Optimal cascaded predictive control for photovoltaic systems: application based on predictive emulator

The present study sheds new light on advanced control methods of photovoltaic (PV) emulators using finite set model predictive control (FS-MPC). In the first part of the study, a predictive PV emulator (P-PVE) based on a Buck converter is proposed and tested under hard climatic conditions and load variations. The high performance of the P-PVE in terms of dynamic response, reference tracking, accuracy, simplicity, and efficiency is confirmed experimentally when compared with those of the commonly used one based PI controller. The second part of the study proposes an efficient cascaded predictive control (CPC) method applied on two topologies of PV systems, namely the stand-alone system and the grid-connected system. In each topology, the P-PVE is cascaded to a maximum power point tracking Boost converter in order to track efficiently the maximum power point. In addition to the high performance offered by the FS-MPC, the proposed control strategy allows to control all cascaded converters at the same time in one stage instead of controlling them separately, thus providing more flexibility and simple controllability. Extensive experimental results are done confirming the correctness and the effectiveness of the proposed CPC under hard climatic conditions, even in the presence of distorted grid voltage.