Predictive Control Based On RBF-ARX Model Applied In A Level Control System

Predictive control based on RBF-ARX is a new forward problem on study in the field of complex nonlinear industrial process control. This paper, according to predictive control theory, discusses RBF-ARX model's structure and predictive control based on RBF-ARX after analyzing two kinds of basic predictive control strategies. RBF-ARX model and predictive control based on RBF-ARX are applied to a level control system successfully.Firstly, the background, the development, the tendency and Problems of predictive control are summarized briefly. The origin and applications of RBF-ARX are also reviewed. Aiming at the typical predictive control algorithm (model algorithm control, generalized predictive control), the basic structure and theory of the predictive control are discussed. A detailed analysis including predictive model, methods of revising feedback and receding horizon optimization is made.Then, RBF-ARX model's structure and the theory of predictive control based on RBF-ARX are discussed strongly. RBF-ARX is an ARX model with Gaussian radial basis function (RBF) coefficients depending on the working points of a system. The RBF-ARX model is constructed as a global model, and is estimated off-line by SNPOM which is a fast-converging hybrid method composed of LMM and LSM. The predictive control based on RBF-ARX also does not require on-line parameter estimation and the quadratic programming is used to compute the controller's outputs. Because the local linearization of the system at each working-point may be easily obtained from the global RBF-ARX model, so there is no need to use nonlinear programming techniques.Finally, RBF-ARX model and predictive control based on RBF-ARX are applied to a level control system successfully. The results are prefect, the SNPOM method was fast-converging and the predictive errors of RBF-ARX model of the system are so small that we can conclude that the RBF-ARX model of the system can describe the system's dynamical characters fully. In real control, the control performance of predictive control based on RBF-ARX is much better than the control performance of predictive control based on local linear ARX model, predictive control based on global linear ARX model and PID.