Physics Informed Neural Networks for Modeling Large-Scale Wind Driven Ocean Circulation
This study investigates the application of the physics informed neural network as a meshfree collocation method for approximating solutions to large-scale wind driven ocean circulation models. By integrating the Stommel and Stommel-Munk models into the neural network framework, the neural network pr...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10849527/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832576765527064576 |
|---|---|
| author | Boohyun An Mohammad Z. Shanti Chan Yeob Yeun Ernesto Damiani Sungmun Lee Tae-Yeon Kim |
| author_facet | Boohyun An Mohammad Z. Shanti Chan Yeob Yeun Ernesto Damiani Sungmun Lee Tae-Yeon Kim |
| author_sort | Boohyun An |
| collection | DOAJ |
| description | This study investigates the application of the physics informed neural network as a meshfree collocation method for approximating solutions to large-scale wind driven ocean circulation models. By integrating the Stommel and Stommel-Munk models into the neural network framework, the neural network provides a viable alternative to traditional numerical methods for simulating ocean circulation. The architecture of the neural network was systematically optimized through hyperparameter tuning, including the selection of optimizers, activation functions, network configurations, and learning rate schedulers to ensure stable convergence and minimize fluctuations in training loss. The effects of different training point distributions, such as uniform, uniform-refined, random, and random-refined, were also examined. The results show that refining the distribution of training points near the western boundary layer can achieve similar accuracy and training performance even with fewer points. This approach highlights the potential of the physics informed neural network to address more complex oceanographic models, where conventional numerical methods may be constrained by data availability and computational cost. |
| format | Article |
| id | doaj-art-78363b033583489191a9ede1e174f513 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-78363b033583489191a9ede1e174f5132025-01-31T00:01:00ZengIEEEIEEE Access2169-35362025-01-0113187731879710.1109/ACCESS.2025.353266910849527Physics Informed Neural Networks for Modeling Large-Scale Wind Driven Ocean CirculationBoohyun An0https://orcid.org/0000-0002-1927-2474Mohammad Z. Shanti1https://orcid.org/0000-0002-6112-7967Chan Yeob Yeun2https://orcid.org/0000-0002-1398-952XErnesto Damiani3https://orcid.org/0000-0002-9557-6496Sungmun Lee4Tae-Yeon Kim5https://orcid.org/0000-0003-4743-6023Department of Computer Science, C2PS, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesDepartment of Computer Science, C2PS, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesDepartment of Computer Science, C2PS, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesDepartment of Computer Science, C2PS, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesDepartment of Biomedical Engineering and Biotechnology, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesCivil and Environmental Engineering, Khalifa University of Science and Technology, Abu Dhabi, United Arab EmiratesThis study investigates the application of the physics informed neural network as a meshfree collocation method for approximating solutions to large-scale wind driven ocean circulation models. By integrating the Stommel and Stommel-Munk models into the neural network framework, the neural network provides a viable alternative to traditional numerical methods for simulating ocean circulation. The architecture of the neural network was systematically optimized through hyperparameter tuning, including the selection of optimizers, activation functions, network configurations, and learning rate schedulers to ensure stable convergence and minimize fluctuations in training loss. The effects of different training point distributions, such as uniform, uniform-refined, random, and random-refined, were also examined. The results show that refining the distribution of training points near the western boundary layer can achieve similar accuracy and training performance even with fewer points. This approach highlights the potential of the physics informed neural network to address more complex oceanographic models, where conventional numerical methods may be constrained by data availability and computational cost.https://ieeexplore.ieee.org/document/10849527/Deep learningocean currentsphysics informed neural networkquasi-geostrophicStommelStommel-Munk |
| spellingShingle | Boohyun An Mohammad Z. Shanti Chan Yeob Yeun Ernesto Damiani Sungmun Lee Tae-Yeon Kim Physics Informed Neural Networks for Modeling Large-Scale Wind Driven Ocean Circulation IEEE Access Deep learning ocean currents physics informed neural network quasi-geostrophic Stommel Stommel-Munk |
| title | Physics Informed Neural Networks for Modeling Large-Scale Wind Driven Ocean Circulation |
| title_full | Physics Informed Neural Networks for Modeling Large-Scale Wind Driven Ocean Circulation |
| title_fullStr | Physics Informed Neural Networks for Modeling Large-Scale Wind Driven Ocean Circulation |
| title_full_unstemmed | Physics Informed Neural Networks for Modeling Large-Scale Wind Driven Ocean Circulation |
| title_short | Physics Informed Neural Networks for Modeling Large-Scale Wind Driven Ocean Circulation |
| title_sort | physics informed neural networks for modeling large scale wind driven ocean circulation |
| topic | Deep learning ocean currents physics informed neural network quasi-geostrophic Stommel Stommel-Munk |
| url | https://ieeexplore.ieee.org/document/10849527/ |
| work_keys_str_mv | AT boohyunan physicsinformedneuralnetworksformodelinglargescalewinddrivenoceancirculation AT mohammadzshanti physicsinformedneuralnetworksformodelinglargescalewinddrivenoceancirculation AT chanyeobyeun physicsinformedneuralnetworksformodelinglargescalewinddrivenoceancirculation AT ernestodamiani physicsinformedneuralnetworksformodelinglargescalewinddrivenoceancirculation AT sungmunlee physicsinformedneuralnetworksformodelinglargescalewinddrivenoceancirculation AT taeyeonkim physicsinformedneuralnetworksformodelinglargescalewinddrivenoceancirculation |