Identification of geological conditions based on stacking algorithm during slurry shield tunneling

Abstract

Geological detection ahead of shield cutterhead is a challenge in the construction of large-diameter shield tunnels. This study proposed a framework for geological conditions identification based on stacking algorithm in large-diameter shield tunnel excavation of Chunfeng tunnel project. The relationships among shield parameters were analyzed using the Spearman rank correlation coefficient. Principal component analysis was employed to extract essential information from the parameters during shield tunneling. Then, the K-means+ +clustering algorithm was used to construct a correlation dataset between the principal components of shield parameters and the categories of geological conditions. Subsequently, the Stacking algorithm was applied to recognize the geological categories ahead of the shield cutterhead, and the results were compared with those of optimized random forest, support vector machine, and gradient boosting decision tree algorithms. The proposed method achieved superior performance, accurately classifying geological features ahead of the shield cutterhead with an accuracy of 0.984. Moreover, the stacking algorithm achieves higher accuracy in geological category recognition compared to the optimized single-machine learning algorithms. This research results can provide valuable geological conditions for adjusting shield parameters, gesture control, and risk reduction during the large-diameter shield tunneling.

Keywords

Large-diameter tunnel / Shield construction / Geological identification / Machine learning / Stacking algorithm

References

[1] Li Set al.. An overview of ahead geological prospecting in tunneling. Tunnelling Underground Space Technol, 2017, 63: 69-94

[2] Zhang Q, Liu Z, Tan J (2019) Prediction of geological conditions for a tunnel boring machine using big operational data. Automation in Construction 100:73-83. https://doi.org/10.1016/j.autcon.2018.12.022

[3] Broere W. Urban underground space: solving the problems of today’s cities. Tunnelling Underground Space Technol, 2016, 55: 245-248

[4] Qin S, Cheng Y, Zhou WH (2023) State-of-the-art review on pressure infiltration behavior of bentonite slurry into saturated sand for TBM tunneling. Smart Construction and Sustainable Cities 1:14. https://doi.org/10.1007/s44268-023-00018-y

[5] Wang ZZet al.. Predicting geological interfaces using stacking ensemble learning with multi-scale features. Can Geotech J, 2023, 60(7): 1036-1054

[6] Zhou Z, Irizarry J, Zhou J. Development of a database exclusively for subway construction accidents and corresponding analyses. Tunnelling Underground Space Technol, 2021, 111: 103852

[7] Zhang N, Shen S-L. Prediction of TBM cutter wear in heterogeneous ground under high ambient pressure. Comput Geotech, 2025, 177: 106837

[8] Zhang Net al.. Identification of cutter flat wear for rock TBMs using vibration signal of shield. Mech Syst Signal Process, 2025, 223: 111885

[9] Paraskevopoulou Cet al.. Assessing the failure potential of tunnels and the impacts on cost overruns and project delays. Tunnelling Underground Space Technol, 2022, 123: 104443

[10] Sousa RL, Einstein HH. Lessons from accidents during tunnel construction. Tunnelling Underground Space Technol, 2021, 113: 103916

[11] Kneib G, Kassel A, Lorenz K (2000) Automatic seismic prediction ahead of the tunnel boring machine. First Break, 18(7):295-302. https://doi.org/10.1046/j.1365-2397.2000.00079.x

[12] Yamamoto Tet al.. Imaging geological conditions ahead of a tunnel face using three-dimensional seismic reflector tracing system. International Journal of the JCRM, 2011, 6(1): 23-31

[13] Wang Y-Net al.. New index to identify karst caves based on operational parameters during shield tunnelling. Can Geotech J, 2025, 62: 1-24

[14] Yan T, Shen S-L, Zhou A. GFII: a new index to identify geological features during shield tunnelling. Tunnelling Underground Space Technol, 2023, 142: 105440

[15] Zhang N, Shen S-L, Zhou A. A new index for cutter life evaluation and ensemble model for prediction of cutter wear. Tunnelling Underground Space Technol, 2023, 131: 104830

[16] Yan Tet al.. Prediction of geological characteristics from shield operational parameters by integrating grid search and K-fold cross validation into stacking classification algorithm. J Rock Mech Geotech Eng, 2022, 14(4): 1292-1303

[17] Portugal I, Alencar P, Cowan D. The use of machine learning algorithms in recommender systems: a systematic review. Expert Syst Appl, 2018, 97: 205-227

[18] Zhou Cet al.. Deep learning technologies for shield tunneling: challenges and opportunities. Autom Constr, 2023, 154: 104982

[19] Liu H-C, Zhang N, Yin Z-Y (2025) Probabilistic stratigraphic modelling from sparse boreholes based on deep learning. Geotechnique 75(11): 1457-1469

[20] Yan T, Shen S-L, Zhou A. Data augmentation-assisted muck image recognition during shield tunnelling. Underground Space, 2025, 21: 370-383

[21] Shan Fet al.. Applications of machine learning in mechanised tunnel construction: a systematic review. Eng, 2023, 4(2): 1516-1535

[22] Geng Zet al.. An uncertainty weighted kriging-random forest model for real-time ground property prediction during earth pressure balance shield tunneling. Appl Soft Comput, 2025, 180: 113417

[23] Ayawah PEAet al.. A review and case study of artificial intelligence and machine learning methods used for ground condition prediction ahead of tunnel boring machines. Tunnell Underground Space Technol, 2022, 125: 104497

[24] Fu Xet al.. Data-driven real-time advanced geological prediction in tunnel construction using a hybrid deep learning approach. Autom Constr, 2023, 146: 104672

[25] Wang Yet al.. Prediction of TBM operation parameters using machine learning models based on BPSO. Adv Eng Inform, 2023, 56: 101955

[26] Wu Yet al.. Real-time prediction of tunnel boring machine thrust based on multi-resolution analysis and online learning. Computer-Aided Civil and Infrastructure Engineering, 2025, 40(28): 5212-5227

[27] Wold S, Esbensen K, Geladi P. Principal component analysis. Chemometr Intell Lab Syst, 1987, 2(1): 37-52

[28] Vassilvitskii S, Arthur D. (2006) k-means++: The advantages of careful seeding. In Proceedings of the eighteenth annual ACM-SIAM symposium on Discrete algorithms

[29] Toy T, Das P, Jagirdar R et al (2025) Prediction of mechanical properties of eco-friendly concrete using machine learning algorithms and partial dependence plot analysis. Smart Construction and Sustainable Cities 3:2. https://doi.org/10.1007/s44268-025-00048-8

[30] Zeng Yet al.. Risks analysis of large diameter slurry shield tunneling in urban area. Undergr Space, 2023, 13: 281-300

[31] Hartigan JA, Wong MA. Algorithm AS 136: a k-means clustering algorithm. Appl Stat, 1979, 28(1): 100-108

[32] Rousseeuw PJ. Silhouettes: a graphical aid to the interpretation and validation of cluster analysis. J Comput Appl Math, 1987, 20: 53-65

RIGHTS & PERMISSIONS

The Author(s)