Transfer Learning for Fault Diagnosis

迁移学习 故障诊断 深度神经网络

This repository is for the transfer learning or domain adaptive with fault diagnosis.

It should be notice that we use the tensorflow 1.15. If one use the lastest tensorflow, there will be some errors.

The paper is as follow:

Domain Adaptation with Multilayer Adversarial Learning for Fault Diagnosis of Gearbox under Multiple Operating Conditions

A Deep Transfer Model With Wasserstein Distance Guided Multi-Adversarial Networks for Bearing Fault Diagnosis Under Different Working Conditions

Research on variational mode decomposition in rolling bearings fault diagnosis of the multistage centrifugal pump

Citation

If you use this code and datasets for your research, please consider citing:

@inproceedings{zhang2019domain,
  title={Domain Adaptation with Multilayer Adversarial Learning for Fault Diagnosis of Gearbox under Multiple Operating Conditions},
  author={Zhang, Ming and Lu, Weining and Yang, Jun and Wang, Duo and Bin, Liang},
  booktitle={2019 Prognostics and System Health Management Conference (PHM-Qingdao)},
  pages={1--6},
  year={2019},
  organization={IEEE}
}

@ARTICLE{8713860, 
author={M. {Zhang} and D. {Wang} and W. {Lu} and J. {Yang} and Z. {Li} and B. {Liang}}, 
journal={IEEE Access}, 
title={A Deep Transfer Model With Wasserstein Distance Guided Multi-Adversarial Networks for Bearing Fault Diagnosis Under Different Working Conditions}, 
year={2019}, 
volume={7}, 
number={}, 
pages={65303-65318}, 
keywords={Fault diagnosis;Rolling bearings;Data models;Wavelength division multiplexing;Convolution;Employee welfare;Training;Transfer learning;fault diagnosis;convolutional neural network;multi-adversarial networks}, 
doi={10.1109/ACCESS.2019.2916935}, 
ISSN={2169-3536}, 
month={},}

@article{zhang2017research,
  title={Research on variational mode decomposition in rolling bearings fault diagnosis of the multistage centrifugal pump},
  author={Zhang, Ming and Jiang, Zhinong and Feng, Kun},
  journal={Mechanical Systems and Signal Processing},
  volume={93},
  pages={460--493},
  year={2017},
  publisher={Elsevier}
}

Contact

If you have any problem about our code, feel free to contact:

[email protected] or [email protected]

or describe your problem in issues