Dynamic Multi-swarm Particle Swarm Optimization with Center Learning Strategy

In this paper, we propose a novel variant of particle swarm optimization, called dynamic multi-swarm particle swarm optimization with center learning strategy (DMPSOC). In DMPSOC, all particles are divided into several subswarms. Then, a center-learning strategy is designed, in which each particle within the sub-swarms will learn from the historical optimal position of a particle or the center position in a sub-swarm. Also, an alternative learning factor is given to determine the particle learning strategy, which can be classified as center-learning or optimum-learning. Four benchmark functions are used in order to compare the performance of DMPSOC algorithm with the standard particle swarm optimization (SPSO). Experiments conducted illustrate that the proposed algorithm outperform SPSO in terms of convergence rate and solution accuracy.