1887
Volume 67 Number 4
  • E-ISSN: 1365-2478
PDF

Abstract

ABSTRACT

Machine learning methods including support‐vector‐machine and deep learning are applied to facies classification problems using elastic impedances acquired from a Paleocene oil discovery in the UK Central North Sea. Both of the supervised learning approaches showed similar accuracy when predicting facies after the optimization of hyperparameters derived from well data. However, the results obtained by deep learning provided better correlation with available wells and more precise decision boundaries in cross‐plot space when compared to the support‐vector‐machine approach. Results from the support‐vector‐machine and deep learning classifications are compared against a simplified linear projection based classification and a Bayes‐based approach. Differences between the various facies classification methods are connected by not only their methodological differences but also human interactions connected to the selection of machine learning parameters. Despite the observed differences, machine learning applications, such as deep learning, have the potential to become standardized in the industry for the interpretation of amplitude versus offset cross‐plot problems, thus providing an automated facies classification approach.

Loading

Article metrics loading...

/content/journals/10.1111/1365-2478.12682
2018-09-13
2020-03-31
Loading full text...

Full text loading...

/deliver/fulltext/gpr/67/4/gpr12682.html?itemId=/content/journals/10.1111/1365-2478.12682&mimeType=html&fmt=ahah

References

  1. BaldiP. and HornikK.1989. Neural networks and principal component analysis: learning from examples without local minima. Neural Networks2, 53–58.
    [Google Scholar]
  2. BergstraJ. and BengioY.2012. Random search for hyper‐parameter optimization. Journal of Machine Learning Research13, 281–305.
    [Google Scholar]
  3. CastagnaJ.P., BatzleM.L. and EastwoodR.L.1985. Relationships between compressional‐wave and shear‐wave velocities in clastic silicate rocks. Geophysics50, 571–581.
    [Google Scholar]
  4. HeatonJ.2008. Introduction to Neural Networks for Java. Heaton Research.
  5. HintonG.E., OsinderoS. and TehY‐W.2006. A fast learning algorithm for deep brief nets. Neural Computation18, 1527–1554.
    [Google Scholar]
  6. HintonG.E. and SalakhutdinovR.R.2006. Reducing the dimensionality of data with neural networks. Science313, 504–507.
    [Google Scholar]
  7. HsuC.‐W., ChangC.‐C. and LinC.‐J.2003. A practical guide to support vector classification. Technical report, Department of Computer Science, National Taiwan University.
  8. KemperM. and GunningJ.2014. Joint impedance and facies inversion: seismic inversion redefined. First Break, 32, 89–95.
    [Google Scholar]
  9. KuzmaH.A.2003. A support vector machine for AVO interpretation. 73rd SEG meeting, Dallas, USA, Expanded Abstracts, 181–184.
  10. LiG., YouJ. and LiuX.2015. Support vector machine (SVM) based prestack AVO inversion and its applications. Journal of Applied Geophysics120, 60–68.
    [Google Scholar]
  11. LiJ. and CastagnaJ.2004. Support vector machine (SVM) pattern recognition to AVO classification. Geophysical Research Letters31, L02609.
    [Google Scholar]
  12. LiZ., JinL., YangC. and ZhungZ.2015. Hyperparameter search for deep convolutional neural network using effect factors. IEEE China Summit and International Conference on Signal and Information Processing, Chengdu, China, pp. 782–786. IEEE.
  13. MetelR.M.2017. Mini‐batch stochastic gradient descent with dynamic sample size. arXiv:1708.00555.
  14. MockusJ., TiesisV. and ZilinskasA.1978. The application of Bayesian methods for seeking the extremum. Towards Global Optimization2, 117–129.
    [Google Scholar]
  15. MudgeD.C.2014. Regional controls on Lower Tertiary sandstone distribution in the North Sea and NE Atlantic margin basins. Geological Society, London, Special Publications, 403, 17–42.
    [Google Scholar]
  16. OkaforE., SmitR., SchomakerL. and WieringM.2017. Operational data augmentation in classifying single aerial images of animals. IEEE International Conference on INnovations in Intelligent SysTems and Applications (INISTA), pp. 354–360. IEEE.
    [Google Scholar]
  17. RosenblattF.1958. The perceptron: a probabilistic model for information storage and organization in the brain. Psychological Review65, 65–386.
    [Google Scholar]
  18. RumelhartD.E., HintonG.E. and WilliamsR.J.1986. Learning representations by back‐propagating errors. Nature323, 533–536.
    [Google Scholar]
  19. SchölkopfB., MikaS., BurgesC.J.C., KnirschP., MullerK.‐R., RatschG. and SmolaA.J.1999. Input space versus feature space in kernel‐based methods. IEEE Transactions on Neural Networks10, 1000–1017.
    [Google Scholar]
  20. SchölkopfB., SungK.‐K., BurgesC.J.C., GirosiF., NiyogiP., PoggioT. and VapnikV.1997. Comparing support vector machines with Gaussian kernels to radial basis function classifiers. IEEE Transactions on Signal Processing45, 2758–2765.
    [Google Scholar]
  21. SmolaA.J. and SchölkopfB.2004. A tutorial on support vector regression. Statistics and Computing14, 199–222.
    [Google Scholar]
  22. SnoekJ., RippelO., SwerskyK., KirosR., SatishN., SundaramN., PatwaryM.M.A., Prabhat and AdamsR.P.2015. Scalable Bayesian optimization using deep neural networks. Proceedings of the 32nd International Conference on Machine Learning, pp. 2171–2180.
    [Google Scholar]
  23. SvetlanaS.P. and Solov'evA.D.1997. The origin of the method of steepdescent. Historia Mathematica24, 361–375.
    [Google Scholar]
  24. VapnikV.1995. The Nature of Statistical Learning Theory. Springer.
    [Google Scholar]
  25. VapnikV. and ChervonenkisA.1974. Theory of Pattern Recognition (in Russian). Nauka.
    [Google Scholar]
  26. VapnikV. and LernerA.1963. Pattern recognition using generalized portrait method. Automation and Remote Control24, 774–780.
    [Google Scholar]
  27. WesselP. and SmithW.H.F.1991. Free software helps map and display data. EOS Transactions AGU72, 441–448.
    [Google Scholar]
  28. Zabihi NaeiniE. and ExleyR.2017. Quantitative interpretation using facies‐based seismic inversion. Interpretation5, SL1–SL8.
    [Google Scholar]
http://instance.metastore.ingenta.com/content/journals/10.1111/1365-2478.12682
Loading
/content/journals/10.1111/1365-2478.12682
Loading

Data & Media loading...

  • Article Type: Research Article
Keyword(s): Elastics , Interpretation , Inversion , Reservoir geophysics and Rock physics
This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error