EXTENDED ROBUST MODEL PREDICTIVE CONTROL FOR INTEGRATING SYSTEMS

This paper proposes a robust model predictive control for systems with stable and integrating poles. The approach combines the methods developed in Odloak (2004) and Carrapiço and Odloak (2005) to obtain a robust controller for integrating systems when multi-plant uncertainty is considered. The key idea in this development is to separate the control problem in two sub-problems, each of which takes into account the required robust constraints. Nominal stability is achieved by using an infinite output horizon, and the whole method is based on a state space model formulation that leads to an offset free MPC. The simulation examples illustrate the performance and robustness of the proposed approach and demonstrate that it can be implemented in real applications.