Intelligent fault diagnosis of rotating machinery based on continuous wavelet transform-local binary convolutional neural network
作者：
Highlights：
• A local binary convolutional neural network-based intelligent fault diagnosis approach is proposed for rotating machinery, where no manually extracted features are required.
• By utilizing local binary convolution layer, the proposed continuous wavelet transform-local binary convolution neural network has much lower model complexity compared with tradition CNN.
• The proposed method can be applied on single and compound faults diagnosis of rotating machinery, which has been proved in the bearing fault diagnosis case and gearbox compound fault diagnosis case.
摘要
•A local binary convolutional neural network-based intelligent fault diagnosis approach is proposed for rotating machinery, where no manually extracted features are required.•By utilizing local binary convolution layer, the proposed continuous wavelet transform-local binary convolution neural network has much lower model complexity compared with tradition CNN.•The proposed method can be applied on single and compound faults diagnosis of rotating machinery, which has been proved in the bearing fault diagnosis case and gearbox compound fault diagnosis case.
论文关键词：Intelligent fault diagnosis,Local binary convolution neural network,Continuous wavelet transform,Rotating machinery,Deep learning
论文评审过程：Received 16 January 2020, Revised 17 January 2021, Accepted 19 January 2021, Available online 22 January 2021, Version of Record 27 January 2021.
论文官网地址：https://doi.org/10.1016/j.knosys.2021.106796
