To reveal the influence mechanism of geometric parameters of structural planes on the damage evolution of surrounding rock and the quality of section formation during tunnel blasting excavation, taking the Xiamen Haicang Expressway Tunnel project as a prototype, a three-dimensional numerical model was established using LS-DYNA to conduct numerical experiments. The influence of different structural plane dip angles, widths, and spacing on the distribution of blasting-induced damage in the tunnel surrounding rock was investigated. The results indicated that structural planes exerted a pronounced directional control effect on blasting damage. Damage preferentially extended along the dip direction of the structural planes, while being significantly constrained in the normal direction. As the dip angle increased from 0° to 90°, the cross-sectional damage pattern transitioned from approximately symmetrical to an asymmetrical expansion dominated by directional guidance, exhibiting localized and discrete characteristics due to segmentation and blocking effects. When the structural plane width increased from 0.1 m to 0.7 m, the interface weakening effect and stress wave scattering energy dissipation were significantly enhanced, leading to a notable increase in overbreak area and maximum overbreak. Conversely, as the structural plane spacing increased from 0.6 m to 2.4 m, overbreak decreased and contour smoothness improved. The research findings provided a basis for understanding energy control, parameter optimization, and the improvement of formation quality in tunnel blasting.

To accurately evaluate the safety state of tunnel lining cracked sections, optimize structural design, and determine maintenance timing scientifically, a stability evaluation system was established in this study by integrating fracture mechanics with the ultimate bearing capacity theory. Numerical simulations were employed to systematically calculate the safety coefficients for three key lining parts—namely, the vault, haunch, and side wall—under varying crack depths. A dual-index system, comprising the stability coefficient P (indicating crack stability)and the bearing capacity safety coefficient K (assessing whether the sectional bearing capacity meets requirements), was introduced to analyze the influence mechanisms of various factors on structural safety. The results showed that the post cracking stability of the lining was primarily governed by two factors: first, whether the cracks continued to propagate, and second, the weakening of bearing capacity resulting from the reduction in effective section thickness due to cracking. Furthermore, cracking in one location was found to have a negligible impact on the bearing capacity safety coefficient of other locations, confirming the independence of local crack assessment. Critical crack depths and angles corresponding to different damage grades were also identified. This study provides a quantitative basis for the safety evaluation of tunnel lining cracks, offering valuable references for optimizing structural design theory and scientifically planning maintenance schedules.

The improvement effect of fiber reinforced cementitious matrix(FRCM)on the bearing and deformation performance of shield tunnels was investigated in this paper. The FRCM, which consisted of engineered cementitious composites(ECC)with internal polyvinyl alcohol(PVA)fibers and basalt fiber-reinforced polymer(BFRP)grids, was adopted to strengthen the segment joints of shield tunnels. Based on the test results of the full-scale bearing capacity model of shield tunnels, the rationality of the three-dimensional(3D)finite element analysis model for shield tunnels was verified from the perspectives of the failure modes and load-deformation curves. The improvement of FRCM on the overall bearing performance of the tunnel was first analyzed using the developed 3D elastoplastic constitutive model of PVA-ECC by the authors. The influence of sensitive parameters, including the strengthened location, strengthened area, number of fiber mesh layers, and thickness of the ECC, on the improvement of the load-carrying capacity was clarified from the perspectives of damage mode, overall structural deformation, and joint opening. Furthermore, the optimal parameters of FRCM for strengthening the shield tunnel segments were determined by comprehensively considering the tunnel damage, load-carrying capacity, and economic considerations. The findings from this study are of great significance for improving the load-carrying capacity of shield tunnels by using FRCM.

During the launching and receiving phases of shield tunneling, as well as during the traversal of building pile foundations, the shield cutter is required to cut through heterogeneous composite materials like reinforced concrete. This process often leads to technical challenges including cutterhead chipping, accelerated wear, and steel bar-induced clogging of the screw conveyor, which ultimately compromises tunneling efficiency. To explore the influence of internal steel bars on the dynamic failure process of concrete structures and the cutting force of the cutterhead during the cutting process, a coupling method combining the discrete element-based particle flow program and continuum mechanics was adopted to study the dynamic failure process of the cutterhead cutting reinforced concrete. Firstly, the contact parameters for the simulated steel bars and concrete materials were calibrated, and a coupling simulation analysis model of the cutterhead cutting reinforced concrete was established. The variation laws of the cutting force of the C-type(single-sided blade)and the F-type(double-sided blade)were studied under different cutting depths and cutter spacings when cutting concrete and reinforced concrete respectively. The results showed that when the F-type tool cuts concrete, the relationship between the cutting depth and the cutting force could be fitted using the arctangent function, and the cutting force was less affected by the tool spacing; when cutting reinforced concrete, the average cutting force was approximately 5 to 7 times higher than that for plain concrete. The F-type tool exhibited a stronger pulling effect on the reinforcing bars, but the average cutting force of the C-type tool was approximately 21% to 39% higher than that of the F-type tool.

Multi-arch tunnels without middle drift were widely applied in regions with complex terrain and unfavorable geological conditions. However, the mechanisms of structural stress imbalance and mutual disturbance between the two caves during construction remained insufficiently understood. Based on Xiazhai Tunnel Project of Xuanhui Expressway in Yunnan, numerical simulations validated by field monitoring data were employed to analyze the mechanical response throughout the construction process of a multi-arch tunnel without a middle drift under different burial depths and bias pressures. The results indicated that the right shoulder, middle wall and right arch foot of the first tunnel were identified as dangerous areas, and the displacement of the first tunnel was significantly impacted by the construction of the following tunnel. When the bias angle was 25°, the vertical displacement of the right shoulder of the first tunnel caused by the upper step excavation of the following tunnel reached 25.9%. The horizontal convergence of the lower step line of the first tunnel was found to be significantly higher than that of the upper step line, and the middle wall and the right arch foot were identified as key stress concentration areas. Increasing burial depth was shown to enhance the self-weight stress of surrounding rock and thus increase the absolute deformation of the structure, while the strengthened confinement reduced the relative contribution of disturbance induced by following-tunnel excavation. Under increasing bias angle, the displacement on both sides of the bench line exhibited a left-leaning tendency, and the structural load-transfer path was altered. The initial support and secondary lining were most unfavorable under the 13° and 25° bias conditions respectively. The results demonstrated that when the buried depth exceeded 40 m or the bias was greater than 13°, the support of the middle wall and arch foot should be strengthened. This research provided a quantitative basis and engineering reference for stability control of multi-arch tunnels without a middle drift under complex geological conditions.

Frequent rockbursts and stress-induced collapses during the construction of deep-buried TBM tunnels under high in-situ stress were investigated through failure characterization, initial stress-field inversion, and graded rockburst control. Based on field investigations, 139 cases of high-stress damage were collected. The results showed that rockburst craters and collapse cavities were jointly controlled by high in-situ stress, structural planes, and surrounding rock properties, whereas their boundary morphology was mainly determined by the number, scale, and spatial combination of structural planes. A three-dimensional geological model considering surface topography and fault-bedding structures was established. Constrained by in-situ stress data from eight hydraulic-fracturing measurement points, five boundary conditions, including self-weight, horizontal extrusion, and shear, were applied, and the regional three-dimensional initial stress field was inverted using multiple linear regression. Good agreement was obtained between the inverted and measured results, with a multiple correlation coefficient of 0.91. Stress distribution along the tunnel axis was extracted, and zones of abrupt stress increase near faults were identified for risk classification. For a typical section with a maximum burial depth of about 1 687 m, significant stress concentration was observed near the tunnel face, while post-excavation stress release and time-dependent redistribution were identified. Timely support after the surrounding rock exited the shield was shown to effectively reduce stress release and rockburst risk. Finally, a construction strategy involving advance prediction, graded control, dynamic monitoring, and safety protection was proposed, together with corresponding measures for slight, moderate, and intense rockbursts.

The construction of closely-spaced tunnels, characterized by the limited clearance between adjacent drifts, posed significant challenges as the excavation of the subsequent tunnel could induce considerable disturbances that jeopardized the structural integrity of the pilot tunnel. Quantifying the resilience of the pilot tunnel-defined as its capacity to withstand construction-induced disturbances from the subsequent tunnel and to recover its functionality thereafter-was therefore crucial for ensuring safety during the construction of closely-spaced tunnels. This study investigated the Hankou Spiral Tunnels, a closely-spaced tunnel complex within the Xinjin Expressway project. Finite element models for nine distinct scenarios, encompassing three varying burial depths and three clear distances in a Grade V rock mass, were developed using the ABAQUS platform. The validity of the numerical model was confirmed through comparison with field monitoring data, specifically tunnel crown settlement and convergence deformation. Analysis revealed that crown settlement induced by subsequent tunnel excavation led to more pronounced functional degradation of the pilot tunnel compared to horizontal convergence, establishing it as the more appropriate indicator for resilience quantification. Based on the identified resilience evaluation indicators and structural performance evolution patterns, both resistance resilience index and recovery resilience index were calculated for all working conditions. The results demonstrated that increasing the pillar width from 10 m to 30 m under constant burial depth conditions enhanced the resistance and recovery resilience indices by 1.0%-9.0%, while reducing the burial depth from 150 m to 50 m under constant pillar width conditions improved these indices by 4.0%-12.0%. Generally, larger pillar widths and shallower burial depths resulted in higher resilience indices, with only the scenario combining 150 m burial depth and 10 m pillar width classified as medium resilience level, while all other configurations achieved high resilience level. This research provided a quantitative assessment framework for resilience-oriented design and construction of closely-spaced tunnels, offering valuable references for similar underground engineering projects.

In surface settlement prediction induced by shield tunneling, traditional empirical formulas and conventional single machine learning models often fell short in adequately capturing the nonlinear spatiotemporal characteristics embedded in multi-source parameters. To address this challenge, a deep learning-based predictive model was proposed in this research for effectively representing the dynamic interdependencies of stratum disturbance responses under complex geological conditions. Specifically, a hybrid VMD-CNN-BiLSTM-Attention framework was constructed by synergistically integrating variational mode decomposition(VMD), convolutional neural network(CNN), bidirectional long short-term memory(BiLSTM), and the attention mechanism. Within the proposed framework, VMD was first applied to decompose the raw settlement time series into multiple intrinsic mode functions, thereby facilitating noise reduction and latent pattern extraction. Subsequently, CNN was employed to extract spatial correlation features from both shield operational parameters(e.g., thrust, torque)and geotechnical properties(e.g., elastic modulus, cohesion, permeability coefficient). The high-level feature representations were then processed by the BiLSTM module to model bidirectional long-term temporal dependencies, while the attention mechanism was utilized to dynamically assign adaptive weights to salient time steps, thus enhancing both predictive accuracy and model interpretability. The model was trained and validated using field-monitored surface settlement data obtained during the shield-driven construction of the eastern extension of Changsha Metro Line 6, where the tunnel was successfully underpassed a critical airport runway. Comparative evaluation against in-situ measurements demonstrated that the proposed model achieved a low root mean square error(E_RMS)on the test set and a high coefficient of determination(R²>0.96), which significantly outperformed conventional approaches in prediction accuracy. The results confirmed that the VMD-CNN-BiLSTM-Attention framework provided a robust and reliable tool for real-time settlement forecasting, thereby offering strong theoretical support for settlement control and early risk warning in shield tunneling projects.

To enhance the precision and efficiency of mechanized tunneling in TBM projects, this research investigated a boreability classification method based on tunneling parameters and muck characteristics. Based on a TBM construction section of a tunnel in western Sichuan, tunneling parameters and muck samples were employed as the primary research objects. The time-series data of TBM tunneling parameters were preprocessed using techniques including redundant data elimination, change-point detection and conditional filtering, outlier removal based on boxplot analysis, and smoothing filters. The average total thrust, average cutterhead torque, and average penetration rate per excavation cycle were selected as clustering variables. By employing the silhouette coefficient, BIC(Bayesian information criterion, B_IC)and the elbow method, the optimal number of clusters was determined to be four. Subsequently, a Gaussian mixture model(GMM)unsupervised clustering algorithm was applied to classify the tunneling cycles. A boreability classification method was then established by integrating the clustering results with muck morphology characteristics and field geological surveys. The clustering results were validated through subsampling validation and external validation, both of which confirmed a high level of consistency. These findings demonstrated the robustness of the clustering outcomes and the effectiveness of the proposed boreability classification method. This research provides a data-driven method for TBM boreability assessment, offering enhanced guidance for design and improved adaptability in construction.