A MODEL PREDICTING TRANSIENT RESPONSES OF PROTON-EXCHANGE MEMBRANE FUEL-CELLS

There has been a recent interest in modelling the transient behaviour of proton exchange membrane (PEM) fuel cells. In the past, there have been several electrochemical models which predicted the steady-state b ehaviour of fuel cells by estimating the equilibrium cell voltage for a particular set of operating conditions. These operating conditions i ncluded reactant gas concentrations and pressures, and operating curre nt. Steady-state behaviour is very common and in some cases is conside red as the normal operating standard. Unsteady-state behaviour, howeve r, is becoming more of an issue, especially for the transportation app lications of fuel cells where the operating conditions will normally c hange with time. For example, system start-up, system shut-down, and l arge changes in the power level may be accompanied by changes in the s tack temperature, as well as changes in the reactant gas concentration s at the electrode surface. Therefore, both mass and heat transfer tra nsient features must be incorporated into an electrochemical model to form an overall model predicting transient responses by the stack. A t hermal model for a Ballard Mark V 35-cell 5 kW PEM fuel cell stack has been developed by performing mass and energy balances on the stack. T he thermal characterization of the stack included determining the chan ges in the sensible heat of the anode, cathode, and water circulation streams, the theoretical energy release due to the reaction, the elect rical energy produced by the fuel cell, and the heat loss from the sur face of the stack. This thermal model was coupled to a previously-deve loped electrochemical model linking the power produced by the stack an d the stack temperature to the amount and method of heat removal from the stack. This electrochemical model calculates the power output of a PEM fuel cell stack through the prediction of the cell voltage as a c omplex function of operating current, stack temperature, hydrogen and oxygen gas flowrates and partial pressures. Initially, a steady-state overall dynamic model (electrochemical model coupled with the thermal model) was developed. This was then transformed into a transient model which predicts fuel cell performance in terms of cell voltage output and heat losses as a function of time due to various changes imposed o n the system.