• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2013, Vol. 35 ›› Issue (7): 95-101.

• 论文 • 上一篇    下一篇

一种求解约束优化问题的进化算法及其工程应用

朱高峰1,伍铁斌2,3,张艳蕾1,成运2,刘云连2   

  1. (1.湖南人文科技学院物理与信息工程系,湖南 娄底 417000;
    2.湖南人文科技学院通信与控制工程系,湖南 娄底 417000;
    3.中南大学信息科学与工程学院,湖南 长沙 410083)
  • 收稿日期:2012-11-29 修回日期:2013-02-25 出版日期:2013-07-25 发布日期:2013-07-25
  • 基金资助:

    国家自然科学基金资助项目(61273185);湖南省自然科学基金资助项目(12JJ2040);湖南省重点建设学科资助项目;湖南省教育厅重点项目资助(09A046);湖南人文科技学院青年基金资助项目(2010QN16,2012QN07)

Evolutionary algorithm for constrained optimization
problem and its engineering applications          

ZHU Gaofeng1,WU Tiebin2,3,ZHANG Yanlei1,CHENG Yun2,LIU Yunlian2   

  1. (1.Department of Physics and Information Engineering,
    Hunan University of Humanities Science and Technology,Loudi 417000;
    2.Department of Communications and Control Engineering,
    Human University of Humanities Science and Technology,Loudi 417000;
    3.School of Information Science and Engineering,Central South University,Changsha 410083,China)
  • Received:2012-11-29 Revised:2013-02-25 Online:2013-07-25 Published:2013-07-25

摘要:

提出一种改进的用于求解约束优化问题的进化算法。该算法利用混沌方法初始化个体以保证其均匀分布在搜索空间中。在进化过程中,将种群分为可行子种群和不可行子种群,分别采用不同的交叉和变异操作,以平衡算法的全局和局部搜索能力。标准测试问题的实验结果表明了改进算法的有效性。最后将改进算法应用到两个工程优化设计问题中,得到了满意的结果。

关键词: 约束优化问题, 进化算法, 交叉, 变异, 工程应用

Abstract:

A modified evolutionary algorithm (MEA) is proposed to solve constrained optimization problems. Chaotic sequence method is introduced to construct the initialization population that is scattered uniformly over the entirely search space in order to maintain the diversity. In the evolution process, our algorithm is based on individual feasibility; the population is divided into feasible subpopulation and infeasible subpopulation, which evolve with different crossover operator and different mutation operator, respectively. Numerical simulation results on four benchmark problems demonstrate the effectiveness and robustness of the proposed algorithm. Several engineering optimization problems are designed to test the MEA, and the results show that the MEA can solve different constrained optimization problems.

Key words: constrained optimization problem;evolutionary algorithm;crossover;mutation;engineering application