«Onshore and offshore oil and gas well construction»

Theoretical background of using machine learning to solve problems of wells cementing

UDC: 622.245
DOI: 10.33285/0130-3872-2023-12(372)-23-26

Authors:

¹ Tyumen Industrial University, Tyumen, Russia

Keywords: digitalization, well cementing, machine learning, probabilistic modeling, mathematic modeling

Annotation:

The authors of the article presents analytically determined the methodological aspects of production processes optimization using machine learning. The experience obtained in the course of using modern data analysis tools of such companies as "Gazprom Neft" PJSC, Skolkovo Institute of Science and Technology and Google, a generally accepted approach to solving problems based on digital technologies has been determined. The fundamental provisions of the methodology are as follows: data collection and their verification, visualization of input information, mathematical modeling based on standard methods and machine learning algorithms with subsequent comparison, expert analysis of the results. The main problems when using machine learning such as ensuring the required quality and volume of input data, the impossibility of logical explanation of the calculation results obtained in the course of a number of methods application and the lengthy setup of algorithms to minimize the likelihood of errors are highlighted. According to the conducted analysis of information on well cementing by leading oil and gas producing companies, it was determined that the accumulated volume of the data obtained is sufficient for creating software products based on machine learning in order to increase a well cementing tightness.

Bibliography:

1. Logachev Yu.L., Shiryaev A.B., Logachev A.Yu. Otsenka vliyaniya profilya skorostey potoka na vytesnyayushchuyu sposobnost' vyazkoplastichnykh zhidkostey // Stroitel'stvo neftyanykh i gazovykh skvazhin na sushe i na more. – 2017. – № 3. – S. 27–31.
2. Chernyshov S.E., Kunitskikh A.A. Tamponazhnye sostavy dlya tsementirovaniya obsadnykh kolonn skvazhin s upravlyaemoy kinetikoy rasshireniya // Neftyanaya provintsiya. – 2017. – № 3(11). – S. 59–74. – DOI: 10.25689/NP.2017.3.59-74
3. Nikolaev N.I., Lyu Kh., Kozhevnikov E.V. Issledovanie vliyaniya polimernykh bufernykh zhidkostey na prochnost' kontakta tsementnogo kamnya s porodoy // Vestn. PNIPU. Geologiya. Neftegazovoe i gornoe delo. – 2016. – T. 15, № 18. – S. 16–22. – DOI: 10.15593/2224-9923/2016.18.2
4. Sovershenstvovanie kachestva tsementirovaniya obsadnykh kolonn na Yuzhno-Priobskom mestorozhdenii / R.M. Fetsenets, S.N. Lyakh, A.Yu. Kuznetsov, E.V. Naykov // Burenie i neft'. – 2009. – № 12. – S. 20–23.
5. Basarygin Yu.M., Bulatov A.I., Proselkov Yu.M. Zakanchivanie skvazhin: ucheb. posobie dlya vuzov. – M.: Nedra, 2000. – 666 s.
6. Spravochnik inzhenera-neftyanika: v 3 t. T. 2. Inzhiniring bureniya: per. s angl. / pod red. A.G. Shatrovskogo, S.O. Borozdina. – M.-Izhevsk: In-t komp'yuter. issled, 2014. – 1064 s.
7. Alpaydin E. Introduction to machine learning. – Cambridge, Massachusetts: MITpress, 2009. – XXII, 616 p.
8. Zinkevich M. Pravila mashinnogo obucheniya: luchshie praktiki dlya mashinnogo obucheniya. – URL: https://developers.google.com/machine-learning/guides/rules-of-ml?hl=ru
9. Primenenie metodov mashinnogo obucheniya dlya povysheniya kachestva krepleniya skvazhin / D.V. Shalyapin, D.L. Bakirov, M.M. Fattakhov [i dr.] // Izv. vuzov. Neft' i gaz. – 2020. – № 5(143). – S. 81–93. – DOI: 10.31660/0445-0108-2020-5-81-93
10. Data-driven model for the identification of the rock type at a drilling bit / N. Klyuchnikov, A. Zaytsev, A. Gruzdev [et al.] // J. of Petroleum Science and Engineering. – 2019. – Vol. 178. – P. 506–516. – DOI: 10.1016/j.petrol.2019.03.041
11. Machine Learning for Improved Directional Drilling / J. Pollock, Z. Stoecker-Sylvia, V. Veedu [et al.] // Offshore Technology Conf., Houston, Texas, USA, April 30 – May 3. – 2018. – DOI: 10.4043/28633-MS
12. Martinelli M., Colombo I., Russo E.R. Predict Geomechanical Parameters with Machine Learning Combining Drilling Data and Gamma Ray // SPE Middle East Oil & Gas Show and Conf., Nov. 28 – Dec. 1, 2021, event canceled. – DOI: 10.2118/204688-MS
13. Sovershenstvovanie metodiki analiza dannykh po krepleniyu skvazhin / D.V. Shalyapin, D.L. Bakirov, M.M. Fattakhov, V.G. Kuznetsov // Stroitel'stvo neftyanykh i gazovykh skvazhin na sushe i na more. – 2022. – № 5(353). – S. 36–39. – DOI: 10.33285/0130-3872-2022-5(353)-36-39
14. Issledovanie fakticheskoy informatsii po krepleniyu skvazhin s primeneniem mashinnogo obucheniya i neyronnykh setey / D.V. Shalyapin, D.L. Bakirov, M.M. Fattakhov [ i dr.] // Izv vuzov. Neft' i gaz. – 2021. – № 3(147). – S. 108–119. – DOI: 10.31660/0445-0108-2021-3-108-119
15. Shalyapin D.V., Bakirov D.L., Kuznetsov V.G. Otsenka effektivnosti vnedreniya tekhnologicheskikh resheniy po povysheniyu kachestva krepleniya skvazhin, razrabotannykh na osnove tsifrovykh tekhnologiy // Izv vuzov. Neft' i gaz. – 2023. – № 3(159). – S. 68–83. – DOI: 10.31660/0445-0108-2023-3-68-83