Evaluation of an improved Particle Swarm Optimization algorithm on MATLAB

Author

Taj, T.A. ; Khan, T.A. ; Asif, Muhammad Kamran ; Ijaz, Imran

Author_Institution

Dept. of Electr. Eng., King Saud Univ., Riyadh, Saudi Arabia

fYear

2013

fDate

29-31 Aug. 2013

Firstpage

101

Lastpage

105

Abstract

An improved Particle Swarm Optimization(IPSO) algorithm is proposed in this paper. In the algorithm instead of using inertial weight and the particles velocity, two new factors are introduced, gamma which changes itself with respect to the number of iteration currently in progress. Gamma has a certain range which is a function of iteration; in our case it ranges from 0.94 to 0. A premature estimate mechanism is implemented so that after every iteration a new value of the gamma changes itself. Here the concept of the occurrence probability is also introduced and its values is set as 0.05. The benchmarks we used to check the algorithm are Sphere, Ackley, Rosenbrock, Schewfel´s 2.26 and Rastrigin.

Keywords

iterative methods; particle swarm optimisation; probability; Ackley benchmark; IPSO algorithm; MATLAB; Rastrigin benchmark; Rosenbrock benchmark; Schewfel´s 2.26 benchmark; Sphere benchmark; improved particle swarm optimization algorithm; inertial weight; iteration function; iteration number; occurrence probability; particle velocity; premature estimate mechanism; Algorithm design and analysis; Benchmark testing; Convergence; Mathematical model; Optimization; Particle swarm optimization; Standards; BenchMarks and probabilty; Improved PSO;

fLanguage

English

Publisher

ieee

Conference_Titel

Innovative Computing Technology (INTECH), 2013 Third International Conference on

Conference_Location

London

Print_ISBN

978-1-4799-0047-3

Type

conf

DOI

10.1109/INTECH.2013.6653679

Filename

6653679