Efficient Prediction of Fuel Cell Performance Using Global Modeling Method
A global modeling method is developed to describe the relationship between multi-type parameters and fuel cell performance, which significantly contributes to the efficient performance prediction of fuel cell systems. The multi-type parameters include operating parameters, geometric parameters of th...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-11-01
|
Series: | Energies |
Subjects: | |
Online Access: | https://www.mdpi.com/1996-1073/15/22/8549 |
_version_ | 1797465416903163904 |
---|---|
author | Qinwen Yang Gang Xiao Tao Liu Bin Gao Shujun Chen |
author_facet | Qinwen Yang Gang Xiao Tao Liu Bin Gao Shujun Chen |
author_sort | Qinwen Yang |
collection | DOAJ |
description | A global modeling method is developed to describe the relationship between multi-type parameters and fuel cell performance, which significantly contributes to the efficient performance prediction of fuel cell systems. The multi-type parameters include operating parameters, geometric parameters of the graphite end plates, and the membrane electrolyte assembly physical parameters. An adaptive sampling method integrated with the Kriging method is newly developed for global modeling. Experiments are designed and implemented for model construction and evaluation. The results show the local development and global development in the whole design space can be balanced during the adaptive sampling process. Meanwhile, the prediction capability of accuracy and sensitivity for the global model is reliable in the whole design space. The prediction accuracy is improved by nearly 26% compared to the fuel cell model built for optimization with the same sample size. The prediction sensitivity also proved that the global model could follow the experimental variations under small input deviations. |
first_indexed | 2024-03-09T18:21:09Z |
format | Article |
id | doaj.art-029392c190614be3bed7e48298ba4c6f |
institution | Directory Open Access Journal |
issn | 1996-1073 |
language | English |
last_indexed | 2024-03-09T18:21:09Z |
publishDate | 2022-11-01 |
publisher | MDPI AG |
record_format | Article |
series | Energies |
spelling | doaj.art-029392c190614be3bed7e48298ba4c6f2023-11-24T08:14:48ZengMDPI AGEnergies1996-10732022-11-011522854910.3390/en15228549Efficient Prediction of Fuel Cell Performance Using Global Modeling MethodQinwen Yang0Gang Xiao1Tao Liu2Bin Gao3Shujun Chen4College of Mechanical and Vehicle Engineering, State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha 410082, ChinaCollege of Mechanical and Vehicle Engineering, State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha 410082, ChinaCollege of Mechanical and Vehicle Engineering, State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha 410082, ChinaCollege of Mechanical and Vehicle Engineering, State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha 410082, ChinaCollege of Mechanical and Vehicle Engineering, State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha 410082, ChinaA global modeling method is developed to describe the relationship between multi-type parameters and fuel cell performance, which significantly contributes to the efficient performance prediction of fuel cell systems. The multi-type parameters include operating parameters, geometric parameters of the graphite end plates, and the membrane electrolyte assembly physical parameters. An adaptive sampling method integrated with the Kriging method is newly developed for global modeling. Experiments are designed and implemented for model construction and evaluation. The results show the local development and global development in the whole design space can be balanced during the adaptive sampling process. Meanwhile, the prediction capability of accuracy and sensitivity for the global model is reliable in the whole design space. The prediction accuracy is improved by nearly 26% compared to the fuel cell model built for optimization with the same sample size. The prediction sensitivity also proved that the global model could follow the experimental variations under small input deviations.https://www.mdpi.com/1996-1073/15/22/8549fuel cellenergy conversion efficiencyglobal modeling methodadaptive sampling method |
spellingShingle | Qinwen Yang Gang Xiao Tao Liu Bin Gao Shujun Chen Efficient Prediction of Fuel Cell Performance Using Global Modeling Method Energies fuel cell energy conversion efficiency global modeling method adaptive sampling method |
title | Efficient Prediction of Fuel Cell Performance Using Global Modeling Method |
title_full | Efficient Prediction of Fuel Cell Performance Using Global Modeling Method |
title_fullStr | Efficient Prediction of Fuel Cell Performance Using Global Modeling Method |
title_full_unstemmed | Efficient Prediction of Fuel Cell Performance Using Global Modeling Method |
title_short | Efficient Prediction of Fuel Cell Performance Using Global Modeling Method |
title_sort | efficient prediction of fuel cell performance using global modeling method |
topic | fuel cell energy conversion efficiency global modeling method adaptive sampling method |
url | https://www.mdpi.com/1996-1073/15/22/8549 |
work_keys_str_mv | AT qinwenyang efficientpredictionoffuelcellperformanceusingglobalmodelingmethod AT gangxiao efficientpredictionoffuelcellperformanceusingglobalmodelingmethod AT taoliu efficientpredictionoffuelcellperformanceusingglobalmodelingmethod AT bingao efficientpredictionoffuelcellperformanceusingglobalmodelingmethod AT shujunchen efficientpredictionoffuelcellperformanceusingglobalmodelingmethod |