Reproduction of Reservoir Pressure in Oil Field Development: Prospects and Problems of Using Methods Machine Learning

Currently, models based on the application of artificial intelligence methods are actively developed and applied in solving a variety of problems, including in the practice of petroleum engineering. Evaluation of the accuracy and reliability of the developed models is usually reduced to defining standard statistical criteria, while the developers do not always use a separate examination sample. This article presents the results of the study, which are reduced to multivariate testing of the neural network model previously developed by the authors to determine the dynamic reservoir pressure in the selection zones of oil wells. The model is characterized by a number of advantageous characteristics, including minimal requirements for the amount of initial data, which determines its relevance and practical demand. However, the closed nature of computational algorithms related to the “black box” category does not allow us to reasonably formulate the conditions and criteria for applying the model, the reliability of the retro- and prospective forecast of the reservoir pressure. Three oil deposits of one field with different geological and physical conditions were selected as the object of study. The availability of a large number of actual reservoir pressure determinations by means of hydrodynamic well testing at the field allowed testing the model under a variety of scenarios, for each of which the forecast error was estimated and analyzed. As a result, high estimates of the model for retro- and prospective reservoir pressure reproduction were confirmed. It was found that forecast errors are reduced to zero in the presence of a large number of actual reservoir pressure determinations. However, to perform the calculation for each well, a single measurement for the entire history is sufficient. It was found that a sharp change in the well flow rate should also be accompanied by an actual determination of reservoir pressure with the entry of the obtained value into the model. In the absence of even a single reservoir pressure measurement for the wells, the model reliably reproduces its value using the kriging procedure used in the algorithms.

1. Akhmetdinov A.M., Marchenko A.V., Pavlov D.V., Khabarov A.V., Oblekov R.G., Bakalo N.Yu. (2024). Improving the methods for interpreting hydrodynamic studies of oil wells based on phase segregation analysis after well shutdown. Gazovaya promyshlennost’, S2(866), pp. 66–73. (In Russ.)

2. Campos D., Wayo D.D.K., De Santis R.B., Martyushev D.A., Yaseen Z.M., Duru U.I., Saporetti C.M., Goliatt L. (2024). Evolutionary automated radial basis function neural network for multiphase flowing bottom-hole pressure prediction. Fuel, 377, 132666. https://doi.org/10.1016/j.fuel.2024.132666

3. Cao S., Wang C., Niu Q., et al. (2024). Enhancing pore pressure prediction accuracy: A knowledge-driven approach with temporal fusion transformer. Geoenergy Science and Engineering, 238, 212839. https://doi.org/10.1016/j.geoen.2024.212839

4. Davoodi S., Thanh H.V., Vanovskiy V., Wood D.A, Al-Shargabi M., Mehrad M. (2025). Hybrid machine-learning and optimization models for precise determination of pore pressure changes in subsurface reservoirs. Computational Geosciences, 29, 47. https://doi.org/10.1007/s10596-025-10387-1

5. Davoodi S., Wood D.A., Al-Shargabi M., Vanovskiy V., Rukavishnikov V., Burnaev E. (2026). Emerging applications of physics-informed and physicsguided machine learning in geoenergy science: A review. Communications in Nonlinear Science and Numerical Simulation, 154, 109551. https://doi.org/10.1016/j.cnsns.2025.109551

6. Deng S., Pan H.-Y., Wang H.-G., et al. (2024). A hybrid machine learning optimization algorithm for multivariable pore pressure prediction. Petroleum Science, 21(1), pp. 535–550. https://doi.org/10.1016/j.petsci.2023.09.001

7. Gadilshina V.R., Morozov P.E., Shamsiev M.N., Khairullin M.Kh. (2023). Hydrodynamic studies of oil wells after short-term formation disturbance. Tomsk State University Journal of Mathematics and Mechanics, 85, pp. 90–100. (In Russ.) https://doi.org/10.17223/19988621/85/7

8. Harp D.R., O’Malley D., Yan B., Pawar R. (2021). On the feasibility of using physics-informed machine learning for underground reservoir pressure management. Expert Systems with Applications, 178, 115006. https://doi.org/10.1016/j.eswa.2021.115006

9. Jia L., Peng S., Xu J., Yan F. (2021). Interlayer interference during coalbed methane coproduction in multilayer superimposed gas-bearing system by 3D monitoring of reservoir pressure: An experimental study. Fuel, 304, 121472. https://doi.org/10.1016/j.fuel.2021.121472

10. Kaleem W., Tewari S., Fogat M., Martyushev D.A. (2024). A Hybrid Machine Learning Approach Based Study of Production Forecasting and Factors Influencing the Multiphase Flow through Surface Chokes. Petroleum, 10(2), pp. 354–371. https://doi.org/10.1016/j.petlm.2023.06.001

11. Kanin E., Garipova A., Boronin S., Vanovskiy V., Vainshtein A., Afanasyev A., Osiptsov A., Burnaev E. (2025). Combined mechanistic and machine learning method for construction of oil reservoir permeability map consistent with well test measurements. Petroleum Research, 10(2), pp. 247–265. https://doi.org/10.1016/j.ptlrs.2024.09.001

12. Karsakov A.V., Zyatikov P.N., Sharf I.V. (2025). Development of oil and gas industry technologies using artificial intelli-gence. Bulletin of the Tomsk Polytechnic University. Geo Assets Engineering, 336(5), pp. 216–228. (In Russ.) https://doi.org/10.18799/24131830/2025/5/5048

13. Khankishieva T.U. (2023). Method for determining bottomhole pressure without stopping a well equipped with a borehole sucker rod pump. Neftyanoe Khozyaystvo = Oil industry, 3, pp. 54–57. (In Russ.) https://doi.org/10.24887/0028-2448-2023-3-54-57

14. Kuznetsova E.A., Nikulin S.E., Shilov A.V., Filatov M.A. (2023). Experience in applying analytical methods for determining reservoir pressure. Neftepromyslovoe delo, 6(654), pp. 12–16. (In Russ.) https://doi.org/10.33285/0207-2351-2023-6(654)-12-16

15. Levitina E.E., Inyakina E.I., Paklinov N.M. (2023). Assessment of the dynamics of reservoir pressure and drainable gas reserves based on well survey data. News of higher educational institutions. Oil and Gas, 5(161), pp. 67–76. (In Russ.) DOI: 10.31660/0445-0108-2023-5-67-76

16. Li Q., Etzel T.M., Konstantinou A.G., Mazumdar P. (2024). Rock physics and basin modeling nexus for predicting pore pressure. Geoenergy Science and Engineering, 234, 212575. https://doi.org/10.1016/j.geoen.2023.212575

17. Martyushev D.A., Ponomareva I.N., Shadrov T.A., Khaitin R.A. Data Stream Analytics, a service for automating processes of hydrodynamic modeling and control of oil field development (module “Assessment and forecast of the energy state of the deposit”). Certificate of registration of the computer program RU 2023614383, 01.03.2023. Application No. 2023613550 dated 01.03.2023. (In Russ.)

18. Martyushev D.A., Ponomareva I.N., Zakharov L.A., Shadrov T.A. (2021). Application of machine learning for forecasting formation pressure in oil field development. Bulletin of the Tomsk Polytechnic University, Geo Assets Engineering, 332(10), pp. 140–149. (In Russ.) https://doi.org/10.18799/24131830/2021/10/3401

19. Nie Y., Xian C., Luo J. et al. (2025). Bagging machine learning algorithms for rapid identification, classification, evaluation and upscaling in unconventional reservoir. Geoenergy Science and Engineering, 246, 213545. https://doi.org/10.1016/j.geoen.2024.213545

20. Nigametyanova G.A., Ishkin D.Z. (2024). Forecasting the duration of the pressure build-up based on reservoir parameters and well completion. Neftyanaya provintsiya, 1(37), pp. 89–97. (In Russ.)

21. Peters B., Haber E., Lensink K. (2024). Fully invertible hyperbolic neural networks for segmenting large-scale surface and sub-surface data. Artificial Intelligence in Geosciences, 5, 100087. https://doi.org/10.1016/j.aiig.2024.100087

22. Ponomareva I.N., Martyushev D.A., Govindarajan S.K. (2024). A new approach to predict the formation pressure using multiple regression analysis: Case study from Sukharev oil field reservoir – Russia. Journal of King University – Engineering Sciences, 36(8), pp. 694–700. https://doi.org/10.1016/j.jksues.2022.03.005

23. Pwavodi J., Kelechi I.N., Angalabiri P., et al. (2023). Pore pressure prediction in offshore Niger delta using data-driven approach: Implications on drilling and reservoir quality. Energy Geoscience, 4(3), 100194. https://doi.org/10.1016/j.engeos.2023.100194

24. Savchenko V.O., Goridko K.A., Kartavtseva I.A., Abdullaev R.A., Khodakov I.O., Simonov M.V. (2023). Approach to assessing reservoir pressure and the type of pressure build-up curve during short-term shutdowns of oil wells equipped with electric centrifugal pumps. Neftyanoe Khozyaystvo = Oil industry, 12, pp. 40–44. (In Russ.) https://doi.org/10.24887/0028-2448-2023-12-40-44

25. Sergeev V.L., Dong V.H., Pham D.A. (2019). Adaptive interpretation of the results of hydrodynamic studies of horizontal wells using predictive models. Bulletin of Tomsk Polytechnic University. Geo Assets Engineering, 330(1), pp. 165–172. (In Russ.) https://doi.org/10.18799/24131830/2019/1/62

26. Shao R., Wang H., Xiao L. (2024). Reservoir evaluation using petrophysics informed machine learning: A case study. Artificial Intelligence in Geosciences, 5, 100070. https://doi.org/10.1016/j.aiig.2024.100070

27. Shi F., Liao H., Qu F., Liu J., Wu T. (2024). Collaborative-driven reservoir formation pressure prediction using GAN-ML models and well logging data. Geoenergy Science and Engineering, 242, 213271. https://doi.org/10.1016/j.geoen.2024.213271

28. Tang H., Fu P., Jo H., Jiang S., et al. (2022). Deep learning-accelerated 3D carbon storage reservoir pressure forecasting based on data assimilation using surface displacement from InSAR. International Journal of Greenhouse Gas Control, 120, 103765. https://doi.org/10.1016/j.ijggc.2022.103765

29. Wang F., Hou X. (2025). Machine learning-based prediction of physical parameters in heterogeneous carbonate reservoirs using well log data. Energy Geoscience, 6(2), 100383. https://doi.org/10.1016/j.engeos.2025.100383

30. Xi H., Luo Z., Guo Y. (2025). Reservoir evaluation method based on explainable machine learning with small samples. Unconventional Resources, 5, 100128. https://doi.org/10.1016/j.uncres.2024.100128

31. Yu H., Chen G., Gu H. (2020). A machine learning methodology for multivariate pore-pressure prediction. Computers & Geosciences, 143, 104548. https://doi.org/10.1016/j.cageo.2020.104548

32. Zakharov L.А., Martyushev D.А., Ponomareva I.N. (2022). Predicting dynamic formation pressure using artificial intelligence methods. Journal of Mining Institute, 253, pp. 23–32. https://doi.org/10.31897/PMI.2022.11

33. Zhang G., Davoodi S., Band S.S., et al. (2022). A robust approach to pore pressure prediction applying petrophysical log data aided by machine learning techniques. Energy Reports, 8, pp. 2233–2247. https://doi.org/10.1016/j.egyr.2022.01.012

34. Zhang X., Lu Y.-H., Jin Y., Chen M., Zhou B. (2024). An adaptive physics-informed deep learning method for pore pressure prediction using seismic data. Petroleum Science, 21(2), pp. 885–902. https://doi.org/10.1016/j.petsci.2023.11.006

35. Zong P., Xu H., Tang D., Zhao T. (2023). A dynamic prediction model of reservoir pressure considering stress sensitivity and variable production. Geoenergy Science and Engineering, 225, 211688. https://doi.org/10.1016/j.geoen.2023.211688