Identification and management of an electrical storage system for application in photovoltaic installations

The aim of this paper is to propose a MPC (Model Predictive Control)-based hierarchical architecture for microgrids characterized by photovoltaics and storage systems. The first part of the work is focused on the storage model's identification during the charge and discharge phases with the aim of defining appropriate constraints to be inserted in the decision model. In the higher level, an optimization problem is developed and solved under a MPC scheme. The lower level, implemented on a PLC, receives references from the higher level and applies simple heuristics rules for real time control, on the basis of weather forecasts and data measured in real time. The developed algorithms have been applied to a real case study: a portion of the Savona Campus (University of Genoa) polygeneration microgrid that includes a photovoltaic field, an electric storage system, a load, and a connection with the electric grid.