BAYESIAN DEEP OPERATOR LEARNING FOR HOMOGENIZED TO FINE-SCALE MAPS FOR MULTISCALE PDE
Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review
Author(s)
Related Research Unit(s)
Detail(s)
Original language | English |
---|---|
Pages (from-to) | 956-972 |
Number of pages | 17 |
Journal / Publication | Multiscale Modeling & Simulation |
Volume | 22 |
Issue number | 3 |
Online published | 17 Jul 2024 |
Publication status | Published - Sept 2024 |
Link(s)
Abstract
We present a new framework for computing fine-scale solutions of multiscale partial differential equations (PDEs) using operator learning tools. Obtaining fine-scale solutions of multiscale PDEs can be challenging, but there are many inexpensive computational methods for obtaining coarse-scale solutions. Additionally, in many real-world applications, fine-scale solutions can only be observed at a limited number of locations. In order to obtain approximations or predictions of fine-scale solutions over general regions of interest, we propose to learn the operator mapping from coarse-scale solutions to fine-scale solutions using observations of a limited number of (possible noisy) fine-scale solutions. The approach is to train multi-fidelity homogenization maps using mathematically motivated neural operators. The operator learning framework can efficiently obtain the solution of multiscale PDEs at any arbitrary point, making our proposed framework a mesh-free solver. We verify our results on multiple numerical examples showing that our approach is an efficient mesh-free solver for multiscale PDEs. © 2024 by SIAM.
Research Area(s)
- neural operator, neural homogenization, multiscale finite element method, dis- cretization invariant, multi-fidelity
Citation Format(s)
BAYESIAN DEEP OPERATOR LEARNING FOR HOMOGENIZED TO FINE-SCALE MAPS FOR MULTISCALE PDE. / ZHANG, Zecheng; MOYA, Christian; LEUNG, Wing Tat et al.
In: Multiscale Modeling & Simulation, Vol. 22, No. 3, 09.2024, p. 956-972.
In: Multiscale Modeling & Simulation, Vol. 22, No. 3, 09.2024, p. 956-972.
Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review