Abstract
Directing against the poor performance of motor bearing fault diagnosis, a fault diagnosis method of motor bearing based on Gaussian filter denoising, Hilbert transform envelope extraction, and Convolutional Neural Network is proposed. Gaussian filtering and Hilbert transform envelope extraction are used to preprocess the original data. The feature calculation of time-domain, frequency-domain, and wavelet singular entropy is carried out, and then standardized processing is carried out. Finally, Convolutional Neural Network is used for classification. The Convolutional Neural Network model is experimentally verified using the data set of the Case Western Reserve University Bearing Data Center. The experimental results can testify that the algorithm can achieve an accuracy of 98% for an 8-type motor fault diagnosis.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call