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 (5.7 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

An adaptive physics-informed deep learning method for pore pressure prediction using seismic data

Xin Zhanga,b, Yun-Hu Lua,b,c( ), Yan Jina,b,c, Mian Chena,b, Bo Zhoud
State Key Laboratory of Petroleum Resources and Prospecting, China University of Petroleum (Beijing), Beijing, 102249, China
College of Petroleum Engineering, China University of Petroleum (Beijing), Beijing, 102249, China
College of Artificial Intelligence, China University of Petroleum (Beijing), Beijing, 102249, China
PetroChina Tarim Oilfield Company, Korla, Xinjiang, 841000, China

Edited by Jie Hao and Meng-Jiao Zhou

Show Author Information

Abstract

Accurate prediction of formation pore pressure is essential to predict fluid flow and manage hydrocarbon production in petroleum engineering. Recent deep learning technique has been receiving more interest due to the great potential to deal with pore pressure prediction. However, most of the traditional deep learning models are less efficient to address generalization problems. To fill this technical gap, in this work, we developed a new adaptive physics-informed deep learning model with high generalization capability to predict pore pressure values directly from seismic data. Specifically, the new model, named CGP-NN, consists of a novel parametric features extraction approach (1DCPP), a stacked multilayer gated recurrent model (multilayer GRU), and an adaptive physics-informed loss function. Through machine training, the developed model can automatically select the optimal physical model to constrain the results for each pore pressure prediction. The CGP-NN model has the best generalization when the physics-related metric λ = 0.5. A hybrid approach combining Eaton and Bowers methods is also proposed to build machine-learnable labels for solving the problem of few labels. To validate the developed model and methodology, a case study on a complex reservoir in Tarim Basin was further performed to demonstrate the high accuracy on the pore pressure prediction of new wells along with the strong generalization ability. The adaptive physics-informed deep learning approach presented here has potential application in the prediction of pore pressures coupled with multiple genesis mechanisms using seismic data.

Keywords

Pore pressure predictionSeismic data1D convolution pyramid poolingAdaptive physics-informed loss functionHigh generalization capability

References

 
Agrawal, V., Singh, R.K., Singh, J., et al., 2011. Role of formation pressure and sonic measurements while drilling in pre-drill well design validation and syn-drill well design optimization - case studies from deepwater, India. In: SPE Middle East Oil and Gas Show and Conference, Manama, Bahrain. https://doi.org/10.2118/142540-MS.
Crossref
 

Boer, L., Sayers, C.M., Nagy, Z.R., et al., 2006. Pore pressure prediction using well-conditioned seismic velocities. First Break 24 (5), 43-49. https://doi.org/10.3997/1365-2397.2006012.
Crossref Google Scholar
 

Bowers, G.L., 1995. Pore pressure estimation from velocity data: accounting for overpressure mechanisms besides undercompaction. SPE Drill. Complet. 10 (2), 89-95. https://doi.org/10.2118/27488-PA.
Crossref Google Scholar
 
Bowers, G.L., 2001. Determining an appropriate pore-pressure estimation strategy. In: Offshore Technology Conference, Houston, Texas. https://doi.org/10.4043/13042-MS.
Crossref
 

Chen, F.B., Zong, Z.Y., 2022. PP-wave reflection coefficient in stress-induced anisotropic media and amplitude variation with incident angle and azimuth inversion. Geophysics 87 (6), C155-C172. https://doi.org/10.1190/geo2021-0706.1.
Crossref Google Scholar
 

Chen, F.B., Zong, Z.Y., Yin, X.Y., 2023. Monitoring the change in horizontal stress with multi-wave time-lapse seismic response based on nonlinear elasticity theory. Petrol. Sci. 20 (2), 815-826. https://doi.org/10.1016/j.petsci.2022.09.022.
Crossref Google Scholar
 

Cho, K., Van Merrienboer, B., Bahdanau, D., et al., 2014. On the properties of neural machine translation: encoder-decoder approaches. https://doi.org/10.48550/arXiv.1409.1259.
Crossref
 

Chung, J., Gulcehre, C., Cho, K., et al., 2014. Empirical evaluation of gated recurrent neural networks on sequence modeling. https://doi.org/10.48550/arXiv.1412.3555.
Crossref
 

Cibin, P., 1999. What seismic velocity field for pore pressure prediction. SEG Tech. Progr. Expand. Abstr. 23 (1). https://doi.org/10.1190/1.1851132.
Crossref Google Scholar
 

Dong, C., Loy, C.C., He, K., et al., 2016. Image super-resolution using deep convolutional networks. IEEE Trans. Pattern Anal. Mach. Intell. 38 (2), 295-307. https://doi.org/10.1109/TPAMI.2015.2439281.
Crossref Google Scholar
 

Dutta, N., 2002. Geopressure prediction using seismic data: current status and the road ahead. Geophysics 67 (6), 2012-2041.
Crossref Google Scholar
 
Eaton, B.A., 1975. The equation for geopressure prediction from well logs. In: The Society of Petroleum Engineers of AIME, Dallas, Texas. https://doi.org/10.2118/5544-MS.
Crossref
 

Farsi, M., Mohamadian, N., Ghorbani, H., et al., 2021. Predicting formation pore-pressure from well-log data with hybrid machine-learning optimization algorithms. Nat. Resour. Res. 30, 3455-3481. https://doi.org/10.1007/s11053-021-09852-2.
Crossref Google Scholar
 

Gao, Y., Chen, M., Jiang, H., 2021. Influence of unconnected pores on effective stress in porous geomaterials: theory and case study in unconventional oil and gas reservoirs. J. Nat. Gas Sci. Eng. 88, 103787.
Crossref Google Scholar
 

Gutierrez, M.A., Braunsdorf, N.R., Couzens, B.A., 2006. Calibration and ranking of pore-pressure prediction models. Lead. Edge 25 (12), 1516-1523.
Crossref Google Scholar
 
He, K., Zhang, X., Ren, S., et al., 2015. Spatial pyramid pooling in deep convolutional networks for visual recognition. In: IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 1904-1916.
Crossref
 

Hochreiter, S., Schmidhuber, J., 1997. Long short-term memory. Neural Comput. 9 (8), 1735-1780.
Crossref Google Scholar
 

Hottmann, C.E., Johnson, R.K., 1965. Estimation of formation pressures from log-derived shale properties. J. Petrol. Technol. 17 (6), 717-722.
Crossref Google Scholar
 
Huang, G., Liu, Z., van der Maaten, L., et al., 2017. Densely connected convolutional networks. In: 30th IEEE Conference on Computer Vision and Pattern Recognition, New York, pp. 2261-2269.
Crossref
 

Huang, H., Li, J., Yang, H., et al., 2022. Research on prediction methods of formation pore pressure based on machine learning. Energy Sci. Eng. 10 (6), 1886-1901. https://doi.org/10.1002/ese3.1112.
Crossref Google Scholar
 

Jin, J., Jin, Y., Lu, Y., et al., 2022. Image processing and machine learning based cavings characterization and classification. J. Petrol. Sci. Eng. 208, 109525. https://doi.org/10.1016/j.petrol.2021.109525.
Crossref Google Scholar
 

Keshavarzi, R., Jahanbakhshi, R., 2013. Real-time prediction of pore pressure gradient through an artificial intelligence approach: a case study from one of middle east oil fields. Euro. J. Environ. Civil Eng. 17 (8), 675-686. https://doi.org/10.1080/19648189.2013.811614.
Crossref Google Scholar
 

Kim, Y., 2014. Convolutional neural networks for sentence classification. https://doi.org/10.48550/arXiv.1408.5882.
Crossref
 

LeCun, Y., Bengio, Y., Hinton, G., 2015. Deep learning. Nature 521 (7553), 436-444. https://doi.org/10.1038/nature14539.
Crossref Google Scholar
 

Li, Q.F., Fu, J.H., Chi, P., et al., 2023. A deep learning approach for abnormal pore pressure prediction based on multivariate time series of kick. Geoenergy Sci. Eng. 226, 211715. https://doi.org/10.1016/j.geoen.2023.211715.
Crossref Google Scholar
 

Lopez, J.L., Rappold, P.M., Ugueto, G.A., et al., 2004. Integrated shared earth model: 3D pore-pressure prediction and uncertainty analysis. Lead. Edge 23 (1), 52-59.
Crossref Google Scholar
 

Matinkia, M., Amraeiniya, A., Behboud, M.M., et al., 2022. A novel approach to pore pressure modeling based on conventional well logs using convolutional neural network. J. Petrol. Sci. Eng. 211, 110156. https://doi.org/10.1016/j.petrol.2022.110156.
Crossref Google Scholar
 

Najibi, A.R., Ghafoori, M., Lashkaripour, G.R., et al., 2017. Reservoir geomechanical modeling: in-situ stress, pore pressure, and mud design. J. Petrol. Sci. Eng. 151, 31-39. https://doi.org/10.1016/j.petrol.2017.01.045.
Crossref Google Scholar
 
Opara, A.I., Onuoha, K.M., 2009. Pre-drill pore pressure prediction from 3-D seismic data in parts of the onshore Niger delta basin. In: SPE International Technical Conference and Exhibition in Abuja, Abuja, Nigeria. https://doi.org/10.2118/128354-ms.
Crossref
 

Oprea, S., Martinez-Gonzalez, P., Garcia-Garcia, A., et al., 2022. A review on deep learning techniques for video prediction. IEEE Trans. Pattern Anal. Mach. Intell. 44 (6), 2806-2826. https://doi.org/10.1109/TPAMI.2020.3045007.
Crossref Google Scholar
 

Oughton, R.H., Wooff, D.A., Hobbs, R.W., et al., 2018. A sequential dynamic bayesian network for pore-pressure estimation with uncertainty quantification. Geophysics 83 (2), D27-D39. https://doi.org/10.1190/geo2016-0566.1.
Crossref Google Scholar
 

Sayers, C.M., Johnson, G.M., Denyer, G., 2002. Predrill pore-pressure prediction using seismic data. Geophysics 67 (4), 1286-1292. https://doi.org/10.1190/1.1500391.
Crossref Google Scholar
 
Subrahmanyam, D., Rao, P.H., 2008. Seismic attributes-a review. In: 7th International Conference and Exposition on Petroleum Geophysics. Hyderabad.
 
Szegedy, C., Liu, W., Jia, Y., et al., 2015. Going deeper with convolutions. In: IEEE Conference on Computer Vision and Pattern Recognition, New York. https://doi.org/10.1109/cvpr.2015.7298594.
Crossref
 

Tjur, T., 2009. Coefficients of determination in logistic regression models-a new proposal: the coefficient of discrimination. Am. Statistician 63 (4), 366-372. https://doi.org/10.1198/tast.2009.08210.
Crossref Google Scholar
 

Wang, Z., Wang, R., 2015. Pore pressure prediction using geophysical methods in carbonate reservoirs: current status, challenges and way ahead. J. Nat. Gas Sci. Eng. 27, 986-993. https://doi.org/10.1016/j.jngse.2015.09.032.
Crossref Google Scholar
 

Weinzierl, W., Wiese, B., 2021. Deep learning a poroelastic rock-physics model for pressure and saturation discrimination. Geophysics 86 (1), MR53-MR66.
Crossref Google Scholar
 

Wessling, S., Bartetzko, A., Tesch, P., 2013. Quantification of uncertainty in a multistage/multiparameter modeling workflow: pore pressure from geophysical well logs. Geophysics 78 (3), WB101-WB112.
Crossref Google Scholar
 

Yang, J., Zhang, D., Frangi, A.F., et al., 2004. Two-dimensional PCA: a new approach to appearance-based face representation and recognition. IEEE Trans. Pattern Anal. Mach. Intell. 26 (1), 131-137.
Crossref Google Scholar
 

Ye, M., Shen, J., Lin, G., et al., 2022. Deep learning for person re-identification: a survey and outlook. IEEE Trans. Pattern Anal. Mach. Intell. 44 (6), 2872-2893. https://doi.org/10.1109/TPAMI.2021.3054775.
Crossref Google Scholar
 

Yu, F., Jin, Y., Chen, K.P., et al., 2014. Pore-pressure prediction in carbonate rock using wavelet transformation. Geophysics 79 (4), D243-D252. https://doi.org/10.1190/geo2013-0277.1.
Crossref Google Scholar
 

Zhang, J., 2011. Pore pressure prediction from well logs: methods, modifications, and new approaches. Earth Sci. Rev. 108 (1–2), 50-63. https://doi.org/10.1016/j.earscirev.2011.06.001.
Crossref Google Scholar
 

Zhang, J., Yin, S., 2017. Real-time pore pressure detection: indicators and improved methods. Geofluids 2017, 3179617. https://doi.org/10.1155/2017/3179617.
Crossref Google Scholar
 

Zhang, Z., Sun, B., Wang, Z., et al., 2022. Formation pressure inversion method based on multisource information. SPE J. 27 (2), 1287-1303. https://doi.org/10.2118/209206-PA.
Crossref Google Scholar
 

Zhao, J., 2019. Speech emotion recognition using deep 1D and 2D CNN LSTM networks. Biomed. Signal Process Control 47, 312-323. https://doi.org/10.1016/j.bspc.2018.08.035.
Crossref Google Scholar
Petroleum Science
Volume 21 Issue 2,
April 2024
Pages 885-902
DOI: 10.1016/j.petsci.2023.11.006
Cite this article:
Zhang X, Lu Y-H, Jin Y, et al. An adaptive physics-informed deep learning method for pore pressure prediction using seismic data. Petroleum Science, 2024, 21(2): 885-902. https://doi.org/10.1016/j.petsci.2023.11.006

5

Views

0

Downloads

0

Crossref

0

Web of Science

0

Scopus

0

CSCD

Altmetrics
Received: 12 April 2023
Revised: 11 October 2023
Accepted: 06 November 2023
Published: 19 November 2023
© 2023 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