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 Multichannel MN-GCN for Wheelset-Bearing System Fault Diagnosis

Multichannel Dynamic Graph Convolutional Network-Based Fault Diagnosis and Its Application in Blast Furnace Ironmaking Process

A Sound and Vibration Fusion Method for Fault Diagnosis of Rolling Bearings under Speed-Varying Conditions.

Fault Diagnosis of Wheeled Robot Based on Prior Knowledge and Spatial-Temporal Difference Graph Convolutional Network

Rolling Bearing Fault Diagnosis Based on Multi-Scale Weighted Visibility Graph and Multi-Channel Graph Convolution Network

A multi-scale graph convolutional network with contrastive-learning enhanced self-attention pooling for intelligent fault diagnosis of gearbox

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

Graph Convolutional Network Based on CQT Spectrogram for Bearing Fault Diagnosis

Graph neural network-based bearing fault diagnosis using Granger causality test

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

Semi-supervised fault diagnosis of wheelset bearings in high-speed trains using autocorrelation and improved flow Gaussian mixture model

Digital Twin-Driven Graph Convolutional Memory Network for Defect Evolution Assessment of Rolling Bearings

Improved GNN based on Graph-Transformer: A new framework for rolling mill bearing fault diagnosis

Multi-Scale Deep Graph Convolutional Networks for Intelligent Fault Diagnosis of Rotor-Bearing System Under Fluctuating Working Conditions

A Method for Rolling Bearing Fault Diagnosis Based on GSC-MDRNN with Multi-Dimensional Input

Rolling bearing fault diagnosis method based on generative adversarial enhanced multi-scale convolutional neural network model

Fault Diagnosis of Wheelset Bearings in High-Speed Trains Using Logarithmic Short-time Fourier Transform and Modified Self-calibrated Residual Network

A graph-guided collaborative convolutional neural network for fault diagnosis of electromechanical systems

Multi-scale Residual Neural Network with Enhanced Gated Recurrent Unit for Fault Diagnosis of Rolling Bearing

Lightweight Convolutional Neural Network and Its Application in Rolling Bearing Fault Diagnosis under Variable Working Conditions.

An Adaptive Weighted Multiscale Convolutional Neural Network for Rotating Machinery Fault Diagnosis under Variable Operating Conditions