A two-stage domain alignment method for multi-source domain fault diagnosis. (15th June 2023)

• Record Type: Journal Article
• Authors: Cao, Wei; Meng, Zong; Sun, Dengyun; Liu, Jingbo; Guan, Yang; Cao, Lixiao; Li, Jimeng; Fan, Fengjie
• Is Part Of: Measurement. Volume 214(2023)
• Publication Date: 2023-06-15
• Subjects: Multi-source -- Domain adaptation -- Fault diagnosis -- Pseudo label; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2023.112818
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: A two-stage network framework containing feature extraction from multiple domains. A two-stage domain adaptation method based on global and class granularity. A pseudo-tag filtering method based on high-confidence multi-classifier voting. Abstract: The issue of restricted target domain tags and constrained information offered by a single source domain in the intelligent fault diagnosis may be successfully resolved by multi-source domain adaptation. Compared with traditional domain adaptation methods, multisource domain adaptation methods face more difficult challenges: the differences between domains are more complex. Hence, a two-stage domain alignment method for multi-source domain fault diagnosis is proposed. The method is accomplished by developing a common feature extractor and several domain feature extractors and classifiers, along with the two-stage distribution adaptation method. Finally, the classifier is used to forecast target samples, while the voting mechanism is utilized to construct the target samples' pseudo labels. With the resulting pseudo labels, the final model training is completed.