作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2009, Vol. 35 ›› Issue (23): 187-189,. doi: 10.3969/j.issn.1000-3428.2009.23.065

• 人工智能及识别技术 • 上一篇    下一篇

基于指标的多目标进化算法研究

张景成,戴光明   

  1. (中国地质大学计算机学院,武汉 430074)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2009-12-05 发布日期:2009-12-05

Research on Indicator-based Multi-objective Evolutionary Algorithm

ZHANG Jing-cheng, DAI Guang-ming   

  1. (School of Computer, China University of Geosciences, Wuhan 430074)
  • Received:1900-01-01 Revised:1900-01-01 Online:2009-12-05 Published:2009-12-05

摘要: 基于指标的进化算法(IBEA)是一个出色的多目标优化算法。IBEA具有良好的收敛性,但在保持解的多样性方面对于某些问题却表现较差。对IBEA进行研究,分析其适应度分配原理,针对其缺点进行改进,并将IBEA与其他2个算法进行了测试比较。测试结果表明改进后的IBEA在保持了原算法优点的情况下使其在解的多样性方面有了较大改观。

关键词: 多目标优化, 进化算法, 性能指标

Abstract: Indicator Based Evolutionary Algorithm(IBEA) is an excellent multi-objective optimism algorithm. IBEA has outstanding performance for convergence, but it has poor performance for maintaining diversity on some test problems. This paper studies the fitness assignment strategy of IBEA and improves it for its inferiors. Test and comparison indicate that the improved IBEA maintains the advantage of the original IBEA and shows better performance for maintaining diversity than original IBEA.

Key words: multi-objective optimism, evolutionary algorithm, quality indicator

中图分类号: