One of the most common failure types that rapidly deteriorates track geometry is sleeper voids or unsupported sleepers. Ballast pulverization, or so-called “white spots”, is a sign that indicates the presence of the high sleeper voids in the track. However, the objective estimation of the size and form of voids is possible by time- and cost-consuming track-side measurements at many points along the track. The study presents an efficient model-based approach for the identification of the void geometry by the track-side experimental measurements of rail deflection in one point. A robust 3-beam track model with a two-mass vehicle model together with a time-effective surrogate optimization algorithm is used for the multidimensional search of the void geometry that is fitting to experimental data. The results show that the void geometry could be found precisely with one-point measurements, significantly reducing the time and cost involved in the process. Therefore, a practical, simplified method of determining the void zone's depth and length is proposed. It is based on analyzing the relation between void sizes and the rail deflection wave sizes. Unlike the void depth, the void length cannot be found by the simple difference between the deflection waves in the void and the reference zones. The proposed method assumes wave estimation by applying deflection thresholds, ensuring a practical and reliable approach. The reliability of the proposed method in estimating void length and depth instills confidence in the effectiveness of the approach. Finally, it was used to estimate the void length and depth for many problem zones in ballast tracks that were inspected with track-side measurements. The result analysis shows that the void length and depth are subjected to a non-linear relation: the long-length voids have unproportionally higher depths than short voids. The results indicate that the settlement intensity of the neighbor-to-void sleepers is much lower than that of the hanging sleepers.

The cross-level and twist irregularities are the most dangerous irregularity types that could cause wheel unloading with the risk of derailments and additional maintenance expenses. However, the mechanism of the irregularities initiation and development is unclear. The motivation of the present study was the previous experimental studies on the application of wide sleepers in the ballasted track. The long-term track geometry measurements with wide sleepers show an enormous reduction of the vertical longitudinal irregularities compared to the conventional track. However, wide sleepers had higher twist and cross-section level irregularities. The present paper aims to explain the phenomenon by discrete element method (DEM) modeling the development process of sleeper inhomogeneous support at cross-level depending on the sleeper form. The DEM simulations show that the maximal settlement intensity is up to 3.5 times lower for a wide sleeper in comparison with the conventional one. Nevertheless, the cross-level differential settlements are almost the same for both sleepers. The particle loading distribution after all loading cycles is concentrated on the smaller area, up to the half sleeper length, with fully unloaded zones under sleeper ends. Ballast flow limitation under the central part of the sleeper could improve the resilience of wide sleepers to the development of cross-level irregularities. The mechanism of initiation of the cross-level irregularity is proposed, which assumes the loss of sleeper support under sleeper ends. The further growth of inhomogeneous settlements along the sleeper is assumed as a result of the interaction of two processes: ballast flow due to dynamic impact during void closing and on the other side high pressure due to the concentration of the pressure under the middle part of the sleeper. The DEM simulation results support the assumption of the mechanism and agree with the experimental studies.

The high influence of impact and vibration on the behavior of crushed stone and ballast materials has been known for a long time. The zones with unsupported sleepers, which are always present in transition zones, crossings, welds, etc., are typically characterized by impact interaction, ballast full unloading, and additional preloading. However, no studies on ballast layer settlements consider impact vibration loading. Moreover, the influence of the cyclic loading on the ballast settlement intensity is considered ambiguously, with both decelerating and accelerating trends. The comprehensive literature review presents the influence of factors on settlement intensity. The present study aims to estimate the long-term processes of sleeper settlement accumulation depending on the loading factors: impact, cyclic loading, and preloading. The typical for a void zone ballast loading pattern was determined for various void sizes and the position along the track by using a model of vehicle-track interaction that was validated by experimental measurements. The loading patterns were parametrized with four parameters: maxima of the cyclic loading, impact loading, sleeper acceleration, and minimal preloading. A specially prepared DEM simulation model was used to estimate the ballast settlement intensity after initial settlement stabilization for more than 100 loading patterns of the void zone cases. The settlement simulation results clearly show that even a low-impact loading pattern causes many times increased settlement intensity than ordinary cyclic loading. Moreover, the initial preloading in the neighbor-to-void zones can cause even a decrease in the settlement intensity compared to the full ordinary or partial unloading. A statistical analysis using a machine learning approach and an analytic one was used to create the model for the intensity prediction regarding the loading patterns. The analytic approach demonstrates somewhat lower prediction quality, but it allows to receive plausible and simple analytic equations of the settlement intensity. The results show that the maximal cyclic loading has a nonlinear influence on the settlement intensity that corresponds to the 3–4 power function, and the impact loading is expressed by the linear to parabolic function. The ballast’s minimal preloading contributes to the reduction of the settlement intensity, especially for high cyclic loadings that are typical for neighbor-to-void zones. The results of the present study could be used for the complementing of the present phenomenological equations with the new factors and further application in the algorithms of the settlements accumulation prediction.

It is essential to assess the rolling contact fatigue (RCF) of turnouts and maintain them in advance. It saves a lot of money while protecting the safety of railway operations. In Germany, the damage on rails, especially crossing noses, mainly depends on the subjective judgment of experts. There are no objective and comprehensive evaluation criteria. This paper presents the application of image processing and supervised machine learning algorithms to crossing nose fatigue judgment. The fatigue characteristics of the crossing nose rolling contact surface along the life cycle of the crossing nose are analyzed. The study used crack information from magnetic particle inspection (MPI) images of crossing nose surfaces. It uses basic image processing methods to collect physical information about features of fatigue cracks in images. Existing feature selection methods are used to exclude irrelevant features and retain valuable features. And we select the best feature selection method through the regression results. Statistically significant crack features and combinations that depict the surface fatigue state are found. In this paper, by comparing several usually machine learning regression algorithms, it is found that the supervised learning of support vector machine regression (SVR) has achieved the best results in the regression fitting of the crack feature data in this paper. The regression results form a simple system to evaluate the life cycle of crossing nose. The system finds the location of cracks that can create dangerous defects in the crossing nose surface. The research result consists of the early prediction of rail contact fatigue.

The share of rail transport in world transport continues to rise. As the number of trains increases, so does the load on the railway. The rails are in direct contact with the loaded wheels. Therefore, it is more easily damaged. In recent years, domestic and foreign scholars have conducted in-depth research on railway damage detection. As the weakest part of the track system, switches are more prone to damage. Assessing and predicting rail surface damage can improve the safety of rail operations and allow for proper planning and maintenance to reduce capital expenditure and increase operational efficiency. Under the premise that functional safety is paramount, predicting the service life of rails, especially turnouts, can significantly reduce costs and ensure the safety of railway transportation. This paper understands the evolution of contact fatigue on crossing noses through long-term observation and sampling of crossing noses in turnouts. The authors get images from new to damaged. After image preprocessing, MPI (Magnetic Particle Imaging) is divided into blocks containing local crack information. The obtained local texture information is used for regression prediction using machine-supervised learning and LSTM network (Long Short-Term Memory) methods. Finally, a technique capable of thoroughly evaluating the wear process of crossing noses is proposed.

Railway damage detection is of great significance in ensuring railway safety. The cracks on the rail surface play a key role in studying the formation and development process of rail damage, predicting the occurrence of rail defects, and then improving the service life of the rail. However, due to the small shape of the cracks, the typical detection method is relatively complicated, and the speed is quite slow. Although traditional magnetic particle inspection technology is fairly accurate at detection, it is costly and inconvenient to carry and install, while also limiting the detection speed and affecting the system’s operation. In this paper, a semantic segmentation detection method is developed by using various collected rail surface crack data and deep learning through a neural network. By comparing the inspection of the same rail surface with magnetic particle inspection technology, only inexpensive cameras are used and the inspection speed is increased while maintaining relatively high accuracy. In addition, the method can achieve fast detection speeds if it is extended to be combined with high-frequency cameras. It is an economical, efficient, and environmentally friendly method for future rail surface detection.

The present paper deals with the experimental investigation of interlocking effect of crushed stone ballast material, assessing it as the relationship with the residual and dynamic stresses under the ballast layer during laboratory dynamic tests with the consideration of different boundary conditions. The laboratory experiments were executed with a scaled model of ballast under the sleeper. The measured pressure at the bottom surface of the ballast has two parts: dynamic and residual. The dynamic part depends on the external loading; the residual part remains after unloading. The measured residual stress was observed up to 3 times higher than the stress due to cyclic external loading. The relationship of the residual stress and interlocking effect to ballast particles angularity is analyzed. A simple interpretation of the distribution of residual stress is proposed, that depends on the measured cyclic stress and the elasticity of bounding walls. The study of interlocking effect of ballast could be potentially useful for many practical problems of railway track design as well as for the track maintenance issues.