Cite this article
1
Download
44
Views
Journal Browser
Volume | Year
Issue
Search
News and Announcements
View All
ARTICLE

Sandstone porosity prediction based on gated recurrent units

HUI SONG1 WEI CHEN*2,3 HUA ZHANG4 YANG WANG5,6 YAJUAN XUE7
Show Less
1 College of Geophysics and Petroleum Resources, Yangtze University, Daxue Road 111, Caidian District, Wuhan 430100, P.R. China. 201400567@yangtzeu.edu.cn,
2 Key Laboratory of Exploration Technology for Oil and Gas Resources of Ministry of Education, Yangtze University, Daxue Road 111, Wuhan 430100, P.R. China.,
3 Hubei Cooperative Innovation Center of Unconventional Oil and Gas, Daxue Road 111, Caidian District, Wuhan 430100, P.R. China.,
4 School of Geophysics and Measurement-control Technology, East China University of Technology, Nanchang 330013, P.R. China.,
5 Faculty of Materials Science & Engineering, Hubei University, Wuhan 430062, China.,
6 Tianshu New Energy Material Industry Research and Design Institute, Hubei University, Wuhan 430062, P.R. China.,
7 School of Communication Engineering, Chengdu University of Information Technology, Chengdu 610225, P.R. China.,
JSE 2020, 29(4), 371–388;
Submitted: 1 March 2019 | Accepted: 28 February 2020 | Published: 1 August 2020
© 2020 by the Authors. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution -Noncommercial 4.0 International License (CC-by the license) ( https://creativecommons.org/licenses/by-nc/4.0/ )
Abstract

Song, H., Chen, W., Zhang, H., Wang, Y. and Xue, Y.J., 2020. Sandstone porosity prediction based on gated recurrent units. Journal of Seismic Exploration, 29: 371-388. Sandstone porosity prediction is a difficult task because of the heterogeneity of reservoir rock. Deep learning has been widely used in various fields, but it is rarely used for sandstone porosity prediction. Recurrent neural networks (RNNs) are currently very popular algorithms for deep learning and have achieved good performance in processing sequence data, such as speech recognition and machine translation. Since sandstone porosity prediction belongs to sequence data prediction, this paper proposes to use the gated recurrent units (GRUs), which is a variant of RNNs, to predict sandstone porosity using the well logs data. Six well logs data are divided into training set, validation set and testing set. We apply three deep learning architectures including convolutional neural networks (CNNs), recurrent neural networks (RNNs) and gated recurrent units (GRUs) to predict sandstone porosity. Results show that GRUs can extract the nonlinear characteristics of data more effectively and are more suitable for sandstone porosity prediction.

Keywords
convolutional neural networks (CNNs)
recurrent neural networks (RNNs)
gated recurrent units (GRUs)
sandstone porosity prediction
References
  1. Ahmadi, M.A., Zendehboudi, S., Lohi, A., Elkamel, A. and Chatzis, I, 2013. Reservoirpermeability prediction by neural networks combined with hybrid genetic algorithm andparticle swarm optimization. Geophys. Prosp., 61, 582-598.
  2. Bengio, Y., Simard, P. and Frasconi, P., 1994. Learning long-term dependencies with gradientdescent is difficult. IEEE Transact. Neural Netw., 5: 157-166.
  3. Chen, W., Xie, J., Zu, S., Gan, S. and Chen, Y., 2016a. Multiple-reflection noise attenuationusing adaptive randomized- order empirical mode decomposition. IEEE Geosci. RemoteSens. Lett., 14: 18-22.
  4. Chen, W., Yuan, J., Chen, Y. and Gan, S., 2017a. Preparing the initial model for iterativedeblending by median filtering. J. Seismic Explor, 26: 25-47.
  5. Chen, Y., Hill, J., Lei, W., Lefebvre, M., Tromp, J., Bozdag, E. and Komatitsch, D., 2017b.
  6. Automated time-window selection based on machine learning for full-waveforminversion. Expanded Abstr., 87th Ann. Internat. SEG Mtg., Houston: 1604-1609.
  7. Chen, Y., Huang, W., Zhang, D. and Chen, W., 2016b. An open-source matlab code package forimproved rank-reduction 3D seismic data denoising and reconstruction. Comput. Geosci., 95:59-66.
  8. Chung, J., Gulcehre, C., Cho, K. and Bengio, Y. 2014. Empirical evaluation of gatedrecurrent neural network on sequence modeling. arXiv preprint arXiv:1412.3555.
  9. Cracknell, M.J. and Reading, A.M. 2013. The upside of uncertainty: Identification oflithology contact zones from airborne geophysics and satellite data using random forests andsupport vector machines. Geophysics, 78(3): WB113-WB126.
  10. Fattahi, H. and Karimpouli, S., 2016. Prediction of porosity and water saturation usingpre-stack seismic attributes: a comparison of Bayesian inversion and computationalintelligence methods: Comput. Geosci., 20: 1075-1094.
  11. Helle, H.B., Bhatt, A. and Ursin, B., 2001. Porosity and permeability prediction fromwire-line logs using artificial neural networks: a north sea case study. Geophys. Prosp., 49:431-444.
  12. Hinton, G., Deng, L., Yu, D., Dahl, G.E., Mohamed, A.-R., Jaitly, N., Senior, A., Vanhoucke, V.,
  13. Nguyen, P. and Sainath, T.N., 2012. Deep neural networks for acoustic modeling in speechrecognition: The shared views of four research groups. IEEE Sign. Process. Mag., 29: 82-97.
  14. Hinton, G.E., Osindero, S. and Teh, Y.-W., 2006. A fast learning algorithm for deep beliefnets. Neural Comput., 18: 1527-1554.
  15. Hochreiter, S. and Schmidhuber, J., 1997. Long short-term memory. Neural Comput., 9:1735-1780.
  16. Iturrar’an-Viveros, U. and Parra, J.O., 2014. Artificial neural networks applied to estimatepermeability, porosity and intrinsic attenuation using seismic attributes and well-log data. J.Appl. Geophys., 107: 45-54.
  17. Kavukcuoglu, K., Sermanet, P., Boureau, Y.-L., Gregor, K., Mathieu, M. and Cun, Y.L., 2010.
  18. Learning convolutional feature hierarchies for visual recognition. Adv. Neur. Inform.Process. Syst., 1090-1098.
  19. Liu, W. and Chen, W., 2019. Recent advancements in empirical wavelet transform and itsapplications. IEEE Access, 7: 103770-103780.
  20. Mikolov, T., Chen, K., Corrado, G. and Dean, J., 2013. Efficient estimation of wordrepresentations in vector space. arXiv preprint arXiv:1301.3781.
  21. Paitz, P., Gokhberg, A. and Fichtner, A., 2017. A neural network for noise correlationclassification. Geophys. J. Internat., 212: 1468-1474.
  22. Pal, M., 2006. Support vector machines-based modelling of seismic liquefaction potential.
  23. Internat. J. Numer. Analyt. Meth. Geomechan., 30: 983-996.
  24. Reynen, A. and Audet, P., 2017. Supervised machine learning on a network scale: Application toseismic event classification and detection. Geophys. J. Internat., 210: 1394-1409.
  25. Rogers, S., Chen, H., Kopaska-Merkel, D. and Fang, J., 1995. Predicting permeability fromporosity using artificial neural networks. AAPG Bull., 79: 1786-1796.
  26. Siahsar, M.A.N., Gholtashi, S., Torshizi, E.O., Chen, W. and Chen, Y., 2017. Simultaneousdenoising and interpolation of 3-D seismic data via damped data-driven optimal singularvalue shrinkage. IEEE Geosci. Remote Sens. Lett., 14: 1086-1090.
  27. Yuan, S., Liu, J., Wang, S.. Wang, T. and Shi, P., 2018. Seismic waveform classification andfirst-break picking using convolution neural networks. IEEE Geosci. Remote Sens. Lett. 15:272-276.
  28. Zhang, G., Wang, Z. and Chen, Y., 2018a. Deep learning for seismic lithology prediction.Geophys. J. Internat., 215: 1368-1387.
  29. Zhang, G., Wang, Z., Li, H., Sun, Y., Zhang, Q. and Chen, W., 2018b. Permeability prediction ofisolated channel sands using machine learning. J. Appl. Geophys., 159: 605-615.
  30. Zhang, K., Zuo, W., Chen, Y., Meng, D. and Zhang, L., 2017. Beyond a Gaussian denoiser:
  31. Residual learning of deep CNN for image denoising. IEEE Transact. Image Process., 26:3142-3155.
  32. Zhao, T., Jayaram, V., Marfurt, K.J. and Zhou, H., 2014. Lithofacies classification in barnettshale using proximal support vector machines. Expanded Abstr., 84 Ann. Internat. SEGMtg., Denver: 1491-1495.
Share
Back to top
Journal of Seismic Exploration, Electronic ISSN: 0963-0651 Print ISSN: 0963-0651, Published by AccScience Publishing