Integrating deep learning and discrete cosine transform for surface waves full-waveform inversion

Интеграция глубокого обучения и дискретного косинусного преобразования для полноформатной инверсии поверхностных волн
Felipe Rincón, Sean Berti, Mattia Aleardi, A. Tognarelli, E. Stucchi
2024-11-08

Monte Carlo uncertainty estimationS-wave velocity modelsdeep learningdiscrete cosine transformsurface-wave full-waveform inversion
SUMMARY Accurate estimations of near-surface S-wave velocity (Vs) models hold particular significance in geological and engineering investigations. On the one hand, the popular multichannel analysis of surface waves (MASWs) is limited to the 1-D and the plane wave assumptions. On the other hand, the more advanced and computationally expensive full-waveform inversion (FWI) approach is often solved within a deterministic framework that hampers an accurate uncertainty assessment and makes the final predictions heavily reliant on the starting model. Here we combine deep learning with discrete cosine transform (DCT) to solve the FWI of surface waves and to efficiently estimate the inversion uncertainties. Our neural network approach effectively learns the inverse non-linear mapping between DCT-compressed seismograms and DCT-compressed S-velocity models. The incorporation of DCT into the deep learning framework provides several advantages: it notably reduces parameter space dimensionality and alleviates the ill-conditioning of the problem. Additionally, it decreases the complexity of the network architecture and the computational cost for the training phase compared to training in the full domain. A Monte Carlo simulation is also used to propagate the uncertainties from the data to the model space. We first test the implemented inversion method on synthetic data to showcase the generalization capabilities of the trained network and to explore the implications of incorrect noise assumptions in the recorded seismograms and inaccurate wavelet estimations. Further, we demonstrate the applicability of the implemented method to field data. In this case, available borehole information is used to validate our predictions. In both the synthetic and field applications, the predictions provided by the proposed method are compared with those of a deterministic FWI and the outcomes of a network trained in the full data and model spaces. Our experiments confirm that the implemented deep-learning inversion efficiently and successfully solves the FWI problem and yields more accurate and stable results than a network trained without the DCT compression. This opens the possibility to efficiently train a neural network that provides accurate instantaneous predictions of Vs near-surface models and related uncertainties.
1
DCT compression reduces parameter-space dimensionality, alleviates ill-conditioning, and lowers neural-network complexity and training cost compared with full-domain learning.
2
Monte Carlo simulations propagate data uncertainties into model-space uncertainty estimates, addressing limitations of deterministic full-waveform inversion.
3
Synthetic and field experiments compare the proposed method with deterministic FWI and full-domain networks, with field predictions validated against borehole information.
4
The method combines deep learning with discrete cosine transform (DCT) to invert surface-wave seismograms for near-surface S-wave velocity models.
5
The neural network learns the nonlinear mapping between DCT-compressed seismograms and DCT-compressed S-wave velocity models.

near-surface S-wave velocity models inferred from surface-wave seismograms

full-waveform inversion of surface waves, including uncertainty estimation and the effects of noise and wavelet assumptions

Publication Details
Publication Date
2024-11-08
Journal
Publisher
ISSN
Cited by
24
Access Type
Author Information
Authors
Felipe Rincón
Sean Berti
Mattia Aleardi
A. Tognarelli
E. Stucchi
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%