A hybrid niching PSO enhanced with recombination-replacement crowding strategy for multimodal function optimization

Abstract

This paper presents a hybrid niching algorithm based on the PSO to deal with multimodal function optimization problems. First, we propose to evolve directly both the particle population and memory population (archive population), called the P&A pattern, to enhance the efficiency of the PSO for solving multimodal optimization functions, and investigate illustratively the niching capability of the PSO and the PSO P&A. It is found that the global version PSO is disable, but the local version PSO P&A is able, to niche multiple species for locating multiple optima. Second, the recombination-replacement crowding strategy that works on the archive population is introduced to improve the exploration capability, and the hybrid niching PSO P&A (HN-PSO P&A) is developed. Finally, experiments are carried out on multimodal functions for testing the niching efficiency and scalability of the proposed method, and it is verified that the proposed method has a sub-quadratic scalability with dimension in terms of fitness function evaluations on specific MMFO problems.