期刊文献+

多目标决策下Pareto非支配解的快速构造方法 被引量:11

Fast construction method of Pareto non-dominated solution for multi-objective decision-making problem
原文传递
导出
摘要 有效构造非支配解集可加快Pareto前沿的求解速度,提升多目标决策的质量和效率.在非支配解定义和性质分析基础上,推导出支配关系传递性引理,非支配解集构造定理及引理,并据此提出一种基于性质定理的非支配解集构造方法.基于所提方法,分析其循环次数和比较次数,推导出在最坏情况下能算出确定值的复杂度计算公式.最后证明该方法的正确性与完备性,分析最坏情形下其构造集的结构特征,并通过ZDT1~ZDT3测试函数进行检验.结果表明:所提方法比排除法和选举法的计算复杂度更低,构造速度更快. Formulating non-dominated solution set effectively can speed up the solving process of the Pareto front,and can improve the quality and efficiency of multi-objective decision-making.Therefore,based on the definition and feature of the non-dominated solutions,the lemma of dominations relation transitivity,the theorems and lemma of non-dominated solution set construction are deduced.Depending upon the proposed theorems and lemmas,a novel non-dominated solution set construction method is first proposed.Then the frequency of the comparison and the number of iteration in the new method are counted,and a novel formula is put forward to calculate the deterministic value of computational complexity in the worst case.Finally,the correctness and completeness of the new method are proved in theory,and the structural features of the construction set at the maximum degree of complexity are given,what’s more,the performance test of the new method is carried out through the ZDT1 ~ ZDT3 test functions.The results show that the new method is lower in computational complexity and faster in construction speed than the exclusions method and electoral law method.
出处 《系统工程理论与实践》 EI CSSCI CSCD 北大核心 2016年第2期454-463,共10页 Systems Engineering-Theory & Practice
基金 湖北省教育厅青年项目(Q20151104) 国家自然科学基金面上项目(51275366) 国家自然科学基金青年项目(51305311) 国家重点基础研究发展计划项目(973计划)(2014CB046705)~~
关键词 多目标决策 Pareto非支配解 构造方法 复杂度 multi-objective decision-making Pareto non-dominated solution construction method computational complexity
  • 相关文献

参考文献8

二级参考文献66

  • 1姚咏,范文慧,熊光楞.复杂产品开发过程仿真及其优化方法研究[J].系统仿真学报,2006,18(3):726-730. 被引量:20
  • 2郑金华,蒋浩,邝达,史忠植.用擂台赛法则构造多目标Pareto最优解集的方法[J].软件学报,2007,18(6):1287-1297. 被引量:54
  • 3Kuhn H W, Tucker A W. Nonlinear programming[ C]. In j Neyman editor Proceedings of the Second Berkeley Symposium on Mathematical Statistics and Probability, Berkeley, Califomia: University of Califomia Press, 1951:481 - 192. 被引量:1
  • 4Schaffer J D. Some experiments in machine learning using vector evaluated genetic algorithms[ C]. Unpublished doctoral dissertation, Vanderbilt University, 1984. 被引量:1
  • 5Zilzler E,Thiele L. Multi - objective optimization using evolutionary algorithms - A comparative case study[C]. Parallel Problem Solving from Nature, Springer, Berlin: 1984:292 - 301. 被引量:1
  • 6Deb K, Agrawal S, Pratap A,et al. A Fast Elitist Non - Dominated Sorting Genetic Algorithm for Multi - Objective Optimization: NSGA - Ⅱ .Technical Report No 2000001. Kanpur: Indian Institute of Technology Kanpur, 2000. 被引量:1
  • 7Kuhn H W,Tucker A W. Nonlinear programming[C]. In J Neyman editor Proceedings of the Second Berkeley Symposium on Mathematical Statistics and Probability, Berkeley, California:University of California Press,1951:481-192. 被引量:1
  • 8Schaffer J D.Some experiments in machine learning using vector evaluated genetic algorithms[C].Unpublished doctoral dissertation, Vanderbilt University,1984. 被引量:1
  • 9Zilzler E,Thiele L.Multi-objective optimization using evolutionary algorithms-A comparative case study[C].Parallel Problem Solving from Nature,Springer,Berlin:1984:292-301. 被引量:1
  • 10Deb K, Agrawal S, Pratap A,et al.A Fast Elitist Non-Dominated Sorting Genetic Algorithm for Multi-Objective Optimization: NSGA-Ⅱ.Technical Report No.2000001.Kanpur:Indian Institute of Technology Kanpur,2000. 被引量:1

共引文献155

同被引文献91

引证文献11

二级引证文献102

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部