Abstract: A method for determining if more hydrogen has been added to a fuel cell system than a predetermined threshold amount to detect leaks in an anode subsystem or a cathode subsystem of a fuel cell system. The method includes determining a quantity of hydrogen added to the fuel cell system for a given period of time during a predetermined operating condition of the fuel cell system and determining whether the quantity of hydrogen added is more than the predetermined threshold amount. The method also includes adapting an anode subsystem reactant gas concentration model if the quantity of hydrogen added to the fuel cell system is more than the predetermined threshold amount to provide precise control of pressure in the anode subsystem and the cathode subsystem of the fuel cell system.

Abstract: A system and method for determining a possible failure of membranes in the fuel cells for a fuel cell stack. The method includes monitoring the stack current density and the minimum cell voltage of the fuel cells in the stack. If both the minimum cell voltage and the stack current density are below predetermined values, then the method multiplies scaling factors of the minimum cell voltage and the stack current density to provide a membrane failure factor. If the membrane failure factor is greater than a threshold, then an indication is given of a possible membrane failure.

Abstract: A system and method for detecting and predicting low performing cells in a fuel cell stack. When the fuel cell stack is running and certain data validity criteria have been met, an algorithm collects the data, such as stack current density, average cell voltage and minimum cell voltage. This information is used to estimate predetermined parameters that define the stack polarization curve. The system defines a predetermined minimum current density that is used to identify a low performing cell. The system then calculates an average cell voltage and a minimum cell voltage at the minimum current density set-point, and calculates a cell voltage difference between the two. If the cell voltage difference is greater than a predetermined low voltage threshold and the minimum cell voltage is less than a predetermined high voltage threshold, the algorithm sets a flag identifying a potential for a low performing cell.

Abstract: A system and method for detecting and predicting low performing cells in a fuel cell stack. When the fuel cell stack is running and certain data validity criteria have been met, an algorithm collects the data, such as stack current density, average cell voltage and minimum cell voltage. This information is used to estimate predetermined parameters that define the stack polarization curve. The system defines a predetermined minimum current density that is used to identify a low performing cell. The system then calculates an average cell voltage and a minimum cell voltage at the minimum current density set-point, and calculates a cell voltage difference between the two. If the cell voltage difference is greater than a predetermined low voltage threshold and the minimum cell voltage is less than a predetermined high voltage threshold, the algorithm sets a flag identifying a potential for a low performing cell.