A fuel cell system may include a valve, a hydrogen source, an anode loop, a blower, a pressure sensor, and an anode controller. The valve is communicably coupled to a hydrogen source and configured to supply hydrogen to an anode loop. The blower is arranged to supply recycled hydrogen to the anode loop. The pressure sensor is configured to sense an anode inlet pressure. The anode controller is configured to determine a target anode inlet pressure, according to a current demand. The anode controller is configured to execute a feedback control loop, using the anode inlet pressure, to control the blower and the valve, to supply hydrogen to the anode loop.

A system and method for controlling power for a fuel cell are disclosed. The system includes: a fuel cell; a load device electrically connected to the fuel cell; a stack controller configured to set a stack limit current on the basis of a current output current of the fuel cell, the stack limit current configured to limit an output current of the fuel cell on the basis of an output voltage of the fuel cell; and a load controller configured to set a consumption limit current on the basis of the set stack limit current, the consumption limit current configured to limit a consumption current of the load device, the load controller being configured to control the consumption current of the load device to a value equal to or lower than the set consumption limit current.

This application provides a method for controlling an anode purge valve of a fuel cell, a device, a medium, and a product, and relates to the field of fuel cell control technologies. The method includes: acquiring a system state of a fuel cell system and a corresponding reward value; inputting the system state of the fuel cell system and the corresponding reward value into a trained prediction model, to obtain a control action; the trained prediction model is a neural network model based on a reinforcement learning algorithm; and controlling an anode purge valve of the fuel cell system based on the control action. In this application, the reinforcement learning technology is introduced into the control of the anode purge valve of the fuel cell.

A fuel-cell-applied system includes a control device configured to control power generation of the fuel cell, a power conversion device configured to convert power generated by the fuel cell into power for input to a load, an isolation device configured to irreversibly disconnect electrical connection between the fuel cell and the power conversion device, a first detector configured to detect an output power state form the fuel cell, and a second detector configured to detect an input power state to the power conversion device, wherein the control device detects a disconnected state of the electrical connection caused by the isolation device on a basis of the output power state and the input power state respectively detected by the first detector and the second detector.

A method for maintaining a target electrochemical impedance (ECI) for a fuel cell, which corresponds to a target hydration state for the fuel cell. The method includes determining a target electrochemical impedance (ECI) for the fuel cell based on current operating conditions. The method further includes determining actual ECI for the fuel cell and comparing actual ECI to the target ECI. The method further includes adjusting a cathode flow to the fuel cell based on a deviation of the actual ECI from the target ECI.