Model predictive control of power converters: theory and practice

This thesis studies the use of model predictive control (MPC) to handle power converters. The focus is on technical and theoretical issues regarding the use of a particular class of predictive strategy called Finite Control Set MPC (FCS-MPC). The main advantage of this predictive strategy comes from the fact that switching actions are explicitly taken into account as constraints on the control input of the system in the problem formulation. Consequently, modulation stages (to handle the converter switches) are not required. Throughout this thesis, we illustrate how to implement this predictive technique to improve the performance, in terms of power quality and dynamic response, in some classes of power converters. We show that FCS-MPC allows power converters to operate near to their limits of achievable performance when. In this thesis we also show that the benefits of using FCS-MPC can be obtained not only to handle power converters under normal operating conditions but also to achieve fault tolerance of these devices. Additionally, in this thesis we present a presented a stability analysis of MPC for linear time-invariant (LTI) systems with discrete input alphabets. Based on our theoretical results, we also analysed stability and performance of power converters when governed by FCS-MPC. The latter study is focused on power converters that can be modelled as LTI systems with quantised inputs.