Wayside monitoring is a promising cost-effective alternative to predict damage in the rolling stock. The main goal of this work is to present an unsupervised methodology to identify out-of-roundness (OOR) damage wheels, such as wheel flats and polygonal wheels. This automatic damage identification algorithm is based on the vertical acceleration evaluated on the rails using a virtual wayside monitoring system and involves the application of a two-step procedure. The first step aims to define a confidence boundary by using (healthy) measurements evaluated on the rail constituting a baseline. The second step of the procedure involves classifying damage of predefined scenarios with different levels of severities. The proposed procedure is based on a machine learning methodology and includes the following stages: (1) data collection, (2) damage-sensitive feature extraction from the acquired responses using a neural network model, i.e., the sparse autoencoder (SAE), (3) data fusion based on the Mahalanobis distance, and (4) unsupervised feature classification by implementing outlier and cluster analysis. This procedure considers baseline responses at different speeds and rail irregularities to train the SAE model. Then, the trained SAE is capable to reconstruct test responses (not trained) allowing to compute the accumulative difference between original and reconstructed signals. The results prove the efficiency of the proposed approach in identifying the two most common types of OOR in railway wheels.
The dynamic load distribution within in-service axlebox bearings of high-speed trains is crucial for the fatigue reliability assessment and forward design of axlebox bearings. This paper presents an in situ measurement of the dynamic load distribution in the four rows of two axlebox bearings on a bogie wheelset of a high-speed train under polygonal wheel–rail excitation. The measurement employed an improved strain-based method to measure the dynamic radial load distribution of roller bearings. The four rows of two axlebox bearings on a wheelset exhibited different ranges of loaded zones and different means of distributed loads. Besides, the mean value and standard deviation of measured roller–raceway contact loads showed non-monotonic variations with the frequency of wheel–rail excitation. The fatigue life of the four bearing rows under polygonal wheel–rail excitation was quantitatively predicted by compiling the measured roller–raceway contact load spectra of the most loaded position and considering the load spectra as input.
High-speed trains typically utilize helical gear transmissions, which significantly impact the bearing load capacity and fatigue service performance of the gearbox bearings. This paper focuses on the gearbox bearings, establishing dynamic models for both helical gear and herringbone gear transmissions in high-speed trains. The modeling particularly emphasizes the precision of the bearings at the gearbox’s pinion and gear wheels. Using this model, a comparative analysis is conducted on the bearing loads and contact stresses of the gearbox bearings under uniform-speed operation between the two gear transmissions. The findings reveal that the helical gear transmission generates axial forces leading to severe load imbalance on the bearings at both sides of the large gear, and this imbalance intensifies with the increase in train speed. Consequently, this results in a significant increase in contact stress on the bearings on one side. The adoption of herringbone gear transmission effectively suppresses axial forces, resolving the load imbalance issue and substantially reducing the contact stress on the originally biased side of the bearings. The study demonstrates that employing herringbone gear transmission can significantly enhance the service performance of high-speed train gearbox bearings, thereby extending their service life.
Although train modeling research is vast, most available simulation tools are confined to city- or trip-scale analysis, primarily offering micro-level simulations of network segments. This paper addresses this void by developing the NeTrainSim simulator for heavy long-haul freight trains on a network of multiple intersecting tracks. The main objective of this simulator is to enable a comprehensive analysis of energy consumption and the associated carbon footprint for the entire train system. Four case studies were conducted to demonstrate the simulator’s performance. The first case study validates the model by comparing NeTrainSim output to empirical trajectory data. The results demonstrate that the simulated trajectory is precise enough to estimate the train energy consumption and carbon dioxide emissions. The second application demonstrates the train-following model considering six trains following each other. The results showcase the model ability to maintain safe-following distances between successive trains. The next study highlights the simulator’s ability to resolve train conflicts for different scenarios. Finally, the suitability of the NeTrainSim for modeling realistic railroad networks is verified through the modeling of the entire US network and comparing alternative powertrains on the fleet energy consumption.
The accurate assessment of running safety during earthquakes is of significant importance for ensuring the safety of railway lines. Currently, assessment methods based on a single index suffer from issues such as misjudgment of operational safety and difficulty in evaluating operational margin, making them unsuitable for assessing train safety during earthquakes. Therefore, in order to propose an effective evaluation method for the running safety of trains during earthquakes, this study employs three indexes, namely lateral displacement of the wheel–rail contact point, wheel unloading rate, and wheel lift, to describe the lateral and vertical contact states between the wheel and rail. The corresponding evolution characteristics of the wheel–rail contact states are determined, and the derailment forms under different frequency components of seismic motion are identified through dynamic numerical simulations of the train–track coupled system under sine excitation. The variations in the wheel–rail contact states during the transition from a safe state to the critical state of derailment are analyzed, thereby constructing the evolutionary path of train derailment and seismic derailment risk domain. Lastly, the wheel–rail contact and derailment states under seismic conditions are analyzed, thus verifying the effectiveness of the evaluation method for assessing running safety under earthquakes proposed in this study. The results indicate that the assessment method based on the derailment risk domain accurately and comprehensively reflects the wheel–rail contact states under seismic conditions. It successfully determines the forms of train derailment, the risk levels of derailment, and the evolutionary paths of derailment risk.
The issue of low-frequency structural noise radiated from high-speed railway (HSR) box-girder bridges (BGBs) is a significant challenge worldwide. Although it is known that vibrations in BGBs caused by moving trains can be reduced by installing multiple tuned mass dampers (MTMDs) on the top plate, there is limited research on the noise reduction achieved by this method. This study aims to investigate the noise reduction mechanism of BGBs installed with MTMDs on the top plate. A sound radiation prediction model for the BGB installed with MTMDs is developed, based on the vehicle–track–bridge coupled dynamics and acoustics boundary element method. After being verified by field tested results, the prediction model is employed to study the reduction of vibration and noise of BGBs caused by the MTMDs. It is found that installing MTMDs on top plate can significantly affect the vibration distribution and sound radiation law of BGBs. However, its impact on the sound radiation caused by vibrations dominated by the global modes of BGBs is minimal. The noise reduction achieved by MTMDs is mainly through changing the acoustic radiation contributions of each plate of the bridge. In the lower frequency range, the noise reduction of BGB caused by MTMDs can be more effective if the installation of MTMDs can modify the vibration frequency and distribution of the BGB to avoid the influence of small vibrations and disperse the sound radiation from each plate.
During the operation of sandy railways, the challenge posed by wind-blown sand is a persistent issue. An in-depth study on the influence of wind-blown sand content on the macroscopic and microscopic mechanical properties of the ballast bed is of great significance for understanding the potential problems of sandy railways and proposing reasonable and adequate maintenance and repair strategies. Building upon existing research, this study proposes a new assessment indicator for sand content. Utilizing the discrete element method (DEM) and fully considering the complex interactions between ballast and sand particles, three-dimensional (3D) multi-scale analysis models of sandy ballast beds with different wind-blown sand contents are established and validated through field experiments. The effects of varying wind-blown sand content on the microscopic contact distribution and macroscopic mechanical behavior (such as resistance and support stiffness) of ballast beds are carefully analyzed. The results show that with the increase in sand content, the average contact force and coordination number between ballast particles gradually decrease, and the disparity in contact forces between different layers of the ballast bed diminishes. The longitudinal and lateral resistance of the ballast bed initially decreases and then increases, with a critical point at 10% sand content. At 15% sand content, the lateral resistance is mainly shared by the ballast shoulder. The longitudinal resistance sharing ratio is always the largest on the sleeper side, followed by that at the sleeper bottom, and the smallest on the ballast shoulder. When the sand content exceeds 10%, the contribution of sand particles to stiffness significantly increases, leading to an accelerated growth rate of the overall support stiffness of the ballast bed, which is highly detrimental to the long-term service performance of the ballast bed. In conclusion, it is recommended that maintenance and repair operations should be promptly conducted when the sand content of the ballast bed reaches or exceeds 10%.
The technology of pantograph sinking in the cavity is generally adopted in the new generation of high-speed trains in China for aerodynamic noise reduction in this region. This study takes a high-speed train with a 4-car formation and a pantograph as the research object and compares the aerodynamic acoustic performance of two scale models, 1/8 and 1/1, using large eddy simulation and Ffowcs Williams–Hawkings integral equation. It is found that there is no direct scale similarity between their aeroacoustic performance. The 1/1 model airflow is separated at the leading edge of the panhead and reattached to the panhead, and its vortex shedding Strouhal number (St) is 0.17. However, the 1/8 model airflow is separated directly at the leading edge of the panhead, and its St is 0.13. The cavity’s vortex shedding frequency is in agreement with that calculated by the Rooster empirical formula. The two scale models exhibit some similar characteristics in distribution of sound source energy, but the energy distribution of the 1/8 model is more concentrated in the middle and lower regions. The contribution rates of their middle and lower regions to the radiated noise in the two models are 27.3% and 87.2%, respectively. The peak frequencies of the radiated noise from the 1/1 model are 307 and 571 Hz. The 307 Hz is consistent with the frequency of panhead vortex shedding, and the 571 Hz is more likely to be the result of the superposition of various components. In contrast, the peak frequencies of the radiated noise from the 1/8 scale model are 280 and 1970 Hz. The 280 Hz comes from the shear layer oscillation between the cavity and the bottom frame, and the 1970 Hz is close to the frequency at which the panhead vortex sheds. This shows that the scaled model results need to be corrected before applying to the full-scale model.