Investigation on Exhaust Energy Recovery System Using Radial Turbine in High-Power Proton Exchange Membrane Fuel Cells

Abstract With the increased power density and capacity of the proton exchange membrane fuel cell (PEMFC) stack, the intake air pressure level is getting higher and can reach up to 3.5bar in some cases, leading to a higher exergy level of the exhaust gas at the fuel cell cathode. To improve the system efficiency of the PEMFC, the energy of the fuel cell exhaust gas at the cathode side can be recovered with a turbine and partially drive the compressor to save part of the power needed by the electric motor. To evaluate the potential capability of the PEMFC exhaust gas recovery, detailed insights regarding the energy recovering capability of the turbine from exhaust gas energy as a result of gas supply pressure, air stoichiometric ratio, current density and operating temperature are presented in the current study. A systematic model of the fuel cell system is established and validated. The model includes the air supply system, electric motor driven turbocharger, fuel cell stack components and necessary pressure drop as well as the gas dynamic model inside the stack. The model can be used to simulate the recovery process of the exhaust gas energy at arbitrary fuel cell operation loadings. It can be used to predict the system energy recovery efficiency once the loading condition of the fuel cell is known. Results indicate that the energy recovery system performance is determined by the fuel cell operating conditions. Results also show that the gas inlet state, current density, electrochemical reaction process and pressure loss play a key role in the recovery efficiency of the turbine and the ratio of recovery power to fuel cell output power. The recovery efficiency is higher at a larger pressure ratio and current. To verify the accuracy of the system model, a high-power fuel cell is used for validation, the rated power of the proton exchange membrane fuel cell stack is calibrated through the electrochemical model from the energy recovery system and good agreement is achieved between the simulation and experiment.