An Approach For Power Flow Analysis Based On Particle Swarm Optimization
An Approach For Power Flow Analysis Based On Particle Swarm Optimization
Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
An Approach for Power Flow Analysis
Based on Particle Swarm Optimization
Shivam Nanda; Pradeepti Lakra
Department of Electrical Engineering Jabalpur Engineering College Jabalpur (M.P. ), India
Abstract:- Power flow solution is the ultimate solution To solve the challenges mentioned above,
for determining the unknown parameters of a power computational intelligence methods such as Particle swarm
system. The suitable operation of the system depends on optimization (PSO) can be implemented for the voltage
knowing the losses in the system, like those of loads and stability assessment to solve the complexity of active and
transmission lines, before they are installed. This paper reactive power. This paper analyzes the load flow problem
presents a load flow analysis approach that uses particle in power system planning studies to stabilize system bus
swarm optimization algorithm. The method introduced voltage and minimize the corrosponding possible losses.
helps to find the best solution for improving the bus
voltage of the network and minimize corresponding II. LOAD FLOW ANALYSIS
losses. To find the optimal solution to the problem, an
IEEE 6 bus is used to present the case. In a power system, power flows from the generating
station to the load through different branches of the network
Keywords:- Voltage; Global Best; Position Best; Velocity; [1]. Power flow analysis is a fundamental technique used in
Fitness Function; Etc. electrical power systems to determine steady-state operating
conditions by calculating parameters such as voltages,
I. INTRODUCTION currents, active power and reactive power. The load flow
equations include power balance equations for both active
Load flow is an important tool used by power and reactive powers at each bus [5].The load flow problem
engineers for planning the best operation of a power system can be formulated as an optimization problem with the
[1]. The power system consists of an electric circuit that objective to find the voltage magnitudes and angles that
includes generators, transmission network, and distribution minimize the difference between the input and the output
network, where power flows from the generating station to power at each bus.
the load through different branches of the network. Power
flow studies provide a systematic mathematical approach to A. Voltage Stabilization
determine various bus voltages, phase angles, active and Voltage stabilization refers to minimizing the
reactive power flows through different branches, generators, difference in voltage level from a flat voltage profile in an
transformers, and loads under steady-state conditions [1]. In electrical system. The voltage stability problem refers to the
the present scenario, load-flow analysis faces several voltage drop that occurs when the power system experiences
challenges that need to be addressed. Firstly, modern power a heavy load. There are a few tools that can be used to
systems are becoming more complex with the integration of simulate the load flow of the system. In the static voltage
renewable energy sources, distributed generation, and stability study, optimization methods are the main analysis
advanced control technologies [2]. This complexity poses techniques, and they are used to find the voltage stability
challenges in accurately modeling the system components, margin or loading margin of the system [4],[5]. This is
capturing their dynamic behavior, and ensuring accurate important for various reasons, including the improvement of
load-flow solutions. Secondly, voltage stability and reactive the efficiency of the system. By minimizing voltage drop,
power management are critical to the operation of power the voltage level can be maintained closer to the desired
systems. Inadequate reactive power support or voltage value. Therefore, an approach considering an evolutionary
control can lead to voltage violations and even voltage computational technique such as PSO may be interesting to
collapse. Load-flow analysis needs to consider reactive analyze.
power sources, such as capacitor banks, voltage regulators,
and accurately model their impact on system voltage stability B. Loss Minimization
[3]. Thirdly, high penetration of nonlinear loads, such as In an electrical power system, losses occur due to
electronic devices and power electronic converters, is various factors, such as resistance in conductors, transformer
becoming more prevalent in modern bus systems. These losses, reactive power consumption, system inefficiencies,
loads introduce harmonic distortions, voltage fluctuations, and losses in transmission and distribution networks [23].
and power factor issues, which can impact load-flow Due to the rapid increase in demand for electricity,
calculations and require specialized modeling techniques. environmental constraints, and competitive energy market
scenarios, transmission and distribution systems are often
operated under heavily loaded conditions. This causes
concern about distribution system losses. Traditionally,
loss minimization focuses on optimizing network
IJISRT24FEB1283 www.ijisrt.com 1537
Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
configuration or reactive power support through capacitor Artificial Bee Colony Optimization (ABCO): Artificial
placement [21]. Possible remedies to reduce these losses bee colony optimization (ABC) is a swarm intelligence
include the implementation of reactive power compensation algorithm inspired by the intelligent foraging behavior of
devices such as static capacitors or synchronous condensers honey bees. It is a metaheuristic algorithm, which means
that can compensate for reactive power and reduce that it does not guarantee to find the optimal solution to
associated losses. Improving the system voltage profile not a problem, but it is often able to find good solutions in a
only reduces reactive power requirements but also reasonable amount of time. ABC works by maintaining a
demotesassociated losses [24]. population of artificial bees, which represent potential
solutions to the problem. The bees are divided into three
C. Optimization groups: employed bees, onlooker bees, and scout bees
Optimization can be defined as a mechanism through [12],[13]. It is efficient to hybridize ABC with other
which the maximum or minimum value of a given function techniques to increase its performance.
can be found. There are numerous optimization techniques
available across various domains. Optimization using
constraints in terms of reliability is found to be the best
option for optimizing structures with discrete parameters [9].
Fig.1 shows a few examples of optimization techniques, as
described below.
Particle swarm optimization (PSO): It is a computational
algorithm that simulates the social behavior of
individuals in a swarm to optimize solutions to various
problems.
Ant Colony Optimization (ACO): The ant colony
optimization algorithm (ACO) is an evolutionary meta-
heuristic algorithm based on a graph representation that
has been applied successfully to solve various hard Fig. 1: Optimization Techniques
combinatorial optimization problems. Initially proposed
by Marco Dorigo in 1992 [6], the main idea of ACO is to Studies show that particle swarm optimization is the
model the problem as the search for a minimum cost most commonly used optimization technique.
path in a graph. Artificial ants walk through this graph,
looking for good paths. Each ant has rather simple D. System Design and Analysis
behavior, so it will typically only find rather poor-quality An IEEE 6 bus system is used in this paper to analyze
paths on its own. Better paths are found as the emergent the presented load flow analysis method. A simplified
result of the global cooperation among ants in the colony graphical representation of an IEEE 6 bus power system is
[7],[8]. shown in Fig. 2. The bus data of the IEEE 6 bus system is
Genetic Algorithm (GA): A genetic algorithm (GA) is a shown in Table 1 below.
search heuristic that mimics the process of natural
selection to find optimal solutions to problems. GAs
are an example of mathematical technology transfer. By
simulating evolution, one can solve optimization
problems from a variety of sources. The principle of
GAs is to simply imitate genetics and natural selection
using a computer program. A genetic algorithm is a
problem-solving method that uses genetics as its model
of problem solving [9], [10], [11].
Table 1: Bus Data of the IEEE 6 Bus System
Bus No. Bus type Voltage Magnitude(p.u.) Voltage Angle (degrees)
1 SB 1.05 0
2 PV 1.05 0
3 PV 1.07 0
4 PQ 1 0
5 PQ 1 0
6 PQ 1 0
IJISRT24FEB1283 www.ijisrt.com 1538
Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Fig. 2: IEEE 6 Bus System Layout
III. PARTICLE SWARM OPTIMIZATION Swarm: It is apparently not a properly planned and
CONCEPTS AND FORMULATION controlled population of moving particles that seems to
move in a random direction in the problem space.
Many areas in power systems require solving one or Position: The position of each particle is updated by
more indirect optimization problems. While systematic adding the updated velocity to the current position of the
methods might suffer from slow convergence and the curse individual in the swarm. The position of the ith
of amplitude, heuristic-based swarm intelligence can be an individual at (k + 1)th iteration is found in the equation of
efficient alternative. Particle swarm optimization (PSO), is position.
known to effectively solve large- scale nonlinear Velocity: The velocity is updated by considering the
optimization problems. PSO (Particle Swarm Optimization) current velocity of the particle, the best fitness function
was first introduced by Eberhart and Kennedy (1995) and value of that particle, and the best fitness function value
used for the optimization of continuous nonlinear functions. among the particles in the swarm. The velocity of each
This paper presents a detailed overview of the basic particle is modified in each iteration [4].
concepts of PSO. Also, it provides an inclusive review of Inertia: It makes the particle move in the same direction
the power system applications that have benefited from the and with the same velocity. The greater the inertia of the
powerful nature of PSO as an optimization technique. For particle, the more force is required to change its motion.
each application, technical details that are required for Personal Influence: Improves individual particle position
applying PSO, such as its type, particle formulation by making them return to a position that is better than
(solution representation), and the most efficient fitness the current one.
functions, are also discussed [13]. It is discovered through Social Influence: Makes the particle follow the best-
the simulation of the social behavior of bird blocks. The neighbor direction.
swarm is composed of some volume-less particles with
Position best: The best individual position of a particle is
velocities, each of which represents a feasible solution in the defined as position best, which is the best position giving
solution space. The algorithm finds the optimal solution by the particle.
moving the particles in the solution space because of the
Global best: The best position of all the particles is
convenience of realization[14].
defined as the global best. The position of the best
particle among all the particles in the population is called
The basic elements of the PSO [11] technique are
the global best.
briefly stated and defined as follows:
Applications of PSO: PSO is a method of evolutionary
computation that can be realized conveniently [24]. It
Particle: These are the interacting elements for the
uses primary math operators and receives good results in
solutions that are potentially possible in the problem
static, noisy, and continuously changing environments.
space. These agents (particles) constitute a swarm
All these advantages make PSO applicable to more and
moving around in the search space looking for the best
more fields. Classification based on diverse categories of
solution. In standard PSO, each member of the
PSO applications in selected research studies is shown in
population is called a particle. Each particle contains its
Fig. 3.
position and a velocity [2].
IJISRT24FEB1283 www.ijisrt.com 1539
Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Fig. 3: Application of PSO
IV. OPTIMIZATION ALGORITHM Where
A. The Methodology of PSO Vik - Current velocity of particle i at iteration k
The PSO technique is a tool that solves optimization
problems and is based on the movement of swarms [14], w- Inertia weight
[15]. In general, it is investigated that a group of particles
move in a limited, diverse space and in the iterative search r and - Random number in the range [0, 1]
for the optimal solution, the position of each of them is a
potential solution. In each iteration, new positions are C1 and C2 - Acceleration coefficients
generated for the particles, which are obtained by means of a
speed calculated according to the best current position of xtk- Current position of particle i at iteration k
each particle and the best global position of the swarm. The
quality of the position of each particle is obtained by pbestt is the best position of the current particle
evaluating the objective function each time it takes a new achieved so far, while it gbestt is the global best position
position. This process is repetitive and ultimately leads to an achieved by all informants. The new position of each
optimal solution where all the particles converge [16]. The particle is given by Eq. 2.
terminology used are the following:
xtk+1 = xt + vtk+1 (2)
xi = [xi1,xi2,........x in represents the position of the I th
particle, which has m components or coordinates. B. Algorithm and Mathematical Model
In this paper, the particles that constitute the swarm are
The best previous position of the ith particle is: defined by the voltage magnitude of the buses [14]. As the
IEEE 6 bus system is used to present the case, the
pbestt = pbestt1,pbestt2, ... ... , pbesttn the best global population consists of a total of 6 particles that are moving
position of the swarm (gbest) corresponds to the swarm at their own velocities in the search space. The particle
particle with the best search behavior. position is defined by the voltage, magnitide, and phase
angle of the bus. A swarm of particles updates their relative
Vi = Vi1, Vi2, ... . . Vin is the velocity of the ith positions from one iteration to another. To get the optimum
particle. solution, each particle moves towards its prior personal
best position (Pbest) and the global best position (Gbest) in
The swarm of particles evolves in the search space by the swarm [18], [19]. The personal best position of the
modifying their velocites and position of each particle in the particle ′𝑖′ is the best possible value of the particle among
next iteration are calculated by the expressions in eq 1: all the values of its position in different iterations. When
the voltage of a particle′𝑖′approaches near unity, the
(1) particle whose value is closest to unity is considered Pbest.
The global best is the position that is considered the best
among all the particles in the search space. The voltage
that is closest to unity among all the particles is considered
Gbest, as defined by Eq. 2 [11]. The flowchart of PSO
shown in fig.4.
IJISRT24FEB1283 www.ijisrt.com 1540
Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
The inertia weighting factor for the velocity of
particle'i' is defined by the inertial weight approach as in
(5)
Eq 3. wt = wmax − wmax− wminitermax × iter (3) Where NL is the number of transmission lines, gk is
the conductance of branch k between buses i and j, Vi tis
Where w max and w min are the upper and lower
voltage magnitude at i bus, ∅ij is the voltage angle
limits of the inertia weighting factor, respectively. w is
difference between bus i and j. It is observed that the losses
the inertia weight, a parameter that controls the impact of
in the system have changed from 3.984 p.u. to 0.1670 p.u.
the particle's current velocity on its future velocity.
by using the above equation. The optimization of voltage
involves minimizing the load bus voltage difference and,
hence, improving the voltage profile of the system.
(6)
Where w is the weighting factor, with the help of this
equation, we can reduce the voltage deviation. It is observed
that the voltage deviation is reduced from 0.173 p.u. to
0.1095 p.u. as given in Table 3. The objective function is
proposed in order to minimize losses and improve the
voltage profile. The objective function can be expressed in
the below Eq. 7.
(7)
F represents the fitness function, which evaluates the
quality or fitness of a particle's position in the search space.
The goal of PSO is to find the optimal solution by
maximizing or minimizing this fitness function, depending
on the problem. The fitness function F has successfully
decreased from 4.157 to 0.2765.
Each particle in the swarm represents the voltage
magnitude of the IEEE 6 bus system. The particles undergo
a change in velocity as per Eq. 1, and their position changes
Fig. 4: Flowchart of PSO as per Eq. 2 for each iteration. The voltage of each bus is
therefore updated after each iteration. The graphical
The particle swarm optimization algorithm uses a representation of the bus voltages is shown from Fig. 5 to
fitness function (𝑥) to evaluate the quality or fitness of a Fig. 10. The optimization of power loss in a system typically
potential solution. It takes the solution parameters (𝑥) as involves minimizing various sources of dissipation, such as
input and returns a single scalar value representing the resistive losses, switching losses, and other losses associated
fitness of the solution. with components and the operation of the system.
𝐹𝑖𝑡𝑛𝑒𝑠𝑠𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛 = 𝑓(𝑥) (4)
One of the commonly used equations for power loss
optimization is the power loss equation for a component, as
in Eq. 5 [22].
Table 2: Parameters of PSO
Parameters Values
Population size 06
C1 and C2 0.3
Inertia weight 00.3
Iteration 15
IJISRT24FEB1283 www.ijisrt.com 1541
Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Table 3: Summary of calculation results by the proposed method
IEEE bus data Particle Swarm Optimization
F1 (losses in p.u.) 1st iterations 15th iterations
F2 (Voltage Deviation in p.u.) 3.984 0.041 0.1670
Total F 0.173 0.15 0.1095
4.157 0.191 0.2765
V. SIMULATION AND RESULTS The basic configuration of the IEEE 6 bus system is
shown in Figure 2, and according to the parameter
In the PSO simulation, the optimization problem is sensitivity analysis for the IEEE 6 bus system (15 trials),
formulated as finding the best solution in a multi- based on the trials in Fig. 6, in horizontal direction shows
dimensional search space. The solution is represented as a voltage magnitude and the vertical direction shows
particle, and a population of particles (referred to as a iterations. On 13th iteration, the value decreased.
swarm) is created to explore the search space. Each particle
has a position and a velocity, which are updated iteratively
based on its own experience and the collective knowledge of
the swarm. This paper proposes load flow analysis using the
PSO algorithm. The voltage and loss optimization approach
was applied to an IEEE 6 bus power system. The parameters
used are as given in Table 2. The results obtained from PSO
depend on various parameters, such as the specific problem
being solved, the parameter setting of the algorithm (e.g.,
number of particles, maximum iterations, inertia weight,
acceleration coefficients), and the characteristics of the
search space. All these parameters are used in this paper to
reduce losses as well as optimize the voltage profile of the
system.
Fig. 7: Voltage magnitudes for the 3rd bus
The basic configuration of the IEEE 6 bus system is
shown in Figure 2, and according to the parameter
sensitivity analysis for the IEEE 6 bus system (15 trials),
based on the trials in Figure 7, in horizontal direction shows
voltage magnitude and the vertical direction shows
iterations. On 10th iteration, the value decreased.
Fig. 5: Voltage magnitudes for the 1st bus
The basic configuration of the IEEE 6 bus system is
shown in Figure 2, and according to the parameter
sensitivity analysis for the IEEE 6 bus system (15 trials),
based on the trials in Fig. 4, in horizontal direction shows
voltage magnitude and the vertical direction shows
iterations. On 11th iteration, the value decreased.
Fig. 8: Voltage magnitudes for the 4th bus
The basic configuration of the IEEE 6 bus system is
shown in Figure 2, and according to the parameter
sensitivity analysis for the IEEE 6 bus system (15 trials),
based on the trials in Fig. 8, in horizontal direction shows
voltage magnitude and the vertical direction shows
iterations. On 12th iteration, the value decreased.
Fig. 6: Voltage magnitudes for the 2nd bus
IJISRT24FEB1283 www.ijisrt.com 1542
Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Fig. 9: Voltage Magnitudes for the 5th Bus Fig. 10: Voltage Magnitudes for the 6th Bus
The basic configuration of the IEEE 6 bus system is The basic configuration of the IEEE 6 bus system is
shown in Figure 2, and according to the parameter shown in Figure 2, and according to the parameter
sensitivity analysis for the IEEE 6 bus system (15 trials), sensitivity analysis for the IEEE 6 bus system (15 trials),
based on the trials in Fig. 9, in horizontal direction shows based on the trials in Fig. 10, in horizontal direction shows
voltage magnitude and the vertical direction shows voltage magnitude and the vertical direction shows
iterations. On 4th iteration, the value decreased. iterations. On 13th iteration, the value decreased.
Table 4: Parameter Sensitivity Analysis for the IEEE 6 Bus System (15 Trials)
S.N Iterati on No Bus No Velocity Position S.N Iteration No Bus No Velocity Position
1 1 1 0.025∠-136.7 0.981∠-1.037 46 8 4 0.005∠88.36 0.910∠6.891
2 1 2 0.037∠119.4 0.979∠-2.387 47 8 5 0.∠0 0.91∠-5.60
3 1 3 0.018∠-174.1 0.982∠-4.753 48 8 6 0.006∠89.72 0.911∠-7.212
4 1 4 0.004∠-148.9 0.926∠-4.867 49 9 1 0.014∠-45.46 0.886∠6.234
5 1 5 0.∠0 0.91∠-5.60 50 9 2 0.024∠-94.22 0.909∠-0.752
6 1 6 0.006∠144.8 0.934∠-6.431 51 9 3 0.002∠122.8 0.913∠-6.460
7 2 1 0.031∠-131.9 0.961∠-2.434 52 9 4 0.002∠84.63 0.909∠-6.702
8 2 2 0.020∠-176.5 0.959∠-2.509 53 9 5 0.∠0 0.91∠-5.60
9 2 3 0.027∠-138.5 0.963∠-5.939 54 9 6 0.006∠87.26 0.910∠-6.642
10 2 4 0.007∠-130.3 0.920∠-5.22 55 10 1 0.008∠-15.8 0.893∠-63.20
11 2 5 0∠0 0.91∠-5.60 56 10 2 0.024∠-92.75 0.908∠-2.264
12 2 6 0.018∠-113.8 0.928∠-7.496 57 10 3 0.001∠11.34 0.912∠-63.56
13 3 1 0.024∠-133.6 0.944∠-3.566 58 10 4 0.001∠78.23 0.909∠-67.02
14 3 2 0.020∠145.5 0.924∠-1.861 59 10 5 0.0∠0 0.91∠-5.60
15 3 3 0.029∠-127.2 0.948∠-7.442 60 10 6 0.003∠93.36 0.909∠-6.624
16 3 4 0.004∠-136.6 0.916∠-5.034 61 11 1 0.005∠2.511 0.896∠0.013
17 3 5 0.0∠0 0.91∠-5.60 62 11 2 0.018∠-92.16 0.908∠-3.399
18 3 6 0.005∠-179.1 0.922∠-7.534 63 11 3 0.001∠103.8 0.911∠-6.297
19 4 1 0.026∠-118.9 0.933∠-5.030 64 11 4 0.001∠76.81 0.909-6.639
20 4 2 0.016∠-137.7 0.930∠-2.565 65 11 5 0∠0 0.91∠∠-5.60
21 4 3 0.012∠-172.6 0.936∠-7.636 66 11 6 0.002∠82.69 0.909∠-6.515
22 4 4 0.018∠-99.11 0.915∠-6.208 67 12 1 0.018∠-79.05 0.899∠-1.125
23 4 5 0.0∠0 0.91∠-5.60 68 12 2 0.013∠-91.77 0.908∠-4.281
24 4 6 0.005∠125.8 0.918∠-7.284 69 12 3 0.001∠95.14 0.910∠-6.235
25 5 1 0.011∠-135.6 0.860∠-11.06 70 12 4 0.001∠75.35 0.903∠0.104
26 5 2 0.015∠-122.7 0.922∠-3.386 71 12 5 0∠0 0.91∠-5.60
27 5 3 0.009∠-150.8 0.927∠-74.22 72 12 6 0.001∠77.61 0.909∠-6.454
28 5 4 0.023∠-96.75 0.915∠-76.48 73 13 1 0.019∠-83.01 0.901∠-0.076
29 5 5 0∠0 0.91∠--5.60 74 13 2 0.003∠-90.93 0.908∠--4.406
30 5 6 0.016∠-102.1 0.916∠-83.24 75 13 3 0.001∠86.28 0.909∠-6.172
31 6 1 0.024∠-127.4 0.849∠12.55 76 13 4 0.017∠-88.04 0.903∠-0.973
32 6 2 0.012∠-116.3 0.917∠-4.087 77 13 5 0∠0 0.91∠-5.60
33 6 3 0.008∠130.7 0.921∠-7.098 78 13 6 0.001∠75.69 0.909∠-6.39
34 6 4 0.002∠119.6 0.913∠-75.13 79 14 1 0.023∠86.03 0.902∠13.80
IJISRT24FEB1283 www.ijisrt.com 1543
Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
35 6 5 0.∠0 0.91∠-5.60 80 14 2 0.00∠-90.39 0.908∠-4.720
36 6 6 0.004∠105.7 0.914∠-80.83 81 14 3 0.001∠77.33 0.909∠-6.109
37 7 1 0.061∠-79.66 0.862∠-3.991 82 14 4 0.022∠-88.23 0.904∠-2.365
38 7 2 0.027∠-97.12 0.914∠1.679 83 14 5 0∠0 0.91∠-5.60
39 7 3 0.007∠116.6 0.917∠-6.725 84 14 6 0.003∠75.33 0.909∠-6.203
40 7 4 0.005∠94.60 0.911∠-7.205 85 15 1 0.047∠83.21 0.909∠4.310
41 7 5 0.∠0 0.91∠-5.60 86 15 2 0.008∠-59.02 0.912∠-51.27
42 7 6 0.008∠91.38 0.912∠-7.587 87 15 3 0.001∠75.64 0.909∠-6.046
43 8 1 0.029∠-65.34 0.876∠-56.55 88 15 4 0.018∠87.95 0.904∠-12.24
44 8 2 0.015∠-99.76 0.911∠0.725 89 15 5 0∠0 0.91∠-5.60
45 8 3 0.003∠123.8 0.915∠6.582 90 15 6 0.002∠75.28 0.909∠-6.090
VI. CONCLUSION [6]. Ünal, M., Ak, A., Topuz, V., Erdal, H. (2013). Ant
ColonyOptimization (ACO). In: Optimization of PID
Particle swarm optimization is an iterative optimization Controllers Using Ant Colony and Genetic
algorithm that simulates the social behavior of bird flocking Algorithms. Studies in Computational Intelligence,
or fish schooling. By iteratively updating particle position vol 449. Springer, Berlin, Heidelberg.
and velocity, the algorithm aims to converge towards the [7]. Stützle, Thomas, Manuel López-Ibáñez, Paola
optimal solution. This paper has successfully presented a Pellegrini, Michael Maur, Marco Montes de Oca, M.
power flow algorithm that implements PSO, where particles Birattari and Marco Dorigo. “Parameter Adaptation
represent the buses and position represents the voltage and in Ant ColonyOptimization.”
phase angle at each bus. The iterative procedure shows a [8]. Lopez-lbanez, M. (2018). Ant Colony
substantial improvement in the voltage profile of the system Optimization. Intelligent Systems.
as the voltage deviation is reduced from 0.173 p.u. to [9]. Deka, Paresh Chra. “Genetic Algorithm (GA).” A
0.1095 p.u. It is also observed that system line losses change Primer on Machine Learning Applications in
from 3.984 p.u. to 0.1670 p.u. Overall, the fitness function Civil Engineering (2019)
has improved from 4.157 p.u. to 0.2765 p.u. It is concluded [10]. Tang, Wallace Kit-Sang, Kim Fung Man, Sam Tak
that in the context of power flow analysis, PSO can help to Wu Kwong and Qianhua He. “Genetic algorithms
find the optimal or near optimal operating point of an and their applications.” IEEE Signal Process. Mag.
electrical network by adjusting the voltage and angle at 13 (1996): 22- 37.
each bus to minimize a given objective function (e.g., system [11]. Liepins, Gunar E. and Michael Hilliard. “Genetic
losses or voltage deviations). Overall, PSO offers a flexible algorithms: Foundations and applications.” Annals of
and powerful approach for power flow analysis, capable of Operations Research 21 (1990): 31-57.
efficiently exploring solution spaces and providing optimal [12]. Karaboğa, Derviş and Bahriye Basturk. “A powerful
or near-optimal solutions for complex electrical networks. and efficient algorithm for numerical function
optimization: artificial bee colony (ABC) algorithm.”
REFERENCES Journal of Global Optimization 39 (2007): 459-471.
[13]. Y. del Valle, G. K. Venayagamoorthy, S.
[1]. Afolabi, Olukayode & Ali, Warsame & Cofie, P. & Mohagheghi, J. - Hernandez and R. G. Harley,
Fuller, John & Obiomon, Pamela & Kolawole, "Particle Swarm Optimization: Basic Concepts,
Emmanuel. (2015).Analysis of the Load Flow Variants and Applications in Power Systems," in
Problem in Power System Planning Studies. Energy IEEE Transactions on Evolutionary Computation,
and Power Engineering. vol. 12, no. 2, pp. 171-195, April 2008.
[2]. Mageshvaran, R. & Raglend, Jacob & Yuvaraj, V. & [14]. J. Kennedy, E. Russell, "Particle Swarm
Rizwankhan, P. & Vijayakumar, T. & Sudheera,. Optimization", Proceedings of IEEE International
(2008). Implementation of non-traditional Conference on Neural Networks, pp. 1942-1948,
optimization techniques (PSO, CPSO, HDE) for the 1995.
optimal load flow solution. [15]. Akbal, "PSO and CSA to Estimate of Parameter in
[3]. Kothari, I.J. and Nagrath, D.P. (2007) Modern Power Power Line" IEEE 2013.
System Analysis. 3rd Edition, New York. [16]. R. C. Soldevilla and F. A. C. Huerta, "Estimation of
[4]. Thannimalai, Pavithren, et al. "Voltage Stability Transmission Lines Parameters Using Particle
Analysis and Stability Improvement of Swarm Optimization," 2018 IEEE PES Transmission
Power System." International Journal of Electrical & & Distribution Conference and Exhibition - Latin
Computer Engineering (2088-8708) 5.2 (2015). America (T&D-LA), Lima, Peru, 2018, pp. 1-5.
[5]. H. Lee and K. Y. Lee, "A study on voltage collapse [17]. S. A. Jumaat, I. Musirin, O. Muhammad Murtadha,
mechanism in electric power systems," in IEEE and H. Mokhlis, "PSO-based technique for loss
Transactions on Power Systems, vol. 6, no. 3, pp. minimization considering voltage profile and cost
966-974, Aug. 1991, doi: 10.1109/59.119236. function," 2011 5th International Power Engineering
and Optimization Conference, Shah Alam Malaysia,
2011, doi:10.1109/PEOCO.2011.5970456
IJISRT24FEB1283 www.ijisrt.com 1544
Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[18]. R. Mageshvaran, I. Jacob Raglend, V. Yuvaraj, P. G. [28]. Mojtaba Ahmadieh Khanesar, Mohammad
Rizwankhan, T. Vijayakumar, and Sudheera, Teshnehlab, and Mahdi Aliyari Shoorehdeli, "Novel
"Implementation of non-traditional optimization Binary Particle Swarm Optimization," 2007
techniques (PSO, CPSO, and HDE) for the optimal Mediterranean Conference on Control and
load flow solution," TENCON 2008 IEEE Region 10 Automation, Athens, Greece, 2007, pp. 1-6, doi:
Conference, Hyderabad, India, 2008, pp. 1-6, doi: 10.1109/MED.2007.4433821.
10.1109/TENCON.2008.4766839. [29]. Kanika Tyagi and Kirti Tyagi. Article: A
[19]. R. Bhowmik, A. K. Chakraborty, and P. Das, Comparative Analysis of Optimization
"Optimal location of UPFC based on the PSO Techniques. International Journal of Computer
algorithm considering active power loss Applications 131(10):6-12, December 2015.
minimization," 2012 IEEE Fifth Power India [30]. P. Dan Cristian, C. Barbulescu, A. Simo, S. Kilyeni,
Conference, Murthal, India, 2012, pp. 15, doi: and F. Solomonesc, "Load flow computation Particle
10.1109/PowerI.2012.6479489. Swarm Optimization algorithm," 2012 47th
[20]. S. A. Jumaat, I. Musirin, O. Muhammad Murtadha, International Universities Power Engineering
and H. Mokhlis, "PSO-based technique for loss Conference (UPEC), Uxbridge, UK, 2012, pp. 1-6,
minimization considering voltage profile and cost doi: 10.1109/UPEC.2012.6398606.
function," 2011 5th International Power Engineering [31]. E.C. Laskari, K. E. Parsopoulos, and M. N. Vrahatis,
and Optimization Conference, Shah Alam, Malaysia, "Particle swarm optimization for integer
2011, doi:10.1109/PEOCO.2011.5970456. programming," Proceedings of the 2002 Congress on
[21]. J. B. Abugri and M. Karam, "Particle Swarm EvolutionaryComputation,
Optimization for the Minimization of Power Losses doi:10.1109/CEC.2002.104478.
in Distribution Networks," 2015 12th International [32]. C. Laskari, K. E. Parsopoulos, and M. N. Vrahatis,
Conference on Information Technology New Particle swarm optimization for minimax problems,
Generations, Las Vegas, NV, USA, 2015, doi: Proceedings of the 2002 Congress on Evolutionary
10.1109/ITNG.2015.18. Computation. CEC'02 (Cat. No.02TH8600),
[22]. K. Kaur and Y. Kumar, "Swarm Intelligence and its Honolulu, HI, USA, 2002, vol.2, doi:
applications towards Various Computing: A 10.1109/CEC.2002.1004477.
Systematic Review," 2020 International Conference [33]. Altaf Q.H. Badar, B.S. Umre, and A.S. Junghare,
on Intelligent Engineering and Management Reactive power control using dynamic particle
(ICIEM), London, UK, 2020, swarm optimization for real power loss minimization
doi:10.1109/ICIEM48762.2020.9160177. International Journal of Electrical Power & Energy
[23]. S. Chalotra, S. K. Sehra, and S. S. Sehra, "A Systems, Volume 41, Issue 1, 2012, ISSN 0142-0615.
systematic review of applications of Bee Colony [34]. M.A. Abido, Multiobjective particle swarm
Optimization," 2016 International Conference on optimization for environmental economic dispatch
Innovation and Challenges in Cyber Security problems, Electric Power Systems Research, Volume
(ICICCS- INBUSH), Greater Noida, India, 79, Issue 7, 2009, ISSN 0378-7796.
2016, doi: 10.1109/ICICCS.2016.7542297. [35]. Modern Power System Analysis, 5th Edition, by D.
[24]. Y. del Valle, G. K. Venayagamoorthy, S. P. Kothari, I. J. Nagrath, and R. K. Saket, © 2022 |
Mohagheghi, J. C. Hernandez, and R. G. Harley, Published: June 6, 2022
"Particle Swarm Optimization: Basic Concepts, [36]. H.R.E.H. Bouchekara, M.A. Abido, and M.
Variants, and Applications in Power Systems," in Boucherma, Optimal power flow using the teaching-
IEEE Transactions on Evolutionary Computation, learning-based optimization technique, Electric
vol. 12, April 2008, doi: Power Systems Research, Volume 114, 2014, ISSN
10.1109/TEVC.2007.896686. 0378-7796.
[25]. Katoch, S., Chauhan, S.S., & Kumar, V., A review of [37]. Harish Garg, A hybrid PSO-GA algorithm for
genetic algorithms past, present, and future.Multimed constrained optimization problems, Applied
Tools Appl 80, 8091–8126 (2021). Mathematics and Computation, Volume 274, 2016,
https://doi.org/10.1007/s11042-020-10139-6 ISSN 0096-3003.
[26]. Optimal power flow using particle swarm [38]. Mahadevan, K., and Kannan, P.S., Comprehensive
optimization International Journal of Electrical Learning Particle Swarm Optimization for Reactive
Power & Energy Systems, doi:10.1016/S0142- Power Dispatch, INSS 1568-4946.
0615(01)00067-9. [39]. Sameer Singh, Power Loss Reduction in Power
[27]. M. Ahanch, M. S. Asasi, and M. S. Amiri, "A Systems Based on PSO: A Case Study, Research
Grasshopper Optimization Algorithm to Solve Scholar, Department of Electrical Engineering, Jaipur
optimal distribution system reconfiguration and National University, Jaipur, Rajasthan, India
distributed generation placement problems," 2017 DOI:10.5120/ijca2017913710.
IEEE 4th International Conference on Knowledge- [40]. Gad, A.G. Particle Swarm Optimization Algorithm
Based Engineering and Innovation (KBEI), Tehran, and Its Applications: A Systematic Review. Arch
Iran, 2017, doi:10.1109/KBEI.2017.8324880. Computat Methods Eng 29, 2531–2561 (2022).
IJISRT24FEB1283 www.ijisrt.com 1545
Volume 9, Issue 2, February 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
[41]. Liu, Hong Ying et al. “Parallel Ant Colony
Optimization Algorithm.” 2006 6th World Congress
on Intelligent Control and Automation 1 (2006):
3222-3226.
[42]. S. N. Deepa and J. Rizwana, "Power system stability
by reducing power losses using optimization
techniques," 2013 IEEE International Conference on
Computational Intelligence and Computing Research,
Enathi, India, 2013, pp. 1-4, doi:
10.1109/ICCIC.2013.6724152.
[43]. M. F. Mustafar, I. Musirin, M. R. Kalil, and M. K.
Idris, "Ant Colony Optimization (ACO) Based.
Technique for Voltage Control and Loss
Minimization Using Transformer Tap Setting," 2007
5th Student Conference on Research and
Development, Selangor, Malaysia, 2007, pp. 1-6,
doi:10.1109/SCORED.2007.4451444.
