Research On Modeling And Control Strategy Of Pneumatic Control Valve Control System

As an indispensable actuator in the automatic production process of modern chemical plants,the control effect of control valve directly affects other links.The control effect of the regulating valve depends on its hardware performance and the control algorithm of the intelligent valve positioner.With the improvement of the production technology level,the influence of the hardware performance on its control effect is becoming smaller and smaller.At present,researchers focus on the research of the control algorithm.Domestic research on intelligent valve positioner started late,and there is still a certain gap between domestic and foreign products.On the basis of consulting domestic and foreign literature and reference materials,the following work has been carried out:(1)First of all,the working principle and composition of the pneumatic control valve control system are introduced,then the main mechanisms that affect the control accuracy are modeled mathematically,and the factors that affect the control accuracy of the valve are analyzed using the mathematical model.(2)This paper described the principle of the traditional five-step switch control algorithm and the shortcomings in practical work.In view of these shortcomings,an improved control algorithm is proposed.The improved algorithm further divides the control range,adds a deceleration zone after the B-B control,and changes the PWM signal control to the composite pulse modulation signal control.The control parameters and parameter tuning methods needed for the improved algorithm are introduced.The simulink simulation platform of the system is built to verify the feasibility of the improved algorithm.(3)The parameters of pulse modulation control cannot be obtained by valve selftuning.An intelligent algorithm is proposed to establish the mathematical model of valve control,and then the parameter tuning method of optimal control parameters is found in the established mathematical model.The fitting effect of the unoptimized BP neural network is poor.Different methods are used to optimize the BP neural network,and then the optimization effect is compared with the LSTM algorithm.It is found that the fitting effect of the particle swarm optimization neural network is the best.Finally,the particle swarm optimization algorithm is applied to optimize the parameters of the model with better optimization effect.(4)The experimental platform of the pneumatic control valve control system is built,and the hardware used is introduced in detail.This paper introduces the software used for programming and the structure of the program,introduces the function of the main modules of the program,and shows its structure.The traditional five-step switch control algorithm and the improved algorithm are used to control the regulating valve,and the control effect is compared and analyzed.It was found that the performance of the improved valve control algorithm was improved in all aspects.