Physics-Based Learning Models for Ship Hydrodynamics
We present the concepts of physics-based learning models (PBLM) and their relevance and application to the field of ship hydrodynamics. The utility of physics-based learning is motivated by contrasting generic learning models for regression predictions, which do not presume any knowledge of the syst...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
The Society of Naval Architects and Marine Engineers
2015
|
| Online Access: | http://hdl.handle.net/1721.1/97749 https://orcid.org/0000-0003-1273-9964 https://orcid.org/0000-0001-5080-5016 |
| _version_ | 1826214761419767808 |
|---|---|
| author | Weymouth, Gabriel D. Yue, Dick K. P. |
| author2 | Massachusetts Institute of Technology. Department of Mechanical Engineering |
| author_facet | Massachusetts Institute of Technology. Department of Mechanical Engineering Weymouth, Gabriel D. Yue, Dick K. P. |
| author_sort | Weymouth, Gabriel D. |
| collection | MIT |
| description | We present the concepts of physics-based learning models (PBLM) and their relevance and application to the field of ship hydrodynamics. The utility of physics-based learning is motivated by contrasting generic learning models for regression predictions, which do not presume any knowledge of the system other than the training data provided with methods such as semi-empirical models, which incorporate physical insights along with data-fitting. PBLM provides a framework wherein intermediate models, which capture (some) physical aspects of the problem, are incorporated into modern generic learning tools to substantially improve the predictions of the latter, minimizing the reliance on costly experimental measurements or high-resolution high-fidelity numerical solutions. To illustrate the versatility and efficacy of PBLM, we present three wave-ship interaction problems: 1) at speed waterline profiles; 2) ship motions in head seas; and 3) three-dimensional breaking bow waves. PBLM is shown to be robust and produce error rates at or below the uncertainty in the generated data at a small fraction of the expense of high-resolution numerical predictions. |
| first_indexed | 2024-09-23T16:10:44Z |
| format | Article |
| id | mit-1721.1/97749 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T16:10:44Z |
| publishDate | 2015 |
| publisher | The Society of Naval Architects and Marine Engineers |
| record_format | dspace |
| spelling | mit-1721.1/977492022-10-02T06:51:50Z Physics-Based Learning Models for Ship Hydrodynamics Weymouth, Gabriel D. Yue, Dick K. P. Massachusetts Institute of Technology. Department of Mechanical Engineering Weymouth, Gabriel D. Yue, Dick K. P. We present the concepts of physics-based learning models (PBLM) and their relevance and application to the field of ship hydrodynamics. The utility of physics-based learning is motivated by contrasting generic learning models for regression predictions, which do not presume any knowledge of the system other than the training data provided with methods such as semi-empirical models, which incorporate physical insights along with data-fitting. PBLM provides a framework wherein intermediate models, which capture (some) physical aspects of the problem, are incorporated into modern generic learning tools to substantially improve the predictions of the latter, minimizing the reliance on costly experimental measurements or high-resolution high-fidelity numerical solutions. To illustrate the versatility and efficacy of PBLM, we present three wave-ship interaction problems: 1) at speed waterline profiles; 2) ship motions in head seas; and 3) three-dimensional breaking bow waves. PBLM is shown to be robust and produce error rates at or below the uncertainty in the generated data at a small fraction of the expense of high-resolution numerical predictions. United States. Office of Naval Research 2015-07-16T13:16:06Z 2015-07-16T13:16:06Z 2013-03 2012-07 Article http://purl.org/eprint/type/JournalArticle 00224502 15420604 http://hdl.handle.net/1721.1/97749 Weymouth, Gabriel D., and Dick K.P. Yue. “Physics-Based Learning Models for Ship Hydrodynamics.” Journal of Ship Research 57, no. 1 (March 1, 2013): 1–12. https://orcid.org/0000-0003-1273-9964 https://orcid.org/0000-0001-5080-5016 en_US http://dx.doi.org/10.5957/JOSR.57.1.120005 Journal of Ship Research Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf The Society of Naval Architects and Marine Engineers arXiv |
| spellingShingle | Weymouth, Gabriel D. Yue, Dick K. P. Physics-Based Learning Models for Ship Hydrodynamics |
| title | Physics-Based Learning Models for Ship Hydrodynamics |
| title_full | Physics-Based Learning Models for Ship Hydrodynamics |
| title_fullStr | Physics-Based Learning Models for Ship Hydrodynamics |
| title_full_unstemmed | Physics-Based Learning Models for Ship Hydrodynamics |
| title_short | Physics-Based Learning Models for Ship Hydrodynamics |
| title_sort | physics based learning models for ship hydrodynamics |
| url | http://hdl.handle.net/1721.1/97749 https://orcid.org/0000-0003-1273-9964 https://orcid.org/0000-0001-5080-5016 |
| work_keys_str_mv | AT weymouthgabrield physicsbasedlearningmodelsforshiphydrodynamics AT yuedickkp physicsbasedlearningmodelsforshiphydrodynamics |