Systems and Means of Informatics

2017, Volume 27, Issue 3, pp 74-87

GENERATION OF EXPERTLY-INTERPRETED MODELS FOR PREDICTION OF CORE PERMEABILITY

А. М. Bochkarev
I. L. Sofronov
V. V. Strijov

Abstract

This article is devoted to prediction of core permeability. Permeability is one of the main properties for estimation of filtration of gas and liquid in core. To build a permeability model, porosity, density, depth of measurement, and other core physical properties are used. An algorithm for choosing the optimal prediction model is proposed. The model of superpositions of expertly-defined functions is suggested. The proposed method is a superposition of previously obtained optimal expetly-defined functions and a two-layer neural network. The experiment on core analysis, aero- and hydrodynamics datasets was conducted. During the experiment, the optimal expertly-interpreted models for all datasets were derived. The suggested approach is compared to other methods for choosing models, such as Lasso regression, support vector regression (SVR), gradient boosting, and neural network. The error and optimal parameters estimation was conducted using cross-validation. The experiment showed that the proposed approach is competitive with other state-of-the-art methods. Moreover, the number of neurons is significantly reduced with the use of superpositions of expertly-defined functions.

[+] References (23)

Bear, J. 2013. Dynamics of fluids in porous media. Courier Corp. 459 p.
Watson, A.T., P. C. Richmond, and P. D. Kerig. 1988. A regression-based method for estimating relative permeabilities from displacement experiments. SPE Reservoir Eng. 3(3):953-958.
Ibrahim, M.N., andL. F. Koederitz. 2008. Two-phase relative permeability prediction using a linear regression model. SPE Eastern Regional Meeting. Society of Petroleum Engineers. ID: SPE-65631. 11 p. doi: 10.2118/65631-MS.
Cozzi, M., L. Ruvo, and P. Scaglioni. 2007. Core-data preprocessing to improve permeability-log estimation. SPE Reserv. Eval. Eng. 10(5):563-570.
Akande, K. O., S. O. Olatunji, and T. O. Owolabi. 2015. Comparative analysis of feature selection-based machine learning techniques in reservoir characterization. SPE Saudi Arabia Section Annual Technical Symposium and Exhibition. Society of Petroleum Engineers. ID: SPE-178006-MS. 21 p. doi: 10.2118/178006-MS.
Al-Mudhafar, W. J. 2015. Integrating Bayesian model averaging for uncertainty reduction in permeability modeling. Offshore Technology Conference. ID: SPE-25646- MS. 10 p. doi: 10.4043/25646-MS.
Al-Mudhafar, W.J. 2015. Comparison of permeability estimation models through Bayesian model averaging and LASSO regression. Abu Dhabi Petroleum Exhibition and Conference (International). Society of Petroleum Engineers. ID: SPE-177556- MS. 17 p. doi:10.2118/ 177556-MS.
Xue, G., A. Datta-Gupta, and P. Valko. 1997. Optimal transformations for multiple regression: Application to permeability estimation from well logs. SPE Formation Eval. 12(2):85-94.
Al-anazi, A.F., and I.D. Gates. 2010. Support-vector regression for permeability prediction in a heterogeneous reservoir: A comparative study. SPE Reserv. Eval. Eng. 13(3):485-495.
Guler, B., T. Ertekin, and A. S. Grader. 2003. An artificial neural network based relative permeability predictor. J. Can. Petrol. Technol. 42(4):49-57. doi: 10.2118/03 04-02.
Aminian, K., S. Ameri, and A. Oyerokun. 2003. Prediction of flow units and permeability using artificial neural networks. SPE Western Regional/AAPG Pacific Section Joint Meeting. Society of Petroleum Engineers. ID: SPE-83586-MS. 7 p. doi: 10.2118/83586-MS.
Shokir, E. M., A. A. Alsughayer, and A. Al-Ateeq. 2006. Permeability estimation from well log responses. J. Can. Petrol. Technol. 45(11):41-46. doi: 10.2118/06-11-05.
Maslennikova, Y. 2013. Permeability prediction using hybrid neural network modelling. SPE Annual Technical Conference and Exhibition. Society of Petrolum Engineers. ID: SPE-167640-STU. 6 p. doi: 10.2118/167640-STU.
Rudoy, G. I., and V. V. Strijov. 2013. Algoritmy induktivnogo porozhdeniya superpo- zitsiy dlya approksimatsii izmeryaemykh dannykh [Algorithms for inductive generation of superpositions for approximation of experimental data]. Informatika i ee Primeneniya - Inform. Appl. 7(1):44-53.
Airfoil Self Noise. UCI Machine Learning Repository. Available at: https:// archive.ics.uci.edu/ml/datasets/Airfoil+Self-Noise (accessed August 11, 2017).
Concrete Compressive Strength Data Set. UCI Machine Learning Repository. Available at: https: //archive.ics.uci.edu/ml/datasets/Concrete+Compressive+Strength (accessed August 11, 2017).
Yacht Hydrodynamics Data Set. UCI Machine Learning Repository. Available at: https://archive.ics.uci.edu/ml/datasets/Yacht+Hydrodynamics (accessed August 11, 2017).
Bishop, C. M. 2006. Pattern recognition and machine learning. Berlin: Springer. 738 p.
Ghodrat, M. A., T. Givargis, and A. Nicolau. 2005. Equivalence checking of arithmetic expressions using fast evaluation. Conference (International) on Compilers, Architectures and Synthesis for Embedded Systems Proceedings. ACM. 147-156.
Experiment code. Available at: https://sourceforge.net/p/mlalgorithms/code/HEAD/ tree/Group274/Bochkarev2016PermeabilityEstimation/code/Bochkarev2016Permea bilityEstimation.ipynb?format=raw (accessed August 11, 2017).
Lopez, R., E. Balsa-Canto, and E. Onate. 2008. Neural networks for variational problems in engineering. Int. J. Numer. Meth. Eng. 75(11): 1341-1360.
Yeh, I.C. 1998. Modeling of strength of high-performance concrete using artificial neural networks. Cement Concrete Res. Elsevier. 28(12): 1797-1808.
Ortigosa, I., R. Lopez, and J. Garcia. 2007. A neural networks approach to residuary resistance of sailing yachts prediction. Conference (International) on Marine Engineering MARINE Proceedings. 250.

[+] About this article