首页 | 本学科首页   官方微博 | 高级检索  
     

分级变异的动态克隆选择算法
引用本文:胡江强,郭晨,尹建川,李铁山. 分级变异的动态克隆选择算法[J]. 控制与决策, 2007, 22(6): 608-612
作者姓名:胡江强  郭晨  尹建川  李铁山
作者单位:大连海事大学,航海学院,辽宁,大连,116026;大连海事大学,自动化与电气工程学院,辽宁,大连,116026
基金项目:国家自然科学基金项目(60474014);教育部高等学校博士学科点专项基金项目(20040151007);交通部应用基础研究项目(200432922504).
摘    要:基于浮点数编码,提出一种分级变异的动态免疫克隆选择优化算法.根据抗体的亲和度将种群分解为3个子种群,分配以不同的搜索任务,实施不同的变异策略.在进化过程中动态改变种群规模、克隆规模和变异参数,从而加快了全局搜索速度,提高了局部搜索精度.对5个复杂函数的优化仿真实验表明了该算法的有效性。

关 键 词:浮点数编码  克隆选择  进化算法  动态参数  函数优化
文章编号:1001-0920(2007)06-0608-05
收稿时间:2006-02-24
修稿时间:2006-02-242006-06-05

Dynamic clonal selection algorithm with classified mutation
HU Jiang-qiang,GUO Chen,YIN Jian-chuan,LI Tie-shan. Dynamic clonal selection algorithm with classified mutation[J]. Control and Decision, 2007, 22(6): 608-612
Authors:HU Jiang-qiang  GUO Chen  YIN Jian-chuan  LI Tie-shan
Affiliation:a. College of Navigation; b. College of Automation and Electrical Engineering, Dalian Maritime University, Dalian 116026, China.
Abstract:A dynamic immune clonal selection algorithm with classified mutation is proposed based on floating point coding. To speed up the global search and improve the local convergence precision, the following two main strategies are introduced. According to the antibody affinity in relation to the antigen, the antibody population is decomposed into several subsets, and they are submitted to respective mutation processes for their different given tasks. Then, the population size, the clone size and the mutation parameters are dynamically changed with evolution processing. The proposed algorithm is used to optimize 5 complex functions for testing and the results show its effectiveness.
Keywords:Floating point number codes Clonal selection   Evolutionary algorithm   Dynamic parameters   Function optimization
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《控制与决策》浏览原始摘要信息
点击此处可从《控制与决策》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号