Authors : Ikeh, Lesor; Asimiea, N. W.

Volume/Issue : Volume 11 - 2026, Issue 3 - March

Google Scholar : https://tinyurl.com/7t75shzv

Scribd : https://tinyurl.com/u8b7j5cy

DOI : https://doi.org/10.38124/ijisrt/26mar2050

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : The rate of fluid flow through reservoir rocks is determined by permeability, one of the key characteristics of a reservoir. Production forecasts, history matching, and robust reservoir simulation all depend heavily on accurate permeability estimates. It can be difficult to build trustworthy permeability models because of the inherent variety of permeability at various sizes and the scarcity of core data. To get above these obstacles, this work uses a variety of machine learning techniques, including Support Vector Regression (SVR), Random Forest (RF), XGBoost, and LightGBM, to predict lab-measured core permeability from frequently obtained well logs.A datasets that represented a carbonate platform (YField) was considered. The resilience of this technique under various geological settings could be assessed using the Y- field dataset, which consisted of 17 wells spread across a single reservoir. This approach relies heavily on feature engineering, especially when it comes to integrating vertical variability. Taking into account the smoothing effect of well logs over smallscale heterogeneities and the significance of spatial context, measurements from nearby well log readings were added into the models. By taking into consideration nearby depositional environments and shared geological history, this increased prediction accuracy. Results shows that the R2 values in Y- Field's blind tests were as high as 0.64, and the leave-one-wellout cross-validation technique produced validation R2 values as high as 0.8. The Yorla, Kpean, and Teera-ue formations had blind test R2 ratings of up to 0.82, 0.74, and 0.80 for the Y- Field, respectively. Even if these results are satisfactory, they demonstrate how machine-learning techniques can be used to accurately estimate permeability and emphasize the need for feature engineering. This work argues that although automated feature engineering using machine learning shows potential, human intervention more especially, the incorporation of geographical context can still greatly improve predictions. It may be the goal of future developments to incorporate this spatial awareness into machine learning algorithms.

Keywords : Permeability, Formation, Well Logs, Reservoir, Fluid, Porous Rock, Carbonate.

References :

1. Aigbedion, I. (2007). A case study of permeability modeling and reservoir performance in the absence of core data in the niger delta, nigeria. Journal of Applied Sciences, 7(5), 772–776.
2. Asquith, G. B., Krygowski, D., Gibson, C. R., & Asquith, G. B. (2004). Basic well log analysis. American Association of Petroleum Geologists.
3. Brereton, R. G., & Lloyd, G. R. (2010). Support vector machines for classification and regression. The Analyst, 135(2), 230–267.
4. Carcione, J. M., Gei, D., Picotti, S., Misnan, M. S., Rashidi, M. R., Bakar, Z. A., Harith, Z. Z., Bahri, N. H., & Hashim, N. (2020). Porosity and permeability of the overburden from wireline logs: A case study from offshore malaysia. Geomechanics and Geophysics for Geo-Energy and Geo-Resources, 6(3).
5. Chaya. (2022). Random forest regression. https://levelup.gitconnected.
6. Chen, T., & Guestrin, C. (2016). XGBoost. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.
7. Cohen, I., Huang, Y., Chen, J., Benesty, J., Benesty, J., Chen, J., Huang, Y., & Cohen, I. (2009). Pearson correlation coefficient. Noise reduction in speech processing, 1–4.
8. Costa Gomes, J., Geiger, S., & Arnold, D. (2022). The design of an open-source carbonate reservoir model. Petroleum Geoscience, 28(3).
9. Crampin, T. (2008). Well log facies classification for improved regional exploration. Exploration Geophysics, 39(2), 115–123.
10. Deegan, C. E., & Scull, B. J. (1977). A standard lithostratigraphic nomenclature for the central and northern North Sea. UK Institute of Geological Sciences, Report, 1, 36.
11. Diniz Ferreira, E. L., & Torres-Verdin, C. (2012). Improved estimation of pore connectivity and permeability in deepwater carbonates with the construction of multi-layer static and dynamic petrophysical models.
12. Elkatatny, S., Mahmoud, M., Tariq, Z., & Abdulraheem, A. (2017). New insights into the prediction of heterogeneous carbonate reservoir permeability from well logs using artificial intelligence network. Neural Computing and Applications, 30(9), 2673–2683, Equinor. (2022). Volve field data set.
13. Fadairo, A., & Awuyo, O. (2019). Achieving best practices in log pre-processing for facies and permeability modeling. International Journal of Mechanical Engineering and Technology, 10.
14. Feng, R., Grana, D., & Balling, N. (2021). Imputation of missing well log data by random forest and its uncertainty analysis. Computers amp;amp; Geosciences, 152, 104763.
15. Gholanlo, H. H. (2021). Analysis of permeability based on petrophysical logs: Comparison between heuristic numerical and analytical methods. Journal of Petroleum Exploration and Production Technology, 11(5), 2097–2111.
16. Gray, E., Hartley, A., & Howell, J. (2020). The influence of stratigraphy and facies distribution on reservoir quality and production performance in the triassic skagerrak formation of the uk and norwegian central North Sea. Geological Society, London, Special Publications, 494(1), 379–409.
17. Grinsztajn, L., Oyallon, E., & Varoquaux, G. (2022). Why do tree-based models still outperform deep learning on tabular data?
18. Han, D., & Kwon, S. (2021). Application of machine learning method of a data-driven deep learning model to predict well production rate in the shale gas reservoirs. Energies, 14(12), 3629.
19. Hansen, S. (1996). A compaction trend for cretaceous and tertiary shales on the Norwegian shelf based on sonic transit times. Petroleum Geoscience, 2(2), 159–166.
20. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep Residual Learning for Image Recognition.
21. Ibrahem Ahmed Osman, A., Najah Ahmed, A., Chow, M. F., Feng Huang, Y., & El-Shafie, A. (2021). Extreme gradient boosting (xgboost) model to predict the groundwater levels in Selangor malaysia. Ain Shams Engineering Journal, 12(2), 1545–1556.
22. Katahara, K. (2008). What is shale to a petrophysicist? The Leading Edge, 27(6), 738–741.
23. Kavzoglu, T., & Teke, A. (2022). Advanced hyperparameter optimization for improved spatial prediction of shallow landslides using extreme gradient boosting (xgboost). Bulletin of Engineering Geology and the Environment, 81(5).
24. Ke, G., Meng, Q., Finely, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T.-Y. (2017). Lightgbm: A highly efficient gradient boosting decision tree. Advances in Neural Information Processing Systems 30 (NIP 2017).
25. Kieft, R. L., Jackson, C. A.-L., Hampson, G. J., & Larsen, E. (2010). Sedimentology and sequence stratigraphy of the hugin formation, quadrant 15, norwegian sector, south viking graben. Geological Society, London, Petroleum Geology Conference Series, 7(1), 157–176.
26. Kramer, O. (2016). Scikit-learn. In Machine learning for evolution strategies (pp. 45–53). Springer International Publishing.
27. Kumar, A. (2022). Xgboost vs lightgbm.
28. Strohmenger, C. J., Weber, L. J., Ghani, A., Rebelle, M., Al-Mehsin, K., Al- Jeelani, O., Al-Mansoori, A., & Suwaina, O. (2004). High-resolution sequence stratigraphy of the kharaib formation (lower cretaceous, u.a.e.) All Days.
29. Talebi, H., Peeters, L. J., Otto, A., & Tolosana-Delgado, R. (2021). A truly spatial random forests algorithm for geoscience data analysis and modelling. Mathematical Geosciences, 54(1), 1–22.
30. Vapnik, V. (1998). The support vector method of function estimation. In J. A. K. Suykens & J. Vandewalle (Eds.), Nonlinear modeling: Advanced black-box techniques (pp. 55–85). Springer US.