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.

A Fault Detection Strategy for Wheel Flat Scars with Wavelet Neural Network and Genetic Algorithm

Railway Rolling Bearing Faults Diagnosis Based On Wavelet Packet And Probabilistic Neural Network

Improved Wavelet Neural Network Based on Hybrid Genetic Algorithm Applicationin on Fault Diagnosis of Railway Rolling Bearing.

Fault Diagnosis Method for the Rolling Bearing of Railway Vehicle Based on Wavelet Packet Transform and BP Neural Network

Defect Prediction of Railway Wheel Flats based on Hilbert Transform and Wavelet Packet Decomposition

Wheel Defect Detection Using Attentive Feature Selection Sequential Network With Multidimensional Modeling of Acoustic Emission Signals

Automatic Recognition and Positioning of Wheel Defects in Ultrasonic B-Scan Image Using Artificial Neural Network and Image Processing

Wayside Detection of Wheel Minor Defects in High-Speed Trains by a Bayesian Blind Source Separation Method

A Bayesian machine learning approach for online detection of railway wheel defects using track-side monitoring

A Method for Railway Gearbox Faults Detection Based on Time- Frequency Feature Parameters and Genetic Algorithm Neural Network

Improved Wavelet De-Noising Method of Rail Vibration Signal for Wheel Tread Detection

Deep learning-based fault detection in railway wheelsets using time series analysis

Wavelet Scattering and Neural Networks for Railhead Defect Identification

Wheel fault diagnosis model based on multichannel attention and supervised contrastive learning

The Study on The Application of BP Neural Network Based on Genetic Algorithm Optimization in Magnetic Flux Leakage (MFL) of Rail Top Surface Stripping Defects

Detection of Wheel Polygonization Based on Wayside Monitoring and Artificial Intelligence

An Integrated Model of kNN and GBDT for Fault Diagnosis of Wheel on Railway Vehicle

The Detection of Wheel-Flats Based on Fiber Optic Bragg Grating Array

OORNet: A deep learning model for on-board condition monitoring and fault diagnosis of out-of-round wheels of high-speed trains

Rolling Element Bearing Fault Diagnosis Using Wavelet Neural Network

Railway Wheel Flat Detection System Based on a Parallelogram Mechanism