Sand intrusion into ballasted beds seriously threatens their long-term stability and operational safety. Based on wind tunnel experiments and particle image velocimetry (PIV), this study investigated the movement of sand particles around ballasted beds in a wind-sand environment by systematically measuring and analyzing the spatiotemporal evolution of particle velocity fields, directional distributions, and flux transport. The results show that, when the wind-sand flow passes through the ballast-rail system, the flow field structure changes significantly, exhibiting clear velocity stratification and flow direction reorganization. The particle motion direction undergoes a typical evolution process of convergence, deflection, chaos, and recovery along the flow path. The directional concentration decreases from 0.959 on the windward side to 0.200 in the inter-rail region, and then rises to 0.639 on the leeward side. The particle flux attenuates by about 48% along the path, while near-surface deposition is significant, with the proportion of downward-moving particles generally exceeding 60% at all measurement positions. The ballasted bed affects wind-sand transport through the combined mechanisms of energy dissipation and screening: energy dissipation continuously weakens the sand-carrying capacity of the airflow, while the screening effect promotes sand deposition within the ballast layer.

To further enhance the intelligent recognition, assessment, early warning, and active prevention and control capabilities of high-speed railways in responding to risks such as natural disasters, perimeter invasion/foreign object intrusion, and external environmental safety, a method for active perception and early warning of the operational environment safety of high-speed railways is proposed based on the concept of active control of high-speed railway operating environment safety. By analyzing the action mechanism and spatiotemporal evolution patterns of the main influencing factors on the operational environment safety of high-speed railways, the disturbance mechanisms of various risk sources on train operation are revealed. On this basis, a situational awareness method for the operating environment safety across full spatiotemporal scenarios is designed, covering refined forecasting of meteorological disasters, multi-modal fusion-based recognition of perimeter invasion/foreign object intrusion, and intelligent perception of external environmental hazards through air-space-ground collaboration. Corresponding intelligent assessment and early warning models are then constructed, and active control and emergency response strategies are formulated. The results show that the accuracy of refined gale situational awareness for wind speed forecasting reaches 93%. Compared with the existing similar intelligent methods, the transmission delay of alarm information from system generation to train's beyond-visual-range terminal display is reduced from 2.364 s to 1.651 s. This method can provide a systematic solution for engineering applications and demonstrate promising prospects for practical implementation.

To optimize the layout of railway rescue trains and enhance railway emergency rescue efficiency in China,the genetic-simulated annealing hybrid algorithm is improved based on the arc risk quantification. First, a multi-dimensional risk quantification evaluation index system for the railway network is constructed. Through the Entropy Weight-TOPSIS method, risk quantification evaluation is conducted on each arc segment of the network. Then, combined with coverage theory, an optimal layout model for railway rescue trains is established with objectives including network rescue coverage rate, rescue time satisfaction, and rescue train layout cost. Secondly, the Multi-Phase Adaptive Simulated Annealing Genetic Algorithm (MP-ASAGA) is designed to solve the model. The solution process is divided into the exploration phase focusing on searching for the global optimum and the development phase focusing on accelerating convergence, with different evolutionary strategies applied in each phase to improve the algorithm's solving performance. Finally, a case study using actual railway network data from a railway bureau in China is conducted for calculation and validation. The results show that compared with the original layout scheme of the railway bureau in the case study, the optimal railway rescue train layout scheme obtained by the proposed method achieves an improvement of 8.99% in network rescue coverage rate, and an improvement of 11.62% in rescue time satisfaction. This method can provide corresponding theoretical support for the layout optimization of railway rescue trains and the enhancement of rescue efficiency.

To clarify the seismic failure mechanism and develop a seismic performance evaluation method for prefabricated metro station structures in liquefiable sites, this study takes Shuangfeng Station of Changchun Metro Line 2 as an engineering case and establishes a three-dimensional soil-structure interaction numerical model using the finite difference software FLAC3D. The evolution characteristics of soil pore water pressure as well as the response laws of displacement and stress of the prefabricated station structure under different ground motions are investigated. Combined with the quasi-static test results of the prefabricated station structure, the seismic damage evolution process and failure mechanism are analyzed, and a dual-parameter seismic performance evaluation method simultaneously considering the global inter-story drift ratio and the opening amount of mortise-and-tenon joints is proposed. Subsequently, seismic fragility analyses are conducted based on both scalar and vector-valued ground motion intensity parameters. The results indicate that when the peak ground acceleration (PGA) is ≥0.2g (g as gravitational acceleration), significant liquefaction occurs in part of the site, and the onset time of liquefaction is markedly advanced with increasing ground motion intensity; the degree of liquefaction near the structure is generally lower than that in the area far from the structure. Liquefaction-induced stiffness degradation and non-uniform ground deformation significantly alter the structural load-transfer path; structural damage is mainly concentrated in the central column and the mortise-and-tenon joints of the sidewalls, exhibiting a progressive evolution from the ends of the central column toward the sidewalls and the connection zones of the arch roof and bottom slab. Even under a low axial load ratio, the central column remains the most vulnerable component. The proposed dual-parameter evaluation criterion enables a more rational assessment of seismic performance. Compared with conventional scalar intensity measures, vector-valued intensity measures more comprehensively reflect the influence of ground motion amplitude and spectral characteristics on structural failure probability, thus obtaining more reasonable fragility assessment results. The findings can provide references for the seismic design and performance assessment of prefabricated metro station structures in liquefiable sites.

To address the problem where the gauge corner of rails is prone to negative deviation due to excessive grinding, leading to reduced wheel-rail equivalent conicity and impaired running stability of Electric Multiple Units (EMUs), a rail profile optimization method considering grinding deviation is proposed. First, taking a high-speed railway as an example, statistical analysis is conducted on the deviation of measured post-grinding rail profiles to reveal the distribution characteristics of grinding deviation, and the adverse effects of negative deviation on wheel-rail contact performance are verified through vehicle dynamics simulation. Second, the measured grinding profiles are divided into training and validation sets. With the deviation between the degraded profile and the 60N profile controlled within -0.2 mm to +0.2 mm and the nominal equivalent conicity no less than 0.034 as constraint conditions, and with the the optimization objective that the contact stress for the optimized profile matched with the LMA wheel not exceed that of the LMA-60N combination, a genetic algorithm is employed to optimize seven characteristic parameters of the 60N profile, yielding the optimized profile and the degraded profile considering negative deviation caused by excessive grinding. Finally, the performance of the optimized and degraded profiles is verified through wheel-rail contact analysis and vehicle dynamics calculation; and a comparative evaluation is conducted against the standard 60N profile and two sets of field-measured grinding profiles. The results show that: in the gauge corner region (16° - 45°), the profile deviation is predominantly negative, mainly distributed in the range of -0.8 mm to 0 mm with a mean value of approximately -0.4 mm, indicating a prominent excessive grinding problem; compared with the standard 60N profile, the optimized profile matched with the LMA wheel achieves an increased nominal equivalent conicity of 0.036 and a 14% reduction in wheel-rail contact stress, exhibits comparable dynamic performance and shows improved car body lateral stability; the deviation between the degraded profile and the 60N profile is controlled within -0.2 mm to +0.2 mm, with a nominal equivalent conicity of 0.034, satisfying the design constraints and effectively resolving the problem of excessively low equivalent conicity; compared with the two sets of field-measured grinding profiles, the degraded profile demonstrates significant advantages in both dynamic and contact mechanical performance, with reductions of 19% in wheelset lateral acceleration, 7% in bogie frame lateral acceleration, and 13% in car body lateral acceleration, as well as reductions of 19% - 37% in maximum normal contact stress and 33% - 41% in maximum tangential contact stress, along with significantly reduced average contact stress and more concentrated distribution. The proposed profile optimization method provides a reference for field grinding operations and offers guidance for addressing the low conicity hunting problem of EMUs.

To meet the requirements of modern railway bridge emergency repair, a technical scheme for a deployable medium-span emergency repair girder based on telescopic diagonal web members is proposed. The girder utilizes deployable frame units as its basic components, enabling folding and deployment through the extension and retraction of the diagonal web members. This design resolves the technical challenge of balancing assembly efficiency with storage and transportation space in existing repair girders, while also meeting the emergency repair demands of both conventional-speed and high-speed railway bridges. Finite element analysis models and multi-body dynamics models are established to conduct static analysis and vehicle-bridge coupled dynamic response analysis on the deployable railway emergency repair girder. The results indicate that the stress levels and displacements of the deployable repair girder meet the limit requirements of the “Code for Design on Railway Bridge and Culvert”. The member stresses are highest under the loading of mixed passenger and freight railway traffic. The arrangement of the diagonal web members significantly influences the ultimate bearing capacity of the deployable girder, with the inverted V-shaped configuration yielding a higher ultimate bearing capacity. Among the three truss configurations for the 32 m span, the heavy truss emergency repair girder exhibits superior dynamic response indices. The wheel load reduction rate is identified as the key factor controlling train speed, and the speed limit for high-speed trains crossing the 32 m span deployable railway emergency repair girder can be controlled at 120 km · h^-¹.

To address the mechanism analysis and treatment of inverted arch uplift and track slab cracking in an in-service tunnel, a combined approach of field structural disease characteristic analysis and laboratory tests was adopted to explore the disease mechanism of inverted arch uplift deformation and track slab cracking in an in-service tunnel of the Shanghai-Kunming Railway. According to the analysis of structural cracking characteristics and disease mechanism, integrated treatment measures of "grouting anchor pipe installation - bedrock grouting and removal - reconstruction of inverted arch structure" was proposed. These measures were adopted to guide the construction of the background project, and the evolution laws of the contact stress between the bedrock and the inverted arch was monitored and analyzed during the construction process. The results indicated that: the expansion deformation of bedrock upon water exposure was the main cause of local uplift of the inverted arch and other supporting structures, leading to increased internal forces, uneven deformation, and cracking; factors such as bedrock bearing capacity reduction, stress concentration induced by local high in-situ stress, and uneven stiffness and stress distribution in the invert arch structure further exacerbated the risk of non-coordinated deformation and cracking in the bedrock-invert arch system; the contact stress between the bedrock and the reconstructed inverted arch showed a phased evolution pattern, initially increasing slowly and then gradually converging to stability. The average contact stresses at three monitoring sections were 167.83, 169.51 and 165.82 kPa, respectively, which ensured the stability and safety of the tunnel structure. The treatment measures in combination with the disease mechanism analysis effectively prevented and controlled inverted arch uplift and structural cracking in the in-service tunnel. The research results provide a design scheme and engineering application reference for the treatment and prevention of similar engineering diseases.

Given the significant randomness of vehicle-bridge dynamic response for higher-speed railways, this study aims to explore the characteristics and probability distribution of dynamic response of a 400 km · h^-1 train passing through a bridge. A vehicle-bridge coupled random vibration model is established based on the pseudo-excitation method and the whole-process iteration method, and its validity is confirmed through comparison with the simulation results of Monte Carlo method. Based on this model, the time-frequency distribution laws of safety and stability indices of the train running at 400 km · h^-1 are analyzed, and the random characteristics of vehicle-bridge dynamic response under higher speeds on simply-supported beams with different fundamental frequencies are studied. The results show that the statistical values of vehicle-bridge response vary with time, showing typical non-stationary characteristics. The dynamic coefficient of the bridge is mainly controlled by the arrangement of train axle and the wheelbase, and is only slightly affected by the random excitation of track irregularity. Under resonance conditions of simply supported beam, the wheel load reduction rate increases significantly with the increase of speed, and the carbody vibration acceleration is insensitive to the resonance response of the simply supported beam. The fundamental frequency of the simply-supported beam has minor effect on the wheel load reduction rate and carbody vibration acceleration, whereas the randomness of the track irregularities has a significant effect on the vertical vibration acceleration and the wheel load reduction rate of the bridge.

With the continuous advancement of infrastructure construction in western China, research on the bearing mechanisms and design methods of bridge foundations in complex terrain has become increasingly important. Focusing on the mechanical properties and structural design of embedded foundations for railway bridges in mountainous areas, this study investigates the potential failure modes of slope rock mass under combined loads. A theoretical calculation model for the embedded foundation-rock mass system under slope terrain conditions was established, revealing the interaction mechanism between the foundation and the slope rock mass. Based on this, combined with limit equilibrium theory, formulas for the ultimate bearing capacity of vertical embedded foundations and inclined arch-abutment embedded foundations under slope conditions were derived. The design rationality of the embedded foundation for the Zhongjian River Bridge was verified. The results show that the primary failure mode of vertical embedded foundations is overall shear failure of the rock mass at the pile end. As the shear force and bending moment loads outside the slope increase, the foundation-rock mass system is prone to horizontal shear failure. For inclined arch-abutment embedded foundations, the main failure mode involves combined failure at the pile end and along the pile side. The upper part of the pile foundation exhibits significant load-induced deformation, showing flexible characteristics, while the lower part mainly undergoes rigid deformation. Verification results indicate that the design parameters of both types of foundations meet bearing capacity requirements. The results provide a theoretical basis and engineering reference for the design and stability analysis of bridge foundations in mountainous areas.

To address the data security risks arising from the explosive growth of railway passenger transport data, the core lies in achieving intelligent identification and dynamic protection of sensitive information. Then, an intelligent identification technology for sensitive data in railway passenger tickets based on data knowledge base is proposed. Firstly, a three-level knowledge base of "laws and regulations-industry standards-enterprise norms" is constructed. Secondly, combined with historical railway passenger ticket data, a multi-level intelligent identification algorithm for sensitive data is designed, thereby efficiently and accurately identifying sensitive information in multi-modal data. On this basis, the graph technology is finally introduced to construct a data asset and sensitive data lineage graph, and based on the topological relationship of data flow, the efficient propagation of sensitive information labels among related data nodes is achieved. The results show that the sensitive information identification efficiency of the proposed technology reaches about 217 000 messages per second in structured data processing, which is almost twice as high as the traditional solution. In unstructured data processing, through domain knowledge graphs injection, the F₁ value of sensitive entity recognition is increased to 91.24%, and the context misjudgment rate is reduced to 5.88%. The accuracy of text extraction and sensitive information recognition of multimedia images reaches 93.71%. This technology can significantly improve the accuracy and processing efficiency of sensitive data identification in railway passenger tickets.