Nonlinear predictive control based on NARX models with structure identification

摘要

A new nonlinear model predictive control (NMPC) based on NARX modeling with structure detection algorithm is presented in this paper. The scheme employs a nonlinear structure identification procedure known as Kortman-Unbehauen [12-13] algorithm. A suitable NARX model is then stated which enables to compute the prediction used to design the nonlinear model predictive control law. Performance and robustness properties of the new scheme are illustrated by several simulation results.