期刊
KNOWLEDGE-BASED SYSTEMS
卷 232, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.knosys.2021.107488
关键词
Fault diagnosis; Rolling bearing; Few-shot learning; Unsupervised learning; Generative adversarial network
资金
- National Natural Science Foundation of China [U1933101, 51875436, 91960106, 51965013, 51421004]
- China Postdoctoral Science Foundation [2020T130509, 2018M631145]
- Liuzhou Natural Science Foundation [2021AAA0112]
- Guangxi Natural Science Foundation Program [2020GXNSFAA159081]
- Fundamental Research Funds for the Central Universities, China [XZY022021006]
In this study, an unsupervised representation learning method called BIMGAN was proposed for fault diagnosis of rotating machinery under small sample size conditions. By maximizing mutual information estimation and feature matching strategy, the method learns the mapping relationship between samples and their feature representations, achieving average accuracies of 99.73% and 98.36% in case studies.
The abnormal detection of rotating machinery under small sample size conditions is of prime importance in the field of fault diagnosis. In this work, we proposed an unsupervised representation learning method called Bidirectional InfoMax GAN (BIMGAN), which can perform fast and effective feature extraction and fault recognition with few samples. First, we obtain the low-dimensional feature representation by a prior normalized encoder and reconstruction of the sample via the generator. Second, the mapping relationship between the sample and its corresponding feature representation is learned by maximizing mutual information estimation with the constraint of the feature matching (FM) strategy. Different from the general GANs, we are aiming at learning a good feature mapping of an encoder to capture the feature representation instead of reconstructing realistic samples. And then, a supervised pattern recognition task based on the feature representation is conducted for fault diagnosis. Finally, the inverse mapping learned by the encoder is visualized and the effectiveness is demonstrated. And the performance of the proposed method outperforms several advanced unsupervised methods on two case studies of rolling bearings fault recognition with some standard architectures, where the average accuracy can achieve 99.73% and 98.36% respectively. (C) 2021 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据