说明：  最近在学习基于神经网络的电机智能控制，神经网络与PID有两种结合方式：采用单神经元结构，神经元输入权值一一对应PID的三个参数，神经元的输入值为经过比例、积分和微分处理后的偏差值；另一种是在常规PID控制器的基础上增加一个神经网络，用神经网络在线调节PID的三个参数。
(Recently, we are learning the intelligent control of motor based on neural network. There are two ways to combine neural network and PID: using single neuron structure, the input weights of neuron correspond to the three parameters of PID one by one, and the input values of neuron are the deviation values after proportional, integral and differential processing; The other is to add a neural network on the basis of the conventional PID controller, and use the neural network to adjust the three parameters of PID online.)