摘要
该文针对自适应模糊控制器的多参数优化问题,提出一种自适应遗传算法同时优化模糊规则和隶属函数的方法。先对隶属度函数和控制规则进行联合编码,遗传进化前期采用锦标赛精英保留,后期采用基于轮盘赌的非线性选择方法,保留了种群中较优个体,提高种群的多样性。采用一种自适应交叉变异算子,使交叉变异概率根据进化过程不断自动调整,避免算法过早收敛,加快收敛速度。仿真结果表明,优化后的自适应模糊控制器具有良好的动静态特性。
An adaptive genetic algorithm is introduced to optimize the membership functions and the fuzzy control rules in SFPID controller. Joint coding is done with membership functions and rules. With elitist strategy adopted in early genetic evolution stage, a nonlinear selection method based on roulette selection（RS） ,is introduced in last period. Optimum individuals are preserved and population diversity is increased. An adaptive crossover mutation operator makes the probability of crossover and mutation adjusted with evolution stage changing, that prevents prematurity and speeds up convergence. Simulation results on a second-order control object show that the optimized SFPID controller has good dynamic and static characteristics.
出处
《杭州电子科技大学学报（自然科学版）》
2011年第3期58-61,共4页
Journal of Hangzhou Dianzi University：Natural Sciences
基金
浙江省重大科技专项资助项目(C01015-2)
杭电科研基金资助项目(KYF041505004)
关键词
模糊控制器
自适应遗传算法
隶属度函数
控制规则
fuzzy logic PID controller
adaptive genetic algorithm
membership functions
fuzzy control rules
