Abstract
Accounting for geological scenario uncertainty is one of the contemporary challenges in reservoir prediction modelling. Multi-point statistics approach allows distinguishing between different geological scenarios represented by various training images. A set of generated multipoint statistics realisations are mapped in a model and then classified using a machine learning technique to derive relations between the realisations. Then, the space of the uncertain parameters is searched for multiple history matched realizations to be used to quantify uncertainty in reservoir predictions.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Kanevski, M., Pozdnoukhov, A., & Timonin, V. (2009). Machine learning for spatial environmental data. Theory, applications and software (p. 377). Lausanne Switzerland: EPFL Press.
Mohamed, L., Christie, M., & Demyanov, V. (2010). Comparison of stochastic sampling algorithms for uncertainty quantification. SPE Journal, 15(1), 31–38.
Park, H., Scheidt, C., Fenwick, D., Boucher, A., & Caers, J. (2013). History matching and uncertainty quantification of facies models with multiple geological interpretations. Submitted to Computational Geosciences, February (online).
Remy, N., Boucher, A., & Jianbing, W. (2009). Applied geostatistics with SGEMS (p. 264). Cambridge: Cambridge University Press.
Rojas T., Demyanov V., Christie M., & Arnold D. (2011). Use of geological prior information in reservoir facies modelling. Proceedings of IAMG, Salzburg. http://www.cogeo.at/publications/iamg2011/IAMG2011_Salzburg_093.pdf
Scheidt, C., & Caers, J. (2008). A new method for uncertainty quantification using distances and kernel methods: Application to a deepwater turbidite reservoir. SPE Journal, 14(4), 680–692.
Strebelle, S. (2002). Conditional simulation of complex geological structures using multiple-points statistics. Mathematical Geology, 34(1), 1–21.
Wickelmaier, F. (2003). An Introduction to MDS (p. 26). Sound Quality Research Unit, Aalborg University, Denmark.
Acknowledgments
The funding for this work was provided by the industrial sponsors of the Heriot-Watt Uncertainty Project. We would like to thank J. Caers and Stanford University for sharing the MDS use of SGems software [4] for MPS simulations and for providing Stanford VI case study. We would like to thank M. Kanevski and University of Lausanne for using MLOffice for neural network and SVM modelling [1]. We appreciate Epistemy for providing Raven history-matching and uncertainty quantification software. We appreciate Schlumberger for providing of Eclipse reservoir simulator.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Rojas, T., Demyanov, V., Christie, M., Arnold, D. (2014). Learning Uncertainty from Training Images for Reservoir Predictions. In: Pardo-Igúzquiza, E., Guardiola-Albert, C., Heredia, J., Moreno-Merino, L., Durán, J., Vargas-Guzmán, J. (eds) Mathematics of Planet Earth. Lecture Notes in Earth System Sciences. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-32408-6_35
Download citation
DOI: https://doi.org/10.1007/978-3-642-32408-6_35
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-32407-9
Online ISBN: 978-3-642-32408-6
eBook Packages: Earth and Environmental ScienceEarth and Environmental Science (R0)