Hydrogen PEMFC stack performance analysis : a data-driven approach

For low power fuel cells, it is paramount that management of reactants, water and heat, be realized in a passive fashion in order to minimize parasitic losses. Effective fuel, oxygen supply and water management for reliable performance are also greatly affected by cell geometry and materials. Fuel cells are complex systems to optimize on a mere experimental basis. As an aid to this goal, data-driven analysis techniques, requiring no mathematical model to be fixed a priori, are gaining a reputation in other fields of work, where a phenomenological modeling approach might be intractable. This work presents a characterization study of a 12W PEMFC series stack by means of a new data-driven technique, M-NMF. The stack was developed for low temperature operation, uses own designed flow field plates, integrated in a series configuration, and is operated for 12 combinations of hydrogen/air flowrate ratios, generating as many polarization curves. M-NMF is applied, in combination with an alternating least squares algorithm, to the analysis of the overvoltage data matrix derived from the original experimental polarization data. From this analysis, it is possible to group and differentiate data according to similar overvoltage patterns and gain insight into their relative contribution to fuel cell performance immunization