An Energy Management System for the Control of Battery Storage in a Grid-Connected Microgrid Using Mixed Integer Linear Programming

This paper proposes an energy management system (EMS) for battery storage systems in grid-connected microgrids. The charging/discharging power is determined such that the overall consumption cost minimized, considering variation grid tariff, renewable generation and load demand. modeled as economic dispatch optimization problem over a 24 h horizon solved using mixed integer linear programming (MILP). formulation, therefore, requires knowledge of expected production demand next h. To achieve this, long short-term memory (LSTM) network proposed. receding (RH) strategy suggested to reduce impact prediction error enable real-time implementation EMS benefits from actual data on day. At each hour, LSTM predicts h, then charging or discharging command only first hour applied real-time. Real are used update input, process repeated. Simulation results show proposed outperforms offline strategy, reducing operating by 3.3%.