Research purposes: Marine soft soil has unfavorable engineering properties characterized by"three highs and two lows". The shear characteristics of the interface between these soils and mixed cement piles are critical to the load transfer mechanism within the pile structure. In order to study the shear mechanical behaviour of the interface between marine soft soil and deep cement mixing(DCM)pile, an improved direct shear test equipment was used to analyze the shear evolution at the pile-soil interface, considering factors including varying normal stresses, water content, and resting time at the pile-soil interface. And a variable stiffness ideal elastic-plastic model for describing the shear mechanical properties of the marine soft soil-DCM pile interface was proposed.
Research conclusions: (1) The relationship between shear stress and shear displacement at the pile-soil interface undergoes three distinct phases—initial, transitional, and stable—and follows a hyperbolic curve. (2) With the increase of water content, the peak shear stress at the pile-soil interface decreases rapidly, while the peak shear displacement changes in a growing trend. (3) With the increase of normal stress and resting time, the interface peak shear stress increases. The increase is obvious in the resting time of 3-14 d, and the increment turns to slow down after 14-28 d. The peak interfacial shear displacement shows a decreasing trend. (4) Water content and resting time have significant effects on the pile-soil interfacial cohesion and internal friction angle. The interfacial cohesion and internal friction angle decrease with the increase of water content ranging from 8.8 kPa to 1.1 kPa and from 3.7° to 1.1°, respectively, and increase with the increase of resting time, ranging from 1.6 kPa to 5.5 kPa and from 2.3° to 3.2°, respectively, both sets of variations exhibit power-exponential function relationships. The growth of shear strength parameters is significant in the resting time of 3-14 d. (5) Based on the test results, the formula for calculating the interfacial shear stiffness of marine soft soil with different water content under different normal stresses is obtained. (6) The results of this study can provide a reference for the design and construction of DCM piles in marine soft strata.

Research purposes: To explore the impact of groundwater level fluctuations on the stability of the high-speed railway foundation in the mined-out areas, based on the engineering background of the Taijiao high-speed railway passing through the mined-out areas of the Dianshang coal mine, the indoor physical model tests were carried out. With a geometric similarity ratio of 1∶100, four working conditions were set, namely groundwater level at 15 m, 25 m, 35 m, and 45 m below the ground surface. The dynamic response characteristics—including dynamic stress, acceleration, and displacement—of the foundation in the mined-out areas under high-speed railway reciprocating dynamic loading were analyzed, and the evolution law of cumulative deformation of the foundation under different water levels was revealed.
Research conclusions: (1) With the rise of the groundwater table, the dynamic stress and acceleration of the foundation increase, and the dynamic response of the foundation exhibits an amplification effect. (2) The change of groundwater level will cause the compaction of the overlying rock and soil structure, resulting in the cumulative settlement of the foundation in the mined-out areas. (3) The evaluation of effective vibration velocity method shows that the rise of groundwater level will lead to the reduction of the dynamic stability of the foundation, and there is the risk of cumulative deformation exceeding the limit. (4) The research results provide valuable reference and guidance for the safe operation and maintenance of high-speed rail passing through the mined-out areas.

Research purposes: In complex field environments, the fastening torque of railway fastener systems often exhibits significant randomness,which results in a stochastic distribution of the longitudinal resistance of the fasteners. To investigate the influence of the randomness of fastening torque on the seismic response of the bridge-rail system, the WJ-8 type fastener was taken as an example. Based on existing experimental data, the fastening torque distribution was assumed to follow a normal distribution, denoted as X~N(μ,σ²). A formula for the longitudinal resistance of fasteners considering torque randomness was derived. An interaction model of a 7×32 m simply-supported bridge and the track was established to explore the influence of torque randomness on the seismic response of the bridge-rail system.
Research conclusions: (1) The variation amplitude in the seismic response of the bridge-rail system differs under various seismic excitations, but consistently increases with the standard deviation σ of the fastener torque. (2) The rail stress, rail displacement, and rail-girder relative displacement are more sensitive to the randomness of fastening torque, with maximum variation amplitudes of 8.2%, 6.6%, and 8.7%, respectively. (3) In contrast, the girder displacement, pier-top displacement, and internal forces in the pier are less influenced by the randomness of fastening torque, showing maximum variations of 2.0%, 1.9%, and 2.0%, respectively. (4) In the seismic design of ballastless tracks on bridges, considering the influence of fastening torque randomness, a safety factor of 1.1 can be applied to the rail stress, rail displacement, and rail-girder relative displacement. (5) The findings can provide a reference for the seismic design, operational safety, and maintenance of ballastless track structures on railway bridges.

Research purposes: Under continuous high temperature, a significant temperature gradient is prone to form within the slab track, which induces track slab warpage and interlayer separation, and weakens its bearing capacity and service life. In order to reveal the temperature distribution characteristics of CRTS Ⅲ slab track under continuous high temperature, a temperature field monitoring system was established based on a track base in East China to monitor the temperature field data of slab track. Based on the monitoring results, the spatio-temporal evolution law of the temperature field of slab track under high temperature environment was analyzed, and the temperature prediction model of CRTS Ⅲ slab track was established.
Research conclusions: (1) Under continuous high temperature, the temperature field of slab track changes in a daily cycle. The daily temperature difference of rail is the largest, while the daily temperature differences of track slab, self compacting concrete layer and base slab decrease successively. (2) Diurnal variations in the track slab are characterized by alternating positive and negative temperature gradients with marked vertical hysteresis.The ratio of the maximum positive value to the maximum negative value of the temperature gradient in the middle and at the bottom of the track slab ranges from 3.08 to 3.38. The duration of the negative gradient is 1.27 to 1.38 times that of the positive gradient. (3) Under the influence of solar radiation, the transverse temperature distribution of the slab track is asymmetrical. The transverse temperature differences of the track slab and the self-compacting concrete layer are small. However, the maximum transverse temperature difference of the base slab is 8.3 ℃. (4) Based on the monitoring data and statistical regression analysis, a temperature field prediction model of the CRTS Ⅲ slab track is established and validated with good performance. (5) The research results can provide a theoretical basis for the structural design, operation and maintenance of slab track in high temperature areas.

Research purposes: Due to the continuous increase in social transportation demand, the speed-up lines still play an indispensable role in China's transportation system. The increase in train running speed will intensify the interaction force and impact between the wheels and the rails, causing damage to the wheels and the rails, affecting the smoothness of the rails and thereby endangering the stability and safety of train operation. To study the vibration response laws of the track structure under the influence of different measured track irregularities on the speed-up lines, this paper,based on the on-site measured data of track irregularities in a certain section of the Jingguang line, carried out preprocessing such as eliminating trend terms, eliminating outliers, filtering, and determining the stability of the measured data, and used the principle of least squares to fit the power spectrum of track irregularities, proposing a fitting formula for track irregularities on speed-up lines. Based on the characteristics of the track structure of the speed-up lines, a dynamic finite element model of the vehicle-track coupling system was established, and the validity of the model was verified. The influence of track irregularities within different wavelength ranges on the vibration of the vehicle-track coupling system was explored from the time domain perspective.
Research conclusions: (1) The Welch method and the maximum entropy method are used to calculate the power spectrum of track irregularities in the speed-up lines, and the differences between the two methods are compared and analyzed, with the maximum entropy method effectively addressing the limitations of resolution and data length. (2) As the wavelength range increases, the vibration response of the track structure to track irregularities shows an overall decreasing trend, and track irregularities within the 1-30 m wavelength range have a significant impact on this response. (3) The larger the range of track irregularity excitation, the greater the impact on the wheel-rail force, and the larger the wavelength of track irregularities, the more obvious the impact on the wheel-rail force. (4) This study provides theoretical basis and technical reference for the track maintenance and repair, structure optimization, and traffic safety control of speed-up lines.

Research purposes: Thick steel plates are essential structural components in grand bridges, where through-thickness mechanical property gradients critically influence overall performance. The mechanisms underlying the distinct yield behaviors between surface and core regions in thick steel plates remain insufficiently understood. This study investigates a Q420qD grade bridge steel plate and employs layered tensile tests combined with multi-scale microstructural characterization to elucidate the micro-mechanisms underlying through-thickness variations in yielding behavior.
Research conclusions: (1) The surface layer of the experimental steel plate exhibits continuous yielding behavior, while the core layers demonstrate pronounced discontinuous yielding characteristics. (2) Microstructural analysis shows that the full-thickness structure is composed of granular bainite (GB) and polygonal ferrite (PF). From the surface to the core, the average grain size increases from 6.86 μm to 11.57 μm. Dislocation density shows a marked decrease from surface to 1/8 thickness before stabilizing. (3) The mechanistic analysis reveals a dislocation-grain size distribution synergistic mechanism underlying the yield behavior differences. (4) These results offer a theoretical foundation for the process optimization of thick steel plates and provide a reference for their appropriate application in grand bridge engineering.

Research purposes: Corrugated steel webs are the key components of corrugated steel web composite girder bridges. Initial imperfections can easily trigger premature buckling failure, which directly affects the overall load-bearing capacity and structural safety of bridges. To investigate the distribution characteristics of initial imperfections in corrugated steel webs, experimental measurements were conducted on three specimens taken from a real bridge project. Based on the test results, a method for determining the magnitude of initial imperfections was proposed. Furthermore, finite element models incorporating initial imperfections were established for nonlinear analysis, and the rationality of the proposed modeling approach was verified.
Research conclusions: (1) The initial deformation within a single sub-panel of the corrugated steel web is consistent in direction. Adjacent sub-panels along the length exhibit opposite deformation directions. The deformation amplitude of flat sub-panels is larger than that of inclined sub-panels, and the deformation shows a half-wave sinusoidal distribution along the web height. (2) The initial imperfections of corrugated steel webs in real bridges are much smaller than those of previous small-scale test specimens, and are close to the acceptance standard of 1/750 of the web height (h_w) specified in Corrugated Steel Webs for Composite Bridges (JT/T 784—2022). (3) Combined with test results and relevant specifications, a rational method for determining initial imperfections in numerical analysis is proposed: 1/200 of the flat panel width (a_w) for corrugated steel webs controlled by local buckling, and 1/750 of the web height (h_w) for those controlled by interactive and global buckling. (4) Initial imperfections reduce the shear bearing capacity of corrugated steel webs, and the adverse effect on the ultimate bearing capacity becomes more significant with the increase of imperfection amplitude. (5) This study reveals the distribution law of initial imperfections in corrugated steel webs and proposes a feasible method for determining initial imperfections in numerical simulations, which can support further research on the shear performance of such members.

Research purposes: Buckling restrained braces are a type of metal damper with superior performance, but they are currently less commonly used in the field of large-span bridges. The article focuses on a large-tonnage buckling restrained brace for bridge applications. A scaled model is designed and manufactured. By comparing finite element analysis and model experiments, key parameters were analyzed, including the material of the core component of the buckling restrained brace, the filling material of the restraining component, the restraining stiffness of the steel cylinder of the restraining component, the restraining gap between the core component and the restraining component, and the welding forming process of the core section of the core component. Based on these analyses, the value rules for key structural parameters of bridge buckling restrained brace were studied.
Research conclusions: (1) The material of the buckling restrained brace core component should preferably be LY225 material, which has stronger energy dissipation capacity and plastic deformation capacity. (2) UHPC should be selected as the filling material for restraining components to improve their filling density and durability. (3) The wall thickness of the steel cylinder of the restraining component only needs to meet the requirements of the welding structure to meet the restraining stiffness requirements. (4) The restraining gap should be set to 1 mm to ensure good tensile and compressive uniformity of the buckling restrained brace. (5) The core section of the core component should be formed using intermittent spot welding technology to achieve better fatigue performance. (6) The research findings can be applied to buckling restrained braces in fields such as multi-pylon bridge structures and large bridge piers.

Research purposes: To address the issues of high subjectivity and low efficiency in traditional rock mass classification methods that rely heavily on manual identification, this paper proposed an intelligent rock mass classification method for drill-and-blast tunnels that integrates drilling parameters with tunnel face images. Furthermore, feature importance evaluation was conducted to reveal the contribution of different features to the classification results.
Research conclusions: (1) A complete process for the acquisition, preprocessing, and alignment of multi-source geological information was constructed, integrating drilling parameters and images. (2) A 104-dimensional image and drilling parameter fusion feature system was proposed, and a lightweight residual grading model was designed, achieving an overall accuracy of 91.11% on the predicted samples. (3) A three-model joint feature importance evaluation using random forest, extreme gradient boosting, and lightweight gradient boosting machine was introduced to analyze the importance of different features to the rock grading task. (4) The research results can provide reliable data support for real-time rock grading determination and support optimization in drill-and-blast tunnel construction.

Research purposes: With the rapid development of urban underground space utilization and transportation network construction in China, the application of super-large diameter (≥14 m) earth pressure balance (EPB) shields is continuously expanding. Compared with small-diameter shields, the difficulty of face stability control for super-large diameter shields increases significantly, which has become one of the key technical bottlenecks restricting safe and efficient tunneling. This paper systematically reviews the current research status of face stability in super-large diameter EPB shield tunnels, summarizes the applicability and limitations of existing theoretical methods and technical approaches, analyzes the multi-factor coupling mechanisms and potential risk sources affecting stability, and evaluates the characteristics and shortcomings of current control measures. The study aims to provide theoretical reference and technical support for revealing the instability mechanisms of super-large diameter shield faces, optimizing construction control strategies, and promoting intelligent and refined development.
Research conclusions: (1) Regarding failure mechanism research, traditional theoretical methods such as limit analysis, limit equilibrium and silo theory have been established, along with numerical simulation techniques including finite element method, finite difference method and discrete element method. Model tests based on active unloading and dynamic excavation processes have also been conducted, revealing the fundamental laws of face instability. However, existing methods still have obvious limitations in characterizing soil heterogeneity, dynamic construction processes and multi-field coupling effects. (2) Regarding influencing factors, multiple factors including soil physical and mechanical parameters, structural characteristics, groundwater effects, construction parameters, as well as shield diameter and burial depth have coupled influences on face stability. Among them, the increase in shield diameter significantly alters stress transfer paths and support pressure distribution patterns, imposing higher requirements for stability control. (3) Regarding risk analysis, adverse geological conditions such as soft soil, sandy soil and composite strata, together with construction and management factors including unbalanced chamber pressure, mismatched advance speed and cutterhead rotation speed, groundwater activities, synchronous grouting defects and inadequate monitoring, are the main risk sources inducing face instability disasters. (4) Regarding control measures, stability control methods represented by intelligent chamber pressure regulation, soil conditioning technology and ground reinforcement technology have achieved remarkable results. Intelligent soil conditioning, coordinated coupling of excavation parameters and closed-loop control of intelligent monitoring are becoming important directions for future technological development. (5) The research results can provide theoretical support for face stability analysis and control of super-large diameter EPB shield tunnels, and offer technical references for the design, construction and risk management of related engineering projects.

Research purposes: The deterioration of tunnel surrounding rock in the seasonally frozen area is a progressive process due to freeze-thaw damage, and the existing theoretical models rarely accurately simulate the entire process of progressive freeze-thaw deformation of rocks under the coupling condition of freeze-thaw and loading because of the complex deformation mechanism of the surrounding rock. Based on this,uniaxial compression tests were conducted on the tunnel surrounding rock under freeze-thaw action, taking the Yushuchuan Tunnel as an example. The damage and deterioration characteristics of granite after 600 freeze-thaw cycles were explored. Then, a mechanical constitutive model simulating the whole deformation process of rock was established considering the coupling effect of freeze-thaw and loading, which was validated and evaluated against experimental data.
Research conclusions: (1) The linear elastic modulus of the surrounding rock decreased by 67.413% after 600 freeze-thaw cycles, and the initial compaction stage gradually became more prominent. (2) The error relative coefficient between the established mechanical constitutive model and the experimental curve was greater than 0.95. The simulation accuracy of the proposed mechanical constitutive model was superior to that of existing theoretical models, and this model can be applied to other types of freeze-thawed rocks with strong generalizability. (3) The research findings offer important theoretical guidance for the construction, design, operation and maintenance of tunnels in the seasonally frozen regions.

Research purposes: CTCS-2 train control system has been successfully applied in China's high-speed railways for many years, accumulating rich technical and operational experience, and has become an important choice for overseas railway signal systems. However, there are certain differences between some overseas railway lines and China's high-speed railways in terms of design specifications, preventing the CTCS-2 train control system from fully satisfying operational requirements. This article identifies the main differences between some overseas railways and China's high-speed railways, analyzes the problems faced by CTCS-2 train control system as well as the processing methods of other systems, and proposes an adaptive solution tailored to the characteristics of the CTCS-2 train control system for application in overseas railways.
Research conclusions: (1) The minimum speed limit for all lateral routes within the protection range of signals is added to balise message, which solves the safety and efficiency issues existing in stations with turnouts below No. 12. (2) Identifiers for tracks where accurate stopping at the marker is difficult are added to balise message, and on-board equipment implements release speed monitoring based on these identifiers, thereby resolving the problem of difficulty in stopping precisely at the marker on arrival-departure tracks. (3) Controlled balise groups in stations adopt backup settings, and Driver-Machine Interface (DMI) and Balise Transmission Module (BTM) of on-board equipment adopt dual redundancy configurations, which further improve the availability of CTCS-2 train control system. (4) The research results can provide some references for the application of domestic CTCS-2 train control system and the "going global" strategy of China's high-speed railways.

Research purposes: As the core equipment of China's CTCS-0 train control system for general speed railway, the Train Monitoring Device (LKJ) has played a vital role in ensuring operational safety and improving transportation efficiency. However, due to the limitations of railway infrastructure conditions at its initial design stage and the excessive reliance on manual intervention, LKJ can no longer meet the demands of China's railway development. This paper analyzes the manual operation issues in LKJ during station entry and exit processes—specifically, siding track number selection, branch line number selection, and departure alignment—and proposes an optimized station entry/exit function scheme based on STP for CTCS-0 train control system in general speed railway.
Research conclusions: (1) The research focuses on locomotive positioning technology based on the fusion of multi-source information—namely, BeiDou satellite navigation, LKJ operation data, and interlocking information—to achieve precise tracking of locomotive location. (2) Using the existing interlocking data security interface and data transmission radio of STP, a safety protocol is employed for train-ground data exchange to enable real-time interlocking information onboard. (3) It is the first time that STP station interlocking information is utilized to automatically generate LKJ control information, replacing manual operations related to the station entry and exit processes, thereby providing an integrated solution. (4) The solution keeps the LKJ hardware unchanged, requires only a simple upgrade to STP, and continues to use STP management specifications, thereby reducing costs, facilitating subsequent maintenance and management, and providing a reference for the transformation of train control systems on general speed railways.

Research purposes: Accurately calculating the dropper length of the fully compensated elastic catenary suspension in the overhead contact system (OCS) of high-speed railways is one of the core technologies for high-quality construction of OCS projects and ensuring the pantograph-catenary current collection performance under high-speed operation conditions. To address the issues of cumbersome iterative calculations and convergence dependent on initial values existing in the accurate calculation of OCS dropper length by using simulation or nonlinear methods, a dropper length calculation model based on the exact element method for cable structure analysis is constructed, which can realize the one-time accurate solution of dropper length.
Research conclusions: (1) By integrating loads and constraints, introducing a messenger wire density correction function to compensate linearization errors, a discrete model of the dropper system with polynomial displacement distribution is constructed, linearized nodal element relations are derived, and a universal calculation model for fully compensated straight dropper lengths is established.(2) Compared with the reference model of working elastic catenary spans in European standards, the proposed method achieves complete consistency with the standard values, with zero absolute and relative errors. Compared with measured line data of elastic catenary anchor joints in European standards, the maximum relative errors of the proposed method are 0.102% for working elastic catenary spans and 0.060% for transitional spans (combining elastic and simple catenaries), verifying the accuracy and engineering applicability of the method.(3) Considering the influence of nonlinear characteristics of messenger wires on dropper lengths, the proposed method adapts to engineering scenarios such as large spans and complex loads, and can be directly compatible with the digital design platform of OCS, providing technical support for intelligent construction and safe operation & maintenance of high-speed railway OCS projects.

Research purposes: The traditional power flow algorithm relies on high-performance computing equipment to meet the requirements of online state assessment. An efficient power flow algorithm enables timely evaluation of power supply capacity under strongly stochastic train load conditions, providing technical support for the autonomous optimization of train operations. This paper proposes a hierarchical fast power flow algorithm applicable to multiple time scales for traction power supply systems, which supports parallel computing across multiple CPU cores to improve hardware utilization efficiency. Based on the wiring principles of traction transformers, a decoupling method for the two-side feeding sections is proposed, transforming the overall solution of a large-scale matrix into the iterative solution of smaller matrices. Furthermore, to address the poor adaptability of the traditional chasing method under special conditions, specific improvements are proposed. Based on the above, a modular organization scheme for power flow calculation of the entire line is developed, enhancing the flexibility and rapidity of the online system assessment. Finally, the proposed algorithm is compared and validated against a MATLAB simulation model and traditional algorithms.
Research conclusions: (1) The proposed algorithm can effectively realize the online state evaluation of the system while ensuring accuracy and convergence. (2) Compared with the traditional algorithm, the proposed algorithm has more significant solving efficiency. When the analysis time window is 60 seconds, the computational efficiency is increased by 37.73%. (3) The proposed algorithm provides a theoretical basis for online evaluation and analysis of electrified railway.

Research purposes: Multi-network integration policies in rail transportation serve as an important means for the government to promote coordinated regional development. Identifying the content characteristics of these policies and revealing their design patterns can provide theoretical support for improving the multi-network integration policy system and enhancing the scientific basis of policy formulation. This study combined natural language processing technology with the PMC index model to construct the PMC-NLP index model, systematically analyzing and quantitatively evaluating the texts of China's rail transit multi-network integration policies. The aim is to provide data support and decision-making basis for the optimization of rail transportation policies.
Research conclusions: (1) The content of multi-network integration policies in rail transportation can be categorized into three primary themes: regional transportation integration, rail infrastructure development, and multi-network integration to guide urban development. (2) The national-level policies exhibit a phased evolutionary process characterized by framework design, domain expansion, in-depth advancement, and fine-tuning adjustments. (3) Provincial and municipal policies for rail transit multi-network integration show significant spatial clustering, with both the volume and quality of policy demonstrating convergence in geographically adjacent regions. (4) The results of this study provide a reference for the precise optimization of rail transportation multi-network integration policies.

Research purposes: Against the strategic background of metropolitan area integration and infrastructure interconnection in the Guangdong-Hong Kong-Macao Greater Bay Area, Dongguan, as a core node city connecting Guangzhou and Shenzhen, has its rail transit development lagging behind the needs of regional economic development. It has prominent problems such as insufficient network density, low hub level, and poor multimodal transport connection, showing a significant gap with similar domestic cities. Drawing on the experience of rail transit development at home and abroad, this study focused on the development bottlenecks of Dongguan's rail transit, systematically sorted out the current development situation and core problems, and put forward targeted optimization countermeasures to solve the development difficulties and support the high-quality and sustainable development of Dongguan's rail transit.
Research conclusions: (1) Dongguan's rail transit has formed a multi-level network structure, and its connectivity with cities in the GBA has improved, but it still remains in the stage of "initial framework formation and pending function optimization", with the core shortcomings of insufficient hub level and lagging urban rail construction.(2) Optimizing the hub layout is the key. It is necessary to introduce the second Guangzhou-Shenzhen High-speed Railway and build a main hub in the central urban area to realize the transformation to a hub city. (3) This study proposes to construct a "two vertical and two horizontal" intercity framework and promote the first phase construction of Urban Rail Transit Line 3 to realize networked rail operation and improve the current situation of insufficient passenger flow. (4) Measures such as legalization of TOD planning and establishment of a revenue feedback mechanism can be adopted to solve the problems of fund raising and project promotion, and promote the coordinated development of rail transit and land use. (5) The results of this study can be applied to the planning and construction of rail transit in Dongguan, providing practical examples and references for similar node cities in the GBA.

Research purposes: Corrosion fatigue poses severe threats to the service safety of in-service steel bridges, making the development of coating-free weathering steel bridges a highly promising direction.To investigate the fatigue performance of weathering high-toughness high-strength steel Q420qENH and its welded joints in corrosive environments, experimental studies were conducted on specimens including unwelded base metal and two types of commonly used welded joints. The research encompassed both corrosion tests and post-corrosion fatigue performance evaluations. The obtained fatigue test data after corrosion exposure were subsequently fitted and analyzed using S-N curve methodology.
Research conclusions: (1) The welding impact on Q420qENH weathering steel during actual application cannot be underestimated. For highway coating-free weathering steel bridge design, it is recommended to adopt Fatigue Detail Category 100 for butt joints without weld reinforcement grinding, and Category 125 for base metal specimens. For railway coating-free weathering steel bridge design, it is suggested to reduce the fatigue allowable stress range category to Class Ⅶ for butt joints and Class Ⅲ for base metal specimens. (2) When using cruciform welding, the outside-to-inside welding sequence is not recommended. (3) Q420qENH components subjected to high-stress regions require periodic maintenance or localized coating treatment prior to service. (4)The results of this study can provide data support and technical reference for the theoretical research and fatigue performance design of coating-free weathering steel bridges in the marine atmospheric environment.

Research purposes: With the accelerated development of global high-speed rail construction and evolving economic conditions, the international high-speed rail EPC (Engineering, Procurement and Construction) projects face complex supply chain risk scenarios and a lack of adaptive strategies. Thus, this study identifies supply chain risk management strategies via the LDA (Latent Dirichlet Allocation) topic model, constructs a strategy framework by integrating dynamic capability theory, and uses fsQCA (Fuzzy-set Qualitative Comparative Analysis) to analyze strategy combinations under different risk scenarios.
Research conclusions: (1) LDA analysis identifies five supply chain risk management strategies: dynamic perception, digital-driven, standardization, collaborative planning and resource reconfiguration. Integrating these strategies with dynamic capability theory, this research develops a closed-loop risk management framework of "risk perception-rapid response-continuous adaptation". (2) The fsQCA results show that under strong internal and weak external risks, general contractors should prioritize the combination of collaborative planning and resource reconfiguration; under weak internal and strong external risks, the focus should shift to the combination of dynamic perception, digital-driven, and standardization; under strong internal and external risks, they should adopt the combination of all five strategies. (3) This study provides practical guidance for general contractors of international high-speed rail EPC projects to implement supply chain risk management and ensure the safe and stable operation of projects.