Paper contribution

Feature extraction and classification of spatially varying rock mass around a circular tunnel using CNN

Pratishtha Mishra; Debarghya Chakraborty

Proceedings
EUROCK 2026
Pages
1349-1355
Submitted
28 January 2026
Accepted
15 June 2026
Published
12 August 2026

Summary

Abstract

This study aims to quantify the rotational anisotropic behavior of spatially variable rock slope mass in the presence of a tunnel. The rock mass is assumed to follow the Generalized Hoek-Brown (GHB) failure criterion, adhering to rock strength metrics such as the Geological Strength Index (GSI), the uniaxial compressive strength of intact rock (σci), and the intact rock constant (mi). The random fields of uniaxial compressive strength (σci) of rock mass are prepared using lower-bound random finite element limit analysis (LB-RFELA). The random fields generated are assumed to follow a log-normal distribution, and the rock properties are horizontally stratified. These fields are prepared using the Markovian autocorrelation function and the expansion optimal linear estimation (EOLE). Initially, a full-fledged numerical approach is performed to develop the random fields using random field theory (RFT) and Monte Carlo Simulation (MCS). The surcharge (qs) over the rock slope domain is optimized, and a dataset of input random field realizations, along with their corresponding stability numbers, is prepared. A convolutional neural network (CNN) is developed that extracts complex features, including erratic stratifications and their interactions with varying slope angles. Henceforth, the model predicts the stability of the rock slope domain embedded with a tunnel. The results show that the model reliably identifies the tunnel, slope angles, and autocorrelation lengths. Because the study emphasizes probabilistic behavior, key stratification parameters, particularly autocorrelation lengths, were incorporated during CNN development. The model’s predictions were compared with those from the rigorous numerical procedure, and the close agreement confirms the CNN’s efficiency in capturing spatial variability and anisotropic effects.

Contributors

Authors

  1. Pratishtha Mishra INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR, India ORCID 0009-0002-7996-4066
  2. Debarghya Chakraborty INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR, India ORCID 0000-0002-9383-9539

Index terms

Keywords

Full text

Attachments

Works cited

References

  1. Al-Bittar, T., & Soubra, A. H. (2017). Bearing capacity of spatially random rock masses obeying Hoek–Brown failure criterion. Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards, 11(2), 215-229.
  2. Ali, A., Lyamin, A. V., Huang, J., Sloan, S. W., & Cassidy, M. J. (2017). Undrained stability of a single circular tunnel in spatially variable soil subjected to surcharge loading. Computers and Geotechnics, 84, 16-27.
  3. Babu, G. S., & Srivastava, A. (2008). Response surface methodology (RSM) in the reliability analysis of geotechnical systems. Proc., 12th International Association for Computer Methods and Advances in Geomechanics, 4147-4165.
  4. Bharti, J. P., Mishra, P., Moorthy, U., Sathishkumar, V. E., Cho, Y., & Samui, P. (2021). Slope stability analysis using Rf, gbm, cart, bt and xgboost. Geotechnical and Geological Engineering, 39(5), 3741-3752.
  5. Chakraborty, D., & Kumar, J. (2013). Stability of a long unsupported circular tunnel in soils with seismic forces. Natural Hazards, 68(2), 419-431.
  6. Chen, F., Wang, L., & Zhang, W. (2019). Reliability assessment on stability of tunnelling perpendicularly beneath an existing tunnel considering spatial variabilities of rock mass properties. Tunnelling and Underground Space Technology, 88, 276-289.
  7. Craig, R. F. (2013). Soil mechanics. Springer.
  8. Der Kiureghian, A., & Ke, J. B. (1988). The stochastic finite element method in structural reliability. Probabilistic engineering mechanics, 3(2), 83-91.
  9. Fu, Y., Lin, M., Zhang, Y., Chen, G., & Liu, Y. (2022). Slope stability analysis based on big data and convolutional neural network. Frontiers of Structural and Civil Engineering, 16(7), 882-895.
  10. Goh, A. T. C., Zhang, W., Zhang, Y., Xiao, Y., & Xiang, Y. (2018). Determination of earth pressure balance tunnel-related maximum surface settlement: a multivariate adaptive regression splines approach. Bulletin of Engineering Geology and the Environment, 77(2), 489-500.
  11. Lambe, T. W., & Whitman, R. V. (2008). Soil mechanics SI version. John Wiley & Sons.
  12. Mishra, P., & Chakraborty, D. (2024, September). Meta-model Based Probabilistic Study of a Strip Footing Placed over a Spatially Varying Rock Mass. In Asian Rock Mechanics Symposium (pp. 285-294). Singapore: Springer Nature Singapore.
  13. Mishra, P., Samui, P., & Mahmoudi, E. (2021). Probabilistic design of retaining wall using machine learning methods. Applied Sciences, 11(12), 5411.
  14. Pan, Y., Zhang, L., Wu, X., Qin, W., &Skibniewski, M. J. (2019). Modeling face reliability in tunneling: A copula approach. Computers and Geotechnics, 109, 272-286.
  15. Rahaman, O., & Kumar, J. (2020). Stability analysis of twin horse-shoe shaped tunnels in rock mass. Tunnelling and Underground Space Technology, 98, 103354.
  16. Saikia, P., Baruah, R. D., Singh, S. K., & Chaudhuri, P. K. (2020). Artificial Neural Networks in the domain of reservoir characterization: A review from shallow to deep models. Computers & Geosciences, 135, 104357.
  17. Sloan, S. W. (1988). Lower bound limit analysis using finite elements and linear programming. International journal for numerical and analytical methods in geomechanics, 12(1), 61-77.
  18. Sloan, S. W., & Assadi, A. (1991). Undrained stability of a square tunnel in a soil whose strength increases linearly with depth. Computers and Geotechnics, 12(4), 321-346.
  19. Spanos, P. D., & Ghanem, R. (1989). Stochastic finite element expansion for random media. Journal of engineering mechanics, 115(5), 1035-1053.
  20. Sudret, B., & Der Kiureghian, A. (2000). Stochastic finite element methods and reliability: a state-of-the-art report (pp. 18-34). Berkeley: Department of Civil and Environmental Engineering, University of California.
  21. Vanmarcke, E., & Grigoriu, M. (1983). Stochastic finite element analysis of simple beams. Journal of engineering mechanics, 109(5), 1203-1214.
  22. Vořechovský, M. (2008). Simulation of simply cross correlated random fields by series expansion methods. Structural safety, 30(4), 337-363.
  23. Wang, Z. Z., & Goh, S. H. (2021). Novel approach to efficient slope reliability analysis in spatially variable soils. Engineering Geology, 281, 105989.
  24. Yamamoto, K., Lyamin, A. V., Wilson, D. W., Sloan, S. W., & Abbo, A. J. (2011). Stability of a circular tunnel in cohesive-frictional soil subjected to surcharge loading. Computers and Geotechnics, 38(4), 504-514.
  25. Zhang, J., Phoon, K. K., Zhang, D., Huang, H., & Tang, C. (2021). Deep learning-based evaluation of factor of safety with confidence interval for tunnel deformation in spatially variable soil. Journal of Rock Mechanics and Geotechnical Engineering, 13(6), 1358-1367.
  26. Zhang, W., Li, H., Li, Y., Liu, H., Chen, Y., & Ding, X. (2021). Application of deep learning algorithms in geotechnical engineering: a short critical review. Artificial Intelligence Review, 54(8), 5633-5673.
  27. Zhang, Y., Li, M., Han, S., Ren, Q., & Shi, J. (2019). Intelligent identification for rock-mineral microscopic images using ensemble machine learning algorithms. Sensors, 19(18), 3914.