›› 2012, Vol. 25 ›› Issue (5): 119-.

• 论文 • 上一篇    下一篇

基于双向搜索差分进化的多目标优化算法

宋通,庄毅,郭云   

  1. (南京航空航天大学 计算机科学与技术学院,江苏 南京 210016)
  • 出版日期:2012-05-15 发布日期:2012-05-24
  • 作者简介:宋通(1987—),男,硕士研究生。研究方向:多目标优化。
  • 基金资助:

    航空科学基金资助项目(2010ZC13012)

A Multi-objective Optimization Algorithm Based on Differential Evolution with the Bidirectional-search Mechanism

 SONG Tong, ZHUANG Yi, GUO Yun   

  1. (School of Computer Science and Technology,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)
  • Online:2012-05-15 Published:2012-05-24

摘要:

针对差分进化算法求解多目标优化问题时易陷入局部最优的问题,设计了双向搜索机制以增强DE(Differential Evolution,DE)算法的局部搜索能力。一方面降低了算法陷入局部最优的风险,另一方面可增强Pareto解集的多样性,使Pareto前沿面的解集分布更为均匀。实验结果表明,相比于NSGA-II等同类算法,提出的方法在搜索Pareto最优解时效率更高,并且Pareto最优解集的精度及分布程度比前者更好。

关键词: 差分进化, 多目标优化, 双向搜索

Abstract:

In order to avoid the situation of falling into local optimum when solving the Multi-objective Optimization Problem (MOP) with Differential Evolution Algorithm (DE),we design a bidirectional search mechanism which can improve the ability of local search of the DE and reduce the risk of local optimum,as well as make the Pareto fronts more evenly distributed.Experimental results show that the proposed method is more efficient than similar algorithms such as NSGA-II with better precision and distribution of Pareto optimal solution.

Key words: differential evolution;multi-objective optimization;bidirectional search

中图分类号: 

  • TP301.6