计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (33): 54-57.

• 学术探讨 • 上一篇    下一篇

基于随机算子的快速多目标遗传算法

彭 琰,郑金华   

  1. 湘潭大学 信息工程学院,湖南 湘潭 411105
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-11-21 发布日期:2007-11-21
  • 通讯作者: 彭 琰

Fast genetic multi-objective genetic algorithm based on random operator

PENG Yan,ZHENG Jin-hua   

  1. College of Information Engineering,Xiangtan University,Xiangtan,Hunan 411105,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-21 Published:2007-11-21
  • Contact: PENG Yan

摘要: 在NSGA-Ⅱ算法的基础上,对NSGA-Ⅱ构造非支配集的方法进行了改进,用擂台赛法则构造非支配集,当非支配集小于种群大小时,采用随机算子在可行域内随机产生新的解个体填充到下一代父种群中,形成了一种新的多目标遗传算法。在实验部分将改进后的算法和NSGA-II进行了性能比较,实验结果表明改进后的算法具有良好的分布性,算法运行效率也较高。

关键词: 多目标遗传算法, 非支配集, 擂台赛法则, 随机算子

Abstract: This paper improves the approach of constructing non-dominated set in NSGA-Ⅱ.We use AP(Arena’s Principle) to construct non-dominated set.When non-dominated set is smaller than the size of the population,a random operator is used to generate new solutions in feasible area to fill with the next population.In the experiments,the improved algorithm is compared with NSGA-Ⅱ.Finally the experimental results demonstrate that the improved algorithm has a good diversity and is more efficient than NSGA-Ⅱ.

Key words: multi-objective genetic algorithm, non-dominated set, AP, random operator