Using neural networks to assess porosity and permeability of mountain rocks
DOI:
https://doi.org/10.31471/1993-9868-2025-2(44)-46-54Keywords:
porosity; rock permeability; geophysical research; artificial intelligence; machine learning; neural network; data analysis; research interpretation.Abstract
The advent of artificial intelligence (AI) and its application in various industries has made it possible to automate and optimise many processes, including those in geophysics. Using AI in geophysical well surveys could transform the way data is collected, processed and interpreted. AI is expected to become a key tool in the oil and gas industry, advancing data modelling and analysis and offering significant benefits in terms of accuracy and efficiency. This paper presents an approach to estimating the porosity and permeability of rocks based on geophysical survey results, using a neural network as the primary tool. Porosity and permeability are key characteristics of reservoir formations and important parameters for design calculations and the development of productive horizons in oil and gas fields. The main difficulty in determining these parameters is that they are dependent on many other factors. Emerging information technologies offer opportunities to identify complex relationships between diverse geological and geophysical parameters. An artificial neural network model has been developed to predict the porosity and permeability coefficients of rocks, based on an in-depth analysis of geophysical data. This approach enables hidden relationships between parameters to be identified that would go unnoticed when applying traditional analytical methods. The study achieved high accuracy in predicting porosity and permeability values, as confirmed by their alignment with the results of laboratory experiments. This demonstrates the effectiveness of using machine learning to automate the analysis of geophysical data, and the proposed approach has significant advantages in terms of speed and accuracy of analysis. This is especially important in situations where laboratory research is impossible or limited.The use of neural networks contributes to a deeper understanding of the impact of various factors on rock porosity and permeability, which is important for decision-making in the process of reservoir development.The novelty of this work is the creation of an integrated approach that combines classical geophysical methods with innovative machine learning algorithms. This combination provides fast and efficient determination of rock porosity and permeability, opening up new opportunities for optimising the processes of field exploration and development.
Downloads
References
1. Schon, J. H. (2011). Physical properties of rocks: A Workbook. (Vol. 8). Elsevier. http://ndl.ethernet.edu.et/bitstream/123456789/419/1/97.pdf
2. Bernabe, Y., Mok, U., & Evans, B. (2006). A note on oscillation flow method for measuring rock permeability. International Journal of Rock Mechanics & Mining Sciences, 43(2), 311–316. https://doi.org/10.1016/j.ijrmms.2005.04.013
3. Tiab, D., & Donaldson, E. (2011). Petrophysic: Theory and practice of measuring reservoir rock and fluid transport properties (2nd ed.). Gulf Professional Publishing. https://www.geokniga.org/books/40073
4. Vyzhva, S. A., Bezrodna, I. M., & Kozionova, O. O. (2012). Analiz filtratsiino-iemnisnykh vlastyvostei karbonatnykh porid karbonu Rudenkivsko-Proletarskoho rehionu za rezultatamy petro-fizychnykh doslidzhen [Analysis of filtration-volumetric properties of Carboniferous carbonate rocks of the Rudenkivsko-Proletarskyi region based on petrophysical research results]. Heoinformatyka, (1), 17–25. http://jnas.nbuv.gov.ua/article/UJRN-0000039914 [in Ukrainian].
5. Vyzhva, S. A., Mykhailov, V. A., Onyshchuk, D. I., & etc. (2013). Petrofizychni parametry netradytsiinykh porid-kolektoriv Pivdennoho naftohazovoho rehionu [Petrophysical parameters of unconventional reservoir rocks of the Southern oil and gas region]. Heoinformatyka, (3), 17–25. http://jnas.nbuv.gov.ua/article/UJRN-0000025433 [in Ukrainian].
6. Ahmadi, M. A., & Chen, Z. (2019). Comparison of machine learning methods for estimating permeability and porosity of oil reservoirs via petrophysical logs. Petroleum, 5(3), 271–284. https://www.scribd.com/document/419961181/Comparison-of-Machine-Learning-Methods-for-Estimating-Permeabilityand-Porosity-of-Oil-Reservoirs-via-Petro-physical-Logs
7. Heidari, A. A., Faris, H., Aljarah, I., & etc. (2019). An efficient hybrid multilayer perceptron neural network with grasshopper optimization. Soft Computing, 23(17), 7941–7958. https://doi.org/10.1007/s00500-018-3424-2
8. Saputro, O. D., Maulana, Z. L., & Latief, F. D. E. (2016). Porosity log prediction using artificial neural network. Journal of Physics: Conference Series, 739(1), 012092. https://doi.org/10.1088/1742-6596/739/1/012092
9. Abdel Azim, R., & Aljehani, A. (2022). Neural Network Model for Permeability Prediction from Reservoir Well Logs. Processes, 10(12), 2587. https://doi.org/10.3390/pr10122587
10. Matinkia, M., Hashami, R., Mehrad, M., & etc. (2023). Prediction of permeability from well logs using a new hybrid machine learning algorithm. Petroleum, 9(1), 108–123. https://doi.org/10.1016/j.petlm.2022.03.003
11. Antoniuk, V. V. (2023). Prohnozuvannia filtratsiino-iemnisnykh vlastyvostei ta klasyfikatsiia porid-kolektoriv za kompleksom karotazhnykh ta petrofizychnykh danykh... [Prediction of filtration and storage properties and classification of reservoir rocks...] (Doctoral dissertation). Taras Shevchenko National University of Kyiv. Https://scc.knu.ua/upload/iblock/037/5sz8xiks1jzqeyupptx7ftprxybqf62x/Dis_Antoniuk_V.pdf [in Ukrainian].
12. Dmytriieva, I. S., & Dmytrenko, A. M. (2025). Metody shtuchnoho intelektu dlia prohno-zuvannia pokladiv vuhlevodniv z tryvymirnykh seishmichnykh zobrazhen [Artificial intelligence methods for predicting hydrocarbon deposits from three-dimensional seismic images]. Systemni tekhnolohii, (4(159)), 158–165. https://doi.org/10.34185/1562-9945-4-159-2025-16 [in Ukrainian].
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Oil and Gas Power Engineering
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
.png)
1.png){kind=logo}
