Khabarovsk Federal Research Center, Far Eastern Branch, Russian Academy of Sciences, Khabarovsk, Russia

I. Yu. Rasskazov, Director, Academician of the Russian Academy of Sciences, Doctor of Engineering Sciences, adm@igd.khv.ru

Institute of Mining, Far Eastern Branch, Russian Academy of Sciences, Khabarovsk, Russia

A. V. Konstantinov, Researcher, Candidate of Engineering Sciences
A. P. Grunin, Senior Researcher, Candidate of Engineering Sciences
Yu. V. Fedotova, Leading Researcher, Candidate of Engineering Sciences

The problem of predicting hazardous geodynamic phenomena in conditions of high-stress rock mass remains highly relevant. This is due to the increasing depth of mining operations, the complexity of geological conditions, and the growing anthropogenic impact. This study presents an approach to assessing rockburst hazard based on seismoacoustic data recorded by the Prognoz-ADS system at the Yuzhnoye deposit. Time series were constructed for multiple parameters of seismoacoustic signals, events and acoustically active zones. To enhance the informativeness of predictive models, the automatic selection attributes was applied using the Gini criterion. Additionally, the individually calculated weighting coefficients were used to address the pronounced imbalance of classes. Each time series had a unique set of attributes, and no consistent patterns of predominance of one attributes over the other attributes were revealed. Random forest and gradient boosting models were trained on both full dataset and test samples. The results demonstrated high predictive accuracy (up to 84 % at a probability threshold of 0.2) and identified periods with elevated rockburst risk. The developed models were found to effectively recognize precursors of rock pressure. They also show potential for retrospective analysis of the stress–strain behavior of rock mass. To improve the safety of mining operations under conditions of increased rock pressure, the proposed approach is recommended for implementation in production practice for testing and further refinement.
The studies were carried out using the resources of the Shared Use Center for Mineral Resources Research at the Khabarovsk Federal Research Center, Far Eastern Branch, Russian Academy of Sciences, supported by the Russian Federation represented by the Ministry of Science and Higher Education under Project No. 075-15-2025-621.

1. Rahimdel M. J., Aryafar A., Vaziri S. Fuzzy FMEA for the safety risk analysis of underground coal mining (a case study in Iran). Mining Technology: Transactions of the Institutions of Mining and Metallurgy. 2022. Vol. 131, Iss. 2. pp. 104–114.

2. Wojtecki Ł., Iwaszenko S., Apel D. B. An attempt to use machine learning algorithms to predict strong tremors during longwall mining of a coal seam. Journal of Applied Geophysics. 2023. Vol. 218. ID 105210.
3. Papadopoulos D., Benardos A. Enhancing machine learning algorithms to assess rock burst phenomena. Geotechnical and Geological Engineering. 2021. Vol. 39, Iss. 8. pp. 5787–5809.
4. Wang K., Zhuang X., Zhao X., Wu W., Liu B. Roof pressure prediction in coal mine based on Grey Neural Network. IEEE Access. 2020. Vol. 8. pp. 11705–11706.
5. Zhang Y., Fang K., He M., Liu D., Wang J. et al. Rockburst prediction using artificial intelligence techniques: A review. Rock Mechanics Bulletin. 2024. Vol. 3, Iss. 3. ID 100129.
6. Westman E. C., Wempen J. M., Coons D. J., McCarter M. K., Pariseau W. G. Stress redistribution in a longwall yield pillar—A comparison between active seismic tomography and theory. Mining, Metallurgy & Exploration. 2022. Vol. 39, Iss. 3. pp. 1017–1026.
7. Kurlenya M. V., Skazka V. V., Azarov A. V., Serdyukov A. S., Patutin A. V. Using surface waves for monitoring rock mass condition around underground openings and structures. Journal of Mining Science. 2022. Vol. 58, No. 6. pp. 875–885.
8. Ma T.-H., Tang C.-A., Liu F., Zhang S.-C., Feng Z.-Q. Microseismic monitoring, analysis and early warning of rockburst. Geomatics, Natural Hazards and Risk. 2021. Vol. 12. pp. 2956–2983.
9. Eremenko A. A., Mulev S. N., Shtirts V. A. Microseismic monitoring of geodynamic phenomena in rockburst-hazardous mining conditions. Journal of Mining Science. 2022. Vol. 58, No. 1. pp. 10–19.
10. Liu G.-F., Jiang Q., Feng G.-L., Chen D.-F., Chen B.-R. et al. Microseismicity-based method for the dynamic estimation of the potential rockburst scale during tunnel excavation. Bulletin of Engineering Geology and the Environment. 2021. Vol. 80, Iss. 5. pp. 3605–3628.
11. Rasskazov I. Yu., Anikin P. A., Grunin A. P., Konstantinov A. V. Geomechanical monitoring for safe and efficient subsoil use: Methods and tools. Gornyi Zhurnal. 2025. No. 3. pp. 4–11.
12. Konstantinov A. V., Lomov M. A. Application of cluster analysis methods for identifying acoustic active zones. Problemy nedropolzovaniya. 2025. No. 1(44). pp. 78–88.
13. Azuma H., Kunimasa H., Kusumo A. W., Oda Y., Watanabe T. et al. Application of machine learning methods for earthquake detection from high-density temporary observation seismic records on a v olcanic island. Journal of Applied Geophysics. 2024. Vol. 230. ID 105503.
14. Li Y., Yang T.-H., Liu H.-L., Wang H., Hou X.-G. et al. Real-time microseismic monitoring and its characteristic analysis in working face with high-intensity mining. Journal of Applie d Geophysics. 2016. Vol. 132. pp. 152–163.
15. Zhang P., Li X., Chen J. Prediction method for mine earthquake in time sequence based on clustering analysis. Applied Sciences. 2022. Vol. 12, Iss. 21. ID 11101.
16. Nicolis O., Plaza F., Salas R. Prediction of inte nsity and location of seismic events using deep learning. Spatial Statistics. 2021. Vol. 42. ID 100442.
17. Breiman L. Random Forests. Machine Learning. 2001. Vol. 45. pp. 5–32.
18. Wang S., Zhuang J., Zheng J., Fan H., Kong J. et al. Application of Bayesian hyperparameter optimized random forest and XGBoost model for landslide susceptibility mapping. Frontiers in Earth Science. 2021. Vol. 9. ID 712240.
19. Genuer R., Poggi J.-M., Tuleau-Malot C. Variable selection using random forests. Pattern Recognition Letters. 2010. Vol. 31, Iss. 14. pp. 2225–2236.