Artificial-intelligence-guided design of ordered gas diffusion layers for high-performing fuel cells via Bayesian machine learning

Abstract Rational design of gas diffusion layers (GDL) is an example of a long-standing pursuit to increase the power density and reduce the cost of proton exchange membrane fuel cells (PEMFC). However, current state-of-the-art GDLs are designed by trial-and-error, which is a time-consuming endeavor...

Full description

Saved in:
Bibliographic Details
Main Authors: Jing Sun, Pengzhu Lin, Lin Zeng, Zixiao Guo, Yuting Jiang, Cailin Xiao, Qinping Jian, Jiayou Ren, Lyuming Pan, Xiaosa Xu, Zheng Li, Lei Wei, Tianshou Zhao
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Nature Communications
Online Access:https://doi.org/10.1038/s41467-025-61794-y
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849342470383468544
author Jing Sun
Pengzhu Lin
Lin Zeng
Zixiao Guo
Yuting Jiang
Cailin Xiao
Qinping Jian
Jiayou Ren
Lyuming Pan
Xiaosa Xu
Zheng Li
Lei Wei
Tianshou Zhao
author_facet Jing Sun
Pengzhu Lin
Lin Zeng
Zixiao Guo
Yuting Jiang
Cailin Xiao
Qinping Jian
Jiayou Ren
Lyuming Pan
Xiaosa Xu
Zheng Li
Lei Wei
Tianshou Zhao
author_sort Jing Sun
collection DOAJ
description Abstract Rational design of gas diffusion layers (GDL) is an example of a long-standing pursuit to increase the power density and reduce the cost of proton exchange membrane fuel cells (PEMFC). However, current state-of-the-art GDLs are designed by trial-and-error, which is a time-consuming endeavor. Here, we propose a closed-loop workflow of Bayesian machine learning approach to guide the design of GDL structures. With artificial neural network accelerating the calculation of anisotropic transport properties of reconstructed GDLs, Bayesian optimization algorithm identifies optimal structures in only 40 steps, maximizing the PEMFC’s limiting current density. Results suggest that the optimal porous GDL structure consists of highly orientated fibers with moderate diameters, which is successfully fabricated with a controlled electrospinning technique. The PEMFC demonstrates a high power density of 2.17 W cm-2 and a limiting current density of ~7200 mA cm-2, far exceeding that with commercial GDL (1.33 W cm-2 and ~2700 mA cm-2).
format Article
id doaj-art-9ea00e3bbcd34bb598e2853bd62d3e7f
institution Kabale University
issn 2041-1723
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Nature Communications
spelling doaj-art-9ea00e3bbcd34bb598e2853bd62d3e7f2025-08-20T03:43:22ZengNature PortfolioNature Communications2041-17232025-07-0116111310.1038/s41467-025-61794-yArtificial-intelligence-guided design of ordered gas diffusion layers for high-performing fuel cells via Bayesian machine learningJing Sun0Pengzhu Lin1Lin Zeng2Zixiao Guo3Yuting Jiang4Cailin Xiao5Qinping Jian6Jiayou Ren7Lyuming Pan8Xiaosa Xu9Zheng Li10Lei Wei11Tianshou Zhao12Department of Mechanical and Aerospace Engineering, The Hong Kong University of Science and TechnologyDepartment of Mechanical and Aerospace Engineering, The Hong Kong University of Science and TechnologyDepartment of Mechanical and Energy Engineering, Southern University of Science and TechnologyDepartment of Mechanical and Aerospace Engineering, The Hong Kong University of Science and TechnologyDepartment of Mechanical and Aerospace Engineering, The Hong Kong University of Science and TechnologyDepartment of Mechanical and Energy Engineering, Southern University of Science and TechnologyDepartment of Mechanical and Aerospace Engineering, The Hong Kong University of Science and TechnologyDepartment of Mechanical and Aerospace Engineering, The Hong Kong University of Science and TechnologyDepartment of Mechanical and Energy Engineering, Southern University of Science and TechnologyDepartment of Mechanical and Aerospace Engineering, The Hong Kong University of Science and TechnologyDepartment of Mechanical and Aerospace Engineering, The Hong Kong University of Science and TechnologyDepartment of Mechanical and Energy Engineering, Southern University of Science and TechnologyDepartment of Mechanical and Aerospace Engineering, The Hong Kong University of Science and TechnologyAbstract Rational design of gas diffusion layers (GDL) is an example of a long-standing pursuit to increase the power density and reduce the cost of proton exchange membrane fuel cells (PEMFC). However, current state-of-the-art GDLs are designed by trial-and-error, which is a time-consuming endeavor. Here, we propose a closed-loop workflow of Bayesian machine learning approach to guide the design of GDL structures. With artificial neural network accelerating the calculation of anisotropic transport properties of reconstructed GDLs, Bayesian optimization algorithm identifies optimal structures in only 40 steps, maximizing the PEMFC’s limiting current density. Results suggest that the optimal porous GDL structure consists of highly orientated fibers with moderate diameters, which is successfully fabricated with a controlled electrospinning technique. The PEMFC demonstrates a high power density of 2.17 W cm-2 and a limiting current density of ~7200 mA cm-2, far exceeding that with commercial GDL (1.33 W cm-2 and ~2700 mA cm-2).https://doi.org/10.1038/s41467-025-61794-y
spellingShingle Jing Sun
Pengzhu Lin
Lin Zeng
Zixiao Guo
Yuting Jiang
Cailin Xiao
Qinping Jian
Jiayou Ren
Lyuming Pan
Xiaosa Xu
Zheng Li
Lei Wei
Tianshou Zhao
Artificial-intelligence-guided design of ordered gas diffusion layers for high-performing fuel cells via Bayesian machine learning
Nature Communications
title Artificial-intelligence-guided design of ordered gas diffusion layers for high-performing fuel cells via Bayesian machine learning
title_full Artificial-intelligence-guided design of ordered gas diffusion layers for high-performing fuel cells via Bayesian machine learning
title_fullStr Artificial-intelligence-guided design of ordered gas diffusion layers for high-performing fuel cells via Bayesian machine learning
title_full_unstemmed Artificial-intelligence-guided design of ordered gas diffusion layers for high-performing fuel cells via Bayesian machine learning
title_short Artificial-intelligence-guided design of ordered gas diffusion layers for high-performing fuel cells via Bayesian machine learning
title_sort artificial intelligence guided design of ordered gas diffusion layers for high performing fuel cells via bayesian machine learning
url https://doi.org/10.1038/s41467-025-61794-y
work_keys_str_mv AT jingsun artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT pengzhulin artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT linzeng artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT zixiaoguo artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT yutingjiang artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT cailinxiao artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT qinpingjian artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT jiayouren artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT lyumingpan artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT xiaosaxu artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT zhengli artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT leiwei artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning
AT tianshouzhao artificialintelligenceguideddesignoforderedgasdiffusionlayersforhighperformingfuelcellsviabayesianmachinelearning