An Improved Particle Swarm Optimization with Re-initialization Mechanism

An improved Particle Swarm Optimization with re-initialization mechanism, which is based on the estimation of the varieties and activities of the particles, is proposed to balance the global search ability of the Standard Swarm Optimization (SPSO). Firstly the motion behavior of single particle is discussed, including the motion mode, convergence and the relationship between motion characteristic and the performance of SPSO. Then, a new variable named ¿steplength¿ is employed to represent the variety and activity of the particle population. The group of particles which satisfied the re-initialization conditions will be reinitialized in probability so that the variety and activity of the particle population can be hold in a reasonable level. Experiment results indicate that the improved Particle Swarm Optimization proposed in this paper has better performance compared with the other three PSO algorithms.