Joint distribution adaptation network with adversarial learning for rolling bearing fault diagnosis

作者：

Highlights：

• A joint distribution adaptation network with adversarial learning is developed to tackle the transfer issues.

• Improved joint maximum mean discrepancy is proposed to precisely measure the joint distribution discrepancy.

• Substantial transfer cases are carried out to demonstrate the effectiveness of our method.

摘要

•A joint distribution adaptation network with adversarial learning is developed to tackle the transfer issues.•Improved joint maximum mean discrepancy is proposed to precisely measure the joint distribution discrepancy.•Substantial transfer cases are carried out to demonstrate the effectiveness of our method.

论文关键词：Unsupervised cross-domain fault diagnosis,Joint distribution adaptation,Improved joint maximum mean discrepancy,Adversarial domain adaptation

论文评审过程：Received 17 December 2020, Revised 4 March 2021, Accepted 17 March 2021, Available online 24 March 2021, Version of Record 1 April 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.106974