Simulation of energy management system using model predictive control in AC/DC microgrid

This research seeks to enhance energy management systems (EMS) within a microgrid by focusing on the importance of accurate renewable energy prediction and its strong correlation with load curtailment. Analyzing the precision of disturbance predictions, reveals that predicting one hour in advance is more effective than immediate predictions or those made several hours beforehand. Furthermore, the study investigates scheduling load curtailment to manage peak power from renewable energy sources by comparing two distinct strategies: Case 1, which implements curtailments in both morning and afternoon, and Case 2, which focuses solely on midday curtailment. The findings indicate that Case 1 effectively aligns load management with the peak output of photovoltaic (PV) energy, thereby reducing reliance on grid power and enhancing energy efficiency. In contrast, Case 2's focus on midday curtailment results in increased energy purchases from the grid, missing the chance to leverage abundant solar energy. A key finding of this research shows that applying Case 1 for curtailment along with accurate forecasting, improves battery coordination and alleviates stress on the supercapacitor, leading to energy purchases from the grid being reduced. This interdependent relationship between precise forecasting and effective load management not only enhances the efficiency of the hybrid energy system (battery and supercapacitor), but also relies more on renewable energy sources and storage systems, thereby lowering overall energy costs, leading to a more reliable and effective energy management system.