基于混沌免疫克隆选择算法的TSP问题求解模型  

TSP Problem Solving Model Based on Chaos Immune Clonal Selection Algorithm

作　　者：倪志平[1] 余玲[1] 覃溪[1] 

机构地区：[1]广西科技大学鹿山学院电气与计算机工程系,广西柳州545616

出　　处：《科技通报》2016年第10期188-190,215,共4页Bulletin of Science and Technology

基　　金：2016年度广西高校中青年教师基础能力提升项目(KY2016YB842)

摘　　要：针对标准免疫克隆算法在求解TSP问题的过程中还存在收敛性不好、效率低下等问题。本文设计了一种以非线性混沌优化免疫克隆算法为基础的TSP问题求解模型,最先运用混沌变量完成抗体编码,利用混沌机制等产生克隆初始种群,然后对后代进行克隆,并将混沌算法引入到免疫克隆变异中,在进化中将混沌变量映射到实际优化问题中计算抗体的亲和度,之后再优化选择算子,为种群的多样性提供保证。实验仿真结果表明,本文提出的改进IA算法在执行时间和迭代次数上都优于传统免疫算法,在收敛问题上,比传统免疫算法更具有优势。According to the low convergence and efficiency of the standard immune clonal algorithm in solving TSP problems, a TSP problem solving model is designed based on immune clonal algorithm with the optimization of nonlinear chaos. First, chaotic variables are used to complete antibody encoding,using the chaos mechanism producing clones initial population. Then to cloned offspring, and the chaos algorithm is into the immune clone variation, the affinity of the antibody is calculated in the evolution of chaotic variables that is mapped to a practical optimization problems. And then optimizing the selection operators, provide guarantee for the diversity of population. Simulations show that compared with the traditional immune algorithm, the proposed improved IA algorithm has batter performance on the execution time, the number of iterations and convergence.

关 键 词：免疫克隆算法 非线性混沌 抗体亲和度 种群多样性 选择算子 TSP求解 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]