A Neural-network-based multiscale Hybridizable Discontinuous Galerkin method for solving PDEs in porous media

Abstract

We develop a neural-network-accelerated multiscale hybridizable discontinuous Galerkin method for elliptic problems with heterogeneous coefficients. The method preserves the standard MsHDG local-to-global structure: fine-scale HDG problems on coarse blocks define discrete Dirichlet-to-Neumann operators, which are assembled through the standard MsHDG global skeleton equations. To reduce the cost of constructing these local operators, we train a neural network on coefficient fields defined on a reference block and use the predicted operators in place of repeated fine-scale local solves. The numerical experiments assess both the accuracy and online efficiency of the resulting NN-MsHDG method. For two-dimensional binary permeability fields with moderate contrast, the neural method reproduces the standard MsHDG solution with relative modeling errors of a few percent while reducing the total online computational cost by factors of approximately 5 to 16, depending on the coarse trace dimension. In the high-contrast regime, the chosen polynomial coarse trace spaces already yield substantial discretization errors in the standard MsHDG method, indicating the need for more effective coarse spaces, such as coefficient-adapted spectral trace spaces. In addition, the learned local operators introduce modeling errors that become severe as the trace space is enriched. These results demonstrate the potential of neural surrogates for accelerating multiscale HDG computations while also highlighting the need for improved operator representations and a better understanding of error amplification in high-contrast problems.

Hybridizable Discontinuous Galerkin methods for the Keller-Segel model
Silvia Bertoluzza, Gabriella Bretti, Micol Pennacchio +1
arXiv28 Oct 2026PreprintarXiv licenceFull textNumerical analysis & scientific computingarXiv
A Hybridizable Discontinuous Galerkin Method for Time-Modulated Drude--Lorentz Media
Jeonghun J. Lee
arXiv15 Oct 2026PreprintCC BY 4.0Full textNumerical analysis & scientific computingarXiv
A Massively Parallel Hybridizable Discontinuous Galerkin Solver for Direct Numerical Simulation of Compressible Flows on GPUs
Andrew Welter, Thea Collin, Ngoc Cuong Nguyen +1
arXiv28 Aug 2026PreprintarXiv licenceFull textNumerical analysis & scientific computing Fluid dynamicsarXiv
The discontinuous Galerkin method for solving the two-dimensional transport equation with distributed-order memory kernel on triangular meshes
Leilei Wei, Yanping Chen, Fangfang Qin +2
Applied Numerical Mathematics17 May 2026Peer-reviewedLicence not stated8 citedGeneral mathematicsOpenAlex
A Local Discontinuous Galerkin method for the modified Camassa--Holm equation
Xiang-Ke Chang, Yong Liu, Qi Tao
arXiv8 Sep 2026PreprintCC BY 4.0Full textNumerical analysis & scientific computingarXiv

Cite

BibTeX RIS

BibTeX

@misc{haines2026neuralnetworkbased,
  title = {{A Neural-network-based multiscale Hybridizable Discontinuous Galerkin method for solving PDEs in porous media}},
  author = {Tony Haines and Ke Shi},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2608.25850},
  eprint = {2608.25850},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2608.25850},
}

RIS

TY  - UNPB
TI  - A Neural-network-based multiscale Hybridizable Discontinuous Galerkin method for solving PDEs in porous media
AU  - Tony Haines
AU  - Ke Shi
PY  - 2026
DA  - 2026/09/04
PB  - arXiv
DO  - 10.48550/arxiv.2608.25850
UR  - https://arxiv.org/abs/2608.25850
AB  - We develop a neural-network-accelerated multiscale hybridizable discontinuous Galerkin method for elliptic problems with heterogeneous coefficients. The method preserves the standard MsHDG local-to-global structure: fine-scale HDG problems on coarse blocks define discrete Dirichlet-to-Neumann operators, which are assembled through the standard MsHDG global skeleton equations. To reduce the cost of constructing these local operators, we train a neural network on coefficient fields defined on a reference block and use the predicted operators in place of repeated fine-scale local solves. The numerical experiments assess both the accuracy and online efficiency of the resulting NN-MsHDG method. For two-dimensional binary permeability fields with moderate contrast, the neural method reproduces the standard MsHDG solution with relative modeling errors of a few percent while reducing the total online computational cost by factors of approximately 5 to 16, depending on the coarse trace dimension. In the high-contrast regime, the chosen polynomial coarse trace spaces already yield substantial discretization errors in the standard MsHDG method, indicating the need for more effective coarse spaces, such as coefficient-adapted spectral trace spaces. In addition, the learned local operators introduce modeling errors that become severe as the trace space is enriched. These results demonstrate the potential of neural surrogates for accelerating multiscale HDG computations while also highlighting the need for improved operator representations and a better understanding of error amplification in high-contrast problems.
ER  -

Abstract

Related

Cite