An Elitist Learning Particle Swarm Optimization With Scaling Mutation and Ring Topology

Abstract

Particle swarm optimization (PSO) is an evolutionary algorithm that is well known for its simplicity and effectiveness. It usually has strong global search capability but has the drawback of being easily trapped by local optima. A scaling mutation strategy and an elitist learning strategy are presented in this paper. Based on these strategies, an improved PSO variant (LSERPSO) is developed through a local search and ring topology strategy. The new scaling mutation strategy involved an exploration and exploitation balance focusing on mutation operation. A collection of elite individuals is maintained such that an array of current particles can learn from them. A ring topology-based neighborhood structure is adopted to maintain the population diversity and to reduce the possibility of particles being trapped in local optima. Finally, a quasi-Newton-based local search is incorporated to enhance the fine-grained capability. The effects of these proposed strategies and their cooperation are verified step by step. The performance of LSERPSO is comprehensively studied using IEEE CEC2015 benchmark functions.