计算机科学 ›› 2012, Vol. 39 ›› Issue (Z6): 489-492.

• • 上一篇    下一篇

基于Pareto的多目标克隆进化算法

贺 群,程 格,安军辉,戴光明,彭 雷   

  1. (中国地质大学计算机学院 武汉430074)
  • 出版日期:2018-11-16 发布日期:2018-11-16

Pareto-based Multi-object Clonal Evolutionary Algorithm

  • Online:2018-11-16 Published:2018-11-16

摘要: 为了克服部分多目标进化算法中容易出现退化与早熟,造成收敛速度过慢的不足,结合精英保留策略、基于近部规则的环境选择以及免疫克隆算法中的比例克隆等思想,提出一种基于Pareto的多目标克隆进化算法NPCA(Non-dominated Pareto Clonal Algorithm)。通过部分多目标优化测试函数ZDT和DTLZ对算法进行了性能测试,验证了该算法能获得分布更加均匀的Parcto前沿,解的收敛性明显优于典型的多目标进化算法。

关键词: 多目标优化问题,多目标进化算法,多目标优化免疫算法,NPCA算法

Abstract: To overcome the shortcomings of partial multiobjective evolutionary algorithms,we combine some outstanding thoughts in SPEA2 and immune multi-objective optimization algorithm then innovate out a Pareto-based multi-object clonal evolutionary algorithm NPCA(non-dominated Pareto clonal algorithm). And testing the algorithm with the famous multi-objective optimization problems ZDTand DTLZ, the results show that the new algorithm NPCA obviously takes advantages over the typical multi-objective evolutionary algorithms.

Key words: MOP(Multi objective optimization problems),MOEA(Multi-objective evolutionary algorithms),Multi-objective optimization immune algorithms,NPCA(Non-dominated Pareto clonal algorithm)

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!