A Nonlinear Model Predictive Control Based on Least Squares Support Vector Machines Narx Model

Abstract

In the domain of industry process control, the model identification and predictive control of nonlinear systems are always difficult problems. To solve the problems, an identification method based on least squares support vector machines for function approximation is utilized to identify a nonlinear autoregressive external input (NARX) model. The NARX model is then used to construct a novel nonlinear model predictive controller. In deriving the control law, a quasi-Newton algorithm is selected to implement the nonlinear model predictive control (NMPC) algorithm. The simulation result illustrates the validity and feasibility of the nonlinear MPC algorithm.