An adaptive mutation operator for particle swarm optimization

Abstract

Particle swarm optimization (PSO) is an effcient tool for optimization and search problems. However, it is easy to betrapped into local optima due to its in-formation sharing mechanism. Many research works have shown that mutation operators can help PSO prevent prema- ture convergence. In this paper, several mutation operators that are based on the global best particle are investigated and compared for PSO. An adaptive mutation operator is designed. Experimental results show that these mutation operators can greatly enhance the performance of PSO. The adaptive mutation operator shows great advantages over non-adaptive mutation operators on a set of benchmark test problems.