• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2016, Vol. 38 ›› Issue (01): 62-66.

• 论文 • 上一篇    下一篇

一种用于进化算法历史计算数据的高效利用方法

闫盼,谭瑛,张建华   

  1. (太原科技大学工业与系统工程研究所,山西 太原 030024)
  • 收稿日期:2014-11-12 修回日期:2015-04-24 出版日期:2016-01-25 发布日期:2016-01-25
  • 基金资助:

    国家自然科学基金(61472269,61403272)

A method of using historical calculation data
efficiently in evolutionary algorithms 

YAN Pan,TAN Ying,ZHANG Jianhua   

  1. (Institute of Industrial and System Engineering,Taiyuan University of Science and Technology,Taiyuan 030024,China)
  • Received:2014-11-12 Revised:2015-04-24 Online:2016-01-25 Published:2016-01-25

摘要:

进化算法由于其强大的系统建模能力和空间搜索能力已被广泛应用于许多实际问题的求解中。然而,在算法进化的过程中存在个体适应值重复计算的问题,尤其在解决实际工程中的复杂问题时,适应值的计算会消耗大量时间。为此,利用哈希表的高速存取能力,将哈希表用于存取适应值的历史计算数据,从而避免优化过程中适应值的重复计算,并且对优化结果没有任何影响。仿真实验结果验证了此方法的有效性。

关键词: 哈希表, 历史计算数据, 进化算法

Abstract:

Evolutionary algorithms have been widely used in practical problems for their remarkable system modeling capability and spatial searching capability. However, there is a problem of repetitive computation for individual fitness in the process of evolutionary algorithms. Especially when solving complex engineering problems, fitness calculations can expend a large amount of time. The hash table features highspeed access capability, which can be used to access historical data of fitness and solve repetitive computation problem of individual fitness during the optimization process without affecting the results. Simulation results prove the efficiency of the proposal.

Key words: hash table;historical calculation data;evolutionary algorithms