Cross-domain intelligent bearing fault diagnosis under class imbalanced samples via transfer residual network augmented with explicit weight self-assignment strategy based on meta data
作者:
Highlights:
• A novel cross-domain transfer fault diagnosis method for class imbalanced samples is proposed.
• Fault feature extractor based on deep residual network is constructed to avoid gradient disappearance and improve the diagnosis performance.
• Cross-domain transfer is carried out to reduce the degree of difficulty in diagnosis.
• Explicit weight self-assignment strategy based on meta data is adopted to optimize the sample weighting process with class imbalance.
摘要
•A novel cross-domain transfer fault diagnosis method for class imbalanced samples is proposed.•Fault feature extractor based on deep residual network is constructed to avoid gradient disappearance and improve the diagnosis performance.•Cross-domain transfer is carried out to reduce the degree of difficulty in diagnosis.•Explicit weight self-assignment strategy based on meta data is adopted to optimize the sample weighting process with class imbalance.
论文关键词:Fault diagnosis,Transfer learning,Imbalanced data,ResNet,Meta data
论文评审过程:Received 22 March 2022, Revised 11 June 2022, Accepted 11 June 2022, Available online 17 June 2022, Version of Record 28 June 2022.
论文官网地址:https://doi.org/10.1016/j.knosys.2022.109272