Multi-scale modeling of proton exchange membrane fuel cell by coupling finite volume method and lattice Boltzmann method

Research output: Contribution to journalArticlepeer-review

115 Scopus citations

Abstract

A multi-scale modeling framework combining finite volume method (FVM) and lattice Boltzmann method (LBM) previously developed by our group is used to predict electrochemical transport reaction in proton exchange membrane fuel cell (PEMFC) cathode with a parallel gas channel (GC), a gas diffusion layer (GDL) with porous structures and a catalyst layer (CL) with idealized microstructures. In this framework, the PEMFC cathode is divided into two sub-domains, one is GC and the other contains GDL and CL. The FVM is used to simulate transport phenomena in the GC sub-domain, while the LBM is employed for pore-scale transport phenomena in the GDL and CL as well as proton conduction in the CL in the other sub-domain. Two reconstruction operators are adopted to transfer macro density, velocities and concentration in the FVM to density distribution functions and concentration distribution functions in the LBM at the interface between the two sub-domains. Simulation results show that the coupled (hybrid) simulation strategy developed is able to predict transport phenomena in the GC and to capture the pore-scale transport processes in porous GDL and CL. In addition, some techniques to save the computational resources and to improve the efficiency of the coupled (hybrid) simulation strategy are discussed.

Original languageEnglish
Pages (from-to)268-283
Number of pages16
JournalInternational Journal of Heat and Mass Transfer
Volume63
DOIs
StatePublished - 2013

Keywords

  • Boltzmann method
  • Coupled (hybrid) simulation strategy
  • Finite volume method
  • Lattice
  • Multi-scale
  • Porous media
  • Proton exchange membrane fuel cell

Fingerprint

Dive into the research topics of 'Multi-scale modeling of proton exchange membrane fuel cell by coupling finite volume method and lattice Boltzmann method'. Together they form a unique fingerprint.

Cite this