Certainty and Transferability Guided Few-Shot Open-Set Cross-Domain Fault Diagnosis

作     者：An, Yiyao Zhang, Ke Chai, Yi Zhu, Zhiqin Li, Yuanyuan 

作者机构：Chongqing Univ Posts & Telecommun Sch Automat Chongqing 400065 Peoples R China Chongqing Univ Sch Automat Chongqing 400044 Peoples R China 

出 版 物：《IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS》 (IEEE Trans. Ind. Inf.)

年 卷 期：2025年第21卷第4期

页      面：2997-3006页

核心收录：

学科分类：1201[管理学-管理科学与工程(可授管理学、工学学位)] 0808[工学-电气工程] 08[工学] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：National Natural Science Foundation of China [62403091, 62276037] Natural Science Foundation of Chongqing [CSTB2024NSCQ-MSX0923, cstc2021jcyj-msxmX0891] Science and Technology Research Program of Chongqing Municipal Education [KJQN202400618] Special key project of Chongqing technology innovation and application development [CSTB2024TIAD-STX0023, CSTB2023TIAD-KPX0088, CSTB2022TIAD-KPX0039] 

主　　题：Feature extraction Fault diagnosis Employee welfare Prototypes Informatics Few shot learning Degradation Correlation Contrastive learning Vectors Cross-domain fault diagnosis domain adaptation few-shot learning open-set learning prototypical contrastive learning (PCL) 

摘      要：A certainty and transferability guided few-shot domain adaptation network is proposed to address few-shot open-set cross-domain fault diagnosis in this article. The proposed method is composed of a feature extractor, a certainty-guided prototypical contrastive module and a transferability weighting domain adaptation module. The certainty-guided prototypical contrastive module based on samples informative importance is designed to enhance the data sensitivity with limited samples while achieving well class separation for open-set scenarios. The module infers informative importance of samples to guide method learn more effective representations. Meanwhile, correlation and uniformity principles are incorporated to alleviate prototype collapse. The transferability weighting domain adaptation module is designed to address great domain gaps and negative transfer caused by asymmetrical label spaces. The module quantifies sample transferability and down-weights the irrelevant samples based on their transferability scores. Experimental results on few-shot open-set cross-domain bearing fault diagnosis tasks demonstrated the superior and effectiveness of the proposed method.