Maximising self-consumption in energy communities: a comparative study of demand flexibility and dynamic sharing strategies

Energy Communities (ECs) have emerged as a vehicle for actively engaging citizens in the energy transition, enabling them to generate, consume, store and share renewable energy. Maximising collective self-consumption is a key objective of ECs, as it allows members to better capture the benefits of shared energy generation, and several approaches have been developed toward this goal. However, most of existing works aim to maximise ECs’ self-consumption by applying demand side management strategies or increasing the benefits with dynamic energy sharing, but not combining them. This study is one of the first to propose a data-driven framework that leverages machine learning, optimisation, battery management and dynamic energy sharing strategies to increase the self-consumption of ECs. The impact of the optimisation and dynamic energy sharing coefficients on ECs’ self-consumption and electricity costs was examined, together with their combination. The application of the proposed framework to an established Renewable Energy Community (REC) in Spain shows that dynamic energy sharing yields better results compared to load optimisation, and their combination shows more promising results. Permitting more frequent updates of dynamic energy-sharing coefficients can improve the allocation of locally generated energy and support higher member-level self-consumption, particularly in communities with heterogeneous and seasonal demand profiles.