A novel transfer learning fault diagnosis method for rolling bearing based on feature correlation matching
- 8 September 2022
- journal article
- research article
- Published by IOP Publishing in Measurement Science and Technology
- Vol. 33 (12) , 125006
Abstract
As one of the mainstream transfer learning methods, correlation alignment (CORAL) has been widely applied in fault diagnosis field and has achieved certain achievements. However, CORAL ignores the differences between domains in the matching process, which makes it difficult to measure the discrepancies between domains accurately. To compensate the shortcomings of the CORAL, this paper proposes a new feature correlation matching (FCM) method, and further it is applied as the objective function to propose a deep feature correlation matching network (DFCMN). The FCM method focuses on both first-order feature correlation and second-order feature correlation of the source and target domains, which measures the discrepancies between different domains more comprehensively and accurately. With the powerful feature mapping capability of neural network, the DFCMN can improve the feature similarity in different domain centers while reducing the discrepancies of feature distribution between different domains, so as to obtain more reliable shared features and improve the cross-work-conditions diagnosis accuracy. The effectiveness of the proposed method is verified through multiple transfer tasks utilizing public rolling bearing data sets.Keywords
Funding Information
- National Natural Science Foundation of China (52075310)
- Natural Science Foundation of Anhui Provincial Education Department (No.KJ2021A1086)
This publication has 32 references indexed in Scilit:
- Intelligent Bearing Fault Diagnosis Method Combining Compressed Data Acquisition and Deep LearningIEEE Transactions on Instrumentation and Measurement, 2017
- Nonlinear Chirp Mode Decomposition: A Variational MethodIEEE Transactions on Signal Processing, 2017
- Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark studyMechanical Systems and Signal Processing, 2015
- Sequential Fuzzy Diagnosis Method for Motor Roller Bearing in Variable Operating Conditions Based on Vibration AnalysisSensors, 2013
- Chatter identification in end milling process using wavelet packets and Hilbert–Huang transformInternational Journal of Machine Tools and Manufacture, 2013
- Active noise cancellation algorithms for impulsive noiseMechanical Systems and Signal Processing, 2013
- Early fault diagnosis of rotating machinery based on wavelet packets—Empirical mode decomposition feature extraction and neural networkMechanical Systems and Signal Processing, 2012
- Rolling element bearing diagnostics—A tutorialMechanical Systems and Signal Processing, 2011
- Rotating machine fault diagnosis using empirical mode decompositionMechanical Systems and Signal Processing, 2008
- An improved Hilbert–Huang transform and its application in vibration signal analysisJournal of Sound and Vibration, 2005