Nonlinear Predictive Control Based on the Combination of Steady-State Nonlinear Model and Linear ARX Model

By representing the steady-state model of nonlinear system with a map of a set of numerical values and modulating the dynamic gains of ARX(AutoRegressive with eXternal input) model according to the system steady-state gains,a nonlinear model predictive control algorithm based on the combination of steady-state nonlinear model and linear ARX model is proposed.The algorithm carries out on-line identification of the system dynamic model parameters with recursive least square method and solves the objective function with sequential quadratic programming.Simulation is made in a representative nonlinear chemical process,i.e.,the pH neutralization process,to validate the presented algorithm,and the results demonstrate that the proposed method has a better performance in tracking set-point and restraining disturbance than the generalized predictive controller.