Premium
Steady‐state Semi‐empirical Model of a Single Proton Exchange Membrane Fuel Cell (PEMFC) at Varying Operating Conditions
Author(s) -
Labach I.,
Rallières O.,
Turpin C.
Publication year - 2017
Publication title -
fuel cells
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.485
H-Index - 69
eISSN - 1615-6854
pISSN - 1615-6846
DOI - 10.1002/fuce.201600074
Subject(s) - proton exchange membrane fuel cell , ohmic contact , polarization (electrochemistry) , electrolyte , dielectric spectroscopy , relative humidity , voltage , approximation error , electrical impedance , materials science , thermodynamics , analytical chemistry (journal) , mechanics , chemistry , electrochemistry , mathematics , statistics , membrane , electrode , composite material , electrical engineering , physics , layer (electronics) , chromatography , engineering , biochemistry
The purpose of this paper is to develop a physically‐based macroscopic steady‐state model predicting the performances of a polymer electrolyte fuel cell (PEMFC) at various operating conditions. The model is mostly mechanistic: the equation for the V–I curves holds 7 parameters, 6 of which have a physical significance. This work aims to quantify the main voltage losses (namely activation, ohmic and mass‐transport) thanks to their respective sensitivity to the operating variables. The model parameters are here obtained by fitting a set of 16 experimental V–I curves. First a full factorial plan of experiments is carried out on a single cell to characterize the influence of four factors: temperature, pressure, air relative humidity and stoichiometry. Polarization curves are measured by current steps; at the end of each step, a high‐frequency electrochemical impedance spectroscopy (EIS) is achieved in order to extract the ohmic resistance. Cyclic voltammetries are also performed to measure hydrogen crossover. Then the whole experimental database constituted is used to parameterize the theoretical equation of the V–I curve. The maximal relative error between data and fit is 8.2% and the average relative error is 1%. The prediction capability of the single cell model is finally evaluated on three polarization curves.