The prediction of shield tunneling attitude holds significant importance for shield construction. Existing research primarily focuses on the application of deep learning in time-series prediction, while the practical value of traditional machine learning methods and confidence interval prediction in engineering applications is often overlooked. In addition, the determination of the number of modes in modal decomposition remains uncertain, which limits model stability and prediction accuracy. To achieve better prediction of shield attitude, based on the shield construction project of the Jiangyin-Jingjiang Yangtze River Tunnel, SSA (sparrow search algorithm) was applied to jointly optimize the number of modes and iterations of VMD (variational mode decomposition), enabling adaptive and efficient decomposition of non-stationary signals. Furthermore, SSA was utilized to optimize the hyperparameters of SVR (support vector regression), constructing a high-precision time-series prediction model. The results indicate that the test set performance of the SSA-VMD-SVR model achieves MSE=0.484 36, RMSE=0.680 78, and R2=0.969 58, outperforming traditional models such as VMD-BILSTM and EMD-SVR. Moreover, prediction uncertainty is quantified through confidence interval analysis based on kernel density estimation, shifting from “predictable results” to “quantifiable risks”. This study provides a high-precision prediction tool and scientific decision-making support for shield tunnel construction under complex geological conditions.
影响盾构机姿态的因素可分为3大类:隧道几何因素、地质因素、施工参数因素。短距离的隧道轴线可认为是直线,因此不考虑隧道几何因素,地质因素的影响可通过施工参数的侧面影响体现。结合文献中的输入参数选取及现场施工人员建议,选取刀盘转速、刀盘扭矩、推进油缸总推力等57个施工参数作为初步研究参数。为剔除冗余特征,研究采用最大信息系数(maximal information coefficient, MIC)互信息方法进行特征选择,以定量刻画各施工参数与姿态变量之间的统计相关性。与传统相关性指标相比,MIC不仅能够捕捉线性依赖关系,还能识别复杂的非线性关联,因而更适用于盾构施工多源耦合数据的特征筛选。基于图3所示计算结果,选取相互信息值不小于0.2的16个关键特征作为预测模型的输入,以增强姿态预测的准确性与稳健性。
WangShuying, WangLai, PanQiujing, et al. Deep learning model for vertical attitude prediction of shield tunneling machine based on data-driven approach[J]. Journal of Central South University (Science and Technology Edition), 2024, 55(2): 485-497.
[3]
ZhouC, GaoY, ChenE J, et al. Deep learning technologies for shield tunneling: challenges and opportunities[J]. Automation in Construction, 2023, 154: 104982.
DingZhi, DongYuqing, ZhangXiao, et al. Research and prospect of the effect of shield attitude change on segments and control technology[J]. Science Technology and Engineering, 2021, 21(21): 8745-8756.
GaoJunhua, YangGuang, ZhaoSensen, et al. Regularity and prediction of segment floating of shallow buried large diameter shield tunnel in soft soil area: a case study on Shanghai Airport Contact Line[J]. Science Technology and Engineering, 2024, 24(11): 4759-4768.
LiYanchun, DengYongzhong, MaGuoyu, et al. Shield jamming accident of dual-mode shield caused by attitude deflection in composite stratum[J]. Science Technology and Engineering, 2023, 23(11): 4849-4857.
[10]
KongX, TangL, LingX, et al. Development of shield model test system for studying the bias load of shield in soil-rock compound strata[J]. Tunnelling and Underground Space Technology, 2024, 143: 105464.
[11]
ZhengY, WangR, HuZ, et al. Analysis of longitudinal stress and deformation characteristics of shield tunnel under faulting dislocation[J]. Transportation Geotechnics, 2023, 43: 101137.
XiaoMingqing, FengKun, ZhouZiyang, et al. Study on influencing factors of segment dislocation during shield tunneling[J]. Chinese Journal of Geotechnical Engineering, 2023, 45(7): 1347-1356.
HuangXumin, HuangLinzhong, LiangYu. Analysis and application of longitudinal uplift characteristics of shield tunnel segments affected by synchronous grouting during construction[J]. Chinese Journal of Geotechnical Engineering, 2024, 43(9): 1700-1707.
GuanZhenchang, XieLifu, ZhouYuxuan, et al. Research on shield attitude prediction model and control method based on machine learning[J]. Tunnel Construction (Chinese & English), 2024, 44(10): 2032-2040.
CaoHuajin. Shield attitude prediction and optimization based on machine learning and non-dominated sorting genetic algorithm[J]. Railway Construction, 2023, 63(7): 93-97.
[20]
WuX, WangJ, FengZ, et al. Multisource information fusion for real-time prediction and multiobjective optimization of large-diameter slurry shield attitude[J]. Reliability Engineering and System Safety, 2024, 250: 110305.
[21]
ChenH, LiX, FengZ, et al. Shield attitude prediction based on Bayesian-LightGBM machine learning[J]. Information Sciences, 2023, 632: 105-129.
ZhuMeiheng, ChenZhaoqing, ZhangDongmei, et al. Shield tunneling attitude prediction method based on AWPSO-GRU algorithm: a case study of Shanghai Airport Contact Line[J]. Science Technology and Engineering, 2025, 25(14): 6062-6071.
[24]
ZhangW, LiuY, LiJ, et al. Prediction of shield tunnel attitude based on deep learning[J]. Tunnelling and Underground Space Technology, 2023, 115: 104029.
[25]
YaoQ C, AnB H, CaoR S, et al. On the generalization of an attention-based GRU model for shield attitude and position prediction[J]. Journal of Physics: Conference Series, 2024, 2890(1): 012033.
FengTugen, HuJinjian, ZhangJian, et al. Research on attitude prediction of super-large diameter shield based on deep learning[J]. Journal of Central South University (Science and Technology Edition), 2024, 55(4): 1477-1491.
LengWuming, WuZhuolin, YuanLigang, et al. A prediction method for shield attitude of subway tunnel based on EMD-Light-GBM model[J]. Journal of Harbin Institute of Technology, 2025, 57(7): 96-107.
WangHeng, HuJinjian, SunChengguo, et al. Research on grouped prediction of shield attitude based on fixed number of components using EMD-GRU[J]. Modern Tunnelling Technology, 2025, 62(2): 132-140.