Name
Title | ||
|---|---|---|
Understanding and improving deep learning-based rolling bearing fault diagnosis with attention mechanism. |
Abstract | ||
|---|---|---|
•A novel deep learning-based method with attention mechanism is proposed for rolling bearing fault diagnosis.•Efforts are made to understand the signal processing mechanism of deep learning, and the relationship between the well-established fault diagnosis knowledge and the ’black box’ data-driven approach is revealed.•The superiority of the proposed method on extracting the most discriminative features of different bearing health conditions is demonstrated by comparisons with existing methods and latest related researches.•The results of the proposed method are promising even for the difficult cross-domain fault diagnosis tasks with very limited labeled training data. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1016/j.sigpro.2019.03.019 | Signal Processing |
Keywords | Field | DocType |
Rolling element bearing,Fault diagnosis,Deep learning,Envelope spectrum,Attention mechanism | Training set,Black box (phreaking),Mathematical optimization,Bearing (mechanical),Artificial intelligence,Deep learning,Artificial neural network,Discriminative model,Mathematics,Machine learning | Journal |
Volume | ISSN | Citations |
161 | 0165-1684 | 9 |
PageRank | References | Authors |
0.59 | 0 | 3 |