In the long-distance shield tunneling process, the shield tail brush needs to be replaced after wear failure, especially in cross-river and cross-sea tunnel projects. Taking the shield tail brush replacement project of Qingchun Road tunnel in Hangzhou, China as the research background, a three-dimensional sequential thermo-mechanical coupled numerical simulation was carried out to study the shield tail brush replacement with liquid nitrogen freezing by ABAQUS software. The development of temperature fields and displacement fields of soil, and the deformations of the tunnel sections along the longitudinal direction induced by freezing were obtained. The numerical model was further validated by site measurement. In addition, the effects of the frozen wall thickness, buried depth, excavation radius, and coefficients of linear thermal expansion of soils on the frost heave were studied.
The hydraulic thrust system of the shield machine is used for driving forward and pose adjustment. It is of great significance to figure out the thrust mechanism and motion characteristics of the shield machine to ensure the safety of tunnel excavation. This study aims to develop a model to explore the influence of thrust cylinder advancement on the motions of the shield machine. Firstly, the study carried out mechanism analysis of the thrust mechanism for the shield machine and established a method to describe the position and attitude of the shield machine during the tunneling process by the homogeneous transformation matrix. Then, a new inverse kinematic model was proposed to quantify the relationship between the telescoping movements of cylinders and shield machine motions, and the Jacobian matrix was derived to solve the instantaneous kinematics analysis. Furthermore, a virtual prototype model was developed to simulate the kinematic behavior of the shield machine and validate the accuracy of the kinematic model. The model provides the basic constraint relations for the practical position control system and lays a strong foundation for the dynamic model and automatic trajectory tracking control of shield machines for future studies. Based on the proposed model, the displacement, velocity, and acceleration of cylinders that drive the shield machine to the target motions can be solved exactly. It can provide a reference for the pose control of the shield machine during the practical shield tunneling.
Accurate prediction of shield tunneling attitude is crucial for preventing engineering risks such as tunnel misalignment, structural collisions, and ground subsidence in underground construction projects. This study develops a real-time prediction method for shield tunneling parameters using the Qingdao Jiaozhou Bay Second Subsea Tunnel as a case study. We systematically compare three machine learning approaches (XGBoost, Transformer, and LSTM) enhanced with discrete wavelet transform (DWT) for noise reduction and Bayesian optimization (BOA) for hyperparameter tuning. Results show that BOA-XGBoost achieved the R2 is 0.98 with the Mean Absolute Percentage Error (MAPE) errors below 2% for all five output parameters (horizontal/vertical deviations at shield head/tail and roll angle). While BOA-XGBoost outperformed other models in accuracy, BOA-LSTM demonstrated superior computational efficiency—33.44% faster than XGBoost and 96.79% faster than Transformer. Under complex geological conditions, deep learning models (BOA-LSTM and BOA-Transformer) showed better adaptability than BOA-XGBoost, particularly in mixed soil-rock strata. This study provides practical guidance for selecting appropriate prediction models based on geological conditions and computational constraints in tunnel engineering projects.
1 November 2024 in Journal of Central South University
The Discrete Element Method (DEM) is a commonly used numerical analysis tool for simulating the mechanical behavior of clay materials. DEM models require the specification of microscopic virtual parameters that are not measurable and are not directly related to the macroscopic characteristics of the material. Therefore, calibration of these microscopic virtual parameters is necessary to simulate the mechanical behavior of clay. Calibration typically uses an iterative trial-and-error method, which involves running multiple DEM simulations, making the calibration process computationally expensive. In this work, we propose a calibration method based on Response Surface Methodology for designing virtual direct shear and adhesion shear experiments. This method significantly reduces the number of DEM simulations required during the calibration process, thus lowering the associated computational cost. Using this approach, the shear stress–strain curve for clay materials was obtained, and a comparison with experimental data shows that the response surface model generated by this method demonstrates high accuracy and reliability. The application of this method to simulate the shear and adhesion behavior of clay represents a novel approach. Validation was performed through a case study of a tunnel boring machine operating in clay, and the predicted results aligned well with the measured data, confirming the applicability of the method. This approach provides effective theoretical support for studying the interaction between clay and machinery.