Domain Adversarial Graph Convolutional network (DAGCN)

This code is about the implementation of Domain Adversarial Graph Convolutional Network for Fault Diagnosis Under Variable Working Conditions.

Note

The DAGCN consists of a CNN and a MRF_GCN, and the framework of this code is based on Unsupervised Deep Transfer Learning for Intelligent Fault Diagnosis: An Open Source and Comparative Study.

Implementation

python ./DAGCN/train_advanced.py --model_name DAGCN_features --checkpoint_dir ./DAGCN/results/ --data_name CWRU --data_dir D:/Data/西储大学轴承数据中心网站 --transfer_task [3],[0] --last_batch True

Citation

MRF_GCN: @ARTICLE{MRF_GCN, author={T. {Li} and Z. {Zhao} and C. {Sun} and R. {Yan} and X. {Chen}}, journal={IEEE Transactions on Industrial Electronics}, title={Multi-receptive Field Graph Convolutional Networks for Machine Fault Diagnosis}, year={2020}, volume={}, number={}, pages={1-1}, doi={10.1109/TIE.2020.3040669}}

DAGCN: @ARTICLE{9410617, author={T. {Li} and Z. {Zhao} and C. {Sun} and R. {Yan} and X. {Chen}}, journal={IEEE Transactions on Instrumentation and Measurement}, title={Domain Adversarial Graph Convolutional Network for Fault Diagnosis Under Variable Working Conditions}, year={2021}, volume={70}, number={}, pages={1-10}, doi={10.1109/TIM.2021.3075016}}