Weighted quantile discrepancy-based deep domain adaptation network for intelligent fault diagnosis
作者：
Highlights：
• A novel WQD metric is proposed to measure the discrepancy of distributions.
• It is then introduced to develop a DWQDAN model for intelligent fault diagnosis.
• The weights in DWQDAN are flexible to focus on different quantiles of distributions.
• The efficacy of DWQDAN is illustrated by the public CWRU and Paderborn datasets.
• The results show the DWQDAN model is promising and suitable for fault diagnosis.
摘要
•A novel WQD metric is proposed to measure the discrepancy of distributions.•It is then introduced to develop a DWQDAN model for intelligent fault diagnosis.•The weights in DWQDAN are flexible to focus on different quantiles of distributions.•The efficacy of DWQDAN is illustrated by the public CWRU and Paderborn datasets.•The results show the DWQDAN model is promising and suitable for fault diagnosis.
论文关键词：Domain adaptation,Deep learning,Intelligent fault diagnosis
论文评审过程：Received 5 August 2021, Revised 26 December 2021, Accepted 3 January 2022, Available online 10 January 2022, Version of Record 24 January 2022.
论文官网地址：https://doi.org/10.1016/j.knosys.2022.108149
