Tensorflow Code for our paper "Multi-mode data augmentation and fault diagnosis of rotating machinery using modified ACGAN designed with new framework".

Paper Link: (https://doi.org/10.1016/j.aei.2022.101552)

If you find our work useful in your research, please consider citing:

@inproceedings{MACGAN-for-fault-diagnosis-of-rotating-machinery,
  title={Multi-mode data augmentation and fault diagnosis of rotating machinery using modified ACGAN designed with new framework},
  author={Wei Li, Xiang Zhong, Haidong Shao, Baoping Cai, and Xingkai Yang},
  booktitle={Advanced Engineering Informatics},
  year={2022}
}

• Python 3.8
• Tensorflow 2.4.0
• Numpy 1.19.5

the bearing and gear fault datasets are collected from Case Western Reserve University bearing data center, gear vibration dataset of University of Connecticut.

In this paper, a signal-to-image conversion method [50] is used to transform the original 1D vibration signals into 2D gray images. code: signal-to-image conversion [50] L. Wen, X. Li, L. Gao, Y. Zhang, A new convolutional neural network-based data-driven fault diagnosis method, IEEE Trans. Ind. Electron., 65, pp. 5990-5998, July. 2018. Paper Link: (https://ieeexplore.ieee.org/document/8114247)

Before begin, please separate the training data of different health states into folders in foldergray_images\\. The code for training MACGAN and generate gray images istraining and generating.py, you may also customize the parameters in config part in training and generating.py. The generated images will be saved in foldergenerated images\\.

Our SpectralNormalization code is based on (https://github.com/IShengFang/SpectralNormalizationKeras)

If you have any questions about the codes or would like to communicate about intelligent fault diagnosis, fault detection, please contact us: [email protected] Mentor E-mail：[email protected]