Hydrogen fuel cell hybrid ships (HFCEVs) face the adaptability issue of energy management strategies (EMS) under complex operating conditions. This study proposes a hierarchical collaborative optimization strategy (HCOS) for the dynamic response requirements and equipment protection requirements of ship power systems. This method organically integrates the robustness framework of finite state machines with the optimization kernel of minimizing equivalent hydrogen consumption, and achieves comprehensive improvement in system performance through the collaborative cooperation of the operating condition recognition layer and the real-time optimization layer. The simulation verification based on the operating data of the “Three Gorges Hydrogen Boat One” hydrogen fuel cell demonstration ship shows that compared with traditional control methods, the proposed strategy exhibits significant advantages in maintaining power system stability, optimizing the working range of energy storage components, and balancing energy efficiency and equipment life, providing a new solution for energy management of ship hybrid power systems.