电子学报 ›› 2017, Vol. 45 ›› Issue (10): 2343-2347.DOI: 10.3969/j.issn.0372-2112.2017.10.005

• 学术论文 • 上一篇    下一篇

一种基于双空间密度的多目标进化算法

王鹏, 张长胜, 张斌, 吴嘉轩, 刘婷婷   

  1. 东北大学计算机科学与工程学院, 辽宁沈阳 110819
  • 收稿日期:2016-06-20 修回日期:2016-12-22 出版日期:2017-10-25
    • 作者简介:
    • 王鹏,男,1987年生于山东烟台.东北大学计算机应用技术专业博士研究生.研究方向为服务计算、人工智能算法.E-mail:815268711@qq.com;张长胜,男,1980年生于吉林长春.东北大学信息科学与工程学院副教授、硕士生导师.主要研究方向为智能信息处理.
    • 基金资助:
    • 国家自然科学基金 (No.61572116,No.61572117,No.61502089); 国家关键科技研发基金 (No.2015BAH09F02); 中央高校东北大学基本科研专项基金 (No.N150408001,No.N150404009)

A Two-Space-Density Based Multi-objective Evolutionary Algorithm for Multi-objective Optimization

WANG Peng, ZHANG Chang-sheng, ZHANG Bin, WU Jia-xuan, LIU Ting-ting   

  1. School of Computer Science and Engineering, Northeastern University, Shenyang, Liaoning 110819, China
  • Received:2016-06-20 Revised:2016-12-22 Online:2017-10-25 Published:2017-10-25
    • Supported by:
    • National Natural Science Foundation of China (No.61572116, No.61572117, No.61502089); Fund of National Key Science and Technology Research and Development Program of China (No.2015BAH09F02); Fundamental Research Funds for the Central Universities for Northeastern University (No.N150408001, No.N150404009)

摘要: 目前,大多数多目标进化算法的多样性保持机制都只强调目标空间的多样性而忽视决策空间的多样性.这种设置可能导致种群在目标空间拥有良好多样性的同时却在决策空间拥有较差的多样性.为了解决上述问题,本文提出了一种基于双空间密度的多目标进化算法.为了反映个体在决策空间和目标空间的多样性,本文定义了双空间密度指标.基于双空间密度的配对选择操作可以平衡算法的收敛性与多样性;基于双空间密度的个体选择操作可以同时使决策空间和目标空间得到充分的搜索.实验结果表明,本文算法的求解质量明显优于对比算法.

关键词: 人工智能, 进化算法, 空间密度, 决策空间, 目标空间

Abstract: Most of the evolutionary algorithm researches related to diversity maintenance scheme are dedicated to the diversity of objective space and ignore the diversity of decision space.This arrangement could lead to excessive diversity in the objective space but poor diversity in the decision space.To address this issue,this paper proposes a two-space-density based multi-objective evolutionary algorithm.Two-space-density is defined to reflect the diversity in both the objective space and the decision space.Based on two-space-density,TSD-mating selection is presented to balance the convergence and the diversity of population;TSD-selection is designed to fully explore the objective space and the decision space.The experimental results show that our algorithm performs competitively against the chosen state-of-the-art designs.

Key words: artificial intelligence, evolutionary algorithm, space density, decision space, objective space

中图分类号: