Subgrade engineering serves as the critical load-bearing component for pavement structures, which significantly influences the stability, safety and durability of road infrastructure. To further advance the sustainability of subgrade engineering in China, propel its high-quality development toward green low-carbon, sustainability, and intelligent development, and contribute to the national strategy of building a “Transportation Power”, this study systematically synthesizes the latest advancements in scientific and technological innovation within China's subgrade engineering domain in recent years, while comprehensively delineating the priority directions for future research. The research is grounded in an analysis of the industry's current development status and evolving trends, and centers on six thematic pillars, namely, engineering properties of subgrade fillers, durable subgrade design theory, subgrade widening technology, subgrade protection and retaining structures, intelligent construction of subgrade engineering, and subgrade disaster prevention and mitigation. Specifically, it encompasses cutting-edge research areas, including the mechanical behaviors of various subgrade soils (e.g., oversized-grained soil, coarse-grained soil, fine-grained soil, and special soils), subgrade moisture evolution mechanisms and associated design methodologies, approaches for determining subgrade structural modulus, calculation and control criteria for subgrade permanent deformation, indices and standards for uneven settlement control of widened subgrade, splicing techniques for new and existing subgrade segments, waterproofing and drainage systems for reconstructed/expanded subgrade, advanced ground improvement technologies, subgrade slope stability assessment, slope protection measures, anti-slide piles and retaining wall structures, design and rehabilitation of slope anchorage systems and waterproof-drainage facilities, intelligent compaction of subgrades, intelligent detection and real-time monitoring of subgrade performance, classification of subgrade disasters, subgrade defect detection, disaster monitoring and warning systems, prevention and mitigation strategies for subgrade defects and disasters, and evaluation and enhancement of subgrade disaster resilience. For each area, this study analyzes and deliberates on the current state of academic research, prevailing challenges, targeted countermeasures, and future development prospects. This review is intended to provide strategic guidance and reference for the advancement of China's subgrade engineering discipline, while offering novel perspectives and foundational insights for researchers and practitioners in this field.

Maintaining an adequate level of vigilance during driving is crucial for driving safety. Types of vigilance decline during driving can be categorized into fatigue, distraction, and prolonged automated driving monitoring. These types may differ in key features and attention mechanisms, but their heterogeneous characteristics remain unclear. This heterogeneity may contribute to the poor generalization ability and suboptimal performance of current models for detecting impaired driving states. This study systematically examines the characteristics and mechanisms of vigilance decline during driving through a literature analysis on measurement tools, types, features, influencing factors, underlying mechanisms, detection methods, and warning systems. The following conclusions were drawn: ① While measurement tools for vigilance have formed a relatively complete system, their application in traffic scenarios is not yet widespread. ② The characteristics of vigilance decline are generally defined, but research on type differences in fatigue driving is insufficient, and studies on EEG features of cognitive distraction are lacking. The effects of auditory-cognitive distraction, fatigue driving, and their interaction on takeover efficiency need further exploration. ③ Mechanistically, vigilance decline due to sleep-related fatigue is linked to reduced cortical activity, while task-related fatigue, distraction, and automated driving monitoring are associated with insufficient attention resources and arousal levels. ④ Existing detection technologies focus excessively on fatigue driving and visual-manual distraction, with insufficient research on detecting cognitive distraction and comprehensive vigilance assessment. The high cost and complexity of EEG and eye-tracking devices limit their use. ⑤ Current warning systems overlook factors such as driving environment and individual physiological and psychological states, lacking differentiated warning strategies based on vigilance decline mechanisms. The following recommendations are proposed: ① Strengthen interdisciplinary collaboration to develop a vigilance measurement paradigm specific to traffic scenarios and promote empirical research on vigilance measurement tools in transportation. ② Systematically compare the vigilance characteristics across different types of fatigue driving and analyze eye-tracking and EEG feature maps of distracted driving under various cognitive processing combinations. ③ Develop portable, low-invasiveness EEG devices and create real-time monitoring models for cognitive distraction based on eye-tracking features and ERP indicators. ④ Overcome ERP identification challenges through standardized experimental design, innovative data analysis methods, and multimodal data fusion techniques. ⑤ Establish classification-based warning standards for drivers, design personalized warning strategies based on vigilance decline mechanisms, and integrate in-vehicle environmental control warning systems.

Traffic signs are crucial elements ensuring the safe, efficient, and green operation of the transportation system. However, the effectiveness of traffic signs is often restricted by the adaptability of traffic sign setting standards, the effectiveness of testing methods, and the comprehensiveness of utility analysis. This paper focuses on the three fundamental issues of “unable to find,” “difficult to understand,” and “incorrect navigation” caused by traffic signs, which affect travel quality. It systematically sorts out and summarizes the research hotspots of traffic signs using the scientific knowledge map method. Meanwhile, starting from the driver's cognitive decision-making chain of “discovery, understanding, and execution,” this paper comprehensively reviews the experimental testing platforms, basic theories, and key technologies in traffic sign research. The study further emphasizes the need to deeply characterize the complex impacts of traffic signs on drivers' visual perception, cognitive processing, manipulative behaviors, and vehicle operating states, and to explore their implicit influence pathways on driving behavior. Additionally, it conducts utility assessments and optimizations of the traffic sign system from the perspective of drivers' information needs. Based on this, combining the team's research experience, this paper innovatively constructs a research and application paradigm for traffic signs covering six aspects: “scheme design, feature representation, quantitative evaluation, optimized selection, supporting arrangements, and standard guidelines.” By analyzing typical traffic sign research cases, it elaborates on the specific implementation steps of this paradigm and compares it with similar research methods at home and abroad. The results show that the research and application paradigm of traffic signs integrating human factor needs has obvious advantages. It not only provides theoretical support for sign design research and optimization but also offers a solid basis for solving common issues related to safety facilities in the transportation industry and significant engineering applications.

Automated vehicles encounter significant safety and efficiency challenges within mixed traffic flows involving frequent pedestrian interactions. Precise modeling of this pedestrian-vehicle interaction is therefore crucial. Such modeling is fundamental not only to advancing the vehicle's decision-making intelligence but also to building high-fidelity virtual testbeds, collectively enabling safer navigation, enhanced user experiences, and superior traffic throughput. Despite its importance, the field currently suffers from highly fragmented research and a lack of systematic review. This paper addresses this gap by providing a comprehensive survey of the state-of-the-art progress and key challenges in pedestrian-vehicle interaction modeling. It first deconstructs the essential characteristics of this interaction through the lenses of traffic safety, utility maximization, social norms, and information exchange to establish a formal definition. It then systematically reviews dominant behavioral modeling techniques and interaction quantification methods, further examining the unique attributes and modeling paradigms of automated vehicles as interactive agents. The paper concludes with a summary and outlook on future technological trends. Our review identifies several critical limitations in the current literature: Theoretical: An inadequate understanding of pedestrian cognitive mechanisms and unsystematic insights into the role of communication; Modeling: The constraints of existing physics- or utility-based assumptions and a lack of research into hybrid models; Contextual: A general disregard for the heterogeneity of interaction scenarios, vehicles, and participants; Methodological: The persistent bottleneck of poor explainability in data-driven approaches. To overcome these challenges, future work must focus on deepening the understanding of cognitive processes, exploring the coupling of physics- and utility-driven models, and systematically integrating contextual factors. Significantly, emerging technologies like multimodal large language models and theories of embodied cognition are creating new research paradigms. We argue that substantial progress in this field necessitates deep interdisciplinary fusion and novel applications of these technologies, paving the way for a next-generation intelligent transportation system that is safer, more efficient, and fundamentally human-centric.

With the rapid increase in the demand for automated driving testing, quickly selecting critical scenarios from numerous testing scenarios has become a top priority. Due to the scarcity and low occurrence probability of critical scenarios, which result in low testing efficiency, there is an urgent need to develop accelerated methods for critical scenario identification. To provide a comprehensive review of accelerated methods for critical scenario identification in automated driving testing, this review is explored in three dimensions: functional scenarios, logical scenarios, and concrete scenarios. In the functional scenario dimension, the research primarily focuses on scenario configuration, selecting the combinations of factors that constitute critical scenarios. In the logical scenario dimension, the research focuses on the scope of scenarios, selecting the range of values for the factors that define critical scenarios. In the concrete scenario dimension, the research emphasizes scenario instances, selecting the specific values of factors that constitute critical scenarios. It is noteworthy that current research on functional and logical scenarios is still insufficient, requiring more scholars to engage in this area. Furthermore, existing methods face multiple challenges, including insufficient scenario authenticity, limited acceleration effects, and incompatibility with automated driving functions. Future research should focus on addressing these issues, particularly in the areas of functional and logical scenarios, and continuously optimizing the accelerating technology for critical scenario identification to provide robust support for the ongoing advancement of automated driving testing technology.

To improve the decision-making ability of intelligent vehicle in dynamic scenarios, aiming at the decline of decision-making performance caused by insufficient cognition of changes under interactive environment situations, an overtaking decision-making method that considers real-time interactive vehicle prediction information was proposed. Firstly, by analyzing the overtaking behavior of human driver, a decision-making framework for interactive overtaking behavior of intelligent vehicle that integrates multiple factors was constructed. On this basis, a trajectory prediction model that can distinguish the temporal and spatial features of interacting vehicles was established based on Convolutional Neural Network (CNN) and Long Short Term Memory Network (LSTM). Based on Kullback-Leibler (KL) divergence theory, a real-time guidance and correction mechanism combines physical laws of vehicle motion was designed. Thus, the accuracy and interpretability of the prediction model were improved. Furthermore, based on risk field theory and game theory, constraints on the safety and efficiency benefits of ego vehicle were analyzed, and an overtaking behavior decision-making model for intelligent vehicle considering interactive prediction information was proposed. Equilibrium solution of the overtaking game was realized in dynamic scenarios. Finally, overtaking scenario experiments were conducted through numerical simulation and vehicle test. Results show that the proposed decision-making model that considers interactive prediction information, can realize high-precision trajectory prediction. And it can make real-time overtaking decisions based on the prediction results in dynamic scenarios. The trajectory prediction model predicts average displacement errors of 0.993 meter. In addition, the game stability and benefit during the decision-making process increase by 28% and 20%, respectively. The model shows higher decision-making safety and stability performance. This study provides theoretical support for intelligent vehicle trajectory prediction and driving decision-making, and improve the driving decision-making performance of intelligent vehicle.

Current vehicle collision and danger warning system primarily rely on vehicle-road cooperation using LiDAR sensors. However, LiDAR is generally expensive, and the computational demands for point cloud processing are extremely high, resulting in costly implementations. Cameras, characterized by low unit cost and minimal computational requirements, have become an effective platform for collision and danger warning systems. Existing Camera-based collision warning systems operate in the pixel domain, which suffers from low localization accuracy and a high error probability. To address these issues, a novel 3D tracking algorithm utilizing roadside cameras was proposed. The object-level vehicle-road cooperation was executed in order to extend the vehicle's perception area and improve blind-spot detection performance. To optimize the system for roadside scenarios, the 3D object detection module was modified, by introducing normalized training labels based on camera imaging theory. Using supervised learning, the module effectively learned the relationship between target pixels and depth, significantly improving localization accuracy. In the motion prediction module, a multi-class motion prediction method was proposed, which employed specific nonlinear models to describe the movements of different traffic participants. This ensured precise fitting of motion trends. The Generalized Intersection over Union metric was used to measure the motion similarity between different participants, which enhanced differentiation in dense roadside scenarios and improved association accuracy. Spatial alignment of vehicle-road perception results was achieved through affine transformation. The final fusion results was performed by jointly using the similarity metric and the Hungarian matching algorithm. Experimental results on the V2X-Seq dataset demonstrate that our algorithm outperforms other advanced 3D tracking methods in terms of accuracy and effectiveness. By leveraging vehicle-road cooperation, the system correctly triggers crash warnings in “ghost probe” scenarios, significantly improving the warning success rate and enhancing intelligent vehicle safety.

Human factors, such as insufficient driving experience and the tendency toward high-speed driving, are the main causes of brake failure in heavy-duty trucks on Long Steep Downhill (LSD) sections of highways. Guiding driver to maintain reasonable speeds through information-based intervention is an effective solution. This involves optimizing active speed control strategies to meet the information needs of drivers during their sequential decision-making process. Based on reinforcement learning, the paper establishes an Excellent Driver Model (EDM) with sequential decision-making capability, extracts significant differences in driving behavior between naturalistic drivers and the EDM, and maps these differences into an optimized active speed intervention strategy. First, an interactive environment was built based on a real highway LSD section. Second, considering the cumulative effect of driving risks on LSD sections, a single-step reward function and a global reward function were designed, incorporating driving safety, efficiency and comfort. Then, the Dueling Deep Q-Learning Network (DuDQN) was used to train the EDM. Finally, driving simulation experiments were conducted to collect operation data from naturalistic drivers. Spatiotemporal differences in driving behavior between naturalistic drivers and the EDM were identified using speed trend tests, which helped determine the timing and targets of speed intervention. Numerical results show that EDMs based on DuDQN and DQN perform similarly, with only a slight advantage for DuDQN. Compared to naturalistic driving, the EDM improved the safety performance by 4%, and its maximum acceleration change rate was 0.12 m·s^-3, lower than the 0.22 m·s^-3 observed in naturalistic driving. Significant difference in driving behavior between naturalistic driving and the EDM often occurred in sections with frequent slope changes. By extracting these differential features, a reasonable speed intervention strategy can be developed. The proposed framework innovatively supports the formulation of driving speed intervention strategies in a model-driven manner, which will help enhance active safety prevention and control capabilities on highway LSD sections.

To avoid the influence of inter-individual differences on the performance of fatigue driving recognition and improve the personalized level and efficiency of fatigue driving detection, the proposed method combined Principal Component Analysis (PCA) with One Dimensional Convolutional Neural Network (1DCNN) to build a fast and personalized fatigue driving recognition method. Firstly, simulated driving experiments in highway scenarios were conducted, noninvasively collecting driving behavior data, facial landmark coordinates, and fatigue scores (Karolinska Sleepiness Scale, KSS) of 24 participants. Then, the raw experimental data was segmented using a two-layer sliding time window, and 486 fatigue-related measurements were calculated. Finally, a compact deep learning algorithm was used to construct a personalized fatigue driving recognition model based on PCA-1DCNN. We used 5-fold cross-validation to divide the experimental data of each driver, trained and validated personalized fatigue driving recognition models, and obtained personalized fatigue detection thresholds. The proposed personalized fatigue recognition models achieve an average accuracy, sensitivity, and specificity of 99.93%, 99.92%, and 99.94% across all participants, respectively. The recognition performance is relatively stable for all participants, with standard deviations of 0.07%, 0.11%, and 0.09% for accuracy, sensitivity, and specificity, respectively. The proposed method does not require manual feature extraction and use a lightweight 1DCNN to quickly detect fatigue driving, with an average detection time of 0.010 3 seconds. The proposed method aims to avoid the influence of inter-individual differences and improve the accuracy and speed of fatigue driving recognition at an individual driver level, which can support the development of efficient personalized fatigue driving warning systems and promote personalized modeling research on other dangerous driving behaviors.

As a typical high-risk vehicle type, the coordination between truck driving speed and highway geometric design directly affects the operational safety level of expressways. To clarify the influence of highway geometric design on exiting truck speeds, this study analyzed the operating speeds of floating trucks on expressway segments from the mainline to the ramps and constructed a prediction model for exiting truck operating speeds on expressways. First, by examining the speed variation trends of trucks during the exit process and considering highway geometric design factors, the key characteristic cross-sections of truck speed within the diverging area were identified. Subsequently, a variable importance projection analysis was conducted to determine the critical geometric design factors affecting truck speed variation at these cross-sections. Finally, based on the measured data from 12 cases, the partial least squares regression method was employed to establish operating speed prediction models for trucks at three characteristic cross-sections, and the models were validated using the measured data from four cases. The results indicate that truck speeds undergo significant changes at the taper point, diverging point, and gore area, with the initial deceleration typically occurring approximately 500 m upstream of the exit. The results indicate that the speed of the trucks changes significantly at three cross-sections: the taper point, divergence point, and gore area. The deceleration behavior of trucks typically first emerges approximately 500 m upstream of the exit, implying that the geometric design and traffic signs exert a significant influence on truck speed. Critical design factors affecting truck speed variations include the length of the taper section, length of the deceleration lane, taper rate of the transition section, length of the taper line, length of the guide line, and radius of curvature at the end of the gore area. The developed prediction model can relatively accurately describe the law of truck speed changes in the divergence area, with the mean absolute percentage error of its prediction results being less than 10%. These findings offer theoretical support and practical guidance for evaluating alignment consistency, formulating speed-limit schemes, and deploying traffic safety facilities in highway interchange areas.

Scour is one of the primary causes of hydraulic bridge failure. However, although classical scour analysis theory has seen new advancements in numerical simulations and turbulence phenomenological models, it still fails to properly balance computational accuracy and efficiency. The theory struggles to efficiently and accurately compute equilibrium scour depth and spatial morphology, thereby impeding advancements in underwater structural safety design and operational maintenance for bridges. On one hand, mainstream equilibrium scour depth calculation formulas are often overly conservative and differ significantly from actual bridge measurements. On the other hand, classical single-phase-flow computational methods still exhibit considerable differences in morphology and depth compared to flume test observations, while two-phase-flow models that can more accurately describe sediment transport are computationally expensive. We proposed a theoretical system and technical roadmap for intelligent bridge scour analysis that proceeds from three perspectives: optimization of scour calculation formulas, intelligent single-phase flow, and intelligent two-phase flow simulations. This was achieved through the synergistic integration of multi-source data, multi-dimensional domain knowledge, and advanced artificial intelligence algorithms. In related intelligent analysis cases, this theoretical system and technical roadmap demonstrated significant advantages over traditional mainstream empirical formulas and simulation methods across multiple key dimensions, including computational accuracy, efficiency, safety, and generalization capability. It effectively alleviated the long-standing contradiction in traditional scour assessment where accuracy and efficiency were difficult to balance. This approach not only provides a feasible solution for high-accuracy and high-efficiency scour assessment but also offers technical support for full life-cycle bridge design and disaster prevention and mitigation.

This study aims to enhance the seismic performance of seismically isolated bridges by combining the advantages of variable curvature pendulum bearing (VCPB) and shape memory alloy (SMA) cables. We propose a novel SMA variable frequency pendulum bearing (SVFPB) and parameter design method specifically for bridges with SVFPBs. The hysteretic model of the SVFPB is derived, and numerical models for the VCPB and SMA cables are developed using OpenSees software. The accuracy of these numerical models is verified against experimental data, and the SVFPB is further modeled. We then optimally design the parameters of the SVFPB for an isolated bridge based on the validated model and proposed method. We determine the optimum scale and distribution coefficients for the SMA cables, along with their corresponding yield forces. The effectiveness of the proposed method and the performance enhancement of the SVFPB system in bridges are demonstrated through the optimum parameters. The analysis considers the efficacy of the SVFPB system in mitigating seismic responses in bridges. Results show that the optimum model parameters for the SMA cables can be achieved using the proposed method. The seismic performance of bridge equipped with SVFPBs using these optimum parameters shows significant improvement. Specifically, the girder peak displacement, bearing residual displacement, and base forces in piers are fully controlled. The girder peak displacement and bearing residual displacement are reduced, and the reduction reached 6.7% and 51.3%, respectively. The maximum increments in base force and bending moment of the piers are limited to 7.4% and 7.0%, respectively. These findings provide reliable foundation for enhancing the performance of bridges with SVFPBs.

To investigate the mechanical behavior of perfobond leiste (PBL) shear connectors in coarse aggregate ultra-high performance concrete (CA-UHPC) thin slabs, push-out tests were conducted to study the influence and mechanisms of factors such as the presence of through reinforcement, perforated steel plate thickness, hole diameter, concrete strength, and connector configuration. The test results indicate that, in specimens without through reinforcement, the concrete tenons had to independently resist the relative slip forces, resulting in shear failure accompanied by the rapid development of vertical cracks, which prevented the utilization of their excellent compressive strength. Compared to specimens with through reinforcement, the shear capacity and stiffness decreased by 34.5% and 28.8%, respectively, displaying a brittle failure mode. Using the equivalent replacement theory for area and stiffness, the shear contribution of through reinforcements can be represented by their equivalent concrete tenon area. Due to the use of 13 mm and 20 mm steel fibers, increasing the thickness of the perforated steel plate from 12 mm to 20 mm reduced the number of fibers distributed along the shear interface beneath the concrete tenons by 52.9%, weakening the pin resistance and resulting in a 17.2% decrease in shear capacity. However, with the increase in perforated steel plate hole diameter from 40 mm to 50 mm and the compressive strength of CA-UHPC from 146 MPa to 163 MPa, the shear capacity increased by 32.8% and 62.2%, respectively. At the same time, compared with ordinary UHPC, the addition of coarse aggregate promotes the improvement of the compressive bearing capacity and stiffness of CA-UHPC concrete tenon, so that the PBL connector has better shear bearing performance. These two factors were identified as the primary contributors to the improvement of shear capacity and stiffness. This study provides valuable data and theoretical support for the performance design and engineering application of PBL connectors in the field of CA-UHPC.

Optical flow methods in computer vision have significant potential in the field of structural health monitoring. However, traditional optical flow methods often lack robustness and accuracy when faced with complex environmental changes. Therefore, this paper proposes an adaptive feature point recognition algorithm based on computer vision to improve the accuracy of feature point detection and tracking under challenging conditions, such as variations in lighting, perspective, noise interference, and rain. The algorithm dynamically adjusted the parameters for feature point extraction, matching, and optimization to adapt to the motion and variation of feature points in different scenarios and environmental conditions. Experimental results show that the root mean square error (RMSE) of the algorithm is significantly lower than that of the traditional LK optical flow method under various conditions, including lighting variation (bright and low light), perspective variation, noise interference (strong and weak noise), and rain interference (heavy and light rain), thus verifying the robustness and accuracy of the adaptive feature point recognition algorithm in complex environments. Furthermore, a comparison of processing time and the number of feature points demonstrates the algorithm's advantage in real-time performance. This study provides an effective method for the real-time monitoring of bridge structures, contributing to improved construction quality and operational safety.

To achieve the multi-level seismic fortification goal of precast segmental piers, a precast segmental double-column pier with gradient seismic capacity is proposed, and its performance evolution under different earthquake levels is studied. The quasi-static test of precast segmental double-column piers was carried out. In view of the damage evolution characteristics of the gradient seismic piers in the test, the evolution laws of typical seismic performance indexes such as hysteretic characteristics, energy dissipation capacity and residual displacement were analyzed. Based on the improved Park-Ang damage model, the seismic performance index system that characterizes the relationship between the overall performance of the gradient seismic pier and the damage state was obtained. The results show that the gradient seismic piers show the characteristics of graded damage evolution under the action of earthquake. The components are damaged and destroyed in turn under different loading levels, and the typical seismic performance index shows obvious gradient evolution characteristics; The proposed seismic performance index can describe the damage and evolution of piers at different stages；The seismic performance index of the gradient seismic piers shows a significant gradient change, and its performance index is 3%-16% higher than that of the ordinary precast segmental piers. The research results can provide a reference for the graded seismic design of bridge piers.

When a data-driven stochastic, subspace, modal identification method processes large-scale data, the optimization of the construction parameters of the Hankel matrix is key to achieving efficient and accurate modal identification. In this study, a method is proposed to determine the optimal parameters of the Hankel matrix to improve the efficiency and accuracy of the automatic identification of dense modals in large-span bridges. First, a Hankel matrix was constructed from the acceleration response data of the numerical simulation, and the optimal number of block rows in the Hankel matrix was determined using the condition number of the projection matrix. Subsequently, we redesigned the division ratio of the subspace of the Hankel matrix and determined the optimal division ratio of the subspace by analyzing the effects of different division ratios of the subspace of the Hankel matrix on the automatic modal identification results. Subsequently, the density-based spatial clustering of applications with a noise (DBSCAN) algorithm with an optimal minimum number of points (MinPts) was used to separate completely the dense physical modes from the spurious modes, and the coefficient of variation was analyzed to determine the effective modal parameters. Finally, the applicability of the optimal parameters of the Hankel matrix in the dense modal identification of large-span bridges was verified based on the measured data from the suspension bridge health monitoring system. The results show that determining the optimal number of rows in the Hankel matrix improves the quality of the physical modal feature information accumulated by the projection matrix, and improves considerably the accuracy of modal identification. The DBSCAN algorithm with optimal MinPts can determine accurately the physical mode order and successfully extract dense physical modes in the frequency range of 0-0.5 Hz. Moreover, an automatic modal identification method based on the optimal division ratio of the Hankel matrix subspace can effectively reduce the dispersion of the frequency and damping-ratio modal identification results. The selection of reasonable Hankel matrix parameters can improve the accuracy of the automatic identification of dense modes in large-span bridges that is of great significance for bridge health monitoring applications.

In order to improve the bonding performance of weak surfaces between layers in 3D printed concrete structures and enhance the flexural capacity of printed bridges, especially the main girder components, and promote the further application of 3D printing technology in bridge engineering, a herringbone (HB) stiffener was proposed to strengthen the interlayer interface of 3D printed concrete. Various reinforced specimens were designed and fabricated with subsequent tests, including direct tensile test, tensile splitting test and direct shear test respectively. Results show that the implantation of HB stiffeners significantly enhanced the tensile and shear properties of the printed specimens. Compared with the unreinforced control group specimens, the tensile strength of the specimens in the direction perpendicular to the interlayer interfaces (Z-direction) increased from 0.832 MPa to 1.823 MPa, with an increase of 119%. The splitting strength along the printing path direction (X-direction) increased from 1.262 MPa to 2.387 MPa, with an increase of 89%. The splitting tensile strength in the direction perpendicular to the printing path (Y-direction) increased from 1.179 MPa to 2.212 MPa, with an increase of 87%. The shear strength in the X-direction increased from 1.487 MPa to 3.819 MPa, with an increase of 156%; and the shear strength in the Y-direction increased from 0.897 MPa to 3.350 MPa, with an increase of 273%. It is demonstrated that the use of HB stiffener can effectively improve the interfacial bonding performance of printed components, especially in tensile and shear resistance. In addition, the typical loading feature of unreinforced and reinforced structure were revealed through load-displacement curve analysis. It is further proved that the implantation of stiffeners not only improves the mechanical strength of the interlayer interface, but also enhances the structural toughness of the 3D printed specimen. Finally, based on the results, calculation formulas for the tensile and shear strength of 3D printed reinforced specimens were derived with the semi-theoretical, semi-empirical approach, providing theoretical support for the engineering applications of 3D printed reinforced concrete structures.

In corrosive environments, prestressed concrete (PC) structures frequently suffer from varying degrees of localized corrosion, which reduces their load-bearing capacity. To investigate the impact of the location and degree of localized corrosion on the flexural performance of PC beams, three post-tensioned PC beams were designed, with two beams subjected to localized electrochemically accelerated corrosion tests at the mid- and quarter-span positions. After achieving the designed corrosion levels in the localized regions, bending tests were performed on all three beams. The effects of localized corrosion on beam deflection, crack propagation, section strain, and failure modes were analyzed, and numerical simulations were conducted using 3D scanning technology and ABAQUS models. The results indicate that localized corrosion significantly reduces the bending stiffness of the PC beams, weakening their load-bearing capacity, ductility, and deflection performance. The effect of mid-span corrosion on bending performance was greater than that of quarter-span corrosion. Additionally, localized corrosion significantly reduced the crack resistance of the corroded regions, leading to an increased number of cracks and a faster crack propagation rate. At the mid-span, localized corrosion primarily affects the width of the cracks in the flexural zone of the beam, whereas quarter-span corrosion mainly influences the width of the diagonal cracks. However, mid-span corrosion has a more significant effect on the maximum crack width at the ultimate load. Compared to the uncorroded beams, the cracking load in the localized corrosion areas decreased by 10%-20%, and the number of cracks increased with the corrosion level. The maximum crack width at the ultimate load decreased by 32%-50%. In corroded beams, nonuniform corrosion tends to facilitate local stress concentration more than uniform corrosion, leading to more severe crack damage in localized areas, which influences crack development and width. The nonuniform corrosion model obtained through 3D scanning more accurately reflects the impact of corrosion on the local mechanical properties of the beams.

This paper endeavors to enhance the crack resistance of the negative moment regions in continuous composite beams by integrating Hybrid Fiber Reinforced Concrete (HFRC) on the basis of prestressed application. Four composite beams with different concrete slab were initially designed and fabricated: conventional concrete slab (CB-1), HFRC slab (CB-2), prestressed conventional concrete slab (PCB-1), and prestressed HFRC slab (PCB-2), to conduct static load tests. During the test, the accelerometers were utilized to acquire the frequency and corresponding mode shapes at three static load stages: LS-1 (0 kN), LS-2 (100 kN), and LS-3 (ultimate load). The results revealed that the maximum crack widths in specimens CB-2, PCB-1, and PCB-2 were reduced by 19.2%, 42.5%, and 59.2% respectively, while the cracking loads increased by 43%, 129%, and 189% respectively. At the LS-1 stage, relative to specimen CB-1, the first-order frequencies of specimens CB-2, PCB-1, and PCB-2 increased by 2.4%, 2.9%, and 7.6% respectively, the second-order frequencies by 3.6%, 5.9%, and 6.8% respectively, and the third-order frequencies by 0.6%, 1.4%, and 3.5% respectively. This indicates that the integration of prestressing technology and hybrid fibers can improve the crack resistance and stiffness of the negative moment regions in continuous composite beams, take advantage of the synergy of the prestressing technique and HFRC. Subsequently, the damage extent in the negative moment regions of composite beams was quantified using the enhanced Coordinated Modal Assurance Criterion (eCOMAC), where the eCOMAC values at mid-span of specimens CB-2, PCB-1, and PCB-2 showed a decrease of 22.5%, 34.3%, and 54.8% respectively compared to specimen CB-1, aligning well with the experimental results. Finally, the finite element software ABAQUS and eCOMAC were used to investigate the effect of prestressing and the volume content of steel fiber and polypropylene fiber on the cracking performance of the negative moment zone. The results show that PF has a slight improvement on concrete damage, while SF has a more obvious improvement effect. And prestressing is best.

Tunnel structures crossing active fault zones face threats from fault dislocation deformation. The application of variable-stiffness flexible joints is one of the effective measures to enhance the tunnel's resistance to fault dislocation. However, there remains a lack of longitudinal stiffness optimization design methods specifically tailored for tunnel structures with variable-stiffness joints. To address this gap, this paper proposes a longitudinal stiffness optimization design method for variable-stiffness joint tunnel structures traversing active faults. First, multi-level resilience design objectives for fault-crossing tunnels are established, and the design concept and process for variable stiffness flexible joint resilient structures are proposed. Second, based on the structural characteristics and key design parameters of variable stiffness flexible joints, the tunnel structure system is simplified into a multi-parameter beam-spring model. A backtracking analysis method based on a random search algorithm is applied, using the multi-level fault-displacement resilience design objectives as the reference, to obtain the longitudinal stiffness optimization distribution function and key design parameters of the flexible joint tunnel structure. Finally, four typical examples are designed to compare and verify the advantages and reliability of this design method. The variation patterns of the angle and shear stiffness optimization distribution functions of the tunnel's variable stiffness flexible joints under different fault displacement magnitudes are explored. The analysis results show that a variable stiffness resilient structure with an optimized longitudinal stiffness distribution function can better adapt to fault displacement. It reduces the internal forces of the tunnel structure while ensuring the smoothness of the tunnel axis. Near the fault displacement surface, reducing the stiffness of the tunnel joint helps improve the tunnel's fault-displacement resilience, while further from the displacement surface, the joint stiffness exhibits a “combination of rigidity and flexibility” distribution along the longitudinal direction. Additionally, the design of the tunnel's fault-displacement resilience should focus on the angle stiffness of the joint to accommodate the large-scale dislocation deformations that may occur in the fault zone. This study provides a theoretical basis for the fault-displacement resilience evaluation and the optimization design of variable stiffness joints in tunnels crossing active fault zones.

During metro station construction, holes are created in the main structure of the station to meet functional requirements such as interchange and ventilation, leading to a greater weakening of the bearing capacity of the main structure in the hole-opening area. To investigate the influence of a hole on the arch structure of a metro station constructed using the pile-beam-arch method on the overall mechanical properties, this study considered a newly built metro station in Guangzhou. First, the damage mode, crack extension, and deformation law of the opening in the arch structure of the metro station constructed using the pile-beam-arch method were studied in depth through indoor model tests. Subsequently, the stability and influence of the full-scale opening in the arch structure were investigated using numerical simulation. The results of the study show that although the beam in the region of the opening is in three-dimensional torsion, its horizontal bending stiffness is increased, with vertical bending as the main deformation characteristic. Moreover, the longitudinal girder without an arch causes a 'Y’ type tensile-shear composite penetration crack to form in the center of the longitudinal section, and a 45° penetration diagonal crack forms in the middle of the span. Furthermore, significant stress concentrations are observed at the opening boundaries, the overall force transmission path of the station structure changes, and the center arch does not directly influence the rest of the structure in the opening. Compared with the complete structure, the influence range of the station arch opening is 80% times that of the length of the opening, with a deformation difference of less than 10% as the criterion. The order of influence of each structural internal force is as follows: beam， arch，floor slab， side wall. These results provide a reference for underground engineering construction under similar geological conditions.

Sunken tunnels are low-lying sections of urban roads. It is crucial to minimize or even eliminate urban flood disasters and ensure traffic safety by preventing the accumulation of excessive water under heavy rainfall conditions. In this study, road surface conditions and rainfall loss were considered as the starting points, and a formula was derived to predict the water depth in low-lying sections characterized by inflow, drainage capacity, and road slope functions. Furthermore, the storm water management model was used to construct a drainage network model to predict the water depth under different rainfall and drainage conditions and to verify the feasibility of the theoretical formula. Finally, based on the simulation and prediction results, a risk-level assessment of the water depth within the tunnel was conducted. The research results show the following. Under ideal drainage conditions, the maximum water depths in the tunnel for rainfall recurrence periods of 10, 50, and 100 years are 14.93, 20.85, and 22.74 cm, respectively. The simulated water depth prediction curves are consistent with the results obtained using the theoretical formula. The curves initially exhibit a rapid increase, followed by a gradual increase, reaching a peak value, and then a rapid decrease. The peak water depth occurs after peak rainfall intensity. At a pipe blockage rate of 25%, the water accumulation risk level is III when the rainfall recurrence period is 10 years and IV for both the 50- and 100-year recurrence periods. The research findings provide a reference for the monitoring, early warning, and risk assessment of water accumulation depth in sunken tunnels.

In order to improve the construction efficiency, terrain adaptability and structural resilience of tunnel sheds along mountainous highways, a steel tunnel shed composed of multiple modular units is proposed, each of which contains a flexible buffer structure consisting of flexible protection net system and sand cushion. Based on the full-scale impact test with a protective kinetic energy of 500 kJ, combined with the dynamic nonlinear numerical analysis, the working mechanism and force transfer path are discussed, the deformation response and internal force distribution characteristics are indicated, and the effects of sand cushion thickness and rockfall mass on the protection performance are further studied. The results show that the dual protective mechanism of the flexible buffer structure can effectively intercept the rockfall. Regarding the energy consumption mode, the internal energy takes the dominant position, in which the proportion of ring mesh, energy dissipating device and sand cushion is 9.50%, 68.72%, and 21.78%, respectively. Compared to the model with only the sand cushion, the intrusion depth of sand cushion in the flexible buffer structure decreases by 61% and the impact force by 71%. Although the peak impact force gradually decreases as the thickness of the sand cushion increases, considering the structural weight bearing and protective effect, it is suggested that the thickness of sand cushion is 0.4 m. Moreover, under the same impact energy level, the internal force of wire-ring mesh and other components reduces with the increasing mass of the rockfall, while the energy dissipating device are activated even more fully. On this basis, the evaluation criteria of the service energy level and the ultimate energy level of the flexible buffer structure are defined, and then the service energy level and the ultimate energy level of the tested structure is clarified as 500 kJ and 1 500 kJ, respectively.