Asset Details
Do you wish to reserve the book?
Rolling Bearing Fault Diagnosis Method Based on Attention CNN and BiLSTM Network
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?
Rolling Bearing Fault Diagnosis Method Based on Attention CNN and BiLSTM Network
Do you wish to request the book?
Rolling Bearing Fault Diagnosis Method Based on Attention CNN and BiLSTM Network
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
Rolling Bearing Fault Diagnosis Method Based on Attention CNN and BiLSTM Network
2023
Overview
To solve the problems that existing bearing fault diagnosis methods cannot adaptively select features and are difficult to deal with noise interference, an end-to-end fault diagnosis method is proposed based on attention CNN and BiLSTM (ACNN-BiLSTM). In the proposed method, the raw vibration acceleration signal of the bearing is taken as the input, the short-term spatial features are extracted through a one-dimensional wide convolutional neural network, and the batch normalization algorithm is used to improve the stability of the data distribution. Following, a convolutional block attention module is introduced to redistribute the weights between different feature dimensions, enhancing the model's attention to important features. Finally, the attention-weighted features are sent to BiLSTM for further feature extraction, and the softmax classifier is used for fault diagnosis. The proposed method is compared with advanced algorithms such as WCNN-BiGRU on the CWRU public dataset. The experimental results show that ACNN-BiLSTM has the highest accuracy, recall, and F1-Measure. Even under the extreme noise interference condition of SNR = 10 dB, ACNN-BiLSTM can achieve a diagnostic accuracy of 96.58%. In addition, the proposed method is also used for fault diagnosis of bearing measured data of the VALENIAN-PT500 test bench. The results show that the average diagnostic accuracy of ACNN-BiLSTM is up to 99.79%, which has strong generality and is superior to other advanced comparison methods.
Publisher
Springer US,Springer Nature B.V
Subject
/ Entropy
