Existing fault diagnosis methods based on signal processing rely on manual feature extraction, the deep learning method can not learn the time domain and frequency domain characteristics of vibration signal at the same time. The time domain and frequency domain can be considered by short-time Fourier transform. Based on this principle and the Gate Recurrent Unit network (GRU) and Short time Fourier transform (STFT), this paper propose an improved GRU fault diagnosis model. The model can learn the characteristics of both time and frequency domain directly from the original vibration signal, and realize the end-to-end fault diagnosis. Using the Western Reserve University Bearing Data (CWRU) for training and testing, the test accuracy can reach 98.6%. Finally, compared with the traditional LSTM and GRU, the effectiveness and advantages of the improved model are proved.
Rolling Bearing Fault Diagnosis Based on Improved GRU
Lect. Notes Electrical Eng.
International Conference on Electrical and Information Technologies for Rail Transportation ; 2021 October 21, 2021 - October 23, 2021
Proceedings of the 5th International Conference on Electrical Engineering and Information Technologies for Rail Transportation (EITRT) 2021 ; Chapter : 48 ; 430-440
2022-02-23
11 pages
Article/Chapter (Book)
Electronic Resource
English
Rolling Bearing Fault Diagnosis Based on Improved GRU
British Library Conference Proceedings | 2022
|Rolling Bearing Fault Diagnosis Based on Improved GRU
TIBKAT | 2022
|Autoregressive Model-Based Vibration Fault Diagnosis of Rolling Bearing
Online Contents | 2010
|A fault diagnosis method based on improved parallel convolutional neural network for rolling bearing
SAGE Publications | 2023
|The Fault Diagnosis Method of Rolling Bearing Based on CEEMDAN
British Library Conference Proceedings | 2022
|