№84-5

Adaptive control of exploration drilling parameters based on machine learning to increasing core recovery

Ye. Koroviaka1,    https://orcid.org/0000-0002-2675-6610

V. Khomenko1,     https://orcid.org/0000-0002-3607-5106

O. Pashchenko1,  https://orcid.org/0000-0003-3296-996X

S. Shevchenko1,  https://orcid.org/0000-0003-3994-1927

O. Dreshpak1        https://orcid.org/0000-0003-1019-4382

1Dnipro University of Technology, Dnipro, Ukraine

Coll.res.pap.nat.min.univ. 2026, 84:66–79

Full text (PDF)

https://doi.org/10.33271/crpnmu/84.066

ABSTRACT

Purpose. The goal is to dynamically adjust drilling parameters (bit axial load and rotation speed) according to predicted risks of core integrity disruption, which improves core quality and the reliability of geological data for critical mineral resource assessment.

The methods. The proposed approach combines high-frequency sensor data collection with predictive machine learning models in a closed-loop control system. Drilling parameters (WOB, RPM, torque, mechanical drilling speed, vibrations) were recorded at 100 Hz. The XGBoost classifier, trained on the labeled time series, predicted the probability of core loss and initiated an automatic safe drilling mode with WOB reduction and RPM adjustment.

Findings. The model achieved an accuracy of 0.92 and a completeness of 0.91, predicting core failure an average of 4.4 seconds before it occurred. Adaptive control increased the average core selection ratio from 83.5% to 94.0%, with the largest effect observed in fractured rocks (from 65% to 88%). This was accompanied by a 12% reduction in mechanical drilling speed (from 4.6 to 4.05 m/h) as the system prioritized core integrity in high-risk areas.

The originality. The dependence of core losses during exploratory drilling on changes in drilling technological parameters (axial load on the bit, rotation frequency, torque, mechanical passage speed and vibrations) and structural features of rocks, in particular heterogeneous and fractured formations, has been established. It has been shown that the use of the proposed system allows to increase the core selection ratio by 10% and significantly reduce its losses in difficult geological conditions.

Practical implementation. The proposed system improves the quality of geological data, reduces the costs of re-drilling intervals with substandard core output, reduces project risks and can be integrated into modern drilling rig control systems as an intelligent tool to support exploratory drilling.

Keywords: adaptive drilling control, machine learning, core selection, geological exploration drilling, real-time optimization, critical minerals.

References

1. Araujo, L. M., Esteves, P. B., Fabbro, S., Kuffa, M., & Wegener, K. (2023). Geometric-kinematic model for wear simulation of diamond-impregnated segments in concrete core drilling. Procedia CIRP, 117, 38–43. https://doi.org/10.1016/j.procir.2023.03.008

2. Khomenko, V. L., Ratov, B. T., Pashchenko, O. A., Davydenko, O. M., & Borash, B. R. (2023). Justification of drilling parameters of a typical well in the conditions of the Samskoye field. IOP Conference Series: Earth and Environmental Science, 1254(1), 12052. https://doi.org/10.1088/1755-1315/1254/1/012052

3. Kumar, Ch. V., Vardhan, H., Murthy, Ch. S. N., & Karmakar, N. C. (2019). Estimating rock properties using sound signal dominant frequencies during diamond core drilling operations. Journal of Rock Mechanics and Geotechnical Engineering, 11(4), 850–859. https://doi.org/10.1016/j.jrmge.2019.01.001

4. Pashchenko, O., Ratov, B., Khomenko, V., Gusmanova, A., & Omirzakova, E. (2024). Methodology for optimizing drill bit performance. 24th International Multidisciplinary Scientific GeoConference Proceedings SGEM 2024, Science and Technologies in Geology, Exploration and Mining, Vol 24, Issue 1.1, 24, 623–632. https://doi.org/10.5593/sgem2024/1.1/s06.78

5. Basarir, H., & Karpuz, C. (2015). Preliminary estimation of rock mass strength using diamond bit drilling operational parameters. International Journal of Mining, Reclamation and Environment, 30(2), 145–164. https://doi.org/10.1080/17480930.2015.1025561

6. Wang, J., Feng, P., Zhang, J., & Guo, P. (2018). Reducing cutting force in rotary ultrasonic drilling of ceramic matrix composites with longitudinal-torsional coupled vibration. Manufacturing Letters, 18, 1–5. https://doi.org/10.1016/j.mfglet.2018.08.002

7. Carr-Wilson, S., Pattanayak, S. K., & Weinthal, E. (2024). Critical mineral mining in the energy transition: A systematic review of environmental, social, and governance risks and opportunities. Energy Research & Social Science, 116, 103672. https://doi.org/10.1016/j.erss.2024.103672

8. Aitkazinova, S. (2016). Methodology of assessment and prediction of critical condition of natural technical systems. 16th International Multidisciplinary Scientific GeoConference SGEM2016, Informatics, Geoinformatics and Remote Sensing. https://doi.org/10.5593/sgem2016/b22/s09.001

9. Serban, S. G., & Kiss, I. (2021). Identifying the Possibilities for Superior Recovery by Pelletization of Industry Related Small and Powdery Iron Containing Waste. Acta Polytechnica Hungarica, 18(4), 79–104. https://doi.org/10.12700/aph.18.4.2021.4.5

10. Liu, Y., French, P., Skinner, G., Gleizes, L., Bisset, P., Gajare, S., Kelleher, D., & Razzetto, M. (2024). Integrated Casing Wear Management in Well Design and Drilling Operations with a Field Case Analysis. Apogce 2024. https://doi.org/10.2118/221231-ms

11. Benberber, M. R., De Barros, A. J. C., De Melo, A. B., Iskander, M., Aghayarova, N., Manga, P., Bertheux, W. B., & Ledevin, P. (2024). Specialized High Tensile Strength and Quick Disconnect Tool Used to Achieve 6,500-ft Plug Length Saves 9 Days of Rig Time. Adipec.https://doi.org/10.2118/222761-ms

12. Mwale, W., Liu, Z., & Chipusu, K. (2025). A Hybrid AI Framework for Integrated Predictive Maintenance and Mineral Quality Assessment in Mining. Applied Sciences, 15(22), 12222. https://doi.org/10.3390/app152212222

13. Chen, X., Zheng, L., Jiang, Y., & Jiang, C. (2020). Transformation of minerals at the boundary of magma-coal contact zone: case study from Wolonghu Coal Mine, Huaibei Coalfield, China. International Journal of Coal Science & Technology, 8(1), 168–175. https://doi.org/10.1007/s40789-020-00373-6

14. Cai, J., Jiao, X., Wang, H., He, W., & Xia, Y. (2024). Multiphase fluid-rock interactions and flow behaviors in shale nanopores: A comprehensive review. Earth-Science Reviews, 257, 104884. https://doi.org/10.1016/j.earscirev.2024.104884

15. Perlin, A. P., Gomes, C. M., Kneipp, J. M., & Motke, F. D. (2018). Practices of energy use management and sustainable performance in Brazilian mineral industries. Environmental Quality Management, 27(4), 143–153. https://doi.org/10.1002/tqem.21566

16. Chen, C., Wang, L., Liu, R., Yu, J., Liu, H., & Wu, J. (2023). Chloride Penetration of Recycled Fine Aggregate Concrete under Drying–Wetting Cycles. Materials, 16(3), 1306. https://doi.org/10.3390/ma16031306

17. Luckanachai, P., Martino, A., & Boone, K. (2024). Permanent DAS/DTS Monitoring for SAGD Production – Design, Results, and Recommendations. SPE Thermal Well Integrity and Production Symposium. https://doi.org/10.2118/223127-ms

date of first submission of the article to the publication – 01/12/2026
date of acceptance of the article for publication after review – 02/14/2026
date of publication – 03/30/2026