Computational Technology Resources - CCC

B. Vuillod^1,2, M. Zani², L. Hallo¹, E. Panettieri² and M. Montemurro²

¹French Atomic Energy Commission, Le Barp, France
²Universite de Bordeaux, Arts et Metiers Institute of Technology, CNRS, INRA, Bordeaux INP, HESAM Universite, I2M UMR, Talence, France

doi:10.4203/ccc.5.1.10

B. Vuillod, M. Zani, L. Hallo, E. Panettieri, M. Montemurro, "A surrogate model based on NURBS entities for engineering problems", in P. Iványi, J. Logo, B.H.V. Topping, (Editors), "Proceedings of the Sixth International Conference on Soft Computing, Machine Learning and Optimisation in Civil, Structural and Environmental Engineering", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 5, Paper 1.10, 2023, doi:10.4203/ccc.5.1.10

Keywords: metamodel, gradient-based optimisation, fitting, NURBS hyper-surfaces, multiple-input-multiple-output, resources gain.

Abstract

Surrogate models are increasingly used in many sectors due to their ability to reproduce structural/system responses starting from numerical or experimental results. The main goal of a surrogate model is to preserve the same accuracy as the original model (within a certain interval) by considerably reducing the computational cost and, possibly, the required resources. Among the methods available in the literature, the one proposed in this article is based on Non-Uniform Rational Basis Spline (NURBS) entities. In this context, these entities appear promising, as they are continuous, versatile, able to adapt to Multiple- Input-Multiple-Output problems, and can be modified locally without impacting the precision of the metamodel elsewhere in the definition domain, i.e., possess capabilities for local support. Conversely, the off-line tasks that allows generating the NURBS-based surrogate model can be relatively heavy. In this paper, we propose an optimisation strategy for reducing the amount of data required to drive the NURBS metamodel while still maintaining a good accuracy level.

download the full-text of this paper (PDF, 12 pages, 523 Kb)

go to the previous paper
go to the next paper
return to the table of contents
return to the volume description

	Computational & Technology Resources an online resource for computational, engineering & technology publications
	not logged in - login
Front Page Browse CCC CCP CSETS CTR IJRT Other Authors Search Purchase Guide FAQ Contact us	Civil-Comp Conferences ISSN 2753-3239 CCC: 5 PROCEEDINGS OF THE SIXTH INTERNATIONAL CONFERENCE ON SOFT COMPUTING, MACHINE LEARNING AND OPTIMISATION IN CIVIL, STRUCTURAL AND ENVIRONMENTAL ENGINEERING Edited by: P. Iványi, J. Logo and B.H.V. Topping Paper 1.10 A surrogate model based on NURBS entities for engineering problems B. Vuillod^1,2, M. Zani², L. Hallo¹, E. Panettieri² and M. Montemurro² ¹French Atomic Energy Commission, Le Barp, France ²Universite de Bordeaux, Arts et Metiers Institute of Technology, CNRS, INRA, Bordeaux INP, HESAM Universite, I2M UMR, Talence, France doi:10.4203/ccc.5.1.10 Full Bibliographic Reference for this paper B. Vuillod, M. Zani, L. Hallo, E. Panettieri, M. Montemurro, "A surrogate model based on NURBS entities for engineering problems", in P. Iványi, J. Logo, B.H.V. Topping, (Editors), "Proceedings of the Sixth International Conference on Soft Computing, Machine Learning and Optimisation in Civil, Structural and Environmental Engineering", Civil-Comp Press, Edinburgh, UK, Online volume: CCC 5, Paper 1.10, 2023, doi:10.4203/ccc.5.1.10 Keywords: metamodel, gradient-based optimisation, fitting, NURBS hyper-surfaces, multiple-input-multiple-output, resources gain. Abstract Surrogate models are increasingly used in many sectors due to their ability to reproduce structural/system responses starting from numerical or experimental results. The main goal of a surrogate model is to preserve the same accuracy as the original model (within a certain interval) by considerably reducing the computational cost and, possibly, the required resources. Among the methods available in the literature, the one proposed in this article is based on Non-Uniform Rational Basis Spline (NURBS) entities. In this context, these entities appear promising, as they are continuous, versatile, able to adapt to Multiple- Input-Multiple-Output problems, and can be modified locally without impacting the precision of the metamodel elsewhere in the definition domain, i.e., possess capabilities for local support. Conversely, the off-line tasks that allows generating the NURBS-based surrogate model can be relatively heavy. In this paper, we propose an optimisation strategy for reducing the amount of data required to drive the NURBS metamodel while still maintaining a good accuracy level. download the full-text of this paper (PDF, 12 pages, 523 Kb) go to the previous paper go to the next paper return to the table of contents return to the volume description
Back to top	©Civil-Comp Limited 2023 - terms & conditions