Title: Barebones particle swarm for multi-objective optimisation problems

Authors: Yong Zhang, Dun-wei Gong, Ya-nan Jiang

Addresses: School of Information and Electronic Engineering, China University of Mining and Technology, Xuzhou Jiangsu 221008, China. ' School of Information and Electronic Engineering, China University of Mining and Technology, Xuzhou Jiangsu 221008, China. ' Department of Mathematical Sciences, University of Nevada, Las Vegas, Las Vegas 89154-4020, USA

Abstract: Control parameters, inertia weight and acceleration coefficients influence strongly performance of multi-objective particle swarm optimisation (MOPSO) algorithms. To eliminate the need for tuning of these parameters for different optimisation problems, this paper presents an almost parameter-free MOPSO algorithm, in which the concept of barebones particle swarm is incorporated into MOPSO. A special mutation operator that enriches the exploratory capabilities of our algorithm is also introduced. The proposed algorithm is validated using several benchmark test problems and four standard metrics. Results indicate that the proposed algorithm is highly competitive, and that can be considered a viable alternative to solving multi-objective optimisation problems.

Keywords: multi-objective optimisation; particle swarm optimisation; barebones PSO; mutation operators; control parameters; inertia weight; acceleration coefficients.

DOI: 10.1504/IJICA.2009.031779

International Journal of Innovative Computing and Applications, 2009 Vol.2 No.2, pp.86 - 99

Published online: 24 Feb 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article