With the large-scale adoption of electric vehicles and hydrogen fuel cell vehicles, electricity-hydrogen coupled distribution networks are playing an increasingly critical role in multi-energy coordinated scheduling. To address the limitations of existing systems in terms of economic performance, operational reliability, and responsiveness to load uncertainty, this paper proposes a multi-time scale bi-level optimization scheduling model based on a virtual power plant, which integrates and coordinates distributed energy resources including photovoltaic generation, energy storage systems, electrolyzers, hydrogen storage tanks, and fuel cells. The upper-level model aims to minimize day-ahead operational costs through a centralized optimization approach for global scheduling. The lower-level model incorporates a rolling-window optimization mechanism to dynamically respond to real-time fluctuations in the loads of electric vehicles and hydrogen fuel cell vehicles. Simulation results based on the IEEE 33-bus distribution system demonstrate that the proposed model significantly reduces operational costs, enhances scheduling flexibility, and improves the utilization efficiency of renewable energy.