一种模糊控制小生境遗传算法的应用研究  

Application study on a fuzzy-controlled niche genetic algorithm

作　　者：牟在根[1] 梁杰[1] 隋军[2] 颜谋[1] 

机构地区：[1]北京科技大学土木与环境工程学院,北京100083 [2]广州市市政工程设计研究院,广州510060

出　　处：《北京科技大学学报》2006年第3期299-302,共4页Journal of University of Science and Technology Beijing

基　　金：国家自然科学基金(No.50078004)

摘　　要：基于遗传算法的基本原理,提出一种改进的遗传算法,将模糊控制思想与小生境技术引入到其中,从而保护种群的多样性,同时使每代最优解得以保存．遗传算法加入小生境技术后虽可保持种群群体的多样性,但是不可避免的会产生部分个体的早熟以及陷入局部最优,于是加入模糊控制思想,对种群的交叉概率 P_c 和变异概率 P_m 进行模糊控制,以此为基础,形成了一种新型的模糊控制小生境遗传算法．最后通过对三个典型函数的数值分析证明了该方法的有效性和可行性．Based on the keystone of genetic algorithm （GA）, improvements are made to simple genetic algorithm （SGA） in two aspects. The theory of fuzzy control and the niche technique are introduced into the GA, for the purpose of enhancing the population diversity and maintaining the best part of each generation. In order to avoid premature convergence and occurrence of minimal deceptive problems, which is caused by the niche technique, fuzzy control is presented for the controlling of the crossover probability Pc and mutation probability Pm. Above all, that is the new type of algorithm-fuzzy controlled niche genetic algorithm （FNGA）. Through comparisons to FGA and NGA with the optimization of several functions, the result of the new algorithm shows its feasibilityand reliability.

关 键 词：遗传算法 小生境技术 模糊控制 交叉概率 变异概率 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]