摘要： 为提高PSO算法的搜索能力，提出一种协同粒子群算法CPSO-ADS。引入种群分布熵及群落差异度评价，用以有效初始化群落。给出趋向向量修正粒子的位置向量，提高算法收敛速度。运用占优子空间概念，通过评价子空间搜索价值确定种群的迁移方向。实验结果表明，该算法搜索性能稳定，能以大概率收敛到全局最优。

关键词: 种群分布熵, 趋向向量, 占优子空间, 协同进化, 粒子群优化算法

Abstract: This paper proposes a novel cooperative Particle Swarm Optimization(PSO) algorithm(CPSO-ADS) to improve the search ability of PSO algorithm. To initialize the cluster effectively, population scatter entropy strategy and cluster differential degree strategy are introduced. To improve the convergence rate, it amends the position vector of a particle by producing an appulsive vector. And to ascertain the migration direction of a population, it proposes the concept of dominant subspace to evaluate the value of the special subspace. Experimental result shows that algorithm has stable search ability and can converge to the global optimum with large probability.

Key words: population scatter entropy, appulsive vector, dominant subspace, co-evolution, Particle Swarm Optimization(PSO) algorithm

中图分类号: 

• TP18