广东工业大学学报 ›› 2014, Vol. 31 ›› Issue (1): 46-50.doi: 10.3969/j.issn.1007-7162.2014.01.009

• 综合研究 • 上一篇    下一篇

基于最优变异的万有引力算法

徐怀祥, 刘伟   

  1. 广东工业大学 应用数学学院,广东 广州 510006
  • 收稿日期:2012-11-26 出版日期:2014-03-29 发布日期:2014-03-29
  • 作者简介:徐怀祥(1985-),男,硕士研究生,主要研究方向为智能计算及其应用.
  • 基金资助:

    国家自然科学基金资助项目(60974077)

The Gravitational Algorithm Based on the Optimal Mutation

Xu Huai-xiang, Liu Wei   

  1. School of Applied Mathematics,Guangdong University of Technology, Guangzhou 510006,China
  • Received:2012-11-26 Online:2014-03-29 Published:2014-03-29

摘要: 针对标准引力算法存在的收敛速度慢、容易在进化过程中陷入停滞的缺点,给出了一种基于最优变异的万有引力算法. 该算法引入了一种变异算子使得算法能够跳出局部最优解,同时还能提高算法的收敛速度;对变异后不可行物质给出了约束处理方法.数值实验的结果表明新算法不仅具有较好的寻优能力,而且在一定程度上提高了算法的稳定性,是一种有效的算法.

关键词: 万有引力算法;变异;搜索

Abstract: Considering the low convergence speed of the standard gravitational algorithm, and its being stalled easily in the evolutionary process, it proposes the gravitational algorithm, based on the optimal mutation. A mutation operator was introduced, which made it possible for the algorithm to escape from local optima, thus accelerating the convergence speed. A constraint handling mechanism was proposed for the infeasible mass after mutation. The simulation results show that the new algorithm not only has better ability of optimization, but also improves to some extent the stability. It is an effective algorithm.

Key words: gravitational algorithm;mutation;search

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!