基于遗传算法优化的自适应FNN-PID控制器研究  

Reseach of Self-adaptive FNN-PID Controller Optimized by Genetic Algorithm

作　　者：高相铭[1] 刘付斌[1] 

机构地区：[1]安阳师范学院物理与电气工程学院,安阳455000

出　　处：《科学技术与工程》2012年第27期6949-6954,共6页Science Technology and Engineering

摘　　要：针对传统PID控制器参数整定后因无法在线自动调整而导致控制效果不理想的问题,提出了一种基于遗传算法优化模糊神经网络(FNN,FUZZY NEURAL NETWORK)的自适应FNN-PID控制器模型。该模型结合了模糊神经网络良好的自适应自学习能力和遗传算法强大的全局搜索能力。利用遗传算法对模糊神经网络的参数进行优化与训练,使PID控制器能够根据被控对象的变化而适时在线调整自身参数KP,KI和KD,从而达到理想的控制性能。将该控制器应用于异步电动机控制系统进行仿真实验,结果表明:基于遗传算法优化的自适应FNN-PID控制器具有较好的自适应能力和鲁棒性,控制效果明显优于传统PID控制器。Conventional methods of designing the parameters of PID controller need a precise mathematic model of object,and moreover,the controller can not adjust itself on line to the variation of the surroundings and the object.In contrast,fuzzy neural network（FNN） has fine adaptive ability and learning ability not requiring any mathematic model of object.genetic algorithm（GA） is a new optimization method with global random searching ability.So a kind of FNN-PID control method is proposed in which genetic algorithm and fuzzy neural networks are mixed.The genetic algorithm is used to optimize the parameters of the FNN’s membership function,and the back propagation algorithm is used to optimize the connection coefficients of fuzzy neural network.The simulation results show the system based on adaptive FNN-PID controller has better adaptability and robustness,and has the advantages of higher ability to resist disturbance and adaptability to parameters changing than conventional PID controller.

关 键 词：FNN-PID 遗传算法 异步电动机 

分 类 号：TP273.21[自动化与计算机技术—检测技术与自动化装置]