vs PSS time constant(s)- T

Figure 10. Γζi for PSS parameters K = 10, T = Ti,1 = 0.2–0.6 s.

Γω vs PSS time constant(s)- T

Γ vs PSS time consant(s)- T

ΓP vs PSS time constant(s) -T

Γ1 vs PSS time constant(s) -T

Γ1 vs PSS time constant(s) -T

i Ki,l1 α [s−1] ±jω [rad s−1] ξ

i Ki,u1 α [s−1 ] ±jω [rad s−1 ] ξ

Energies 2021, 14, x FOR PEER REVIEW 16 of 30

Generator speed [rads-1]

Further, according to (14):

0 10 0.507928 0.9999982622 0.9938148445 0.9764722440 3.8037042787

ΓΓωω vs PSS gain

1.0005 5.4. Case Study 2—Disturbance Duration 150 ms

maxΓi,um (i ) < maxΓi,ln (i )

Figure Figure 33.ζi,l1Γζand

Static compensation devices parameters:

You might also like