Journal
SIGNAL PROCESSING
Volume 161, Issue -, Pages 136-154Publisher
ELSEVIER
DOI: 10.1016/j.sigpro.2019.03.019
Keywords
Rolling element bearing; Fault diagnosis; Deep learning; Envelope spectrum; Attention mechanism
Categories
Funding
- Fundamental Research Funds for the Central Universities [N170503012, N170308028, N180708009]
- National Natural Science Foundation of China [61871107]
Ask authors/readers for more resources
In the recent years, deep learning-based intelligent fault diagnosis methods of rolling bearings have been widely and successfully developed. However, the data-driven method generally remains a black box to researchers and there is a gap between the emerging neural network-based methods and the well-established traditional fault diagnosis knowledge. This paper proposes a novel deep learning-based fault diagnosis method for rolling element bearings. Attention mechanism is introduced to assist the deep network to locate the informative data segments, extract the discriminative features of inputs, and visualize the learned diagnosis knowledge. Experiments on a popular rolling bearing dataset intuitively show the effectiveness of the proposed method, which is able to provide reliable diagnosis even with very few training data. The experimental results suggest this research offers a promising tool for intelligent fault diagnosis and provides effort in understanding the underlying mechanism of deep neural network. (C) 2019 Elsevier B.V. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available