«Geology, geophysics and development of oil and gas fields»

Optimization of hydrocarbon reservoir models with application of machine learning methods

UDC: 622.(276+279).1/.4.001.57
DOI: -

Authors:

¹ Northern (Arctic) Federal University named after M.V. Lomonosov, Arkhangelsk, Russia

Keywords: neural network, geological and hydrodynamic models of hydrocarbon reservoir, machine learning, modeling

Annotation:

The article represents a reviewed scope of literature containing information on geological, hydrodynamic modelling, applicable machine learning methods and describes a concept of deterministic model creation of hydrocarbon reservoir.
The authors reviewed the main stages of geological, hydrodynamic modelling, which take specific set of parameters defining the accuracy of modelling results. Features of the modeling stages specify further directions of optimization, where machine learning-based methods (ML methods) reapplicable.
The authors studied the following ML methods: neural networks, k-nearest neighborhood, evolutionary algorithms, adaptive neural-fuzzy inference systems, which are preferable for classification or regression of different data.
Different algorithms are applicable for optimization of ML methods while training model or increasing accuracy of modeling results.

Bibliography:

1. Osnovy geologicheskogo modelirovaniya: ucheb. posobie / V.A. Belkina, S.R. Bembel’, A.A. Zaboeva, N.V. San’kova. – Tyumen’: TyumGNGU, 2015. – T. 1. – 168 s. – URL: https://www.geokniga.org/bookfiles/geokniga-osnovy-geologicheskogo-modelirovaniya.pdf
2. Derduga A.V., Nshuti M.I. Postroenie geologicheskoy i gidrodinamicheskoy modeley Vladimirskogo mestorozhdeniya // Nauka. Tekhnika. Tekhnologii (politekhnicheskiy vestnik). – 2018. – № 4. – S. 265–279. – URL: http://www.id-yug.com/images/id-yug/SET/2018/4/2018-4-266-280.pdf
3. Bozhenyuk N.N. Metody adaptatsii i snizheniya neopredelennostey pri geologo-gidrodinamicheskom modelirovanii terrigennykh kollektorov na primere ryada mestorozhdeniy Zapadnoy Sibiri: dis. … kand. geol.-mineral. nauk. – Tyumen’. 2018. – 163 s.
4. Kaarov Zh.Z., Gadzhiev M.Zh. Etapy postroeniya trekhmernoy tsifrovoy geologicheskoy modeli verkhneyurskikh otlozheniy gruppy plastov YuV // Mezhdunarodnyy zhurnal gumanitarnykh i estestvennykh nauk. – 2020. – № 3(1). – S. 68–72. – DOI: 10.24411/2500-1000-2020-1020
5. Sun Sh., Zhang T. Reservoir Simulation: Machine Learning and Modeling. – Gulf Professional Publishing, 2020. – 340 p.
6. Davidenko V.D. Razrabotka determinirovannykh modeley povyshennoy tochnosti i programmnykh kompleksov dlya pryamogo modelirovaniya fizicheskikh protsessov v yadernykh reaktorakh: dis. … d-ra tekhn. nauk. – M.: Nats. issled. tsentr "Kurchatovskiy institut", 2017. – 253 s.
7. Vliyanie geologicheskoy izuchennosti razrabatyvaemykh mestorozhdeniy na prirost zapasov nefti na primere Sokolkinskogo neftyanogo mestorozhdeniya / A.V. Lobusev, M.A. Lobusev, L.N. Salakhova, A.I. Koshkina // Geologiya, geofizika i razrabotka neftyanykh i gazovykh mestorozhdeniy. – 2023. – № 6(378). – S. 5–11. – DOI: 10.33285/2413-5011-2023-6(378)-5-11
8. Metodicheskiy podkhod pri postroenii geologicheskikh modeley zalezhey uglevodorodov sredneyurskikh otlozheniy Zapadnoy Sibiri v usloviyakh anomal’no vysokikh plastovykh davleniy / O.A. Smirnov, V.N. Borodkin, A.V. Lukashov [i dr.] // Geologiya, geofizika i razrabotka neftyanykh i gazovykh mestorozhdeniy. – 2023. – № 2(374). – S. 5-15. – DOI: 10.33285/2413-5011-2023-2(374)-5-15