Transfer learning for predicting source terms of principal component transport in chemically reactive flow
Transfer learning has been highlighted as a promising framework to increase the accuracy of the data-driven model in the case of data sparsity, specifically by leveraging pretrained knowledge to the training of the target model. The objective of this study is to evaluate whether the number of requis...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Cambridge University Press
2024-01-01
|
| Series: | Data-Centric Engineering |
| Subjects: | |
| Online Access: | https://www.cambridge.org/core/product/identifier/S2632673624000509/type/journal_article |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850061439555862528 |
|---|---|
| author | Ki Sung Jung Tarek Echekki Jacqueline H. Chen Mohammad Khalil |
| author_facet | Ki Sung Jung Tarek Echekki Jacqueline H. Chen Mohammad Khalil |
| author_sort | Ki Sung Jung |
| collection | DOAJ |
| description | Transfer learning has been highlighted as a promising framework to increase the accuracy of the data-driven model in the case of data sparsity, specifically by leveraging pretrained knowledge to the training of the target model. The objective of this study is to evaluate whether the number of requisite training samples can be reduced with the use of various transfer learning models for predicting, for example, the chemical source terms of the data-driven reduced-order modeling (ROM) that represents the homogeneous ignition of a hydrogen/air mixture. Principal component analysis is applied to reduce the dimensionality of the hydrogen/air mixture in composition space. Artificial neural networks (ANNs) are used to regress the reaction rates of principal components, and subsequently, a system of ordinary differential equations is solved. As the number of training samples decreases in the target task, the ROM fails to predict the ignition evolution of a hydrogen/air mixture. Three transfer learning strategies are then applied to the training of the ANN model with a sparse dataset. The performance of the ROM with a sparse dataset is remarkably enhanced if the training of the ANN model is restricted by a regularization term that controls the degree of knowledge transfer from source to target tasks. To this end, a novel transfer learning method is introduced, Parameter control via Partial Initialization and Regularization (PaPIR), whereby the amount of knowledge transferred is systemically adjusted in terms of the initialization and regularization schemes of the ANN model in the target task. |
| format | Article |
| id | doaj-art-27f4f9d7772843ac93efc4eaf4bfd339 |
| institution | DOAJ |
| issn | 2632-6736 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | Cambridge University Press |
| record_format | Article |
| series | Data-Centric Engineering |
| spelling | doaj-art-27f4f9d7772843ac93efc4eaf4bfd3392025-08-20T02:50:14ZengCambridge University PressData-Centric Engineering2632-67362024-01-01510.1017/dce.2024.50Transfer learning for predicting source terms of principal component transport in chemically reactive flowKi Sung Jung0https://orcid.org/0000-0003-4356-1353Tarek Echekki1https://orcid.org/0000-0002-0146-7994Jacqueline H. Chen2Mohammad Khalil3Combustion Research Facility, Sandia National Laboratories, Livermore, CA, USA Department of Mechanical Engineering, Pukyong National University, Busan, Republic of KoreaDepartment of Mechanical and Aerospace Engineering, North Carolina State University, Raleigh, NC, USACombustion Research Facility, Sandia National Laboratories, Livermore, CA, USACombustion Research Facility, Sandia National Laboratories, Livermore, CA, USATransfer learning has been highlighted as a promising framework to increase the accuracy of the data-driven model in the case of data sparsity, specifically by leveraging pretrained knowledge to the training of the target model. The objective of this study is to evaluate whether the number of requisite training samples can be reduced with the use of various transfer learning models for predicting, for example, the chemical source terms of the data-driven reduced-order modeling (ROM) that represents the homogeneous ignition of a hydrogen/air mixture. Principal component analysis is applied to reduce the dimensionality of the hydrogen/air mixture in composition space. Artificial neural networks (ANNs) are used to regress the reaction rates of principal components, and subsequently, a system of ordinary differential equations is solved. As the number of training samples decreases in the target task, the ROM fails to predict the ignition evolution of a hydrogen/air mixture. Three transfer learning strategies are then applied to the training of the ANN model with a sparse dataset. The performance of the ROM with a sparse dataset is remarkably enhanced if the training of the ANN model is restricted by a regularization term that controls the degree of knowledge transfer from source to target tasks. To this end, a novel transfer learning method is introduced, Parameter control via Partial Initialization and Regularization (PaPIR), whereby the amount of knowledge transferred is systemically adjusted in terms of the initialization and regularization schemes of the ANN model in the target task.https://www.cambridge.org/core/product/identifier/S2632673624000509/type/journal_articleartificial neural networkchemical kineticsprincipal component analysisreduced order modelingtransfer learning |
| spellingShingle | Ki Sung Jung Tarek Echekki Jacqueline H. Chen Mohammad Khalil Transfer learning for predicting source terms of principal component transport in chemically reactive flow Data-Centric Engineering artificial neural network chemical kinetics principal component analysis reduced order modeling transfer learning |
| title | Transfer learning for predicting source terms of principal component transport in chemically reactive flow |
| title_full | Transfer learning for predicting source terms of principal component transport in chemically reactive flow |
| title_fullStr | Transfer learning for predicting source terms of principal component transport in chemically reactive flow |
| title_full_unstemmed | Transfer learning for predicting source terms of principal component transport in chemically reactive flow |
| title_short | Transfer learning for predicting source terms of principal component transport in chemically reactive flow |
| title_sort | transfer learning for predicting source terms of principal component transport in chemically reactive flow |
| topic | artificial neural network chemical kinetics principal component analysis reduced order modeling transfer learning |
| url | https://www.cambridge.org/core/product/identifier/S2632673624000509/type/journal_article |
| work_keys_str_mv | AT kisungjung transferlearningforpredictingsourcetermsofprincipalcomponenttransportinchemicallyreactiveflow AT tarekechekki transferlearningforpredictingsourcetermsofprincipalcomponenttransportinchemicallyreactiveflow AT jacquelinehchen transferlearningforpredictingsourcetermsofprincipalcomponenttransportinchemicallyreactiveflow AT mohammadkhalil transferlearningforpredictingsourcetermsofprincipalcomponenttransportinchemicallyreactiveflow |