The demand for intelligent fault diagnosis algorithms has increased dramatically in the field of aeroengines. Traditional bearing fault diagnosis algorithms mainly extract features manually, and then input them into the classification model for fault identification. As the scale of condition monitoring for mechanical equipment and the sampling frequency gradually increase, how to automatically extract useful features from massive amounts of data and accurately diagnose fault types has become a research hotspots. Due to the powerful feature extraction capabilities of deep learning, this study trained a TCN (temporal convolutional network) model to identify the vibration signals of rolling bearings with 10 different types of faults. The characteristics of different fault signals are extracted through dilated convolution, dropout and residual structure. Finally the generalization ability of the trained model is tested by the cross-validation method. The results show that the model can reach an accuracy of 98.7%, which proves that the method can effectively identify the fault types of rolling bearings.
Research on Fault Diagnosis Method of Rolling Bearing Based on TCN
2021-07-16
3476836 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Rolling Bearing Fault Diagnosis Based on Improved GRU
British Library Conference Proceedings | 2022
|Rolling Bearing Fault Diagnosis Based on Improved GRU
TIBKAT | 2022
|Rolling Bearing Fault Diagnosis Based on Improved GRU
Springer Verlag | 2022
|Autoregressive Model-Based Vibration Fault Diagnosis of Rolling Bearing
Online Contents | 2010
|The Fault Diagnosis Method of Rolling Bearing Based on CEEMDAN
Springer Verlag | 2022
|