Advertisement

An AI-based workflow for estimating shale barrier configurations from SAGD production histories

  • Jingwen Zheng
  • Juliana Y. Leung
  • Ronald P. Sawatzky
  • Jose M. Alvarez
Original Article
  • 182 Downloads

Abstract

An artificial intelligence (AI)-based workflow is deployed to develop and test procedures for estimating shale barrier configurations from SAGD production profiles. The data employed in this project are derived from a set of synthetic SAGD reservoir simulations based on petrophysical properties and operational constraints representative of Athabasca oil sands reservoirs. Initially, a two-dimensional reservoir simulation model is employed. The underlying model is homogeneous. Its petrophysical properties, such as the porosity, permeability, initial oil saturation and net pay thickness, have been taken from average values for several pads in Suncor’s Firebag project. Reservoir heterogeneities are simulated by superimposing sets of idealized shale barrier configurations on the homogeneous model. The location and geometry of each shale barrier is parameterized by a unique set of indices. The resulting heterogeneous model is subjected to flow simulation to simulate SAGD production. Next, a two-step workflow is followed: (1) a network model based on AI tools is constructed to match the output of the reservoir simulation (shale indices are inputs, while production rate is the output) for a known training set of shale barrier configurations; (2) for a new SAGD production history generated via reservoir simulation with a shale barrier configuration that is unknown to the AI model generated in Step 1, an optimization scheme based on a genetic algorithm approach is adopted to perturb the shale indices until the difference between the target production history and the production history predicted from the AI model is minimized. A number of cases have been tested. The results show a good agreement between the shale barrier configurations predicted by the AI model with the configurations used to generate production histories in the reservoir simulation model (i.e., the “true” model). Thus, this optimization workflow offers potential to become an alternative tool for indirect inference of the uncertain distribution of shale barriers in SAGD reservoirs from data capturing field performance. This work highlights the potential of an AI-based workflow to infer the presence and distribution of heterogeneous shale barriers from field SAGD production time series data. It presents an innovative parameterization scheme suitable for representing heterogeneous characteristics of shale barriers. If this approach proves to be successful, it could allow the distribution of shale barriers to be inferred together with the impact of these barriers on SAGD performance. This would provide a basis for developing operating strategies to reduce the impact of the barriers.

Keywords

Reservoir engineering Heavy oil recovery processes Genetic Algorithm Data-driven proxy 

List of symbols

C

An objective function that compares the similarity between two production profiles

hi

Historical production profile at time i

i

A single time point (monthly) along the production profile

m

Total number of time points in the production profile

fi

Forecasted production profile at time i

x

Grid size in \(x\)-direction

z

Grid size in \(z\)-direction

Acronyms

AI

Artificial intelligence

ANN

Artificial neural network

BP

Back propagation

ELM

Extreme learning machine

EnKF

Ensemble Kalman filter

GA

Genetic algorithm

LA

Levenberg–Marquardt

LI

Lean zone indicator

MLP

Multilayer perceptron

NMSE

Normalized mean squared error

RBF

Radial basis function

SA

Simulated annealing

SAGD

Steam-assisted gravity drainage

SASP

Simultaneous perturbation stochastic approximation

SI

Shale indicator

SOR

Steam-to-oil ratio

Notes

Acknowledgements

This research was supported by InnoTech Alberta under a PhD Pilot Program administered by the University of Alberta. Academic licenses for STARS are provided by Computer Modeling Group (CMG).

References

  1. 1.
    Akin S (2005) Mathematical modeling of steam assisted gravity drainage. SPE Reserv Eval Eng 8(05):372–376CrossRefGoogle Scholar
  2. 2.
    Amirian E, Leung JY, Zanon S, Dzurman P (2015) Integrated cluster analysis and artificial neural network modeling for steam-assisted gravity drainage performance prediction in heterogeneous reservoirs. Expert Syst Appl 42(2):723–740CrossRefGoogle Scholar
  3. 3.
    Ballester PJ, Carter JN (2007) A parallel real-coded genetic algorithm for history matching and its application to a real petroleum reservoir. J Pet Sci Eng 59(3):157–168CrossRefGoogle Scholar
  4. 4.
    Brun B, Gosselin O, Barker JW (2004) Use of prior information in gradient-based history matching. SPE J 9(01):67–78CrossRefGoogle Scholar
  5. 5.
    Butler R, McNab G, Lo H (1981) Theoretical studies on the gravity drainage of heavy oil during in-situ steam heating. Can J Chem Eng 59(4):455–460CrossRefGoogle Scholar
  6. 6.
    Butler RM (1985) A new approach to the modelling of steam-assisted gravity drainage. J Can Pet Technol 24(03):42–51CrossRefGoogle Scholar
  7. 7.
    Butler RM (1991) Thermal recovery of oil and bitumen. Prentice Hall, Englewood CliffsGoogle Scholar
  8. 8.
    Chen Q, Gerritsen MG, Kovscek AR (2008) Effects of reservoir heterogeneities on the steam-assisted gravity-drainage process. SPE Reserv Eval Eng 11(05):921–932CrossRefGoogle Scholar
  9. 9.
    CMG (2015) STARS: users’ guide, advanced processes & thermal reservoir simulator (version 2015). Computer Modeling Group Ltd, CalgaryGoogle Scholar
  10. 10.
    de Sousa SHG (2007) Scatter search metaheuristic applied to the history matching problem. Paper presented at SPE annual technical conference and exhibition. Anaheim, California, USAGoogle Scholar
  11. 11.
    Deutsch CV, Journel AG (1998) GSLIB: Geostatistical software library and user’s guide. Oxford University Press, New YorkGoogle Scholar
  12. 12.
    Ding S, Guo L, Hou Y (2017) Extreme learning machine with kernel model based on deep learning. Neural Comput Appl 28(8):1975–1984CrossRefGoogle Scholar
  13. 13.
    Elzwayie A, El-Shafie A, Yaseen ZM, Afan HA, Allawi MF (2017) RBFNN-based model for heavy metal prediction for different climatic and pollution conditions. Neural Comput Appl 28(8):1991–2003CrossRefGoogle Scholar
  14. 14.
    Fedutenko E, Yang C, Card C, Nghiem LX (2014) Time-dependent neural network based proxy modeling of SAGD process. Paper presented at the SPE heavy oil conference-Canada, Calgary, Alberta, CanadaGoogle Scholar
  15. 15.
    Feizabadi SA, Zhang XK, YangP (2014). An integrated approach to building history-matched geomodels to understand complex long lake oil sands reservoirs, part 2: simulation. Paper presented at the SPE heavy oil conference-Canada, Calgary, CanadaGoogle Scholar
  16. 16.
    Gao G, Li G, Reynolds AC (2007) A stochastic optimization algorithm for automatic history matching. SPE J 12(2):196–208CrossRefGoogle Scholar
  17. 17.
    Ghasemi M, Whitson CH (2013) Modeling steam-assisted gravity drainage with a black-oil proxy. SPE Reserv Eval Eng 16(02):155–171CrossRefGoogle Scholar
  18. 18.
    Hadamard J (1902) Sur les problèmes aux dérivées partielles et leur signification physique. Princeton Univ Bull 13:49–52Google Scholar
  19. 19.
    Hajizadeh Y, Christie M, Demyanov V (2011) Ant colony optimization for history matching and uncertainty quantification of reservoir models. J Pet Sci Eng 77(1):78–92CrossRefGoogle Scholar
  20. 20.
    Haykin S (2008) Multilayer perceptron. In: Horton MJ, Disanno S (eds) Neural networks and learning machines, 3rd edn. Prentice Hall, Englewood Cliffs, pp 197–199Google Scholar
  21. 21.
    Hiebert AD, Morrish IC, Card C, Ha H, Porter S, Kumar A, Sun F, Close JC (2013) Incorporating 4D seismic steam chamber location information into assisted history matching for A SAGD simulation. Paper presented at the SPE heavy oil conference-Canada, Calgary, CanadaGoogle Scholar
  22. 22.
    IHS Energy (2015) AccuMap software, 321 Inverness Drive South Englewood CO 80112, USA. Retrieved June 25, 2015, from http://www.ihsenergy.com
  23. 23.
    Ito Y, Chen J (2010) Numerical history match of the Burnt Lake SAGD process. J Can Pet Technol 49(5):40–49CrossRefGoogle Scholar
  24. 24.
    Jacquard P (1965) Permeability distribution from field pressure data. Soc Pet Eng J 5(04):281–294CrossRefGoogle Scholar
  25. 25.
    Jia X, Cunha L, Deutsch C (2009) Investigation of a stochastic optimization method for automatic history matching of SAGD processes. J Can Pet Technol 48(01):14–18CrossRefGoogle Scholar
  26. 26.
    Kabanikhin SI (2008) Definitions and examples of inverse and ill-posed problems. J Inverse Ill Posed Probl 16(4):317–357MathSciNetCrossRefMATHGoogle Scholar
  27. 27.
    Kingsford C, Salzberg SL (2008) What are decision trees? Nat Biotechnol 26(9):1011–1013CrossRefGoogle Scholar
  28. 28.
    Kisman K.E. & Yeung K.C. (1995). Numerical study of the SAGD process in the Burnt Lake oil sands lease. Paper presented at the SPE international heavy oil symposium, Calgary, Alberta, CanadaGoogle Scholar
  29. 29.
    Kruger WD (1961) Determining areal permeability distribution by calculations. J Pet Technol 13(07):691–696CrossRefGoogle Scholar
  30. 30.
    Le Ravalec M, Morlot C, Marmier R, Foulon D (2009) Heterogeneity impact on SAGD process performance in mobile heavy oil reservoir. Oil Gas Sci Technol Rev IFP 64(4):469–476CrossRefGoogle Scholar
  31. 31.
    Liu F, Wang J, Morris P (2013) A practical approach to history-matching large, multi-well SAGD simulation models: a mackay river case study. Paper presented at the SPE heavy oil conference-Canada, Calgary, CanadaGoogle Scholar
  32. 32.
    Liu J, Jaiswal A, Yao K, Raghavenda CS (2015) Autoencoder-derived features as inputs to classification algorithms for predicting well failures. Paper presented at the SPE western regional meeting, Garden Grove, California, USAGoogle Scholar
  33. 33.
    Liu N, Oliver DS (2003) Evaluation of Monte Carlo methods for assessing uncertainty. SPE J 8(02):188–195CrossRefGoogle Scholar
  34. 34.
    Liu N, Oliver DS (2005) Ensemble Kalman filter for automatic history matching of geologic facies. J Pet Sci Eng 47(3):147–161CrossRefGoogle Scholar
  35. 35.
    Ma Z, Leung JY, Zanon S, Dzurman P (2015) Practical implementation of knowledge-based approaches for steam-assisted gravity drainage production analysis. Expert Syst Appl 42(21):7326–7343CrossRefGoogle Scholar
  36. 36.
    Maschio C, Vidal AC, Schiozer DJ (2008) A framework to integrate history matching and geostatistical modeling using genetic algorithm and direct search methods. J Pet Sci Eng 63(1):34–42CrossRefGoogle Scholar
  37. 37.
    McDonald JH (2009) Handbook of biological statistics. Sparky House Publishing, BaltimoreGoogle Scholar
  38. 38.
    Mojarad M, Dehghanpour H (2016) Analytical modeling of emulsion flow at the edge of a steam chamber during a steam-assisted-gravity-drainage process. SPE J 21:353–363CrossRefGoogle Scholar
  39. 39.
    Noble WS (2006) What is a support vector machine? Nat Biotechnol 24(12):1565–1567CrossRefGoogle Scholar
  40. 40.
    Patel RG, Rahim S, Li Z (2015) Initial sampling of ensemble for steam-assisted-gravity-drainage-reservoir history matching. J Can Pet Technol 54:424–441CrossRefGoogle Scholar
  41. 41.
    Pooladi-Darvish M, Ali SM (1994). Steam heating of fractured formations containing heavy oil: basic premises and a single-block analytical model. Paper presented at the SPE annual technical conference and exhibition, New Orleans, Louisiana, USAGoogle Scholar
  42. 42.
    Ouenes A, Brefort B, Meunier G, Dupere S (1993). A new algorithm for automatic history matching: application of simulated annealing method (SAM) to reservoir inverse modeling. Paper SPE 26297Google Scholar
  43. 43.
    Oyedotun OK, Khashman A (2017) Deep learning in vision-based static hand gesture recognition. Neural Comput Appl 28(12):3941–3951CrossRefGoogle Scholar
  44. 44.
    Reis JC (1992) A steam-assisted gravity drainage model for tar sands: linear geometry. J Can Pet Technol 31(10):14–20CrossRefGoogle Scholar
  45. 45.
    Rose PE (1993). The steam-assisted gravity drainage of oil sand bitumen. Ph.D. thesis, Department of Chemicals and Fuels Engineering, The University of Utah, USAGoogle Scholar
  46. 46.
    Scheidt C, Caers J (2009) Representing spatial uncertainty using distances and kernels. Math Geosci 41(4):397–419CrossRefMATHGoogle Scholar
  47. 47.
    Sharma J, Gates ID (2010) Multiphase flow at the edge of a steam chamber. Can J Chem Eng 88(3):312–321Google Scholar
  48. 48.
    Suncor Energy (2012) Suncor Firebag 2012 ERCB performance presentation. Retrieved July 16, 2015, from https://www.aer.ca/documents/oilsands/insitu-presentations/2012AthabascaSuncorFirebagSAGD8870.pdf
  49. 49.
    Suncor Energy (2013) Suncor Firebag 2013 ERCB performance presentation. Retrieved July 16, 2015, from https://www.aer.ca/documents/oilsands/insitu-presentations/2013AthabascaSuncorFirebagSAGD8870.pdf
  50. 50.
    Suncor Energy (2014) Suncor Firebag 2014 AER Performance Presentation. Retrieved July 16, 2015, from https://www.aer.ca/documents/oilsands/insitu-presentations/2014AthabascaSuncorFirebagSAGD8870.pdf
  51. 51.
    The MathWorks Inc., MATLAB and Global optimization toolbox release 2014b, Natick, Massachusetts, USAGoogle Scholar
  52. 52.
    TOP Analysis (2015) TOP Analysis software, 144 - 4th Avenue SW Calgary AB T2P 3N4, Canada. Retrieved Aug 20, 2015, from http://www.topanalysis.com
  53. 53.
    Wang C, Leung J (2015) Characterizing the effects of lean zones and shale distribution in steam-assisted-gravity-drainage recovery performance. SPE Reserv Eval Eng 18:329–345CrossRefGoogle Scholar
  54. 54.
    Whitley D (1994) A genetic algorithm tutorial. Stat Comput 4:65–85CrossRefGoogle Scholar
  55. 55.
    Yang C, Nghiem LX, Card C, Bremeier M (2007). Reservoir model uncertainty quantification through computer-assisted history matching. In: SPE annual technical conference and exhibition. Anaheim, California, USAGoogle Scholar
  56. 56.
    Yang G, Butler R (1992) Effects of reservoir heterogeneities on heavy oil recovery by steam-assisted gravity drainage. J Can Pet Technol 31(08)Google Scholar
  57. 57.
    Yu T, Wilkinson D, Castellini A (2008) Constructing reservoir flow simulator proxies using genetic programming for history matching and production forecast uncertainty analysis. J Artif Evol Appl 2008:2Google Scholar
  58. 58.
    Zhang XK, Feizabadi SA, Yang P (2014). An integrated approach to building history-matched geomodels to understand complex long lake oil sands reservoirs, part 1: geomodeling. Paper presented at the SPE heavy oil conference-Canada, Calgary, CanadaGoogle Scholar
  59. 59.
    Zhang L, Tian F (2014) Performance study of multilayer perceptrons in a low-cost electronic nose. IEEE Trans Instrum Meas 63(7):1670–1679CrossRefGoogle Scholar
  60. 60.
    Zhang L, Zhang D (2016) Robust visual knowledge transfer via extreme learning machine-based domain adaptation. IEEE Trans Image Process 25(10):4959–4973MathSciNetCrossRefGoogle Scholar
  61. 61.
    Zhang L, Zhang D (2017) Evolutionary cost-sensitive extreme learning machine. IEEE Transon Neural Netw Learn Syst 28(12):3045–3060CrossRefGoogle Scholar
  62. 62.
    Zheng J, Leung JY, Sawatzky RP, Alvarez JM (2016) A proxy model for predicting SAGD production from reservoirs containing shale barriers. Paper presented at the SPE Canada heavy oil technical conference, Calgary, CanadaGoogle Scholar

Copyright information

© The Natural Computing Applications Forum 2018

Authors and Affiliations

  1. 1.University of AlbertaEdmontonCanada
  2. 2.InnoTech AlbertaEdmontonCanada

Personalised recommendations