Intelligent fault diagnosis of rolling bearings using a semi-supervised convolutional neural network
- PDF / 1,103,952 Bytes
- 17 Pages / 595.276 x 790.866 pts Page_size
- 70 Downloads / 213 Views
ORIGINAL SUBMISSION
Intelligent fault diagnosis of rolling bearings using a semi-supervised convolutional neural network Yaochun Wu 1,2
&
Rongzhen Zhao 1 & Wuyin Jin 1 & Tianjing He 1 & Sencai Ma 1 & Mingkuan Shi 1
Accepted: 5 October 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract The success of convolutional neural networks (CNNs) in intelligent fault diagnosis is largely dependent on massive amounts of labelled data. In a real-world case, however, massive amounts of labelled data are difficult or costly to collect, whereas abundant unlabelled data are often available. To utilize such unlabelled data, a novel method using a semi-supervised convolutional neural network (SSCNN) for intelligent fault diagnosis of bearings is proposed. First, a 1-d CNN is applied to learn class space features and generate class probabilities of unlabelled samples, based on which a class probability maximum margin criterion (CPMMC) method is used to construct the loss function of unlabelled samples. Then, the constructed loss function, which aims to maximise the inter-class distance of class space features and minimise the intra-class distance of class space features, is integrated into the cross-entropy loss function of the CNN, and the SSCNN is established. Finally, the SSCNN model is applied to analyse the vibration signals collected from rolling bearings, and a novel intelligent fault diagnosis method using the SSCNN is proposed. Two datasets are employed to validate the effectiveness of the proposed methodology. The results show that the established SSCNN can effectively utilise unlabelled samples to train the model and enhance its fault diagnosis performance. Through a comparison with commonly used semi-supervised deep learning methods, the superiority of the proposed method is validated. Keywords Convolutional neural network . Semi-supervised learning . Maximum margin criterion . Intelligent fault diagnosis . Rolling bearing
1 Introduction Rolling bearings, as machine parts used to connect and fix objects, often work in complex and variable environments for a long time [1–3]. A sudden failure of rolling bearings may lead to an unexpected breakdown, enormous economic losses, and even casualties. Therefore, it is meaningful to develop effective intelligent fault diagnosis methodologies to monitor the condition of rolling bearings [4–6].
* Rongzhen Zhao [email protected] Yaochun Wu [email protected] 1
School of Mechanical and Electrical Engineering, Lanzhou University of Technology, Lanzhou 730050, China
2
School of Mechanical Engineering, Anyang Institute of Technology, Anyang 455000, China
As a rising star in intelligent fault diagnosis, deep learning has attracted increasing attention in recent years [7–10]. This is because deep learning can perform multi- layer mapping of features, and it has feature adaptive learning abilities and especially obvious advantages in processing mechanical failure big data. Successful applications of convolutional neural networks (CNNs), as key d
Data Loading...
