差分进化微粒群优化算法—DEPSO

A Differential Evolution Particle Swarm Optimizer—DEPSO

微粒群优化算法是一种新的进化计算技术,具有良好的优化性能,但是对于高维多模态函数,因进化后期微粒多样性的降低导致算法早熟收敛。文章提出的差分进化微粒群优化算法(DEPSO),拓宽了微粒信息传递的途径,增加了微粒的多样性,保证了算法的全局收敛。实验结果表明,DEPSO比PSO有更好的性能。

Particle swarm optimization algorithm is a new evolutionary computation technology and exhibits good performance on optimization.However the algorithm,to the highly dimentions and highly muti-modal function,will fall into premature covergence due to the decrease of poputation diversity in the evolution later.This paper intoduce a differential evolution particle swarm optimizer（DEPSO） which enlarge the tansfering approach of the particles information and increase the particles diversity and guarantee global covergence. Experiments on benchmark functions shows DEPSO outperform basic PSO.