Combining the improved delayed detached eddy simulation and Ffowcs Williams–Hawkings equation, a numerical study is conducted to explore the potential of base-frame fairings in aerodynamic noise reduction of high-speed pantographs and deepen the understanding of related flow physics. The fairing models for noise control are designed without changing the bottom structures of the pantograph. The aerodynamic and acoustic results indicate that the flow deflection and acceleration effects caused by the fairings when shielding the bottom components of the pantograph as well as the self-noise generated by the interaction between the wake of unshielded components and the fairings may compromise the noise reduction effects. Compared with the solid fairing, the perforated fairing has additional advantages in noise reduction. The presence of through-holes leads to a flow redistribution around the fairing, alleviating the flow deflection and acceleration effects. Besides, the airflow ejected from the holes on leeward side can suppress the formation of vortex structures in the fairing wake and push them downstream, thereby effectively weakening the flow field fluctuation near the fairing tail. The investigation of the aerodynamic drag of the pantograph and lift fluctuation of the strip further confirms the superiority of the perforated fairing over the solid one.

Current seismic damage assessments for high-speed railway (HSR) bridges primarily focus on the overall structural safety, lacking evaluations from multiple performance perspectives, which affects the post-earthquake traffic decision-making for the bridges. This study proposes a performance-based comprehensive functional damage probability assessment framework for high-speed railway simply supported bridges (HSRSSBs) under earthquakes. The framework categorizes the functions of HSR bridges into three levels: post-earthquake traffic function (PTF), structural bearing function (SBF), and collapse resistance function (CRF), corresponding to the operational, structural safety, and structural integrity requirements of HSRSSB, respectively. By analyzing the damage states of key bridge components during earthquakes, the functional damage probability assessment indicators and classification thresholds are established according to various performance requirements. Damage probability calculations are conducted using the probability density evolution method and vulnerability method. Finally, based on the relationship between damage probabilities at different functional levels, a comprehensive damage probability assessment framework considering the three-level performance requirements of HSRSSBs is developed, and the influence of varying pier heights on the functional damage probability relationship is examined. The results indicate that current HSRSSB designs meet all performance requirements under frequent earthquakes. Under design-level earthquake conditions, the SBF remains in a slight damage state, while the PTF exhibits varying degrees of damage, which worsens as pier height increases. The pier structure satisfies seismic demands even under rare earthquake conditions.

Nonuniform track support and differential settlements are commonly observed in bridge approaches where the ballast layer can develop gaps at crosstie-ballast interfaces often referred to as a hanging crosstie condition. Hanging crossties usually yield unfavorable dynamic effects such as higher wheel loads, which negatively impact the serviceability and safety of railway operations. Hence, a better understanding of the mechanisms that cause hanging crossties and their effects on the ballast layer load-deformation characteristics is necessary. Since the ballast layer is a particulate medium, the discrete element method (DEM), which simulates ballast particle interactions individually, is ideal to explore the interparticle contact forces and ballast movements under dynamic wheel loading. Accurate representations of the dynamic loads from the train and track superstructure are needed for high-fidelity DEM modeling. This paper introduces an integrated modeling approach, which couples a single-crosstie DEM ballast model with a train–track–bridge (TTB) model using a proportional–integral–derivative control loop. The TTB–DEM model was validated with field measurements, and the coupled model calculates similar crosstie displacements as the TTB model. The TTB–DEM provided new insights into the ballast particle-scale behavior, which the TTB model alone cannot explore. The TTB–DEM coupling approach identified detrimental effects of hanging crossties on adjacent crossties, which were found to experience drastic vibrations and large ballast contact force concentrations.

The issue of fatigue damage to rails has become increasingly prominent with the rise in subway traffic and speed. The hazardous space of the turnout frog significantly intensifies the dynamic interaction between the vehicle and the frog rail, leading to more pronounced fatigue damage in the turnout rail. This paper focuses on the No. 9 turnout fixed frog commonly used in subway lines. A three-dimensional explicit transient rolling contact finite element model of the fixed frog is established. The dynamic response of wheel–rail rolling contact is analyzed under various speeds and vertical stiffness conditions. Rolling contact fatigue crack locations, angles, and initiation life were investigated. The research indicates that the 30 mm top width cross-section of the nose rail is most susceptible to fatigue cracks, which initiate on the rail surface. The angle between the crack initiation surface and the lateral direction is between 70° and 95°. Higher speeds result in shorter fatigue life, while the vertical stiffness of the fastener has less of an effect. The simulation results align with findings from field surveys. The established model and research conclusions can provide theoretical support for optimizing fixed frog structures and predicting fatigue life.

To ensure the compatibility between rolling stock and infrastructure when dynamically assessing railway bridges under high-speed traffic, the damping properties considered in the calculation model significantly influence the predicted acceleration amplitude at resonance. However, due to the normative specifications of EN 1991-2, which are considered to be overly conservative, damping factors that are far below the actual damping have to be used when predicting vibrations of railway bridges, which means that accelerations at resonance tend to be overestimated to an uneconomical extent. Comparisons between damping factors prescribed by the standard and those identified based on in situ structure measurements always reveal a large discrepancy between reality and regulation. Given this background, this contribution presents a novel approach for defining the damping factor of railway bridges with ballasted tracks, where the damping factor for bridges is mathematically determined based on three different two-dimensional mechanical models. The basic principle of the approach for mathematically determining the damping factor is to separately define and superimpose the dissipative contributions of the supporting structure (including the substructure) and the superstructure. Using the results of a measurement campaign on 15 existing steel railway bridges in the Austrian rail network, the presented mechanical models are calibrated, and by analysing the energy dissipation in the ballasted track, guiding principles for practical application are defined. This guideline is intended to establish an alternative to the currently valid specifications of EN 1991-2, enabling the damping factor of railway bridges to be assessed in a realistic range by mathematical calculation and thus without the need for extensive in situ measurements on the individual structure. In this way, the existing potential of the infrastructure with regard to the damping properties of bridges can be utilised. This contribution focuses on steel bridges, but the mathematical approach for determining the damping factor applies equally to other bridge types (concrete, composite, or filler beam).

In recent years, the issue of structure-borne noise generated by steel–concrete composite (SCC) bridges has become increasingly severe. To control this noise by adjusting the cross section parameters of SCC bridges, this study first established a numerical model based on the hybrid finite element–statistical energy analysis (FE-SEA) method. The overall sound pressure levels calculated by numerical model are compared with field measurements, showing discrepancies of 0.4 dB and 1.1 dB, respectively. The comparison confirms the accuracy of the numerical model. Then, a high-accuracy radial basis function neural network (RBFNN) was trained using samples generated from the numerical model with uniform design. To achieve greater noise reduction with lower costs, the non-dominated sorting genetic algorithm (NSGA-II) was used for multi-objective constrained optimization, resulting in the Pareto frontier for sound power levels (SWLs) and material cost. Finally, the solution set was evaluated using the technique for order preference by similarity to an ideal solution method, and the optimal combination of cross sectional parameters was obtained. This combination resulted in a 5 dB reduction in the SWL of the structure and a 23.9% reduction in material cost.

Since the view that the localized rail third-order bending mode can cause high-order polygonization (mainly 18–23) of high-speed train wheels was put forward in 2017, many scholars have attempted to link a connection between the localized rail bending modes and wheel polygonization phenomenon and polygonal wheel passing frequency. This paper first establishes a flexible track model considering the structural and parametric characteristics of fasteners, verifies the model by using vehicle tracking test data, then investigates the influence of fastener parameter matching on the localized rail bending modes, and obtains the following conclusions: (1) There is nearly a 1:1 mapping relationship between the localized rail bending modal frequency and polygonal wheel passing (PWP) frequency, which supports that the localized rail bending mode is one of the causes of wheel polygonization. (2) The iron plate of the fastener system plays a role of dynamic vibration absorber in the vehicle-rail coupled system, and the fastener parameters significantly influence the localized rail bending modal vibration. Finally, this paper proposes a design principle of a high-frequency vibration-absorbing fastener, which provides a feasible solution to mitigate the localized rail bending modal vibration and high-order wheel polygonization. Meanwhile, it points out that this measure may induce other high-frequency vibration problems, e.g., aggravating modal vibration above 800 Hz. Further, this paper proposes a concept of differentiated arrangement of fasteners, suggesting that different high-frequency vibration-absorbing fasteners be installed in different sections of the whole line to make the localized rail bending modal frequency of the whole line disordered, thus disrupting and further mitigating the development of the wheel polygonization.

Turnout irregularity significantly affects the stochastic vibration behavior of vehicle–turnout structures. This study proposes a fitting formula for the turnout irregularity spectrum and develops a turnout irregularity full information expression model (TIFIEM) using a stochastic harmonic function. The model is applied to vehicle–turnout structure stochastic vibration and reliability analysis. Findings suggest that the Hamming window method, with a window length of 4096 points, is optimal for estimating the turnout irregularity spectrum. It is recommended to fit the power spectral density (PSD) using a 5th-order polynomial for better accuracy. The TIFIEM effectively addresses randomness in amplitude, frequency, and phase. An analysis of 250 irregularity samples is sufficient for the desired accuracy. Additionally, the PSD amplitude at various frequency points follows a Chi-square distribution with 2° of freedom. Regions 3–7 m from the tip of the switch rail on the straight switch rail and 53–54 m on the point rail are most susceptible to wear. When the vehicle passes through the turnout at 300 km/h, the reliability of vehicle–turnout structures at the crossing panel decreases to 95.8%.

Railway systems are critical components of transportation networks requiring consistent maintenance. This paper proposes a novel data-driven approach to detect various maintenance needs of railway track systems using acceleration data obtained from a passenger train in operation. The framework contains four modules. Firstly, data pre-processing and cleansing are performed to extract useful data from the whole dataset. Then, condition-sensitive features are extracted from the raw data in three different domains of time, frequency, and time–frequency. In the third module, the best subset of measurement features that characterize the state of the tracks are selected using the analysis of variance (ANOVA) algorithm which eliminates irrelevant characteristics from the feature set of responses. Finally, a multilabel classification algorithm based on the cascade feed-forward neural network (CFNN) is used to classify the type of maintenance needs of the track. An open-access dataset from a field study in Pennsylvania, USA, is used in this study for validation of the proposed method. The results indicate that employing a CFNN can achieve 95% accuracy in identifying two maintenance activities, tamping and surfacing, using time-domain features. Moreover, an extensive analysis has been conducted to evaluate the influence of various feature extraction and selection methods, diverse classification algorithms, and different types of accelerometers (uni-axial and tri-axial) on the accuracy of the proposed method.

The diversion effect caused by the linked structure in a metro tunnel with cross-passage complicates the impact of longitudinal fire source location on the smoke backflow layering behavior that has not been clarified, despite the fact that the scenario exists in practice. A series of laboratory-scale experiments were conducted in this study to investigate the smoke back-layering length in a model tunnel with cross-passage. The heat release rate, the velocity of longitudinal air flow, and the location of the fire source were all varied. It was found that the behavior of smoke backflow for the fire source located at the upstream of bifurcation point resembles a single-hole tunnel fire. As the fire source’s position shifts downstream from the bifurcation point, the length of smoke back-layering progressively increases. A competitive interaction exists between airflow diversion and smoke diversion during smoke backflow, significantly affecting the smoke back-layering length in the main tunnel. The dimensionless smoke back-layering length model was formulated in a tunnel featuring a cross-passage, taking into account the positions of longitudinal fire sources. The dimensionless smoke back-layering length exhibits a positive correlation with the 17/18 power of total heat release rate Q and a negative correlation with the 5/2 power of longitudinal ventilation velocity V.