Abstract:Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, Ahead of Print. The existing data-driven fault diagnosis methods face some significant problems in practical applications. Many traditional methods rely on a large number of high-quality labeled data for training, but in the industrial environment, the actual fault data obtained is often limited and unbalanced. This data scarcity seriously limits the diagnostic ability of the model and is prone to insufficient diagnostic accuracy. In addition, the data-driven method has a strong dependence on data, and it is prone to misjudgment in the face of complex environments such as noise interference and equipment state changes. These problems jointly restrict the application effect of fault diagnosis methods in industrial actual scenarios. Based on this, this paper proposes a new method of rolling bearing fault diagnosis based on digital twin technology and improved convolutional neural network (CNN)-Transformer deep learning model. Firstly, the geometric characteristics and motion mechanism of rolling bearings are analyzed in depth, and a high-fidelity virtual twin model is established. A balanced simulation data set is generated by numerical simulation. Secondly, we improve the traditional CNN, combined with the Transformer deep learning framework, to enhance the ability of the network to extract features. By performing wavelet transform on the test data obtained from the rolling bearing acceleration test bench and the simulation data generated by the twin model, a dual-channel signal of parallel convolution is formed, and a fault diagnosis model based on dual-channel parallel CNN-Transformer is constructed. Finally, the effectiveness of the proposed method is verified by ablation experiments. The results show that the proposed method can accurately and efficiently identify different rolling bearing fault modes and has superior diagnostic performance. At the same time, the model can also be further extended to related fields to provide new ideas and technical references for fault diagnosis of other mechanical equipment.

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

A fault diagnosis method for rolling bearings based on graph neural network with one-shot learning

Exploiting Graph Neural Network with One-Shot Learning for Fault Diagnosis of Rotating Machinery

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

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

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

Intelligent fault diagnosis of rolling bearings based on the visibility algorithm and graph neural networks

Transfer Learning for Bearing Fault Diagnosis basedon Graph Neural Network with Dilated KNN andAdversarial Discriminative Domain Adaptation

Rolling bearing fault diagnosis method using time-frequency information integration and multi-scale TransFusion network

Multi-input Parallel Graph Neural Network for Semi-supervised Rolling Bearing Fault Diagnosis

A Novel Method for Rolling Bearing Fault Diagnosis Based on Gramian Angular Field and CNN-ViT

Spatial-temporal graph feature learning driven by time–frequency similarity assessment for robust fault diagnosis of rotating machinery

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

An Intelligent Fault Diagnosis Method of Rolling Bearings Based on Short-Time Fourier Transform and Convolutional Neural Network

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

Research on digital twin-assisted dual-channel parallel convolutional neural network-transformer rolling bearing fault diagnosis method

A hybrid fault diagnosis method for rolling bearings based on GGRU-1DCNN with AdaBN algorithm under multiple load conditions

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

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

MSGAFN: Multi-scale graph attention fusion network for machine fault diagnosis

ASG-HOMGAT: A high-order multi-head graph attention network with adaptive small graph structure for rolling bearing fault diagnosis