Model Predictive Control for Nonlinear Singularly Perturbed System Based on Linearization Method

A new model predictive control design method based on state feedback linearization is proposed for a class of nonlinear singularly perturbed system. Firstly, the system is decomposed into slow subsystem and fast subsystem by singularly perturbed theory. Secondly, the decomposed system is transformed into a linear system by feedback linearization. Then the model predictive controller is derived by predictive control theory, such that the quadratic performance index achieved minimum value. This controller design method selects different quadratic performance indicators for fast and slow subsystems respectively and avoids solving complex nonlinear equations. Finally, simulation results verify the effectiveness of the proposed method and the states of the closed-loop system are convergent.