Physics-based surrogate modeling of parameterized PDEs for optimization and uncertainty analysis
Physics-based surrogate modeling of parameterized PDEs for optimization and uncertainty analysis
This paper presents physics-based surrogate modeling algorithms for systems governed by parameterized partial differential equations (PDEs) commonly encountered in design optimization and uncertainty analysis. We first outline unsupervised learning approaches that leverage advances in the machine learning literature for a meshfree solution of PDEs. Subsequently, we propose continuum and discrete formulations for systems governed by parameterized steady-state PDEs. We consider the case of both deterministically and randomly parameterized systems. The basic idea is to embody the design variables or uncertain parameters in additional dimensions of the governing PDEs along with the spatial coordinates. We show that the undetermined parameters of the surrogate model can be estimated by minimizing a physics-based objective function derived using a multidimensional least-squares collocation or the Bubnov-Galerkin scheme. This potentially allows us to construct surrogate models without using data from computer experiments on a deterministic analysis code. Finally, we also outline an extension of the present approach to directly approximate the density function of random algebraic equations.
AIAA-2002-1586
American Institute of Aeronautics and Astronautics
Nair, P.B.
d4d61705-bc97-478e-9e11-bcef6683afe7
2002
Nair, P.B.
d4d61705-bc97-478e-9e11-bcef6683afe7
Nair, P.B.
(2002)
Physics-based surrogate modeling of parameterized PDEs for optimization and uncertainty analysis.
In Proceedings of 43rd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference.
American Institute of Aeronautics and Astronautics.
11 pp
.
(AIAA-2002-1586).
Record type:
Conference or Workshop Item
(Paper)
Abstract
This paper presents physics-based surrogate modeling algorithms for systems governed by parameterized partial differential equations (PDEs) commonly encountered in design optimization and uncertainty analysis. We first outline unsupervised learning approaches that leverage advances in the machine learning literature for a meshfree solution of PDEs. Subsequently, we propose continuum and discrete formulations for systems governed by parameterized steady-state PDEs. We consider the case of both deterministically and randomly parameterized systems. The basic idea is to embody the design variables or uncertain parameters in additional dimensions of the governing PDEs along with the spatial coordinates. We show that the undetermined parameters of the surrogate model can be estimated by minimizing a physics-based objective function derived using a multidimensional least-squares collocation or the Bubnov-Galerkin scheme. This potentially allows us to construct surrogate models without using data from computer experiments on a deterministic analysis code. Finally, we also outline an extension of the present approach to directly approximate the density function of random algebraic equations.
Text
nair_02d.pdf
- Accepted Manuscript
More information
Published date: 2002
Venue - Dates:
43rd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, Denver, Colorado, USA, 2002-04-22 - 2002-04-25
Identifiers
Local EPrints ID: 21992
URI: http://eprints.soton.ac.uk/id/eprint/21992
DOI: AIAA-2002-1586
PURE UUID: 8e21abc3-a248-40e6-ba09-b2676ebf545e
Catalogue record
Date deposited: 29 Mar 2006
Last modified: 15 Mar 2024 06:34
Export record
Altmetrics
Contributors
Author:
P.B. Nair
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics