ISSN 1000-1239 CN 11-1777/TP

计算机研究与发展 ›› 2015, Vol. 52 ›› Issue (9): 2123-2134.doi: 10.7544/issn1000-1239.2015.20140472

• 人工智能 • 上一篇    下一篇

差分选择策略在复杂多目标优化问题中的研究

郑金华1,3 , 刘磊1,2,3, 李密青1,3, 尹呈1,3, 王康1,3   

  1. 1(湘潭大学信息工程学院 湖南湘潭 411105); 2(邵阳市烟草公司 湖南邵阳 422900); 3(智能计算与信息处理教育部重点实验室(湘潭大学) 湖南湘潭 411105) (jhzheng@xtu.edu.cn)
  • 出版日期: 2015-09-01
  • 基金资助: 
    基金项目:国家自然科学基金项目(61379062,61403326);湖南省教育厅重点科研基金项目(12A135);湖南省自然科学基金项目(14JJ2072);湖南省科技支撑计划项目(2014GK3027);湖南省研究生创新基金项目(CX2013A011)

Difference Selection Strategy for Solving Complex Multi-Objective Problems

Zheng Jinhua1,3, Liu Lei1,2,3, Li Miqing1,3, Yin Cheng1,3, Wang Kang1,3   

  1. 1(Department of Information Engineering, Xiangtan University, Xiangtan, Hunan 411105); 2(China Tobacco of Shaoyang, Shaoyang, Hunan 422900); 3(Key Laboratory of Intelligent Computing & Information Processing(Xiangtan University), Ministry of Education, Xiangtan, Hunan 411105)
  • Online: 2015-09-01

摘要: 在多目标进化算法中,如何提高生成解的质量一直是研究的热点与难点.为解决以上问题,该算法从差分进化算法与计算资源分配策略2个方向进行了研究.根据多目标问题从决策空间到目标空间的映射关系以及差分进化算法基本原理,提出了一种基于双种群的多目标差分选择策略.它利用2个种群来区分个体间收敛性差别,在调整差分参数以适应多目标算法特性的基础上,以收敛性差别为依据选择参与差分运算的个体,从而提高差分算法性能,加快子代个体收敛.另外,根据子代个体收敛速率的不同,动态调整计算资源的分配,进一步提高算法收敛性.与ε-MOEA和MOEA/D-DRA在一系列复杂的多目标优化问题上进行了对比实验,结果表明了所提策略的有效性.

关键词: 差分进化, 选择策略, 动态分配计算资源, 多目标进化算法, 复杂Pareto解集

Abstract: Since the emergence of complex multi-objective problems in the finance and economics areas, dealing with multi-objective problems has gained increasing attention. How to improve the quality of generating solutions is the key in solving such problems. Although a number of MOEAs (multi-objective evolution algorithms) have been proposed over the last several years to solve the complex financial and economic multi-objective problems, not much effort has been made to deal with generating solutions in multi-objective optimization. Recently, we have suggested a MODEA_DACR (multi-objective difference evolution algorithm via dynamic allocation of computational resource) to improve the quality of generating solutions. The proposed algorithm uses two populations with different convergence rates to extract convergence information for the Pareto set, and then adjusts the parameter and difference evolution selection strategy according to the obtained convergence rate.In addition,based on the convergence rate of the population the proposed algorithm dynamically allocates the computational resources. The proposed algorithm is compared with two state-of-the-art algorithms, ε-MOEA and MOEA/D-DRA, on a suite of test problems with a complex Pareto set. Experimental results have shown the effectiveness of the proposed algorithm.

Key words: difference evolution, selection strategy, dynamic allocation of computational resource, multi-objective evolutionary algorithms, complex Pareto set

中图分类号: