A two-stage distributionally robust optimization model for optimizing water-hydrogen complementary operation under multiple uncertainties

Under the pressures of fossil energy depletion and the “Carbon peak and neutrality” target, the development of clean energies such as hydropower and hydrogen has received widespread attention. The integration of hydropower, power-to-hydrogen/hydrogen-to-power and energy storage (forming a water-hydrogen complementary system) can improve the water resource utilization and obtain additional benefits by selling hydrogen etc. However, random fluctuations in market electricity prices, water flow and electric load seriously interfere with the complementarity of water and hydrogen, hindering the acquisition of the above benefits. To this end, this paper proposes a two-stage distributionally robust optimization model to solve the operation scheduling issue of the water-hydrogen complementary system under multiple uncertainties. Specifically, the uncertain distribution of market electricity prices, water flow and electric load forecasting errors are depicted with a moment-based ambiguity set. In the first stage, electricity and hydrogen are coordinately scheduled based on the forecast information to maximize the operation profit of the complementary system. In the second stage, the operations of flexibility resources are linearly adjusted from the first stage to resist the interference of the “worst-case” distribution in the ambiguity set. Finally, the model is equivalently reformulated into a mixed integer linear programming for solution feasibility. Simulation verifies that: 1) the model is conducive to the complementary system operation, such as 43.7% profit improvement (compared with scheduling ignoring uncertainties), 97.70% water utilization and effectively resisting uncertainties; 2) the model keeps low conservativeness and computational complexity compared with the stochastic and robust optimizations.