Title: Particle swarm optimisation with adaptive selection of inertia weight strategy

Authors: Hindriyanto Dwi Purnomo; Hui-Ming Wee

Addresses: Department of Information Technology, Satya Wacana Christian University, Diponegoro St. 52-60, Salatiga, Central Java, 50711, Indonesia ' Department of Industrial and System Engineering, Chung Yuan Christian University, No. 200, Chung Pei Rd., Chungli, 32023, Taiwan

Abstract: Particle swarm optimisation (PSO) is a powerful metaheuristics method that is motivated by the collective behaviour of the intelligence swarms. The lack of velocity controls is a major drawback of the PSO. Inertia weight is a parameter that is commonly used to control the particle speed. In this paper, an adaptive selection of inertia weight strategy is proposed. A set of inertia weight strategy is placed in a candidate pool. Each strategy will be chosen by a probability that is based on its success rate. Empirical studies on the ten unconstrained continuous benchmark problems show that the proposed method can improve the ability to avoid local optima, however it does not increase its convergence speed.

Keywords: particle swarm optimisation; PSO; adaptive selection; inertia weight; candidate pool; success rate; velocity control; particle speed.

DOI: 10.1504/IJCSE.2016.077731

International Journal of Computational Science and Engineering, 2016 Vol.13 No.1, pp.38 - 47

Received: 06 Jul 2013
Accepted: 20 Dec 2013

Published online: 14 Jul 2016 *

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