Reservoir Property Estimation Using the Seismic Waveform and Feedforword Neural Networks

  • Ping An
  • Wooil M. Moon
  • Fotis Kalantzis
Part of the Modern Approaches in Geophysics book series (MAGE, volume 21)


Feedforward neural networks are used to estimate reservoir properties. The neural networks are trained using known reservoir properties from well-log data, and the seismic waveform at the well locations. The trained neural networks are then applied to data from the complete seismic survey to generate a map of predicted reservoir properties. The theoretical analysis and tests with synthetic models, show that the neural networks are adaptive to coherent noise, and that random noise in the training samples can increase the robustness of the trained neural networks. This approach has been applied to a mature oil field in Northern Alberta, Canada, to explore the Devonian reef edge oil by estimating the product of porosity and net pay thickness. The resulting prediction map was used to select new well locations and design horizontal well trajectories. Four wells were drilled based on the prediction, and all were successful. This increased oil field production by about 20%.


Seismic Data Feedforward Neural Network Horizontal Well Reservoir Property Seismic Attribute 
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. An, P., and Moon, W. M., 1993, Reservoir characterization using feedforward neural networks: 63rd Ann Internat. Mtg., Soc. Expl. Geophys., Expanded Abstracts, 258–262.Google Scholar
  2. An, P., 1994, The effect of random noise in lateral reservoir characterization using feedforward neural networks: 64th Ann. Internat. Mtg., Soc. Expl. Geophys., Expanded Abstracts, 787–790.Google Scholar
  3. Kalantzis, F., and An, P., 1996, Porosity mapping in a Devonian reef using 3-D seismic waveform and neural networks: 58th Ann. Internat. Mtg., Eur. Assn. Geosci. Eng., Extended Abstracts, X004.Google Scholar
  4. Hornik, K., 1991, Approximation capabilities of multilayer feedforward networks: Neural Networks, 4, 251–257.Google Scholar
  5. Matsuoka, K., 1992, Noise injection into inputs in back-propagation learning: IEEE Trans. Syst., Man, Cybern., 22, 436–440.CrossRefGoogle Scholar
  6. Rumelhart, D. E., and Williams, R. J., 1986, Parallel distributed processing: exploration in microstructures of cognition: Vol. 1, MIT Press, Cambridge, NJ.Google Scholar
  7. Schultz, P. S., Ronen, S., Mantran, P., Hattori, M., and Corbett, C., 1994, Seismic-guided estimation of log properties, Part 3: a controlled study: The Leading Edge, 13, No. 7, 770–776.CrossRefGoogle Scholar
  8. Uffen, J. D., 1995, Swan Hills Unit #1: adding value with seismic through reservoir delineation and characterization: 656 Ann. Internat. Mtg., Soc. Expl. Geophys., Expanded Abstracts, 115–117.Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2003

Authors and Affiliations

  • Ping An
    • 1
  • Wooil M. Moon
    • 2
  • Fotis Kalantzis
    • 3
  1. 1.Schlumberger-GeoQuestHoustonUSA
  2. 2.Department of Geological SciencesThe University of ManitobaWinnipegCanada
  3. 3.Geophysical Specialist (PGeoph.)CalgaryCanada

Personalised recommendations