Neftyanaya Provintsiya
electronic peer-reviewed scholarly publication
Neftyanaya provintsiya No. 3(39), 2024

Reservoir pressure calculation in producing wells using machine learning methods

A.A. Gaysin, N.K. Isroilov, A.Kh. Gilyazov
DOI: https://doi.org/10.25689/NP.2024.3.123-136

Abstract


Reservoir pressure is a critical factor that determines energy potential of a producing reservoir and overall well and reservoir productivity. Reservoir pressure refers to the pressure of hydrocarbons (oil, gas, and water) contained within the reservoir voids. Changes in reservoir pressure should be continuously monitored. Once reservoir pressure decreases, methods for supplementing the natural reservoir energy are applied; particularly, reservoir pressure maintenance. Reservoir pressure decline rate depends on fluid (oil, water, and gas) production rates according to field development plan and implementation of reservoir pressure maintenance methods, if any. According to Operational Guidelines RD 153-39.0-109-01, downhole reservoir pressure measurements should be conducted every six months. However, some of these measurements may turn out to be unreliable, thus reducing the available reservoir energy state data. Moreover, direct measurements of reservoir pressure may require extended well shutdown periods to result in oil production losses and potential technical issues during well startup.
In this paper, machine learning methods for reservoir pressure prediction are considered. The present research effort is peculiar in that it presents comparative analysis of a variety of machine learning methods for specific production target, as well as reveals an optimal set of features for model training.
The results of the present research can be used to analyze the development status of hole sections in absence of reservoir pressure measurements, justify initial data adjustments for reservoir simulation modeling, and to prepare a list of wells for reservoir pressure studies aimed at objective evaluation of producing reservoir energy state while minimizing oil production losses.

Key words:

machine learning, reservoir pressure, downhole pressure, algorithms, features

References

1. Zakharov L.A., Martyushev D.A., Ponomareva I.N. Prognozirovanie dinamicheskogo plastovogo davleniya metodami iskusstvennogo intellekta [Forecasting dynamic reservoir pressure by artificial intelligence methods]. Zapiski Gornogo Instituta [Journal of Mining Institute]. 2022, Vol. 253, pp. 23-32. (in Russian)

2. Kharisov M.N., Yunusova E.A., Maysky R.A. Algoritm kosvennogo opredeleniya plastovogo davleniya s ispolzovaniem metodov data mining [Algorithm for indirect determination of reservoir pressure using data mining methods]. Problemy Sbora, Podgotovki i Transporta Nefti i Nefteproduktov [Problems of Gathering, Treatment and Transportation of Oil and Oil Products]. 2018, Vol. 3, pp. 40-48. (in Russian)

3. Hazim H. Al-Attar. A simple method for estimating average reservoir pressure and well flow efficiency in volumetric oil reservoirs. Second ADRAC, Abu Dhabi, 2014. 10 P. (in English)

4. Kuznetsova E.A., Nikulin S.E., Shilov A.V., Filatov M.A. Opyt primeneniya analiticheskikh metodov opredeleniya plastovogo davleniya [The experience of using analytical methods for determining reservoir pressure]. Neftepromyslovoe Delo [Petroleum Engineering]. 2023, Vol. 6, pp. 12-16. (in Russian)

5. Secondary analysis of electronic health records – MIT critical data. Massachusetts Institute of Technology. 435 P. DOI 10.1007/978-3-319-43742-2. (in English)

6. Hancock, J.T., Khoshgoftaar, T.M. CatBoost for big data: an interdisciplinary review. Journal of Big Data, No.7, 2020, p. 45. (in English)

Authors

A.A. Gaysin, PhD Candidate, Chair of Oil and Gas Fields Development and Operation, Almetyevsk State University of Technology – Higher Petroleum School, TatNIPIneft Institute – PJSC TATNEFT
2, Lenin Str., Almetyevsk, 423462, Russian Federation
E-mail: GaysinAA@tatneft.ru

N.K. Isroilov, PhD Candidate, Chair of Oil and Gas Fields Development and Operation, Almetyevsk State University of Technology – Higher Petroleum School, TatNIPIneft Institute – PJSC TATNEFT
2, Lenin Str., Almetyevsk, 423462, Russian Federation
E-mail: IsroilovNK@tatneft.ru

A.Kh. Gilyazov, PhD Candidate, Chair of Oil and Gas Fields Development and Operation, Almetyevsk State University of Technology – Higher Petroleum School, TatNIPIneft Institute – PJSC TATNEFT
2, Lenin Str., Almetyevsk, 423462, Russian Federation
E-mail: GilyazovAH@tatneft.ru

For citation:

A.A. Gaysin, N.K. Isroilov, A.Kh. Gilyazov Raschot plastovogo davleniya v dobyvayushchikh skvazhinakh pri pomoshchi metodov mashinnogo obucheniya [Reservoir pressure calculation in producing wells using machine learning methods]. Neftyanaya Provintsiya, No. 3(39), 2024. pp. 123-136. DOI https://doi.org/10.25689/NP.2024.3.123-136. EDN LOBMLY (in Russian)

© Non-governmental organization Volga-Kama Regional Division of the Russian Academy of Natural Science, 2015-2024 All the materials of the journal are available under the Creative Commons Attribution 4.0 License (https://creativecommons.org/licenses/by/4.0/)