Data-driven analysis of the spatial dependence of grouting efficiency during tunnel excavation

Abstract

Grouting with water-cement mixtures is the most widely used and cost-effective method for managing excess water inflow during tunnel construction. Due to uncertain geological and hydrological conditions, current grouting design relies heavily on the experience of onsite engineers. Recent advances in machine learning offer a promising alternative to traditional design to predict grout volume and improve grouting efficiency. Here, an artificial neural network (ANN) model was developed using the data set from an operation tunnel of Jurong Rock Caverns in Singapore to showcase an efficient and physics-guided training strategy. The ANN model was refined by incorporating the spatial scenarios, including the number of grouting holes in four quadrants of tunneling faces, the sequence of grouting screens along the tunnel axis, and the order of grouting rounds on the tunneling faces. The results indicate that an improved training strategy should encompass the grouting process, from Round 1 with grouting holes uniformly distributed around the tunnel periphery, to Round 2 with grouting holes drilled midway between neighboring first-round holes, and to Round 3 with grouting holes determined by onsite engineers. This model, trained based on the order of grouting rounds, performs better than the other models, highlighting the importance of establishing machine learning models grounded in physical principles. The finding was verified by the data set from another operation tunnel and concluded with a perspective on future grouting research.

Keywords

artificial neural networks / grout volume / spatial dependence / tunnel excavation

References

[1] Corson L, Reid C, Lunn RJ, et al. Field validation of a detectable, magnetic, cementitious grout for rock fracture grouting. Int J Rock Mech Min Sci. 2021; 145: 104853.

[2] Fang Z, Wu W. Dynamic change in dominant factor controls the injection-induced slip behaviors of rock fractures. Int J Rock Mech Min Sci. 2024; 183: 105887.

[3] Gustafson G, Stille H. Prediction of groutability from grout properties and hydrogeological data. Tunnel Undergr Space Technol. 1996; 11(3): 325-332.

[4] Hornik K, Stinchcombe M, White H. Multilayer feedforward networks are universal approximators. Neural Net. 1989; 2: 359-366.

[5] Jolliffe IT. Principal Component Analysis. 2nd edition. Springer; 2002.

[6] Kang H, Li W, Gao F, Yang J. Grouting theories and technologies for the reinforcement of fractured rocks surrounding deep roadways. Deep Undergr Sci Eng. 2023; 2: 2-19.

[7] Li S, Liu R, Zhang Q, Zhang X. Protection against water or mud inrush in tunnels by grouting: a review. J Rock Mech Geotech Eng. 2016; 8: 753-766.

[8] Liu H, Low BK. System reliability analysis of tunnels reinforced by rockbolts. Tunnel Undergr Space Technol. 2017; 65: 155-166.

[9] Liu Q, Xiao F, Zhao Z. Grouting knowledge discovery based on data mining. Tunnel Undergr Space Technol. 2020; 95: 103093.

[10] Meng W, Wu W. Machine learning-aided prediction of the mechanical properties of frozen fractured rocks. Rock Mech Rock Eng. 2023; 56: 261-273.

[11] Meng W, Xu N, Zhao Z, Wu W. Excavation-induced fault instability: a machine learning perspective. Rock Mech Rock Eng. 2024; 57: 5251-5265.

[12] Öge İF. Prediction of cementitious grout take for a mine shaft permeation by adaptive neuro-fuzzy inference system and multiple regression. Eng Geol. 2017; 228: 238-248.

[13] Pedregosa F, Varoquaux G, Gramfort A, et al. Scikit-learn: machine learning in python. J Mach Learn Res. 2011; 12: 2825-2830.

[14] Qin Y, Song D, Chen H, Cheng W, Jiang G, Cottrell G. A dual-stage attention-based recurrent neural network for time series prediction. ArXiv. 2017; 1704: 02971.

[15] Rafi J, Stille H. A method for determining grouting pressure and stop criteria to control grout spread distance and fracture dilation. Tunnel Undergr Space Technol. 2021; 112: 103885.

[16] Stille B, Gustafson G. A review of the Namntall tunnel project with regard to grouting performance. Tunnel Undergr Space Technol. 2010; 25(4): 346-356.

[17] Sun L, Grasselli G, Liu Q, Tang X. Coupled hydro-mechanical analysis for grout penetration in fractured rocks using the finite-discrete element method. Int J Rock Mech Min Sci. 2019; 124:104138.

[18] Tian M, Chen S, Han L, Xiao H. Development and application of novel high-efficiency composite ultrafine cement grouts for roadway in fractured surrounding rocks. Deep Undergr Sci Eng. 2024; 3: 53-69.

[19] Tie J, Meng W, Wei M, Wu W. Loading rate and mineralogical controls on tensile strength of rocks: a machine learning view. Rock Mech Rock Eng. 2023; 56: 6119-6125.

[20] Tie J, Wu W. Intelligent detection of dynamic cracking along an interface of brittle material using high-speed photography assisted by data augmentation and machine learning. Int J Rock Mech Min Sci. 2024; 179: 105784.

[21] Wanniarachchi WAM, Wu W. Permeability evolution of rock-concrete interfaces in underground lined storage systems. Int J Rock Mech Min Sci. 2021; 143: 104792.

[22] Xiao F, Liu Q, Zhao Z. Information and knowledge behind data from underground rock grouting. J Rock Mech Geotech Eng. 2021; 13: 1326-1339.

[23] Zhao J, Liu Q, Lee KW, Choa V, Teh CI. Underground cavern development in the Jurong sedimentary rock formation. Tunnel Undergr Space Technol. 1999; 14: 449-459.

RIGHTS & PERMISSIONS

2025 The Author(s). Deep Underground Science and Engineering published by John Wiley & Sons Australia, Ltd on behalf of China University of Mining and Technology.