J4 ›› 2010, Vol. 37 ›› Issue (4): 737-742.doi: 10.3969/j.issn.1001-2400.2010.04.028

• 研究论文 • 上一篇    下一篇

动态改变邻域空间和搜索步的自由搜索算法

李团结;曹玉岩;孙国鼎   

  1. (西安电子科技大学 机电工程学院,陕西 西安  710071)
  • 收稿日期:2009-05-17 出版日期:2010-08-20 发布日期:2010-10-11
  • 通讯作者: 李团结
  • 作者简介:李团结(1972-),男,教授,博士,E-mail: tjli@mail.xidian.edu.cn.
  • 基金资助:

    国家自然科学基金资助项目(50875193)

Free search algorithm with the variable neighbourhood and step

LI Tuan-jie;CAO Yu-yan;SUN Guo-ding   

  1. (School of Mechano-electronic Engineering, Xidian Univ., Xi'an  710071, China)
  • Received:2009-05-17 Online:2010-08-20 Published:2010-10-11
  • Contact: LI Tuan-jie

摘要:

针对群体智能优化方法——自由搜索算法后期寻优效率降低、特别是多维空间寻优效果不佳的问题,提出一种动态改变邻域空间和搜索步的自由搜索算法,该算法前期邻域空间和搜索步变化不大,进行全局搜索;后期邻域空间和搜索步变化较大,进行局部寻优.给出了动态调整邻域空间及搜索步的方法.通过对4个经典的函数进行测试实验,结果表明,该算法在平均最优值和成功率上都有所提高,而且收敛速度快、精度高,尤其对多维多峰函数效果更加明显.

关键词: 自由搜索, 全局优化, 邻域空间, 算法, 搜索步

Abstract:

To overcome the problem of population-based optimisation algorithms-free search (FS) to find the best solution with the low efficient later stage, especially multi-dimensional search space, FS with the variable neighbourhood and step method is proposed. The previous neighbourhood space and step do not change much in searching for the overall situation and change relatively greatly in the later stage in local exploration by the algorithm. The methods for the varying neighbourhood and step are given and simulation results with four traditional functions show that the algorithm has a better probability of finding the global optimum and mean best value, with quick convergence and high precision, especially for the multi-dimensional and multimodal function.

Key words: free search, global optimization, neighbourhood space, algorithms, search step