Asset Details
Do you wish to reserve the book?
Bearing fault diagnosis method based on SAGAN and improved ResNet
Hey, we have placed the reservation for you!
By the way, why not check out events that you can attend while you pick your title.
Oops! Something went wrong.
Looks like we were not able to place the reservation. Kindly try again later.
Are you sure you want to remove the book from the shelf?
Bearing fault diagnosis method based on SAGAN and improved ResNet
Do you wish to request the book?
Bearing fault diagnosis method based on SAGAN and improved ResNet
Please be aware that the book you have requested cannot be checked out. If you would like to checkout this book, you can reserve another copy
We have requested the book for you!
Your request is successful and it will be processed during the Library working hours. Please check the status of your request in My Requests.
Oops! Something went wrong.
Looks like we were not able to place your request. Kindly try again later.
Journal Article
Bearing fault diagnosis method based on SAGAN and improved ResNet
2025
Overview
Traditional rolling bearing fault diagnosis methods struggle to adaptively extract features under complex industrial environments, and obtaining large and rich fault data under real operating conditions is difficult and expensive. Aiming at these issues, a bearing fault diagnosis method based on Self-Attention Generative Adversarial Networks (SAGAN) and Improved Deep Residual Networks (IResNet) was proposed (SAGAN_IResNet). Firstly, the original vibration signals are transformed into two-dimensional time–frequency images using continuous wavelet transform, providing both time domain and frequency domain information. Secondly, SAGAN is used to generate new samples similar to the original sample distribution, thereby expanding the data. Furthermore, a bearing fault diagnosis model is constructed using an improved residual network that incorporates the Multi-head Self-Attention (MHA) to adaptively obtain the global feature information, alleviate the problem of gradient dispersion and network degradation, and enhance the model’s diagnostic performance in the presence of strong noise and variable load conditions. Experimental verification is conducted using bearing datasets from Case Western Reserve University, Southeast University and Jiangnan University. The results show that the method proposed in this paper has strong bearing fault diagnosis performance under the condition of few samples, strong noise and variable load.
Publisher
Nature Publishing Group UK,Nature Publishing Group,Nature Portfolio
Subject
