电子学报

• 学术论文 • 上一篇    下一篇

基于集合的高维多目标优化问题的进化算法

巩敦卫, 季新芳, 孙晓燕   

  1. 中国矿业大学信息与电气工程学院, 江苏徐州 221008
  • 收稿日期:2012-11-26 修回日期:2013-03-16 出版日期:2014-01-25
    • 作者简介:
    • 巩敦卫 男,1970年3月出生,江苏铜山人.博士,教授,博士生导师,1992年、1995年和1999年分别在中国矿业大学、北京航空航天大学和中国矿业大学获理学学士、工学硕士和工学博士学位.主要从事基于搜索的软件工程、智能优化与控制等方面的研究. E-mail:dwgong@vip.163.com 季新芳 女,1987年8月出生,江苏泰州人.2010获中国矿业大学工学学士学位.现为中国矿业大学控制理论与控制工程硕士研究生,主要从事多目标优化方面的研究. E-mail:mimosa-615615@126.com
    • 基金资助:
    • 中央高校基本科研业务费专项资金资助 (No.2013XK09); 国家自然科学基金 (No.61105063)

Solving Many-Objective Optimization Problems Using Set-Based Evolutionary Algorithms

GONG Dun-wei, JI Xin-fang, SUN Xiao-yan   

  1. School of Information and Electrical Engineering, China University of Mining and Technology, Xuzhou, Jiangsu 221008 , China
  • Received:2012-11-26 Revised:2013-03-16 Online:2014-01-25 Published:2014-01-25
    • Supported by:
    • Fundamental Research Funds for the Central Universities (No.2013XK09); National Natural Science Foundation of China (No.61105063)

摘要: 由于高维多目标优化问题包含的目标很多,已有的方法往往难以解决该问题.本文提出一种有效解决该问题的基于集合的进化算法,该方法以超体积、分布度,以及延展度为新的目标,将原优化问题转化为3目标优化问题;定义基于集合的Pareto占优关系,设计体现用户偏好的适应度函数;此外,还提出集合进化策略.将所提方法应用于4个基准高维多目标优化问题,并与其他2种方法比较,实验结果表明了所提方法的优越性.

关键词: 集合进化, 用户偏好, 进化算法, 高维多目标优化, 集合进化, 用户偏好

Abstract: Previous methods are difficult to tackle a many-objective optimization problem since it contains many objectives.A set-based evolutionary algorithm was proposed to effectively solve the above problem in this study.In the proposed method,the original optimization problem was first transformed into a tri-objective one by taking such indicators as hyper-volume,distribution and spread as three new objectives;thereafter,a set-based Pareto dominance relation was defined,and a fitness function reflecting a user's preference was designed;additionally,set-based evolutionary strategies were suggested.The proposed method was applied to four benchmark many-objective optimization problems and compared with the other two methods.The experimental results show its advantages.

Key words: set-based evolution, user preference, evolutionary algorithm, many-objective optimization, set-based evolution, user preference

中图分类号: