The Research For Neural Network PID Control System

In order to enhance the performance of the controller, this paper combined the neural network and PID control, and deeply studied the neural network PID controller based on single-variable and multi-variable system.For single-variable system, the neural network PID controller based multi-step predictive performance target function combined the predictive control idea and the neural network PID control idea. It uses multi-step predictive performance target function to train the weights of neural network PID controller, This paper improved the system: It uses new multi-step predictive performance target function to train the weights, and it uses the dynamic recursion neural network instead of multiplayer feed forward neural network that is furthermore fit for real-time control to identify the part of neural network identification. The simulating results shows that this method has better response performance than the neural network PID control method with identificationFor multi-variable systems, at first, This paper studied the neural networks PID controller based multi-variable systems using multi-step predictive performance target function, After studying the system's simulating instances, I got the results; Then this paper studied the structure and arithmetic of the PID neural network multivariable controller. It is made up of paratactic multi-sub-network, if there are n controlled variables in controlling system, the sub-networks then will have n too. The input layer of Each sub-network accepted the present signal of the system and the output signal of controlled object; The hidden layer that is made up of proportion, integral and differential three parts realizes PID operation; The output layer realizes the integration of the rules; and its output layer's weights were adjusted using the least mean squares in stead of grads arithmetic in order to quicken the regulative speed of the weights, the results show that the system has much higher performance of self-studying and self-adapting.