摘要
克隆选择算法随机产生种群的方式,将容易导致数字的取值非均匀的分布在解的空间,从而增加数据冗余的现象。为了克服克隆选择算法的缺点,将克隆选择算法和混沌优化相结合,提出一种用于函数优化的混沌克隆优化算法.该算法利用混沌的随机性、遍历性和规律性来避免陷入局部极小值,同时引入等价划分的策略,减少了可能出现的数据冗余现象。仿真实验显示了所设计的算法能以较快的速度完成给定范围的搜索和全局优化任务。
The way of clonal selection algorithm randomly generated population that will easily lead to numbers of non-uniform distribution of values in the solution space, thus increasing the data redundancy phenomenon. To overcome the shortcomings of clonal selection algorithm, a chaotic clonal optimization algorithm for function optimizing is proposed by combining clonal selection algorithm, chaos optimization. This algorithm uses chaotic characteristics-randomness, ergodicity and regularity to avoid trapping around local optimal. Equivalent division strategy is introduced by reducing the possible data redundancy phenomenon. The simulation results show that the proposed algorithm can converge to the global optimum at quicker rate in a given range.
出处
《重庆大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2012年第8期34-41,共8页
Journal of Chongqing University
基金
国家自然科学基金资助项目(60803095,60975050
60902053)
高等学校博士学科点专项科研基金资助项目(20070486081)
中央高校基本科研业务费专项资金(ZY11005)
关键词
混沌优化
克隆选择算法
等价划分
数据冗余
chaotic optimization
clonal selection algorithm
equivalence division
data redundancy