欢迎访问中国科学院大学学报,今天是

中国科学院大学学报 ›› 2003, Vol. 20 ›› Issue (3): 316-320.DOI: 10.7523/j.issn.2095-6134.2003.3.009

• 论文 • 上一篇    下一篇

一种引入强制变异的改进遗传算法

孔祥蕾1,2, 张先燚2, 罗晓琳1,2, 李海洋1,2   

  1. 1. 中国科学院安徽光学精密机械研究所环境光谱学实验室, 合肥 230031
    2. 中国科学院研究生院, 北京 100039
  • 收稿日期:2002-07-15 修回日期:2002-11-01 发布日期:2003-05-10
  • 通讯作者: 李海洋,Email:hli@aiofm.ac.cn
  • 基金资助:

    国家自然科学基金资助项目(20073042)

An Improved Genetic Algorithm with Enforced Mutation

KONG XiangLei1,2, ZHANG XianYi2, LUO XiaoLin1,2, LI HaiYang1,2   

  1. 1. Laboratory of Environmental Spectroscopy, Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Hefei 230031, China ;
    2. Graduate School of the Chinese Academy of Sciences, Beijing 100039, China
  • Received:2002-07-15 Revised:2002-11-01 Published:2003-05-10

摘要:

针对常用遗传算法存在容易产生过早收敛的问题,提出了一种将强制变异、最佳解保留和自适应交叉变异参数调整相结合的改进遗传算法。这种方法将进化过程中群体的平均适应度与最大适应度进行比较,以确定是否需要对群体实施强制变异或采用自适应交叉、变异概率调整。数值模拟的结果表明,这种方法可有效地克服早熟现象,提高全局优化能力

关键词: 遗传算法, 早熟, 强制变异, 自适应参数

Abstract:

An improved genetic algorithm is developed, with the addition of three new operations :enforced mutation;direct preserving of best chromosome and using adaptive parameters.The method compares average fitness of pop with maximum fitness, and makes sure the necessity of enforcedmutation or crossover and mutation with adaptive parameters.The simulating results indicate the improvedmethod can prevent premature and realize global-optimization effectively.

Key words: genetic algorithm, premature, enforced mutation, adaptive parameters

中图分类号: