Research On Control Of PH Neutralization Process Based On Gauss Process Model

PH neutralization process reveals the quality of chemical reaction in the fields of chemical and medicine industry and sewage treatment. The level of control effects the quality and quantity of production directly, so it takes an important position in process control. However, pH neutralization process is a typical kind of nonlinear complex process. The special feature of time-varying,nonlinearity and wide lag make it difficult to build accurate mathematical model, thus as recognized, this is one of the most complex control object. As reported it can not make a quick and precise response with regular control methods, they may even cause concussion in the control process sometimes. Optimal control will be even more impossible.Compared with the typical nonlinear model, Gaussian process model in can be better in identification of the prediction model and can give the uncertainty of the probability measure.Based on Gaussian Process,the author Modelingthe control model of the pH value based on the confection process of sodium nitrate during the manufacture of acrylon.and Depth discussing the problem of their control.First,based on the reaction mechanism of electrochemistry, the reaction situation of acid-base neutralization had been analyzed deeply. And according to the equation of charge balance, the mechanism model of acid-base neutralization, which is designed for the research of sodium nitrate confection process, has been derived. Based on the mechanism model, the emulation model is created with the help of software module--Simulink in MATLAB. The issue of pH value control in neutralization process is researched, regarding the emulation model as the control object.Gauss process model is chosen in my study. Based on the structure feature of the model and the principle of least-squares, the off-line recursive identification for parameter estimation method is discussed. At last, according to the strategy of predictive control, real-time control emulation is adopted to the control object with reasonable parameters. This is also compared with the PID control method and linear predictive control method in the aspect of control effect. The result indicates that, according to the pH neutralization of the experiment equipment of sodium nitrate confection process, the predictive controller based on Gauss process model has shorter steady duration, less overshoot and static error. In that case, the nonlinear control problem of pH value has been solved well. The future potential tasks are pointed out in the end.