The demand for faster, longer, and safer railway networks has intensified the need for robust condition monitoring systems, particularly those capable of detecting rolling stock defects such as wheel flats, bearing faults, hunting movement, overloading, and unbalanced loads. Effective monitoring and timely maintenance are essential to mitigate these issues, enhancing both reliability and safety in railway operations. This paper presents a comprehensive review of wayside condition monitoring (WCM) systems. The study begins with a structured overview of WCM architectures, followed by a bibliometric analysis that highlights recent research trends in machine learning applications for railway condition monitoring. A detailed classification of rolling stock defects is provided to establish the foundation for condition assessment. The review then presents an extensive survey of commercial wayside monitoring systems currently in use, including hot axle box detectors, wheel impact load detectors, acoustic detection systems, and weigh-in-motion technologies. The various sensor types integrated into these systems are also described in detail. To bridge the gap between raw data collection and actionable insights, the paper includes a dedicated section on knowledge extraction from WCM systems. This section outlines key approaches for fault detection, diagnosis, and classification using advanced data processing methods, including signal processing and machine learning techniques. Finally, the paper identifies critical challenges such as data quality, real-time processing constraints, infrastructure limitations, and the need for generalizable models. It also discusses research gaps and suggests future directions, including the integration of edge computing, digital twin technology, and self-diagnostic capabilities to support the development of next-generation intelligent WCM systems.
The sliding contact of the pantograph–overhead catenary system (PCS) causes wear of the contacting surfaces, which is a major issue for the maintenance of the rolling stock and overhead equipment. This phenomenon involves complex interrelations between mechanical, electrical, and thermal effects, posing significant challenges in modeling the wear process and finding mitigation measures. This paper reviews recent research concerned with wear in the PCS. The laboratory test rigs used for experimental investigations are described and compared. Then, the wear mechanism is analyzed to outline the effects of critical parameters such as contact force, electric current, arcs, ambient factors, and materials. The definition and use of wear models, as well as numerical methods for simulating the long-term wear effects in the PCS are summarized. Finally, gaps in present research and future research needs are identified, with particular emphasis on addressing the effects of uncertainties related to environmental factors and finding measures to mitigate abnormal wear phenomena recently reported in railway systems.
Structural discontinuities in railway tracks have proven challenging from a maintenance perspective. These discontinuities can lead to uneven settlements, reducing serviceability of the railway network and increasing the track’s dynamic loading. To optimize the long-term performance of railway structures, it is essential to evaluate different design solutions under varying loading conditions to identify potential risk factors early. Consequently, this study proposes a novel computational model for simulating the dynamic long-term behavior of ballasted railway tracks. The proposed model enables computationally efficient simulation and provides an innovative mathematical framework for analyzing the mechanical behavior of structural discontinuities, allowing detailed consideration of substructure and subsoil properties, including their variations along the longitudinal direction of the track. Simulations were conducted to investigate the effects of bridge transition zones and rail defects on the short- and long-term behavior of the track for two vehicle types. In addition, extensive field measurement data were utilized for model verification. Based on simulations, the axle load appears to be the primary factor influencing the long-term performance of railway transition zones. However, for more localized defect types, the significance of driving speed and the unsprung mass of rolling stock becomes more pronounced. Overall, the findings highlight the nonlinear relationship between vehicle loading and structural deterioration, emphasizing its strong dependence on track properties.
Infrastructure managers rely on diagnostic trains that periodically measure track geometry and vehicle accelerations to ensure the safety of the railway network. Their runs are scheduled depending on the line priority, in order to safely monitor the evolution of track defects. However, sudden and unpredictable defect growth may happen and be missed between successive runs. Therefore, condition monitoring systems have been installed on in-service vehicles. In fact, these trains run every day along the same line, so they can provide additional information useful for maintenance practices. When trains run along conventional lines, their speed significantly changes depending on the line characteristics, and vehicle accelerations strongly depend on speed. Therefore, monitoring systems that rely on vehicle accelerations should carefully take this effect into account. In this paper, a methodology to estimate the track longitudinal level using bogie accelerations from an in-service vehicle is presented. The recorded accelerations were double-integrated to account for the speed variation, and a model-based strategy was adopted to reduce the filtering action of the primary suspension. Data were recorded during a two-year monitoring campaign along an Italian railway line. The methodology allowed for the estimation of the longitudinal level along specific track sections, considering statistical measures like the peak value. A maximum error of 1 mm was found between the estimated values and those measured by the diagnostic train (considering a defect with magnitude of 7.5 mm). Therefore, the results showed that it is possible to estimate the peak longitudinal level between the two rails using one single vertical accelerometer installed on the bogie of an in-service vehicle. The results of this research may be used to support the current maintenance strategy with daily estimations of track longitudinal level. It should be noted that specific attention was given only to this type of track geometry parameter, since it often drives maintenance operations. In the future, the possibility to extend the methodology to the estimation of different type of defects, like cross-level and twist, could be considered.
Train-induced ground vibration impacts the environment and has attracted increased attention. Prediction of this vibration is subject to considerable uncertainty and is affected by many factors, particularly when building foundations are situated within the transmission path between the vibration source and the target building. A hybrid computational model for predicting ground vibration has been developed to assess the impact of building foundations of various types located in the transmission path. The model is used to investigate four types of foundations: raft, strip, pile, and box. The ground responses are determined by combining the semi-analytical model ‘modelling of train-induced vibration’ (MOTIV) for train-induced vibration with a finite element model of the ground including the building. MOTIV is used to determine the excitation in terms of the force density for a train running either on a surface railway or in a tunnel. The finite element model is used to determine line source transfer mobilities in the presence of the building foundations by applying a set of incoherent unit loads to the ground. The results indicate that building foundations in the transmission path can mitigate ground vibration, particularly for excitation at the ground surface, for which reductions in overall vibration level of between 4 and 8 dB are found. Deep foundations provide greater vibration reduction than shallow foundations, notably in cases involving surface railways. The underground railway case demonstrates less significant insertion loss than the surface railway case, with values based on overall velocity level between −2 and 2 dB, rising to 4 dB for the box foundation. Furthermore, it is observed that ground vibration mitigation is more pronounced in the near field than in areas further from the building.
A transmittance function-based diagnosis methodology for robust detection, isolation, and precise estimation of railway vehicle suspension components degradation under varying travelling conditions is presented. The methodology is based on transmittance function data-driven stochastic functionally pooled models, which are capable of representing the vehicle dynamics under varying operating conditions. This characteristic ensures robustness in fault/degradation diagnosis as the effects of varying travelling conditions on vehicle dynamics may fully or partly obscure the degradation effects. Degradation detection is accomplished in an unsupervised manner, while isolation and estimation are supervised. The travelling conditions are considered non-measurable in the inspection phase, where the methodology is assessed via different scenarios and the transmittance function is employed for the regularization of track excitation uncertainty. The methodology is validated through thousands of Monte Carlo simulations with a Simpack-based 54-degree-of-freedom vehicle model of the Attiko Metro company using acceleration signals from two points on the vehicle. A wide range of travelling speeds, passenger payloads, and different levels of degradation (through stiffness reduction) of the primary suspension conical rubber spring and the air spring of the secondary suspension are considered, combined with different realizations of track irregularities following the ERRI B176 standard. The results indicate excellent detection performance for even early stage degradation levels, while correct isolation is achieved for approximately 94% and 99% of the test cases considered in the vehicle’s primary and secondary suspension components, respectively. The corresponding degradation level estimation accuracy for these components is demonstrated via the mean absolute error, which is 2.45% ± 2.92% and 4.05% ± 3.66%. The methodology’s performance is additionally compared with that of a state-of-the-art data-driven method, showing an overall superior performance.
The complex mechanical configuration and non-streamlined design of the bogie adversely affect the aerodynamic performance of high-speed trains. Optimizing the underbody flow structure is a key strategy to reduce aerodynamic drag. This study introduces two types of bogie bottom fairings designed for the head car and establishes four computational models with varying covering structures. Aerodynamic performance is analyzed using the improved delayed detached eddy simulation (IDDES) method, based on the shear–stress transport (SST) k–ω turbulence model. The numerical simulations are validated through wind tunnel testing of a full-scale bogie, with an error margin below 13.46%. Results demonstrate that the proposed covering schemes substantially reduce aerodynamic drag. Specifically, integrating flat bottom and side fairings reduces the drag coefficient of the head car by 22.7% and that of the entire train by 10.6%. These findings provide valuable guidance for the implementation of bogie covering structures and offer significant insights into the aerodynamic design optimization of high-speed trains.
Efficient train operation relies on optimal traction at the wheel–rail interface, which can be compromised by factors such as water and/or contamination (e.g. leaf, sand, oil, and surface abrasion debris). This study introduces a finite element model to assess adhesion enhancement at the wheel–rail interface, with a focus on the impact of sand particles during the sanding process. Surface roughness is initially introduced to quantify its effect on adhesion, followed by the inclusion of rail plastic deformation. By integrating these two factors, the model provides a comprehensive framework for evaluating the complex mechanisms influencing adhesion at the wheel–rail interface, particularly in real-world train operations where surface conditions interact with contaminants such as sand fragments. This approach addresses the existing gap in understanding how rail surface condition and rail plastic deformation contribute to adhesion enhancement during the sanding process, offering new insights for optimising railway maintenance strategies.
To improve the accuracy and efficiency of the vehicle-track dynamic interaction under the framework of traditional finite element method, the hierarchical function has been introduced into the track structure modeling and coupled by multi-scale coupling method in substructural interaction and wheel–rail contacts with high modeling versatility. The reliability of the proposed method is validated by comparing to the results obtained by the FEM model from aspects of the dynamic response in the time- and frequency-domain and the characteristic frequencies of structural vibration. In addition, the efficiency and accuracy of hierarchical function in vehicle–track dynamics simulation are also proved. In the numerical examples, the errors of FEM-based and hierarchical function-based track structure models are analyzed from the perspective of characteristic frequency; the computational efficiency and accuracy of different hierarchical function-based coupled dynamics models of vehicle-track are compared; finally, the shear effect of rails and track slabs on the dynamic response of the system is analyzed. Numerical analysis indicated that, in comparison with other high-order function, the calculation efficiency can be increased by 56%. Furthermore, the shear effect predominantly influences vertical dynamic response, but has little effect on displacement and lateral dynamic response.
In order to analyze the impact of axle spacing on the vibration response of railway bridges, it is essential to use moving loads with non-uniform intervals, namely a non-equidistant moving load model of the train. In this paper, by incorporating this model and using the Fourier transform and modulus extraction, the corresponding moving load amplitude spectrum (MLAS) related to the moving loads and the vibration modes of bridges is deduced in the frequency domain, which can thus effectively reflect the displacement amplitude laws of the bridge free vibration. Hence, the resonance and cancelation phenomena of simply supported bridges under trains are comprehensively investigated using the presented MLAS. On the one hand, the corresponding resonance speeds obtained by the MLAS are not only related to the length of carriages, bridge frequencies, and spans, but also to the number of carriages, presenting a supplement to the traditional resonance speed, which sometimes does not cause the maximum displacement response of bridges. On the other hand, the cancelation phenomena in this paper are divided into type I and type II, and the closed-form solutions of the corresponding cancelation speeds are derived from the proposed MLAS, thus providing a more comprehensive understanding of the influencing factors of cancelation phenomena. In addition, through the MLAS comparison of different train models, it is found that the non-equidistant moving loads cannot be completely replaced by the equidistant moving loads in analyzing the resonance phenomenon of railway bridges, which is related to the ratio of the carriage length to the bridge span. The type II cancelation speeds are consistent whether moving loads are equidistant or non-equidistant. However, the type I cancelation speeds are inconsistent, and the difference is reflected in the influence of the distance between the wheel centers of each bogie and the distance between the bogie centers. These research results have been validated by time domain analyses in numerical examples. Therefore, the presented frequency domain method based on the MLAS is very valuable for comprehensively investigating the resonance and cancelation phenomena of simply supported bridges under non-equidistant moving loads.