The Design And Application Of PH Intellectualized Controller System Based On Conduit Blender

Neutralization process is a typical nonlinear process with pure time delay. As we know, this process can't be controlled effectively with conventional linear control method. The nonlinear gain compensation PID method is employed in the practical application. In this paper the realization of this control system and its result are introduced. Because the process of the reaction is complicated, it's difficult to model, so the nonlinear gain PID method can't identify the changes of the reaction quickly and accurately, and optimal control can't be got from this method.This paper introduced several typical methods used in pH control and something about the waste water treatment. From that, we know, a usual method is using neural networks to simulate the state memory mechanism of dynamic system and predictive control strategy to realize the optimal control. Neural networks is viewed as a universal approximator for nonlinear functions, but there are some fatal bugs with it, so it is necessary to find another method more right for neutralization process.Support Vector Machines (SVM) is an important area of SLT, which developed after the neural networks. In this paper, predictive control strategy based on support vector machines (SVM) is presented. And the theory of support vector machines and the reality of regression modeling are described. We have done some experiments to demonstrate the validity of the SVM learning method. Besides, a Model Algorithmic Control (MAC) based on SVM is presented, too. The model uses nonlinear least squares Levenberg-Marquadt (LM) as method of receding, and the demonstration shows that results of SVM MAC are better than the nonlinear gain PID and SVM MAC has more rejection capacities for unknown disturbance.At last, a summary of the paper is given and further research interests are also proposed.