Aiming at the problem that the signal features of rolling bearing fault diagnosis need to be extracted and selected manually, which affects the accuracy of fault classification, a rolling bearing fault diagnosis method based on LSTM Auto-Encoder is proposed in this paper. The Auto-Encoder can automatically learn useful features from the vibration signal. LSTM is used to process time series data. The LSTM network is used as the encoder and decoder of Auto-Encoder. Meanwhile, KL divergence is introduced to improve the loss function to better reconstruct the signal. Experiment shows that the proposed algorithm has good performance in multiclass classification, and puts forward a new direction for the automation and intelligence of rolling bearing fault diagnosis.
Bearing Fault Diagnosis Method of Bearing Based on LSTM Auto-Encoder
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 : 65 ; 582-591
2022-02-23
10 pages
Article/Chapter (Book)
Electronic Resource
English
Bearing Fault Diagnosis Method of Bearing Based on LSTM Auto-Encoder
British Library Conference Proceedings | 2022
|Fault diagnosis system and method combining convolution auto-encoder and logistic regression
European Patent Office | 2020
|