Track geometry monitoring is essential for track maintenance. Dedicated track inspection vehicles are scheduled to measure track geometry irregularities throughout the railway network, so cannot inspect each line frequently. It is desirable to inspect geometry much higher frequently using in-service trains. One possible way is estimating track geometry from vehicle–body accelerations because the accelerators can be easily installed in vehicle–body. However, inverting track geometry from vehicle–body acceleration is a pending issue. Up to now, most research has focused on vehicle dynamics modelling and simulation. In this paper, we solve the problem using deep learning method and realistic measurement data. The training and test data are the inspection data acquired by comprehensive inspection trains from three main high-speed railways in China. The proposed AM–CNN–GRU model combines an Attention Mechanism (AM), a Convolutional Neural Network (CNN) and a Gated Recurrent Unit (GRU). The model’s inputs are vertical and lateral accelerations, and vehicle speed. The outputs are two vertical track irregularities, with wavelengths of 3–42m and 3–120 m, respectively. We evaluated the model by comparing the estimated irregularities with the actual measurements. We discussed different models, high-speed lines, sub-rail infrastructures and speed to validate the model effectiveness.

Ballast fouling occurs when the voids of granular particles are wholly or partially filled by fine particles. Railway ballast fouling is always an inevitable issue related to railway performance and safety in ballasted tracks. This paper presents an overview of the railway ballast fouling mechanism, current works, perspective, and innovation. The paper first reviews the fouling mechanism with different fouling sources and their influence, followed by the fouling indices. Then, fouling inspection methods are reviewed, and the ground penetrating radar is highlighted due to its high efficiency and accuracy. The proposed solutions, including rehabilitation, prevention, reinforcement, and stabilization in the ballasted track, are introduced. Additionally, several perspectives and works for the future are mentioned. Ballast cleaning is a thorough and effective treatment. Ground-penetrating radar inspection and proper evaluation can provide a preview and perspective on ballast cleaning determination.

The load generated during train operation is transmitted to sleepers through rails, and then from sleepers to trackbeds. As one of the important components of the track structure, sleepers are very important to the service-life and safety of railways. The running speed of the train will produce different load frequencies on the rails. To explore the response of sleepers under cyclical train loads, this research used the discrete element method to simulate rails and sleepers, and realizes the loading of sleepers by applying cyclic loads to the rails. The discrete element method was used to simulate the compression-failure test of the concrete sample to verify the reliability of parameters of the discrete element model. On this basis, the sleeper was simulated by the discrete element, and the change law of the mechanical parameters such as particle displacement, contact force, and stress of the sleeper under different loading frequencies were analyzed. The research results show that the sleeper particles under the rail had a large vertical displacement, and the maximum displacement and the maximum contact force of the particles in the sleeper increased with increasing loading frequency. The stress change trend of the sleeper under the left and right rails was the same, and both showed periodic changes with cyclic loading. The loading frequency had little effect on the stress of the sleeper. The vertical displacement, velocity, and acceleration amplitude of the rail all increased with increasing cyclic loading frequency, thereby increasing the force between the rail and the sleeper, and increasing the dynamic response of the sleeper.

By analyzing dynamic track geometric irregularity data of high-speed railway (HSR) with ballast track in China, there is a peak of [5, 11] m wavelength for the longitudinal level at the beam joint of simply supported girder bridges. The difference between the absolute value of track longitudinal level at the beam joint and that at the midspan is one to three mm. Under annual temperature cycle, the ballast at the beam joint loosens because of the bridge–track interaction, which results in the longitudinal level at the beam joint gets larger in winter and smaller in summer, that means the longitudinal level is negatively correlated with temperature, and at the same time, the development of the track longitudinal level and Track Quality Index (TQI) with the temperature are given. In addition, the vertical acceleration of axle box and bogie, and wheel–rail vertical force are larger when the train passes the beam joint than when it passes the midspan, which threatens running safety significantly. This article established a three-dimensional vehicle–track–bridge finite element model (FEM), thus the thresholds of track longitudinal level at the beam joint for simply supported girder bridge with ballast track in 250 km/h HSR were proposed, i.e., the maximum of dynamic track longitudinal level is 14 mm, and the maximum of static value is 11 mm. Once one of the two thresholds exceeds the corresponding limit, the train speed should be reduced to 160 km/h or below.

Metal sheets have good performance and have been widely used. Different kinds of defects can be generated during the preparation and service of metal plates, which will cause the structural performance of the metal plates to decline, thus requiring structural health monitoring (SHM). This study proposes an electromagnetic ultrasonic (EMUS) surface wave detection technique for metal sheet defects via simulation. The numerical results show that after the excitation parameters of the EMUS transducer are optimized through orthogonal experimental design, the amplitude of the EMUS signal generated is increased by about 80%. The power spectrum density (PSD) of the EMUS response signal is used to detect defects. Compared with the peak-to-peak detection, the accuracy is higher, and the reliability is better. The accuracy of the proposed “central zero-point” method for measuring the time delay of the EMUS signal wave packet is higher than that of the “peak-to-peak amplitude” method and the “vibration starting point” method and is close to the accuracy of the “cross-correlation” method.

Ballastless tracks have a high smoothness, but the corresponding laying requirements are strict. Therefore, the maximum span of cable-stayed bridges that can accommodate ballastless tracks is 392 m. For laying ballastless track structures over larger spans, the deformation characteristics of long-span, cable-stayed bridges under complex loads are incompletely understood, and the interaction between them and long-span track–bridge structures is unclear. The influence of the wavelength of the cosine wave on the track–bridge mapping of different orbital structures was explored. The wavelength characteristics of vertical deformation under complex loads were investigated. The track–bridge integrated model for the cable-stayed bridge was established to analyze the mapping relationship between the rail and the bridge and the wavelength characteristics of deformation. Based on the mapping relationships and the wavelength characteristics of deformation, the train–track–bridge dynamics simulation model was simplified. The results show that, when the minimum wavelength of bridge deformation surpassed 6 m, 10 m, and 16 m, the rail deformation in the ballasted track, the longitudinal-connected track, and the unit slab-type ballastless track accurately mirrored the deformation of the bridge. For the span of bridges ranging from 200 m to 600 m, the wavelength of vertical deformation ranged from 21 to 1270 m under complex loads. During local loads, the vertical deformation below the 200 m wavelength constituted a significant proportion near the pie. Considering the influence of the deformation on the train vibration response, the train–bridge dynamic coupling model can be employed to treat the track structure as a load to reduce the complexity of the model and enhance the calculation efficiency.