一种合作型协同生物地理学优化算法  

Cooperative Coevolutionary Biogeography-based Optimization

作　　者：王晓光[1,2] 郑向伟[1,2] 

机构地区：[1]山东师范大学信息科学与工程学院,山东济南250014 [2]山东省分布式计算机软件新技术重点实验室,山东济南250014

出　　处：《小型微型计算机系统》2014年第8期1837-1841,共5页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(61373149;61272094)资助;山东省中青年科学家奖(BS2010DX033)资助

摘　　要：生物地理学优化算法(Biogeography-Based Optimization,BBO)是一种基于迁移共享信息的智能算法,其结构简单有效,但解决复杂问题时效果不佳.基于分而治之策略和协同进化框架,提出一种合作型协同生物地理学优化算法(Cooperative Coevolutionary BBO,CBBO).首先将种群划分为多个子种群,采用BBO演化每个子种群,然后选择其它子种群中最好的栖息地,为当前子种群的栖息地构建上下文向量,并通过目标函数进行评价.针对6个基准函数测试CBBO算法,并与BBO、PSO、ACO、DE算法进行比较,结果表明,CBBO算法在收敛速度和收敛精度方面优于BBO等其它算法.Biogeography-Based Optimization （BBO） is an intelligent optimization algorithm inspired by biogeography mathematics. The structure of BBO is simple and efficient and BBO has its unique migration operator and mutation operator, but has poor perform- ance when is used to solve complex problems. Therefore, a new Cooperative Coevolutionary Biogeography-based Optimization （ CB- BO） combining divide-and-conquer paradigm and coevolution is proposed in this paper. In CBBO, whole population is divided into several sub-populations and each is evolved with BBO. Habitats in current sub-population are evaluated with context vectors construc- ted by selecting the best habitat of the other sub-populations. Tests on six benchmark functions and comparison with BBO and several other evolutionary algorithms show that CBBO demonstrates better performance in convergence rate and accuracy.

关 键 词：生物地理学优化算法 迁移算子 协同进化 上下文向量 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]