Modeling and control of systems with signal dependent nonlinear danymics

A modeling and control method for a class of nonlinear systems whose working-point varies with some process variable are presented. Based on two global NARMAX (nonlinear ARMAX) models, called the ExpARMAX (exponential ARMAX) models, a multi-step predictive control algorithm without resorting to on-line parameter estimation for the nonlinear systems are proposed. The one of the models is used as the internal model of the predictive controller, and another is applied to predict future values of measurable disturbances used in the predictive controller. The global ExpARMAX models only require off-line identification, and their local linearization forms are similar to linear ARMAX models. Case studies on a Nitrogen Oxide (NOx) decomposition (de-NOx) process of thermal power plants verified the validity.