多模态函数优化的人工免疫应答优化算法  

Artificial Immune Optimization Algorithm and Its Application to Multi-modal Function Optimization

作　　者：续欣莹[1] 张志军[1,2] 谢刚[1] 谢克明[1] 

机构地区：[1]太原理工大学信息工程学院,山西太原030024 [2]山西财经大学信息管理学院,山西太原030006

出　　处：《太原理工大学学报》2009年第4期333-336,共4页Journal of Taiyuan University of Technology

基　　金：高等学校博士学科点专项科研基金资助项目(2006112005);山西省青年科学基金资助项目(2007021018)

摘　　要：针对传统的人工免疫算法收敛速度较慢,搜索精度较低的问题,基于免疫应答原理提出一种多模态函数优化的人工免疫应答优化算法。该算法将初始抗体群分配为记忆抗体群和一般抗体群,对记忆抗体群实施高频变异算子进行局部精细搜索,并对一般抗体群执行趋同过程算子进行全局粗搜索,从而加快了进化过程。通过典型多模态函数对该算法进行性能测试,实验结果表明,该算法具有更快的收敛速度和更高的搜索精度。Traditional artificial immune algorithm had slow convergence and low searching precision. To overcome these disadvantages, an artificial immune response optimization algorithm （AIROA） based on immune response principle was proposed in this paper. Initial antibody population was allocated as memory antibody population and temporary antibody population. Memory antibody population was driven by hypermutation operator for local searching. Temporary antibody population was driven by similartaxis procedure operator for global searching, which can accelerate the evolutionary process. Compared with CSA, simulation results on benchmark multimodal function showed that AIROA presented in this paper had advantages in convergence speed and precision.

关 键 词：免疫应答 优化算法 多模态函数优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]