Optimization of the Algorithm for Searching for the Most Reliable Implementations of a Geological Model

Author details

Abstract

One of the most important tasks in the oil and gas industry is to predict the values of geological parameters of productive formations in the interwell space. The accuracy of estimating the effective oil-saturated thickness or the values of filtration and capacity properties at the locations of the project well stock directly affects the efficiency of oil and gas assets development and economic indicators. The task of predicting the values of geological parameters is complicated by the fact that the geological environment has been studied by wells in fragments, and sources of information about the interwell space, despite continuous technological progress, have limited accuracy. In addition, the real geological structure in most cases is much more complex than our understanding of it. The vertical and lateral heterogeneity of productive formations and the high degree of geological properties variability do not allow the effective use of interpolation methods. This paper presents the results of testing the author's methodology aimed at finding the most reliable implementations of a three-dimensional lithology model. The proposed approach is based on the use of Bayesian optimization to determine the most optimal values of variogram ranges along the X and Y axes during the modeling of a three-dimensional lithology cube. The mean absolute error of the predicted total effective thickness of the reservoir, calculated using cross-validation, was employed as the primary metric to evaluate the reliability of the three-dimensional lithology model. The results demonstrate the advantages of applying Bayesian optimization compared to the classical grid search method for parameter optimization. Firstly, the proposed approach enabled the creation of a three-dimensional lithology model with higher predictive capability. Secondly, the developed methodology significantly reduced the computational resources required for the calculations.

Keywords

geostatistics3D geological modelalluvial depositsstochastic modelingcross-validationBayesian optimizationsequential indicator modelinguncertainty assessment

Bibliography

  1. 1.Potapov D.V., Efremova E.I. Analiz geologicheskikh riskov na etape otsenki zapasov s tsel'iu povysheniia effektivnosti osvoeniia aktiva [Analysis of geological risks at the stage of reserves assessment in order to increase the asset development efficiency]. Neftegazovaia geologiia. Teoriia i praktika, 2023, vol. 18, no. 1, pp. 13-29. DOI: 10.17353/2070-5379/2_2023 Scholar
  2. 2.Priezzhev I.I. et al. Priamoi neirosetevoi prognoz kollektorskikh svoistv plasta po dannym seismorazvedki na primere klinoformnykh otlozhenii Zapadnoi Sibiri [Neural network prediction of reservoir properties of the reservoir according to seismic data on the example of clinoform deposits of Western Siberia]. PRONEFT'. Professional'no o nefti, 2023, vol. 8, no. 2, pp. 28-39. DOI: 10.51890/2587-7399-2023-8-2-28-39 Scholar
  3. 3.Egorov D.V., Belozerov B.V. Analiz primenimosti algoritmov mashinogo obuchenia dlia zadach interpolatsii i prognoza geologicheskikh svoistv v mezhskvazhnimom prostranstve [Analysis of machine learning algorithms applicability for tasks of interpolation and geological properties forecasting within interwell space]. PRONEFT. Professional'no o nefti, 2022, no. 4, pp. 13-16. DOI: 10.24887/2587-7399-2018-4-13-16 Scholar
  4. 4.Platov B.V., Khairutdinova R.I., Kadirov A.I. Otsenka tolshchiny malomoshchnykh plastov pri pomoshchi dannykh seismorazvedki na primere tul'sko-bobrikovskikh otlozhenii Respublikii Tatarstan [Evaluating the thickness of thin-bed seams using seismic data on the example of the Tula-Bobrikovian sediments in the Republic of Tatarstan]. Izvestiia vysshikh uchebnykh zavedenii. Geologiya i razvedka, 2021, vol. 63, no. 4, pp. 80-87. DOI: 10.32454/0016-7762-2020-63-4-80-87 Scholar
  5. 5.Galkin S.V. Metodologiya ucheta geologicheskikh riskov na etape poiskov i razvedki neftianykh mestorozhdenii [Accounting methods of geological risks on the stage of oil fields exploration]. Nedropol'zovanie, 2012, no. 4, pp. 23-32. Scholar
  6. 6.Dem'ianov V.V., Savel'eva E.A. Geostatistika: teoriia i praktika [Geostatistics: Theory and Practice]. Ed. R.V. Arutiunian, Institut problem bezopasnogo razvitia atomnoi energetiki Rossiiskoi akademii nauk. Moscow: Nauka, 2010, 327 p. Scholar
  7. 7.Materon Zh. Osnovy prikladnoi geostatistiki [Fundamentals of Applied Geostatistics]. Eds. D.Ia. Surazhskii, A.M. Margolin. Moscow: Rossiiskaia gosudarstvennaia biblioteka, 2008, 408 p. Scholar
  8. 8.Emery X. Properties and limitations of sequential indicator simulation. Stochastic Environmental Research and Risk Assessment, 2004, vol. 18, pp. 414-424. DOI: 10.1007/s00477-004-0213-5 Scholar
  9. 9.Deutsch C.V. A sequential indicator simulation program for categorical variables with point and block data: BlockSIS. Computers & Geosciences, 2006, vol. 32, no. 10, pp. 1669-1681. DOI: 10.1016/j.cageo.2006.03.005 Scholar
  10. 10.Juang K.W., Chen Y.S., Lee D.Y. Using sequential indicator simulation to assess the uncertainty of delineating heavy-metal contaminated soils. Environmental Pollution, 2004, Vol. 127, no. 2, pp. 229-238. DOI: 10.1016/j.envpol.2003.07.001 Scholar
Download PDF

Article information

Published
24 February 2025
Issue date
24 February 2025
Issue
Vol. 25, No. 1 (2025)
Share