Railway turnout (RT) equipment is an important component of railway signaling systems, and its reliable operation is essential for railway safety. To improve maintenance efficiency and ensure equipment reliability, this study proposes a fault prediction method based on RT action current curves collected by the microcomputer monitoring system, combining Sparse Autoencoder (SAE), Squeeze and Excitation Networks (SE), and Bidirectional Long Short-Term Memory (Bi-LSTM). Based on the 7 fault and 1 normal modes, the pre-fault degradation current data of the ZD6 RT were collected in chronological order, and the labels were assigned according to the fault types subsequently confirmed by field maintenance records. Secondly, an RT fault prediction model based on SAE fusion SE and Bi-LSTM was constructed. Because hyperparameters directly affect the performance of the prediction model, Adam algorithm is used to optimize the model parameters, and the Bayesian Optimization (BO) is used to optimize the hyperparameters of the RT fault prediction model. The proposed model is designed to learn the stage-dependent characteristics and sequential relationships embedded in each complete turnout action current curve and to predict the corresponding fault state and fault type. Finally, the experiment proved that this method can accurately predict the fault status of the RT, so that on-site maintenance personnel can predict the fault in advance and take maintenance measures, improve the maintenance efficiency of the RT and ensure the safe operation of the train.
Future high-speed railway systems rely heavily on secure and efficient wireless communication for train operation, passenger services, positioning, and safety-related functions. Antenna systems play a critical role in establishing these communication links; however, their design and implementation in high-speed rail environments present unique challenges compared to conventional vehicular and terrestrial platforms. Factors such as high mobility, large metallic train structures, aerodynamic constraints, harsh propagation conditions, and stringent reliability requirements impose significant demands on antenna integration and performance. This paper provides a comprehensive overview of antenna systems for high-speed railways from both engineering and deployment perspectives. Particular emphasis is placed on implementation challenges, including train-mounted and trackside antennas, with detailed discussion on practical integration aspects such as platform effects, radome design, electromagnetic compatibility, and maintenance. Key propagation scenarios in high-speed railway operations—such as open track, tunnels, viaducts, and stations—are analyzed in terms of their impact on antenna radiation behavior and overall system performance. Furthermore, state-of-the-art antenna designs are systematically classified based on radiation characteristics, frequency bands, and functional roles, ranging from omnidirectional and multi-antenna systems to dual-band and emerging high-frequency solutions. Beyond performance metrics, this review also examines validation methodologies, comparing simulation-based studies with real-world measurements from operational railway environments. Finally, future challenges and research directions are outlined in the context of next-generation railway communication standards and emerging technologies. Overall, this review aims to bridge the gap between theory and practice by providing practical insights for researchers and railway engineers involved in the design, integration, and deployment of antenna systems for high-speed railways.
Virtual Coupling (VC) aims to reduce inter-train following distances by relying on relative braking and vehicle-to-vehicle information exchange. Although VC train-following control is commonly addressed using model-based approaches such as Model Predictive Control, maintaining safe operation under communication delay, adhesion degradation, and positioning uncertainty remains challenging. This study proposes a safety-shielded Deep Reinforcement Learning (DRL) architecture for VC train following. A Soft Actor-Critic policy generates continuous nominal control commands for speed tracking, headway regulation, comfort, and actuation effort, while a braking-distance-based safety shield supervises the hard minimum-distance constraint through a one-step action-projection mechanism whenever the projection remains feasible. Therefore, the DRL policy is not treated as a standalone safety-certifying controller; safety is provided by the combined policy–shield architecture. The proposed method is evaluated under nominal operation, 10% adhesion loss, a communication-delay sweep (τ ∈ {0.0, 0.5, 1.0, 2.0, 5.0, 10.0} s), and a combined-uncertainty stress test involving delay, adhesion loss, and position noise. All scenarios are assessed over 50 randomized seeds using worst-case and distributional statistics. Additional diagnostics, including safety-shield activation, raw-action feasibility, correction magnitude, and shield-off evaluation, are used to distinguish the role of the learned policy from that of the safety shield. The results show safe operation under nominal conditions, 10% adhesion loss, and communication delays up to 2.0s in the considered setting. In contrast, delays of 5.0s and 10.0s and the combined-uncertainty stress test lead to safety violations and early termination, identifying the feasibility boundary of the considered VC formulation. Overall, the study provides a reproducible safety-shielded DRL baseline for VC train following and clarifies the delay regimes under which the proposed architecture remains applicable.
This study investigates the heat of hydration and internal temperature field of concrete in the precast construction stage of the 32m box girder for high-speed rail. The aim is to guide early thermal insulation and maintenance practices for the box girder, aiming to prevent temperature cracks or internal concrete damage resulting from significant temperature differentials or rapid cooling rates. Through testing, the adiabatic temperature rise of concrete was measured. By combining numerical analysis and actual measurements, this study analyzed the variations in the temperature field during the precast stage of the box girder and fitted an exponential function to represent the internal temperature variations over time. Results indicate that the internal temperature of the box girder concrete sharply increases from 5 to 15hours after casting, peaking between 15 to 20hours at around 61.0-61.3°C with a maximum rise of approximately 41.3°C. Furthermore, it was observed that the internal heat dissipation of concrete in the web of the box girder is inefficient, with a slow cooling rate averaging about 0.209°C/h. To mitigate temperature differentials, it is recommended to provide adequate insulation, especially during extreme cooling conditions or winter construction. This study highlights the significance of the 20-60hours period post-casting as a critical phase for curing, necessitating appropriate measures to prevent temperature-related cracks due to internal/external temperature disparities. Finally, the research findings demonstrate that the internal temperature variations in various sections of the box girder adhere to an exponential function, with MATLAB-developed models offering improved predictions for guiding construction activities and thermal insulation maintenance. These research results offer a scientific foundation for enhancing precast construction methods and maintenance practices, ultimately reducing the risk of early concrete cracking.
To address tracking difficulties in weak-texture regions and the limited robustness of traditional vision-based structural vibration measurement methods to visual interference, this paper proposes an approach combining prior physical assumptions with deep learning-based point tracking. First, CoTracker3, a dense point tracking model based on the Transformer architecture, is introduced to construct virtual displacement sensors, which effectively overcomes tracking failures caused by weak-texture features in complex environments. Second, based on the physical prior that local regions of a bridge structure can be approximated as rigid bodies, a point tracking mask generation algorithm based on affine consistency is proposed to identify unreliable trajectories caused by visual interference. Finally, leveraging the physical low-rank property of the vibration observation matrix in structural dynamics, an alternating least squares method with mask constraints is employed for matrix completion, achieving reconstruction of the displacement data affected by visual interference. Free-vibration experiments on a cable-stayed bridge model demonstrate that the proposed method can identify structural displacements with high precision under both the normal condition and four simulated visual-interference conditions lasting 5s, including Gaussian-noise occlusion, vehicle occlusion, reflection interference, and severe shadows. In addition, systematic comparisons with representative vision-based methods are conducted under the normal condition and multiple visual-interference conditions. These comparisons quantitatively evaluate displacement accuracy, grid-point consistency, and data-recovery performance, thereby demonstrating the comprehensive performance of the proposed framework.
This study investigates the economic impacts of high-speed rail (HSR) with a focus on their temporal heterogeneity and underlying mechanisms. Employing a spatial difference-in-differences (spatial DID) model that accounts for staggered city-level openings, we find that HSR generates significant direct economic benefits beginning in the opening year, with effects strengthening over time. Spatial spillovers to neighboring cities, however, emerge after a three-year delay but remain persistent. Notably, HSR particularly stimulates economic development in less-developed cities, contributing to a reduction in regional disparities. A dual-channel mechanism is identified: HSR promotes local growth by enhancing factor mobility (capital and labor), yet simultaneously induces industry siphoning from less-developed to more developed areas. These findings underscore HSR's role in fostering equitable regional development through both immediate and lagged spillover effects, and suggest that policymakers should leverage delayed spillovers and implement coordinated industrial policies to mitigate adverse siphoning effects during the initial phase.
To address the underutilization of high-speed rail (HSR) infrastructure alongside growing large-scale freight demand, this paper proposes an integrated optimization framework for HSR passenger-freight mixed operations. The objective is to minimize total system costs and passenger schedule disruptions while strategically inserting dedicated freight trains to satisfy designated large-scale freight demand. Adopting a modular transport organization approach, origin-destination (OD) demands are consolidated into transport blocks. A multi-constraint scheduling model is developed, incorporating constraints related to freight train composition, loading/unloading operations, train capacity, and transfer coordination. The model introduces flexible running time allowances and restricted overtaking rules, permitting freight trains to receive priority at stations. Through joint optimization of block selection and schedule adjustment, the framework demonstrates robustness by flexibly combining strategies, including fine-tuning passenger schedules, adjusting freight dwell times, and arranging strategic yielding. A case study on the Nanjing-Hangzhou HSR line validates the model’s effectiveness, with sensitivity analysis offering quantitative insights for parameter tuning, such as increasing passenger train intervals and freight train speeds to alleviate capacity tension. This study provides both methodological and practical support for efficient HSR mixed-traffic organization and scheduling.
Deep neural network-based intelligent methods have brought new breakthroughs to fault diagnosis. However, their ‘black box’ nature hinders their implementation and widespread adoption in safety-critical equipment such as railway tracks. The Explainable Convolutional Neural Networks Fuzzy Inference System (X-DCFR) combines the advantages of high-dimensional feature extraction from CNNs and the interpretability of FIS rules, seeking a balance between the ability to express deep-level fault relationships and the interpretability of fault diagnosis mechanisms. A customized training dataset was constructed using a track circuit data simulation system, and the model was trained through a staged strategy consisting of ‘CNN pre-training → independent FIS layer training → joint fine-tuning of the entire network’. Simulation results show that X-DCFR outperforms other algorithms in identifying eight typical fault states in railway tracks, achieving an average accuracy of 96.1%. Furthermore, the model’s excellent generalization ability and robustness were verified on actual equipment, providing an engineering reference for the practical application of deep neural network-based fault diagnosis in railway tracks.
In heterogeneous railway cooperative transmission systems, accurate link-state prediction (e.g., QoS metrics like throughput and latency) is crucial for multipath scheduling. However, collecting large-scale real communication data in operational railway systems is costly, while conventional tool-based simulation pipelines struggle to reproduce the complex coupling between network evolution and link-state dynamics. This results in a scarcity of effective data for intelligent scheduling and performance prediction. To address this issue, this study proposes Rail-GAN, a domain-aware adaptation of the TimeGAN sequence-generation framework for heterogeneous railway cooperative transmission. Rail-GAN introduces QoS-specific preprocessing, full-dimensional moment matching, engineering consistency post-filtering, and source-to-target fine-tuning to generate multivariate QoS sequences under severe target-domain data scarcity. The generated data are evaluated from both statistical fidelity and downstream practical utility perspectives. Experimental results show that Rail-GAN achieves substantially better discriminative and predictive scores than the original TimeGAN baseline. In downstream LSTM-based prediction tasks, augmenting a scarce target-domain dataset with Rail-GAN synthetic samples reduces the MAE for TCP rate and packet loss rate by 31.19% and 26.61%, respectively, while RTT exhibits a mild 4.20% degradation. This result suggests that the current smoothing and moment-matching design attenuates rare RTT bursts, which remains an important limitation and motivates future burst-aware generative modeling. Overall, Rail-GAN provides a practical synthetic-data foundation for intelligent operation, congestion control, and resource allocation in highly dynamic railway communication networks.
As vehicular networks evolve from closed architectures to open cloud-edge-end computing paradigms, traditional perimeter-based security defenses are becoming increasingly ineffective. The dynamic coupling and decoupling of trains, combined with the stringent deterministic latency requirements of Time-Sensitive Networking (TSN), render conventional Public Key Infrastructure (PKI) and centralized authentication mechanisms unsuitable. To address these challenges, this paper proposes a lightweight zero-trust architecture for dynamic vehicular computing networks. First, a decentralized identity (DID) management framework rooted in Physical Unclonable Functions (PUF) is established, preventing identity forgery and single-point failures. Second, to handle the authentication storm during train coupling, a reputation-based consensus algorithm (RE-PBFT) is introduced. It selects high-reputation nodes for committee voting, reducing communication complexity from O(N2) to O(k2). Third, for cross-domain access control, a service-oriented Capability Verifiable Credential (VC) mechanism is designed and integrated with a zero-interaction verification protocol over TSN frames. By encapsulating a full VC in the initial handshake packet and an 8-byte session hash in subsequent packets, this protocol reduces packet fragmentation and preserves the worst-case delay (WCD) boundaries of underlying control flows. Simulation results demonstrate that the PUF key recovery rate exceeds 99.9% under electromagnetic noise, while the RE-PBFT algorithm bounds consensus latency to milliseconds as network scale expands. The proposed architecture secures future autonomous train operations with low latency.
High-speed railways constitute a critical component of modern transportation infrastructure. However, the longitudinal rail force, influenced by cyclic train loads and environmental conditions, poses considerable safety risks if not accurately monitored. Traditional longitudinal rail force detection techniques primarily rely on physical testing and periodic manual inspections, which significantly limit the potential for real-time and continuous monitoring. To address these limitations, this study introduces an innovative approach employing Large Language Models to predict longitudinal rail force based on historical monitoring data. The proposed method is validated through extensive long-term field monitoring of longitudinal rail force on high-speed railway lines, thereby confirming its practical applicability. In contrast to conventional time series forecasting large language models, the proposed method evaluates the prompt-free architecture. When applied to real-world longitudinal rail force data from the Beijing-Shanghai High-Speed Railway, the model achieves an average coefficient of determination (R2) of 0.932 and a root mean square error of 3.537 kN, outperforming traditional deep learning models. Furthermore, the model exhibits strong robustness under conditions of intermittent data loss. The proposed framework is seamlessly integrated into a localized intelligent system using Langchain-Chatchat, enabling expert-level recommendations based on domain-specific documentation. Overall, this study presents a practical, efficient, and scalable solution for intelligent railway monitoring, offering an advancement toward safer and more intelligent high-speed railway operations.
This study presents an innovative decision-support framework for selecting optimal railway track superstructures (ballasted vs. ballast-less) by integrating technical feasibility assessment with life-cycle cost analysis. In this matter Trapezoidal Fuzzy TOPSIS-CRITIC (TraF TOPSIS-CRITIC) methodology was developed that addresses two critical gaps in existing approaches: (1) systematic handling of data uncertainty through fuzzy logic and confidence intervals, and (2) objective criteria weighting to minimize selection bias. The framework evaluates six key technical parameters (speed, traffic load, tunnel/bridge length, settlement, and groundwater) alongside 100-year life-cycle costs. Applied to a 24km Iranian high-speed rail case study, the model demonstrates strong preference for ballast-less tracks in bridge/tunnel-intensive sections, and ballasted solutions favorable in limiting geotechnical conditions (groundwater <1.5m, differential settlement >1mm).Sensitivity analyses reveal economic selection thresholds dependent on inflation-discount rate relationships, with validation against IRS 70727 benchmarks confirming <5% deviation. Key innovations include the first successful implementation of TraF TOPSIS-CRITIC in rail infrastructure. For infrastructure managers, this research provides a quantifiable, bias-free decision tool particularly valuable for networks transitioning to higher operational speeds, while establishing new best practices for holistic track system evaluation that balance technical constraints with long-term economic performance.
Low-temperature superconducting Maglev trains are considered a promising candidate for future transportation due to their superior levitation height and robust self-stability. The integrated Propulsion, Levitation, and Guidance (PLG) system not only achieves multi-function integration but also significantly enhances the system's cost-effectiveness. However, owing to the structure of the PLG system, the induced electromotive force contains significant 5th and 7th order harmonics. Consequently, 6th-order thrust ripples occur during the propulsion process. To address this, this paper establishes a thrust ripple model considering these harmonics and proposes a control strategy based on active harmonic current injection within the current loop to generate compensating thrust and suppress the ripples. Finally, the proposed control strategy was validated through MATLAB-Simulink simulations and Real-Time Closed-Loop Validation on the StarSim real time platform. The results demonstrate that the proposed strategy can effectively suppress the propulsion thrust ripples of the PLG system.
To mitigate the progressive interfacial deterioration between the track slab and the self-compacting concrete (SCC) under complex service environments in high-speed railways, a material strategy employing a gradient polymer distribution within the SCC was proposed and implemented. A finite element model (FEM) of the slab track system featuring this gradient polymer-modified SCC was developed to analyze its mechanical performance and interfacial damage under various conditions, including temperature gradient, train load, and interfacial initial debonding. Results indicated that the maximum vertical displacement of track slab decreased with the increasing polymer dosage under high temperature gradients. The gradient distribution of polymer could significantly reduce the peak stresses of SCC with an appropriate polymer dosage across various conditions, and the maximum interface damage initiation factors under temperature gradient and train load were also significantly reduced. Additionally, as the initial debonding existed, the original bonding interface had experienced damage without polymer incorporation; however, it had not yet experienced damage and its maximum damage initiation factor was far smaller than 1 with gradient polymer-modified SCC, indicating that the damage propagation progress could be effectively inhibited as initial debonding existed. Therefore, the stresses of track structure could be reduced and the interfacial damage could also be delayed with an appropriate polymer dosage.
Critical welded structures in rail vehicles are subjected to prolonged alternating loads, making structural fatigue failure a primary safety concern. This study develops a quantitative crack diagnosis technology based on piezoelectric intelligent layer sensing. Through integrated simulation of weld structures and piezoelectric sensing, we investigate the coupling mechanism between piezoelectric wave signals and crack propagation, and design a comprehensive system architecture for real-time crack monitoring. Ground validation tests were conducted on typical welded specimens with incrementally introduced cracks. The combined simulation and experimental approach demonstrates the feasibility of quantitative diagnosis technology for weld structures, providing a robust framework for enhancing railway safety.
Aiming to jointly build the Belt and Road Initiative with high quality and to better advance the construction of overseas railway projects, this paper aims at the reality of overseas railway projects in areas with fragile supply chains, such as lack of overseas railway resources, complex mountainous terrain, and harsh transportation environment. It studies the location selection of logistics bases, and provides an improvement path for the supply chain resilience of overseas railway projects from the perspective of optimizing the layout of logistics bases. Based on extensive field investigation and literature review, the influencing factors of logistics base location are identified and an evaluation index system is established. The Fuzzy Analytic Hierarchy Process (FAHP) is employed, introducing fuzzy numbers at the weighting calculation stage to capture the uncertainty in expert judgments. Within a unified fuzzy analytical framework, both the weights of evaluation indices and the scores of alternative bases are calculated, and defuzzification is applied to obtain comparable and ranked comprehensive scores. Subsequently, considering constraints such as transportation demand, transshipment distance, and design capacity of each base, the FAHP scores are integrated into a quantitative optimization model together with objectives related to freight cost and investment cost, to produce the final optimal layout. The proposed method is validated using the overseas section of the China–Laos Railway, and the results match the actual logistics base locations, demonstrating that the approach is both scientifically rigorous and practically applicable. This research provides reference for the supply chain resilience construction and material support of the“the Belt and Road”railway project in the future overseas supply chain vulnerable areas.
The stress time-history at the rib-to-deck (RTD) connections of orthotropic steel bridge decks (OSBDs) is crucial for the near-real-time safety assessment and early warning of highway bridge structures. To improve the accuracy and efficiency of stress prediction, this study develops a hybrid Temporal Convolutional Network – Long Short-Term Memory (TCN-LSTM) model. The training dataset is constructed primarily from finite element simulations of stress responses, supplemented by real-world vehicle and temperature data collected via a Weigh-in-Motion (WIM) system and temperature sensors within the structural health monitoring framework of the Nanxi Yangtze River Bridge. Utilizing vehicle volume, speed, axle weight, and ambient temperature data collected and statistically processed by the WIM system within the structural health monitoring framework of the Nanxi Yangtze River Bridge, combined with finite element model simulations, a time-series training dataset was constructed. The TCN-LSTM hybrid architecture captures both local features and long-term dependencies within the time-series data. The model's performance was rigorously compared against five established machine learning models Backpropagation Neural Network (BP), Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), Radial Basis Function Neural Network (RBF), and Random Forest (RF) using four key statistical evaluation metrics: Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and the Coefficient of Determination (R²). Results demonstrate that the TCN-LSTM model achieves the most accurate and stable predictions for stress time-history. Compared to the best-performing traditional neural network model, it significantly reduces MSE, RMSE, and MAE by 62.9%, 38.4%, and 62.1%, respectively, while improving R² by 6.4%. The stress time-series data used for model training and validation are predominantly generated through finite element simulations, incorporating realistic vehicle loading and temperature conditions derived from field measurements. The proposed TCN-LSTM model provides an efficient tool for stress prediction in OSBDs, significantly reducing computational and experimental costs.
Railway infrastructure demands continuous monitoring to ensure operational safety and prevent catastrophic failures. Conventional inspection approaches, including manual visual examination and non-destructive testing, exhibit constraints in scalability, processing speed, and automation capability. This investigation presents a rigorous evaluation of three advanced deep learning frameworks—YOLOv11, YOLO-NAS, and Roboflow 3.0 (Robolow’s default model) on roboflow platform—for detecting multiple rail defect categories, with particular attention to addressing class imbalance challenges. A real-world dataset containing 7,587 images with 35,549 annotations spanning ten defect categories was assembled from North Western Railways, India. The research implements targeted augmentation methodologies and class-weighted training protocols to address severe imbalance (ranging from 0.4% for Broken-Rail to 17.6% for Shelling). Performance assessment across accuracy metrics, computational efficiency, and deployment viability demonstrates YOLOv11 achieving superior mean Average Precision (89.8%), while YOLO-NAS exhibits competitive speed-accuracy trade-offs. Critical examination of cross-regional generalization, temporal robustness, and real-time deployment constraints delivers practical insights for railway authorities implementing AI-based inspection frameworks.
Slab track is highly sensitive to subgrade deformation, particularly at subgrade-bridge transition section, where abrupt stiffness changes govern structural response and durability. In this study, a three-dimensional slab track model on the subgrade-bridge transition (ST-SBT model) is developed to simulate the coupled response of a slab track, subgrade and bridge. The plastic damage of the track concrete is represented by a concrete damaged plasticity model. This model is used to study the vertical deformation, interlayer separation, and damage evolution under the combined effects of semi-cosine and folded-angle non-uniform settlement, overall abutment settlement, and the combined action of abutment settlement and train loads. The results indicate that a short transition section significantly amplifies structural responses. For a transition length of 5m and a folded-angle settlement of 20mm, the track slabs and base slabs will suffer severe tensile damage. However, when the transition length exceeds approximately 10-15m, both deformation and damage will significantly decrease. Overall abutment settlement leads to a high concentration of deformation and damage at the beam ends and the abutment-subgrade joints. As the settlement continues to increase, strip-like cracks will appear in this section of slab track. Coupled train loads will further increase vertical displacement, uplift force, and interlayer separation, and may even induce tensile damage at relatively small levels of settlement. These findings from the ST-SBT model provide a quantitative basis for the rational design of transition section lengths, control of abutment settlement limits, and monitoring and maintenance of slab track in subgrade-bridge transitions.
For variable-section high-speed railway bridges, such as continuous girders and rigid-frame bridges, tapered 3D beam elements are frequently utilized to model bridge components. However, current high-speed railway design software and vehicle-bridge coupling analysis software tend to use prismatic beams instead, which leads to insufficient accuracy. In this paper, the cross-sectional stiffness matrix and flexibility matrix without rigid body displacement are established utilizing a generalized coordinate system with cantilever beam constraint and force interpolation function based on the equilibrium relation. Based on the force-based finite element method in combination with the virtual work principle, the shape function matrices for the cases with and without considering the shear effect are deduced, respectively. Then, the tapered 3D beam element consistent mass matrix is derived. To verify the accuracy of the proposed force-based finite element method, the derived matrix is degenerated into a prismatic beam to obtain the prismatic 3D beam element consistent mass matrix. Furthermore, compared with the commercial software Midas, the maximum natural frequency error for a linearly varying simply supported beam is less than 0.16 %. Both the theoretical degenerate solution and the numerical verification case prove that the stiffness matrix and consistent mass matrix of the tapered 3D beam element derived in this paper are highly accurate. Moreover, the force-based derivation method is proven to be reliable for deriving the tapered 3D beam element dynamic property matrix.
