Impedance-Based, Multi-Physical DC-Performance-Model for a PEMFC Stack

Abstract

Performance prediction for large-sized polymer electrolyte membrane fuel cell (PEMFC) stacks necessitates consideration of spatially deviating operating conditions on the nonlinear electrochemical behaviour. This interaction between operating conditions and electrochemistry is best described by complex CFD-models. But high computing power excludes stack and system modelling in real time applications.We address this challenge by a multi-physical stack model, which couples (i) the non-linear electrochemistry within the cell, (ii) the fluid pressure drop along the gas channels and (iii) the thermal behaviour within the stack. The spatial resolution focusses on the most relevant directions and thus limits the computational effort. Simulation runtime is further reduced by modelling the electrochemical behaviour by a physico-chemically meaningful equivalent circuit model (ECM) [1], which relies on a data set of electrochemical impedance spectroscopy (EIS) measurements performed on incremental cells [2]. Individual impedance contributions are identified by the distribution of relaxation times (DRT). ECM model parameters are subsequently quantified by a CNLS-fitting procedure [3,4] and transferred to a nonlinear, zero-dimensional DC-performance-model. The magnitude of the modelled pressure depends on the gas flow within the channel and considers the change of gas composition, whereas the local gradients in current density cause gradients in released heat within the cell itself. This effect along the gas flow, the convection between fluids and solid parts of the stack (bipolar plates and cell) and the internal heat conduction between the solid control volumes are considered in the modelled thermal behaviour.In this contribution a multi-physical stack model considering gradients in temperature, pressure and gas composition is presented. The interdisciplinary interactions and dependencies within the different physical domains, especially the influence of pressure and temperature on the non-linear electrochemical model, are shown. A concise validation based on measured data and conclusions for the possibilities of further applications as a system model will be discussed.[1] D. Klotz et. al., ECS Transactions 25, pp. 1331-1340 (2009)[2] M. Heinzmann et al., J. Power Sources 402, pp. 24-33 (2018).[3] H. Schichlein et al., J. Appl. Electrochem. 32, pp. 875-882 (2002).[4] S. Dierickx et al. Electrochimica Acta 355, 136764 (2020)