计算机应用 ›› 2011, Vol. 31 ›› Issue (04): 1084-1089.DOI: 10.3724/SP.J.1087.2011.01084

• 人工智能 • 上一篇    下一篇

求解高维优化问题的扰动混沌蚁群优化算法

葛方振1,2,魏臻1,田一鸣1,陆阳1   

  1. 1. 合肥工业大学 计算机与信息学院,合肥 230009
    2. 淮北师范大学 计算机科学与技术学院,安徽 淮北 235000
  • 收稿日期:2010-10-08 修回日期:2010-11-10 发布日期:2011-04-08 出版日期:2011-04-01
  • 通讯作者: 葛方振
  • 作者简介:葛方振(1975-),男,安徽淮北人,讲师,博士研究生,主要研究方向:分布式计算、计算智能;
    魏臻(1965-),男,安徽合肥人,教授,博士生导师,博士,主要研究方向:分布式控制;
    田一鸣(1982-),男,安徽阜阳人,博士研究生,主要研究方向:无线Ad Hoc网络;
    陆阳(1967-),男,安徽合肥人,教授,博士生导师,博士,主要研究方向:无线Ad Hoc网络、分布式控制。
  • 基金资助:
    国家自然科学基金资助项目(60873195;61070220);高等学校博士学科点专项科研基金资助项目(20090111110002);安徽省教育厅省级自然科学基金资助项目(KJ2009B115)

High-dimensional optimization problems via disturbance chaotic ant swarm algorithm

Fang-zen GE1,2,Zhen WEI1,Yi-ming TIAN1,Lu-yang LU1   

  1. 1. School of Computer and Information, Hefei University of Technology, Hefei Anhui 230009, China
    2. School of Computer Science and Technology, Huaibei Normal University, Huaibei Anhui 235000, China
  • Received:2010-10-08 Revised:2010-11-10 Online:2011-04-08 Published:2011-04-01
  • Contact: Fang-zen GE

摘要: 针对新型混沌蚁群优化算法(CAS)求解高维优化问题时存在的计算复杂和搜索精度低问题,提出了扰动混沌蚂蚁群(DCAS)算法。通过建立蚂蚁最佳位置更新贪婪规则和随机邻居选择方法有效地降低了计算复杂度;另外引入自适应扰动策略改进CAS算法,使蚂蚁增强局部搜索能力,提高了原算法的搜索精度。通过一组高维测试函数对DCAS算法的性能进行了高达1000维的仿真实验。测试结果表明,新算法对复杂的高维优化问题可行有效。

关键词: 群智能, 混沌蚁群, 全局搜索, 高维优化问题, 函数优化

Abstract: To resolve the problems of computational complexity and search precision existing in Chaotic Ant Swarm (CAS), a Disturbance CAS (DCAS) algorithm was proposed to significantly improve the performance of the original algorithm. DCAS algorithm reduced computational complexity by a new greedy method of updating ant's best position and a random neighbor selection method. Furthermore, a self-adaptive disturbance strategy was introduced to improve the precision of DCAS by developing ant's local search. Extensive computational studies were also carried out to evaluate the performance of DCAS on a new suite of benchmark functions with up to 1000 dimensions. The results show clearly that the proposed algorithm is effective as well as efficient for the complex high-dimensional optimization problems.

Key words: swarm intelligence, Chaotic Ant Swarm (CAS), global search, high-dimensional optimization problem, function optimization

中图分类号: