Fast electric vehicle (EV) charging stations often face significant constraints related to limited grid connection capacity. To mitigate these limitations, many stations integrate on-site renewable energy generation and battery energy storage systems (BESS) to supplement grid power. Additionally, BESS installations can provide ancillary services such as peak shaving and energy arbitrage, improving the overall economic viability of the system. However, most implementations rely on monotype BESS solutions, which can lead to unnecessary oversizing or limitation of the system’s ability to fully exploit ancillary service opportunities.

Hybrid battery energy storage systems (HBESS), combining high‑energy (HE) and high‑power (HP) battery technologies, have demonstrated economic advantages in applications such as marine transportation electrification. HP batteries have higher power density and better cycling performance, but lower energy density and higher cost compared with HE batteries. So, HE batteries are normally used for applications that need continuous power over long periods, while HP batteries are preferred for applications requiring short term high-power demand.

This paper investigates the use of an HBESS in a stationary application to support both peak shaving and energy arbitrage, which are two services with opposing energy and power requirements. The study is based on a real fast‑charging station located in Norway and includes a detailed description of its architecture, comprising a photovoltaic system, grid connection, and realistic load profile, as well as its operational framework via power and energy management systems.

Two storage configurations are evaluated: (i) an HBESS combining second‑life NMC modules (high energy) with LTO modules (high power), and (ii) a monotype BESS consisting solely of second‑life NMC modules. In the specific case study, the monotype BESS presents an acquisition cost approximately 8% lower than the HBESS. However, the analysis examines whether the hybrid approach can justify its additional cost through improved performance in ancillary service provision, potentially reducing electricity costs and shortening payback time. Both peak shaving and energy arbitrage are simulated on a Matlab/Simulink platform under realistic operating conditions, and the resulting economic performance is evaluated for each configuration. The Norwegian context adds particular relevance to the study, as electricity bills comprise both spot market prices and peak demand charges. Additionally, battery system losses for both configurations may be assessed to provide a more comprehensive and realistic comparison. The results aim to offer useful insights and guidance for future EV charging‑station planning and HBESS sizing.

This research is part of the HiHelios project funded by European Union’s Horizon Europe Research and Innovation Programme under Grant Agreement № 101137626.