Application of convolutional neural networks for rock stability identification and failure time prediction

Abstract

The identification of rock stability and the prediction of failure time are crucial for the early warning and prevention of sudden geological disasters such as landslides and collapses. To address these challenges, this study proposes three convolutional prediction models: CNN-LSTM-Attention, CNN-BiLSTM-Attention, and CNN-GRU-Attention. The displacement coordination coefficient (DCC) index and stress curves were employed as input variables to evaluate the performance of each model in discriminating rock stability states under different data structures and input configurations. Furthermore, an innovative methodology for predicting rock failure time utilizing convolutional models was developed. The experimental results demonstrate that the CNN-LSTM-Attention model, utilizing a 10 × 10 × 2 data structure, exhibits superior performance in rock stability state discrimination, achieving an accuracy of 95.25 % on the validation set and a recall rate of 96 % for samples in high-risk areas. Furthermore, when the DCC index was used as the input variable, the CNN-LSTM-Attention model achieved recall rates of 95.8 % and 86.5 % for medium- and high-risk areas, respectively, in the validation set. These findings indicate that the proposed convolutional models can effectively identify rock stability states by leveraging surface deformation characteristics. The CNN-LSTM-Attention model, with the DCC index as the input variable, is capable of predicting the final rock failure time in real-time once the DCC abrupt change exceeds 0.78. For different rocks, the model can predict the failure time within 20 s after the DCC reaches 0.78, with an error rate of less than 9 %. The convolutional neural network model, developed based on the DCC index, provides a novel methodological approach for geohazard early warning research, facilitating slope instability monitoring and earthquake precursor identification using GNSS and other displacement measurement techniques.

Keywords

Rock failure time / Convolution model / Precursor of rock failure / Displacement coordination coefficient

CRediT authorship contribution statement

Weiyang Li: Writing - review & editing, Writing - original draft, Visualization, Validation, Investigation, Data curation, Conceptualization. Yongxing Shen: Writing - review & editing, Writing - original draft, Visualization, Validation, Supervision, Methodology. Zengchao Feng: Supervision, Project administration, Investigation. Xuchen Guo: Visualization, Investigation.

Declaration of competing interest

Zengchao Feng is an editorial board member for Geohazard Mechanics and was not involved in the editorial review or the decision to publish this article. All other authors declare that there are no competing interests.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (No.52474106).

References

[1] Q. Xu, R.Q. Huang, X.Q. Xiang, Prediction of geological disasters in time and space, J. Mt. Sci. (S1) (2000) 112-117.

[2] Z. Feng, Z.X. Lv, Y.S. Zhao, Research progress on short-impending prediction of rock failure—Comments on the competition of short-impending prediction of rock failure, J. Rock Mech. Geotech. Eng. 41 (12) (2022) 2522-2529.

[3] Y.S. Zhao, Retrospection on the development of rock mass mechanics and the summary of some unsolved centennial problems, J. Rock Mech. Geotech. Eng. 40 (7) (2021) 1297-1336.

[4] S.J. Zhou, Research on Critical power-law Singularity Precursor and Failure Time Prediction, Yanshan University, 2020.

[5] D.Z. Song, Q. Liu, L.M. Qiu, J.G. Zhang, Khan Majid, Y.J. Peng, Y.J. Zhao, M. Wang, M. G. Guo, T.T. Hong, Experimental study on resistivity evolution law and precursory signals in the damage process of gas-bearing coal, Fuel 362 (2024) 130798.

[6] Y.S. Pan, Y.M. Song, C.L. Zhu, H. Ren, H.L. Xu, Localization method of coal rock deformation for rock burst prediction, J. China Coal Soc. 48 (1) (2023) 185-198.

[7] J. Wang, J.W. Gong, Z.Y. Zhao, Position, direction of strain localization of rock-like specimens under uniaxial compression and its application to early-warning, Rock Soil Mech. 39 (S2) (2018) 186-194.

[8] X.B. Wang, W.T. Hou, W. Dong, Evaluation of failure precursors of strain field in uniaxial compressed coal samples based on digital image correlation method, J. Saf. Environ. 18 (4) (2018) 1237-1245.

[9] Y.X. Shen, Z.C. Feng, Study on precursor information of rock instability based on displacement increments measured at multiple points, Nat. Hazards 113 (2022) 1713-1727.

[10] H.M. Cheng, Study on Energy Conversion Mechanism and State Identification During Rock Deformation and Instability, China University of Mining & Technology, Beijing, 2022.

[11] Y.X. Shen, Z.C. Feng, D. Zhou, B.C. Zhang, Different rock failure precursors using displacement coordination coefficient, Environ. Earth Sci. 83 (2024) 490.

[12] A. Arif, C.L. Zhang, M.H. Sajib, M.N. Uddin, M. Habibullah, R.M. Feng, M.J. Feng, M. S. Rahman, Y. Zhang, Rock slope stability prediction: a review of machine learning techniques, Geotech. Geol. Eng. 43 (2025) 124.

[13] F.M. Huang, H.W. Xiong, S.C. Chen, Z.T. Lv, J.S. Huang, Z.L. Chang, F. Catani, Slope stability prediction based on a long short-term memory neural network: comparisons with convolutional neural networks, support vector machines and random forest models, Int. J. Coal Sci. Technol. 10 (2023) 18.

[14] M.S Lin, L.M. Zeng, S Teng, G.F. Chen, B Hu, Prediction of stability of a slope with weak layers using convolutional neural networks, Nat Hazards 120 (2024) 12081-12105.

[15] F. Chen, Z. Liang, A. Cao, ConvLSTM for predicting short-term spatiotemporal distribution of seismic risk induced by large-scale coal mining, Nat. Resour. Res. 32 (2023) 1459-1479.

[16] J.Y. Li, D. Chen, E.Y. Wang, H.Z. Yang, S.L. Zhao, H.S. Jia, Y.W. Sun, Y. Song, X.Y. Fan, Enhanced convolutional neural network-based prediction and source mechanism interpretation of high-energy microseismic events in coal mines, Phys. Fluids 37 (4) (2025) 046606.

[17] Y.X. Zhao, H.Q. Qiao, R.H. Xie, J.H. Guo, Lightweight 3D convolution model for failure prediction based on acoustic emission of coal under uniaxial compression, J. Rock Mech. Geotech. Eng. 41 (8) (2022) 1567-1580.

[18] Dengke Zhang, Yang Han, Chuanle Wang, Lei Gao, Hui Lu, Liang Chen, Erbing Li, Time series prediction of tunnel surrounding rock deformation using CPO-CLA integrated model, J. Rock Mech. Geotech. Eng. 17 (12) (2025) 7915-7930.

[19] Xuejiao Li, Chunyu Liu, Dezhi Zhou, Liang Yu, Xingcai Lu, A unified deep learning model for thermoacoustic instability detection in swirl premixed flames based on spatial-temporal features, Aero. Sci. Technol. 159 (2025) 109994.

[20] Qing Wei, Huijin Zhang, Ju Yang, Bin Niu, Zuxin Xu Pm2, 5 concentration prediction using a whale optimization algorithm based hybrid deep learning model in beijing. China, Environ. Pollut. 371 (2025) 125953.

[21] Seyed Mahdi Miraftabzdeh, Mohammed Ali Khan, Navid Bayati, Dario Zaninelli, Deep learning methods and evaluation of the extensive carbon emission predictive solution for Danish grid, Sustain. Energy Technol. Assessments 75 (2025) 104242.

[22] M.J. Shi, B.H. Yang, R. Chen, D.S. Ye, Logging curve prediction method based on CNN-LSTM-attention, Earth Sci. Inform. 15 (2022) 2119-2131.

[23] L.M. Qiu, J.T. Dang, J.G. Zhang, M. Wang, Q. Liu, L.L. Si, Z.Y. Jiang, M. Khan, Investigating nonlinear resistivity characteristics and mechanisms of coal during various loading stages, J. Appl. Geophys. 238 (2025) 105705.

[24] Y.J. Wei, M.F. Xia, F.J. Ke, X.C. Yin, Y.L. Bai,Evolution-induced catastrophe and its predictability, in: Microscopic and Macroscopic simula-tion: towards Predictive Modelling of the Earthquake Process, Springer, 2000, pp. 1945-1957.

[25] M.F. Xia, Y.J. Wei, F.J. Ke, Y.L. Bai,Critical sensitivity and trans-scale fluctuations in catastrophic rupture. Earthquake Processes: Physical Modelling, Numerical Simulation and Data Analysis Part II, Springer, 2002, pp. 2491-2509.

[26] Y.L. Bai, H.Y. Wang, M.F. Xia, F.J. Ke, Statistical mesomechanics of solid, linking coupled multiple space and time scales, Appl. Mech. Rev. 58 (6) (2005) 372-388.

[27] Y. LeCun, L. Bottou, Y. Bengio, P. Haffner, Gradient-based learning applied to document recognition, Proc. IEEE 86 (11) (1998) 2278-2324.

[28] Y. LeCun, B. Boser, J.S. Denker, D. Henderson, R.E. Howard, W. Hubbard, L.D. Jackel, Backpropagation applied to handwritten zip code recognition, Neural Comput. 11 (4) (1989) 541-551.

[29] S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Comput. 9 (8) (1997) 1735-1780.

[30] K. Cho, B.V. Merri€enboer, C. Gulcehre, D. Bahdanau, F. Bougares, H. Schwenk, Y. Bengio, Learning phrase representations using RNN encoder-decoder for statistical Ma-chine translation, Comput. Sci. (2014) 1724-1734.

[31] V. Mnih, N. Heess, A. Graves, K. Kavukcuoglu,Recurrent models of visual attention, in:Proceedings of the 27th International Conference on Neural Information Processing Systems, 2014, pp. 2204-2212.

[32] R.M. Neal, Markov chain sampling methods for dirichlet process mixture models,J. Comput. Graph Stat. 9 (2) (2000) 249-265.

[33] R.A. Fisher, The use of multiple measurements in taxonomic problems, Ann. Eugen. 7 (2) (1936) 179-188.

[34] Z.C. Feng, Y.X. Shen, Y.S. Zhao, Research on earthquake prediction methodology utilizing the displacement coordination coefficient of rock failure and GNSS, Chin.J. Geophys. 67 (11) (2024) 4204-4219, in Chinese.