基于遗传粒子群算法的高维复杂函数优化方法  

Genetic and particle swarm algorithm-based optimization solution for high-dimension complex functions

作　　者：于万霞[1] 张维存[2] 郑宏兴[1] 

机构地区：[1]天津工程师范学院电子工程系,天津300222 [2]河北工业大学管理学院,天津300130

出　　处：《计算机工程与应用》2007年第36期31-33,共3页Computer Engineering and Applications

基　　金：国家自然科学基金(the National Natural Science Foundation of China under Grant No.60671009)

摘　　要：针对高维复杂函数优化的特点,提出了一种遗传算法与粒子群算法相结合的主-从结构算法。算法中,主级为全局搜索的遗传算法;从级为局部邻域搜索的粒子群算法。通过主-从协调机制和从级转换函数设计,使算法不依赖复杂的编码方式和进化算子进行全局精确搜索。通过仿真和比较实验,验证了算法对高维复杂函数优化的有效性。A hybrid of genetic and particle swarm algorithm is proposed to solve the higen-dimension complex functions optimization.The algorithm is formulated in a form of hierarchical structure.The global search is performed at the master level by genetic algorithm,while the local search is carried out at the slave level by particle swarm optimization.Through the harmonizing mechanism between master and slave level,and special translation function designed for the slave level,the algorithm can execute global exact search without relying on complex coding and complex evolving operators.The simulation and results from comparison with other algorithms demonstrate the effectiveness of the proposed algorithm for high-dimension complex functions optimization.

关 键 词：遗传算法 粒子群算法 算法结构 转换函数 优化 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]