In order to solve the internal contradiction between tunnel deformation control and tunneling efficiency improvement when the double shield tunnel boring machine operate under complex geological conditions, this study proposes a data-driven optimization model to optimize TBM tunneling parameters. The model integrates CatBoost, XGBoost and sequential least squares quadratic programming (SLSQP) algorithm. CatBoost servers as the core algorithm to accurately predict tunnel deformation. XGBoost captures the nonlinear mapping relationship between TBM parameters, and the SLSQP algorithm optimizes tunneling parameters under constraints. The proposed mode was developed and validated using field data of a mountain rail transit project in Sichuan, China. The results show that the R2 of the CatBoost deformation prediction model is 0.923, the MSE is 0.023, and the prediction result is better. After optimization, the average propulsion speed is increased by 17.6% compared with that before optimization, the predicted deformation value is basically maintained near the original level, and the change rate is controlled below 3.45%. The model significantly improves the propulsion efficiency under the premise that the tunnel deformation does not exceed the safety threshold. The research conclusions provide reference for TBM construction decision-making.
WangShuhong, YangRunsheng, DongFurui, et al. Prediction of settlement of shield tunnel subway tunnel roof based on WOA-BP neural network[J].Journal of Basic Science and Engineering, 2025, 33(3): 770-778.
[3]
ChenH Y, LiuJ, ShenG Q, et al. Control of existing tunnel deformation caused by shield adjacent undercrossing construction using interpretable machine learning and multiobjective optimization[J]. Automation in Construction, 2025, 170: 105943.
WangPing'an. Deformation prediction model for soft rock tunnels based on IAO-SVM[J]. Railway Construction Technology, 2024(8): 93-96, 144.
[6]
LiuW L, LiA, LiuC J. Multi-objective optimization control for tunnel boring machine performance improvement under uncertainty[J]. Automation in Construction, 2022, 139: 104310.
WengXiaoin, LiShixin, YangXiaodong, et al. Comprehensive evaluation of flat rock tunnel bottom deformation risk based on optimal weighting-extension cloud model[J]. Journal of Liaoning Technical University (Natural Science), 2023, 42(6): 656-663.
GuoXinxin, WangBo, WangZhenyu, et al. Methods and practices for deformation prediction in high-stress soft rock tunnels considering creep characteristics[J]. Chinese Journal of Geotechnical Engineering, 2023, 45(3): 652-660.
LiJinkui, ZhuangWen. Analysis of deformation of power tunnel passing through pile foundation and underneath at the same time[J]. Journal of Liaoning Technical University (Natural Science), 2023, 42(6): 664-670.
SongYang, ChangYongtao, LiYongqi, et al. Analysis of surrounding rock deformation and slope stability of tunnel section in small clear distance tunnel[J]. Journal of Liaoning Technical University (Natural Science), 2020, 39(1): 24-33.
WangTianning, WangLining, XueYadong, et al. Wireless sensing and prediction method for convergence deformation of mountain tunnels[J]. Chinese Journal of Geotechnical Engineering, 2020, 42(): 224-228.
ZengPeng, ZhangZhiqiang, LiTianbin, et al. Dynamic selection of optimal tunnel convergence prediction model for a probabilistic deformation prediction[J]. Bulletin of Geological Science and Technology, 2024, 43(6): 15-25.
PengJishen, HaoMing, SongLiye, et al. TBM tunneling speed prediction based on TSSA-SVR algorithm[J]. Journal of Liaoning Technical University (Natural Science), 2023, 42(5): 634-640.
[21]
ZhangX M, WangM H, DanF, et al. Risk classification assessment and early warning of large deformation of soft rock in tunnels based on CNN-LSTM model[J]. Scientific Reports, 2024, 14: 29944.
ChenLibo, ZhangMingshu, ChenHaiyong, et al. Analysis and optimal control of shield tunnel construction parameters using categorical boosting-nondominated sorting genetic algorithm-Ⅲ[J]. Tunnel Construction, 2024, 44(8): 1587-1598.
FuLei, WuHuiming, HuangHongwei, et al. Settlement prediction and construction parameters optimization induced by shield tunneling based on random forest and multi-objective particle swarm optimization algorithm-crowding distance[J]. Tunnel Construction, 2025, 45(): 318-329.
[28]
BoY, GuoX G, LiuQ S, et al. Prediction of tunnel deformation using PSO variant integrated with XGBoost and its TBM jamming application[J]. Tunnelling and Underground Space Technology, 2024, 150: 105842.
FeiRuizhen, ShiChenghua, ZhangChunlei, et al. Prediction model for disturbance induced by shield tunneling underneath existing tunnels and optimization of tunneling parameters[J]. Chinese Journal of Geotechnical Engineering, 2026, 48(3): 590-598.
RenXudong, ZhangFengkai, DingWantao, et al. Prediction of ground settlement and optimization of tunneling parameters of slurry shield based on neural network[J]. Science Technology and Engineering, 2025, 25(5): 2090-2099.
[33]
ProkhorenkovaL, GusevG, VorobevA, et al. CatBoost: unbiased boosting with categorical features[J]. Advances in Neural Information Processing Systems,2018,31:6639-6649.
[34]
ChenT Q, GuestrinC. XGBoost: a scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco California USA. ACM, 2016: 785-794.