Study On Intelligent PID Control

PID control is widely used in process control systems. The performance of the controllers is directly related to the stability of the processes and the quality of product, so it is very valuable for their optimal design.The character of PID controllers is simple structure, goodadaptability and great robustness, especially they don't depend on the precise models of the objects. They can also solve the difficulty of precise modeling in industrial process. Since they are used for a long time, the control engineers have already accumulated a great deal of tuning experience of PID controllers' parameters. But the simple PID controllers can't get the satisfied degree, especially for the time-varying objects and non-linear systems, the traditional PID controllers can do nothing for them.The methods of intelligent PID control are studied systematically in this paper. Firstly, PID control and its basic parameters auto-tuning methods are introduced in detail. Secondly, the method of relay feedback auto-tuning of PID controllers used widely is also studied. Then, single neural element and BP neural network are studied deeply, PID control parameters fuzzy optional algorithm and PID control parameters optional algorithm based on genetic algorithm are also studied. At last, a kind of single neural element self-adaptation control structure based on RBF neural network identification is proposed. The simulation results showed that this method had achieved good control effect.