Toward Reusable Surrogate Models: Graph-Based Transfer Learning on Trusses
- 21 September 2021
- journal article
- research article
- Published by ASME International in Journal of Mechanical Design
- Vol. 144 (2), 1-12
- https://doi.org/10.1115/1.4052298
Abstract
Surrogate models are often employed to speed up engineering design optimization; however, they typically 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 paper proposes Graph-based Surrogate Models (GSMs) for space frame structures. 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 space frames. To further promote flexible surrogate models, the paper 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 space frame structures.Keywords
This publication has 34 references indexed in Scilit:
- Data-driven approximation algorithms for rapid performance evaluation and optimization of civil structuresAutomation in Construction, 2016
- Convolutional Neural Networks for Steady Flow ApproximationPublished by Association for Computing Machinery (ACM) ,2016
- Geodesic Convolutional Neural Networks on Riemannian ManifoldsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2015
- Machine learning for neuroimaging with scikit-learnFrontiers in Neuroscience, 2014
- A Survey on Transfer LearningIEEE Transactions on Knowledge and Data Engineering, 2009
- Review of Metamodeling Techniques in Support of Engineering Design OptimizationJournal of Mechanical Design, 2006
- Surrogate-based analysis and optimizationProgress in Aerospace Sciences, 2005
- Structural optimization using evolution strategies and neural networksComputer Methods in Applied Mechanics and Engineering, 1998
- Spatial prediction and ordinary krigingMathematical Geology, 1988
- Numerical Procedures for Surface Fitting of Scattered Data by Radial FunctionsSIAM Journal on Scientific and Statistical Computing, 1986