Deep Neural Network for Prediction of Adsorbent Selectivity on Hydrogen Purification

Chenglong Li; Chengsi Xie; Yi Zong; Richard Chahine; Tianqi Yang; Feng Ye; Jinsheng Xiao

doi:10.1007/978-981-99-8631-6_24

Deep Neural Network for Prediction of Adsorbent Selectivity on Hydrogen Purification

Chenglong Li^*
, Chengsi Xie
, Yi Zong
, Richard Chahine
, Tianqi Yang
, Feng Ye
, Jinsheng Xiao

^*Corresponding author for this work

Wuhan City Polytechnic
Wuhan University of Technology
Université du Québec à Trois-Rivières
Jianghan University

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

169 Downloads (Orbit)

Abstract

With emergence of new materials, more and more materials are available for adsorption and separation processes. The adsorption selectivity of adsorbent to adsorbate is one of the important indicators in choosing materials. Because the adsorption experiment of the mixture is time-consuming and difficult, the selectivity of the adsorbent is generally calculated by the ideal adsorbed solution theory (IAST). Taking the CO₂/H₂ gas mixture as an example, this paper proposes a new adsorption selectivity calculation method based on a deep neural network (DNN) with 5 hidden layers, which takes the molar fraction of CO₂, adsorption pressure and Langmuir adsorption isotherm parameters as the inputs of DNN. Combining the DNN and the NIST/ARPA-E database to quickly and accurately calculate the adsorption selectivity, the hydrogen purification and carbon dioxide storage materials can be quickly screened.

Original language	English
Title of host publication	Proceedings of the 10th Hydrogen Technology Convention
Editors	Hexu Sun, Wei Pei, Yan Dong, Hongmei Yu, Shi You
Number of pages	8
Volume	1
Publisher	Springer
Publication date	2024
Pages	214-221
ISBN (Print)	978-981-99-8630-9, 978-981-99-8633-0
ISBN (Electronic)	978-981-99-8631-6
DOIs	https://doi.org/10.1007/978-981-99-8631-6_24
Publication status	Published - 2024
Event	10^th Hydrogen Technology Convention - Foshan, China Duration: 22 May 2023 → 26 May 2023 Conference number: 10

Conference

Conference	10^th Hydrogen Technology Convention
Number	10
Country/Territory	China
City	Foshan
Period	22/05/2023 → 26/05/2023

Series	Springer Proceedings in Physics
Volume	393
ISSN	0930-8989

Keywords

Hydrogen purification
Ideal absorbed solution theory
Langmuir isotherm
Selectivity
Deep neural network

Access to Document

10.1007/978-981-99-8631-6_24Licence: CC BY

FulltextFinal published version, 767 KBLicence: CC BY

OpenUrl availability

Check availability in DTU Findit

Cite this

@inproceedings{453681efc6a14282ae0bfb8757d1e385,

title = "Deep Neural Network for Prediction of Adsorbent Selectivity on Hydrogen Purification",

abstract = "With emergence of new materials, more and more materials are available for adsorption and separation processes. The adsorption selectivity of adsorbent to adsorbate is one of the important indicators in choosing materials. Because the adsorption experiment of the mixture is time-consuming and difficult, the selectivity of the adsorbent is generally calculated by the ideal adsorbed solution theory (IAST). Taking the CO2/H2 gas mixture as an example, this paper proposes a new adsorption selectivity calculation method based on a deep neural network (DNN) with 5 hidden layers, which takes the molar fraction of CO2, adsorption pressure and Langmuir adsorption isotherm parameters as the inputs of DNN. Combining the DNN and the NIST/ARPA-E database to quickly and accurately calculate the adsorption selectivity, the hydrogen purification and carbon dioxide storage materials can be quickly screened.",

keywords = "Hydrogen purification, Ideal absorbed solution theory, Langmuir isotherm, Selectivity, Deep neural network",

author = "Chenglong Li and Chengsi Xie and Yi Zong and Richard Chahine and Tianqi Yang and Feng Ye and Jinsheng Xiao",

year = "2024",

doi = "10.1007/978-981-99-8631-6\_24",

language = "English",

isbn = "978-981-99-8630-9",

volume = "1",

series = "Springer Proceedings in Physics",

publisher = "Springer",

pages = "214--221",

editor = "Hexu Sun and Wei Pei and Yan Dong and Hongmei Yu and Shi You",

booktitle = "Proceedings of the 10th Hydrogen Technology Convention",

note = "10<sup>th</sup> Hydrogen Technology Convention ; Conference date: 22-05-2023 Through 26-05-2023",

}

Li, C, Xie, C, Zong, Y, Chahine, R, Yang, T, Ye, F & Xiao, J 2024, Deep Neural Network for Prediction of Adsorbent Selectivity on Hydrogen Purification. in H Sun, W Pei, Y Dong, H Yu & S You (eds), Proceedings of the 10th Hydrogen Technology Convention. vol. 1, Springer, Springer Proceedings in Physics, vol. 393, pp. 214-221, 10^th Hydrogen Technology Convention, Foshan, China, 22/05/2023. https://doi.org/10.1007/978-981-99-8631-6_24

Deep Neural Network for Prediction of Adsorbent Selectivity on Hydrogen Purification. / Li, Chenglong; Xie, Chengsi; Zong, Yi et al.
Proceedings of the 10th Hydrogen Technology Convention. ed. / Hexu Sun; Wei Pei; Yan Dong; Hongmei Yu; Shi You. Vol. 1 Springer, 2024. p. 214-221 (Springer Proceedings in Physics, Vol. 393).

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

TY - GEN

T1 - Deep Neural Network for Prediction of Adsorbent Selectivity on Hydrogen Purification

AU - Li, Chenglong

AU - Xie, Chengsi

AU - Zong, Yi

AU - Chahine, Richard

AU - Yang, Tianqi

AU - Ye, Feng

AU - Xiao, Jinsheng

N1 - Conference code: 10

PY - 2024

Y1 - 2024

N2 - With emergence of new materials, more and more materials are available for adsorption and separation processes. The adsorption selectivity of adsorbent to adsorbate is one of the important indicators in choosing materials. Because the adsorption experiment of the mixture is time-consuming and difficult, the selectivity of the adsorbent is generally calculated by the ideal adsorbed solution theory (IAST). Taking the CO2/H2 gas mixture as an example, this paper proposes a new adsorption selectivity calculation method based on a deep neural network (DNN) with 5 hidden layers, which takes the molar fraction of CO2, adsorption pressure and Langmuir adsorption isotherm parameters as the inputs of DNN. Combining the DNN and the NIST/ARPA-E database to quickly and accurately calculate the adsorption selectivity, the hydrogen purification and carbon dioxide storage materials can be quickly screened.

AB - With emergence of new materials, more and more materials are available for adsorption and separation processes. The adsorption selectivity of adsorbent to adsorbate is one of the important indicators in choosing materials. Because the adsorption experiment of the mixture is time-consuming and difficult, the selectivity of the adsorbent is generally calculated by the ideal adsorbed solution theory (IAST). Taking the CO2/H2 gas mixture as an example, this paper proposes a new adsorption selectivity calculation method based on a deep neural network (DNN) with 5 hidden layers, which takes the molar fraction of CO2, adsorption pressure and Langmuir adsorption isotherm parameters as the inputs of DNN. Combining the DNN and the NIST/ARPA-E database to quickly and accurately calculate the adsorption selectivity, the hydrogen purification and carbon dioxide storage materials can be quickly screened.

KW - Hydrogen purification

KW - Ideal absorbed solution theory

KW - Langmuir isotherm

KW - Selectivity

KW - Deep neural network

U2 - 10.1007/978-981-99-8631-6_24

DO - 10.1007/978-981-99-8631-6_24

M3 - Article in proceedings

SN - 978-981-99-8630-9

SN - 978-981-99-8633-0

VL - 1

T3 - Springer Proceedings in Physics

SP - 214

EP - 221

BT - Proceedings of the 10th Hydrogen Technology Convention

A2 - Sun, Hexu

A2 - Pei, Wei

A2 - Dong, Yan

A2 - Yu, Hongmei

A2 - You, Shi

PB - Springer

T2 - 10<sup>th</sup> Hydrogen Technology Convention

Y2 - 22 May 2023 through 26 May 2023

ER -

Deep Neural Network for Prediction of Adsorbent Selectivity on Hydrogen Purification

Abstract

Conference

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this