1887

Abstract

Summary

The manuscript discusses advancements in Full Waveform Inversion (FWI) for seismic imaging, focusing on reducing computational burdens through machine learning (ML). It introduces an m-step autoregressive neural operator framework for efficient simulation of forward and adjoint wavefields in 2D acoustic media. The study compares single-step and autoregressive approaches, highlighting the latter’s superior accuracy and temporal consistency. Using Fourier Neural Operators (FNO) and U-shaped FNO (UFNO), the research demonstrates significant reductions in computational time and improved performance across SEAM-type velocity models. Benchmarking results show that while UFNO achieves lower training loss, FNO offers better throughput and GPU memory efficiency, making it the preferred model for geophysical applications.

© EAGE Publications BV
Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.2025643025
2025-10-06
2026-02-15

  • From This Site

    /content/papers/10.3997/2214-4609.2025643025
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

References

  1. Virieux, J., and Operto, S., 2009, An overview of full-waveform inversion in exploration geophysics: Geophysics, 74, no. 6, WCC1–WCC26.
     [Google Scholar]
  2. Herrmann, F. J., Witte, P. A., and Li, X., 2019, Learned iterative solvers for full-waveform inversion: SEG Technical Program Expanded Abstracts, 1442–1446.
     [Google Scholar]
  3. Li, Z., Kovachki, N., Azizzadenesheli, K., and Anandkumar, A., 2020, Fourier Neural Operator for Parametric Partial Differential Equations: arXiv preprint arXiv:2010.08895.
     [Google Scholar]
  4. Zhang, X., and Alkhalifah, T., 2023, Deep learning for wavefield simulation: Geophysics, 88, no. 1, T1–T12.
     [Google Scholar]
  5. Brandstetter, J., Worrall, D. E., and Welling, M., 2022, Message passing neural PDE solvers: ICLR.
     [Google Scholar]
  6. Li, Z., Azizzadenesheli, K., Bhattacharya, K., and Anandkumar, A., 2021, Fourier Neural Operator for Parametric Partial Differential Equations: ICLR.
     [Google Scholar]
/content/papers/10.3997/2214-4609.2025643025
Loading
Auto-regressive Neural Operators for Efficient Wavefield Simulation
Conference Proceedings 2025, 1 (2025); https://doi.org/10.3997/2214-4609.2025643025
/content/papers/10.3997/2214-4609.2025643025
/content/papers/10.3997/2214-4609.2025643025
Loading

Data & Media loading...

Most Cited This Month Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error