Abstract

One of the factors that has hindered the commercialization of Polymer Electrolyte Membrane Fuel Cells (PEMFCs) has been the high cost per kW from the high cost of platinum catalyst. Hence it is important to optimize the use of the platinum catalyst. In many previous PEMFC models, the catalyst distribution in Cathode Catalyst Layer (CCL) was assumed to be uniform, and there are very few models in which the variation of catalyst loading across the CCL is taken into account. This paper aims to enhance PEMFC power density by optimally distributing the catalyst used in CCL employing a computational fluid dynamic (CFD) simulation in conjunction with agglomerate model of the CCL. First, a numerical approach is presented to evaluate the effect of variant catalyst loading in the CCL of a PEMFC. A two-dimensional, steady state, isothermal implicit model for PEMFC has been developed. A spherical flooded agglomerate model is considered for CCL in which transport of both chemical species and charges are also taken into account. The numerical domain includes channel, land, Gas Diffusion Layer (GDL), and CCL. The developed model is used to determine the optimum catalyst loading distribution along CCL using a novel algorithm based on combining the numerical model and Nelder-Mead Simplex optimization method. In this novel algorithm, the optimum catalyst distribution function is assumed as different order polynomial functions with unknown coefficients. Unknown coefficients are calculated by considering the numerical model as a set of constraints for the optimization method. Nelder-Mead optimization method which is an unconstraint minimization heuristic method is employed to find the unknown coefficients. It is seen that in the optimum case, the maximum PEMFC density is increased about 7 percent. This improvement is a huge leap toward the commercialization of PEMFC.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call