Abstract
As wheels are important components of train operation, diagnosing and predicting wheel faults are essential to ensure the reliability of rail transit. Currently, the existing studies always separately deal with two main types of wheel faults, namely wheel radius difference and wheel flat, even though they are both reflected by wheel radius changes. Moreover, traditional diagnostic methods, such as mechanical methods or a combination of data analysis methods, have limited abilities to efficiently extract data features. Deep learning models have become useful tools to automatically learn features from raw vibration signals. However, research on improving the feature-learning capabilities of models under noise interference to yield higher wheel diagnostic accuracies has not yet been conducted. In this paper, a unified training framework with the same model architecture and loss function is established for two homologous wheel faults. After selecting deep residual networks (ResNets) as the backbone network to build the model, we add the squeeze and excitation (SE) module based on a multichannel attention mechanism to the backbone network to learn the global relationships among feature channels. Then the influence of noise interference features is reduced while the extraction of useful information features is enhanced, leading to the improved feature-learning ability of ResNet. To further obtain effective feature representation using the model, we introduce supervised contrastive loss (SCL) on the basis of ResNet + SE to enlarge the feature distances of different fault classes through a comparison between positive and negative examples under label supervision to obtain a better class differentiation and higher diagnostic accuracy. We also complete a regression task to predict the fault degrees of wheel radius difference and wheel flat without changing the network architecture. The extensive experimental results show that the proposed model has a high accuracy in diagnosing and predicting two types of wheel faults.
Highlights
In recent years, rail transportation business has developed rapidly in China, but the safety situation is increasingly serious, and guaranteeing the reliability of the rail system is still the top priority of the rail transportation business.[1]
Some recent papers applied deep residual networks (RESNet) to fault diagnoses; for example, Zhao et al.[29] built a multiwavelet regularized deep residual network (MWR-Deep residual network (DRN)) model and verified that the model can effectively improve the performance of fault diagnoses
residual networks (ResNets) + squeeze and excitation (SE) + supervised contrastive loss (SCL) makes it easier to distinguish the features of different fault classes, further improving the model’s noise resistance and fault diagnosis capabilities
Summary
Rail transportation business has developed rapidly in China, but the safety situation is increasingly serious, and guaranteeing the reliability of the rail system is still the top priority of the rail transportation business.[1]. If the operation status of key train components can be monitored, the transformation to a condition-based maintenance will be achieved to fulfill the requirements of reliability, availability, maintainability, and safety (RAMS). For the fault diagnosis and condition monitoring of trains, the faulting of key equipment is the biggest threat to the safe operation of rail vehicles, so studying the key equipment, such as bearings and wheels in train operation, has become the top priority of fault diagnosis.[2,3] Currently, bearing faults have been diagnosed in depth,[4,5,6] but the wheel, as an important component, is rarely examined in fault diagnosis. Wheels wear severely, and fault diagnosis is important
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
