摘要
云模型在超熵变大时,体现出雾化特性,云滴离散程度变大但靠近概念核心的云滴不失数量优势.雾化特性有利于表示进化算法中的遗传与变异,在云进化算法(CBEA)中,基于超熵变化的控制进化策略能够合理的调整选择压力,从而决定进化方向.实验表明,算法在寻求函数最优解问题上表现出良好的效率和精度.
The drops of the cloud spread around while the hyper entropy (He) is increasing. But many drops still stand in the central area of the cloud. Atomized feature of the cloud model can be used to adjust the strategies of the evolution. In cloud based evolutionary algorithm, changing He to affect the selection pressure and lead to different evolution result.By the experiments of nine typical test functions' optimization, the precision, stability and convergence rate of the algorithm are well proved.
出处
《电子学报》
EI
CAS
CSCD
北大核心
2009年第8期1651-1658,共8页
Acta Electronica Sinica
基金
国家973重点基础研究发展规划(No.2007CB310803)
关键词
进化算法
云模型
进化策略
雾化特性
genetic algorithm
cloud model
evolution strategies
atomized feature