With the decentralization of renewable energy sources, multi-energy systems (MES) integrating the electrical sector with mobility or thermal sectors are increasingly preferred for their sustainability and flexibility. However, optimizing such complex systems is challenging, requiring a balance of profitability, asset aging reduction, and sustainability.
This study examines a real-life MES setup for residential buildings housing 150 residents, equipped with 220 kWp of photovoltaics and 9 kW of wind power. The system includes a hybrid energy storage system (HESS) with a Vanadium redox flow battery (VFB) and a Lithium-Ion battery (LIB) for the electrical sector. The VFB is also fitted with an innovative thermal coupling module (TCM), enabling dual use as heat storage.
The research applies a mixed-integer linear programming (MILP) method, using linear models to capture variable losses and auxiliary loads of the HESS. The optimization aims to enhance profitability, reduce operational losses, and mitigate asset aging. Tests show that MILP reduces losses by up to 50% over a day’s operation, primarily by strategically operating the HESS to avoid the high auxiliary load of the VFB. Additionally, the MILP schedules LIB charging to avoid prolonged high state-of-charge (SOC) periods, reducing calendric aging by up to 58%. These objectives are achieved while maintaining lower operational costs compared to rule-based operation.
A key focus is determining the optimal forecast horizon for effective MES dispatch. The system was tested under scenarios of increasing and decreasing RES trends, with 24h, 36h, and 48h forecast horizons. Results indicate that longer forecast horizons enhance sustainability across electrical and thermal sectors, while shorter horizons are more effective for immediate loss reduction.