AI Chat Paper
Note: Please note that the following content is generated by AMiner AI. SciOpen does not take any responsibility related to this content.
{{lang === 'zh_CN' ? '文章概述' : 'Summary'}}
{{lang === 'en_US' ? '中' : 'Eng'}}
Chat more with AI
Powered by AMiner. Clicking this button will take you away from SciOpen.
Home Petroleum Science Article
PDF (7.7 MB)
Cite
EndNote(RIS) BibTeX
Collect
Submit Manuscript AI Chat Paper
Show Outline
Outline
Show full outline
Hide outline
Outline
Show full outline
Hide outline
Original Paper | Open Access

Probabilistic seismic inversion based on physics-guided deep mixture density network

Qian-Hao Suna,b, Zhao-Yun Zonga,b( ), Xin Lic
National Key Laboratory of Deep Oil and Gas, China University of Petroleum (East China), Qingdao, 266580, Shandong, China
Laoshan Laboratory, Qingdao, 266580, Shandong, China
CNOOC Research Institute Ltd., Beijing, 100028, China

Edited by Jie Hao and Meng-Jiao Zhou

Show Author Information

Abstract

Deterministic inversion based on deep learning has been widely utilized in model parameters estimation. Constrained by logging data, seismic data, wavelet and modeling operator, deterministic inversion based on deep learning can establish nonlinear relationships between seismic data and model parameters. However, seismic data lacks low-frequency and contains noise, which increases the non-uniqueness of the solutions. The conventional inversion method based on deep learning can only establish the deterministic relationship between seismic data and parameters, and cannot quantify the uncertainty of inversion. In order to quickly quantify the uncertainty, a physics-guided deep mixture density network (PG-DMDN) is established by combining the mixture density network (MDN) with the deep neural network (DNN). Compared with Bayesian neural network (BNN) and network dropout, PG-DMDN has lower computing cost and shorter training time. A low-frequency model is introduced in the training process of the network to help the network learn the nonlinear relationship between narrowband seismic data and low-frequency impedance. In addition, the block constraints are added to the PG-DMDN framework to improve the horizontal continuity of the inversion results. To illustrate the benefits of proposed method, the PG-DMDN is compared with existing semi-supervised inversion method. Four synthetic data examples of Marmousi Ⅱ model are utilized to quantify the influence of forward modeling part, low-frequency model, noise and the pseudo-wells number on inversion results, and prove the feasibility and stability of the proposed method. In addition, the robustness and generality of the proposed method are verified by the field seismic data.

Keywords

Deep learningProbabilistic inversionPhysics-guidedDeep mixture density network

Electronic Supplementary Material

Download File(s)
petsci-21-3-1611_ESM.pdf (726.3 KB)

References

 

Alfarraj, M., AlRegib, G., 2019. Semisupervised sequence modeling for elastic impedance inversion. Interpretation 7 (3), SE237-SE249. https://doi.org/10.1190/Int-2018-0250.1.
Crossref Google Scholar
 
Bishop, C.M., 1994. Mixture Density Networks. Neural Computing Research Group Technical Report NCRG/94/004. Aston University.
 

Biswas, R., Sen, M.K., Das, V., Mukerji, T., 2019. Prestack and poststack inversion using a physics-guided convolutional neural network. Interpretation 7 (3), SE161-SE174. https://doi.org/10.1190/INT-2018-0236.1.
Crossref Google Scholar
 

Buland, A., More, H., 2003. Bayesian linearized AVO inversion. Geophysics 68 (1), 185-198. https://doi.org/10.1190/1.1543206.
Crossref Google Scholar
 

Cao, D., Su, Y., Cui, R., 2022. Multi-parameter pre-stack seismic inversion based on deep learning with sparse reflection coefficient constraints. J. Petrol. Sci. Eng. 209, 109836. https://doi.org/10.1016/j.petrol.2021.109836.
Crossref Google Scholar
 

Chen, Y., Saygin, E., 2021. Seismic inversion by hybrid machine learning. J. Geophys. Res. Solid Earth 126 (9), 126. https://doi.org/10.1029/2020JB021589.
Crossref Google Scholar
 

Cho, K., Merrienboer, B.V., Gulcehre, C., Hdanau, D.B., Bougares, F., Schwenk, H., Bengio, Y., 2014. Learning phrase representations using RNN encoder-decoder for statistical machine translation. Computer Science. https://doi.org/10.3115/v1/D14-1179.
Crossref Google Scholar
 

Das, V., Mukerji, T., 2020. Petrophysical properties prediction from prestack seismic data using convolutional neural networks. Geophysics 85 (5), N41-N55. https://doi.org/10.1190/GEO2019-0650.1.
Crossref Google Scholar
 

Das, V., Pollack, A., Wollner, U., Mukerji, T., 2019. Convolutional neural network for seismic impedance inversion. Geophysics 84 (6), R869-R880. https://doi.org/10.1190/GEO2018-0838.1.
Crossref Google Scholar
 

Di, H., Abubakar, A., 2021. Estimating subsurface properties using a semisupervised neural network approach. Geophysics 87 (1), IM1-IM10. https://doi.org/10.1190/geo2021-0192.1.
Crossref Google Scholar
 

Downton, J.E., 2005. Seismic Parameter Estimation from AVO Inversion. University of Calgary.
 

Earp, S., Curtis, A., 2020. Probabilistic neural network-based 2D travel-time tomography. Neural Comput. Appl. 32 (22), 17077-17095. https://doi.org/10.1007/s00521-020-04921-8.
Crossref Google Scholar
 

Earp, S., Curtis, A., Zhang, X., Hansteen, F., 2020. Probabilistic neural network tomography across Grane field (North Sea) from surface wave dispersion data. Geophys. J. Int. 223 (3), 1741-1757. https://doi.org/10.1093/gji/ggaa328.
Crossref Google Scholar
 

Feng, R., Grana, D., Balling, N., 2021. Variational inference in Bayesian neural network for well-log prediction. Geophysics 86 (3), M91-M99. https://doi.org/10.1190/geo2020-0609.1.
Crossref Google Scholar
 

Gao, L., Chen, P.Y., Yu, S., 2016. Demonstration of convolution kernel operation on resistive cross-point array. IEEE Electron. Device Lett. 37 (7), 870-873. https://doi.org/10.1109/LED.2016.2573140.
Crossref Google Scholar
 

Gao, Z., Li, C., Zhang, B., Jiang, X., Xu, Z., 2020. Building large-scale density model via a deep learning based data-driven method. Geophysics 86 (1), M1-M15. https://doi.org/10.1190/geo2019-0332.1.
Crossref Google Scholar
 
Ge, M., Wang, W., Zheng, W., 2021. Semi-supervised Impedance Inversion by Bayesian Neural Network Based on 2-d CNN Pre-training. https://doi.org/10.48550/arXiv.2111.10596.
 

Gholami, A., 2015. Nonlinear multichannel impedance inversion by total-variation regularization. Geophysics 80 (5), R217-R224. https://doi.org/10.1190/geo2015-0004.1.
Crossref Google Scholar
 

Gholami, A., 2016. A fast automatic multichannel blind seismic inversion for high-resolution impedance recovery. Geophysics 81 (5), V357-V364. https://doi.org/10.1190/GEO2015-0654.1.
Crossref Google Scholar
 

Grana, D., 2016. Bayesian linearized rock-physics inversion. Geophysics 81 (6), D625-D641. https://doi.org/10.1190/geo2016-0161.1.
Crossref Google Scholar
 

Grana, D., Rossa, E.D., 2010. Probabilistic petrophysical-properties estimation integrating statistical rock physics with seismic inversion. Geophysics 75 (3), O21-O37. https://doi.org/10.1190/1.3386676.
Crossref Google Scholar
 

Hamid, H., Pidlisecky, A., 2015. Multitrace impedance inversion with lateral constraints. Geophysics 80 (6), M101-M111. https://doi.org/10.1190/geo2014-0546.1.
Crossref Google Scholar
 

Hochreiter, S., Schmidhuber, J., 1997. Long short-term memory. Neural Comput. 9 (8), 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735.
Crossref Google Scholar
 
Hu, W., Jin, Y., Wu, X., Chen, J., 2020. Physics-guided self-supervised learning for low frequency data prediction in FWI. SEG Technical Program Expanded Abstracts, pp. 875-879. https://doi.org/10.1190/segam2020-3423396.1.
Crossref
 

Ioffe, S., Szegedy, C., 2015. Batch normalization: accelerating deep network training by reducing internal covariate shift. International conference on machine learning, PMLR, 448-456. https://doi.org/10.48550/arXiv.1502.03167.
Crossref Google Scholar
 

Junhwan, C., Seokmin, O., Joongmoo, B., 2022. Uncertainty estimation in AVO inversion using Bayesian dropout based deep learning. J. Petrol. Sci. Eng. 208, 109288. https://doi.org/10.1016/j.petrol.2021.109288.
Crossref Google Scholar
 

Kingma, D., Ba, J., 2014. Adam: a method for stochastic optimization. Computer Science. https://doi.org/10.48550/arXiv.1412.6980.
Crossref Google Scholar
 

Lecun, Y., Bottou, L., Bengio, Y., Haffner, P., 1998. Gradient-based learning applied to document recognition. Proc. IEEE 86 (11), 2278-2324. https://doi.org/10.1109/5.726791.
Crossref Google Scholar
 

Li, K., Yin, X., Zong, Z., 2017. Pre-stack Bayesian cascade AVA inversion in complex-Laplace domain and its application to the broadband data acquired at East China. J. Petrol. Sci. Eng. 158, 751-765. https://doi.org/10.1016/j.petrol.2017.09.005.
Crossref Google Scholar
 

Li, K., Yin, X., Zong, Z., 2020. Facies-constrained prestack seismic probabilistic inversion driven by rock physics. Science China (Earth Sciences) 63 (6), 822-840. https://doi.org/10.1007/s11430-019-9578-1.
Crossref Google Scholar
 

Maiti, S., Tiwari, R.K., 2010. Automatic discriminations among geophysical signals via the Bayesian neural networks approach. Geophysics 75 (1), E67-E78. https://doi.org/10.1190/1.3298501.
Crossref Google Scholar
 

Martin, G.S., Wiley, R., Marfurt, K.J., 2006. Marmousi2: an elastic upgrade for Marmousi. Lead. Edge 25 (2), 156-166. https://doi.org/10.1190/1.2172306.
Crossref Google Scholar
 

Röth, G., Tarantola, A., 1994. Neural networks and inversion of seismic data. J. Geophys. Res. Solid Earth 99 (B4), 6753-6768. https://doi.org/10.1029/93JB01563.
Crossref Google Scholar
 

Shahraeeni, M.S., Curtis, A., 2011. Fast probabilistic nonlinear petrophysical inversion. Geophysics 76 (2), E45-E58. https://doi.org/10.1190/1.3540628.
Crossref Google Scholar
 

Shahraeeni, M.S., Curtis, A., Chao, G., 2012. Fast probabilistic petrophysical mapping of reservoirs from 3D seismic data. Geophysics 77 (3), O1-O19. https://doi.org/10.1190/GEO2011-0340.1.
Crossref Google Scholar
 

Song, L., Yin, X., Zong, Z., Jiang, M., 2022. Semi-supervised learning seismic inversion based on Spatio-temporal sequence residual modeling neural network. J. Petrol. Sci. Eng. 208, 109549. https://doi.org/10.1016/j.petrol.2021.109549.
Crossref Google Scholar
 

Spikes, K., Mukerji, T., Dvorkin, J., Mavko, G., 2007. Probabilistic seismic inversion based on rock-physics models. Geophysics 72 (5), R87-R97. https://doi.org/10.1190/1.2760162.
Crossref Google Scholar
 

Sun, H., Demanet, L., 2020. Extrapolated full-waveform inversion with deep learning. Geophysics 85 (3), R275-R288. https://doi.org/10.1190/geo2019-0195.1.
Crossref Google Scholar
 

Sun, J., Innanen, K.A., Huang, C., 2021. Physics-guided deep learning for seismic inversion with hybrid training and uncertainty analysis. Geophysics 86 (3), R303-R317. https://doi.org/10.1190/geo2020-0312.1.
Crossref Google Scholar
 

Sun, Q.H., Zong, Z., 2019. Amplitude variation with incident angle inversion for fluid factor in the depth domain. Ann. Geophys. 62 (5), SE562. https://doi.org/10.4401/ag-7882.
Crossref Google Scholar
 
Wang, M., Xu, S., Zhou, H., 2020. Self-supervised learning for low frequency extension of seismic data. Society of Exploration Geophysicists SEG Technical Program Expanded Abstracts, pp. 1501-1505. https://doi.org/10.1190/segam2020-3427086.1.
Crossref
 

Wang, Y., Niu, L., Zhao, L., Wang, B., He, Z., Zhang, H., Chen, D., Geng, J., 2021. Gaussian mixture model deep neural network and its application in porosity prediction of deep carbonate reservoir. Geophysics 87 (2), M59-M72. https://doi.org/10.1190/geo2020-0740.1.
Crossref Google Scholar
 

Werbos, P.J., 1990. Backpropagation through time: what it does and how to do it. Proc. IEEE 78 (10), 1550-1560. https://doi.org/10.1109/5.58337.
Crossref Google Scholar
 

Wu, B., Meng, D., Zhao, H., 2021. Semi-supervised learning for seismic impedance inversion using generative adversarial networks. Rem. Sens. 13 (5), 909. https://doi.org/10.3390/rs13050909.
Crossref Google Scholar
 
Yu, F., Koltun, V., 2016. Multi-Scale Context Aggregation by Dilated Convolutions. https://doi.org/10.48550/arXiv.1511.07122.
 

Yuan, S., Jiao, X., Luo, Y., Sang, W., Wang, S., 2021. Double-scale supervised inversion with a data-driven forward model for low-frequency impedance recovery. Geophysics 87 (2), R165-R181. https://doi.org/10.1190/geo2020-0421.1.
Crossref Google Scholar
 

Zhang, X., Curtis, A., 2020. Seismic tomography using variational inference methods. J. Geophys. Res. Solid Earth 125 (4), e2019JB018589. https://doi.org/10.1029/2019JB018589.
Crossref Google Scholar
 

Zhang, X., Curtis, A., 2021. Bayesian full-waveform inversion with realistic priors. Geophysics 86 (5), A45-A49. https://doi.org/10.1190/geo2021-0118.1.
Crossref Google Scholar
 

Zhao, X., Curtis, A., Zhang, X., 2022. Bayesian seismic tomography using normalizing flows. Geophys. J. Int. 228 (1), 213-239. https://doi.org/10.1093/gji/ggab298.
Crossref Google Scholar
 

Zong, Z., Yin, X., Wu, G., 2012. AVO inversion and poroelasticity with P- and S-wave moduli. Geophysics 77 (6), N17-N24. https://doi.org/10.1190/geo2011-0214.1.
Crossref Google Scholar
 

Zong, Z., Wang, Y., Li, K., Yin, X., 2018. Broadband seismic inversion for low-frequency component of the model parameter. IEEE Trans. Geosci. Rem. Sens. 56 (9), 5177-5184. https://doi.org/10.1109/TGRS.2018.2810845.
Crossref Google Scholar
Petroleum Science
Volume 21 Issue 3,
June 2024
Pages 1611-1631
DOI: 10.1016/j.petsci.2023.12.015
Cite this article:
Sun Q-H, Zong Z-Y, Li X. Probabilistic seismic inversion based on physics-guided deep mixture density network. Petroleum Science, 2024, 21(3): 1611-1631. https://doi.org/10.1016/j.petsci.2023.12.015

31

Views

2

Downloads

0

Crossref

0

Web of Science

0

Scopus

0

CSCD

Altmetrics
Received: 21 February 2023
Revised: 11 October 2023
Accepted: 20 December 2023
Published: 28 December 2023
© 2024 The Authors.

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Return