Enhancing Prediction Performance in Tunnel Water Inflow with the Improved Stacking Machine Learning Framework

Abstract:

Accurate prediction of tunnel water inflow (WI) is critical for preventing construction hazards and supporting engineering decision-making. This study proposes an enhanced stacking machine learning framework to achieve reliable WI prediction. SMOGN technology was used to address the issue of imbalanced data distribution. On this basis, the Bayesian optimization tool Optuna was utilized for automated hyperparameter tuning of various base learners, including elastic net (Enet), multilayer perceptron (MLP), support vector regression (SVR), and extreme gradient boosting (XGBoost). Base learner outputs were combined with original features to form augmented inputs, with ridge regression as the meta-learner. Gaussian process regression (GPR) modeled residuals for uncertainty quantification, while the Sobol method assessed feature importance. Results show that feature augmentation and residual modeling improve prediction performance. The proposed stacking model outperforms individual base models and achieves state-of-the-art results. From an engineering perspective, the framework can be used to forecast potential high-inflow zones, guide advance grouting or drainage measures.

• Periodical: International Journal of Engineering Research in Africa (Volume 77)
• Pages: 85-105
• DOI: https://doi.org/10.4028/p-hbPUx6
• Online since: December 2025
• Authors: Xiao Duo Ou, Yufang Wu*, Jie Jiang
• Keywords: Drill-and-Blast Tunnel, Machine Learning, Residual Modeling, Stacking Method, Water Inflow

References

[1] Liu X, Shen S, Xu Y, et al.. (2018). Analytical approach for time‐dependent groundwater inflow into shield tunnel face in confined aquifer. International Journal for Numerical and Analytical Methods in Geomechanics, 42(4), 655-673.

DOI: 10.1002/nag.2760

[2] Li X, Zhang Q, Zhang X, et al.. (2018). Detection and treatment of water inflow in karst tunnel: A case study in Daba tunnel. Journal of Mountain Science, 15(7), 1585-1596.

DOI: 10.1007/s11629-018-4919-0

[3] Wu J, Jia C, Zhang LW. (2021). Expansion of Water Inrush Channel by Water Erosion and Seepage Force. International Journal of Geomechanics, 21(7), 04021121.

DOI: 10.1061/(ASCE)GM.1943-5622.0001985

[4] Li Z, Chen Z, He C, et al.. (2022). Seepage field distribution and water inflow laws of tunnels in water-rich regions. Journal of Mountain Science,19(2), 591-605.

DOI: 10.1007/s11629-020-6634-x

[5] Farhadian H, Katibeh H. (2017). New empirical model to evaluate groundwater flow into circular tunnel using multiple regression analysis. International Journal of Mining Science and Technology, 27(3), 415-421.

DOI: 10.1016/j.ijmst.2017.03.005

[6] Ying H, Zhu C, Shen H, et al.. (2018). Semi-analytical solution for groundwater ingress into lined tunnel. Tunnelling and Underground Space Technology, 76, 43-47.

DOI: 10.1016/j.tust.2018.03.009

[7] Gholizadeh H, Behrouj Peely A, Karney BW, et al.. (2020). Assessment of groundwater ingress to a partially pressurized water-conveyance tunnel using a conduit-flow process model: a case study in Iran. Hydrogeology Journal, 28(7), 2573-2585.

DOI: 10.1007/s10040-020-02213-y

[8] Wang Z, Bi L, Kwon S, et al.. (2020). The effects of hydro-mechanical coupling in fractured rock mass on groundwater inflow into underground openings. Tunnelling and Underground Space Technology, 103, 103489.

DOI: 10.1016/j.tust.2020.103489

[9] Jiang C, Han H, Xie H, et al.. (2021). Karst aquifer water inflow into tunnels: an analytical solution. Geofluids, 2021(1), 6672878.

DOI: 10.1155/2021/6672878

[10] Su Y, Huang Y, Shen H, et al.. (2022). An analytical method for predicting the groundwater inflow to tunnels in a fractured aquifer. Hydrogeology Journal, 30(4), 1279-1293.

DOI: 10.1007/s10040-022-02485-6

[11] Park K, Owatsiriwong A, Lee J-G. (2008). Analytical solution for steady-state groundwater inflow into a drained circular tunnel in a semi-infinite aquifer: A revisit. Tunnelling and Underground Space Technology, 23(2), 206-209

DOI: 10.1016/j.tust.2007.02.004

[12] Su K, Zhou Y, Wu H, et al.. (2017). An Analytical Method for Groundwater Inflow into a Drained Circular Tunnel. Ground water, 55(5), 712-721.

DOI: 10.1111/gwat.12513

[13] Li L, Chen H, Li J, et al.. (2021). A semi-analytical solution to steady-state groundwater inflow into a circular tunnel considering anisotropic permeability. Tunnelling and Underground Space Technology, 116, 104115.

DOI: 10.1016/j.tust.2021.104115

[14] Farhadian H, Katibeh H, Huggenberger P. (2016). Empirical model for estimating groundwater flow into tunnel in discontinuous rock masses. Environmental Earth Sciences, 75(6), 471.

DOI: 10.1007/s12665-016-5332-z

[15] Cheng P, Zhao L, Zhang S, et al.. (2017). Water inflow forecasting for tunnel considering nonlinear variation of permeability coefficient. Journal of Central South University, 24(7), 1612-1618.

DOI: 10.1007/s11771-017-3566-x

[16] Lin P, Li SC, Xu ZH, et al.. (2019). Water Inflow Prediction during Heavy Rain While Tunneling through Karst Fissured Zones. International Journal of Geomechanics, 19(8), 04019093.

DOI: 10.1061/(ASCE)GM.1943-5622.0001478

[17] Wu J, Lu Y, Wu L, et al.. (2023). Numerical Investigation of Water Inflow Characteristics in a Deep-Buried Tunnel Crossing Two Overlapped Intersecting Faults. Water, 15(3), 479.

DOI: 10.3390/w15030479

[18] Frenelus W, Peng H, Zhang J. (2021). Evaluation methods for groundwater inflows into rock tunnels: a state-of-the-art review. International Journal of Hydrology, 5(4), 152-168.

DOI: 10.15406/ijh.2021.05.00277

[19] Zabidi H, Rahim A, Trisugiwo M. (2019). Structural controls on groundwater inflow analysis of hardrock TBM. Cogent Geoscience, 5(1), 1637556.

DOI: 10.1080/23312041.2019.1637556

[20] Park YJ, Hwang HT, Suzuki S, et al.. (2020). Improving precision in regional scale numerical simulations of groundwater flow into underground openings. Engineering Geology, 274, 105727.

DOI: 10.1016/j.enggeo.2020.105727

[21] Bahrami S, Doulati Ardejani F, Baafi E. (2016). Application of artificial neural network coupled with genetic algorithm and simulated annealing to solve groundwater inflow problem to an advancing open pit mine. Journal of Hydrology, 536, 471-484.

DOI: 10.1016/j.jhydrol.2016.03.002

[22] Sedghi M, Zhan H. (2021). On Inflow to a Tunnel in a Fractured Double-Porosity Aquifer. Groundwater, 59(4), 562-570.

DOI: 10.1111/gwat.13079

[23] Kardani N, Bardhan A, Kim D, et al.. (2021). Modelling the energy performance of residential buildings using advanced computational frameworks based on RVM, GMDH, ANFIS-BBO and ANFIS-IPSO. Journal of Building Engineering, 35, 102105.

DOI: 10.1016/j.jobe.2020.102105

[24] Baghbani A, Choudhury T, Costa S, et al.. (2022). Application of artificial intelligence in geotechnical engineering: A state-of-the-art review. Earth-Science Reviews, 228, 103991.

DOI: 10.1016/j.earscirev.2022.103991

[25] Li C, Zhou J, Tao M, et al.. (2022). Developing hybrid ELM-ALO, ELM-LSO and ELM-SOA models for predicting advance rate of TBM. Transportation Geotechnics, 36, 100819.

DOI: 10.1016/j.trgeo.2022.100819

[26] Zhang N, Zhou A, Jin Y, et al.. (2023). An enhanced deep learning method for accurate and robust modelling of soil stress-strain response. Acta Geotechnica, 18(8), 4405-4427.

DOI: 10.1007/s11440-023-01813-8

[27] Zhang P, Wu H, Chen R, et al.. (2020). Hybrid meta-heuristic and machine learning algorithms for tunneling-induced settlement prediction: A comparative study. Tunnelling and Underground Space Technology, 99, 103383.

DOI: 10.1016/j.tust.2020.103383

[28] Tang L, Na S. (2021). Comparison of machine learning methods for ground settlement prediction with different tunneling datasets. Journal of Rock Mechanics and Geotechnical Engineering, 13(6), 1274-1289.

DOI: 10.1016/j.jrmge.2021.08.006

[29] Kim D, Kwon K, Pham K, et al.. (2022). Surface settlement prediction for urban tunneling using machine learning algorithms with Bayesian optimization. Automation in Construction, 140, 104331.

DOI: 10.1016/j.autcon.2022.104331

[30] Liu L, Zhou W, Gutierrez M. (2022). Effectiveness of predicting tunneling-induced ground settlements using machine learning methods with small datasets. Journal of Rock Mechanics and Geotechnical Engineering, 14(4), 1028-1041.

DOI: 10.1016/j.jrmge.2021.08.018

[31] Ye X, Jin T, Chen Y. (2022). Machine learning-based forecasting of soil settlement induced by shield tunneling construction. Tunnelling and Underground Space Technology, 124, 104452.

DOI: 10.1016/j.tust.2022.104452

[32] Zhang D, Guo X, Shen Y, et al.. (2024). Data- and experience-driven neural networks for long-term settlement prediction of tunnel. Tunnelling and Underground Space Technology, 147, 105669.

DOI: 10.1016/j.tust.2024.105669

[33] Yang Z. (2016). Risk prediction of water inrush of karst tunnels based on bp neural network. 2016 4th International Conference on Mechanical Materials and Manufacturing Engineering, Atlantis Press.

DOI: 10.2991/mmme-16.2016.74

[34] Li SC, He P, Li LP, et al. (2017). Gaussian process model of water inflow prediction in tunnel construction and its engineering applications. Tunnelling and Underground Space Technology, 69, 155-161.

DOI: 10.1016/j.tust.2017.06.018

[35] Reddy YR. (2019). An empirical study on the best prediction model of water inflow into drill and blast tunnels among several machine learning techniques. International Journal of Emerging Technologies and Innovative Research, 6(2), 888-891.

[36] Mahmoodzadeh A, Mohammadi M, M Gharrib Noori K, et al.. (2021). Presenting the best prediction model of water inflow into drill and blast tunnels among several machine learning techniques. Automation in Construction, 127, 103719.

DOI: 10.1016/j.autcon.2021.103719

[37] Mahmoodzadeh A, Ghafourian H, Hussein Mohammed A, et al.. (2023). Predicting tunnel water inflow using a machine learning-based solution to improve tunnel construction safety. Transportation Geotechnics, 40, 100978

DOI: 10.1016/j.trgeo.2023.100978

[38] Chen J, Zhou M, Zhang D, et al.. (2021). Quantification of water inflow in rock tunnel faces via convolutional neural network approach. Automation in Construction, 123, 103526.

DOI: 10.1016/j.autcon.2020.103526

[39] Liu D, Xu Q, Tang Y, et al.. (2021). Prediction of Water Inrush in Long-Lasting Shutdown Karst Tunnels Based on the HGWO-SVR Model. IEEE Access, 9, 6368-6378.

DOI: 10.1109/ACCESS.2020.3047626

[40] Zhou J, Zhang Y, Li C, et al.. (2023). Enhancing the performance of tunnel water inflow prediction using Random Forest optimized by Grey Wolf Optimizer. Earth Science Informatics, 16(3), 2405-2420.

DOI: 10.1007/s12145-023-01042-3

[41] Samadi H, Mahmoodzadeh A, Elhag AB, et al.. (2025). Application of hybrid-optimized and stacking-ensemble labeled neural networks to predict water inflow in drill-and-blast tunnels. Tunnelling and Underground Space Technology, 156, 106273.

DOI: 10.1016/j.tust.2024.106273

[42] Kaveh A. (2024). Applications of artificial neural networks and machine learning in civil engineering. (Vol. 1168). Cham, Switzerland: Springer.

[43] Koodiani HK, Jafari E, Majlesi A, et al.. (2024). Machine learning tools to improve nonlinear modeling parameters of RC columns. Journal of Building Engineering, 84, 108492.

DOI: 10.1016/j.jobe.2024.108492

[44] Ekanayake IU, Meddage DPP, Rathnayake U. (2022). A novel approach to explain the black-box nature of machine learning in compressive strength predictions of concrete using Shapley additive explanations (SHAP). Case Studies in Construction Materials, 16, e01059.

DOI: 10.1016/j.cscm.2022.e01059

[45] Wolpert DH, Macready WG. (1997). No free lunch theorems for optimization. IEEE Transactions on Evolutionary Computation, 1(1), 67-82.

DOI: 10.1109/4235.585893

[46] Branco P, Torgo L, Ribeiro RP. (2017). SMOGN: a Pre-processing Approach for Imbalanced Regression. Proceedings of the First International Workshop on Learning with Imbalanced Domains: Theory and Applications. T Paula Branco Luís, Nuno M. Proceedings of Machine Learning Research, PMLR. 74: 36-50.

DOI: 10.1016/j.neucom.2018.11.100

[47] Xie C, Nguyen H, Bui X, et al.. (2021). Predicting roof displacement of roadways in underground coal mines using adaptive neuro-fuzzy inference system optimized by various physics-based optimization algorithms. Journal of Rock Mechanics and Geotechnical Engineering, 13(6), 1452-1465.

DOI: 10.1016/j.jrmge.2021.07.005

[48] Chen Y, Yong W, Li C, et al.. (2023). Predicting the Thickness of an Excavation Damaged Zone around the Roadway Using the DA-RF Hybrid Model. CMES - Computer Modeling in Engineering and Sciences, 136(3), 2507-2526

DOI: 10.32604/cmes.2023.025714

[49] Zhou J, Chen Y, Li C, et al.. (2023). Machine learning models to predict the tunnel wall convergence. Transportation Geotechnics, 41, 101022.

DOI: 10.1016/j.trgeo.2023.101022

[50] Sun Q, Dias D. (2021). Global sensitivity analysis of probabilistic tunnel seismic deformations using sparse polynomial chaos expansions. Soil Dynamics and Earthquake Engineering, 141, 106470.

DOI: 10.1016/j.soildyn.2020.106470

[51] Zhao J, DeJong M. (2023). Three-dimensional probabilistic assessment of tunneling induced structural damage using Monte-Carlo method and hybrid finite element model. Computers and Geotechnics, 154, 105122.

DOI: 10.1016/j.compgeo.2022.105122

[52] Saltelli A, Annoni P, Azzini I, et al.. (2010). Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity index. Computer Physics Communications, 181(2), 259-270.

DOI: 10.1016/j.cpc.2009.09.018

[53] Zhang N, Niu M, Wan F, et al.. (2024). Hazard Prediction of Water Inrush in Water-Rich Tunnels Based on Random Forest Algorithm. Applied Sciences, 14(2), 867.

DOI: 10.3390/app14020867