Open AccessAnalysing the Migration Period Parameter in Parallel Multi-Swarm Particle Swarm Optimization
Author(s) -
Şaban Gülcü,
Halife Kodaz
Publication year - 2016
Publication title -
international journal of computer science and information technology/international journal of computer science and information technology (chennai. print)
Language(s) - English
Resource type - Journals
eISSN - 0975-4660
pISSN - 0975-3826
DOI - 10.5121/ijcsit.2016.8303
Subject(s) - particle swarm optimization , computer science , swarm behaviour , period (music) , multi swarm optimization , metaheuristic , mathematical optimization , algorithm , artificial intelligence , mathematics , physics , acoustics
In recent years, there has been an increasing interest in parallel computing. In parallel computing, multiple computing resources are used simultaneously in solving a problem. There are multiple processors that will work concurrently and the program is divided into different tasks to be simultaneously solved. Recently, a\udconsiderable literature has grown up around the theme of metaheuristic algorithms. Particle swarm optimization (PSO) algorithm is a popular metaheuristic algorithm. The parallel comprehensive learning particle swarm optimization (PCLPSO) algorithm based on PSO has multiple swarms based on the masterslave paradigm and works cooperatively and concurrently. The migration period is an important parameter in PCLPSO and affects the efficiency of the algorithm. We used the well-known benchmark functions in the experiments and analysed the performance of PCLPSO using different migration period