Decoupled Multimodel Predictive control for Discrete-Time Nonlinear Singularly Perturbed System

The present paper deals with a multimodel predictive control for discrete-time nonlinear singularly perturbed systems based on decoupled multimodel structure. For each local model a partial predictive control is synthesized. the weighted sum of these partial controls gives the global control law to be applied to the nonlinear singularly perturbed system. An algorithm is established to synthetisize the adequate multimodel predictive control which garantee to the system the closed-loop desired performances. Simulations results are presented to illustrate the effectiveness of the proposed multimodel predictive control strategy.