MWGCN

This repo is the source code of the paper Multiscale Wavelet-Driven Graph Convolutional Network for Blade Icing Detection of Wind Turbines

1 Prerequisites

• Python 3.6.12
• PyTorch 1.0.0
• math, sklearn, numpy

2 Datasets

The experimental data are SCADA data from two wind turbines installed in the Inner Mongolia Autonomous Region, China, which were provided by the Chinese Ministry of Industry and Information Technology (MIIT). For the dataset acquisition, please refer to http://106.38.3.217/pages/datadown/dataset/dataSet06.jsp, contact the data holder industrial-bigdata@caict.ac.cn for authorisation.

3 Running

For model training, please run train.py. For model testing, please run test.py

For more information and usage questions, please contact us: ryanlai.cs@gmail.com

4 Reference

If you use MWGCN for your research, please cite the following paper.

     
@article{lai2022multiscale,
  title={Multiscale Wavelet-Driven Graph Convolutional Network for Blade Icing Detection of Wind Turbines},
  author={Lai, Zhichen and Cheng, Xu and Liu, Xiufeng and Huang, Lizhen and Liu, Yongping},
  journal={IEEE Sensors Journal},
  year={2022},
  publisher={IEEE}
}