The Control Algorithm Study About The Variable Parameter Pressure Object

With the industrial production and scientific technology is the progress of the controlled objects, then the nonlinear time-varying,hysteresis quality more and more obvious, and the performance of these objects with analytic expression to precise mathematic model. So with classical control algorithm is difficult to achieve satisfactory control effect.In this paper, the industrial production of pressure, variable parameter is proposed based on PID control algorithm of neural network which control based on PID control algorithm. By using the neural network, according to the real-time target condition changes the control parameters adjustment PID algorithm. This algorithm has two steps which is crude adjust and fine adjust. Adjustable phase in the offline using RBF networks, identify the mathematical model of the controlled objects. It can instead of practical control parameters of the controlled object preliminary setting. In the online fine tuning process, due to the control parameters have been preliminarily adjustment, so that is not accused of fluctuation, improve the system of quality control.This paper introduced the basic theory of neural network algorithm, and studies emphatically introduces the paper used BP algorithm and RBF algorithm. Then introduces several common neural network identification model and neural network control strategy, and then presents the hardware structure of the system and the main component selection and software into problems. The control algorithm and the design program, the experimental research and simulation studies. Simulation shows that the structure, the paper puts forward the control algorithm than classical control algorithm has good control effect.This research work for the variable parameters of control algorithm research object of pressure offers a new way.