Rockburst is an engineering phenomenon characterized by the release of elastic strain energy due to the dynamic failure of deep surrounding rock. The existing rockburst proneness indexes primarily focus on rock materials, failing to encompass rock mass quality and engineering excavation disturbance. On the basis of the potential elastic strain energy released by rock failure, five kinds of rockburst proneness indexes of surrounding rock are established considering the rock mass quality and excavation disturbance factor. Firstly, the linear relationship between elastic modulus and residual elastic energy of rock materials (A_EF), the relationships between elastic and deformation moduli, as well as the link with rock mass quality evaluation indexes (i.e., rock mass rating (RMR), basic quality index of rock mass (BQ), and geological strength index (GSI)) and deformation modulus, were used to derive five assessment model of rockburst proneness for surrounding rock. Secondly, the rockburst proneness degree for three grades of surrounding rock (I: excellent rock, II: good rock, and III: fair rock) was assessed utilizing the RMR₈₉, BQ, and GSI indices, and the influence of excavation disturbances on the residual elastic energy of surrounding rock () was analysed. In general, the higher the quality of rock mass and the lesser the disturbance factor, the greater the rockburst proneness degree of surrounding rock. The accuracy of proposed rockburst proneness indexes was validated by using the field data from 27 rockburst cases. The results demonstrate that the discriminant grade of rockburst index based on GSI is basically consistent with the actual occurrence grade of rockburst cases, with an accuracy of 93%, which can be used as a recommended method for evaluating the rockburst proneness degree of surrounding rock. Finally, the shortcomings of rockburst proneness assessment model are discussed, and the improvement direction is elucidated.

During the construction of segmental tunnels, unexpected leakage poses a significant safety hazard to the tunnel structures, potentially leading to collapse. Worldwide, accidents caused by leakage during the construction of shield tunnels have resulted in substantial losses. However, existing studies have not clearly elucidated the mechanism behind tunnel collapse induced by leakage, making it challenging to propose effective prevention or control measures. To address this issue, a series of model tests on tunnel collapse induced by leakage were designed and conducted. These tests replicated the tunnel collapse process and revealed three stages: seepage erosion, soil cave formation and destabilization, and soil impact. The soil caves develop upward, leading to a redistribution of external pressure on the tunnels. Ultimately, the structural collapse of the tunnel occurs due to soil impact from the unstable soil cave. Comparing tunnel entrance/exit accidents with connecting passage accidents highlights that both accident types share the same underlying mechanism but differ in boundary conditions.

In recent decades, there have been numerous reports of damage cases involving tunnels crossing active faults. The mechanical response and failure mechanisms of cross-fault tunnels have become a key issue in the field of tunnel engineering. This study established a continuum-discrete coupling model comprising intact rock mass, fault zones, and tunnel. In this model, the tunnel and intact rock are modeled as continuous media, while the fault zone is modeled as a discrete medium. The non-uniform fault displacement is adopted to simulate the mechanical response and damage patterns of tunnels crossing active faults under reverse faulting. The simulation results are validated by comparison with the damage of Longchi tunnel observed from 2008 Wenchuan earthquake in China, as well as the experimental phenomenon from the model test. The results demonstrate that the proposed coupling model effectively reproduces the tunnel failure modes caused by reverse faulting. In addition, the high consistency between the simulation results and experimental data further confirms computational accuracy and reliability of the coupling model. A parametric analysis based on the Xianglushan tunnel in China is carried out to investigate the effects of fault displacements, fault widths, dip angles and fault zone rock mass qualities on damage patterns of crossing-fault tunnels. This study provides a valuable reference for seismic fortification of the tunnel crossing reverse faults.

This study aims to develop and evaluate a natural gradient boosting (NGBoost) model optimized with Optuna for estimating ground settlement during tunnel excavation, incorporating Shapley additive explanations (SHAP) to perform interpretability analysis on the model’s estimation results. The model’s predictive performance was comprehensively assessed using datasets from two earth pressure balance shield tunneling projects in Changsha and Zhengzhou, China. Comparative analyses demonstrated the superior accuracy and generalization capability of the Optuna-NGBoost-SHAP model (training set: R² = 0.9984, MAE = 0.1004, RMSE = 0.4193, MedAE = 0.0122; validation set: R² = 0.9001, MAE = 1.3363, RMSE = 3.2992, MedAE = 0.3042; test set: R² = 0.9361, MAE = 0.9961, RMSE = 2.5388, MedAE = 0.2147). SHAP value analysis quantitatively evaluated the contributions of input features to the model’s estimations, identifying geometric factors (distance from the shield machine to the monitoring section and cover depth) as the most important features. The findings provide robust decision support for safety management during tunnel construction and demonstrate the reliability and efficiency of the Optuna-NGBoost-SHAP framework in estimating complex ground settlement scenarios.

A reasonable seismic capacity model is crucial for establishing the seismic performance level system and evaluating the seismic reliability of subway station structures. However, the deterministic structural and geotechnical mechanical parameters are usually applied to calibrate the seismic performance levels of subway station structures in the traditional seismic capacity analysis, which ignores the stochasticity of the soil-subway station interaction system. To overcome the challenge caused by the stochastic interaction system, the probability space partition method and stochastic pushover analysis method are combined to develop a calibration strategy of seismic performance levels considering the complete probabilistic information of the stochastic interaction system, and the non-parametric probabilistic seismic capacity models of the subway station structure are further established based on the principle of probability conservation in this paper. A subway station is also taken as the prototype to investigate the applicability of the proposed strategy and the influence of system randomness on the seismic capacity of the subway station structure. The results demonstrate that the seismic performance levels calibrated according to the proposed strategy can effectively consider the complete probabilistic information of the interaction system, which are more rigorous than the existing performance levels. Meanwhile, the probability density evolution of the bearing capacity of the subway station structure is essentially a non-stationary stochastic process, and the non-parametric probability density curves of seismic capacity display noticeable multi-peak characteristic. Moreover, the seismic capacity for L_P1 and L_P2 levels is more sensitive to the variability of geotechnical parameters above and below the structure, while the former for L_P3 and L_P4 levels is more sensitive to that on both sides of the structure. The relevant conclusions can provide some guidance for seismic design and improvement of the performance limits of underground structures in the related codes.

The soil arching effect induced by deep-buried shield tunneling strongly influenced the ground stress and displacement. Therefore, revealing the evolution mechanism of the soil arching effect is a prerequisite for accurately predicting the tunnel load, which has not been understood in deep-buried conditions. Three model tests and eight numerical simulations were carried out to enhance the understanding of the soil arching evolution, in which the stress field, displacement field, and strain field were analysed. The experimental and numerical results indicated that the ground reaction curve presented a two-stage development process of an initially linear decrease followed by a gradual decrease. Compared with the theoretical tunnel loads, the measured and numerical values are relatively larger due to the loosening earth pressure theory ignoring the evolution process of the soil arching effect. The soil arching height decreases with the increase in stress level, measuring 1.75D (the initial diameter of the model tunnel), 1.65D, and 1.61D, respectively, which results from the lagging evolution of the soil arching effect under high-stress conditions. The formation of the shear band was affected by the stress-dependent dilatancy of the soil. At low stress levels, the shear band develops vertically upward. In contrast, at higher stress levels, the shear bands tilt towards the lateral side.

This study employs computer vision and deep learning techniques to execute the refined extraction and quantification of rock mass information in tunnel faces. The integration of contact measurement data and surrounding environmental parameters leads to a proposal for rock mass quality prediction, utilizing integrated machine learning techniques. Subsequently, a 3D model is established by incorporating tunnel face features and environmental data. The safety factor of rock mass excavation is calculated through the utilization of the strength reduction method, and the analysis of rock mass stability on the continuous tunnel face is performed, considering factors such as rock stress and joint sliding. The investigation of variation patterns of excavation safety factors, influenced by multiple modelling factors, is conducted through the utilization of a response surface design method in 46 experimental studies. The research reveals the accurate characterization of complex fissure occurrence obtained in the field through a discrete fracture network. Furthermore, a negative correlation between the safety factor of tunnel excavation and the grade of surrounding rock is observed, with an increase in grade resulting in a decrease in the safety factor. The response surface method effectively discloses polynomial correlations between various parameters such as inclination angle, dip direction, spacing, density, number of groups, and the safety factor. This elucidates the impact patterns of these parameters and their coupling states on the safety factor. The study provides significant insights into the intelligent evaluation of safety for continuous tunnel excavation.

This study explores an integrated framework combining in-situ test-based numerical and data-driven modeling to assess the performance of a deep excavation-tunnel system. To achieve the goal, a case history of deep excavations adjacent to existing tunnels in silt/sand-dominated sediments is introduced to establish a base three-dimensional finite element (3D-FE) model. In-situ tests such as cone penetration test (CPT/CPTU) and seismic dilatometer test (DMT/SDMT), as an alternative to laboratory testing, are used to determine a set of advanced constitutive model parameters. The established excavation-tunnel numerical model is then validated against filed monitoring data. A dataset from numerical simulation is created for training and testing four machine learning models (i.e., artificial neural network (ANN), support vector machines (SVM), random forest (RF), and light gradient boosting machine (LightGBM)), which predict the maximum wall deflection, ground surface settlement, horizontal and vertical displacements of the tunnel. Results show that the ANN model outperforms other models in prediction capacity. Its generalization ability in practice is further enhanced by comparing field measurement data and empirical equations. The findings suggest that, with the integrated in-situ tests, FE and ANN modeling could be used to predict deformation responses of deep excavations close to existing tunnels in soft soil. The present study is useful and valuable for practical risk assessment and mitigation decisions.

The seismic performance of tunnel structure can be examined by fragility analysis, which determines the probability that demand will exceed capacity for a given hazard intensity. Although it is commonly understood that earthquake uncertainties dominate fragility features, the implication of ground motion characteristics on the shield tunnel fragility analysis has not been comprehensively explored. Thus, this study aims to compare the effects of various earthquake characteristics on the fragility of the investigated shield tunnels. To this end, a typical shield tunnel was chosen and modelled using the finite element software. In addition, to account for typical ground motion characteristics, various ground motion sets, including near-field no plus motions (NFNP), near-field motions with a pulse (NFP), and far-field motions (FF), are selected, and a fragility analysis was assessed for every set of ground motion. The fragility curves were generated employing peak ground acceleration (PGA) as the intensity measure (IM) and tunnel drift as the damage measure (DM). The findings indicate that shield tunnels subjected to NFP may be more vulnerable compared to those subjected to NFNP and FF ground motions. This study’s findings highlight the vital role of ground motion characteristics in evaluating the fragility of shield tunnels. Moreover, the results may inform future seismic risk and resiliency evaluations regarding the importance of considering or disregarding the impacts of ground motion characteristics on tunnel vulnerability.

Scientific development and utilization of urban underground space is an inevitable choice for sustainable urban development. However, in the previous suitability evaluation of underground space in coastal cities, the development potential of underground space in the sea area is not considered. Therefore, this study takes the coastal zone of Jiaozhou bay as the study area, establishes evaluation index systems for land and sea areas separately, and explores a new model for evaluating the suitability of underground space in coastal cities by integrating land and sea. In addition, an underground space suitability evaluation model based on the integration of Pythagorean fuzzy sets and Bayesian network is proposed. Firstly, the Pythagorean triangular fuzzy numbers are used to expand the fuzzy range of expert opinions. Then the Pythagorean triangular fuzzy hybrid geometric operator is used to realize the aggregation of expert opinions to solve the difficulty of obtaining the node conditional probability table by the traditional Bayesian network model of underground space suitability evaluation. Finally, the Pythagorean fuzzy Bayesian network is used as an evaluation tool to carry out the underground space suitability evaluation. Based on the evaluation result and urban planning, the overall planning and functional zoning guidelines for underground space development in the study area are given and the suitability and engineering construction difficulty analysis on the second subsea tunnel of Jiaozhou bay is conducted. The research results can provide a valuable reference for the coastal city planning department to develop and utilize underground space.

This case study examines a landmark engineering project in Suzhou, China, involving the construction of two large-diameter (13.2 m) shield tunnels beneath an active high-speed railway (HSR) bridge. This pioneering project is the first of its kind in both China and the world. Advanced numerical simulations were conducted to rigorously assess construction risks. To ensure the operational safety of the existing HSR bridge, an innovative protective system, consisting primarily of segmental steel casing concrete pile barriers, was employed. A comprehensive network of monitoring sensors was strategically deployed to track soil, pile barrier, and pier displacements throughout both the protective and tunnelling phases. Simulation results indicated that tunnelling without protective measures could cause pier displacements of up to 3.1 mm along the bridge—exceeding the maximum allowable displacement of 2 mm in accordance with regulations. Monitoring data revealed that the maximum pier displacement during protective scheme installation was limited to 0.5 mm. With these protective measures, pier displacement during each tunnelling phase remained consistently below 0.5 mm, representing an approximate 80% reduction compared to the unprotected scenario, thereby ensuring the continued safety of the HSR bridge.

To reduce the impact of potential strength outliers on parameter estimation, statistical methods based on the least median square and least absolute deviation have been proposed as alternatives to the traditional least square method. However, little research has been conducted to compare the performance of these different statistical methods. This study introduces a novel procedure for evaluating the three statistical approaches across six selected rock failure criteria, constrained by various rock strength datasets. The consistency of the best-fitting failure criterion and the robustness of the strength parameter estimations serve as the primary benchmarks for evaluation. Based on the benchmark analysis, the following conclusions are drawn. First, both the least square and least absolute deviation methods perform equivalently in identifying the best-fitting failure criterion for a given rock strength dataset, whereas the least median square method does not. Second, when estimating the strength parameters in a two-dimensional failure criterion with the conventional test data of low complexity, the least absolute deviation method is recommended for obtaining robust parameter estimations. Third, as the complexity of conventional test data increases or when true triaxial test data are used to estimate strength parameters for a three-dimensional failure criterion, it is essential to evaluate the outlier-proneness by analyzing the prediction error. If the kurtosis of the prediction error is less than 3, the least square method is preferred. Otherwise, the least absolute deviation method should be used to mitigate the influence of potential strength outliers. This benchmark study offers valuable insights for the future application of different statistical methods in rock mechanics.

Challenges arise in automate design with building information modeling (BIM) in underground space. Industry foundation classes (IFC) standard lacks detailed entity objects for describing excavation retaining structures and geological information, and automated design based on BIM models is not yet for practical application. This study presents a novel automated framework. It integrates the extended IFC standard with mechanical analysis and BIM modeling, significantly advancing structural optimization and rebar detailing. Direct 3D model generation streamlines complex excavation projects, aligning with the trend towards automated, precision-driven design. Key contributions include: (1) the extension of the IFC standard to support excavation retaining structures with objects like IfcBracedPit and IfcPitWall, improving interoperability between geotechnical models and BIM systems; (2) the integration of heuristic algorithms for automated optimization of deformation control parameters, reducing manual intervention; and (3) the promotion of design methodology that bypasses two-dimensional modeling and directly generates three-dimensional models, enhancing efficiency and allowing engineers to focus on high-level decision-making. However, the framework is primarily suited for standard cross-section projects like subway stations and tunnels. Future work will focus on refining the framework for more complex geotechnical projects, addressing software independence and improving design robustness and independence.

To advance resilient infrastructure, this study explores unreinforced shield tunnel segment technologies, a critical but under-researched area. It conducted experiments on ECC-based unreinforced segments (ECCUS), comparing them with ECC-based reinforced segments (ECCRS) and reinforced concrete segments (RCS), focusing on their mechanical properties, including material characteristics, segmental deflection, joint behavior, bolt strain, damage propagation, failure modes, joint toughness, and ductility. Key findings include: (1) ECCUS joints exhibited significantly enhanced bearing capacity, with ultimate strength 34% higher than RCS and 29% higher than ECCRS. In terms of initial cracking strength, ECCUS outperformed RCS by 200% and ECCRS by 34%. (2) The absence of reinforcement cages in ECCUS reduced stiffness but improved overall segment coordination and deformation, leading to deflections 100% greater than RCS and 85% than ECCRS. (3) ECCUS and ECCRS displayed numerous, fine cracks under 200 µm wide, while RCS showed fewer, wider cracks over 3 mm, leading to significant spalling. Cracks in ECCUS were densely distributed across shear and compression zones, in contrast to RCS and ECCRS where they concentrated on compression areas. (4) ECCUS joints exhibited remarkable toughness, with elastic phase toughness 13.47 times that of RCS and 1.91 times that of ECCRS. In the normal serviceability phase, the toughness of ECCUS was 12.17 times that of RCS and 2.53 times that of ECCRS. (5) Considering multi-scale mechanical effects, ECCUS joints amplified the material advantages of ECC over RC more than 11 times during the elastic stage. These findings offer valuable insights for future resilient infrastructure development based on unreinforced construction technologies.

This paper presents a study on multi-objective optimization (MOO) of shield operational parameters (SOPs) for soft ground tunneling using a tunnel boring machine (TBM) in an urban environment, focusing on the case study of the MRT Blue Line in Bangkok. The investigation aims to determine the optimal combination of SOPs, consisting of face pressure ($F_p$), thrust force ($T_f$), grout pressure ($G_p$), and percent grout filling ($G_f$), along with relevant environmental factors, including tunnel depth ($T_d$), inverted groundwater level ($W_i$), and type of surrounding soil ($T_s$). The primary objective is to enhance the penetration rate ($P_{avg}$), in terms of average value), as cost consideration, while mitigating ground surface settlement ($S$), as safety (serviceability) consideration. Using long short-term memory (LSTM) neural networks as predictive models, the results yield coefficient of determination (R²) values of 0.81 and 0.96, root mean square error (RMSE) values of 5.91 mm/min and 3.09 mm, and average bias factor values of 0.99 and 0.88 for the $P$ and $S$ predictive models, respectively, based on validation datasets. This integrated framework, which combines the non-dominated sorting genetic algorithm (NSGA-II) with LSTM neural networks, is applied to MOO to identify the optimal SOPs, while accounting for their influence on $S$ variation as a time-series over 11 timesteps, as considered in this study. For simplification and practical field implementation, the same set of SOP values is applied across all 11 timesteps during the optimization process. Using the proposed optimization framework, the optimal results demonstrate improvements in $P_{avg}$, increasing by up to 109.8% (from 13.99 to 29.35 mm) and in $S$, reducing up to 79.6% (from 34.55 to 7.06 mm) when MOO is conducted as a time series using the simplified method. This finding provides a valuable approach to effectively address the sequential uncertainties of relevant factors in soft ground tunneling for similar projects.

Predicting surface settlement can identify potential risks associated in shield construction. However, in the construction of under-crossing existing structures, the surface settlement is minimal due to the high stiffness of the existing structure, making it unsuitable as a basis for risk assessment. Therefore, interlayer soil settlement was used as an evaluation index in this paper, which was predicted by the developed multi-parameter time series (MPTS) model. This model establishes new dataset, including time, effective stress ratio (ESR), mechanical fluctuation coefficient (MFC), and interlayer soil settlement, where ESR and MFC take into account the changing geological conditions. This study proposes a novel MPTS model, integrating grid search (GS), nonlinear particle swarm optimization (NPSO), and support vector regression (SVR) algorithms to predict interlayer soil settlement during under-crossing construction. It utilizes GS and NPSO to obtain the optimal hyperparameters for SVR. Sensitivity analysis based on MPTS model was used to identify important parameters and propose specific improvement measures. A real under-crossing tunnel project was adopted to verify the effectiveness of the MPTS. The results show that the new input parameters proposed in this paper reduce mean absolute error (MAE) by 20.3% and mean square error (MSE) by 46.7% of prediction results. Compared with the other three algorithms, GS-NPSO-SVR has better prediction performance. Through Sobol sensitivity analysis, previous settlement, ESR and MFC in fully weathered mudstone and moderately weathered mudstone are identified as the primary parameters affecting the interlayer soil settlement. The improvement measures based on analysis results reduce the accumulated settlement by 79.97%. The developed MPTS model can accurately predict the interlayer soil settlement and provide guidance for water stopping or reinforcement construction.

Anthropogenic greenhouse gas emissions stand as the primary catalyst of climate perturbations. A precise evaluation of these emissions holds paramount importance in realizing energy conservation and emission reduction goals. Urban underground highway tunnel facilities emerge as a promising recourse for ameliorating traffic congestion and advancing energy conservation and emission mitigation endeavours. Nonetheless, the methodologies for quantifying its carbon emissions remain scant. This study ventures into the realm of carbon footprint appraisal within the lifecycle paradigm of underground highway tunnel facilities. Tailored to the characteristics, functionalities, and design intricacies of urban underground highway tunnel facilities, the physical boundaries and scopes are meticulously calibrated. Subsequently, a carbon emission computational model adept at encapsulating the emission characteristics throughout the entire lifecycle is formulated. Meanwhile, a detailed database is established for emission factors of various carbon emission activities. Leveraging insights garnered from a specific project case, the overarching carbon emission profiles of the urban underground highway tunnel facility, both in aggregate and individual stages, are elucidated. Concomitantly, bespoke recommendations and strategies aimed at energy preservation and emission abatement are proffered, attuned to the idiosyncratic attributes of carbon emissions across distinct stages.

Muck clogging during shield tunneling often leads to reduced construction efficiency, increased costs and potential safety hazards. Traditional methods for predicting muck clogging primarily rely on the operator’s experience and conventional risk maps, but have limitations in dealing with complex construction conditions. To address these issues, this study presents a Monte-Carlo dropout (MCD)-assisted multi-fidelity neural network (MFNN) framework for effective prediction of muck clogging risk. First, a low-fidelity model is trained based on synthesized data using clogging risk maps. Subsequently, in-situ tunneling data are used as high-fidelity data to train multi-fidelity models. MCD serves to evaluate the uncertainty of the MFNN’s inference, combined with an active learning strategy to refine the low-fidelity model via iterative training of the high-fidelity model. Experimental results show that the MCD-assisted MFNN framework captures clogging features more effectively than traditional machine learning models that use only single-fidelity data, especially in scenarios with imbalanced data. This study provides a viable solution for complex problems in shield tunneling by fully utilizing both experiential knowledge accumulated in engineering practice and field monitoring data, demonstrating the potential of integrating knowledge and data in tackling some challenges that were previously unresolved.

Tunnelling in layered rock with high geostress can cause large deformation disasters, and the reasonable countermeasures rely on a full understanding of the self-bearing capacity of the surrounding rock. In this article, the structural ring concept was introduced to represent the load-bearing capacity of the horizontal layered surrounding rock, whose formation mechanism and determination method were analyzed. Firstly, the mechanical response characteristics of the horizontal layered surrounding rock due to excavation were analyzed. Based on the stress transfer mechanism, the new concept of the structural ring which is a closed structure with a certain thickness was presented. Taking the stress element as the basic analytical model, the maximum increase ratio of the compressive stress was adopted to characterize the structural ring. Then the determination method and its implementation algorithm of the structural ring boundaries were proposed, based on which the beam-arch property of the layered rock was investigated. A series of model tests were carried out to validate the proposed structural ring concept and its determination method. Parametric studies were conducted to illustrate the effect of geological conditions and tunnel geometry on the position and shape of structural rings. Furthermore, the application of the structural ring concept in support design was discussed. It was found that the structural ring was usually oval-shaped with the major axis direction consistent with the dominant in-situ stress. Rock layers had a significant effect on the structural ring, and the beam-arch property was affected by the interlayers and bedding spacing. The support system was beneficial for the formation of the structural ring, which should be designed to balance the support capacity and the stability of the structural ring.