摘要
针对标准粒子群优化算法易出现早熟收敛及寻优精度低等缺陷,提出一种基于双质心和自适应指数惯性权重的改进粒子群算法(DCAEPSO)。算法使用粒子搜到的最优解和当前解构造加权的种群质心和最优个体质心,结合使用自适应指数惯性权重调整了速度更新公式。通过几个典型测试函数仿真及Friedman和Holm检验,实验结果显示DCAEPSO比其他粒子群算法寻优能力强。
This paper proposes a new Particle Swarm Optimization（PSO）algorithm based on two aspects of improvement in standard PSO to avoid the problems about premature convergence and low precision. It adjusts velocity updating formula by embedding self-adaptive exponential inertia weight function and two weighted centroids, which are called the population centroid and the best individual centroid. Through the simulation of several typical benchmark functions, Friedman＇s tests and Holm＇s tests, the experimental results indicate that the proposed algorithm not only has advantages of convergence property over standard PSO and some other modified PSO algorithms, but also outperforms other algorithms proposed in this paper for searching global optimal solution.
出处
《计算机工程与应用》
CSCD
北大核心
2015年第5期58-64,250,共8页
Computer Engineering and Applications
基金
安徽省高校优秀青年人才基金项目(No.2012SQRL154)
滁州学院科研启动基金资助项目(No.2014qd007)
关键词
粒子群算法
质心
自适应指数惯性权重
particle swarm optimization algorithm
centroid
self-adaptive exponential inertia weight
