Single and Multi-label Fault Classification in rotors from unprocessed multi-sensor data through deep and parallel CNN architectures

作者：

Highlights：

• Novel Deep CNN Architecture working directly on raw sensor data for fault diagnosis.

• Eliminates need to pre-process data statistically, or into FFT for feeding to CNN.

• Shallow–Parallel–Multiple-Binary architecture for Multi-Label Fault Classification.

摘要

•Novel Deep CNN Architecture working directly on raw sensor data for fault diagnosis.•Eliminates need to pre-process data statistically, or into FFT for feeding to CNN.•Shallow–Parallel–Multiple-Binary architecture for Multi-Label Fault Classification.

论文关键词：Deep learning,Multi-label,Fault diagnosis,Multi-sensor,Raw data,Rotor-Bearing

论文评审过程：Received 10 March 2021, Revised 25 May 2021, Accepted 4 July 2021, Available online 16 July 2021, Version of Record 23 July 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115565