Zero-shot learning for compound fault diagnosis of bearings

作者：

Highlights：

• A novel compound fault diagnosis method based on zero-shot learning.

• Model trained with the vibration data of single faults to identify unknown compound fault.

• A semantic vector for the single and compound faults that require no expert knowledge.

摘要

•A novel compound fault diagnosis method based on zero-shot learning.•Model trained with the vibration data of single faults to identify unknown compound fault.•A semantic vector for the single and compound faults that require no expert knowledge.

论文关键词：Fault diagnosis,Zero-shot Learning,Compound fault diagnosis,Semantics

论文评审过程：Received 21 June 2021, Revised 22 September 2021, Accepted 4 November 2021, Available online 17 November 2021, Version of Record 23 November 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116197