基于变异策略的粒子群算法-JournalofNortheasternUniversity.PDF

第31卷第11期 东 北 大 学 学 报 ( 自 然 科 学 版 ) Vol31, No. 11 2010 年 11 月 Journal of Northeastern U niversity( Natural cience) Nov . 2 0 1 0 高立群, 吴沛锋, 邹德旋 ( , 110004) : , , ( MP O) , , 4 1 1 , , M P O , : ; ; ; : TP 301. 6 : A : 1005-3026(2010) 11- 1530-04 Particle Swarm Optimization Based on Mutation Strategy GA O L i-qun , W U Pei-f eng, ZO U De-x uan ( chool of Information cience Engineering, Northeastern University, henyang 110004, China. Correspondent: WU Pe-i feng, E-mail: wuzheng820807 @ yahoo. cn) Abstract : After studying the features of P O ( particle sw arm optimization) , the mutation factor w as integrated into the algorithm to form its new version, i. e. , the mutational particle sw arm optimization ( MP O) in w hich a mutation strategy was incorporated. With the mutation factor the algorithm can improve the developability of solution space so as to decrease the possibility that the P O falls into local optima. The tested results of four w ell-know n unconstrained benchmark optimization problems, a constrained problem and an actual application problem revealed that MP O can successfully tackle both the high-dimensional unconstrained problems and the high- dimensional problems w ith linear constraints. Moreover, MP O as a promising optimization algorithm has strong convergence and high stability. Key words: P O; high-dimensional problem; mutation factor; premature [ 1] Kennedy 1995 , ,