Multi-fidelity knowledge inheritance with active querying for data-driven clogging prediction during mechanized tunneling

Abstract

Muck clogging during shield tunneling often leads to reduced construction efficiency, increased costs and potential safety hazards. Traditional methods for predicting muck clogging primarily rely on the operator’s experience and conventional risk maps, but have limitations in dealing with complex construction conditions. To address these issues, this study presents a Monte-Carlo dropout (MCD)-assisted multi-fidelity neural network (MFNN) framework for effective prediction of muck clogging risk. First, a low-fidelity model is trained based on synthesized data using clogging risk maps. Subsequently, in-situ tunneling data are used as high-fidelity data to train multi-fidelity models. MCD serves to evaluate the uncertainty of the MFNN’s inference, combined with an active learning strategy to refine the low-fidelity model via iterative training of the high-fidelity model. Experimental results show that the MCD-assisted MFNN framework captures clogging features more effectively than traditional machine learning models that use only single-fidelity data, especially in scenarios with imbalanced data. This study provides a viable solution for complex problems in shield tunneling by fully utilizing both experiential knowledge accumulated in engineering practice and field monitoring data, demonstrating the potential of integrating knowledge and data in tackling some challenges that were previously unresolved.

Keywords

Machine learning / Muck clogging / Monte-Carlo dropout / Active learning / Multi-fidelity neural networks

Data availability

The data and code used in this study are hosted on the Clear Data Bay platform (https://www.cleardatabay.com/index.php?c=show&id=37).

CRediT authorship contribution statement

Xiao Yuan: Methodology, Investigation, Writing - review & editing, Writing - original draft, Validation, Supervision, Conceptualization. Shuying Wang: Supervision, Project administration, Methodology, Funding acquisition, Conceptualization. Tongming Qu: Visualization, Validation, Software, Investigation, Data curation, Conceptualization. Huanhuan Feng: Resources, Investigation, Validation. Pengfei Liu: Validation, Supervision, Resources. Junhao Zeng: Software, Investigation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgement

This study is supported by the National Natural Science Foundation of China (Grant No. 52022112).

References

[1] Basnet P. M. S., Mahtab S., & Jin A. B. (2023). A comprehensive review of intelligent machine learning based predicting methods in long-term and short-term rock burst prediction. Tunnelling and Underground Space Technology, 142, 105434.

[2] Bozorgzadeh N., Harrison J. P., & Escobar M. D. (2019). Hierarchical Bayesian modelling of geotechnical data: Application to rock strength. Géotechnique, 69(12), 1056-1070.

[3] Cachim P., & Bezuijen A. (2019). Modelling the torque with artificial neural networks on a tunnel boring machine. KSCE Journal of Civil Engineering, 23(10), 4529-4537.

[4] Cai Y. M., Ching J., & Phoon K. K. (2024). Tailored clustering method to identify quasi-regional sites. Engineering Geology, 333, 107490.

[5] Chen R. P., Zhang P., Kang X., Zhong Z. Q., Liu Y., & Wu H. N. (2019). Prediction of maximum surface settlement caused by earth pressure balance (EPB) shield tunneling with ANN methods. Soils and Foundations, 59(2), 284-295.

[6] Chen X. S., Shen J., Bao X. H., Wu X. L., Tang W. C., & Cui H. Z. (2023). A review of seismic resilience of shield tunnels. Tunnelling and Underground Space Technology, 136, 105075.

[7] Conti P., Guo M. W., Manzoni A., & Hesthaven J. S. (2023). Multifidelity surrogate modeling using long short-term memory networks. Computer Methods in Applied Mechanics and Engineering, 404, 115811.

[8] Gal Y., & Ghahramani Z. (2016). Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning. In Proceedings of the 33rd International Conference on Machine Learning (pp. 1050-1059). New York, USA.

[9] He G. F., Zhang P., & Yin Z. Y. (2024). Active learning inspired multifidelity probabilistic modelling of geomaterial property. Computer Methods in Applied Mechanics and Engineering, 432, 117373.

[10] Hou S. K., Liu Y. R., & Yang Q. (2022). Real-time prediction of rock mass classification based on TBM operation big data and stacking technique of ensemble learning. Journal of Rock Mechanics and Geotechnical Engineering, 14(1), 123-143.

[11] Huang H. W., Chang J. Q., Zhang D. M., Zhang J., Wu H. M., & Li G. (2022a). Machine learning-based automatic control of tunneling posture of shield machine. Journal of Rock Mechanics and Geotechnical Engineering, 14(4), 1153-1164.

[12] Huang X., Zhang Q. T., Liu Q. S., Liu X. W., Liu B., Wang J. J., & Yin X. (2022b). A real-time prediction method for tunnel boring machine cutter-head torque using bidirectional long short-term memory networks optimized by multi-algorithm. Journal of Rock Mechanics and Geotechnical Engineering, 14(3), 798-812.

[13] Jiang X., Zhu H. H., Yan Z. G., Zhang F. S., Ye F., Li P. N., Zhang X. H., Dai Z. R., Bai Y., & Huang B. S. (2023). A state-of-art review on development and progress of backfill grouting materials for shield tunneling. Developments in the Built Environment, 16, 100250.

[14] Kong X. X., Ling X. Z., Tang L., Tang W. C., & Zhang Y. F. (2022). Random forest-based predictors for driving forces of earth pressure balance (EPB) shield tunnel boring machine (TBM). Tunnelling and Underground Space Technology, 122, 104373.

[15] Li D. Y., Liu Z. D., Xiao P., Zhou J., & Armaghani D. J. (2022). Intelligent rockburst prediction model with sample category balance using feedforward neural network and Bayesian optimization. Underground Space, 7(5), 833-846.

[16] Li J. B., Chen Z. Y., Li X., Jing L. J., Zhang Y. P., Xiao H. H., Wang S. J., Yang W. K., Wu L. J., Li P. Y., Li H. B., Yao M., & Fan L. T. (2023). Feedback on a shared big dataset for intelligent TBM Part I: Feature extraction and machine learning methods. Underground Space, 11, 1-25.

[17] Li J. H., Guo D., Chen Z. Y., Li X., & Li Z. F. (2024). Transfer learning for collapse warning in TBM tunneling using databases in China. Computers and Geotechnics, 166, 105968.

[18] Liu D. J., Zhang W. P., Dai Q. Q., Chen J. Y., Duan K., & Li M. Y. (2024). Safety evaluation method for operational shield tunnels based on semi-supervised learning and a stacking algorithm. Tunnelling and Underground Space Technology, 153, 106027.

[19] Luo H., Fang Y., Wang J. F., Wang Y. B., Liao H., Yu T., & Yao Z. G. (2023). Combined prediction of rockburst based on multiple factors and stacking ensemble algorithm. Underground Space, 13, 241-261.

[20] Qu T. M., Guan S. H., Feng Y. T., Ma G., Zhou W., & Zhao J. D. (2023a). Deep active learning for constitutive modelling of granular materials: From representative volume elements to implicit finite element modelling. International Journal of Plasticity, 164, 103576.

[21] Qu T. M., Zhao J. D., Guan S. H., & Feng Y. T. (2023b). Data-driven multiscale modelling of granular materials via knowledge transfer and sharing. International Journal of Plasticity, 171, 103786.

[22] Rezvani S., & Wang X. Z. (2023). A broad review on class imbalance learning techniques. Applied Soft Computing, 143, 110415.

[23] Su M. M., Guo N., & Yang Z. X. (2023). A multifidelity neural network (MFNN) for constitutive modeling of complex soil behaviors. International Journal for Numerical and Analytical Methods in Geomechanics, 47(18), 3269-3289.

[24] Thewes M., & Hollmann F. (2016). Assessment of clay soils and clay-rich rock for clogging of TBMs. Tunnelling and Underground Space Technology, 57, 122-128.

[25] Tran N., Chen H. H., Bhuyan J., & Ding J. H. (2022). Data curation and quality evaluation for machine learning-based cyber intrusion detection. IEEE Access, 10, 121900-121923.

[26] Wang L., Pan Q. J., & Wang S. Y. (2024). Data-driven predictions of shield attitudes using Bayesian machine learning. Computers and Geotechnics, 166, 106002.

[27] Wang S. Y., Liu P. F., Hu Q. X., & Zhong J. Z. (2020). Effect of dispersant on the tangential adhesion strength between clay and metal for EPB shield tunnelling. Tunnelling and Underground Space Technology, 95, 103144.

[28] Wang S. Y., Liu T. Y., Zheng X. C., Yang J. S., & Yang F. (2024a). Dynamic collapse characteristics of the tunnel face induced by the shutdown of earth pressure balance shields (EPB): A 3D material point method study. Underground Space, 16, 164-182.

[29] Wang S. Y., Ni Z. L., Qu T. M., Wang H. B., & Pan Q. J. (2022). A novel index to evaluate the workability of conditioned coarse-grained soil for EPB shield tunnelling. Journal of Construction Engineering and Management, 148(6), 04022028.

[30] Wang S. Y., Yuan X., & Qu T. M. (2024b). Machine learning-informed soil conditioning for mechanized shield tunneling. Computer-Aided Civil and Infrastructure Engineering, 39(17), 2662-2682.

[31] Wang S. Y., Zhu H. B., Liu P. F., & Qu T. M. (2024c). Soil slaking under the effect of dispersants: Characteristics and mechanism. Environmental Earth Sciences, 83, 397.

[32] Xu Q. H., Huang X., Zhang B. G., Zhang Z. X., Wang J. H., & Wang S. F. (2023). TBM performance prediction using LSTM-based hybrid neural network model: Case study of Baimang River tunnel project in Shenzhen, China. Underground Space, 11, 130-152.

[33] Yuan X., Wang S. Y., Qu T. M., Feng H. H., Liu P. F., Zeng J. H., & Chen X. S. (2024). Learning the hard-to-learn: Active learning for imbalanced datasets in data-centric tunnel engineering. Computers and Geotechnics, 174, 106629.

[34] Zendaki Y., Cao B. T., Alsahly A., Freitag S., & Meschke G. (2024). A simulation-based software to support the real-time operational parameters selection of tunnel boring machines. Underground Space, 14, 176-196.

[35] Zeng R. K., Wang S. Y., Zhang Y., & Qu T. M. (2025). CFD-DEM modeling of seepage in foam-conditioned soil. Computers and Geotechnics, 177, 106818.

[36] Zhang D. M., Fu L., Huang H. W., Wu H. M., & Li G. (2023a). Deep learning-based automatic detection of muck types for earth pressure balance shield tunneling in soft ground. Computer-Aided Civil and Infrastructure Engineering, 38(7), 940-955.

[37] Zhang P., Sheil B., & Cheng Q. (2024a). Multi-fidelity learned emulator for waves and porous coastal structures interaction modelling. Computers and Geotechnics, 176, 106718.

[38] Zhang P., Yin Z. Y., & Sheil B. (2024b). Multifidelity constitutive modeling of stress-induced anisotropic behavior of clay. Journal of Geotechnical and Geoenvironmental Engineering, 150(3), 04024003.

[39] Zhang P., Yin Z. Y., Jin Y. F., Yang J., & Sheil B. (2022). Physicsinformed multifidelity residual neural networks for hydromechanical modeling of granular soils and foundation considering internal erosion. Journal of Engineering Mechanics, 148(4), 04022015.

[40] Zhang W. S., Liu X. Y., Zhang L. Y., & Wang Y. D. (2023b). Intelligent real-time prediction of multi-region thrust of EPB shield machine based on SSA-LSTM. Engineering Research Express, 5(3), 035013.