Article

Interline Power Flow Controller Allocation for Active Power

Losses Enhancement Using Whale Optimization Algorithm
Ahmed M. Alshannaq 1 , Mohammed A. Haj-ahmed 1, * , Mais Aldwaik 2 and Dia Abualnadi 1

1 Electrical Engineering Department, The University of Jordan, Amman 11942, Jordan

2 Civil Engineering Department, Philadelphia University, Jarash 19392, Jordan
* Correspondence: [email protected]

Abstract: Transmission networks face continuous electrical and mechanical stresses due to increasing
system challenges and power losses. Transmission networks require special focus and detailed
studies each time a load or a generator emerges to the grid. The interline power flow controller
(IPFC) is a relatively new scheme that is implemented in the transmission network to improve
transmission efficiency, decrease transmission losses, and enhance voltage profile. In this paper, the
interline power flow controller’s impact on transmission network performance is investigated as
it is implemented within the IEEE 5-bus, 14-bus, and IEEE 57-bus systems. In addition, the whale
optimization algorithm (WOA) is used to optimize the interline power flow controller locations
within the system to achieve optimal transmission system performance. WOA performance is also
compared to genetic algorithm (GA) and particle swarm optimization (PSO) algorithms, and the
superiority of the proposed WOA-based control is proved. The robustness of the optimized system
against load variations is investigated and the results introduced affirm the capability of the interline
power flow controller to enhance transmission network efficiency.

Keywords: interline power flow controller; transmission network; whale optimization algorithm

Citation: Alshannaq, A.M.; 1. Introduction

Haj-ahmed, M.A.; Aldwaik, M.;
Flexible alternating current transmission system (FACTS) devices were introduced
Abualnadi, D. Interline Power Flow
to enhance system stability, improve load sharing, minimize power losses, manage line
Controller Allocation for Active
loading, and increase power transfer capability [1,2]. Various types of FACTS devices,
Power Losses Enhancement Using
such as static VAR compensators (SVCs), static synchronous series compensators (SSSCs),
Whale Optimization Algorithm.
unified power flow controllers (UPFCs), and IPFC, have been widely adopted.
Energies 2024, 17, 6318. https://
doi.org/10.3390/en17246318
IPFC is a voltage source converter (VSC)-based FACTS device designed for multiline
transmission system series compensation [3]. It consists of at least two VSCs connected to
Academic Editor: Miguel Castilla two or more transmission lines [4], allowing control of power flow and management of line
Received: 29 October 2024
loading. The performance of IPFC varies significantly depending on the placement of these
Revised: 28 November 2024 VSCs within the system [5,6].
Accepted: 12 December 2024 The application of IPFC in enhancing power system performance has gained significant
Published: 15 December 2024 attention in recent years. For instance, the authors in [7–11] investigated the potential
of IPFC implementation to improve power system stability. Specifically, the study in
reference [7] introduced a coordinated droop control strategy that enhances frequency
stability in power grids with high renewable energy penetration. Similarly, research in [8]
Copyright: © 2024 by the authors. examined the integration of IPFC with HVDC systems, demonstrating its capability to
Licensee MDPI, Basel, Switzerland.
maintain voltage stability and manage reactive power in grids dominated by renewable
This article is an open access article
generation. Moreover, the authors in [9] proposed a current injection model for IPFC,
distributed under the terms and
utilizing PSO to design a damping controller aimed at improving dynamic stability under
conditions of the Creative Commons
varying operating conditions.
Attribution (CC BY) license (https://
Transmission system losses were also extensively studied with the implementation of
creativecommons.org/licenses/by/
4.0/).
IPFC within the grid [10,11]. For example, the authors in [10] proposed an optimal power

Energies 2024, 17, 6318. https://doi.org/10.3390/en17246318 https://www.mdpi.com/journal/energies

Energies 2024, 17, 6318 2 of 20

flow control method using IPFC to mitigate the impacts of overloading in a 22-bus test sys-
tem, demonstrating IPFC’s effectiveness in improving reliability and reducing transmission
losses. Similarly, reference [12] presented a theoretical framework for analyzing energy
dissipation and storage in power networks with IPFC, concluding that IPFC effectively
reduces transmission losses by controlling active and reactive power flow while enhanc-
ing voltage profiles. Furthermore, in reference [11], the authors allocated system losses
precisely to active and reactive power injections, mitigating overall system losses.
Building on the proven benefits of FACTS devices in power systems, researchers
have increasingly applied optimization techniques to further enhance power flow and
determine optimal FACTS device locations and parameters. Recently, various optimization
techniques have been employed in these studies, including GA, PSO, and WOA. While
each method has its unique strengths, WOA has gained significant interest due to its
superior convergence speed, efficiency, and simplicity in addressing complex power system
optimization problems [9,13–16]. For instance, reference [15] demonstrated WOA’s ability
to outperform GA in terms of achieving higher power loss reduction and an improved
voltage profile with fewer iterations. GA, although effective at exploring large search
spaces, incurs high computational costs due to the necessity of crossover, mutation, and
selection operations. Its convergence speed can also be relatively slow, requiring numerous
iterations to achieve optimal or near-optimal solutions. PSO, on the other hand, offers
simpler implementation and faster convergence for continuous optimization problems,
but it may sometimes experience premature convergence, particularly in non-linear and
complex systems, leading to suboptimal solutions. WOA, by contrast, is highly suitable
for solving complex nonlinear power system problems because of its ability to effectively
balance exploration (global search) and exploitation (local search) [16].
It is worth noting that the aforementioned algorithms are not the only ones utilized for
optimizing power system performance. Other approaches, such as gray wolf optimization
(GWO), PSO, and differential evolution (DE), have also been applied [17] to enhance the
system’s dynamic stability by optimal allocation of IPFC. In addition, the authors in [14]
combined GA with the cuckoo search algorithm (CSA) to study the effects of IPFC on system
power flow. Similarly, reference [18] employed moth flame optimization (MFO), PSO, and
biogeography-based optimization (BBO) to minimize the active power loss and enhance
voltage profiles considering different loading conditions. In another study, reference [19]
utilized entropy theory to identify the optimal location for IPFC within an electric system.
WOA has been extensively used to address various electric power engineering prob-
lems, as demonstrated in reference [15] for fault location estimation and in reference [20]
for minimizing power losses and enhancing voltage stability. However, its application in
optimizing IPFC locations in transmission networks needs more investigation.
The integration of IPFCs in power systems has been extensively studied to optimize
performance, reduce losses, and enhance stability through metaheuristic algorithms. Sig-
nificant advancements have been made in this domain. For instance, the authors in [21]
proposed the teaching–learning-based optimization (TLBO) algorithm to optimize the reac-
tance of the implementation of thyristor control series capacitors (TCSC) for transmission
congestion management in deregulated power systems. Although the approach demon-
strated improved efficiency on IEEE-30 bus systems, its scalability to larger systems remains
untested. The authors in [22] explored the classification and optimization trends of FACTS
devices using hybrid metaheuristic approaches, addressing parameter tuning challenges.
The study demonstrated theoretical promise but offered limited practical insights into load
variation and voltage profile impacts. The authors in [23] leveraged the modified whale
optimization algorithm (MWOA) for FACTS placement, achieving optimized solutions;
however, the approach necessitated extensive parameter fine-tuning, limiting real-time
applicability. In addition, it does not consider load sensitivity analysis.
The authors in [24] combined a fuzzy logic controller with a genetic algorithm to
identify the optimal IPFC placement. This hybrid approach effectively maintained voltage
stability and minimized transmission losses, though its scalability to larger grids was not
Energies 2024, 17, 6318 3 of 20

fully assessed. In reference [25], the authors utilized the bees algorithm for optimal IPFC
placement to reduce power loss, demonstrating a significant improvement in efficiency
when compared to PSO. However, this study’s results were based on static test systems
and did not account for dynamic grid conditions.
The study in [26] introduced the firefly algorithm to optimize IPFC placement for
power loss reduction in transmission lines, with successful testing on IEEE 30-bus and
NTPS 23-bus systems. Despite its computational efficiency, the approach lacked an analysis
of adaptability under varying load conditions, which is crucial for real-world applications.
Additionally, the research in [19] proposed an entropy-based method for IPFC placement,
employing a power flow transfer entropy index (PFTEI) to evaluate the impact of IPFC
on power distribution. This approach demonstrated effectiveness in a large-scale power
grid, yet it depended heavily on specific fault scenarios and required further validation for
general use.
Recent studies on IPFC allocation and optimization in power systems have achieved
significant advancements, yet several critical challenges remain unaddressed. The limita-
tions primarily center on scalability to larger power systems, adaptability under dynamic
loading conditions, and computational efficiency. Many current metaheuristic approaches,
such as GA and PSO, while effective for small to medium-sized systems, often fall short in
handling the complexities and variations encountered in extensive and highly dynamic
grid environments. Additionally, some methods require extensive parameter tuning and
fail to maintain robust performance under fluctuating load conditions. These constraints
highlight the need for an optimization method that combines high computational efficiency
with resilience in large, variable systems.
This study aims to cover these gaps by introducing WOA for IPFC allocation. WOA
has been shown to balance exploration and exploitation effectively, making it particularly
suited for complex, nonlinear optimization problems in power systems. By implementing
WOA, this research provides an approach that enhances scalability, computational efficiency,
and adaptability under variable loading conditions. Key contributions of this study are
outlined as follows:
1. Enhanced scalability: WOA is applied to optimize IPFC placement in IEEE 5-bus,
IEEE-14 bus, and IEEE-57 bus systems, demonstrating its effectiveness across diverse
grid sizes.
2. Improved load adaptability: WOA’s inherent ability to explore and exploit the search
space allows it to maintain optimal performance under varying load profiles, which is
critical for real-world power systems with dynamic demand.
3. Computational efficiency: The study leverages WOA’s faster convergence properties,
reducing computational overhead compared to traditional methods like GA and
PSO. This efficiency is particularly advantageous for real-time applications and larger
system deployments.
4. Comparative validation: To validate the robustness of the proposed approach, the
results obtained using WOA are compared with those obtained from GA and PSO,
demonstrating WOA’s superiority in achieving lower active power losses and im-
proved voltage stability.
WOA’s efficiency and accuracy in power system optimization have been noted in
prior studies, including steady-state performance analysis of IPFC placements. In this
study, WOA is strategically employed to optimize active power losses and improve voltage
profiles within multi-bus systems by identifying optimal IPFC locations. This approach not
only addresses the persistent limitations in existing literature but also sets a benchmark
for future studies in IPFC allocation by ensuring WOA’s potential to enhance system
performance under complex and dynamic conditions.
The rest of the paper is organized as follows: Section 2 introduces the methodology
and the mathematical background for WOA and power flow and how they are applied
to allocate IPFC. Section 3 introduces the results and discussion, and finally, the paper is
concluded in Section 4.
 benchmark for future studies in IPFC allocation by ensuring WOA’s potential to enhance
system performance under complex and dynamic conditions.
The rest of the paper is organized as follows: Section 2 introduces the methodology
and the mathematical background for WOA and power flow and how they are applied to
Energies 2024, 17, 6318 4 of 20
allocate IPFC. Section 3 introduces the results and discussion, and finally, the paper is
concluded in Section 4.

2. Methodology of
2. Methodology of WOA
WOA forfor IPFC
IPFC Allocation
Allocation
IPFC
IPFC is constructed from
is constructed from aa minimum
minimum ofof two
two DCDC to AC back-to-back
to AC back-to-back converters. These
converters. These
converters are connected to the transmission lines in series using a coupling series
converters are connected to the transmission lines in series using a coupling series trans- trans-
former. The two converters are connected by a DC common link as shown in Figure
former. The two converters are connected by a DC common link as shown in Figure 1 [27]. 1 [27].
The equations that govern the system with the addition of the IPFC is discussed
The equations that govern the system with the addition of the IPFC is discussed next. next.

lines [27].
Figure 1. Schematic diagram of IPFC connected to two transmission lines [27].

2.1. Power Flow

2.1. Power Flow Equations
Equations
Adding IPFCtotothe
Adding IPFC thesystem
system means
means thatthat series
series voltage
voltage sources
sources are added
are added totrans-
to each each
transmission line, which modifies the transmission line parameters. The equations
mission line, which modifies the transmission line parameters. The equations of the power of the
power injected into the system once the IPFC converters are connected to the system
injected into the system once the IPFC converters are connected to the system can be pre- can be
presented as follows:
sented as follows:
𝑃 ==∑∑
Psli n
j=1 Vi𝑉
V𝑉si G𝐺ij cos
cos(δ𝛿i −−δ𝛿si ) ++B𝐵ij sin
sin
(δi𝛿−−δsi𝛿)) ) (1)

n
sli ==∑∑
Q𝑄 j=1 V𝑉
i V𝑉 𝐺ij sin
si G −δ𝛿si ) −−B𝐵ij cos
sin(δ𝛿i − cos(δi𝛿−−δsi𝛿)) (2)
where P sli and Qsli
Psli are the
sli are the injected
injected active
active and
and reactive
reactive powers
powers to to the transmission
transmission line (l)
from
from the VSC s where s is a subscript for the VSC number, j is a number between
the VSC s where s is a subscript for the VSC number, j is a number between 11 toto n
n
(number
(number of buses) when the series voltage source is connected to the transmission line i,i,
of buses) when the series voltage source is connected to the transmission line
si
si is
is the
the converter
converter numbernumber i,i, V Vii is
is the
the ith
ith bus
bus voltage,
voltage, V Vsisi is
is the
the voltage
voltage source
source converter
converter
connected
connected to to bus
bus i,i, δδii is
is the
the ith
ith bus
bus voltage
voltage phase, is the
phase, δδsisi is the voltage
voltage phase
phase of
of voltage
voltage source
source
converter connected to the bus i, and G and B is the conductance
converter connected to the bus i, and Gij and Bij is the conductance and the susceptance
ij ij and the susceptance
connecting
connecting bus bus ii to
to bus
bus j.j.
The contribution
The contribution of of IPFC
IPFC toto thethe power
power flow
flow can can bebe concluded
concluded by by their
their effect
effect on
on the
the
current flow between system buses as shown in
current flow between system buses as shown in the following equations: the following equations:

𝑉 δ∠𝛿
Vi ∠ i −−
Vs1𝑉∠δ∠𝛿s1 −−𝑉
V j ∠∠𝛿
δj
Iij 𝐼= = (3)
Zij𝑍

V𝑉 ∠∠𝛿
δ −−Vs2
𝑉 ∠∠𝛿
δs2 −−V𝑉
k ∠∠𝛿
δk
Iik𝐼= = i i (4)
𝑍
Zik
where
where IIijij and
andIIikik are
are the
thecurrents
currentsflowing
flowingininlines
lines ij and
ij and ik,ik,
respectively, Zij is
respectively, Zijthe transmission
is the transmis-
impedance of the line connecting bus i to bus j, and Z is the transmission
sion impedance of the line connecting bus i to bus j, and Zik is the transmission impedance
ik impedance of the
line connecting bus i to bus k. The above equations are used to quantify the active power
losses in the transmission lines within the system, as shown in the following subsection.

2.2. Mathematical Formulation of the Objective Function

The optimization model for IPFC allocation in this study is formally presented as a
minimization problem, with the objective of reducing the total active power loss across
Energies 2024, 17, 6318 5 of 20

the transmission lines in the power system. This objective is essential to enhance system
efficiency and minimize energy dissipation.

2.2.1. Objective Function

The objective function for minimizing active power loss (TL) is defined as:

TL = ∑in=1 ∑nj=1 Pji + Pij

(5)

where TL represents the total active power loss in the power system, n is the total number
of buses, Pij is the active power loss in the line from bus i to bus j, and Pji is the active power
loss in the line from bus j to bus i. This function accounts for directional power losses in
each transmission line, capturing the cumulative active power dissipation throughout the
system. These terms do not imply that the losses are different for each direction; rather,
they represent the computation of power losses under different power flow conditions,
which accounts for all power flow scenarios to ensure consistency in the loss calculations.

2.2.2. Design Variables

The design variables in this optimization problem are essential components that affect
the performance and placement of the IPFC. The design variable of this study is the IPFC
placement locations, which achieved identifying the optimal buses or transmission lines
where IPFC devices are installed.

2.2.3. Constraints
Optimization is subject to both equality and inequality constraints, ensuring power
balance, operational limits, and system stability:
1. Power balance constraints: To ensure that active and reactive power injections match
the power consumed by the loads at each bus, the following equality constraints must
be satisfied:
∆Pi = Pisch − Pcal
i
=0 (6)
∆Qi = Qisch − Qical = 0 (7)
where i represents the bus number, Pi sch and Qi sch are the scheduled active and reactive
power injections at bus i, and Pi cal and Qi cal are the calculated active and reactive power
demand at bus i.
2. IPFC power injection constraints: The IPFC must not alter the overall active power
injection at each bus, maintaining network stability. This constraint is represented by:
PE = ∑in=1 Psi = 0 (8)
where PE is the total active power injected by the series voltage sources of the IPFC across
the transmission lines, Psi is the power injected by each VSC, and n is the number system
buses.
3. Voltage magnitude and phase angle constraints: To ensure that IPFC operation remains
within safe and effective limits, the voltage magnitude and phase angle for each IPFC
voltage source are constrained as follows [28]:
Vsimin ≤ Vsi ≤ Vsimax (9)
δsimin ≤ δsi ≤ δsimax (10)
where Vsi denotes the voltage magnitude of the SVC associated with bus i, Vsimin and Vsimax
are the minimum and maximum voltage magnitude limits of the series voltage converter,
δsi is the phase angle of the converter injected voltage at bus i, and δsimin and δsimax are the
minimum and maximum limits for the voltage converter phase angles.
Energies 2024, 17, 6318 6 of 20

2.2.4. Formal Problem Definition

In summary, the optimization problem
can be formally stated as:
Minimize TL = ∑in=1 ∑nj=1 Pij + Pji
Subject to:
• Power balance constraints: ∆Pi = 0 and ∆Qi = 0
• IPFC active power injection constraint: PE = 0
• Voltage and phase angle limits: Vsimin ≤ Vsi ≤ Vsimax and δsimin ≤ δsi ≤ δsimax , voltage
magnitudes were constrained from 0.90 p.u. to 1.10 p.u. and the angle was set between
+180◦ and −180◦ .

2.3. Allocation Methodology

The allocation of IPFCs within the power system plays a primary role in reducing
system losses and improving overall performance. The proposed methodology utilizes
WOA to determine the optimal placement and control settings of IPFC devices. The process
is outlined as follows:
(1) Initialization:
The algorithm begins by generating a random initial population of candidate solutions,
where each candidate represents a potential IPFC placement (location) and its correspond-
ing control parameters (injected voltage magnitude and phase angle). These candidate
solutions are evaluated based on the total active power losses and voltage profile across
the system.
(2) Objective Evaluation:
For each candidate solution, the algorithm computes the total active power losses in
the system using the predefined objective function in Equation (5) subject to the constraints
in equations from (6)–(10). The performance of each candidate is assessed, and the best
performing solution in the current iteration is identified as the “leader”. The leader in this
context represents the location of the IPFCs that caused minimum active power losses and
kept the voltage profile within acceptable limits.
(3) Exploration and Exploitation:
WOA mimics the bubble-net hunting strategy of humpback whales to explore and
exploit the search space effectively:
Exploration phase: New candidate solutions are generated by positioning the IPFC at
different locations and adjusting its control parameters. This phase allows the algorithm to
broadly search for promising regions in the solution space.
Exploitation phase: Once promising solutions are identified, the algorithm refines
these solutions by narrowing the search around the current best solution (leader). This
ensures that the optimization converges toward the global minimum.
(4) Iteration and Convergence:
The algorithm iteratively updates the candidate solutions, evaluating their perfor-
mance in each iteration. The process continues until convergence criteria are met, such
as reaching a maximum number of iterations or achieving a negligible improvement in
system losses.
(5) Optimal IPFC Allocation:
At the end of the optimization process, the best-performing solution is selected as the
optimal IPFC location. This solution specifies the transmission line or bus where the IPFC
should be placed and its control settings, ensuring the lowest total active power losses.
(6) Validation of Results:
The final solution is validated to ensure it satisfies all operational constraints, includ-
ing power balance and voltage magnitudes and phases. Comparisons with alternative
Energies 2024, 17, 6318 7 of 20

methods (e.g., GA and PSO) are performed to demonstrate the effectiveness of the proposed
approach.
By systematically evaluating and refining potential IPFC placements, WOA identifies
the optimal configuration that minimizes active power losses while maintaining system
voltage stability and operational limits. The iterative nature of WOA ensures a balance
between exploring new possibilities and exploiting known good solutions, making it a
robust approach for solving the IPFC allocation problem in complex power systems.

2.4. The Whale Optimization Algorithm

The WOA is a nature-inspired type of meta-heuristic algorithms that simulates the
social behavior of humpback whales and is mainly involved in the bubble-net hunting
strategy [29]. WOA is used to solve different optimization problems, especially engineering
problems by determining the best values for different system parameters to select the
optimal solution for the problem. WOA is widely used as an optimizer for different
engineering problems because of its advanced features. These features include the ability
to solve engineering problems with growing complexity, which can support flexibility
and shorten the time needed for decision-making [30]. Other WOA advantages are the
simplicity in structure, the lesser operators required, and faster speed of convergence,
which gives it priority in solving different engineering problems.
This optimizer is inspired by one of the smartest mammal creatures and one of the
most famous type of whales which is called humpback whales. These types of whales use
a special and unique hunting method called the bubble-net feeding method. Furthermore,
humpback whales prefer to hunt a group of fish near the surface by diving nearly twelve
meters down the surface and then start making bubbles along the spiral path as a circle or
9-shape while swimming up toward the surface [27]. The mathematical modeling of WOA
is discussed now.

2.4.1. Whales Encircling Prey

The prey location is selected and encircled by humpback whales. WOA suggests the
target as the best or optimal candidate solution. After the best search agent is selected, the
other search agents try to update and close their positions to the best search agent position
as below:
→ → → →
D = C .X ∗ (t) − X (t) (11)

→ → → →
X ( t + 1) = X ∗ ( t ) − A . D (12)
→ → →
where A and C are vector coefficients, t is the current iteration, X∗ is the best solution until
→
now and updated when the best solution is found in any iteration, and X is the vector
→ →
position. Vectors A and C can be found as follows:
→ → → →
A = 2a · r − a (13)
→ →
C = 2· r (14)
→ →
where a is decreased from 2 to 0 linearly by iteration number, and r is a randomly selected
vector in the range of {0, 1}. The WOA’s exploration-exploitation balance is controlled by
→
the parameter a , which decreases linearly from 2 to 0 during the optimization process. This
ensures that the algorithm begins with a broad search of the solution space (exploration) and
gradually narrows its focus (exploitation) as it converges toward an optimal solution. The
→
linear decrease of a is common in WOA applications and does not require customization
unless specific system dynamics require a different approach.
The above equations can be applied to a 2-dimensional problem and then they can be
spread to be applied to an n-dimensional problem. Any position of the search agent can be
Energies 2024, 17, 6318 8 of 20

updated with respect to the best agent position, and this can give multiple positions around
→ →
the best one according to the values of vectors A and C . For 3-dimensional problems,
→
vector r must be defined randomly which gives the ability to set the vector position at any
point in the search space.

2.4.2. Bubble-Net Hunting Method

The humpback whale’s bubble-net behavior is specified as an exploitation process
and must be modeled mathematically. To achieve this model, a method to decrease the
→
vector a shall be used depending on the number of iterations. This shrinkage must lead to
→
a decrease in the values of vector A as appeared in Equation (13). Then the position must
be updated by the following equations:
→ → →
X (t + 1) = D ′ ·ebl ·cos(2πl ) + X ∗ (t) (15)
→ → →
D ′ = X ∗ (t) − X (t) (16)

→
where D ′ is the ith whale distance, b is the logarithmic spiral constant, and l is a random
number between [−1, 1]. These default settings are generally kept constant in WOA
applications, ensuring that the algorithm can exploit optimal regions effectively. No further
tuning may be required unless a unique system requires a more aggressive exploration
strategy.
The mathematical model that represents the bubble-net hunting behavior can be
concluded from the above steps as follows:
→ → →
→  X ∗ (t) − A ·D , p < 0.5
X ( t + 1) = → → (17)
 D ′ ·ebl ·cos(2πl ) + X ∗ (t) , p ≥ 0.5

where p is a randomly defined number between [0, 1].

2.4.3. Finding the Prey

In nature, humpback whales search randomly for prey depending on the other hump-
backs’ position, in what is called the exploration process. Therefore, to make the search
→
agent explore randomly and not according to the reference whales, A must be selected
greater than 1 or less than −1 to update the agent position randomly. This can support
WOA to proceed to the global search. The prey finding mathematical representation is
given as follows:
→ → −−−−→ →
D = C · Xrand − X (18)

→ −−−−→ → →
X (t + 1) = Xrand − A · D (19)
−−−−→
where Xrand is a random whale that is defined as a random vector position.

2.5. Representation of Humpback Whale Agents for IPFC Allocation

In WOA, each whale agent represents a potential solution to the IPFC allocation prob-
lem. The search space is defined by the candidate locations for IPFC placement and their
associated parameters, such as voltage magnitude and phase angle. This representation
ensures that every agent explores a unique configuration of IPFC allocation, contributing
to the optimization process.
Each whale agent is encoded as a vector in the search space. The vector is composed of
two main components, the IPFC locations and IPFC control parameters. The IPFC location
nergies 2024, 17, x FOR PEER REVIEW 9 of 21

Energies 2024, 17, 6318 Each whale agent is encoded as a vector in the search space. The vector is composed
9 of 20
of two main components, the IPFC locations and IPFC control parameters. The IPFC loca-
tion represents the indices of transmission lines and buses where the IPFC devices are to
be allocated. On the the
represents other hand,of
indices thetransmission
IPFC controllines
parameters include
and buses wherethethe
operational set- are to be
IPFC devices
tings of the IPFC devices, such as voltage magnitude and phase.
allocated. On the other hand, the IPFC control parameters include the operational settings
The position of adevices,
of the IPFC whale agent
such asinvoltage
the search space can
magnitude andbephase.
mathematically expressed
as: The position of a whale agent in the search space can be mathematically expressed as:
𝑥 = 𝑙 (20)
x i = [ li ] (20)
where xi is the position vector of the i-th whale agent, li = [li1, li2, …, liM] represents the
locations where
of M IPFCxi isdevices.
the position vector of
This location the is
vector i-th whaletoagent,
subject li = [li1 , in
the constraints li2 ,the
. . .,param-
liM ] represents
the locations of M IPFC devices. This location vector is subject to the constraints in the
eters vector:
parameters vector:
𝑦 = 𝑝 y = [p ] (21) (21)
i i
where yi is the constraint
where vector of the
yi is the constraint i-th whale
vector of theagent, pi = [pagent,
i-th whale i1, pi2, …, piM] which repre-
pi = [pi1 , pi2 , . . ., piM ] which
sents the represents
control parameters (voltage magnitude and
the control parameters (voltage magnitude and phase) for each IPFC for
phase) device.
each In thisdevice. In
IPFC
study, thethis
optimization problem focuses solely on the placement locations
study, the optimization problem focuses solely on the placement locations of the IPFC of the IPFC
devices, denoted
devices, as the position
denoted as the vector
position xi. vector
These x locations directly influence the system’s
i . These locations directly influence the system’s
performance objectives, such as minimizing active
performance objectives, such as minimizing active power power losses andlosses
maintaining voltage voltage
and maintaining
profiles. While theWhile
profiles. voltage themagnitude Vsn and phase
voltage magnitude Vsn and angle δsn angle
phase of the δIPFC are essential
sn of the IPFC are essential
parameters, they are treated as operational constraints rather
parameters, they are treated as operational constraints rather than design than design variables
variables as as stated
stated in Equations (9) and (10). These constraints ensure that the IPFC operates
in Equations (9) and (10). These constraints ensure that the IPFC operates within its feasible within its
feasible range,
range,maintaining
maintainingstability
stabilityand
andcompliance
compliancewith withsystem
systemrequirements.
requirements.
During the optimization
During process, the
the optimization position
process, theof each whale
position agent
of each is iteratively
whale up-
agent is iteratively up-
dated based on WOA’s
dated based onmathematical model. The
WOA’s mathematical updates
model. Theare influenced
updates by the encircling
are influenced by the encircling
behavior,behavior,
bubble-net hunting strategy,
bubble-net hunting and search
strategy, and forsearch
prey, which
for prey, ensure
which that the agents
ensure that the agents
converge converge
toward the globalthe
toward optimal
globalsolution. Each update
optimal solution. Eachmodifies the components
update modifies of li
the components of li
and ensures that the parameters in p are met, effectively exploring different locations
and ensures that the parameters in pi are met, effectively exploring different locations and
i and
control parameters for IPFC devices.
control parameters for IPFCThe flowchart
devices. in Figure in
The flowchart 2 summarizes the steps the
Figure 2 summarizes of steps of
WOA to find
WOA optimal
to findIPFC locations.
optimal IPFC locations.

Figure 2. WOA flowchart

Figure 2. WOA for IPFC allocation.
flowchart for IPFC allocation.

3. Results3.and
Results and Simulation
Simulation
After solving the power flow equations when IPFC is implemented to the system
using Newton Raphson method, WOA is applied to the IEEE 5, IEEE 14, and IEEE 57
bus systems [31] and the results are presented. Before discussing the results it is worth
mentioning that some assumptions were considered in this paper. Firstly, it is assumed that
Energies 2024, 17, 6318 10 of 20

the PFC voltage source converters are modeled as ideal. This means that switching losses
and operational limits were not considered. On the other hand, the generations and loads
modeled are considered symmetrical. In addition, the transient behaviors of the generators
and the IPFCs themselves, are out of the context of this paper. Finally, the load is kept
constant for most parts of the paper. However, load variations within the IEEE 14-bus
systems are considered to ensure WOA efficiency for the load variations. The simulations
were conducted on a machine with the following specifications: Intel Core i7 processor
(2.6 GHz, 8 cores), 16 GB RAM, and a Windows 10 operating system. The implementation
of the three optimization algorithms was carried out using MATLAB R2024b.
WOA parameters selection is an important part in making the algorithm efficient
and fast. These parameters include the population size, the number of iterations, and the
conversion criteria. The initial population size plays a crucial role in balancing exploration
and exploitation processes. The population size that is used in the paper is 50 agents. This
choice provides enough diversity to explore the search space effectively while ensuring the
required computational efficiency. However, to understand the sensitivity of changing the
population size, this size is changed from 10 to 50 agents to ensure that the choice of 50
is sufficient for the paper purpose of reducing the active power losses in the transmission
system. The number of iterations in the WOA is 100 iterations. It is to be noted that in
other systems or problems, the number of iterations may reach 500. However, for the IPFC
allocation within the 5, 14, and 57 bus systems, 100 iterations were sufficient. Finally, the
conversion criteria used was 10−7 .
The parameter settings for GA and PSO were carefully chosen to provide a fair
comparison with WOA. For GA, a population size of 50, 100 generations, a crossover
rate of 0.8, and a mutation rate of 0.1 were used, aligning with standard configurations
in similar optimization studies. For PSO, the inertia weight was 0.7 over 100 iterations,
with cognitive and social parameters set to 2.0. These configurations ensure consistent and
robust performance, enabling a meaningful evaluation of the comparative effectiveness
of WOA.
These optimization parameters are used to optimally allocate IPFC in IEEE 5, IEEE 14,
and IEEE 57 bus power systems in order to reduce the active power losses and enhance
the voltage profile. Then the results produced by WOA are compared with GA and PSO to
prove the superiority of WOA. The IEEE 5-bus system WOA results are compared with
similar GA results, while the IEEE 14-bus system WOA results are compared to GA and
PSO enhancement results. The enhancements that WOA make ensure that the constraints
of the voltage magnitude and its angle will follow the constraints of the algorithms as
stated in equations 8–10 in the previous section.

3.1. Technical Analysis of IPFC Implementation

The 14-bus system is considered first. WOA is used to find the best locations for
IPFC in this system. The active power losses before and after IPFC is used in the power
system are presented in Table 1 and shown in Figure 3 for each one of the 14 possible
locations of IPFC in the system. Figure 3 illustrates a radar chart that shows the baseline
active power losses across the system before applying the IPFC allocation optimization
(the blue line). After applying IPFC, the figure shows that the losses decreased (orange
line). It is to be noted that as the curve gets closer to the origin (center of the circle), the
losses become less (better). This figure serves as a reference for evaluating the impact of
the optimization process. The values depicted highlight the initial inefficiencies in the
power system, demonstrating significant opportunities for improvement through optimal
placement of IPFCs. These baseline losses provide a benchmark against which the results
of the WOA are compared in subsequent figures, demonstrating the effectiveness of the
proposed optimization approach. Table 1 shows the location of the IPFC that achieved the
most active power enhancement.
 GA. It is also worth mentioning that the cost function in this study focuses on the reduc-
tion of active power losses.

Energies 2024, 17, 6318 Table 1. Active power losses with and without using IPFC for IEEE 14-bus system. 11 of 20

IPFC Position Transmission Line Loss

Location VSC 1 VSC 2 Without-IPFC With-IPFC Improvement
Table 1. Active power losses with and without using IPFC for IEEE 14-bus system.
Between Between Ploss (MW) Ploss (MW) %
1 1-2 IPFC Position
1-3 1.300 Transmission
1.300 Line Loss 0.008
2
Location 2-1VSC 1 2-3 VSC 2 1.300
Without-IPFC 1.290
With-IPFC 0.769
Improvement
3 2-1
Between 2-4 Between 1.300
Ploss (MW) 1.290
Ploss (MW) 0.769
%
41 2-1 1-2 2-5 1-3 1.3001.300 1.274
1.300 1.977
0.008
52 2-3 2-1 2-4 2-3 1.3001.300 1.290
1.282 0.769
1.385
3 2-1 2-4 1.300 1.290 0.769
64 2-3 2-1 2-5 2-5 1.3001.300 1.275
1.274 1.923
1.977
75 2-4 2-3 2-5 2-4 1.3001.300 1.282
1.276 1.385
1.846
6 2-3 2-5 1.300 1.275 1.923
87 3-1 2-4 3-2 2-5 1.3001.300 1.294
1.276 0.462
1.846
98 3-1 3-1 3-4 3-2 1.3001.300 1.293
1.294 0.538
0.462
9 3-1 3-4 1.300 1.293 0.538
10
10
3-2 3-2
3-4 3-4
1.300 1.300
1.292
1.292
0.615
0.615
11
11 4-2 4-2 4-3 4-3 1.3001.300 1.292
1.292 0.615
0.615
12
12 4-2 4-2 4-5 4-5 1.3001.300 1.285
1.285 1.154
1.154
13 4-3 4-5 1.300 1.289 0.846
13
14 4-3 5-2 4-5 5-4 1.3001.300 1.289
1.280 0.846
1.538
14 5-2 5-4 1.300 1.280 1.538

1
14 1.30 2
1.28
13 1.26 3
1.24
12 1.977 % 4
1.22
1.20
11 5

10 6

9 7
8
Without IPFC With IPFC

Figure3.3.IEEE
Figure IEEE14-bus
14-bussystem
systemactive
activepower
powerlosses
losseswith
withand
andwithout
withoutusing
usingIPFC.
IPFC.

TableAfter WOA
2. Active waslosses
power used comparison
to reduce the active
between power
WOA andlosses
GA. in the transmission lines, a
comparison between WOA results was performed with other algorithms to present the
LineofLoss GA WOA
relative superiority using WOA in this study. This comparison was made between WOA
andLocation
GA on basisWithout
of active powerWith-IPFC Improvement
losses reduction With-IPFC
in IEEE 14-bus system. Improvement
When GAIPFC (MW)
was used (MW)
to optimize the location of%IPFCs in the (MW) % it can
IEEE 14-bus system,
be observed
1 that the improvement
1.300 achieved by WOA
1.300 0.000 in the active power loss reduction
1.300 0.008 is
better 2than the improvement
1.300 of GA when
1.289 both algorithms
0.846 were used
1.290 to optimize a similar
0.769
system with similar parameters. These differences are shown in Table 2 and can be visually
3 1.300 1.293 0.538 1.290 0.769
illustrated in Figure 4.
4 1.300 1.280 1.538 1.274 1.977
5 1.300 1.284 1.231
Table 2. Active power losses comparison between WOA and GA. 1.282 1.385
6 1.300 1.278 1.692 1.275 1.923
7 1.300
Line Loss 1.279 GA 1.615 1.276 WOA 1.846
Location Without With-IPFC Improvement With-IPFC Improvement
8 1.300 1.300 0.000 1.294 0.462
IPFC (MW) (MW) % (MW) %
9 1.300 1.300 0.000 1.293 0.538
1 1.300 1.300 0.000 1.300 0.008
2 1.300 1.289 0.846 1.290 0.769
3 1.300 1.293 0.538 1.290 0.769
4 1.300 1.280 1.538 1.274 1.977
Energies 2024, 17, 6318 12 of 20

Table 2. Cont.

Line Loss GA WOA

Location Without With-IPFC Improvement With-IPFC Improvement
IPFC (MW) (MW) % (MW) %
Energies 2024, 17, x FOR PEER REVIEW 5 1.300 1.284 1.231 1.282 1.385
13 of 21
6 1.300 1.278 1.692 1.275 1.923
7 1.300 1.279 1.615 1.276 1.846
8 1.300 1.300 0.000 1.294 0.462
9 1.300 1.300 0.000 1.293 0.538
10
10 1.300
1.300 1.293
1.293 0.538
0.538 1.292
1.292 0.615
0.615
11
11 1.300
1.300 1.291
1.291 0.692
0.692 1.292
1.292 0.615
0.615
12 1.300 1.289 0.846 1.285 1.154
12
13
1.300
1.300
1.289
1.292
0.846
0.615
1.285
1.289
1.154
0.846
13
14 1.300
1.300 1.292
1.299 0.615
0.077 1.289
1.280 0.846
1.538
14 1.300 1.299 0.077 1.280 1.538

1
14 2
13 3
1.977 %
12 4

11 5
Without IPFC 1.692 %
10 6
With IPFC-GA
9 7
With IPFC-WOA 8

Figure4.4.IEEE
Figure IEEE14-bus
14-bussystem
systemactive
activelosses
lossescomparison
comparisonbetween
betweenGA
GAand
andWOA.
WOA.

TableFigure
3. IEEE4 5-bus
displays system a radar chart
voltage comparing
values the percentage
with and without power loss across various
using IPFC.
buses in three different scenarios: without IPFC (in red), with IPFC using the GA optimiza-
Voltage (pu) Improvement
tion (in orange), and withBus IPFC using the WOA (in dashed black). The chart highlights the
Bus No.
effectiveness of IPFC placement Without With
Voltage in reducing power losses across the system. %
The red line shows the highest losses IPFC when no IPFC IPFC
is installed, indicating relative
1
inefficiencies in power flow 1.060management.1.060 1.060
With the implementation of IPFC0.00
optimized by
GA (orange2 line), a noticeable 1.045 reduction in 1.045
power losses is 1.045
observed. However, 0.00the dashed
black line3 representing IPFC 1.000placement optimized
1.044 by WOA shows the lowest0.12
1.042 losses among
the three4scenarios, demonstrating1.000 the superiority
1.037 of WOA1.037in minimizing system
0.07 losses,
with WOA 5 achieving the1.000 highest power loss 1.03enhancement
6 percentage of 1.977%
1.033 0.24compared
to 1.692% for GA.
This Figure underscores 1 the effectiveness of WOA in optimizing IPFC placement,
achieving a more efficient 1.0700 power flow and lower system losses than GA or no IPFC. From
this Figure, it can be1.0500 shown that the improvement of the WOA is relatively greater than
GA in reducing the active 1.0300 power losses of the system. The relative improvement of WOA
compared to GA in the 1.0100
power losses for optimal IPFC location is around 15%.
0.9900
The next
5 step involves analyzing the bus 2 voltages of the IEEE 5-bus system, with
0.9700
the WOA employed0.9500 to allocate the IPFC. It is important to emphasize that the primary
objective of this study is not to improve the voltage profile but to ensure that the voltage at
0.9300
each bus remains within acceptable limits during the algorithm’s search for the optimal
IPFC locations to minimize active power losses.
The IEEE 5-buses voltages with and without IPFC controllers are shown in Table 3
and Figure 5. Figure 5 presents a radar chart comparing voltage levels at key buses under
two different scenarios:
4 bus voltage without3 IPFC (in blue), and bus voltage with IPFC (in
dashed black).Reference
The figure
Voltage Busillustrates the effect ofWithout-IPFC
Bus Voltage IPFC placement on voltage stability across
the buses. This figure
Bus Voltage highlights the effectiveness of IPFC in maintaining voltage levels
With-IPFC

Figure 5. IEEE 5-bus system voltage values with and without using IPFC.
 1.977 %
12 4

11 5
Energies 2024, 17, 6318 Without IPFC 1.692 % 13 of 20
10 6
With IPFC-GA
9 7
With IPFC-WOA 8
closer to the desired reference, reducing deviations and thus improving the overall voltage
stability
Figure 4.and
IEEEperformance
14-bus systemofactive
the power system. between GA and WOA.
losses comparison

Table3.3.IEEE
Table IEEE5-bus
5-bussystem
systemvoltage
voltagevalues
valueswith
withand
andwithout
withoutusing
usingIPFC.
IPFC.
Voltage
Voltage (pu)(pu) Improvement
Improvement
Bus No.
Bus
Bus
Bus No. Voltage Without
Without WithWith
Voltage IPFC %%
IPFC IPFCIPFC
11 1.060
1.060 1.060
1.060 1.0601.060 0.000.00
22 1.045
1.045 1.045
1.045 1.0451.045 0.000.00
3 1.000 1.044 1.042 0.12
4
3 1.000
1.000
1.04 4
1.037
1.0421.037 0.120.07
54 1.000
1.000 1.03 7
1.036 1.0371.033 0.070.24
5 1.000 1.036 1.033 0.24

1
1.0700
1.0500
1.0300
1.0100
0.9900
5 0.9700 2
0.9500
0.9300

4 3
Reference Voltage Bus Bus Voltage Without-IPFC
Bus Voltage With-IPFC

Figure5.5.IEEE
Figure IEEE5-bus
5-bussystem
systemvoltage
voltagevalues
valueswith
withand
andwithout
withoutusing
usingIPFC.
IPFC.

On the other hand, the IEEE 5-buses bus voltages are studied when GA and WOA
algorithms were used to allocate IPFC. The buses voltages without IPFC, and with IPFC
allocated using each algorithm separately are shown in Table 4 and Figure 6.

Table 4. IEEE 5-bus system voltage values comparison between WOA and GA.

GA WOA
With IPFC With IPFC
Bus No. Voltage Bus
Bus Deviation Bus Deviation
Voltage pu % Voltage pu %
1 1.060 1.060 0.00 1.060 0.00
2 1.045 1.045 0.00 1.045 0.00
3 1.000 1.043 4.33 1.042 4.23
4 1.000 1.037 3.71 1.037 3.67
5 1.000 1.035 3.49 1.033 3.34

Figure 6 compares the voltage profiles of the IEEE-5 bus system under three scenarios:
the reference voltage (red line), voltage after IPFC allocation using the GA (orange line),
and voltage after IPFC allocation using the WOA (dashed black line). The chart shows
that both GA and WOA improve the voltage profile, bringing bus voltages closer to the
reference values. However, the WOA-optimized allocation demonstrates better alignment,
particularly at buses 1, 2, and 5, indicating superior performance in maintaining voltage
stability while achieving optimization objective of reducing active power losses.
 Voltage pu % Voltage pu %
1 1.060 1.060 0.00 1.060 0.00
2 1.045 1.045 0.00 1.045 0.00
3 1.000 1.043 4.33 1.042 4.23
Energies 2024, 17, 6318 4 1.000 1.037 3.71 1.037 3.67
14 of 20
5 1.000 1.035 3.49 1.033 3.34

1
1.1000
1.0500
1.0000
0.9500
5 0.9000 2
0.8500
0.8000

4 3
Reference GA WOA

Figure6.6.IEEE
Figure IEEE5-bus
5-bussystem
systemvoltage
voltagevalues
valuescomparison
comparisonbetween
betweenGA
GAand
andWOA.
WOA.

To examine
These tables and the figures
WOA algorithm
show a slight withimprovement
larger systems, application
in the of the for
voltage profile proposed
WOA
optimization
over GA algorithm. technique was
It is to be performed
noted that the on IEEE
deviation 14-bus and IEEE as
is enhanced 57-bus systems.
the values get Firstly,
closer
toa comparison
1 p.u. It alsobetween
has to beWOA notedand thatPSO techniques
while these results based areonrelatively
active powerclose,losses reduction
one shall take
into
andaccount
voltage the valuessuperiority of theisWOA
of the buses lies in its
conducted. Thefaster computational
bus voltage values of timethecompared
IEEE 14-bus to
GA. It is also
systems were worth
studied mentioning
considering thatthe
thevoltage
cost function
valuesinachieved
this study focuses
when PSOon andtheWOA
reduction
were
of active
used. power
Table losses.the voltage values at each bus in the IEEE 14-bus system. As can be
5 shows
To examine
seen from the table, the both
WOAalgorithms
algorithmwere withablelarger systems,
to keep application
the buses voltages ofwithin
the proposed
accepta-
optimization technique was performed on IEEE 14-bus and
ble limits. Nonetheless, WOA shows a very slight improvement in the voltage values. IEEE 57-bus systems. Firstly,
It ais
comparison
to be noted between
that the cost WOA and PSO
function techniques
presented in thebased on active
previous power
section lossesonreduction
focuses lowering
and
the voltage
active powervalueslosses.
of the Table
buses 6isshows
conducted. The bus voltage
a comparison for activevalues
power of the IEEE
losses. 14-bus
Since the
systems
numberwere studiedscenarios
of possible considering the voltage
is large, only some values achieved
selected when
results PSO
are and WOA
shown in the were
table.
used.
WhileTable
PSO 5is shows
noticedthe to voltage
reduce the values at power
active each bus in the
losses IEEE
in the 14-busWOA
system. system. As can be
outperformed
seen from the table, both algorithms were able to keep the buses
PSO in some cases as shown in the table. The base active power loss in IEEE 14-bus system voltages within acceptable
limits. Nonetheless,
is 486.14 MW. It is toWOA shows
be noted thata very slight improvement
the superiority of WOA isinshown the voltage
when values.
the IEEEIt57- is
to be noted that the
bus system is discussed below.cost function presented in the previous section focuses on lowering
the active powerthe
To ensure losses. Table 6ofshows
robustness a comparison
IPFC allocation for active
process poweragainst
using WOA losses. load
Sincevaria-
the
number of possible scenarios is large, only some selected results
tions, the technique was examined with reducing and increasing the load in the IEEE-14 are shown in the table.
While PSO is The
bus system. noticedloads to in
reduce the active
the IEEE 14-buspower
system losses in thefrom
is varied system.
75%WOA to 150% outperformed
of the rated
PSO in some cases as shown in the table. The base active
load. With different load levels, the WOA was applied to optimally allocated power loss in IEEE 14-bus system
IPFCs, the
isresults
486.14areMW. It is to be noted
displayed in Tables 7–9. that the superiority of WOA is shown when the IEEE 57-bus
system is discussed
From the results below.
shown in these tables, it can be concluded that load variation does
not cause any sensitivity or problems in the application of the algorithm. The system was
Table 5. IEEE 14-bus system voltage values comparison between PSO and WOA.
successful in maintaining the minimum active power losses while keeping the voltage
values within limits. PSO WOA
Bus With IPFC With IPFC
Bus No. Voltage
pu Bus Deviation Bus Deviation
Voltage pu % Voltage pu %
1 1.060 1.060 0 1.060 0
2 1.045 1.045 0 1.045 0
3 1.010 1.010 0 1.010 0
4 1.018 1.018 0.04 1.018 0.06
5 1.020 1.020 0.05 1.010 0.1
6 1.070 1.070 0 1.070 0
7 1.062 1.062 0.02 1.062 0.03
8 1.090 1.090 0 1.090 0
9 1.056 1.056 0.02 1.056 0.02
10 1.051 1.051 0.02 1.051 0.02
11 1.057 1.057 0.01 1.057 0.01
12 1.055 1.055 0 1.055 0
13 1.050 1.050 0 1.050 0
14 1.036 1.036 0.01 1.036 0.01
Energies 2024, 17, 6318 15 of 20

Table 6. Active power losses enhancement comparison between PSO and WOA in IEEE 14-bus
system.

PSO WOA
VSC 1 VSC 2 With IPFC With IPFC
Bus No. Location Location
Power Improvement Power Improvement
Loss (kW) % Loss (kW) %
1 2–3 2–4 477.23 1.84 477.62 1.76
2 1–5 5–6 476.91 1.90 477.7 1.74
3 4–5 12–13 477.43 1.80 477.05 1.88
4 12–13 7–8 477.19 1.84 476.92 1.90
5 7–8 7–8 477.68 1.74 477.8 1.72
6 2–5 4–9 477.75 1.73 477.7 1.74
7 6–13 13–14 477.26 1.83 477.51 1.78
8 1–2 13–14 477.46 1.79 477.07 1.87
9 1–5 7–8 476.85 1.92 477.08 1.87
10 5–6 2–3 477.61 1.76 477.52 1.78
11 1–2 9–10 477.09 1.87 477.48 1.79
12 7–8 4–9 477.38 1.81 477.71 1.74
13 4–9 3–4 477.33 1.82 477.07 1.87
14 13–14 4–5 477.49 1.78 477.68 1.75
15 1–5 11–10 477.63 1.76 477.13 1.86
16 9–10 9–10 476.94 1.90 476.92 1.90
17 2–4 1–5 477.27 1.83 477.3 1.82
18 12–13 9–14 476.88 1.91 477.06 1.87
19 1–2 9–10 477.08 1.87 477.67 1.75
20 4–5 12–13 477.07 1.87 477.73 1.73

To ensure the robustness of IPFC allocation process using WOA against load variations,
the technique was examined with reducing and increasing the load in the IEEE-14 bus
system. The loads in the IEEE 14-bus system is varied from 75% to 150% of the rated load.
With different load levels, the WOA was applied to optimally allocated IPFCs, the results
are displayed in Tables 7–9.

Table 7. Active power losses enhancement with 25% load increase in IEEE 14-bus system.

Loss Percentage
VSC1 VSC2 Base
Case with IPFC Enhancement
Location Location Loss (kW)
(kW) %
1 1–5 1–5 452.30 444.57 1.71
2 4–5 4–5 452.30 447.60 1.04
3 3–4 3–5 452.30 447.65 1.03
4 2–5 2–5 452.30 447.84 0.99
5 4–9 4–9 452.30 447.95 0.96
6 9–14 9–14 452.30 448.20 0.91
7 6–11 6–11 452.30 448.31 0.88
8 10–11 10–11 452.30 448.38 0.87
9 6–13 6–13 452.30 448.39 0.86
10 7–8 7–8 452.30 448.43 0.86

From the results shown in these tables, it can be concluded that load variation does
not cause any sensitivity or problems in the application of the algorithm. The system was
successful in maintaining the minimum active power losses while keeping the voltage
values within limits.
From the results in the above tables, it can be concluded that IPFC allocation using
WOA is not sensitive to load variations. However, when compared to the results obtained
from PSO algorithm, WOA achieved slightly better performance in active power losses.
This can be interpreted to the efficiency of PSO in relatively small systems. For IEEE 14-bus
systems, both PSO and WOA achieved 1.71% losses reduction while GA achieved only 1%.
However, WOA proves its superiority when applied to larger systems. PSO and WOA were
Energies 2024, 17, 6318 16 of 20

applied to IEEE 57-bus system. WOA achieved a total enhancement of 19% over PSO which
presents reduced enhancement for the same system. Table 10 presents these enhancement
results; it also compares the computation time needed for convergence for each algorithm.
It can be concluded that WOA achieves best active power losses enhancement compared to
PSO and GA.
Table 8. Active power losses enhancement with 50% load increase in IEEE 14-bus system.

Percentage
VSC1 VSC2 Base Loss with
Case Enhancement
Location Location Loss (kW) IPFC (kW)
%
1 1–5 1–5 712.88 700.24 1.77
2 4–5 4–5 712.88 705.26 1.07
3 3–4 3–4 712.88 705.37 1.05
4 2–5 2–5 712.88 705.61 1.02
5 4–9 4–9 712.88 705.85 0.99
6 6–11 6–11 712.88 706.36 0.92
7 6–13 6–13 712.88 706.46 0.9
8 7–8 7–8 712.88 706.52 0.89
9 5–6 5–6 712.88 706.59 0.88
10 7–9 7–9 712.88 706.81 0.85

Table 9. Active power loss enhancement with 25% load decrease in IEEE 14-bus system.

Loss Percentage
VSC1 VSC2 Base
Case with IPFC Enhancement
Location Location Loss (kW)
(kW) %
1 1–5 1–5 352.18 346.26 1.68
2 4–5 4–5 352.18 348.57 1.03
3 3–4 3–4 352.18 348.6 1.02
4 2–5 2–5 352.18 348.76 0.97
5 6–11 6–11 352.18 349.13 0.87
6 7–8 7–8 352.18 349.22 0.84
7 13–14 13–14 352.18 349.36 0.80
8 7–9 7–9 352.18 349.37 0.80
9 2–4 2–4 352.18 350.12 0.59
10 1–2 1–2 352.18 350.63 0.44

Table 10. Performance and computational time comparison of GA, PSO, and WOA for IEEE 57-bus
system.

Technique Base Case GA PSO WOA

Active Power
673.51 667.14 664.14 661.75
Losses (kW)
Computational
N/A 239.45 502.92 528.38
Time (s)

From the above table, it can be concluded that PSO is the fastest optimization technique
for smaller systems. However, it does not ensure relative optimal results for larger systems.
On the other hand, WOA shows better convergence when compared to GA which makes it
the optimal fit for large power systems applications. The results of the study demonstrate
that the WOA outperformed both the GA and PSO in minimizing active power losses
and maintaining voltage stability across all test systems. The solutions produced by WOA
consistently demonstrated lower total active power losses compared to those generated by
GA and PSO. This improved performance is attributed to WOA’s effective balance between
exploration and exploitation, achieved through its adaptive bubble-net hunting strategy,
which prevents premature convergence to suboptimal solutions. Unlike GA, which relies
 study demonstrate that the WOA outperformed both the GA and PSO in minimizing ac-
tive power losses and maintaining voltage stability across all test systems. The solutions
produced by WOA consistently demonstrated lower total active power losses compared
to those generated by GA and PSO. This improved performance is attributed to WOA’s
Energies 2024, 17, 6318 effective balance between exploration and exploitation, achieved through its adaptive 17 of 20
bubble-net hunting strategy, which prevents premature convergence to suboptimal solu-
tions. Unlike GA, which relies on crossover and mutation, and PSO, which involves ve-
locity
on updates,
crossover andWOA’s simpler
mutation, and mathematical
PSO, which involves foundation and updates,
velocity parameterWOA’s adaptation
simpleren-
hance its ability
mathematical to avoid local
foundation and optima
parameter and identify
adaptation globally optimal
enhance solutions.
its ability Theselocal
to avoid find-
ings highlight
optima the robustness,
and identify adaptability,
globally optimal and computational
solutions. These findingsefficiency
highlightofthe WOA, making
robustness,
it a reliable and
adaptability, andeffective optimization
computational tool for
efficiency of IPFC
WOA,placement
making itinapower
reliablesystems.
and effective
The WOA
optimization toolsensitivity for population
for IPFC placement size issystems.
in power studied. It was proven that when the pop-
TheisWOA
ulation greatersensitivity for algorithm
than 20, the populationconverged,
size is studied.
and theIt results
was proven thatdoes
obtained when notthebe
population is greater
enhanced further. Inthan
this 20, the algorithm
paper, the population converged,
size wasand50.
theFigure
results7obtained does not
below depicts the
be enhanced further.
performance behaviorInofthisthepaper,
WOAthe population
versus population size size.
was Figure
50. Figure 7 below the
7 illustrates depicts
rela-
the performance
tionship betweenbehavior of the WOA
the population size of versus
the WOA population size. achievable
and the most Figure 7 illustrates
enhancement the
relationship between the population
in system performance. The x-axis size of the WOA
represents and the most
the population achievable
size, ranging enhancement
from 10 to 50
in system
agents, performance.
while the y-axis The x-axisthe
indicates represents
improvement the population size, ranging
in performance from 10
metric, which to 50
is active
agents, while the y-axis indicates the improvement in performance
power loss reduction. While the figure shows that performance improvement begins to metric, which is active
power
saturateloss reduction.
after Whilesize
a population the of
figure shows
20, the thatofperformance
choice 50 agents inimprovement
the paper wasbegins madeto to
saturate after a population
ensure sufficient exploration size of 20,
of the the space,
search choice particularly
of 50 agentsfor in larger
the paper was made
and more complex to
ensure
systems sufficient
such asexploration
the IEEE-57ofbus thesystem.
search space,
Using particularly for largersize
a larger population andallows
more complex
the algo-
systems such as the IEEE-57 bus system. Using a larger population size
rithm to avoid premature convergence and ensures robust performance across all test sys- allows the algorithm
to avoid
tems, premature
even at the costconvergence
of slightly and ensures computational
increased robust performance effort.across all test systems,
This choice reflects a
even at the cost of slightly increased computational effort. This choice
trade-off prioritizing solution quality and consistency for scalability across different reflects a trade-off
prob-
prioritizing
lem sizes. solution quality and consistency for scalability across different problem sizes.

2
Most achievable
enhancement

1.5

0.5

0
10 20 30 40 50
Population size

Figure7.7.Active
Figure Activepower
powerlosses
lossesenhancement
enhancementconvergence
convergenceofofWOA
WOAversus
versuspopulation
populationsize.
size.

3.2. Economic Analysis of IPFC Implementation

To complement the technical evaluation of IPFC placement, an economic framework is
introduced to assess the feasibility of implementation. The analysis includes an estimation
of the investment costs for IPFC devices and the corresponding benefits from power loss
reduction.
1. Investment costs: The total investment cost for IPFC implementation Cinv is calculated
as:
Cinv = NIPFC ∗ Cunit (22)
where NIPFC is the number of IPFC devices installed, and Cunit is the unit cost of an IPFC
device, including installation and maintenance. Based on industry data, Cunit is assumed to
range between $50,000 and $400,000 [32], depending on the device capacity and its control
scheme features.
2. Annual savings: The annual savings Sannual from reduced power losses are estimated
using:
Sannual = ∆Ploss ∗ Cenergy ∗ H (23)
where Ploss is the reduction in power losses (in MW), Cenergy is the cost of energy, assumed
to be $100/MWh, and H is the number of operating hours per year, assumed to be 8760 h.
3. Payback period: The payback period Tpayback for IPFC investment is calculated as:
Energies 2024, 17, 6318 18 of 20

Cinv
Tpayback = (24)
Sannual
This metric provides insight into the time required to recover the investment costs
through energy savings. For example, if 2 IPFC (with $100,000 each) were implemented in
the IEEE 57-bus system, and if IPFC installation reduces the losses by 0.1 MW, this means
an 876 MWh a year. This makes the payback period around 2.5 years.
The results presented for the IEEE-57 bus system are provided as a representative
example to demonstrate the effectiveness and feasibility of IPFC implementation. This
choice does not result in any loss of generality for the smaller IEEE-5 and IEEE-14 bus
systems. Similar analyses conducted on these smaller systems would yield comparable
outcomes, with proportional reductions in active power losses and voltage profile im-
provements. Furthermore, the payback periods for the IPFC investment in these systems
are expected to fall within a similar range, reflecting the scalability and robustness of the
proposed optimization methodology. This consistency underscores the applicability of
the results across various system sizes, making the findings broadly relevant for practical
implementation in diverse network configurations.
The results indicate that the economic benefits of IPFC implementation, derived from
reduced power losses, justify the investment costs within a reasonable payback period.
Furthermore, the enhanced voltage profile and improved system stability provide addi-
tional operational benefits. In addition to the economic benefits, the implementation of
IPFCs offers significant technical advantages that contribute to the overall performance
and reliability of the power system. Beyond reducing power losses, IPFCs enhance sys-
tem voltage stability by improving the dynamic response to disturbances and ensuring
consistent power flow across transmission lines. Moreover, the optimized placement of
IPFCs helps maintain bus voltages within acceptable limits, thereby ensuring compliance
with operational standards. These benefits underscore the dual role of IPFCs in providing
both economic value and improved system reliability, making them a vital component for
modern power systems.

4. Conclusions
This paper demonstrated the effectiveness of the IPFC in reducing active power losses
and maintaining voltage profiles within transmission systems. Using WOA, optimal loca-
tions for IPFC deployment were identified in IEEE 5-bus, 14-bus, and IEEE 57-bus systems,
achieving significant improvements over other optimization techniques such as GA and
PSO. WOA demonstrated superior performance, with faster convergence and greater re-
ductions in active power losses, particularly in larger systems where its advantages were
clearer. Additionally, the robustness of the algorithm was validated under varying loading
conditions, highlighting its ability to maintain voltage stability in dynamic environments.
A financial study was also conducted, evaluating the investment costs of IPFC implemen-
tation relative to the achieved reduction in power losses. The results indicate that the
economic benefits of reduced power losses, combined with improved system reliability,
justify the deployment of IPFCs, with a reasonable payback period further supporting
their feasibility. Looking forward, future research could explore extending this method-
ology to even larger and more complex systems to assess wider scalability. Additionally,
integrating other FACTS devices or multi-objective optimization approaches that balance
power loss reduction with cost and reliability metrics would broaden the applicability of
this work. Further exploration of hybrid optimization methods and real-time implementa-
tion in dynamically evolving power systems, particularly those incorporating renewable
energy sources, could also provide new insights and improve system adaptability. These
directions ensure the continued evolution of IPFC and optimization techniques like WOA
in addressing the challenges of modern power grids.
Energies 2024, 17, 6318 19 of 20

Author Contributions: Conceptualization, A.M.A., M.A.H.-a., M.A. and D.A.; methodology, A.M.A.
and M.A.H.-a.; software A.M.A., M.A.H.-a. and M.A.; validation, M.A.H.-a., M.A. and D.A.; formal
analysis, M.A.H.-a., M.A. and D.A.; investigation, M.A.H.-a., M.A. and D.A.; resources, A.M.A.,
M.A.H.-a., M.A. and D.A.; data curation, A.M.A., M.A.H.-a., M.A. and D.A.; writing—original
draft preparation, A.M.A. and M.A.H.-a.; writing—review and editing, M.A.H.-a., M.A. and D.A.;
visualization, A.M.A. and M.A.H.-a.; supervision, M.A.H.-a., M.A. and D.A.; project administration,
M.A.H.-a.;. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Data Availability Statement: The original contributions presented in this study are included in the
article. Further inquiries can be directed to the corresponding author.
Acknowledgments: This paper was done during a sabbatical leave that was given by the University
of Jordan to both Mohammed A. Haj-ahmed and Dia Abualnadi.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Hingorani, N.G.; Gyugyi, L. Understanding FACTS: Concepts and Technology of Flexible AC Transmission Systems; IEEE Press: New
York, NY, USA, 2000; pp. 1–35.
2. Ahmad, A.A.; Sirjani, R. Optimal placement and sizing of multi-type FACTS devices in power systems using metaheuristic
optimization techniques: An updated review. Ain Shams Eng. J. 2020, 11, 611–628. [CrossRef]
3. Gyugyi, L.; Sen, K.; Schauder, C. The interline power flow controller concept: A new approach to power flow management in
transmission systems. IEEE Trans. Power Deliv. 1999, 14, 1115–1123. [CrossRef]
4. Chen, J.; Lie, T.; Vilathgamuwa, D. Basic control of interline power flow controller. In Proceedings of the Winter Meeting of the
Power Engineering Society, New York, NY, USA, 27–31 January 2002; pp. 521–525.
5. Singh, B.; Kumar, R. A comprehensive survey on enhancement of system performances by using different types of FACTS
controllers in power systems with static and realistic load models. Energy Rep. 2020, 6, 55–79. [CrossRef]
6. Alajrash, B.; Salem, M.; Swadi, M.; Senjyu, T.; Kamarol, M.; Motahhir, S. A comprehensive review of FACTS devices in modern
power systems: Addressing power quality, optimal placement, and stability with renewable energy penetration. Energy Rep. 2024,
11, 5350–5371. [CrossRef]
7. Xie, B.; Li, Z.; An, Y.; Zhao, J. The Coordinated Droop Control Strategy for Interline Power Flow Controller. In Proceedings of
the IEEE 6th Conference on Energy Internet and Energy System Integration (EI2), Chengdu, China, 11–13 November 2022; pp.
849–852.
8. Abbasipour, M.; Liang, X. Power Flow Study of MT-HVDC Grid Compensated by Multiport Interline DC Power Flow Controller.
IEEE Trans. Ind. Appl. 2023, 59, 4786–4796. [CrossRef]
9. Rezaei, N.; Kalantar, M.; Shayanfar, H.A.; Alipouri, Y.; Safari, A. Optimal IPFC signal selection and damping controller design
using a novel current injection model in a multi-machine power system. Int. J. Electr. Power Energy Syst. 2013, 44, 461–470.
[CrossRef]
10. Tecrathana, S.; Yokoyama, A.; Nakachi, Y.; Yasumatsu, M. An optimal power flow control method of power system by interline
power flow controller (IPFC). Int. Power Eng. Conf. Singap. 2005, 2, 1075–1080.
11. Ledwich, G.; Palmer, E. Energy function for power systems with transmission losses. IEEE Trans. Power Syst. 1997, 12, 785–790.
[CrossRef]
12. Al-Digs, A.; Chen, Y. Power System Loss Divider. IEEE Trans. Power Syst. 2020, 35, 3286–3289. [CrossRef]
13. Mohamed, K.H.; Rao, K.S.R.; Hasan, K.N.B.M. Optimal power flow and interline power flow controllers using particle swarm
optimization technique. In Proceedings of the TENCON IEEE Region 10 Conference, Singapore, 23–26 November 2009; pp. 1–6.
14. Khabbaz, M.M. Transmission power loss reduction using intelligent techniques-regulated IPFC. In Proceedings of the IEEE
Conference on Energy Conversion (CENCON), Johor Bahru, Malaysia, 13–15 October 2014; pp. 423–428.
15. Balasubbareddy, M. A solution to the multi-objective optimization problem with FACTS devices using NSHCSA including
practical constraints. In Proceedings of the IEEE International Conference on Power, Control, Signals and Instrumentation
Engineering (ICPCSI), Chennai, India, 21–22 September 2017; pp. 2615–2624.
16. Ahmed, A.S.; Attia, M.A.; Hamed, N.M.; Abdelaziz, A.Y. Comparison between genetic algorithm and whale optimization
algorithm in fault location estimation in power systems. In Proceedings of the 2017 Nineteenth International Middle East Power
Systems Conference (MEPCON), Cairo, Egypt, 19–21 December 2017; pp. 631–637.
17. Bohidar, S.; Nahak, N.; Mallick, R.K. Improvement of Dynamic Stability of power system by Optimal Interline Power Flow
Controller. In Proceedings of the Innovations in Power and Advanced Computing Technologies (i-PACT), Vellore, India, 22–23
March 2019; pp. 1–5.
18. Kumar, L.; Kar, M.K.; Kumar, S. Reactive Power Management by Optimal Positioning of FACTS Controllers using MFO Algorithm.
In Proceedings of the Emerging Trends in Industry 4.0 (ETI 4.0), Raigarh, India, 19–21 May 2021; pp. 1–6.
Energies 2024, 17, 6318 20 of 20

19. Li, Q.; Li, B.; Jiang, Q.; Liu, T.; Yue, Y.; Zhang, Y. A Novel Location Method for Interline Power Flow Controllers Based on Entropy
Theory. Prot. Control. Mod. Power Syst. 2024, 9, 70–81. [CrossRef]
20. Naidu T, P.; Venkateswararao, B.; Balasubramanian, G. Whale Optimization Algorithm based Optimal Power Flow: In View of
Power Losses, Voltage Stability and Emission. In Proceedings of the Innovations in Power and Advanced Computing Technologies
(i-PACT), Kuala Lumpur, Malaysia, 27–29 November 2021; pp. 1–6.
21. Gautam, A.; Ibraheem Sharma, G.; Kumawat, M.; Faraz, M.A. A novel solution for the power transmission congestion of
de-regulated power system using TCSC and TLBO algorithm. E-Prime Adv. Electr. Eng. Electron. Energy 2024, 8, 100592. [CrossRef]
22. Mirsaeidi, S.; Devkota, S.; Tzelepis, D.; Said, D.M.; Li, M.; Wang, X.; Zhang, P.; He, J. Optimization of FACTS Devices: Classification,
Recent Trends, and Future Outlook. In Proceedings of the IEEE 4th International Electrical and Energy Conference (CIEEC),
Wuhan, China, 28–30 May 2021; pp. 1–8.
23. Kar, M.K.; Kanungo, S.; Alsaif, F.; Ustun, T.S. Optimal Placement of FACTS Devices Using Modified Whale Optimization
Algorithm for Minimization of Transmission Losses. IEEE Access 2024, 12, 130816–130831. [CrossRef]
24. Reddy, B.R.; Reddy, Y.S.; Sujatha, P. Optimal placement of interline power flow controller (IPFC) to enhance voltage stability. In
Proceedings of the Conference on Power, Control, Communication and Computational Technologies for Sustainable Growth
(PCCCTSG), Kurnool, India, 11–12 December 2015; pp. 78–84.
25. Rajagopalan Thippana, S.V.C.; Parimi, A.M. Optimal Placement of the Interline Power Flow Controller using the Bees Algorithm
to minimize power loss. In Proceedings of the 4th International Conference on Electrical Energy Systems (ICEES), Chennai, India,
7–9 February 2018; pp. 212–217.
26. Thippana, V.C.; Parimi, A.M.; Karri, C. Placement of IPFC for Power Loss Reduction in Transmission lines using Firefly Algorithm.
In Proceedings of the IEEE 17th India Council International Conference (INDICON), New Delhi, India, 10–13 December 2020; pp.
1–6.
27. Vinkovic, A.; Mihalic, R. A current-based model of an IPFC for Newton–Raphson power flow. Electr. Power Syst. Res. 2009, 79,
1247–1254. [CrossRef]
28. Mirjalili, S.; Lewis, A. The Whale Optimization Algorithm. Adv. Eng. Softw. 2016, 95, 51–67. [CrossRef]
29. Gharehchopogh, F.S.; Gholizadeh, H. A comprehensive survey: Whale Optimization Algorithm and its applications. Swarm Evol.
Comput. 2019, 48, 1–24. [CrossRef]
30. Rana, N.; Latiff, M.S.; Abdulhamid, S.M.; Chiroma, H. Whale optimization algorithm: A systematic review of contemporary
applications, modifications, and developments. Neural Comput. Appl. 2020, 32, 16245–16277. [CrossRef]
31. Power Systems Test Case Archive. Available online: https://labs.ece.uw.edu/pstca/ (accessed on 1 November 2022).
32. Nextgen Highways Fact Sheet. Buried High-Voltage Direct Current (HVDC) Transmission is Cost Competitive. Available
online: https://nextgenhighways.org/wp-content/uploads/2023/01/NGH_Buried-HVDC-Cost-Competitive.pdf (accessed
on 1 November 2022).

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.