Based on the combination ofdata- and model-driven approaches, this study expanded the labels of the training set through model inversion results, and added the model inversion objective function to the deep learning algorithm. By constructing a new loss function, this study proposed a seismic impedance optimization inversion method combining model inversion with deep learning inversion. The semi-supervised deep learning network inversion under a pseudo-label was achieved using the RNN network structure. The network inversion results were used as the initial model to participate in the model inversion. The final optimization inversion was completed by continuous iterative optimization of both network and model inversion. The method proposed in this study proves to possess high inversion accuracy and practicability, as demonstrated by the synthesis of the Marmousi model and the actual data.
Chen Y K, Chen H M, Xiang K, et al. Geological structure guided well log interpolation for high-fidelity full waveform inversion[J]. Geophysical Journal International, 2016, 207(2):1313-1331.
Jiang X D, Zhang W, Yang H. The research on Bayesian inference for geophysical inversion[J]. Reviews of Geophysics and Planetary Physics, 2022, 53(2):159-171.
Xing W J, Cao S Y, Chen S Y, et al. Post-stack P-wave impedance inversion based on spectral inversion[J]. Geophysical and Geochemical Exploration, 2023, 47(2):429-437.
Fan Y H, Liu X F, Chen X F. The quasi Newton method in the inversion of the dispersion curve of Rayleigh wave in multilayered media[J]. Geophysical and Geochemical Exploration, 2006, 30(5):456-459.
[6]
Reichstein M, Camps-Valls G, Stevens B, et al. Deep learning and process understanding for data-driven Earth system science[J]. Nature, 2019, 566:195-204.
[7]
Yu S W, Ma J W. Deep learning for geophysics:Current and future trends[J]. Reviews of Geophysics, 2021, 59(3):1-36.
Xiao L Z. The fusion of data-driven machine learning with mechanism models and interpretability issues[J]. Geophysical Prospecting for Petroleum, 2022, 61(2):205-212.
doi: 10.3969/j.issn.1000-1441.2022.02.002
[9]
Das V, Pollack A, Wollner U, et al. Convolutional neural network for seismic impedance inversion[J]. Geophysics, 2019, 84(6):R869-R880.
doi: 10.1190/GEO2018-0838.1
[10]
Alfarraj M, AlRegib G. Semisupervised sequence modeling for elastic impedance inversion[J]. Interpretation, 2019, 7(3):SE237-SE249.
[11]
Biswas R, Sen M K, Das V, et al. Prestack and poststack inversion using a physics-guided convolutional neural network[J]. Interpretation, 2019, 7(3):SE161-SE174.
[12]
Luo R, Chen H Z, Wang B F. Semisupervised seismic impedance inversion with data augmentation and uncertainty analysis[J]. Geophysics, 2023, 88(4):M213-M224.
[13]
Su Y Q, Cao D P, Liu S Y, et al. Seismic impedance inversion based on deep learning with geophysical constraints[J]. Geoenergy Science and Engineering, 2023, 225:211671.
Wang R, Xiong J, Liu Q, et al. Inversion of gravity anomalies based on a deep neural network[J]. Geophysical and Geochemical Exploration, 2022, 46(2):451-458.
Wu S, Ning X B, Yang T W, et al. Neural network-based denoising for ground-penetrating radar data[J]. Geophysical and Geochemical Exploration, 2023, 47(5):1298-1306.
Bian B L. Research and application of reservoir prediction method of channel sands based on high density three-dimensional seismic data[D]. Chengdu: Chengdu University of Technology, 2019.
