Toward Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses
<jats:title>Abstract</jats:title> <jats:p>Surrogate models have several uses in engineering design, including speeding up design optimization, noise reduction, test measurement interpolation, gradient estimation, portability, and protection of intellectual property....
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
ASME International
2022
|
| Online Access: | https://hdl.handle.net/1721.1/145570 |
| _version_ | 1826204628129153024 |
|---|---|
| author | Whalen, Eamon Mueller, Caitlin |
| author2 | Massachusetts Institute of Technology. Department of Architecture |
| author_facet | Massachusetts Institute of Technology. Department of Architecture Whalen, Eamon Mueller, Caitlin |
| author_sort | Whalen, Eamon |
| collection | MIT |
| description | <jats:title>Abstract</jats:title>
<jats:p>Surrogate models have several uses in engineering design, including speeding up design optimization, noise reduction, test measurement interpolation, gradient estimation, portability, and protection of intellectual property. Traditionally, surrogate models require that all training data conform to the same parametrization (e.g., design variables), limiting design freedom and prohibiting the reuse of historical data. In response, this article proposes graph-based surrogate models (GSMs) for trusses. The GSM can accurately predict displacement fields from static loads given the structure’s geometry as input, enabling training across multiple parametrizations. GSMs build upon recent advancements in geometric deep learning, which have led to the ability to learn on undirected graphs: a natural representation for trusses. To further promote flexible surrogate models, this article explores transfer learning within the context of engineering design and demonstrates positive knowledge transfer across data sets of different topologies, complexities, loads, and applications, resulting in more flexible and data-efficient surrogate models for trusses.</jats:p> |
| first_indexed | 2024-09-23T12:58:28Z |
| format | Article |
| id | mit-1721.1/145570 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T12:58:28Z |
| publishDate | 2022 |
| publisher | ASME International |
| record_format | dspace |
| spelling | mit-1721.1/1455702022-10-04T03:23:42Z Toward Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses Whalen, Eamon Mueller, Caitlin Massachusetts Institute of Technology. Department of Architecture <jats:title>Abstract</jats:title> <jats:p>Surrogate models have several uses in engineering design, including speeding up design optimization, noise reduction, test measurement interpolation, gradient estimation, portability, and protection of intellectual property. Traditionally, surrogate models require that all training data conform to the same parametrization (e.g., design variables), limiting design freedom and prohibiting the reuse of historical data. In response, this article proposes graph-based surrogate models (GSMs) for trusses. The GSM can accurately predict displacement fields from static loads given the structure’s geometry as input, enabling training across multiple parametrizations. GSMs build upon recent advancements in geometric deep learning, which have led to the ability to learn on undirected graphs: a natural representation for trusses. To further promote flexible surrogate models, this article explores transfer learning within the context of engineering design and demonstrates positive knowledge transfer across data sets of different topologies, complexities, loads, and applications, resulting in more flexible and data-efficient surrogate models for trusses.</jats:p> 2022-09-26T17:03:28Z 2022-09-26T17:03:28Z 2022 2022-09-26T14:26:44Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/145570 Whalen, Eamon and Mueller, Caitlin. 2022. "Toward Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses." Journal of Mechanical Design, 144 (2). en 10.1115/1.4052298 Journal of Mechanical Design Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf ASME International ASME |
| spellingShingle | Whalen, Eamon Mueller, Caitlin Toward Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses |
| title | Toward Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses |
| title_full | Toward Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses |
| title_fullStr | Toward Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses |
| title_full_unstemmed | Toward Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses |
| title_short | Toward Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses |
| title_sort | toward reusable surrogate models graph based transfer learning on trusses |
| url | https://hdl.handle.net/1721.1/145570 |
| work_keys_str_mv | AT whaleneamon towardreusablesurrogatemodelsgraphbasedtransferlearningontrusses AT muellercaitlin towardreusablesurrogatemodelsgraphbasedtransferlearningontrusses |