Improved Grey Prediction PID Neural Network Application In Valve Position Control System

Pneumatic regulating valve is widely used in many complex process control occasions such as chemical,textile,petroleum,pharmaceutical and metallurgy because of its simple structure,easiness of using,roboust and realible.Valve positioner is the key point for Pneumatic regulating valve to achieve high quality adjustments.Valve positioner is one of the control units of the regulating valve.By using the Pneumatic regulating valve,we can effectively improve the control accuracy,speed and flexibility.The importance of pneumatic regulating valve can never be exaggerated in process industrial fluid control.Research on pneumatic valve positioning system on technical level is as important as on system modeling and other theoretical problem.Modeling the valve positioning system is the foundation of the improvement of its control strategy.In this paper,the first goal is to establish a simulation platform for pneumatic valve positioning system.The structure and working principle of the pneumatic valve positioning system are introduced and analyzed.The system consists of three parts: the valve positioner,the pneumatic actuating mechanism and the position feedback device.On the basis of analyzing the working principle of each parts of the system,the mathematic models of various parts of the system are established and perfected.The model Parameters are determined by the product nameplate and experimental calculation ultimately through making use of the modularization of the mathematical model of various parts in valve positioning system which be achived by Simulink.After all conditions are considered,the simulation platform of the valve positioning system is set up.This paper introduces the PID control and fuzzy PID control strategy of the valve positioning system.The conventional PID strategy strategy has poor performance in real-time and adaptive tuning.Owing to delayed adjust,the fuzzy PID control strategy can not quickly eliminate the error of the valve positioning system.Based on the analysis of the fuzzy control and the gray prediction theory,an improved gray prediction fuzzy PID control strategy is put forward combining with control requirements of the valve positioning system.Fuzzy PID reduces the system deviation by adjusting parametersonline.The gray effect of gray prediction changes from a constant value to a linear function of time,improving the prediction accuracy.In order to realize advance control and solve the problem of poor performance of the aerodynamic system,the improved gray predicting strategy extracts the data sequence of the valve positioning system and predicts the future development trend.In order to test the performance of each module and the effect of the whole simulation platform,the dynamic performance of the simulation platform is verified under 50% and 100% step input.Based on the improvement of the simulation platform,the control effect of the gray prediction fuzzy PID is improved and the control effect of the variable pressure fluctuation and the gas source fluctuation is simulated and contrasted under the main interference condition.