See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/329523088

Impact of Inertia Weight Strategies in Particle Swarm Optimization for

Solving Economic Dispatch problem

Article  in  Indonesian Journal of Electrical Engineering and Computer Science · January 2019

DOI: 10.11591/ijeecs.v13.i1.pp377-383

CITATIONS READS

6 431

4 authors, including:

Mohammed Amine Meziane Youssef Mouloudi

Université Tahri Mohammed Béchar 15 PUBLICATIONS   51 CITATIONS   
5 PUBLICATIONS   12 CITATIONS   
SEE PROFILE
SEE PROFILE

Bousmaha Bouchiba
Université Tahri Mohammed Béchar
39 PUBLICATIONS   65 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Behavior PEM fuel cell for 4WD electric vehicle under different scenario consideration View project

ARTIFICIAL NEURAL NETWORK SLIDING MODE CONTROL FOR MULTI-MACHINE WEB WINDING SYSTEM View project

All content following this page was uploaded by Mohammed Amine Meziane on 19 December 2018.

The user has requested enhancement of the downloaded file.

Indonesian Journal of Electrical Engineering and Computer Science
Vol. 13, No. 1, January 2019, pp. 377~383
ISSN: 2502-4752, DOI: 10.11591/ijeecs.v13.i1.pp377-383  377

Impact of inertia weight strategies in particle swarm

optimization for solving economic dispatch problem

Mohammed Amine Meziane, Youssef Mouloudi, Bousmaha Bouchiba, abdellah Laoufi

Tahri Mohammed University, BP 417, 08000 Bechar, Algeria

Article Info ABSTRACT

Article history: Particle Swarm Optimization (PSO) is a population based stochastic
optimization technique inspired by the social learning of birds or fish. Some
Received Aug 3, 2018 of the appealing facts of PSO are its convenience, simplicity and easiness of
Revised Sep 19, 2018 implementation requiring but few parameters adjustments. Inertia Weight (ω)
Accepted Oct 1, 2018 is one of the essential parameters in PSO, which often significantly the
affects convergence and the balance between the exploration and exploitation
characteristics of PSO. Since the adoption of this parameter, there have been
Keywords: large proposals for determining the value of Inertia Weight Strategy. In order
to show the efficiency of this parameter in the Economic Dispatch
Convergence problem(ED), this paper presents a comprehensive review of one or more
Data From SONELGAZ than one recent and popular inertia weight strategies reported in the related
Economic Dispatch literature. Among this five recent inertia weight four were randomly chosen
Inertia Weight for application and subject to empirical studies in this research, namely,
Particle Swarm Optimization Constant (ω), Random (ω), Global-Local Best (ω), Linearly Decreasing (ω),
which are then compared in term of performance within the confines of the
discussed optimization problem. Morever, the results are compared to those
reported in the recent literature and data from SONELGAZ. The study results
are quite encouraging showing the good applicability of PSO with adaptive
inertia weight for solving economic dispatch problem.
Copyright © 2019 Institute of Advanced Engineering and Science.
All rights reserved.

Corresponding Author:
Mohammed Amine Meziane
Tahri Mohammed University,
BP 417, 08000 Bechar, Algeria.
Email: [email protected]

1. INTRODUCTION
Amongst the different issues of power systems operation, the economic load dispatch (ELD)
problem is one of the key tools in operating and planning of modern electric utility grid. Essentially,
electrical grid systems are interconnected and consist of power generating, transmission and distribution
utilities in order to produce electrical power to consumers, at a low production cost, maximum reliability and
better operating conditions. The ELD is a static problem, it was first discussed by Carpentier in 1962 [1], the
main purpose of ELD is to find the optimal output power of generators to minimize the total generation cost
and satisfy the equality and inequality constraints. To solve this problem many efforts have been made over
the years, various mathematical programming and optimization techniques were used. A survey of literature
on the methods proposed to solve ELD, which can be divided into two categories, the classic (traditional)
methods and the smart (heuristic) methods. It is observed that the traditional methods and heuristic methods
have some limitations to solve ELD problems. The traditional methods suffer with large execution time and
would not be useful when the cost functions are nonlinear. So in some cases, it will be very difficult to
achieve optimal solutions. For this reason, recently, the heuristic methods have been used to overcome this
problem [2-3]. Therefore in recent years, different smart and innovative algorithms such as: Genetic
Algorithm (GA) [4], Particle Swarm Optimization (PSO) [5-6], Evolutionary Programming Algorithm
(EP) [7], Cuckoo Search (CS) [8] , … have been proposed to solve this problem.

Journal homepage: http://iaescore.com/journals/index.php/ijeecs

378  ISSN: 2502-4752

Recently, many researches have been directed towards the application of particle swarm
optimization technique to solve ELD problem [9]. The most important advantages of the PSO are that PSO is
easy to implement and there are few parameters to adjust. In this article, an attempt has been made to solve
economic load dispatch problem using particle swarm optimization by means of minimization of fuel costs
while satisfying physical and operational limitations. However, the prominent model to be discussed in this
paper, are Inertia Weight Strategies, and their effect in PSO for solving the ELD. In order to further illustrate
the effect of such mechanism in PSO for solving ELD, different inertia weight mechanism is reviewed and
experiments are carried out over single objective minimization case in the Real West Algeria 22-bus system
to compare different strategies of setting Inertia Weight. Moreover, the obtained optimal results also
compared with the some reported result found in literature and with Data from SONELGAZ. It found that the
PSO capable to obtain lowest cost as compared to others. Thus, it has great potential to be implemented in
different types of power system optimization problem.

2. PARTICLE SWARM OPTIMIZATION

PSO is a population-based optimization technique which was first introduced by Kennedy and
Eberhart in 1995 [10], inspired by social behavior of bird flocking or fish schooling in search of food. PSO
compared to other existing heuristic optimization strategies such as genetic algorithm, is easier to implement
involving only few parameters to adjust with accurate results in term of calculus. In a PSO system, particles
fly around in a multidimensional search space.
During flight, each particle adjusts its trajectory towards its own previous best position (This value
is called Pbest), and towards the best previous position attained by any member of its neighborhood or
globally, the whole swarm (This value is called Gbest), [11-17]. The two equations which are used in PSO
are velocity update equation (1) and position update equations (2). These are to be modified at each time step,
of PSO algorithm to converge the optimum solution.

V i t  1  V i t   c1 r1Pbest i t   X i t   c 2 r 2 [Gbest i t   X i t  1

X i t  1  X i t   V i t  1 2

Where, i : is the particle index;  : is the inertia coefficient; c1, c 2 are acceleration coefficients
0  c1, c2  2 ; r1, r 2 are random values, 0  r1, r 2  regenerated every velocity update; V i is the particle’s

velocity at time t ; X i is the particle’s position at time t; Pbest is the particle’s individual best solution as of
time t; Gbest is the swarm’s best solution as of time t.
Since 1995 many attempts have been made to improve the performance of the original PSO. For
instance, the maximum velocity V max was introduced to arbitrarily limit the velocities of the particles and
improve the result of the search. The inertia weight (ω) is one of PSO parameters originally proposed by Shi
and Eberhart [18] to bring about a balance between the exploration and exploitation characteristics of PSO.
Since the introduction of this parameter, there have been a number of proposals of different strategies for
determining the value of inertia weight during a course of run.

3. DIFFERENT INERTIA WEIGHT ADAPTATION MECHANISMS

The balance between global and local search throughout the course of a run is critical to the success
of an optimization algorithm [19]. Inertia Weight plays a key role in the process of balance between the
exploration and exploitation characteristics of PSO. In 1998 Shi and Eberhart [18] presented for the first time
the concept of inertia weight by introducing Constant Inertia Weight in which the velocity of each particle is
updated according to the equation (1). They claimed that a large inertia weight facilitates a global search
while a small Inertia Weight facilitates a local search. The following paragraphs represent a review of various
inertia weights in PSO chronologically.
Shi and Eberhart [20] proposed a Constant value of Inertia Weight and experimentally show that w
from [0.8, 1.2] PSO provide the global optimum in a reasonably of iteration. The Random Inertia Weight
strategy [21] is used in dynamic environment to enable PSO to track the optima and increases the
convergence of the algorithm in early iterations.

Indonesian J Elec Eng & Comp Sci, Vol. 13, No. 1, January 2019 : 377 – 383
Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  379

  0.5 
rand ()
3
2

Where rand () is a random number in [0, 1];  is then a uniform random variable in the range [0.5, 1].
In Time Varying Inertia Weight Strategies [22] the value of ω is determined based on iteration
number. These methods can be either a linear or non-linear and increasing or decreasing.
A linearly Decreasing Inertia Weight [23-25] was introduced to improve the performance of PSO.
They suggest that with a ω from the range 0.9 to 0.4 the PSO provides excellent results. In this method, the
value of inertia weight was decreased from max to (min ) according to the following equation:
( )

iter max  iter

iter   max  min   min 4
iter max

Where iter the current iteration of the algorithm and iter max is the maximum number of iterations.
In [26], Global-Local Best Inertia Weight is proposed by Arumugam and Rao. They use the ratio of
the local best and global best of the particles in each generation to determine the adaptive inertia weight in
each iteration.

  1.1 
gbest
pbesti average 5

Feng et al. [27-28] proposed Chaotic Inertia Weight using the merits of chaotic optimization. It
found that the CRIW enhances the performance of PSO in comparison with RIW. The proposed w is as
follows:

iter max  iter

iter   max  min   min  z 6 
iter max

The summary of various inertia weight strategies are displayed in Table 1.

Table 1. Different Inertia Weight Adaptation Strategies

No. IWS NAME OF INERTIA WEIGHT STRATEGIES Formula of inertia weight Reference
1 Constant inertia weight c [20]
rand()
2 Random inertia weight   0.5  [21]
2
iter max  iter
3 Linear Decreasing inertia weight iter   ( max  min)   min [23-25]
iter max
gbest
4 Global-Local Best inertia weight   1.1 
pbesti average [26]

iter max  iter

iter   max  min  min  z
5 Chaotic inertia weight iter max [27-28]
z  4  z  1  z 

4. OBJECTIVE
The inertia weight strategies have been suggested to improve both exploration and exploitation
ability or one of them in PSO. Exploitation means that all particles converge to the same peak of the
objective function and remains there. Furthermore, the exploration characteristic shows the capability of the
algorithm to leave the current peak and looking for better solutions.
Considering the above clarifications, the investigator aim at exploring the impact of inertia weight
on the exploration and exploitation capabilities in PSO and suggest a better strategy for users of this
algorithm within the area of the ELD problem. Experiments have been carried out on four Inertia Weight
Strategies: Constant (ω), Random (ω), Global-Local Best (ω), Linearly Decreasing (ω) in the confine of
economic dispatch optimization problem for 22 bus in power network real, West Algeria.
Impact of inertia weight strategies in particle swarm optimization for … (Mohammed Amine Meziane)
380  ISSN: 2502-4752

4.1. Economic Dispatch

The economic dispatch problem, which is used to minimize the cost of production of real power, can
generally be stated as follows:

 n 
Min  F i P i  7 
i  1 

n
 Pi  P D  P L 8
i 1

n  Pi  P n
P min max 9 
 
Where, generally, F i P i is a quadratic curve;

F i Pi   ai Pi2  bi Pi  ci 10 

Here:
a i , b i and c i are the known coefficient; n : number of generators; P i : real power generation; P D : real

power load; P L : real losses.

4.2. Experiment procedures

In order to test and compare some different inertia weight strategies in PSO reviewed in this
research, important optimization problem such as static economic dispatch for 22 bus in power network real,
West Algeria are used. Inertia weight mechanism’s influence on the ELD problem is tested in terms of
convergence speed and solution quality in the PSO algorithm.
The parameters settings of the experiment are as follows:
Population size (Swarm size) is 100 particles. The maximum iteration allowed number of function
evaluations is 200. The value of acceleration parameters c1 and c2 are taken equal to 2. The experiment
conducted in the ELD investigation was set in 22 bus system of power network real, West Algeria. This latter
consists of 7 thermal units, 15 load buses and 31 transmission lines, 03 compensator var static SVC [3*
(+40Mvar et )10Mvar)]. The total system demand is 856 MW. For implementing these different strategies in
PSO, the programming of the ELD problem using the PSO method has been developed and applied using
MATLAB software environment, tested on a CORE i5, personal computer with 2.20 GHz and 4 GO RAM.

5. SIMULATION RESULT AND DISCUSSION

The four strategies adopted for comparisons: Constant Inertia Weight, Random Inertia Weight,
Global-Local Best Inertia Weight and Linear Decreasing Inertia Weight are shown in Table 2 providing the
best solutions of the ELD problem.

Table 2. Optimization Results of Diffrent Inertia Weight Strategies in PSO for Economic Dispatch
Constant Random Global-Local Best Linear Decreasing
Criterion    
P1 [MW] 320 320 320 182.826
P 2 [MW] 140 140 140 192.257
P 3 [MW] 100 102.5166 100.6703 154.319
P 4 [MW] 104.6458 102.0303 103.9486 150
P 5 [MW] 110 110 110 63.7198
P 6 [MW] 50 50 50 50
P 7 [MW] 80 79.9998 80 79.9986
Transmission Loss 48.6458 48.5482 48.6189 17.12
Total output 904.6458 904.5482 904.6189 873.1204
Load demand 856 856 856 856
Total Cost [$/h] 9548.9 9549 9548.9 8999.34

Indonesian J Elec Eng & Comp Sci, Vol. 13, No. 1, January 2019 : 377 – 383
 Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  381

According to the above table, we notice that Constant and Global-Local Best Inertia Weight gives us
the same production cost and a slightly lower of 0.1 $/h in comparison with Random Inertia Weight,
transmission losses given by Random (ω) is lower than that given by Constant (ω) and Global-Local
Best (ω). In contrast, a Linearly Decreasing Inertia Weight gives a much better production cost of 549.66
[$/h] and minimum transmission loss of 31.5258 [MW], in comparison to other strategies. The difference in
generation cost between these mechanisms and in real power loss clearly shows the advantage of this
mechanism. Figure 1 illustrates convergence characteristics of PSO using the four Inertia Weight Strategies.

10200 10300

10200
10100

10100
10000

10000

Best Cost [$/h]

Best Cost [$/h]

9900
9900

9800
9800

9700
9700

9600 9600

9500 9500
0 20 40 60 80 100 120 140 160 180 200 0 20 40 60 80 100 120 140 160 180 200
Iteration Iteration

(a) (b)

10100

3.967
10
10000
3.965
10

9900
3.963
10
Best Cost [$/h]

Best Cost [$/h]

9800 3.961
10

3.959
9700 10

3.957
10
9600

3.955
10
9500 0 0.5 1 1.5 2 2.5 3
0 20 40 60 80 100 120 140 160 180 200
4
Iteration Iteration x 10
(c) (d)

Figure 1. Convergence characteristic of PSO using four inertia weight adjusting methods on (a) Constant (ω);
(b) Random (ω); (c) Global-Local Best (ω); (d) Linearly decreasing (ω)

These graphs clearly indicate that PSO converges rapidly to a high quality solution at the early
iterations. The minimize cost and power loss obtained by the proposed algorithm is less than value reported
in [29-31] using the evolutionary copulation techniques, genetic algorithm, Ant colony optimization for the
some test systems.
In order to demonstrate the efficiency and the robustness of the proposed PSO and the performance
of usage the inertia weight strategie in PSO for the solution of economic dispatch. The results obtained for
the power network real, West Algeria 220 kV of the 22-bus are compared to those obtained using Data from
SONELGAZ and present in Table 3.

Impact of inertia weight strategies in particle swarm optimization for … (Mohammed Amine Meziane)
382  ISSN: 2502-4752

Table 3. Comparison Results

Criterion Data From SONELGAZ Linear Decreasing 
P1 [MW] 200 182.826
P 2 [MW] 200 192.257
P 3 [MW] 300 154.319
P 4 [MW] 80 150
P 5 [MW] 100 63.7198
P 6 [MW] 100 50
P 7 [MW] 10 79.9986
Transmission Loss 21.40 17.12
Total output 890 873.1204
Load demand 856 856
Total Cost [$/h] 9104.42 8999.34

From the above table, it appears that PSO algorithm when using Linearly Decreasing Inertia Weight
gives much better results than the Data from Sonelgaz. The difference in generation cost and in Real power
loss clearly shows the advantage of this method.

6. CONCLUSION
In this paper, a comparative study on four suggested inertia weight strategies was conducted to
improve their impact on exploration and exploitation abilities in particle swarm optimization algorithm over
economic dispatch problem. These strategies are Constant Inertia Weight, Random Inertia Weight, Global-
Local Best Inertia Weight and Linear Decreasing Inertia Weight. The results verified and proved the main
objective of this study about the impact of inertia weight on the performance of PSO for optimal dispatch. As
an overall outcome of the experiments results carried out assignment, Linear Decreasing Inertia Weight is the
best strategy for a better production cost and a low transmission losses.

ACKNOWLEDGMENT
Authors would like to thank the heads of Laboratory of Analysis, Control and Optimization of
Electro-Energetic Systems (CAOSEE) and Smart Grids the renewable energies (ENERGARID) at the
university TAHRI Mohammed of Bechar (Algeria).

REFERENCES
[1] Hongye Wang, Carlos E. Murillo-Sanchez, Ray D. Zimmerman and Robert J. Thomas. On Computational Issues of
Market-Based Optimal Power Flow. IEEE Transactions on Power Systems, Aug 2007 ; Vol. 22(3) : 1185-1193.
[2] G.Sreenivasan, Dr. C.H.Saibabu, Dr.S.Sivanagaraju. Solution of Dynamic Economic Load Dispatch (DELD)
Problem with Valve Point Loading Effects and Ramp Rate Limits Using PSO. International Journal of Electrical
and Computer Engineering (IJECE). 2011; Vol.1 (1):59-70.
[3] Hossein Shahinzadeh, Sayed Mohsen Nasr-Azadani, Nazereh Jannesari, "Applications of Particle Swarm
Optimization Algorithm to Solving the Economic Load Dispatch of Units in Power Systems with Valve-Point
Effects", International Journal of Electrical and Computer Engineering (IJECE). 2014; Vol. 4, pp. 858~867.
[4] Walter DC, Sheble GB. Genetic algorithm solution of economic dispatch with valve point loading. IEEE
Transactions on Power Systems. 1993; 8(3): 1325-1332.
[5] Jong-Bae Park, Ki-Song Lee, Joong-Rin Shin, Lee KY. A particle swarm optimization for economic dispatch with
non-smooth cost functions. IEEE Transactions on Power Systems. 2005; 20(1): 34-42.
[6] Jiang S, Ji Z, Shen Y, “A novel hybrid particle swarm optimization and gravitational search algorithm for solving
economic emission load dispatch problems with various practical constraints”, Int J Electr Power Energy Syst,
2014; 55: 628–44.
[7] Venkatesh, P Gnanadass R, Padhy NP. Comparison and application of evolutionary programming techniques to
combined economic emission dispatch with line flow constraints. IEEE Transactions on Power Systems. 2003;
18(2): 688-697.
[8] N. Karthik, A.K. Parvathy RA, “Non-convex Economic Load Dispatch using Cuckoo Search Algorith”. Indonesian
Journal of Electrical Engineering and Computer Science (IJEECS), 2017; 5: 48–57.
[9] R. Chakrabarti, P K Chattopadhyay, M Basu, C K Panigrahi, “Particle Swarm Optimization Technique For Dynamic
Economic Dispatch”, July 26, 2005.
[10] J. Kennedy, R.C. Eberhart, Particle swarm optimization. IEEE International Conference on Neuran Networks.1995 ;
pp :1942–1948.

Indonesian J Elec Eng & Comp Sci, Vol. 13, No. 1, January 2019 : 377 – 383
 Indonesian J Elec Eng & Comp Sci ISSN: 2502-4752  383

[11] Djilani Ben Attous, Yacine Labbi. Particle Swarm Optimization based Optimal Power Flow for Units with Non-
Smooth Fuel Cost Functions. International Conference on Electrical and Electronics Engineering. Bursa, Turkey.
2009 :377-381.
[12] Eberhart, R.C., and Shi, Y. Comparing inertial weights and Constriction factor in particle Swarm optimization.
proceeding of the 2000 International Congress on Evaluating Computation, San Diego, California. IEEE Service
Center, Piscataway; 2000: 84-88.
[13] Sajjad Ahmadnia, Ehsan Tafehi. Using Particle Swarm Optimization, Genetic Algorithm, Honey Bee Mating
Optimization and Shuffle Frog Leaping Algorithm for Solving OPF Problem with their Comparison.
TELKOMNIKA Indonesian Journal of Electrical Engineering. 2015. Vol. 15(3):445-451.
[14] Kennedy. J. and Eberhart, R.C. Particle Swarm Optimzation. Proceeding of the 1997 International Conference on
Evaluationary Computation, IEEE service Center, Piscataway. 1997; pp: 303-308.
[15] Youssef MOULOUDI, Mohammed Amine MEZIANE, Abdellah LAOUFI,Bousmaha BOUCHIBA, Othmane
HARISI. A Swarm Algorithm Intelligent Optimization PSO in Power Network Real, West Algeria 220 kV.
ELECTROTEHNICĂ, ELECTRONICĂ, AUTOMATICĂ(EEA). 2016 ; Vol 64(1) : 55-60.
[16] J. C. Bansal, & all. Inertia Weight Strategies in Particle Swarm Optimization. IEEE Third World Congress on
Nature and Biologically Inspired Computing. Salamanca, Spain. 2011.
[17] Samar Bashath, Amelia Ritahani Ismail, Comparison of Swarm Intelligence Algorithms for High Dimensional
Optimization Problems. Indonesian Journal of Electrical Engineering and Computer Science. 2018; Vol. 11(1):
300-307.
[18] Y. Shi and R. Eberhart. A modified particle swarm optimizer. In Evolutionary Computation Proceedings, 1998.
IEEE World Congress on Computational Intelligence. 2002 ; pages 69–73.
[19] Y. Shi, R. Eberhart, Fuzzy adaptive particle swarm optimization. In Congress on Evolutionary Computation. Seoul,
Korea ; 2001.
[20] R.C. Eberhart and Y. Shi. Tracking and optimizing dynamic systems with particle swarms. Proceedings of the 2001
Congress on Evolutionary Computation. 2002. Vol 1 : 94–100.
[21] Y.H. Shi, R.C. Eberhart. A modified particle swarm optimizer. in: IEEE International Conference on Evolutionary
Computation, Anchorage Alaska, 1998, pp. 69–73.
[22] Ahmed Nickabadi et al. A novel particle swarm optimization algorithm with adaptive inertia weight. Applied Soft
Computing. 2011: 3658–3670.
[23] J. Xin, G. Chen, and Y. Hai. A Particle Swarm Optimizer with Multistage Linearly-Decreasing Inertia Weight.
International Joint Conference on Computational Sciences and Optimization. CSO 2009. Vol 1: 505–508.
[24] R.C. Eberhart, Y.H. Shi. Comparing inertia weights and constriction factors in particle swarm optimization. In:
IEEE Congress on Evolutionary Computation. 2000: pp. 84–88.
[25] Y.H. Shi, R.C. Eberhart. Experimental study of particle swarm optimization. In: Conference, Orlando, 2000.
[26] M.S. Arumugam and MVC Rao. On the performance of the particle swarm optimization algorithm with various
Inertia Weight variants for computing optimal control of a class of hybrid systems. Discrete Dynamics in Nature and
Society, 2006.
[27] Y. Feng, G. Teng, A. Wang, Y.M. Yao. Chaotic inertia weight in particle swarm optimization. In: Second
International Conference on Innovative Computing, Information and Control (ICICIC 07), 2007, pp: 475–1475.
[28] Y. Feng, Y.M. Yao, A. Wang. Comparing with chaotic inertia weights in particle swarm optimization. In:
International Conference on Machine Learning and Cybernetics, August 2007. pp: 329–333.
[29] W. Ongsakul, T. Tantimaporn. Optimal powers flow by improved evolutionary programming. Elect. Power Comp.
and Syst. 2006; Vol. 34: pp. 79-95.
[30] J. Yuryevich, K. P. Wong, Evolutionary Programming Based Optimal Power Flow Algorithm. IEEE Transaction on
power Systems. 1999; Vol. 14, No. 4.
[31] C. Thitithamrongchai, B. Eua-arporn. Selfadaptive Differential Evolution Based Optimal Power Flow for Units with
Non-smooth Fuel Cost Functions. J. Electrical Systems. 2007: 88-99.

Impact of inertia weight strategies in particle swarm optimization for … (Mohammed Amine Meziane)

View publication stats