A data-driven approach to estimating post-discovery parameters of unexplored oilfields

Abstract

Consider a typical situation where an investor is considering acquiring an unexplored oilfield. The oilfield has undergone a preliminary geological and geophysical study in which pre-discovery data such as lithology, depth, depositional system, diagenetic overprint, structural compartmentalization, and trap type are available. In this situation, investors usually estimate production rates using a volumetric approach. A more accurate estimation of production rates can be obtained using analytical methods, which require additional data such as net pay, porosity, oil formation volume factor, permeability, viscosity, and pressure. We call these data post-discovery parameters because they are only available after discovery through exploration drilling. A data-driven approach to estimating post-discovery parameters of an unexplored oilfield is developed based on its pre-discovery data by learning from proven reservoir data. Using the Gaussian mixture model, and a data-driven reservoir typology based on the joint probability distribution of post-discovery parameters is established. We came up with 12 reservoir types. Subsequently, an artificial neural network classification model with the resilient backpropagation algorithm is used to find relationships between pre-discovery data and reservoir types. Based on k-fold cross-validation with k = 10, the accuracy of the classification model is stable with an average of 87.9%. With our approach, an investor considering acquiring an unexplored oilfield can classify the oilfield's reservoir into a particular type and estimate its post-discovery parameters' joint probability distribution. The investor can incorporate this information into a valuation model to calculate the production rates more accurately, estimate the oilfield's value and risk, and make an informed acquisition decision accordingly.

Keywords

Data-driven / Pre-discovery data / Post-discovery parameters / Gaussian mixture model / Artificial neural network

References

[1] G.A.C. Lima, S.B. Suslick, Investment decision in oil and gas projects using real option and risk tolerance models, Int. J. Oil Gas Coal Technol. 1 (2008) 3.

[2] J.L. Paddock, D.R. Siegel, J.L. Smith, Option valuation of claims on real assets: the case of offshore petroleum leases, Q. J. Econ. 103 (1988) 479.

[3] Y. Fan, L. Zhu, A real options based model and its application to China's overseas oil investment decisions, Energy Econ. 32 (2010) 627-637.

[4] L. Zhu, Z. Zhang, Y. Fan, Overseas oil investment projects under uncertainty: how to make informed decisions? J. Pol. Model. 37 (2015) 742-762.

[5] B.-J. Tang, H.-L. Zhou, H. Chen, et al., Investment opportunity in China's overseas oil project: an empirical analysis based on real option approach, Energy Pol. 105 (2017) 17-26.

[6] M.A.G. Dias, Valuation of exploration and production assets: an overview of real options models, J. Pet. Sci. Eng. 44 (2004) 93-114.

[7] C. Park, J.M. Kang, B. Min, Compound real options incorporated with a stochastic approach for evaluating an uncertainty in petroleum exploration, Energy Sources B Energy Econ. Plann. 8 (2013) 252-262.

[8] J. Guedes, P. Santos, Valuing an offshore oil exploration and production project through real options analysis, Energy Econ. 60 (2016) 377-386.

[9] M.N. Fonseca, E. de O. Pamplona, VE. de M. Valerio, et al., Oil price volatility: a real option valuation approach in an African oil field, J. Pet. Sci. Eng. 150 (2017) 297-304.

[10] J. Liu, Z. Li, D. Luo, et al., Study on the valuation method for overseas oil and gas extraction based on the modified trinomial tree option pricing model, Math. Probl Eng. 2020 (2020) 1-15.

[11] Makhotin I, Orlov D, Koroteev D, et al. Machine learning for recovery factor estimation of an oil reservoir: a tool for de-risking at a hydrocarbon asset evaluation. Petroleum. Epub ahead of print November 2021. DOI: 10.1016/j.petlm.2021.11.005.

[12] I. Makhotin, D. Orlov, D. Koroteev, Machine learning to rate and predict the efficiency of waterflooding for oil production, Energies 15 (2022) 1199.

[13] Cao Q., Banerjee R., Gupta S., et al. Data driven production forecasting using machine learning. In: Day 2 Thu, June 02, 2016. SPE. Epub ahead of print June 1, 2016. DOI: 10.2118/180984-MS.

[14] Khan MR, Alnuaim S, Tariq Z, et al. Machine learning application for oil rate prediction in artificial gas lift wells. In: Day 3 Wed, March 20, 2019. SPE. Epub ahead of print March 15, 2019. DOI: 10.2118/194713-MS.

[15] A. Tadjer, A. Hong, R.B. Bratvold, Machine learning based decline curve analysis for short-term oil production forecast, Energy Explor. Exploit. 39 (2021) 1747-1769.

[16] Hultzsch P, Lake LW, Gilbert RB. Decision and risk analysis through the life of the field. In: Hydrocarbon Economics and Evaluation Symposium. Society of Petroleum Engineers. Epub ahead of print April 4, 2007. DOI: 10.2118/107704-MS.

[17] A.K. Dixit, R.S. Pindyck, Investment under Uncertainty, Princeton University Press, Princeton, 1994.

[18] M.R. Walsh, L.W. Lake,A Generalized Approach to Primary Hydrocarbon Recovery of Petroleum Exploration & Production, first ed.ed., ume 4, Elsevier, Amsterdam, 2003.

[19] P.F. Worthington, L. Cosentino, The role of cut-offs in integrated reservoir studies, SPE Reservoir Eval. Eng. 8 (2005) 276-290.

[20] P.F. Worthington, The application of cutoffs in integrated reservoir studies, SPE Reservoir Eval. Eng. 11 (2008) 968-975.

[21] P.F. Worthington, Net pay -what is it? What does it do? How do we quantify it? How do we use it? SPE Reservoir Eval. Eng. 13 (2010) 812-822.

[22] T. Yang, Y. Cao, Y. Wang, et al., Determining permeability cut-off values for net pay study of a low-permeability clastic reservoir: a case study of the Dongying Sag, eastern China, J. Pet. Sci. Eng. 178 (2019) 262-271.

[23] Abdul-Majeed GH, Salman NH. An empirical correlation for oil FVF prediction. J. Can. Pet. Technol.; 27. Epub ahead of print November 1, 1988. DOI: 10.2118/88-06-10.

[24] J. Downie, F.E. Crane, Effect of viscosity on relative permeability, Soc. Petrol. Eng. J. 1 (1961) 59-60.

[25] A. Costa, Permeability-porosity relationship: a reexamination of the Kozeny-Carman equation based on a fractal pore-space geometry assumption, Geophys. Res. Lett. 33 (2006), L02318.

[26] Sadeq QM,Bin Wan Yusoff Wi. Porosity and permeability analysis from well logs and Core in fracture, vugy and intercrystalline carbonate reservoirs. J. Aquacult. Res. Dev. 6. Epub ahead of print 2015. DOI: 10.4172/2155-9546.1000371.

[27] G. Chilingar, W. Long, Correlation between porosity and permeability of carbonate rock reservoirs, Energy Sources, Part A Recover Util Environ Eff 39 (2017) 1116-1117.

[28] S. Elmabrouk, A. Zekri, E. Shirif, The prediction of bubble-point pressure and bubble-point oil formation volume factor in the absence of PVT analysis, Petrol. Sci. Technol. 32 (2014) 1168-1174.

[29] A. Hemmati-Sarapardeh, M. Khishvand, A. Naseri, et al., Toward reservoir oil viscosity correlation, Chem. Eng. Sci. 90 (2013) 53-68.

[30] A. Hemmati-Sarapardeh, A. Shokrollahi, A. Tatar, et al., Reservoir oil viscosity determination using a rigorous approach, Fuel 116 (2014) 39-48.

[31] N.B. Alqahtani, A.A. AlQuraishi, W. Al-Baadani, New correlations for prediction of saturated and undersaturated oil viscosity of Arabian oil fields, J. Pet. Explor. Prod. Technol. 8 (2018) 205-215.

[32] T.W. Keelin, The metalog distributions, Decis. Anal. 13 (2016) 243-277.

[33] F. Durante, C. Sempi, Principles of Copula Theory, Chapman and Hall/CRC, Boca Raton, Florida, July 2016, https://doi.org/10.1201/b18674. Epub ahead of print.

[34] McLachlan G, Peel D. Finite Mixture Models. Hoboken NJ, USA: John Wiley & Sons, Inc. Epub ahead of print September 18, 2000. DOI: 10.1002/0471721182.

[35] A.P. Dempster, N.M. Laird, D.B. Rubin, Maximum likelihood from incomplete data via the EM algorithm, J R Stat Soc Ser B 39 (1977) 1-22.

[36] S.M. Diffey, A.B. Smith, A.H. Welsh, et al., A new REML (parameter expanded) EM algorithm for linear mixed models, Aust. N. Z. J. Stat. 59 (2017) 433-448.

[37] BB de Andrade, G.S. Souza, The EM algorithm for standard stochastic frontier models, Pesqui. Oper. 39 (2019) 361-378.

[38] C. Park, A quantile variant of the expectation-maximization algorithm and its application to parameter estimation with interval data, J. Algorithm Comput. Technol. 12 (2018) 253-272.

[39] B. Panić, J. Klemenc, M. Nagode, Improved initialization of the EM algorithm for mixture model parameter estimation, Mathematics 8 (2020) 373.

[40] D.E. Rumelhart, J.E. McClelland, in: Foundations) ume 1, The MIT Press, Cambridge, 1986. Parallel Distributed Processing: Explorations in the Microstructure of Cognition.

[41] Riedmiller M, Braun H. A direct adaptive method for faster backpropagation learning: the RPROP algorithm. In: IEEE International Conference on Neural Networks. IEEE, pp. 586-591.

[42] M. Riedmiller, Advanced supervised learning in multi-layer perceptrons -from backpropagation to adaptive learning algorithms, Comput. Stand. Interfac. 16 (1994) 265-278.

[43] A.D. Anastasiadis, G.D. Magoulas, M.N. Vrahatis, New globally convergent training scheme based on the resilient propagation algorithm, Neurocomputing 64 (2005) 253-270.

[44] S. Theodoridis, Machine Learning:A Bayesian and Optimization Perspective, first ed.ed., Elsevier, London, 2015.

[45] H. Akaike, A new look at the statistical model identification, IEEE Trans. Automat. Control 19 (1974) 716-723.

[46] R.L. Thorndike, Who belongs in the family? Psychometrika 18 (1953) 267-276.

[47] L. Mouselimis, ClusterR, 2020. https://cran.r-project.org/package=ClusterR.

[48] W.N. Venables, B.D. Ripley, Modern Applied Statistics with S, Springer New York, New York, NY, 2002, https://doi.org/10.1007/978-0-387-21706-2. Epub ahead of print.

[49] F. Günther, S. Fritsch, Neuralnet: training of neural networks, R J 2 (2010) 30.

[50] C. Cortes, V. Vapnik, Support-vector networks, Mach. Learn. 20 (1995) 273-297.

[51] D. Heckerman, D. Geiger, D.M. Chickering, Learning bayesian networks: the combination of knowledge and statistical data, Mach. Learn. 20 (1995) 197-243.

[52] Lê S, Josse J, Husson F. FactoMineR : an R package for multivariate analysis. J. Stat. Software; 25. Epub ahead of print 2008. DOI: 10.18637/jss.v025.i01.