Open AccessProbabilistic Neural Network Motor Bearing Fault Diagnosis Based on Improved Feature Extraction
Author(s) -
Fanchao Min,
Jingyu Xue,
Fang Ma
Publication year - 2020
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 - 10.1088/1742-6596/1684/1/012158
Subject(s) - bearing (navigation) , fault (geology) , feature extraction , artificial neural network , computer science , waveform , probabilistic logic , pattern recognition (psychology) , artificial intelligence , wavelet , probabilistic neural network , feature vector , engineering , time delay neural network , seismology , geology , telecommunications , radar
According to the fault characteristics of the rolling bearing in the motor, Db6 wavelet is used to decompose the waveform data, and the feature vector extraction method is improved. The probabilistic neural network is used to diagnose the fault category of the rolling bearing of the motor, and compared with the traditional feature extraction method. Experiments show that the accuracy of the method proposed in this paper reaches 98%, and it has a very broad application prospect in fault diagnosis in actual industrial production.