Abstract
The S-wave velocity is a critical petrophysical parameter in reservoir description, prestack seismic inversion, and geomechanical analysis. However, obtaining the S-wave velocity from field measurements is difficult. When no measured S-wave data are available, petrophysical modelling provides the most accurate S-wave velocity prediction. However, because of the complexity of underground geological structures and diversity of rock minerals, the prediction results of petrophysical modelling are easily affected by factors such as the cognition and experience of the modeller. Therefore, the development of novel robust and simple S-wave velocity inversion and prediction methods independent of the modeller is critical. Inspired by ensemble learning and based on the geologic sedimentation law of reservoirs and their characteristics in logging response, an S-wave velocity inversion and prediction method based on deep hybrid neural network was developed by combining the classical convolution neural network (CNN) with the long short-term memory (LSTM) network. Considering the conventional logging data such as acoustic and density as the input in the proposed method, the CNN was used to establish the nonlinear mapping relationship between the input data and S-wave velocity, and the LSTM network was used to integrate the vertical variation trend of the stratum. Thus, intelligent data-driven inversion and prediction of the S-wave velocity were realised. The experimental results revealed that the proposed method exhibited a strong generalisation ability and could obtain prediction results comparable to those of petrophysical modelling with a single-well data set for training. Thus, a novel methodology for robust and convenient S-wave velocity prediction was devised. The proposed method has considerable academic and application implications.
Similar content being viewed by others
References
Akhundi H, Ghafoori M, Lashkaripour G R. 2014. Prediction of shear wave velocity using artificial neural network technique, multiple regression and petrophysical data: A case study in Asmari Reservoir (SW Iran). Open J Geo, 04: 303–313
Anemangely M, Ramezanzadeh A, Tokhmechi B. 2017. Shear wave travel time estimation from petrophysical logs using ANFIS-PSO algorithm: A case study from Ab-Teymour Oilfield. J Nat Gas Sci Eng, 38: 373–387
Araya-Polo M, Jennings J, Adler A, Dahlke T. 2018. Deep-learning tomography. Leading Edge, 37: 58–66
Asoodeh M, Bagheripour P. 2012. Prediction of compressional, shear, and stoneley wave velocities from conventional well log data using a committee machine with intelligent systems. Rock Mech Rock Eng, 45: 45–63
Azadpour M, Saberi M R, Javaherian A, Shabani M. 2020. Rock physics model-based prediction of shear wave velocity utilizing machine learning technique for a carbonate reservoir, southwest Iran. J Pet Sci Eng, 195: 107864
Bai J, Song Z, Su L, Yang W, Zhu L, Li S. 2012. Error analysis of shear-velocity prediction by the Xu-White model (in Chinese). Chin J Geophys, 55: 589–595
Bengio Y, Simard P, Frasconi P. 1994. Learning long-term dependencies with gradient descent is difficult. IEEE Trans Neural Netw, 5: 157–166
Bergen K J, Johnson P A, de Hoop M V, Beroza G C. 2019. Machine learning for data-driven discovery in solid Earth geoscience. Science, 363: eaau0323
Berryman J G. 1980. Long-wavelength propagation in composite elastic media I. Spherical inclusions. J Acoust Soc Am, 68: 1809–1819
Buland A, Omre H. 2003. Bayesian linearized AVO inversion. Geophysics, 68: 185–198
Cao J. 2017. Deep learning and its application in deep gas reservoir prediction (in Chinese). Comput Tech Geophys Geochem Explor, 39: 775–782
Cao J, Wu S, He X. 2016. Gas reservoir identification basing on deep learning of seismic-print Characteristics. In: AGU Fall Meeting Abstracts, doi: 2016AGUFMNS31A1926C
Cao J, Wu S. 2017. Deep learning: Chance and challenge for deep gas reservoir identification. In: International Geophysical Conference. Qingdao. 17–20
Castagna J P, Batzle M L, Eastwood R L. 1985. Relationships between compressional-wave and shear-wave velocities in clastic silicate rocks. Geophysics, 50: 571–581
Chen W, Yang L, Zha B, Zhang M, Chen Y. 2020. Deep learning reservoir porosity prediction based on multilayer long short-term memory network. Geophysics, 85: WA213–WA225
Chen Y, Zhang D. 2020. Well log generation via ensemble long short-term memory (EnLSTM) network. Geophys Res Lett, 47: e87685
Cui R, Cao D, Liu Q, Zhu Z, Jia Y. 2021. VP and VS prediction from digital rock images using a combination of U-Net and convolutional neural networks. Geophysics, 86: MR27–MR37
Eskandari H, Rezaee M R, Mohammadnia, M. 2004. Application of multiple regression and artificial neural network techniques to predict shear wave velocity from wireline log data for a carbonate reservoir South-West Iran. CSEG Recorder, 42: 40–48
Feng R. 2020. Estimation of reservoir porosity based on seismic inversion results using deep learning methods. J Nat Gas Sci Eng, 77: 103270
Gao H, Wu X, Liu G. 2021. ChannelSeg3D: Channel simulation and deep learning for channel interpretation in 3D seismic images. Geophysics, 86: IM73–IM83
Goltsev A, Rachkovskij D. 2001. A recurrent neural network for partitioning of hand drawn characters into strokes of different orientations. Int J Neur Syst, 11: 463–475
Goodfellow I, Bengio Y, Courville A. 2016. Deep Learning (Vol. 1). Cambridge: MIT Press. 326–366
Graves A, Mohamed A R, Hinton G. 2013. Speech recognition with deep recurrent neural networks. In: 2013 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Vancouver. 6645–6649
Graves A, Schmidhuber J. 2005. Framewise phoneme classification with bidirectional LSTM and other neural network architectures. Neural Networks, 18: 602–610
Greenberg M L, Castagna J P. 1992. Shear-wave velocity estimation in porous rocks: Theoretical formulation, preliminary verification and applications. Geophys Prospect, 40: 195–209
Guo Q, Jin S, Li M, Yang Q, Xu K, Ju Y, Zhang J, Xuan J, Liu J, Su Y, Xu Q, Liu Y. 2020. Application of deep learning in ecological resource research: Theories, methods, and challenges. Sci China Earth Sci, 63: 1457–1474
Han D, Nur A, Morgan D. 1986. Effects of porosity and clay content on wave velocities in sandstones. Geophysics, 51: 2093–2107
Hinton G E, Osindero S, Teh Y W. 2006. A fast learning algorithm for deep belief nets. Neural Comput, 18: 1527–1554
Hochreiter S, Schmidhuber J. 1997. Long short-term memory. Neural Comput, 9: 1735–1780
Imamverdiyev Y, Sukhostat L. 2019. Lithological facies classification using deep convolutional neural network. J Pet Sci Eng, 174: 216–228
Jeong J, Park E, Emelyanova I, Pervukhina M, Esteban L, Yun S T. 2021. Application of conditional generative model for sonic log estimation considering measurement uncertainty. J Pet Sci Eng, 196: 108028
Jørstad A, Mukerji T, Mavko G. 1999. Model-based shear-wave velocity estimation versus empirical regressions. Geophys Prospecting, 47: 785–797
Krief M, Garat J, Stellingwerf J, Ventre J. 1990. A petrophysical interpretation using the velocities of P and S waves (full-waveform sonic). Log Analyst, 31: 355–369
Krizhevsky A, Sutskever I, Hinton G. 2012. ImageNet classification with deep convolutional neural networks. Adv Neural Inform Process Syst, 25: 1097–1105
Kuang L, Liu H, Ren Y, Luo K, Shi M, Su J, Li X. 2021. Application and development trend of artificial intelligence in petroleum exploration and development. Pet Explor Dev, 48: 1–14
Lambrecht B N, Vanderkerken M, Hammad H. 2018. The emerging role of ADAM metalloproteinases in immunity. Nat Rev Immunol, 18: 745–758
Langer H, Falsaperla S, Masotti M, Campanini R, Spampinato S, Messina A. 2009. Synopsis of supervised and unsupervised pattern classification techniques applied to volcanic tremor data at Mt Etna, Italy. Geophys J Int, 178: 1132–1144
LeCun Y, Bengio Y, Hinton G. 2015. Deep learning. Nature, 521: 436–444
Lee S H. 1990. Regression models of shear wave velocities in Taipei basin. J Chin Institute Eng, 13: 519–532
Li D, Liu X, Zha W, Yang J, Lu D. 2020. Automatic well test interpretation based on convolutional neural network for a radial composite reservoir. Pet Explor Dev, 47: 623–631
Lin N, Zhang D, Zhang K, Wang S, Fu C, Zhang J, Zhang C. 2018. Predicting distribution of hydrocarbon reservoirs with seismic data based on learning of the small-sample convolution neural network (in Chinese). Chin J Geophys, 61: 4110–4125
Li N, Xu B, Wu H, Feng Z, Li Y, Wang K, Liu P. 2021. Application status and prospects of artificial intelligence in well logging and formation evaluation (in Chinese). Acta Petrol Sin, 42: 508–522
Li W, Wang H, Yao Z, Liu Y, Chang X. 2009. Shear-wave velocity estimation and fluid substitution by constraint method (in Chinese). Chin J Geophys, 52: 785–791
Lin S S, Shen S L, Zhang N, Zhou A. 2021. Modelling the performance of EPB shield tunnelling using machine and deep learning algorithms. Geosci Front, 12: 101177
Lindberg D V, Rimstad E, Omre H. 2015. Inversion of well logs into facies accounting for spatial dependencies and convolution effects. J Pet Sci Eng, 134: 237–246
Luo S, Yang P, Hu G, Liu S. 2016. S-wave velocity prediction based on the modified P-L model and matrix equation iteration (in Chinese). Chin J Geophys, 59: 1839–1848
Ma X, Tao Z, Wang Y, Yu H, Wang Y. 2015. Long short-term memory neural network for traffic speed prediction using remote microwave sensor data. Tran Res Part C-Emer Technol, 54: 187–197
Mousavi S M, Horton S P, Langston C A, Samei B. 2016. Seismic features and automatic discrimination of deep and shallow induced-micro-earthquakes using neural network and logistic regression. Geophys J Int, 207: 29–46
Nebauer C. 1998. Evaluation of convolutional neural networks for visual recognition. IEEE Trans Neural Netw, 9: 685–696
Olayiwola T, Sanuade O A. 2021. A data-driven approach to predict compressional and shear wave velocities in reservoir rocks. Petroleum, 7: 199–208
Pan J G, Wang H B, Li C, Zhao J G. 2015. Effect of pore structure on seismic rock-physics characteristics of dense carbonates. Appl Geophys, 12: 1–10
Pham N, Wu X, Zabihi Naeini E. 2020. Missing well log prediction using convolutional long short-term memory network. Geophysics, 85: WA159–WA171
Qi Q, Cheng A C H, Li Y E. 2019. Determination of formation shear attenuation from dipole sonic log data. Geophysics, 84: D73–D79
Qi Q, Müller T M, Pervukhina M. 2017. Sonic QP/QS ratio as diagnostic tool for shale gas saturation. Geophysics, 82: MR97–MR103
Rajabi M, Bohloli B, Gholampour Ahangar E. 2010. Intelligent approaches for prediction of compressional, shear and Stoneley wave velocities from conventional well log data: A case study from the Sarvak carbonate reservoir in the Abadan Plain (Southwestern Iran). Comput Geoscis, 36: 647–664
Rezaee M R, Kadkhodaie-Ilkhchi A, Barabadi A. 2007. Prediction of shear wave velocity from petrophysical data utilizing intelligent systems: An example from a sandstone reservoir of Carnarvon Basin, Australia. J Pet Sci Eng, 55: 201–212
Russell B H, Hedlin K, Hilterman F J, Lines L R. 2003. Fluid-property discrimination with AVO: A Biot-Gassmann perspective. Geophysics, 68: 29–39
Saad O M, Chen Y. 2020. Deep denoising autoencoder for seismic random noise attenuation. Geophysics, 85: V367–V376
Sohail G M, Hawkes C D. 2020. An evaluation of empirical and rock physics models to estimate shear wave velocity in a potential shale gas reservoir using wireline logs. J Pet Sci Eng, 185: 106666
Sun Q, Li N, Duan Y, Li H, Tang H. 2021. Logging-while-drilling formation dip interpretation based on long short-term memory. Pet Explor Dev, 48: 978–986
Tan M, Peng X, Cao H, Wang S, Yuan Y. 2015. Estimation of shear wave velocity from wireline logs in gas-bearing shale. J Pet Sci Eng, 133: 352–366
Tian Y, Gao J, Wang D, Chen D. 2021. Removing strong seismic reflection based on the deep neural network (in Chinese). Chin J Geophys, 64: 2780–2794
Tian M, Omre H, Xu H. 2021. Inversion of well logs into lithology classes accounting for spatial dependencies by using hidden markov models and recurrent neural networks. J Pet Sci Eng, 196: 107598
Wang J, Cao J, Yuan S. 2020. Shear wave velocity prediction based on adaptive particle swarm optimization optimized recurrent neural network. J Pet Sci Eng, 194: 107466
Wang K, Huang Q, Wu S. 2020. Application of long short-term memory neural network in geoelectric field data processing (in Chinese). Chin J Geophys, 63: 3015–3024
Wang P, Peng S. 2019. On a new method of estimating shear wave velocity from conventional well logs. J Pet Sci Eng, 180: 105–123
Wu X, Geng Z, Shi Y, Pham N, Fomel S, Caumon G. 2020. Building realistic structure models to train convolutional neural networks for seismic structural interpretation. Geophysics, 85: WA27–WA39
Xi X, Huang J. 2020. Location and imaging of scatterers in seismic migration profiles based on convolution neural network (in Chinese). Chin J Geophys, 63: 687–714
Xing F, Xie Y, Su H, Liu F, Yang L. 2017. Deep learning in microscopy image analysis: A survey. IEEE Trans Neural Netw Learn Syst, 29: 4550–4568
Xu J, Tan M, Wang X, Wu C. 2017. Predicting acoustic-wave velocities and fluid sensitivity to elastic properties in fractured carbonate formation. Interpretation, 5: SB69–SB80
Xu S, Payne M A. 2009. Modeling elastic properties in carbonate rocks. Leading Edge, 28: 66–74
Xu S, White R E. 1996. A physical model for shear-wave velocity pre-diction1. Geophys Prospect, 44: 687–717
Ye S, Zhu H, Li R, Yang Y, Li Q. 2017. Tracing natural gas migration by integrating organic and inorganic geochemical data: A case study of the Jurassic gas fields in western Sichuan Basin, SW China. Pet Explor Dev, 44: 582–594
Yin X Y, Zong Z Y, Wu G C. 2015. Research on seismic fluid identification driven by rock physics. Sci China Earth Sci, 58: 159–171
Yu S, Ma J. 2021. Deep learning for geophysics: Current and future trends. Rev Geophys, 59: e00742
Yu S, Ma J, Wang W. 2019. Deep learning for denoising, Geophysics, 84: V333–V350
Zhang B, Jin S, Liu C, Guo Z, Liu X. 2020. Prediction of shear wave velocity based on a statistical rock-physics model and Bayesian theory. J Pet Sci Eng, 195: 107710
Zhang D, Chen Y, Meng J. 2018. Synthetic well logs generation via Recurrent Neural Networks. Pet Explor Dev, 45: 629–639
Zhang G, Chen H, Wang Q, Yin X. 2013. Estimation of S-wave velocity and anisotropic parameters using fractured carbonate rock physics model (in Chinese). Chin J Geophys, 56: 1707–1715
Zhang G, Wang Z, Chen Y. 2018. Deep learning for seismic lithology prediction. Geophys J Int, 215: 1368–1387
Zhang Y, Yu Z, Hu T, He C. 2021. Multi-trace joint downhole microseismic phase detection and arrival picking method based on U-Net (in Chinese). Chin J Geophys, 64: 2073–2085
Zhang Y, Zhong H R, Wu Z Y, Zhou H, Ma Q Y. 2020. Improvement of petrophysical workflow for shear wave velocity prediction based on machine learning methods for complex carbonate reservoirs. J Pet Sci Eng, 192: 107234
Zhao M, Chen S, Dave Y. 2019. Waveform classification and seismic recognition by convolution neural network (in Chinese). Chin J Geophys, 62: 374–382
Zhou F, Jin L, Dong J. 2017. Review of convolutional neural networks (in Chinese). Chin J Comput, 40: 1229–1251
Zhu K, Wang L, Du Y, Jiang C, Sun Z. 2020. DeepLog: Identify tight gas reservoir using multi-log signals by a fully convolutional network. IEEE Geosci Remote Sens Lett, 17: 568–571
Acknowledgements
We appreciate the valuable comments and revision suggestions from the responsible editor and anonymous reviewers. This work was supported by the National Natural Science Foundation of China (Grant Nos. 42030812, 42042046, 41974160) and the project of SINOPEC Science and Technology Department (Grant No. P20055-6).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Wang, J., Cao, J., Zhao, S. et al. S-wave velocity inversion and prediction using a deep hybrid neural network. Sci. China Earth Sci. 65, 724–741 (2022). https://doi.org/10.1007/s11430-021-9870-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11430-021-9870-8