Application of Neural Networks in Determining Petrophysical Properties from Seismic Survey

  • Bambang Widarsono
  • Suprajitno Munadi
  • Fakhriyadi Saptono
Part of the Studies in Fuzziness and Soft Computing book series (STUDFUZZ, volume 80)


The history of reservoir characterization has shown that considerable efforts have been devoted to establishing knowledge over inter-well correlation of reservoir rock petrophysical properties. Recent developments in seismic technology have attracted our attention back to the possibility of using seismic attributes for determining reservoir rock properties in general. This is to deviate from the traditional use of seismic data merely as support in describing reservoir structures. Recent research and investigations have produced some encouraging progresses.

Recently, a technique for modeling correlation between potentially usable seismic attributes such as P-wave velocity, acoustic impedance, and Poisson ratio, on one side, and petrophysical properties such as porosity and water saturation, on the other side, using data from well-logs, core, and well production tests has been proposed. The plan to apply the technique on real seismic data from a limestone reservoir in Java had revealed potential error that may result in the estimation of the petrophysical properties. The principal problem encountered was in the form of data absence, both from well-log and seismic, which could end up the application of the model in failure. A consideration over the matter had prompted the attention toward soft computing as an approach to minimize the potential error.

Soft computing, particularly neural networks, as a pattern-recognition approach suits well with the task of minimizing the potential error mentioned above. The approach has proved itself useful in various stages of the work especially in providing data otherwise difficult to obtain reliably. The support ranges from assistance in estimating missing log data to generation of Poisson ratio maps required as a support to the acoustic impedance map provided by seismic processing. Use of the estimated data was justified by validation at wells. By applying the technique previously mentioned, with additional support by data from geological and engineering analyses, maps of porosity and water saturation have been produced for the reservoir. Application of neural networks has proved that problems in seismic-based reservoir characterization can be reduced significantly.


Water Saturation Poisson Ratio Acoustic Impedance Reservoir Rock Seismic Survey 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Biot, M.A. (1956). Theory of Propagation of Elastic Waves in A Fluid-saturated Porous Solid. II. Higher Frequency Range. J. Acoust. Soc. Am., 28: 179–191.CrossRefGoogle Scholar
  2. Bishop, C.M. (1995). Neural Network for Pattern Recognition. Oxford University Press, London.Google Scholar
  3. Crampin, S., Evans, R., B., Doyle, M., Davis, J.P., Yegorkina, G.V. and Miller, A. (1980). Observations of Dilatancy-induced Polarization Anomalies and Earthquake Prediction. Nature, 286: 874–877.CrossRefGoogle Scholar
  4. Domenico, S.N. (1976). Effect of Brine-gas Mixture on Velocity in An Unconsolidated Sand Reservoir. Geophysics, 41: 882–894.CrossRefGoogle Scholar
  5. Gassmann, F. (1951) Elastic Waves Through a Packing of Spheres. Geophysics, 16: 673–685.CrossRefGoogle Scholar
  6. Gregory, A.R. (1976). Fluid Saturation Effects on Dynamic Elastic Properties of Sedimentary Rocks. Geophysics, 41, 895–924.CrossRefGoogle Scholar
  7. King, G. (1996). 4-D Seismic Improves Reservoir Management Decisions. Part I. World Oil, March.Google Scholar
  8. King, M.S. (1966). Wave Velocities in Rocks as a Function of Changes in Overburden Pressure and Pore Fluid saturants. Geophysics, 31: 50–73.CrossRefGoogle Scholar
  9. Munadi, S. & Saptono, F. (2000). Rock elastic compressibility as a potential indicator for gas detection in limestone. (in Bahasa Indonesia), Proceeding, 18th National Symposium on Physics, April, 59 — 63.Google Scholar
  10. Murray, R.C. (1960). Origin of porosity in carbonate rocks. Journal of Sedimentary Petrology, vol. 30 no.1: 59 –84.Google Scholar
  11. Shuey, R.T. (1985). A simplification of the Zoeppritz equations. Geophysics, 50: 609–614.CrossRefGoogle Scholar
  12. Widarsono, B. and Saptono, F. (2000a). A New Method in Preparing Laboratory Core Acoustic Data for Assisting Seismic-based Reservoir Characterization, Proceedings, extended abstract presented at the 2000 Symposium of Society of Core Analyst (SCA/SPWLA), Abu Dhabi.Google Scholar
  13. Widarsono, B. and Saptono, F. (2000b). A New Approach in Processing Core and Log Data for Assisting Seismic-based Mapping of Porosity and Water saturation, Proceedings, presented at the 2000 EAGE Conference, Petrophysics meets Geophysics, Paris, France.Google Scholar
  14. Widarsono, B. and Saptono, F. (1997). Acoustic measurement in laboratory: a support in predicting porosity and fluid saturation from seismic survey. (in Bahasa Indonesia) (1997) Proceedings, Symposium and 5th Congress of Association of Indonesian Petroleum Experts (IATMI), Jakarta, Indonesia.Google Scholar
  15. Wong, P.M., Taggart, I.J. and Jian, F.X. (1995). A critical comparison of neural networks and discriminant analysis in lithofacies, porosity and permeability predictions. Journal of Petroleum Geology, 18(2): 191–206.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Bambang Widarsono
    • 1
  • Suprajitno Munadi
    • 1
  • Fakhriyadi Saptono
    • 1
  1. 1.PPPTMGB “LEMIGAS” Research and Development Center for Oil and Gas TechnologyJakartaIndonesia

Personalised recommendations