Particle Swarm Optimization for Multimodal Functions: A Clustering Approach
Alessandro Passaro, Antonina Starita · Journal of Artificial Evolution and Applications · 2008
The particle swarm optimization (PSO) algorithm is designed to find a single optimal solution and needs some modifications to be able to locate multiple optima on a multimodal function. In parallel with evolutionary computation algorithms, these modifications can be grouped in the framework of niching. In this work, we present a new approach to niching in PSO based on clustering particles to identify niches. The neighborhood structure, on which particles rely for communication, is exploited together with the niche information to locate multiple optima in parallel. Our approach was implemented in thek‐means‐based PSO (kPSO), which employs the standardk‐means clustering algorithm, improved with a mechanism to adaptively identify the number of clusters.kPSO proved to be a competitive solution when compared with other existing algorithms, since it showed better performance on a benchmark set of multimodal functions.