模糊神经网络实现的PID参数自调整控制及应用研究

模糊控制学习功能差，神经网络推理功能差，把二者结合可以起到互补作用，保证推理和学习功能的实现。本文用多层ＢＰ神经网络实现了一类模糊推理过程，该网络通过学习记忆ＰＩＤ参数调整的基本规则，实现了ＰＩＤ控制器参数的在线调整。通过对考虑非线性因素的伺服系统控制对象的仿真，说明了该方法的有效性。

Self-Tuning PID Control Utilized by Fuzzy Neural Network and Its Application

Fuzzy control has superiority in inference and neural network has superiority in leaning. The function of inference and learning can be realized at reasonable cost by Combining fuzzy control with neural network. In this paper, one fuzzy inference method has been utilized by multilayer BP neural network. The network is used to remember adjusting rules of PID parameters by learning so the network can adjust PID parameters on line by rules. The effectiveness of the method is demonstrated by simulations in servo-system with nonlinear factor.