With the development of urban metro systems, shield tunnels that pass through existing bridge pile foundations have become an inevitable engineering problem. Therefore, ensuring the stability of the strata and existing bridge piles during tunnel construction is a common goal in engineering practice. Currently, research on the mechanical responses of strata and existing piles under different conditions of upper-soft and lower-hard composite strata during shield tunneling has not been conducted extensively. This paper presents a numerical simulation of a shield tunnel passing through an existing bridge pile foundation in upper-soft and lower-hard composite strata. Subsequently, the surface subsidence and mechanical responses of a single pile were analyzed and evaluated. Additional stresses generated in the soil by existing bridge piles and the selection of grouting pressure were considered to optimize the driving pressure of the slurry shield. Allowable values were proposed to evaluate the construction disturbances caused by the tunnel excavation. The results show that the disturbance to the soil and existing piles is significantly influenced by the hard-rock height ratio, and the surface subsidence increases when the hard-rock height ratio decreases. The displacement and internal force of a single pile are significantly influenced by the load applied to the pile. This study demonstrates the changes in the mechanical responses of a single pile during shield tunnel boring, and provides in-depth insights into the deformation control caused by shield underpassing structures in upper-soft and lower-hard composite strata.

The blockage or failure of the drainage holes will endanger the stability of the slopes and traffic safety of a highway tunnel. This paper studies an algorithm for the automatic classification of drainage hole blockage degree based on convolutional neural network transfer learning to explore the intelligent detection method of drainage hole blockage. The model transfer method is adopted to input drainage hole image samples to retrain the pretrained network to classify new images. Experiments are performed on the collected samples of drainage hole images, and the accuracy of different network models is compared, ResNet-18 being the best. The ResNet-18 performance is compared using different transfer strategies and parameters. The results show that when the SGDM gradient optimisation algorithm is used and the learning rate is 0.0001, the identification effect of these samples is the best. The validation accuracy can reach 91.7%, test accuracy is 90.0%, and the effective classification of drainage hole blockage to different degrees is realised under the transfer learning strategy of ResNet-18 model 1–34 frozen layers. Furthermore, with an expansion of the samples in the future, the identification accuracy will be further improved. The automatic classification system of the blockage degree of drainage hole greatly reduces the cost of manual detection, plays a guiding role in the maintenance of drainage pipes, and effectively improves the safety of highway tunnels and slopes.