Title: Performance-dependent attractive and repulsive particle swarm optimisation
Author: Zhihua Cui
Address: Complex System and Computational Intelligence Laboratory, Taiyuan University of Science and Technology, Taiyuan, Shanxi, 030024, P.R. China
Abstract: Attractive and repulsive particle swarm optimisation (ARPSO) is a novel variant of particle swarm optimisation algorithm aiming to enhance the exploration capability by dynamically changing the population diversity. However, its performance is not always well due to a simple uniform control manner. Therefore, in this paper, a performance-dependent attractive and repulsive particle swarm optimisation (PDARPSO) is designed to improve the performance of ARPSO by introducing the different trajectory for each particle according to its performance. Simulation results show the proposed algorithm can achieve better balance between exploration and exploitation capabilities.
Keywords: particle swarm optimisation; attractive PSO; repulsive PSO; performance-dependent; repulsion; attraction; population diversity.
Int. J. of Modelling, Identification and Control, 2009 Vol.8, No.4, pp.270 - 276
Available online: 09 Dec 2009