Peak shaving with hydrogen energy storage: From stochastic control to experiments on a 4 MWh facility

The formation of power peaks caused by the stochastic nature of the electric vehicles (EVs) charging process is raising concerns related to the stability of the power grid. In this work, we consider an EV charging station equipped with a hydrogen-based energy storage system (HESS) and on-site renewable power generation, and we offer an experimental demonstration of its potential in reducing the power peak of the EV charging station, despite uncertainty in the demand. Our contributions are as follows: (1) we derive a complete system description of a real 4 MWh HESS, and (2) we develop a stochastic model-based receding horizon controller that jointly schedules the charging process and the HESS operation to minimize the power peak. The proposed approach is validated both in simulation and via experiments that demonstrate (i) excellent performance, with an average daily peak reduction of 49.2% with respect to the benchmark, and (ii) scalable run times, making it amenable to real-time operations even in the large EV penetration regime. • The system identification of a 4 MWh hydrogen energy storage system is provided. • A stochastic approach coordinates the schedule of the EVs and of the storage system. • EV probability distributions are used to model the uncertain demand of EVs. • The receding horizon implementation uses online information to refine load forecast. • The proposed approach outperforms other optimization-based methods in simulation. • The experimental validation confirms the performance and scalability of the approach.