Accelerating wave simulations with neural dispersion correctors

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Rincón, Felipe, Fichtner, Andreas, Aleardi, Mattia, Tognarelli, Andrea, Stucchi, Eusebio
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918156458524672
author Rincón, Felipe
Fichtner, Andreas
Aleardi, Mattia
Tognarelli, Andrea
Stucchi, Eusebio
author_facet Rincón, Felipe
Fichtner, Andreas
Aleardi, Mattia
Tognarelli, Andrea
Stucchi, Eusebio
contents We present a Fourier neural operator network, designed to correct dispersion errors in numerical wave simulations. The neural dispersion corrector enables the replacement of a computationally expensive high-accuracy simulation by a less expensive low-accuracy simulation. In contrast to neural network surrogates that fully replace a wave equation, the neural dispersion corrector has only a weak dependence on the distribution of model parameters, such as wave speeds. Consequently, the network can be trained with a significantly smaller dataset, while still generalising to unseen input parameters. Following a description of the network architecture and training, we provide examples for the 3-D elastic wave equation. After training with merely 1$\,$000 examples on one GPU, the neural corrector achieves a speed-up of 16$\times$ compared to a reference spectral-element simulation and a generalisation to a broad range of strongly heterogeneous wave speed distributions.
format Preprint
id arxiv_https___arxiv_org_abs_2510_06881
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Accelerating wave simulations with neural dispersion correctors
Rincón, Felipe
Fichtner, Andreas
Aleardi, Mattia
Tognarelli, Andrea
Stucchi, Eusebio
Geophysics
65M99, 86A15, 35Q68
I.6.8; G.1.8; J.2
We present a Fourier neural operator network, designed to correct dispersion errors in numerical wave simulations. The neural dispersion corrector enables the replacement of a computationally expensive high-accuracy simulation by a less expensive low-accuracy simulation. In contrast to neural network surrogates that fully replace a wave equation, the neural dispersion corrector has only a weak dependence on the distribution of model parameters, such as wave speeds. Consequently, the network can be trained with a significantly smaller dataset, while still generalising to unseen input parameters. Following a description of the network architecture and training, we provide examples for the 3-D elastic wave equation. After training with merely 1$\,$000 examples on one GPU, the neural corrector achieves a speed-up of 16$\times$ compared to a reference spectral-element simulation and a generalisation to a broad range of strongly heterogeneous wave speed distributions.
title Accelerating wave simulations with neural dispersion correctors
topic Geophysics
65M99, 86A15, 35Q68
I.6.8; G.1.8; J.2
url https://arxiv.org/abs/2510.06881