A Data-Based Adaptive Predictive Controller Design for Stabilizing Unknown Nonlinear Systems Featuring Input and Output Saturations

Abstract

This article proposes a Data-Based Adaptive Predictive Control (DBAPC) scheme for a set of unknown nonlinear dynamical systems featuring Input and Output (I/O) saturations. In the beginning, a new dynamic model is provided for a discrete-time unknown nonlinear system by considering I/O saturated data. Furthermore, according to the developed model, a predictive control method is designed for stabilizing the system. Hence, Input and Output saturations are common physical constraints in industrial procedures; the stability analysis and proving the boundedness of the tracking error are provided in the presence of the limitations mentioned above. Proving the stability of the proposed controller in the presence of I/O saturations makes it more applicable than conventional methods of model free adaptive based predictive control. Simulation studies on the Load Frequency Control (LFC) problem of an interconnected three-area power system and one numerical example reveal the advantage and applicability of the proposed controller.