Creation of neural network model package to identify acoustic emission sources at rockburst-hazardous deposits

Modern approaches to geomechanical research include mathematical modeling and machine learning techniques. The maximum progress is achieved in combination of theoretical and in-situ methods with continuous monitoring using intelligent geomechanical systems. Spotlight of this study is on the geoacoustic monitoring system Prognoz-ADS installed in the complex ore mine Yuzhny in the east of the Primorye, which is the most rockburst-hazardous mine in Russia. The system allows recording acoustic waves generated by micro failures in rocks and detecting hazardous geodynamic events at the incipient stage. Efficiency of the system depends in many ways on the identification of useful signals in a general flow of the recorded events. The proposed algorithms based on neural network models enable classification of acoustic signals by their emission sources. Developed in the course of the research, an ensemble machine learning model including a set of neural networks makes it possible to identify the type of an emission source based on the parametric characteristics of acoustic signals. The obtained results prove high efficiency of the developed tools in identification of natural signals in the general acoustic emission flow; such signals in the Yuzhny mine reach 14.8% of the total number of the recorded impulses. The model application improves automation of data processing, enables acquisition of more complete information on geomechanical condition of rock mass, and enhances safety and efficiency of mining.

Konstantinov A. V., Rasskazov I. Yu. Creation of neural network model package to identify acoustic emission sources at rockburst-hazardous deposits. MIAB. Mining Inf. Anal. Bull. 2024;(11):23-36. [In Russ]. DOI: 10.25018/0236_1493_2024_11_0_23.

A.V. Konstantinov, Researcher, Mining Institute of Far Eastern Branch of RAS, 680000, Khabarovsk, Russia, e-mail: alex-sdt@yandex.ru, ORCID ID: 0000-0001-6481-292X,
I.Yu. Rasskazov, Corresponding Member of Russian Academy of Sciences, Dr. Sci. (Eng.), Director, Khabarovsk Federal Research Center of the Far Eastern Branch RAS, 680000, Khabarovsk, Russia, e-mail: rasskazov@igd.khv.ru, ORCID ID: 0000-0002-2215-6642.

A.V. Konstantinov, e-mail: alex-sdt@yandex.ru.

1. Batugin A. S., Batugina I. M. Petukhov I. M. Gornoe delo i okruzhayushchaya sreda. Geodinamika nedr [Mining and the Environment. Geodynamics of the subsurface], Moscow, Nedra, 2012, 121 p.

2. Kocharyan G. G. Occurrence and development of slip processes in continental fault zones under the action of natural and anthropogenic factors. Review of the current state of the issue. Fizika Zemli. 2021, no. 4, pp. 3—41. [In Russ]. DOI: 10.31857/S0002333721040062.

3. Feng J., Wang E., Ding H., Huang Q., Chen X. Deterministic seismic hazard assessment of coal fractures in underground coal mine. A case study. Soil Dynamics and Earthquake Engineering. 2020, vol. 129, article 105921. DOI: 10.1016/j.soildyn.2019.105921.

4. Batugin A. A proposed classification of the Earth's crustal areas by the level of geodynamic threat. Geodesy and Geodynamics. 2021, vol. 12, no. 1, pp. 21—30. DOI: 10.21203/rs.3.rs-62724/v1.

5. Kozyrev A. A., Savchenko S. N., Panin V. I., Semenova I. E., Rybin V. V., Fedotova Yu. V., Kozyrev S. A. Geomekhanicheskie protsessy v geologicheskoy srede gornotekhnicheskikh sistem i upravlenie geodinamicheskimi riskami [Geomechanical processes in the geological environment of mining systems and geodynamic risk management], Apatity, KNTS RAN, 2019, 431 p. DOI: 10.37614/ 978.5.91137.391.7.

6. Oparin V., Tapsiev A., Freidin A. Classification of mining methods for deep orebodies. Advances in Applied Strategic Mine Planning. Springer, Cham, 2018, pp. 559—572. DOI: 10.1007/978-3-31969320-0_32.

7. Shabarov A. N., Tsirel' S. V., Morozov K. V., Rasskazov I. Yu. Concept of integrated geodynamic monitoring at underground mining operations. Gornyi Zhurnal. 2017, no. 9, pp. 59—64. [In Russ]. DOI: 10.17580/gzh.2017.09.11.

8. Trubetskoi K. N., Viktorov S. D., Osokin A. A., Shlyapin A. V. Prediction of rock burst based on control of submicron particle emission during deformation and rock fractures. Gornyi Zhurnal. 2017, no. 6, pp. 16—20. [In Russ]. DOI: 10.17580/gzh.2017.06.03.

9. Zhou J., Zhang Y., Li C., He H., Li X. Rockburst prediction and prevention in underground space excavation. Underground Space. 2023, vol. 14, pp. 70—98. DOI: 10.1016/j.undsp.2023.05.009.

10. Feng X. T., Liu J., Chen B., Xiao Y., Feng G., Zhang F. Monitoring, warning, and control of rockburst in deep metal mines. Engineering. 2017, vol. 3, no. 4, pp. 538—545. DOI: 10.1016/J.ENG. 2017.04.013.

11. Wang A., Qiu L., Liu Y., Lou Q., Sun Z., Wang W. Study on synchronous response law of acoustic and electrical signals of outburst coal rock under load and fracture. Geofluids. 2023, no. 2, pp. 1—11. DOI: 10.1155/2023/1253236.

12. He X., Zhou C., Song D., Li X., He S., Khan M., Cao A. Mechanism and monitoring and early warning technology for rockburst in coal mines. International Journal of Minerals, Metallurgy and Materials. 2021, no. 28, pp. 1097—1111. DOI: 10.1007/s12613-021-2267-5.

13. Manchao H., Fuqiang R., Dongqiao L. Rockburst mechanism research and its control. International Journal of Mining Science and Technology. 2018, vol. 28, no. 5, pp. 829—837. DOI: 10.1016/ j.ijmst.2018.09.002.

14. Rasskazov I. Yu., Kalinov G. A., Anikin P. A., Migunov D. S. Patent na promyshlennyy obrazets RU 129484 [Design patent of the Russian Federation No. 129484], 25.01.2022. [In Russ].

15. Agletdinov E. A. Issledovanie protsessa deformatsii metallicheskikh materialov s primeneniem statisticheskogo podkhoda k analizu vremennykh ryadov akusticheskoy emissii [Study of the deformation process of metallic materials using a statistical approach to the analysis of acoustic emission time series], Candidate’s thesis, Samara TGU, 2021, 24 p.

16. Krasnoyarov N. A., Dmitrieva T. L. Current state of nondestructive testing methods and possibilities of their automation. Uchenye zapiski Komsomol'skogo-na-Amure gosudarstvennogo tekhnicheskogo universiteta. 2022, no. 3(59), pp. 35—42. [In Russ]. DOI: 10.17084/20764359-2022-59-35.

17. Gholizadeh S., Leman Z., Baharudin B. T. H. T. State-of-the-art ensemble learning and unsupervised learning in fatigue crack recognition of glass fiber reinforced polyester composite (GFRP) using acoustic emission. Ultrasonics. 2023, vol. 132, article 106998. DOI: 10.1016/j.ultras.2023.106998.

18. Peng K., Tang Z., Dong L., Sun D. Machine learning based identification of microseismic signals using characteristic parameters. Sensors. 2021, no. 21, 6967. DOI: 10.3390/s21216967.

19. Çakır E., Parascandolo G., Heittola T., Huttunen H., Virtanen T. Convolutional recurrent neural networks for polyphonic sound event detection. IEEE ACM Transactions on Audio, Speech, and Language Processing. 2017, vol. 25, no. 6, pp. 1291—1303. DOI: 10.1109/TASLP.2017.2690575.

20. Liu R. Research on feature fusion method of mine microseismic signal based on unsupervised learning. Shock and Vibration. 2021, vol. 2021. DOI: 10.1155/2021/9544997.

21. Olkhovskiy M., Mullerova E., Martinek P. Impulse signals classification using one dimensional convolutional neural network. Journal of Electrical Engineering. 2020, vol. 71, no. 6, pp. 397—405. DOI: 10.2478/jee-2020-0054.

22. Romanevich K. V., Mulev S. N. Automation of seismic events classification during seismic monitoring at a coal mine using machine learning. Russian Mining Industry Journal. 2023, no. 5S, pp. 58—64. [In Russ]. DOI: 10.30686/1609-9192-2023-5S-58-64.

23. Anikin P. A., Tereshkin A. A., Sidlyar A. V., Rasskazov M. I., Konstantinov A. V., Grunin A. P. Svidetel'stvo o gosudarstvennoy registratsii bazy dannykh RU 2023624340 [Certificate of state registration of the database of the Russian Federation No. 2023624340], 04.12.2023. [In Russ].