Deep Neural Helmholtz Operators for 3D Elastic Wave Propagation and Inversion Article Swipe

MAKE IMPACT

Caifeng Zou , Kamyar Azizzadenesheli , Zachary E. Ross , Robert W. Clayton · YOU? · · 2023 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2311.09608

Numerical simulations of seismic wave propagation in heterogeneous 3D media are central to investigating subsurface structures and understanding earthquake processes, yet are computationally expensive for large problems. This is particularly problematic for full waveform inversion, which typically involves numerous runs of the forward process. In machine learning there has been considerable recent work in the area of operator learning, with a new class of models called neural operators allowing for data-driven solutions to partial differential equations. Recent works in seismology have shown that when neural operators are adequately trained, they can significantly shorten the compute time for wave propagation. However, the memory required for the 3D time domain equations may be prohibitive. In this study, we show that these limitations can be overcome by solving the wave equations in the frequency domain, also known as the Helmholtz equations, since the solutions for a set of frequencies can be determined in parallel. The 3D Helmholtz neural operator is 40 times more memory-efficient than an equivalent time-domain version. We employ a U-shaped neural operator for 2D and 3D elastic wave modeling, achieving two orders of magnitude acceleration compared to a baseline spectral element method. The neural operator accurately generalizes to variable velocity structures and can be evaluated on denser input meshes than used in the training simulations. We also show that when solving for wavefields strictly on the surface, the accuracy can be significantly improved via a graph neural operator layer. In leveraging automatic differentiation, the proposed method can serve as an alternative to the adjoint-state approach for 3D full-waveform inversion, reducing the computation time by a factor of 350.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2311.09608

PDF

https://arxiv.org/pdf/2311.09608

OA Status

green

Cited By

2

Related Works

10

OpenAlex ID

https://openalex.org/W4388787750

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4388787750

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2311.09608

Digital Object Identifier

Title

Deep Neural Helmholtz Operators for 3D Elastic Wave Propagation and Inversion

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2023

Year of publication

Publication date

2023-11-16

Full publication date if available

Authors

Caifeng Zou, Kamyar Azizzadenesheli, Zachary E. Ross, Robert W. Clayton

List of authors in order

Landing page

https://arxiv.org/abs/2311.09608

Publisher landing page

PDF URL

https://arxiv.org/pdf/2311.09608

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2311.09608

Direct OA link when available

Concepts

Helmholtz equation, Helmholtz free energy, Operator (biology), Artificial neural network, Computer science, Polygon mesh, Inversion (geology), Algorithm, Wave equation, Wave propagation, Mathematical analysis, Mathematics, Artificial intelligence, Geology, Physics, Optics, Boundary value problem, Chemistry, Transcription factor, Structural basin, Computer graphics (images), Paleontology, Quantum mechanics, Gene, Biochemistry, Repressor

Top concepts (fields/topics) attached by OpenAlex

Cited by

2

Total citation count in OpenAlex

Citations by year (recent)

2024: 2

Per-year citation counts (last 5 years)

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload