A computational fluid dynamic framework for modeling and simulation of proton exchange membrane fuel cell

Abstract

This thesis describes the development and application of a framework for model and analysis of proton exchange membrane fuel cells (PEMFCs) using computational fluid dynamics (CFD). The developed framework addresses the formulation, solution, and analysis of the PEMFCs systems in a systematic manner. This PEMFCs modelling framework helps to generate problem-system specific models describing a step by step proton exchange membrane (PEM) fuel cell model. Accordingly, the problem-system specific model generation procedure consists of three main steps. In the first step the problems and scope of the study are defined. The PEM fuel cell modelling procedure is done in the next step 2. This second step contains three sub-sections which are geometry definition, model definition and numerical solution and validation. In the step 3, the developed model is validated using available experiment/industrial data. A general, three-dimensional, non-isothermal, multi-phase numerical model has been developed to simulate and examine the fluid flow, heat and mass transfer, species transport, electrochemical reaction, and current density distribution of a PEMFC. The validation results of the PEM fuel cell model developed by using this framework has been successfully done. In addition, applications of the validated model with respect to advanced grid analysis, anisotropic properties investigation, and PEMFCs electrochemistry parameter have been successfully implemented. With respect to grid analysis, the results have shown that grid independence analysis using polarization curve is not accurate, where the concentration of fuel cell reactants and product showed more sensitivity for checking the grid independency. In terms of anisotropic properties investigation, the results have shown that increasing the value of anisotropy in thermal conductivity mitigates the gradient of liquid water between the area underneath the ribs and channels in PEMFCs. With respect to PEMFCs electrochemistry parameter, it has been shown that the new derived of the Kazemi-Jahandideh (K-J) approximation is able to reduce the numerical calculation to find electrochemistry parameters with a higher accuracy.