计算机与现代化 ›› 2012, Vol. 1 ›› Issue (200): 152-03.doi: 10. 3969/j. issn. 1006-2475.2012.04.042

• 算法设计与分析 • 上一篇    下一篇

一种新的分阶段进化的粒子群优化算法

俞智慧1,余绪金2,宋 晔2,吴烈阳2   

  1. 1.上饶师范学院数学与计算机科学学院,江西 上饶 334001; 2.江西省高速公路联网管理中心,江西 南昌 330003
  • 收稿日期:2011-11-14 修回日期:1900-01-01 出版日期:2012-04-16 发布日期:2012-04-16

A New Evolution Algorithm by Stages for Particle Swarm Optimization

YU Zhi-hui1, YU Xu-jin2, SONG Ye2, WU Lie-yang2   

  1. 1. School of Mathematics and Computer Science, Shangrao Normal University, Shangrao 334001, China;2. Highway Network Management Center of Jiangxi, Nanchang 330003, China
  • Received:2011-11-14 Revised:1900-01-01 Online:2012-04-16 Published:2012-04-16

摘要: 针对粒子群优化算法在优化多极值点复杂问题时容易陷入局部极值的不足,提出一种新的分阶段进化的粒子群优化算法。该方法进化过程分为两个阶段,每个阶段对应一个不同的模型,通过结合这两种模型的各自优点有效地降低群体陷入局部最优。仿真实验结果表明,对于复杂多极值函数优化问题,本文算法比标准粒子群算法的寻优能力更强。

关键词: 粒子群优化算法, 局部极值, 进化模型

Abstract: Considering that the standard PSO easily falls into local optimization when it solves the multi-extremum problems, an improved PSO algorithm is proposed. In this algorithm, the evolution process is divided into two stages, while each stage uses a different model, so the possibility of getting local extreme value is reduced by making full use of the respective advantages of the two evolution model. The results of simulation show that the proposed algorithm has the better optimization performance than the standard PSO when solving the multi-extremum problems.

Key words: particle swarm optimization algorithm, local extremum, evolution model