To address the problem of grid-connected fluctuation of wind and photovoltaic power, a capacity optimization allocation strategy for super capacitor-hydrogen hybrid energy storage system based on scenario generation is proposed. The strategy generates a large number of wind and photovoltaic power scenarios by Frank-Copula based on actual data of wind and photovoltaic power, and then obtains typical scenarios by K-means. The scenarios are decomposed by the improved complementary ensemble empirical mode decomposition with adaptive noise method and perform reconstruction, and then, the fluctuations that need to be smoothed out in each typical scenario are obtained. Considering the operating characteristics and constraints of the super capacitor and the proton exchange membrane hydrogen storage system, the capacity allocation model of the super capacitor-hydrogen hybrid energy storage is established in MATLAB. Finally, the Gurobi solver is used to find the best result of the capacity allocation model and to verify the advantage of the proposed strategy. The solution result shows that the model proposed in this paper improves the economy of the energy storage system by 7.09% and reduces the amount of under compensation by 36.848 6 MWh.