Application of objective intelligence for detailed study of hard-to-recover hydrocarbon reserves in source-rock formations (for discussion)

UDK: 557.7.02:004

DOI: 10.24887/0028-2448-2026-4-29-35

Key words: objective intelligence, well log correlation, source-rock formation, hard-to-recover reserves, core geochemical studies, well geophysical logging

Authors: .S. Gutman (IPNE LLC, RF, Moscow); E.V. Kozlova (Skolkovo Institute of Science and Technology, RF, Moscow); M.Yu. Lobova (Bank DOM.RF JSC, RF, Moscow); A.S. Persidskaya (IPNE LLC, RF, Moscow); G.N. Potemkin (IPNE LLC, RF, Moscow; Sergo Ordzhonikidze Russian State University for Geological Prospecting, RF, Moscow); S.A. Rudnev (IPNE LLC, RF, Moscow; Sergo Ordzhonikidze Russian State University for Geological Prospecting, RF, Moscow); M.Yu. Spasennykh (Skolkovo Institute of Science and Technology, RF, Moscow); V.M. Staroverov (IPNE LLC, RF, Moscow; Lomonosov Moscow State University, RF, Moscow); A.V. Shubina (State Comission on Mineral Resources, RF, Moscow)

The paper examines methodological and practical aspects of building artificial intelligence (AI) systems aimed at reducing the impact of the human factor in knowledge-base construction and result interpretation. It is shown that current Russian state standards in the field of AI mainly regulate technical implementation principles while leaving the methodology for knowledge-base filling largely unaddressed, thereby increasing the risks of errors and vulnerabilities. As an alternative, the concept of objective intelligence (OI) is proposed  an approach based on the input of verifiable data, the use of validated regularities, and formalized rules. OI is implemented through an explicitly designed problem-solving model that includes a data input and validation module, a knowledge base, an inference engine, and a results interpretation module. The practical feasibility of the approach is demonstrated using the certified AutoCorr software, applied to construct objective geological models of structurally complex oil and gas-bearing formations. The main workflow stages are described: automated well-log correlation using weighted parameters of the logging-method set, quality control via correlation misfit assessment, and interactive correction. Using the Bazhenov formation in Western Siberia as an example, the study identifies potentially productive intervals and estimates porosity based on the search for multidimensional statistical relationships derived from integrated well-log data and core geochemical analyses. The presented OI workflow implemented in AutoCorr can be used to accelerate interpretation and substantiate reserve-estimation parameters for complex source-rock formations in poorly studied petroleum provinces.

References

1. GOST R 71476-2024 (ISO/MEK 22989:2022). Artificial intelligence. Artificial intelligence concepts and terminology.

2. Gutman I.S., Korrelyatsiya razrezov skvazhin slozhnopostroennykh neftegazonosnykh ob»ektov i geologicheskaya interpretatsiya ee rezul’tatov (Correlation of well sections of complex oil and gas objects and geological interpretation of its results), Moscow: ESOEN Publ., 2022, 336 p.

3. Spasennykh M., Maglevannaia P., Kozlova E. et al., Geochemical trends reflecting hydrocarbon generation, migration and accumulation in unconventional reservoirs based on pyrolysis data (on the example of the Bazhenov formation), Geosciences, 2021, No. 11, DOI: https://doi.org/10.3390/geosciences11080307

4. Kozlova E.V., Fadeeva N.P., Kalmykov G.A. et al., Geochemical technique of organic matter research in deposits enriched in kerogen (the Bazhenov formation, West Siberia) (In Russ.), Vestnik MGU. Seriya 4. Geologiya = Moscow University Geology Bulletin, 2015, no. 5, pp. 44–54.

5. Gutman I.S., Baturin A.Yu., Obgolts A.A. et al., Signs of hard-to-recover unconventional oil-producing rocks established in the process of its integrated study at the exploration and exploitation stages (In Russ.), Neftyanoe khozyaystvo = Oil Industry, 2023, No. 4, pp. 20-25, DOI: https://doi.org/10.24887/0028-2448-2023-4-20-25

6. Kozlova E.V., Bulatov T.D., Leushina E.A. et al., Geokhimicheskaya model’ netraditsionnogo kollektora v paleogenovykh otlozheniyakh Predkavkaz’ya (Geochemical model of an unconventional reservoir in the Paleogene deposits of the Ciscaucasia), Proceedings of 24th International Scientific and Practical Conference on the Exploration and Development of Oil and Gas Fields “Geomodel’ 2022”, Gelendzhik, 5–8 September 2022.