东北大学学报:自然科学版 ›› 2014, Vol. 35 ›› Issue (4): 479-483.DOI: 10.12068/j.issn.1005-3026.2014.04.006

• 信息与控制 • 上一篇    下一篇

基于Species机制的多目标遗传算法

王洪峰1,张迁1,李小将2,3   

  1. (1 东北大学 信息科学与工程学院, 辽宁 沈阳110819; 2 中国民航局第二研究所, 四川 成都610041; 3 民航机场(成都)电子工程设计所, 四川 成都610041)
  • 收稿日期:2013-07-11 修回日期:2013-07-11 出版日期:2014-04-15 发布日期:2013-11-22
  • 通讯作者: 王洪峰
  • 作者简介:王洪峰(1979-),男,辽宁辽阳人,东北大学副教授,博士.
  • 基金资助:
    国家自然科学基金资助项目(71001018);中央高校基本科研业务费专项资金资助项目(N110204005,N110404019);中国博士后基金资助项目(2012T50266).

Speciesbased Genetic Algorithm for Multiobjective Optimization Problems

WANG Hongfeng1, ZHANG Qian1, LI Xiaojiang2,3   

  1. 1School of Information Science & Engineering, Northeastern University, Shenyang 110819, China; 2The Second Research Institute of CAAC, Chengdu 610041,China; 3Civil Aviation Electronic Engineering Design Insititute, Chengdu 610041, China.
  • Received:2013-07-11 Revised:2013-07-11 Online:2014-04-15 Published:2013-11-22
  • Contact: WANG Hongfeng
  • About author:-
  • Supported by:
    -

摘要: 多目标优化算法设计正在成为当前进化算法领域的一个研究热点.考虑将最初用于多峰优化的Species机制引入到多目标遗传算法中,通过借鉴现有多目标算法的相关思想,设计并提出了一种新的Species方法,包括基于Pareto最优性和拥挤度思想的Species种子确定策略及适应性的Species构造策略.一组标准多目标测试函数的仿真实验结果表明,提出的基于Species机制的多目标遗传算法表现出比经典的非支配排序遗传算法Ⅱ(NSGAⅡ)更好的性能.

关键词: 多目标优化问题, 进化多目标优化, 遗传算法, Species机制, 多峰优化

Abstract: Considering that evolutionary multiobjective optimization has been becoming one of research topics in evolutionary algorithm community recently, a speciesbased mechanism was introduced into multiobjective GA, which is initially applied into GA for multimodal optimization problem. And then a new speciesbased method was designed and proposed for multiobjective optimization problem, which comprises a species seed indentifying strategy based on Pareto optimality and crowd degree and an adaptive species constructing scheme. Experimental results showed that the proposed speciesbased multiobjective GA outperforms nondominated sorting GA Ⅱ(NSGAⅡ)on a set of benchmark test problems.

Key words: multiobjective optimization problem, evolutionary multiobjective optimization, genetic algorithm, speciesbased mechanism, multimodal optimization

中图分类号: