Abstract: In accordance with that the standard Particle Swarm Optimization（PSO） has slower convergence during the early period and is prone to becoming trapped in local minima during the later period，this paper presents an effective co-evolutionary PSO based on information entropy and Simulated Annealing（SA）.This approach has been tested on four typical problems used in the literature.In all cases，the results show that the proposed approach is efficient and can reach a higher precision.

Key words: Particle Swarm Optimization（PSO）, information entropy, co-evolution, constrained optimization, Simulated Annealing（SA）

摘要： 针对基本粒子群算法具有搜索初期收敛速度慢，后期易陷入局部极值点的缺陷，引入信息熵衡量粒子群体的适应度值，结合模拟退火算法，提出一种基于信息熵混合协进化粒子群算法，增强了算法的自适应能力。通过4个标准函数对提出的算法进行了测试，仿真结果表明，算法是有效和可行的，且比基本粒子群算法的计算精度高。

关键词: 粒子群优化, 信息熵, 协进化, 约束优化, 模拟退火