Please use this identifier to cite or link to this item:
Title: Particle swarm optimization with re-initialization strategies for continuous global optimization
Authors: Kennedy, D.D.
Zhang, H.
Rangaiah, G.P. 
Bonilla-Petriciolet, A.
Keywords: Benchmark functions
Particle swarm optimization
Re-initialization strategy
Stochastic global optimization
Issue Date: 2013
Citation: Kennedy, D.D.,Zhang, H.,Rangaiah, G.P.,Bonilla-Petriciolet, A. (2013). Particle swarm optimization with re-initialization strategies for continuous global optimization. Global Optimization: Theory, Developments and Applications : 1-42. ScholarBank@NUS Repository.
Abstract: Particle Swarm Optimization (PSO) is a population-based algorithm inspired by social behavior of animals such as bird flocking and fish schooling. PSO algorithm can be easily implemented with few parameters involved, and several modifications have been introduced to the original PSO algorithm to discourage premature convergence in global optimization. In this study, we propose re-initialization strategies for improving the performance of PSO for continuous global optimization. The first re-initialization strategy focuses on individual particle level, where a particle whose pbest fails to improve after certain number of iterations will be reinitialized. The second re-initialization strategy focuses on the population level, where the entire population will be reinitialized. In this case, re-initialization is conducted when the number of pbests having objective function value similar to that of gbest exceeds a certain threshold number. The combination of these two re-initialization strategies has also been studied. The proposed PSO algorithms have been tested using 15 benchmark functions, each with both 10 and 30 variables. The results are compared with the efficient population utilization strategy PSO (EPUS-PSO), comprehensive learning PSO (CLPSO), cooperative PSO (CPSO), unified PSO (UPSO) and unified bare-bones PSO (UBBPSO). © 2013 by Nova Science Publishers, Inc. All rights reserved.
Source Title: Global Optimization: Theory, Developments and Applications
ISBN: 9781624177965
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

Page view(s)

checked on Dec 7, 2018

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.