Prediction of rock unconfined compressive strength for subsurface formation CO2 storage using physics augmented machine learning: A comparative performance evaluation
DOI:
https://doi.org/10.31699/IJCPE.2026.2.4Keywords:
Physics-Augmented Machine Learning; Machine learning; Unconfined Compressive Strength; Zubair field; geomechanical propertyAbstract
The paper introduces a state-of-the-art Physics Augmented Machine Learning framework to predict the critical geomechanical property – Unconfined Compressive Strength (UCS) – in the Zubair reservoir of the Zubair field for mitigating the risk of CO2 storage. The geomechanical results and modeling workflow were obtained via an integrated software system that was developed in Python, while all computational processes and model training were conducted in the Deepnote interactive cloud computing environment; thus, all of the necessary libraries and the high-performance processing capacities were seamlessly integrated. Through the comparison of six ensemble algorithms which are Categorical Boosting (CatBoost), XGBoost )Extreme Gradient Boosting(, RF(Random Forest Regressor), ExtraTrees (Extremely Randomized Trees), GBM (Gradient Boosting Machine), and Deep ANN (Deep Artificial Neural Networks), the research determined the most dependable models, while the Gradient Boosting model stood out by achieving UCS prediction with the best metrics of (R2=0.9995, MAE=3.64). The models are not only providing a precise and reliable but also a scalable assessment tool for subsurface stability, containment integrity, and real-time reservoir monitoring.
Received on 10/02/2026
Received in Revised Form on 01/04/2026
Accepted on 03/04/2026
Published on 30/06/2026
References
[1] J.-Q. Shi, C. Sinayuc, S. Durucan, and A. Korre, "Assessment of carbon dioxide plume behaviour within the storage reservoir and the lower caprock around the KB-502 injection well at In Salah," International Journal of Greenhouse Gas Control, Vol. 7, pp. 115-126, March 2012. https://doi.org/10.1016/j.ijggc.2012.01.002
[2] P. Horsrud, "Estimating Mechanical Properties of Shale from Empirical Correlations," SPE Drilling & Completion, Vol. 16, No. 02, pp. 68–73, June 2001. https://doi.org/10.2118/56017-PA
[3] H. Vo Thanh, Q. Yasin, W. J. Al-Mudhafar, and K.-K. Lee, "Knowledge-based machine learning techniques for accurate prediction of CO2 storage performance in underground saline aquifers," Applied Energy, Vol. 314, Art. No. 118985, May 2022. https://doi.org/10.1016/j.apenergy.2022.118985
[4] Y. Song, S. Jun, Y. Na, K. Kim, Y. Jang, and J. Wang, "Geomechanical challenges during geological CO2 storage: A review," Chemical Engineering Journal, Vol. 452, Art. No. 140968, Jan. 2023. https://doi.org/10.1016/j.cej.2022.140968
[5] S. M. Frailey, "Methods for estimating CO2 storage in saline reservoirs," Energy Procedia, Vol. 1, No. 1, pp. 2769–2776, Feb. 2009. https://doi.org/10.1016/j.egypro.2009.02.048
[6] C. Chang, M. D. Zoback, and A. Khaksar, "Empirical relations between rock strength and physical properties in sedimentary rocks," Journal of Petroleum Science and Engineering, Vol. 51, No. 3–4, pp. 223–237, May 2006. https://doi.org/10.1016/j.petrol.2006.01.003
[7] M. Issa, M. Issa, A. Abdulkareem, and F. Hadi, "Review of Mechanical Rock Properties Measurement and Estimation Techniques," International Journal of Analytical Experimental and Finite Element Analysis (IJAEFEA), Vol. 9, pp. 43–49, 2022.
[8] W. J. Al-Mudhafar, "Integrating well log interpretations for lithofacies classification and permeability modeling through advanced machine learning algorithms," Journal of Petroleum Exploration and Production Technology, Vol. 7, pp. 1023–1033, 2017. https://doi.org/10.1007/s13202-017-0360-0
[9] R. Zhong, M. Tsang, G. Makusha, B. Yang, and Z. Chen, "Improving rock mechanical properties estimation using machine learning," in Proc. Resource Operators Conf. (ROC), University of Wollongong, 2025. https://doi.org/10.71747/uow-r3gk326m.c.7651142.v1
[10] G. E. Karniadakis et al., "Physics-informed machine learning," Nature Reviews Physics, Vol. 3, No. 6, pp. 422–441, June 2021. https://doi.org/10.1038/s42254-021-00314-5
[11] A. Latrach, M. L. Malki, M. Morales, M. Mehana, and M. Rabiei, "A critical review of physics-informed machine learning applications in subsurface energy systems," Geoenergy, Vol. 3, Art. No. 212938, 2024. https://doi.org/10.1016/j.geoen.2024.212938
[12] L. Prokhorenkova, G. Gusev, A. Vorobev, A. V. Dorogush, and A. Gulin, "CatBoost: Unbiased boosting with categorical features," in Advances in Neural Information Processing Systems, Vol. 31, 2018, pp. 6639–6649. https://doi.org/10.48550/arXiv.1706.09516
[13] B. Yuan et al., "Physics-informed machine learning in geotechnical engineering: a direction paper," Geomechanics and Geoengineering, Vol. 20, No. 5, pp. 1128–1159, 2025. https://doi.org/10.1080/17486025.2025.2502029
[14] G. Ke, Q. Meng, T. Finley, T. Wang, W. Chen, W. Ma, Q. Ye, and T.-Y. Liu, "LightGBM: A highly efficient gradient boosting decision tree," in Advances in Neural Information Processing Systems, I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, and R. Garnett, Eds., Vol. 30, pp. 3149–3157, 2017.
[15] A. K. Khassaf, Z. M. Al-hameed, N. R. Al-Mohammedawi, W. J. Al-Mudhafar, D. A. Wood, M. A. Abbas, O. Ameur-Zaimeche, and A. A. Alsubaih, "Physics-Informed Machine Learning for Enhanced Permeability Prediction in Heterogeneous Carbonate Reservoirs," in Proceeding. Offshore Technology Conference, Houston, TX, USA, May 2025, Paper No. OTC-35892-MS. https://doi.org/10.4043/35892-MS
[16] N. Al-Kanaani, "Geomechanical Properties Characterization and Stability Analysis of the Zubair Formation in the Zubair Oilfield Using Well Logs and Borehole Data," Iraqi Geological Journal, Vol. 58, No. 2E, Art. No. 6, 2025. https://doi.org/10.46717/igj.2025.58.2E.6
[17] A. Peter, X. Jin, X. Fan, K. I. Eshiet, Y. Sheng, and D. Yang, "Effect of CO2 Phase on Pore Geometry of Saline Reservoir Rock," Rock Mechanics and Rock Engineering, Vol. 55, No. 4, pp. 1907–1930, Apr. 2022. https://doi.org/10.1007/s00603-021-02658-x
[18] M. A. Abbas, W. J. Al-Mudhafar, A. Anees, and D. A. Wood, "Integrating petrophysical data into efficient iterative cluster analysis for electrofacies identification in clastic reservoirs," Energy Geoscience, Vol. 5, No. 1, Art. No. 100341, Jan. 2024. https://doi.org/10.1016/j.engeos.2024.100341
[19] E. Fjær, R. M. Holt, P. Horsrud, A. M. Raaen, and R. Risnes, Petroleum related rock mechanics, 2nd ed., Vol. 53. Amsterdam, Netherlands: Elsevier, 2008. https://doi.org/10.1016/s0376-7361(07)53002-5
[20] Y. Rimal, N. Sharma, and A. Alsadoon, "The accuracy of machine learning models relies on hyperparameter tuning: student result classification using random forest, randomized search, grid search, bayesian, genetic, and optuna algorithms," Multimedia Tools and Applications, Vol. 83, pp. 74349–74364, 2024. https://doi.org/10.1007/s11042-024-18426-2
[21] S. Sen, A. A. Radwan, M. Leila, A. Abdelmaksoud, and M. Ali, "Geomechanical assessment of the Cenozoic stratigraphy of the Tui area, Taranaki Basin, offshore New Zealand – Implications for geological storage in the Paleocene and Eocene sandstones," International Journal of Greenhouse Gas Control, Vol. 136, Art. No. 104194, 2024. https://doi.org/10.1016/j.ijggc.2024.104194
[22] A. A. Basheer and M. N. Hajmeer, "Artificial neural networks: fundamentals, computing, design, and application," Journal of Microbiological Methods, Vol. 43, No. 1, pp. 3–31, Dec. 2000. https://doi.org/10.1016/s0167-7012(00)00201-3
[23] P. Cosenza, M. Ghoreychi, G. de Marsily, G. Vasseur, and S. Violette, "Theoretical prediction of poroelastic properties of argillaceous rocks from in situ specific storage coefficient," Water Resources Research, Vol. 38, No. 11, pp. 1256–1266, Nov. 2002. https://doi.org/10.1029/2001WR001201
[24] M. Ghafoori, A. Rastegarnia, and G. R. Lashkaripour, "Estimation of static parameters based on dynamical and physical properties in limestone rocks," Journal of African Earth Sciences, Vol. 137, pp. 22–31, Jan. 2018. https://doi.org/10.1016/j.jafrearsci.2017.09.008
[25] T. Chen and C. Guestrin, "XGBoost: A scalable tree boosting system," in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery, 2016, pp. 785–794. https://doi.org/10.1145/2939672.2939785
[26] J. H. Friedman, "Greedy function approximation: a gradient boosting machine," Ann. Statist., Vol. 29, No. 5, pp. 1189–1232, Oct. 2001. https://doi.org/10.1214/aos/1013203451
[27] L. Breiman, "Random Forests," Machine Learning, Vol. 45, No. 1, pp. 5–32, Oct. 2001. https://doi.org/10.1023/A:1010933404324
[28] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, Vol. 521, No. 7553, pp. 436–444, May 2015. https://doi.org/10.1038/nature14539
[29] E. Fjær, "Relations between static and dynamic moduli of sedimentary rocks," Geophysical Prospecting, Vol. 67, No. 1, pp. 128–139, Jan. 2019. https://doi.org/10.1111/1365-2478.12711
[30] Schlumberger Limited, "2015 Annual Report," Houston, TX, USA, Rep. SLB-AR15, 2015.
[31] J. P. Castagna, M. L. Batzle, and R. L. Eastwood, "Relationships between compressional-wave and shear-wave velocities in clastic silicate rocks," Geophysics, Vol. 50, No. 4, pp. 571–581, Apr. 1985. https://doi.org/10.1190/1.1441933
[32] A. R. Najibi, M. Ghafoori, G. R. Lashkaripour, and M. R. Asef, "Reservoir geomechanical modeling: In-situ stress, pore pressure, and mud design," Journal of Petroleum Science and Engineering,Vol. 151, pp. 31–39, Mar. 2017. https://doi.org/10.1016/j.petrol.2017.01.045
[33] J. H. Friedman, "Stochastic gradient boosting," Computational Statistics & Data Analysis, Vol. 38, No. 4, pp. 367–378, 2002. https://doi.org/10.1016/S0167-9473(01)00065-2
Downloads
Published
Issue
Section
License
Copyright (c) 2026 The Author(s). Published by College of Engineering, University of Baghdad.
This work is licensed under a Creative Commons Attribution 4.0 International License.
