Energy Management System Industrialization for Off-Grids Power Systems Based on Data-Driven Machine Learning Models

This article presents the design and industrialization of an Energy Management System (EMS) for off-grid power system composed by a solar PV plant, some gensets, a battery storage and some load consumption. The goal of this article is to provide a clear insight of the different steps needed to achieve industrial maturity for this type of application. An important part of the solution deals with the use of data-driven machine learning models to simulate and optimize the EMS strategies. Several strategies for optimal EMS are presented and compared. The industrialization process including testing on a dedicated Controller-Hardware-in-the-loop (C-HIL) platform is described