The fault diagnosis method of rolling bearing under variable working conditions based on deep transfer learning
- PDF / 3,310,682 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 76 Downloads / 234 Views
(2020) 42:585
TECHNICAL PAPER
The fault diagnosis method of rolling bearing under variable working conditions based on deep transfer learning Shaojiang Dong1 · Kun He1 · Baoping Tang2 Received: 5 December 2019 / Accepted: 1 October 2020 © The Brazilian Society of Mechanical Sciences and Engineering 2020
Abstract The vibration signals of rolling bearing obtained under variable working conditions do not obey the same independent distribution so that the traditional method of bearing fault diagnosis has low accuracy, a fault diagnosis method about rolling bearing based on sparse denoising autoencoder (SDAE) for deep feature extraction combining transfer learning is proposed. First, the bearing vibration signal in the time domain is transformed for frequency domain signal via Fourier transform, which is input into the SDAE for adaptive deep feature extraction. Then, the joint geometrical and statistical alignment is introduced to deal with the deep feature samples for reducing the domain discrepancy both statistically and geometrically. Finally, the k-nearest neighbor classification algorithm is used for completing the fault diagnosis of rolling bearing under variable working conditions. The experimental results show that the method presented in the paper improves the accuracy rate of fault diagnosis about rolling bearing under variable working conditions, verifies its feasibility and effectiveness. Keywords Rolling bearing · Sparse denoising autoencoder · Variable working conditions · Fault diagnosis · Transfer learning
1 Introduction Rolling bearings are an extremely vital part of mechanical equipment, whose running state determines directly the performance of the whole rotating mechanical system, so the research on bearing fault diagnosis is very important [1, 2]. In recent years, the research methods about bearing fault diagnosis are mostly under the assumption of constant working condition [3–5]. When these methods are applied under variable working conditions, the diagnostic effect will be reduced. Therefore, the research on the fault diagnosis and condition monitoring methods of rolling bearing under different working conditions is greatly significant to ensure the healthy operation of mechanical equipment [6].
Technical Editor: Marcelo Areias Trindade. * Kun He [email protected] 1
School of Mechatronics and Vehicle Engineering, Chongqing Jiaotong University, Chongqing 400074, China
State Key Laboratory of Mechanical Transmission, Chongqing University, Chongqing 400044, China
2
The traditional fault diagnosis methods mainly use signal processing methods such as time domain [7], frequency domain and time–frequency domain analysis [8–10], or machine learning methods instancing BP neural network or support vector machine and so on [11]. Despite these methods have achieved good results in bearing fault diagnosis, there still exists some problems that the fault feature can be not fully learned and feature selection transitively depends on subjectivity and diagnosis experience. In recent years, dee
Data Loading...
