生存迁移算法  

Living Migration Algorithm

作　　者：李豆豆[1] 邵世煌[1] 齐金鹏[1] 

机构地区：[1]东华大学信息科学与技术学院,上海201620

出　　处：《系统仿真学报》2008年第8期2034-2038,共5页Journal of System Simulation

基　　金：教育部高校博士点科研基金(20060255006)

摘　　要：基于生物迁移的机理,提出了一种新算法——生存迁移算法。每个生物体都存在着生存极限,生存极限会随着时间不断演化,而生物体为了生存又会在空间上不断地迁移,当全部个体所在位置的生存状况都能满足生存极限时,就认为达到了最佳的生存位置。文章分析了算法的机理与过程,用其解决了四种类型的非线性最优化问题,并进行了比较分析,实验结果说明生存迁移算法具有较快的收敛性,能较好处理一些优化问题。A novel optimization algorithm based on the concept of biomigration was proposed, named Living Migration Algorithm （LMA）. Each individual inheres the minimal life-needs changing with evolvement, and an ability of migrating during the living space for survival. If only the minimal life-needs of all individuals can be satisfied by their life-conditions at certain generation, these space are considered as the optimal living places. The mechanism and the operating flow of LMA were analyzed. It is employed into four non-linear and heterogeneous optimizations for testing its performances. The experiments show that LMA has quick convergence, and can successfully solve many optimal problems.

关 键 词：生物迁移 自适应计算 进化算法 启发式算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]