Migration images guided high-resolution velocity modeling based on fully convolutional neural network
), Abstract
Current data-driven deep learning（DL）methods typically reconstruct subsurface velocity models directly from pre-stack seismic records. However, these purely data-driven methods are often less robust and produce results that are less physically interpretative. Here, the authors propose a new method that uses migration images as input, combined with convolutional neural networks to construct high-resolution velocity models. Compared to directly using pre-stack seismic records as input, the nonlinearity between migration images and velocity models is significantly reduced. Additionally, the advantage of using migration images lies in its ability to more comprehensively capture the reflective properties of the subsurface medium, including amplitude and phase information, thereby to provide richer physical information in guiding the reconstruction of the velocity model. This approach not only improves the accuracy and resolution of the reconstructed velocity models, but also enhances the physical interpretability and robustness. Numerical experiments on synthetic data show that the proposed method has superior reconstruction performance and strong generalization capability when dealing with complex geological structures, and shows great potential in providing efficient solutions for the task of reconstructing high-wavenumber components.
References
Alkhalifah T, Plessix R-É. 2014. A recipe for practical full-waveform inversion in anisotropic media: An analytical parameter resolution study. Geophysics, 79(3): R91-R101.
Corrales M, Hoteit H, Ravasi M. 2024. Seis2rock: A data‐ driven approach to direct petrophysical inversion of pre‐ stack seismic data. Earth and Space Science, 11(2): e2023EA003301.
Das V, Pollack A, Wollner U, et al. 2019. Convolutional neural network for seismic impedance inversion. Geophysics, 84(6): R869-R880.
Downton J, Collet O, Hampson D. 2020. Theory-guided data science-based reservoir prediction of a north sea oil field. Leading Edge, 39(10): 689-764.
Du M, Cheng S J, Mao W J, et al. 2022. Deep-learning-based seismic variable-size velocity model building. IEEE Geoscience and Remote Sensing Letters, 19: 3008305.
Estiri H, Stephens K A, Klann J G, et al. 2018. Exploring completeness in clinical data research networks with DQe-c. Journal of the American Medical Informatics Association, 25(1): 17-24.
Fabien-Ouellet G, Sarkar R. 2020. Seismic velocity estimation: A deep recurrent neural-network approach. Geophysics, 85(1): U21-U29.
Kim Y, Nakata N. 2018. Geophysical inversion versus machine learning in inverse problems. The Leading Edge, 37(12): 894-901.
Lecun Y, Bengio Y, Hinton G. 2015. Deep learning. Nature, 521: 436-444.
Li S C, Liu B, Ren Y X, et al. 2020. Deep-learning inversion of seismic data. IEEE Transactions on Geoscience and Remote Sensing, 58(3): 2135-2149.
Lin Y Z, Theiler J, Wohlberg B. 2023. Physics-guided data-driven seismic inversion: Recent progress and future opportunities in full-waveform inversion. IEEE Signal Processing Magazine, 40(1): 115-133.
Liu B, Ren Y X, Liu H C, et al. 2021. GPRInvNet: Deep learning-based ground-penetrating radar data inversion for tunnel linings. IEEE Transactions on Geoscience and Remote Sensing, 59(10): 8305-8325.
Liu Y, Herrmann F J, Moghaddam P P. 2020. SeisDeNet: An intelligent seismic data denoising network for the internet of things. Journal of Cloud Computing: Advances, Systems and Applications, 9(1): 42.
Lu C, Zhang C. 2023. Seismic velocity inversion via physical embedding recurrent neural networks (RNN). Applied Sciences, 13(24): 13312.
Lu L, Shin Y, Su Y, et al. 2020. Dying ReLU and initialization: Theory and numerical examples. Communications in Computational Physics, 28: 1671-1697.
Mosser L, Dubrule O, Blunt M J. 2020. Stochastic seismic waveform inversion using generative adversarial networks as a geological prior. Mathematical Geosciences, 52: 53-79.
Ovcharenko O, Kazei V, Alkhalifah T A, et al. 2022. Multi-task learning for low-frequency extrapolation and elastic model building from seismic data. IEEE Transactions on Geoscience and Remote Sensing, 60: 4510717.
Rasht-Behesht M, Huber C, Shukla K, et al. 2022. Physics-Informed Neural Networks (PINNs) for wave propagation and full waveform inversions. Journal of Geophysical Research: Solid Earth, 127(5): e2021JB023120.
Reichstein M, Camps-Valls G, Stevens B, et al. 2019. Deep learning and process understanding for data-driven Earth system science. Nature, 566: 195-204.
Robertsson J O A, Laws R, Chapman C, et al. 2006. Modelling of scattering of seismic waves from a corrugated rough sea surface: A comparison of three methods. Geophysical Journal International, 167(1): 70-76.
Schmidhuber J. 2015. Deep learning in neural networks: An overview. Neural Networks, 61: 85-117.
Sun J, Niu Z, Innanen K A, et al. 2020. A theory-guided deep-learning formulation and optimization of seismic waveform inversion. Geophysics, 85(2): R87-R99.
Sun S L, Mao W J, Ouyang W, et al. 2022. Amplitude-preserving imaging condition for scattering-based RTM in acoustic VTI media. IEEE Geoscience and Remote Sensing Letters, 19: 3009005.
Taleb I, Serhani M A, Bouhaddioui C, et al. 2021. Big data quality framework: A holistic approach to continuous quality management. Journal of Big Data, 8(1): 76.
Wang W, Ma J W. 2019. Velocity model building in a crosswell acquisition geometry with image-trained artificial neural networks. Geophysics, 85(2): U31-U46.
Wang Y J, Zong J J, Wang L J, et al. 2023. Physics-driven cycle network for seismic impedance inversion using conditional generative adversarial networks. Journal of Geophysics and Engineering, 20(6): 1337-1347.
Wu B Y, Meng D L, Wang L L, et al. 2020. Seismic impedance inversion using fully convolutional residual network and transfer learning. IEEE Geoscience and Remote Sensing Letters, 17(12): 2140-2144.
Wu X M, Ma J W, Si X, et al. 2023. Sensing prior constraints in deep neural networks for solving exploration geophysical problems. Proceedings of the National Academy of Sciences, 120(23): e2219573120.
Wu Y, Lin Y Z. 2020. InversionNet: An efficient and accurate data-driven full waveform inversion. IEEE Transactions on Computational Imaging, 6: 419-433.
Yang F S, Ma J W. 2019. Deep-learning inversion: A next-generation seismic velocity model building method. Geophysics, 84(4): R583-R599.
Yin Y C, Han L G, Zhang P, et al. 2023. First-break picking of large-offset seismic data based on CNNs with weighted data. Remote Sensing, 15(2): 356.
Yuan C, Zhang X, Jia X, et al. 2020. Time-lapse velocity imaging via deep learning. Geophysical Journal International, 220(2): 1228-1241.
Zhu W Q, Biondi E, Li J X, et al. 2023. Seismic arrival-time picking on distributed acoustic sensing data using semi-supervised learning. Nature Communications, DOI: 10.48550/arXiv.2302.08747
京公网安备11010802044758号