preprints_ui: pu2jx_v1
Data license: ODbL (database) & original licenses (content) · Data source: Open Science Framework
| id | title | description | date_created | date_modified | date_published | original_publication_date | publication_doi | provider | is_published | reviews_state | version | is_latest_version | preprint_doi | license | tags_list | tags_data | contributors_list | contributors_data | first_author | subjects_list | subjects_data | download_url | has_coi | conflict_of_interest_statement | has_data_links | has_prereg_links | prereg_links | prereg_link_info | last_updated |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| pu2jx_v1 | Using Physics-Informed Generative Adversarial Networks to Perform Super-Resolution for Multiphase Fluid Simulations | Computational Fluid Dynamics (CFD) simulations are useful to the field of engineering design as they provide deep insights on product or system performance without the need to construct and test physical prototypes. However, they can be very computationally intensive to run. Machine learning methods have been shown to reconstruct high- resolution single-phase turbulent fluid flow simulations from low-resolution inputs. This offers a potential avenue towards alleviating computational cost in iterative engineering design applications. However, little work thus far has explored the application of machine learning image super-resolution methods to multiphase fluid flow (which is important for emerging fields such as marine hydrokinetic energy conversion). In this work, we apply a modified version of the Super-Resolution Generative Adversarial Network (SRGAN) model to a multiphase turbulent fluid flow problem, specifically to reconstruct fluid phase fraction at a higher resolution. Two models were created in this work, one which incorporates a physics-informed term in the loss function and one which does not, and the results are discussed and compared. We found that both models significantly outperform non-machine learning upsampling methods and can preserve a substantial amount of detail, showing the versatility of the SRGAN model for upsampling multi-phase fluid simulations. However, the difference in accuracy between the two models is minimal indicating that, in the context studied here, the additional complexity of a physics- informed approach may not be justified. | 2022-01-17T14:23:00.627706 | 2022-03-01T18:52:19.396766 | 2022-01-17T15:42:58.062171 | 2022-01-17T05:00:00 | engrxiv | 0 | withdrawn | 1 | 1 | https://doi.org/10.31224/osf.io/pu2jx | CC-BY Attribution-NonCommercial 4.0 International | CFD; ML; computational fluid dynamics; machine learning; physics-informed machine learning; super resolution | ["CFD", "ML", "computational fluid dynamics", "machine learning", "physics-informed machine learning", "super resolution"] | Matthew Li; Christopher McComb | [{"id": "emhbv", "name": "Matthew Li", "index": 0, "orcid": null, "bibliographic": true}, {"id": "9kbjc", "name": "Christopher McComb", "index": 1, "orcid": "0000-0002-5024-7701", "bibliographic": true}] | Matthew Li | Engineering; Mechanical Engineering; Fluid Mechanics | [{"id": "5994df7a54be8100732d43ae", "text": "Engineering"}, {"id": "5994df7a54be8100732d43af", "text": "Mechanical Engineering"}, {"id": "5994df7c54be8100732d441d", "text": "Fluid Mechanics"}] | 0 | no | no | [] | 2025-04-09T20:03:40.615731 |