一种基于家庭聚类思想的遗传算法
A New Family-clustering Genetic Algorithm
-
摘要: 从一个不同角度提出遗传算法的改进思想.这种新的遗传算法——Family Clustering GA(FCGA)和以往的改进方法不同的是,它根据种群中个体的相互关系,采用聚类思想对算法进行综合调控,利用"家庭选择"操作进行微调,使得收敛速度和多样性这两个指标都得到兼顾.从实际的仿真效果来看,效果显著.Abstract: A new improved genetic algorithm is presented in a different aspect. Different from other genetic algorithms and based on the relation between the individuals, this new family-clustering genetic algorithm (FCGA) regulates the optimization with the clustering method and controls in a micro way the individuals with family selection operator. With these operators the speed of convergence and the diversity of the population are well balanced. A good result is achieved in simulations.