Data-driven modelling method and application based on energy multi-layer network structure of energy hub

Energy hub (EH) is a complex system integrating multiple energy sources, playing a crucial role in the Energy Internet (EI). Conventional modelling methods often treat energy sources separately, failing to capture the full dynamic interactions and operational complexities. This paper proposes a novel multi-layer network structure (MNS) for modelling EH, which synchronizes energy flows and optimizes control parameters for energy consumption reduction. The method integrates equipment performance curves into the network, providing a dynamic model that is computationally feasible for real-world applications. In project implementation, the dynamic control method is applied hourly between 8:00 and 17:00, with specific case studies for winter and summer days. The results show that the optimized control strategy can achieve up to 70% energy cost savings in summer and 20% savings in winter while maintaining equipment efficiency above 65% in summer and 60% in winter. The energy consumption costs before and after optimization are significantly reduced, as demonstrated by the comparative analysis. The proposed approach not only enhances system performance but also provides practical implications for optimizing energy hubs in diverse operational conditions.