Research on multi-time scale optimization strategy for PV-electrolyzers hydrogen production system based on MPC

Hydrogen production from new energy power generation is an effective measure to achieve energy transformation, low carbon and clean hydrogen production. To reduce the cost of hydrogen (COH) production, improve the utilization rate of photovoltaic (PV) power generation and deal with PV uncertainty, this paper proposes a multi-time scale optimization strategy for PV-electrolyzers hydrogen production system based on model predictive control (MPC). Firstly, for the hydrogen production system, a rotation operation strategy is proposed based on the characteristics of alkaline electrolyzers (AELs), and an AEL management system(AEMS) is configured to manage the operation and health status of the electrolyzers. The multi-electrolyzer rotation operation strategy is integrated into the optimization strategy. Secondly, with the optimization objective of minimizing the cost, the electricity purchase cost function here is newly defined to guide the system to produce more hydrogen during low electricity price periods. Finally, a two-layer optimization scheduling framework based on MPC is proposed to improve system’s economic efficiency while correcting deviations of day-ahead scheduling. The simulation results show that the power range of the electrolyzer is extended and the improved rotation operation strategy enhances the balance between electrolyzers. The MPC-based optimization strategy effectively coordinates the operation of electric and hydrogen hybrid energy storage. In addition, the proposed strategy increases the average hydrogen production of the hydrogen production system by 20% and reduces the cost of hydrogen (COH) by 3%. • A multi-time-scale dual-layer optimization framework based on MPC is constructed. • A rotation operation control strategy is formulated for multi-electrolyzer array. • The electricity purchase cost function in the objective function is redefined. • The optimization of hydrogen production strategies of “season-day-hour” is realized.