摘要
针对遗传算法易早熟、局部搜索能力弱的问题,采用捕食搜索策略对遗传算法进行改进.在全局搜索中,提出一种基于信息熵的遗传策略,即利用当前实际种群熵与当前期望的种群熵的差值来自适应地改变选择压力、交叉概率和变异概率,以达到调整种群的多样性、改善早熟现象的目的;在局部搜索中,采用缩小算术交叉参数的范围和高斯变异的策略,以达到提高局部搜索能力的目的.通过对4个不同类型复杂函数的优化,结果表明该方法能有效地避免早熟现象的发生,能提高局部搜索能力和搜索速率.
Due to the problem of easy prematureness and the weak capacity of local search of the genetic algorithm,the predatory search algorithm is introduced to improve the genetic algorithm. For the global search,a genetic strategy based on information entropy is presented,that is,selection pressure,crossover probability and mutation probability are changed adaptively according to the difference between current actual population entropy and current expected population entropy,to achieve the purpose of adjusting the population diversity and improving the premature phenomenon. For the local search,the strategy of narrowing parameter rang of arithmetic crossover and the Gaussian mutation is used to achieve the objective of improving local search ability. Through optimization of four different types of complex functions,it is shown that the method can avoid the premature phenomenon effectively,improve the ability of local search and searching speed.
出处
《东南大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2010年第S1期223-227,共5页
Journal of Southeast University:Natural Science Edition
基金
国家高技术研究发展计划(863计划)资助项目(2006AA020301)
关键词
遗传算法
捕食搜索策略
多样性
函数优化
genetic algorithm
predatory search strategy
diversity
function optimization