• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2011, Vol. 33 ›› Issue (10): 120-125.

• 论文 • 上一篇    下一篇

云搜索优化算法

曹〓炬,殷〓哲   

  1. (华中科技大学数学与统计学院,湖北 武汉 430074)
  • 收稿日期:2010-09-08 修回日期:2011-01-06 出版日期:2011-10-25 发布日期:2011-10-25

Clouds Search Optimization

CAO Ju,YIN Zhe   

  1. (School of Mathematics and Statistics,Huazhong University of Science and Technology,Wuhan 430074,China)
  • Received:2010-09-08 Revised:2011-01-06 Online:2011-10-25 Published:2011-10-25

摘要:

本文将云的生成、动态运动、降雨和再生成等自然现象与智能优化算法的思想融合,建立了一种新的智能优化算法云搜索优化算法(CSO)。生成与移动的云可以弥漫于整个搜索空间,这使得新算法具有较强的全局搜索能力;收缩与扩张的云团在形态上会有千奇百态的变化,这使得算法具有较强的局部搜索能力;降雨后产生新的云团可以保持云团的多样性,这也是使搜索避免陷入局优的有效手段。实验表明,基于这三点建立的新算法具有优异的性能,benchmark函数最优值的计算结果以及与已有智能优化算法的比较展现了新算法精确的、稳定的全局求解能力。

关键词: 云搜索优化算法, 智能优化, 函数优化, 全局优化

Abstract:

A new intelligent algorithm called the clouds search optimization(CSO) is proposed by blending the natural phenomena of clouds' generation, dynamic movement, rainfall and regeneration with the ideas of intelligent optimization algorithms. Clouds' generation and movement can permeate the entire search space, which makes this algorithm have strong global search ability. The strange changes in clouds' shapes which come into being by clouds' contraction and expansion bring the algorithm to strong local search ability. Meanwhile, rainfall and regeneration can maintain the clouds’ diversity and void trapping in the local optimum. The optimization experiments of the benchmark functions and the comparison between the new algorithm and other intelligent algorithms show the excellent performance of the new algorithm.

Key words: clouds search optimization algorithm;intelligent optimization;function optimization;global optimization