Lithofacies Identification from Wireline Logs

Bringing Neural Networks to Application
  • Willem J. M. Epping
  • Sandra M. Oudshoff
  • Frances V. Abbots
Conference paper


Neural-network classifiers, of the multi-layered perceptron type, were trained to identify lithofacies (rock types) from wireline logs in two ways: (1) by a semi-autonomous method based on pre-segmented log intervals and (2) by an autonomous method based on non-segmented data. Good results were achieved in reservoirs from two geologically different environments (siliciclastics and carbonates) without having to fine-tune the network parameters. The performance was substantially better than that of linear discriminant and Gaussian classifiers. The standard neural-network algorithm has been modified to increase its robustness e. g. to over-training. A user-friendly neural-network module has been embedded in an existing computing environment to facilitate the use of neural networks by formation analysts.


Neural Network Hide Unit Posteriori Probability Lithofacies Type Gaussian Classifier 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Wolff, M. and Pelissier-Combescure, J.: “Faciolog — automatic electrofacies determination”, SPWLA 23rd Annual Logging Symposium (1982) July 6–9, paper FF.Google Scholar
  2. [2]
    Delfiner, P., Peyret, O. and Serra, O.: “Automatic determination of lithology from well logs”, SPE Formation Evaluation (1987) September, 303–310.Google Scholar
  3. [3]
    Bush, J.M., Fortney, W.G. and Berry, L.N.: “Determination of lithology from well logs by statistical analysis”, SPE Formation Evaluation (1987) December, 412–418.Google Scholar
  4. [4]
    Baldwin, J.L., Bateman, R.M. and Wheatley, C.L.: “Application of a neural network to the problem of mineral identification from well logs”, The Log Analyst (1990) September-October, 279–293.Google Scholar
  5. [5]
    Cardon, H.R.A., Van Hoogstraten, R. and Davies, P.: “A neural network application in geology: identification of genetic facies”, Proceedings of the International Conference on Artificial Neural Networks, Espoo Finland, (1991) June 24–28, p. 809–813.Google Scholar
  6. [6]
    Liu, R.L., Zhou, C.D. and Jin, Z.W.: “Lithofacies sequence recognition from well logs using time-delay neural networks”, SPWLA 33rd Annual Logging Symposium (1992) June 14–17, paper L.Google Scholar
  7. [7]
    Rumelhart, D.E., McClelland, J.L. and the PDP Research Group. Parallel distributed processing, explorations in the microstructure of cognition, Vol. 1: Foundations. MIT Press, Cambridge MA, 1986.Google Scholar
  8. [8]
    Kaiman, B.L. and Kwasny, S.C.: “Why tanh: choosing a sigmoidal function”, Proceedings of the International Joint Conference on Neural Networks, Baltimore MD, (1992) June 7–11, Vol. 4, p. 578–581.Google Scholar
  9. [9]
    Solla, S.A., Levin, E. and Fleisher, M.: “Accelerated learning in layered neural networks”, Complex Systems (1988) 2, 625–640.MathSciNetzbMATHGoogle Scholar
  10. [10]
    Weigend, A.S., Rumelhart, D.E. and Huberman, B.A.: “Generalisation by weight-elimination with application to forecasting”, Advances of Neural Information Processing Systems 3, Morgan Kaufmann, San Mateo CA, (1991), p. 875–882.Google Scholar
  11. [11]
    Heskes, T.M., Slijpen, E.T.P. and Kappen, B.: “Cooling schedules for learning in neural networks”, Physical Review E (accepted).Google Scholar
  12. [12]
    Holmström, L. and Koistinen, P.: “Using additive noise in back-propagation training”, IEEE trans. Neural Networks (1992) 3, 24–38.CrossRefGoogle Scholar
  13. [13]
    Richard, M.D. and Lippmann, R.P.: “Neural network classifiers estimate Bayesian a posteriori probabilities, Neural Computation (1991) 3, 461–483.CrossRefGoogle Scholar
  14. [14]
    Garcia, G. and Whitman, W.W.: “Inversion of a lateral log by using neural networks”, SPE 7th Petroleum Computer Conference, Houston TX, (1992) July 19–22, paper 24454.Google Scholar
  15. [15]
    Wiener, J.M., Rogers, J.A., Rogers, J.R. and Moll, R.F.: “Predicting carbonate permeabilities from wireline logs using a back-propagation neural network”, SEG 61st Annual International Meeting, Houston TX, (1991) November 10–14, p. 285–288.Google Scholar

Copyright information

© Springer-Verlag London Limited 1993

Authors and Affiliations

  • Willem J. M. Epping
    • 1
  • Sandra M. Oudshoff
    • 1
  • Frances V. Abbots
    • 1
  1. 1.Shell Research B.V.Koninklijke/Shell Exploratie en Produktie LaboratoriumAB RijswijkThe Netherlands

Personalised recommendations