Title: A novel particle swarm optimisation with hybrid strategies

Authors: Rongfang Chen; Jun Tang

Addresses: Department of Management Engineering, Hunan Urban Construction College, Hunan 411101, China ' Department of Construction Equipment Engineering, Hunan Urban Construction College, Hunan 411101, China

Abstract: Particle swarm optimisation (PSO) is an efficient optimisation technique, which has shown good search performance on many optimisation problems. However, the standard PSO easily falls into local minima because particles are attracted by their previous best particles and the global best particle. Though the attraction can accelerate the search process, it results in premature convergence. To tackle this issue, a novel PSO algorithm with hybrid strategies is proposed in this paper. The new approach called HPSO employs two strategies: a new velocity updating model and generalised opposition-based learning (GOBL). To test the performance of HPSO, 12 benchmark functions including multimodal and rotated problems are used in the experiments. Computational results show that our approach achieves promising performance.

Keywords: particle swarm optimisation; PSO; hybrid strategies; generalised opposition-based learning; GOBL; global optimisation; velocity updating models.

DOI: 10.1504/IJCSM.2015.069742

International Journal of Computing Science and Mathematics, 2015 Vol.6 No.3, pp.278 - 286

Received: 28 Jul 2014
Accepted: 05 Oct 2014

Published online: 08 Jun 2015 *

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