Battery management for energy communities—Economic evaluation of an artificial intelligence-led system

Storage technologies such as the battery energy storage system are expected to play a critical role in micro-grids. However, current simulation tools underestimate their operating costs, which jeopardises their efficient use and deployment. This paper proposes a strategy based on artificial intelligence and time series prediction for the planning and real-time management of a battery energy storage system in a micro-grid acting as a virtual power plant. An economic analysis was performed, and the findings suggested that the state of charge should be managed to avoid economic losses due to cycle ageing. In addition, the battery should be sized correctly to ensure its economic viability, which indicates that it should be optimised according to the load profile of the micro-grid. The proposed management strategy was demonstrated to avoid the economic losses observed with a non-managed storage system. In addition, ensuring that the micro-grid did not deviate from the load profile agreed upon with the transmission system operator was found to increase economic returns, reduce battery degradation and increase self-consumption with the co-benefit of reducing micro-grid fluctuations.