摘要
针对IAGA自适应遗传算法存在的未成熟收敛问题,提出了一种改进的自适应遗传算法(NIAGA算法),根据自定义判别式判断群体是否出现了未成熟收敛趋势,由不同情况,分别采用宏观调控与微观处理两种方法来设置交叉概率Pc和变异概率Pm,以此促使算法摆脱未成熟收敛.仿真结果表明,新算法有效地改善了IAGA算法的未成熟收敛问题,显示出了更强的全局收敛性.
In view of the problem of the premature convergence of the IAGA adaptive genetic algorithm,an improved adaptive genetic algorithm（NIAGA algorithm） is proposed.The custom discriminant determines whether a group appeard a trend of the premature convergence,then both macroeconomic regulation and control and micro processing method are used to set the crossover probability Pc and mutation probability Pm respectively by the different situation and reduce the possibility of trapping in the premature convergenceThe simulation result shows that the new algorithm improves the problem of the premature convergence of the IAGA algorithm effectively and has better global convergence.
出处
《数学的实践与认识》
北大核心
2015年第19期259-264,共6页
Mathematics in Practice and Theory
关键词
自适应遗传算法
交叉概率
变异概率
收敛性
adaptive genetic algorithm
crossover probability
mutation probability
convergence
