Multi–strategy based bare bones particle swarm for numerical optimisation. (2015)

• Record Type: Journal Article
• Authors: Liu, Jinwei
• Is Part Of: International journal of computing science and mathematics. Volume 6:Number 2(2015)
• Page Start: 178
• Page End: 187
• Publication Date: 2015
• Subjects: particle swarm optimisation -- bare bones PSO -- BPSO -- generalised opposition–based learning -- neighbourhood search -- simulation; Mathematics -- Periodicals; Computer science -- Periodicals; Mathematics -- Data processing -- Periodicals; 510.285
• Languages: English
• ISSNs: 1752-5055

Abstract

Particle swarm optimisation (PSO) is a population–based stochastic search algorithm, which simulates the social behaviour of bird flocking or fish schooling. Many previous studies have shown that PSO is an effective optimisation technique in evolutionary optimisation community. However, the standard PSO still suffers from premature convergence when solving complex multimodal problems. In this paper, we propose a new PSO algorithm called multi–strategy based bare bones PSO (MPSO). The MPSO introduces generalised opposition–based learning (GOBL) and two neighbourhood search strategies into the original bare bones PSO. Simulation study is conducted on 13 well–known benchmark functions. The results show that MPSO achieves better results than the standard PSO and two other PSO algorithms.