基于神经网络的模糊自适应PID控制方法研究

摘要

Because the traditional PID controller can't modify the parameter and its fuzzy rule and rate function depend on experts' experience, this paper puts forward a kind of fuzzy adaptive PID controller based on neural network,which synthesizes the advanta ges of PID control, PID fuzzy control and neural network controller and has wide PID control adaptability, learning and adaptive func tions and the nonlinear control ability of fuzzy control. It is proved that this controller is more frequent and more smooth controller.%针对常规PID控制器不能在线修正参数以及模糊规则和率属函数对专家经验的依赖性,提出了神经网络模糊自适应PID控制器,从而综合了传统PID控制、模糊控制、神经网络控制的优点,使其具有PID控制的广泛适用性和神经网络的自适应和自学习能力,同时又具备模糊控制的非线性控制作用；仿真实验可知该控制器具有更快的响应和更好的平稳性.