计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (32): 25-26.DOI: 10.3778/j.issn.1002-8331.2009.32.008

• 博士论坛 • 上一篇    下一篇

全变异粒子群优化算法

陈建超1,胡桂武1,2,3   

  1. 1.广东商学院 数学与计算科学学院,广州 510320
    2.中国人民大学 教育部数据工程与知识工程重点实验室,北京 100872
    3.中国人民大学 信息学院,北京 100872
  • 收稿日期:2009-08-12 修回日期:2009-09-14 出版日期:2009-11-11 发布日期:2009-11-11
  • 通讯作者: 陈建超

Whole Mutation Particle Swarm Optimization

CHEN Jian-chao1,HU Gui-wu1,2,3   

  1. 1.School of Mathematics & Computational Science,Guangdong University of Business Studies,Guangzhou 510320,China
    2.Key Lab of Data Engineering & Knowledge Engineering for the Ministry of Education,Renmin University of China,Beijing 100872,China
    3.School of Information,Renmin University of China,Beijing 100872,China
  • Received:2009-08-12 Revised:2009-09-14 Online:2009-11-11 Published:2009-11-11
  • Contact: CHEN Jian-chao

摘要: 针对粒子群优化算法容易早熟、收敛精度低等缺点,通过采用全变异策略、最大搜索速度自适应调整等策略得到了一种全变异粒子群优化算法,其中的全变异策略是在陷入早熟的条件下全体粒子参加变异,并且当把粒子看成染色体时,每一个基因等概率地参加变异,可以克服算法的早熟而继续优化,提高了算法的收敛精度。对Shubert函数进行实验的结果表明了算法的有效性。

关键词: 粒子群优化算法, 早熟, 变异, 基因

Abstract: To overcome the premature and low convergence precision of particle swarm optimization,the whole Mutation Particle Swarm Optimization(MPSO) is proposed with whole mutation and the maximum velocity self-adjustment strategy,whole mutation strategy is adopted when PSO encounters premature,the particle is considered as chromosome and every gene has the same probability to be mutated,the MPSO can overcome the local convergence of PSO and improves its convergence precision,the novel algorithm is used to solve the Shubert function optimization problem,the result shows that the algorithm is effective.

Key words: Particle Swarm Optimization(PSO), premature, mutation, gene

中图分类号: