An improved PSO Algorithm with Random Perturbation around Convergence Center

摘要

A new modified strategy is proposed based on the analysis of convergence of particle swarm optimization (PSO). Each particle is manipulated by its convergence center, and then randomly perturbed around it. Some mutation operators are used to retain diversity of population and avoid being plunged to local optimum. Moreover, theoretical analysis has been made to prove that it can more easily converge to the global optimum. Experiment results show that it is superior to basic particle swarm optimization in quality and efficiency.