Open AccessA fault diagnosis equipment of motor bearing based on sound signal and CNN
Author(s) -
Dong Liu,
Binbin Li,
Chaoqun Wang,
Pengyu Cheng,
Bin Jiao
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2010/1/012159
Subject(s) - bearing (navigation) , fault (geology) , spectrogram , computer science , signal (programming language) , convolutional neural network , wavelet transform , sound (geography) , wavelet , speech recognition , acoustics , pattern recognition (psychology) , artificial intelligence , geology , seismology , programming language , physics
As an important part of mechanical equipment, the motor bearing damage rate is very high. In order to realize the fast and accurate diagnosis of motor bearing faults, this paper designs a fault diagnosis equipment based on sound signals. First, perform wavelet transform on the collected sound signal, then use the spectrogram generated by the fast Fourier transform to preliminarily determine whether the motor bearing is faulty, and finally use the convolutional neural network model that has been imported into the processor to diagnose the faulty parts of the motor bearing, The accuracy rate is above 98.41%.