Kunning Tang (1), Quentin Meyer (2), Robin White (3), Ryan T. Armstrong (1), Peyman Mostaghimi (1), Ying Da Wang (1), Shiyang Liu (2), Chuan Zhao (2), Klaus Regenauer-Lieb (4), Patrick Kin Man Tung (1) (5)
Computers and Chemical Engineering. Volume 161 (May 2022). DOI: https://doi.org/10.1016/j.compchemeng.2022.107768

Keywords

Proton exchange membrane fuel cell, X-ray micro-computed tomography, image segmentation, deep learning, convolutional neural network, gas diffusion layer

Abstract

This study demonstrates the benefit of convolutional neural networks to accurately classify the different materials of proton exchange membrane fuel cells using X-ray micro-computed tomography. Nineteen 2D orthoslices from a 3D tomography dataset were segmented with high quality and used to train a novel U-ResNet convolutional neural network (CNN) to segment the complete volume. The results were compared with a 3D manual segmentation performed under time constraints. The CNN segmented all phases with equal or greater accuracy in comparison to the manual segmentation. In particular, the CNN excelled in separating the carbon fibres and binder phase in the gas diffusion layer, which is usually completely avoided due to difficulty. Further, permeability calculations were performed on the binder void space for both segmentations, with the CNN displaying realistic results. Therefore, CNNs have been shown to be a viable and valuable method in segmenting such fuel cells with increased efficiency and accuracy.

Author Affiliation

(1) School of Minerals and Energy Resources Engineering, University of New South Wales, Sydney, Australia
(2) School of Chemistry, University of New South Wales, Sydney, Australia
(3) Carl Zeiss X-ray Microscopy, Pleasanton, USA
(4) WA School of Mines: Minerals, Energy and Chemical Engineering, Curtin University, Perth, Australia
(5) Mark Wainwright Analytical Centre, University of New South Wales, Sydney, Australia