With the gradual increase in the demand for hydrogen, research on the optimal operation of a renewable energy hydrogen production system has become a hot topic. However, the uncertainty of wind and solar renewable energy output makes the optimal operation of these systems complex. To address the impact of wind and solar output uncertainty on system optimization, a multi-time scale optimization strategy for a renewable energy alkaline-proton exchange membrane combined hydrogen production system is proposed, aiming to minimize the system’s daily operating cost. The strategy implements a multi-time scale operational control based on the differences in dynamic response capabilities of different equipment in the system and their specific functions. In the day-ahead stage, a day-ahead distributed robust optimization model of the renewable energy alkaline-proton exchange membrane combined hydrogen production system is constructed. In the intra-day stage, a rolling optimization strategy is adopted. In the real-time stage, supercapacitors are used to smooth out the severe fluctuations in wind and solar power, and then the real-time optimal operation strategy is proposed to minimize the output adjustment variation of each device. The case study results show that the operational economy of the system can be improved by the multi-time scale optimization operation strategy from day-ahead to intra-day and to real-time.