Two-Stage Stochastic Scheduling of Grid-connected Wind-Hydrogen-Storage Systems with Electrolyser Clusters under Wind Power Uncertainties

Hydrogen production from wind farms suffers from the intermittent and uncertain nature of wind generation, which heightens the importance of short-term energy storage units and grid connection in stabilising the power supply to electrolysers and enhancing overall economics. To facilitate multi-unit coordination of a grid-connected wind-hydrogen-storage (WHS) system, this paper proposes a two-stage day-ahead stochastic scheduling model for a transmission-level WHS system with electrolyser clusters subject to uncertain wind power forecasts. First, a total hydrogen production efficiency curve is proposed for electrolyser clusters by combining uniform operation and hot-standby modes of clusters, which enables fast estimations on the optimal number and loads of operating electrolysers from the total cluster load. Then, a two- stage stochastic optimisation model is developed for day-ahead system schedules. In specific, the optimal electrolyser cluster loads are regarded as the first-stage variables that are consistent across random scenarios, while the second-stage stochastic variables including energy storage outputs and energy exchange with grid in each scenario are optimised separately to balance power flows across the system. The stochastic scheduling model proposed here is tested based on a grid-connected WHS system comprising a 76MW wind farm with a 75MW point of common coupling, five 3MW electrolysers and a 15MW/30MWh battery energy storage. Simulation results indicate that the stochastic scheduling achieves stable and efficient operation of electrolyser clusters under all the random scenarios while maximising the overall benefit from green hydrogen production and electricity market trades.