CFs-focused intelligent diagnosis scheme via alternative kernels networks with soft squeeze-and-excitation attention for fast-precise fault detection under slow & sharp speed variations
作者：
Highlights：
• A novel alternative kernel network is proposed for fault detection under different speed variations.
• The implementation procedure is presented in details, while the kernel selection behaviors are revealed.
• Combining with proposed method, the performance of various modified models can be improved by 3%–12%.
• Five case studies indicate that the AkNet is superior to existing models in both identification accuracy and computational efficiency.
摘要
•A novel alternative kernel network is proposed for fault detection under different speed variations.•The implementation procedure is presented in details, while the kernel selection behaviors are revealed.•Combining with proposed method, the performance of various modified models can be improved by 3%–12%.•Five case studies indicate that the AkNet is superior to existing models in both identification accuracy and computational efficiency.
论文关键词：Intelligent fault diagnosis,Non-stationary data analysis,Convolution framework,Adaptive kernel selection
论文评审过程：Received 27 July 2021, Revised 10 November 2021, Accepted 18 December 2021, Available online 24 December 2021, Version of Record 6 January 2022.
论文官网地址：https://doi.org/10.1016/j.knosys.2021.108026
