Controlling the ground settlement and building deformation triggered by shield tunnelling, particularly within water-rich strata, poses a significant engineering challenge. This study conducts a finite element (FE) analysis focusing on the ground settlement and deformation of adjacent structures (with a minimum distance of 2.6 m to the tunnel) due to earth pressure balance (EPB) shield tunnelling. The analysis incorporates the influence of groundwater through a 3D fluid–solid coupling model. This study assesses the effects of tunnelling on the behaviour of nearby buildings and introduces two mitigation strategies: the vertical partition method and the portal partition method. Their effectiveness is compared and evaluated. Our findings reveal that the deformation curves of the stratum and the building are influenced by the accumulation and dissipation of pore pressure. The vertical partition method reduced surface settlement by approximately 70%, while the portal partition method further minimized building deformation but required careful application to avoid issues like uplift. Both methods effectively mitigate the impacts of tunnel construction, with the portal partition method offering superior performance in terms of material use and cost efficiency. This research provides a scientific foundation and technical guidance for similar engineering endeavours, which is vital for ensuring the safety of metro tunnel construction and the stability of adjacent buildings.

The present investigation presents the field measurement and prediction of tunneling-induced surface ground settlement in Tianjin Metro Line 7, China. The cross-section of a metro tunnel exhibits circular symmetry, thereby making it suitable for tunneling with a circular shield machine. The ground surface may deform during the tunneling stage. In the early stage of tunneling, few measurement data can be collected. To obtain a better usable prediction model, two kinds of neural networks according to the model-agnostic meta-learning (MAML) scheme are presented. One kind of deep learning strategy is a combination of the Back-Propagation Neural Network (BPNN) and the MAML model, named MAML-BPNN. The other prediction model is a mixture of the MAML model and the Long Short-Term Memory (LSTM) model, named MAML-LSTM. Founded on several measurement datasets, the prediction models of the MAML-BPNN and MAML-LSTM are successfully trained. The results show the present models possess good prediction ability for tunneling-induced surface ground settlement. Based on the coefficient of determination, the prediction result using MAML-LSTM is superior to that of MAML-BPNN by 0.1.

Freezing can greatly impact the capacity of tunnel linings in soil, making it essential to investigate the changes in stress state caused by frost heave. Therefore, a finite element model for frost-heave analysis is established for a cold-region tunnel portal. The research investigates the impact of lining thickness and tunnel burial depth on the distribution and trends of axial force (N) and bending moment (M) in the tunnel lining before freezing, after freezing, and after a freeze–thaw cycle. The results show that frost-heave-induced internal forces concentrate at the side wall and arch haunch, identifying these as key vulnerable sections of the portal lining. Generally, the axial force and bending moment of the lining increase after freezing and continue to increase after a freeze–thaw cycle. In addition, a composite anti-freezing insulation structure with both thermal-insulation and anti-freezing buffer functions is proposed, and the influence of its thickness on insulation performance is analyzed. Compared with the case without insulation, the composite layer reduces the maximum frozen-zone thickness in the surrounding rock by up to about 75% and decreases the transverse and longitudinal displacements/convergences of the lining by approximately 80–90%. With increasing thickness, the composite layer also lowers the frost-heave force (F_f−h) on the secondary lining and makes its distribution along the lining more uniform.