E.N. Ramazanova1

1 Financial University under the Government of the Russian Federation (Moscow, Russia)
1 Bauman Moscow State Technical University (Moscow, Russia)
1 enramazanova@fa.ru

Formulation of the problem. Geological characteristics of deposits have a significant impact on the selection and application of oil production technologies, and their correct assessment can help optimize these processes. The use of machine learning algorithms makes it possible to more accurately predict and analyze the relationship between geological parameters and the efficiency of production technologies, which in turn can lead to increased productivity and reduced costs. In this connection, the formulation of the problem within the framework of this article is the need to assess the influence of geological characteristics on oil recovery using machine learning methods.

Target – development of a neural network capable of analyzing geological data and predicting its impact on oil recovery.

Results. A database with the geological characteristics of oil fields was created and processed. A neural network was developed and trained based on the provided geological data to predict oil recovery. An analysis of the influence of individual geological characteristics on the productivity of an oil reservoir was carried out. The key factors influencing oil recovery of the reservoir have been identified and their significance has been assessed.

Practical significance. Possibility of improving reservoir oil recovery forecasting using neural networks. The development of such a model will make it possible to more effectively manage the oil production process, optimize work in the fields, as well as reduce operating costs and increase the efficiency of hydrocarbon production.

Ramazanova E.N. Assessing the influence of geological characteristics on oil recovery using neural networks. Nonlinear World. 2025. V. 23. № 3. P. 81–87. DOI: https:// doi.org/10.18127/ j20700970-202503-10 (In Russian)

1. Ibatullin R.R., Ibrarimov N.G., Tahautdinov Sh.F., Hisamov R.S. Uvelichenie nefteotdachi na pozdnej stadii razrabotki mestorozhdenij. Teoriya. Metody. Praktika. M.: OOO «Nedra-Biznescentr». 2004. 292 s. (In Russian).
2. Lenchenkova L.E. Povyshenie nefteotdachi plastov fiziko-himicheskimi metodami. M.: OOO «Nedra-Biznescentr». 1998. 394 s. (In Russian).
3. Surguchev M.V. Vtorichnye i tretichnye metody uvelicheniya nefteotdachi plastov. M.: Nedra. 1985. 308 s. (In Russian).
4. Zheltov Yu.P. Razrabotka neftyanyh mestorozhdenij. M.: Nedra. 1986. 332 s. (In Russian).
5. Mandrik I.E., Shahverdiev A.H., Sulejmanov I.V. Ocenka i prognoz nefteotdachi na osnove modelirovaniya nejronnymi setyami. Neftyanoe hozyajstvo. 2005. №10. S. 36–39 (In Russian).
6. Murav'eva E.A., Konnov V.A. Ocenka vozmozhnosti primeneniya iskusstvennyh nejronnyh setej dlya prognozirovaniya dobychi nefti. Informacionnye tekhnologii. Problemy i resheniya. 2020. № 3(12). S. 59–63 (In Russian).
7. Hasanov I.I., Hasanova Z.R., Shakirov R.A., Nedel'chenko O.I. Obzor primeneniya nejrosetej v oblasti dobychi i transporta nefteproduktov i uglevodorodnogo syr'ya. 2022. № 3–4. S. 11–15.
8. Evsyutkin I.V., Markov N.G. Glubokie iskusstvennye nejronnye seti dlya prognozirovaniya znachenij debita dobyvayushchih skvazhin. Izv. Tomskogo politekhnicheskogo universiteta. Inzhiniring georesursov. 2020. T. 331. № 11. S. 88–95 (In Russian).
9. Lebedeva A.Yu., Skorov D.S. Postroenie modeli iskusstvennoj nejronnoj seti dlya vybora optimal'nogo metoda uvelicheniya nefteotdachi i razrabotka 3D modeli dlya analiza veroyatnosti primeneniya metodov vozdejstviya na mestorozhdenii. V kn.: Novye idei v naukah o Zemle. Materialy XIV Mezhdunar. nauch.-prakt. konf.: v 7 tomah. 2019. S. 449–451 (In Russian).
10. Gimatudinov Sh.K., SHirkovskij A.I. Fizika neftyanogo i gazovogo plasta Izd. 4-e., stereotip. M.: Al'yans. 2005. 311 s. (In Russian).
11. Demyanenko N.A., Pysenkov V.G., Lymar' I.V. i dr. Fiziko-himicheskie metody uvelicheniya nefteotdachi plastov. Neftyanoe hozyajstvo. 2004. № 11. S. 22–23 (In Russian).
12. Dzhulli A., Pal S. Biblioteka Keras – instrument glubokogo obucheniya. Packt Publishing. 2017. 294 c. (In Russian).