Typing pre-Jurassic base rocks by core data and predicting rocks composition by using neural simulation based on Self-Organizing Maps

   The article describes the study of the pre-Jurassic base rocks in the territory of the Kirilkinskaya area of Uvat district in the south of Tyumen region. It was demonstrated that in order to predict net reservoirs in the interwell space within the pre-Jurassic rock complex using 3D seismic CDP data, correct tie-in of the wave field with the material composition (net reservoir vs. non-reservoir) of the rocks is needed. Since the pre-Jurassic interval is usually only fragmentarily studied by the core (at the top and at the bottomhole), the article considers the option of using neural simulation technology based on well logging parameters to restore the material composition of the pre-Jurassic rocks. Since the approaches to the restoration of the material composition of rocks according to well logging data are based on a set of quantitative indicators of the curves for each type of rocks, the approach of dividing the preJurassic rocks into petrotypes is of great importance. In this study, the petrotypes were separated not only on the basis of the material composition of rocks, but the reservoir properties and logging-based properties were also taken into account. Logging-based material composition was estimated in several stages. At the first stage, petrotypes were separated from core data, which allowed to group all types of rocks described in the wells into six main petrotypes. Then, for each petrotype, based on the analysis of log-log cross-plots, a set of optimal logging parameters was identified. This allowed running a neural simulation based on Self-Organizing Maps and restoring the material composition of the pre-Jurassic complex for further net reservoir prediction from seismic data.

1. Moiseev, S. A., Elisheva, O. V., Ryzhkova, S. V., & Grekova, L. S. (2005). Resursy yuga Zapadnoy Sibiri. Neftegazovaya vertikal', (7), 2005, pp. 38-41. (In Russian).

2. Zubkov, M. Yu., & Sitdikov, A. Sh. (1994). Vtorichnye kollektory v doyurskom komplekse Ur'evskogo mestorozhdeniya. Geologiya nefti i gaza, (4), pp. 5-9. (In Russian).

3. Bespalova S. N., & Bakuev, O. V. (1995). Evaluation of fault effect on geological features of pools and gas fields reservoir productivity in Western Siberia. Geologiya nefti i gaza, (7), pp. 16-21. (In Russian).

4. Gerashchenko, A. A. (2003). Petrograficheskie i geokhimicheskie osobennosti porod DYuK Tevlinsko-Russkinskogo mestorozhdeniya. Puti realizatsii neftegazovogo potentsiala KhMAO: sbornik trudov 7-y nauchno-prakticheskoy konferentsii "Puti realizatsii neftegazovogo potentsiala KhMAO". Khanty-Mansiysk, December, 02-05, 2003. Khanty-Mansiysk, IzdatNaukaServis Publ., рр. 229-232. (In Russian).

5. Ukhlova, G. D., Varlamov, S. N., & Kos, I. M. (2006). Postroenie geologicheskikh modeley produktivnykh plastov neokoma tsentral'noy chasti Zapadno-Sibirskoy NGP po kompleksu atributov seysmicheskoy zapisi. Geology, Geophysics and Development of Oil and Gas Fields, (10), pp. 78-91. (In Russian).

6. Bekrenev, R. K., Dugarov, G. A., & Nefedkina, T. V. (2021). Testing of an optimization algorithm for AVAZ inversion on synthetic dataset. Interexpo GEO-Siberia, 2(1), pp. 354-361. (In Russian). DOI: 10.33764/2618-981X-2021-2-1-354-361

7. Marmalevsky, N. Ya., Khromova, I. Yu., & Gornyak, Z. V. (2011). Examples of the use of duplex wave migration for the study of fracture zones. Zbirnik naukovikh prats UkrDGZI, (4), pp. 162-170. (In Russian).

8. Kantemirov, Yu. D., Khabarov, A. V., Oshnyakov, I. O., Koshelev, M. B., Akinshin, A. V., & Kalabin, A. A. (2021). Using methods of computer-aided learning to search for skipped intervals and forecast filtration and capacity properties. Karotazhnik, (8 (314)), pp. 17-26. (In Russian).

9. Izmailov, K. K., Novikova, M. S., Dmitrievsky, M. V., Kantemirov, Yu. D., & Akinshin, A. V. (2021). Up-to-date opportunities of computer-aided learning in well logging data interpretation. Karotazhnik, (8 (314)), pp. 5-16. (In Russian).

10. Kohonen, T. (2001). Self-organizing maps. 3rd edition. Berlin [et al.], Springer, 502 p. (In English). DOI: 10.1007/978-3-642-56927-2

11. Rodina, S. N., & Silkin, K. Yu. (2007). Application of artificial neural net-works for well-log data interpretation. Proceedings of Voronezh State University. Series: Geology, (2), pp. 184-188. (In Russian).

12. Aristodemou, E., Pain, C., Oliveira, C., Goddard, T., & Harris, Ch. (2005). Inversion of nuclear well-logging data using neural networks. Geophysical Prospecting, 53 (1), pp. 103-120. (In English). DOI: 10.1111/j.1365-2478.2005.00432.x

13. Chang, H.-Ch., Kopaska-Merkel, D. C., Chen, H.-Ch., & Durrans, S. R. (2000). Lithofacies identification using multiple adaptive resonance theory neural net-works and group decision expert system. Computer & Geoscience, 26(5), pp. 591-601. (In English). DOI: 10.1016/S0098-3004(00)00010-8