摘要： 粒子群优化算法(PSO)是模拟生物群体智能的优化算法,具有良好优化性能。但是由于信息的单一传递,群体的迅速收缩和群体多样性降低,导致算法早熟收敛。该文采用多样性控制与交叉操作,使粒子群在细化搜索与扩展新区之间进行协调,提出了协调粒子群优化算法HPSO。实验结果表明:HPSO比PSO有更好的性能。

关键词: 协调粒子群优化算法, 多样性, 交叉算子, 伪星型拓扑结构

Abstract: Particle swarm optimization(PSO) algorithm is a new population intelligence based algorithm and exhibits good performance on optimization. However the algorithm will fall into premature convergence due to the decrease of population diversity, caused by the simplex information spread and fast convergence of population. In this paper, crossover operator and diversity control factor are introduced to keep the harmony between attraction (searching in detail) and expand (explore a new area). Experiments on benchmark functions shows HPSO outperform standard PSO.

中图分类号: 

• TP18