Flow Characteristics of Fibrous Gas Diffusion Layers Using Machine Learning Methods
The material characteristics of gas diffusion layers are relevant for the efficient operation of polymer electrolyte fuel cells. The current state-of-the-art calculates these using transport simulations based on their micro-structures, either reconstructed or generated by means of stochastic geometr...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/12/23/12193 |
| _version_ | 1797463643167653888 |
|---|---|
| author | Dieter Froning Jannik Wirtz Eugen Hoppe Werner Lehnert |
| author_facet | Dieter Froning Jannik Wirtz Eugen Hoppe Werner Lehnert |
| author_sort | Dieter Froning |
| collection | DOAJ |
| description | The material characteristics of gas diffusion layers are relevant for the efficient operation of polymer electrolyte fuel cells. The current state-of-the-art calculates these using transport simulations based on their micro-structures, either reconstructed or generated by means of stochastic geometry models. Such transport simulations often require high computational resources. To support material characterization using artificial-intelligence-based methods, in this study, a convolutional neural network was developed. It was trained with results from previous transport simulations and validated using five-fold cross-validation. The neural network enables the permeability of paper-type gas diffusion layers to be predicted. A stochastic arrangement of the fibers, four types of binder distributions, and compression of up to 50% are also considered. The binder type and compression level were features inherent to the material but were not the subject of the training. In this regard, they can be seen as features hidden from the training process. Nevertheless, these characteristics were reproduced with the proposed machine learning model. With a trained machine learning model, the prediction of permeability can be performed on a standard computer. |
| first_indexed | 2024-03-09T17:53:39Z |
| format | Article |
| id | doaj.art-c9958c57119541748028273c64b64477 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T17:53:39Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-c9958c57119541748028273c64b644772023-11-24T10:32:30ZengMDPI AGApplied Sciences2076-34172022-11-0112231219310.3390/app122312193Flow Characteristics of Fibrous Gas Diffusion Layers Using Machine Learning MethodsDieter Froning0Jannik Wirtz1Eugen Hoppe2Werner Lehnert3Forschungszentrum Jülich GmbH, Institute of Energy and Climate Research, Electrochemical Process Engineering (IEK-14), 52425 Jülich, GermanyForschungszentrum Jülich GmbH, Institute of Energy and Climate Research, Electrochemical Process Engineering (IEK-14), 52425 Jülich, GermanyForschungszentrum Jülich GmbH, Institute of Energy and Climate Research, Electrochemical Process Engineering (IEK-14), 52425 Jülich, GermanyForschungszentrum Jülich GmbH, Institute of Energy and Climate Research, Electrochemical Process Engineering (IEK-14), 52425 Jülich, GermanyThe material characteristics of gas diffusion layers are relevant for the efficient operation of polymer electrolyte fuel cells. The current state-of-the-art calculates these using transport simulations based on their micro-structures, either reconstructed or generated by means of stochastic geometry models. Such transport simulations often require high computational resources. To support material characterization using artificial-intelligence-based methods, in this study, a convolutional neural network was developed. It was trained with results from previous transport simulations and validated using five-fold cross-validation. The neural network enables the permeability of paper-type gas diffusion layers to be predicted. A stochastic arrangement of the fibers, four types of binder distributions, and compression of up to 50% are also considered. The binder type and compression level were features inherent to the material but were not the subject of the training. In this regard, they can be seen as features hidden from the training process. Nevertheless, these characteristics were reproduced with the proposed machine learning model. With a trained machine learning model, the prediction of permeability can be performed on a standard computer.https://www.mdpi.com/2076-3417/12/23/12193PEFCmachine learninglattice Boltzmannstochastic modeling |
| spellingShingle | Dieter Froning Jannik Wirtz Eugen Hoppe Werner Lehnert Flow Characteristics of Fibrous Gas Diffusion Layers Using Machine Learning Methods Applied Sciences PEFC machine learning lattice Boltzmann stochastic modeling |
| title | Flow Characteristics of Fibrous Gas Diffusion Layers Using Machine Learning Methods |
| title_full | Flow Characteristics of Fibrous Gas Diffusion Layers Using Machine Learning Methods |
| title_fullStr | Flow Characteristics of Fibrous Gas Diffusion Layers Using Machine Learning Methods |
| title_full_unstemmed | Flow Characteristics of Fibrous Gas Diffusion Layers Using Machine Learning Methods |
| title_short | Flow Characteristics of Fibrous Gas Diffusion Layers Using Machine Learning Methods |
| title_sort | flow characteristics of fibrous gas diffusion layers using machine learning methods |
| topic | PEFC machine learning lattice Boltzmann stochastic modeling |
| url | https://www.mdpi.com/2076-3417/12/23/12193 |
| work_keys_str_mv | AT dieterfroning flowcharacteristicsoffibrousgasdiffusionlayersusingmachinelearningmethods AT jannikwirtz flowcharacteristicsoffibrousgasdiffusionlayersusingmachinelearningmethods AT eugenhoppe flowcharacteristicsoffibrousgasdiffusionlayersusingmachinelearningmethods AT wernerlehnert flowcharacteristicsoffibrousgasdiffusionlayersusingmachinelearningmethods |