Modeling of PEM fuel cell systems including controls and reforming effects for hybrid automotive applications

Due to the nature of fuel cell reactions, fuel cells have the potential of being more fuel efficient while generating fewer harmful emissions than conventional automotive power systems. Additionally, by hybridizing a fuel cell system with a battery, opportunities may exist for significantly improving overall performance.; This study develops models for a stand-alone Proton Exchange Membrane (PEM) fuel cell stack, a direct-hydrogen fuel cell system including auxiliaries, and a methanol reforming fuel cell system for integration into a vehicle performance simulator. Exergetic efficiencies associated with the three models are examined and sources of inefficiency are identified. Fuel cell stack efficiency is highest when operating at low current density. Air compressor power consumption and losses associated with reformer operation significantly lower the overall system efficiency and highlight the importance of low-level control of components within the system.; By incorporating the models developed in this study into the vehicle performance simulator, alternative fuel cell vehicle configurations can be explored using various driving cycles, component sizing, and control strategies to determine effects on overall vehicle performance and fuel economy. For a typical sport utility vehicle operating over the Federal Urban Driving Schedule and Federal Highway Driving Schedule driving cycles, the simulator is used to examine fuel economy in four cases: direct-hydrogen fuel cell vehicle, methanol reforming fuel cell vehicle, direct-hydrogen hybrid (fuel cell system/battery) vehicle, and methanol reforming hybrid vehicle. Results indicate the direct-hydrogen hybrid vehicle shows the strongest potential for high fuel economy.; Additionally, for the direct-hydrogen hybrid vehicle, simple supervisory control strategies for the fuel cell system and battery are used to examine component sizing and operational limits. Dominance filtering is employed to identify component sizing and operational limits that provide the potential for highest fuel economy. Results of this analysis can be used as a point of departure to develop more advanced supervisory and component-level control strategies. Using appropriate supervisory and component-level control strategies to improve total system performance is key to realizing the benefits of fuel cell system integration for automotive applications.