Overview of particle swarm optimization

Abstract

The developments and applications related to particle swarm optimization (PSO) are discussed. Firstly, developments in the particle swarm optimization since 1995 are reviewed. Then parameter settings are analyzed systematically according to some existed testing results. Some improvement methods, such as cluster analysis, selection, neighborhood operator, no-hope/re-hope method, etc., are discussed. Some common test functions and the comparisons between PSO and others evolutionary algorithms are introduced. Finally, applications, both in the developed areas and the promising future application areas, are reviewed.