Open AccessRolling Bearing Fault Diagnosis Method Based on Principal Components Analysis and Probabilistic Neural Network
Author(s) -
Yun Ni,
Dong Dong Ban
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/740/1/012012
Subject(s) - principal component analysis , redundancy (engineering) , artificial neural network , bearing (navigation) , probabilistic neural network , pattern recognition (psychology) , dimensionality reduction , probabilistic logic , computer science , artificial intelligence , dimension (graph theory) , fault (geology) , feature (linguistics) , data mining , mathematics , time delay neural network , linguistics , philosophy , seismology , pure mathematics , geology , operating system
In this paper, a rolling bearing fault diagnosismethod based on PCA and improved PNN network is proposed to solve the problems of high dimension, high redundancy, nonlinearity and nonstationarity of rolling bearingdata. Firstly, the principal components analysis (PCA) algorithm is used to extract the feature information of the original data and obtain the principal component informationafter dimension reduction. Then the principal component information is sent as a feature to the probabilistic neuralnetwork (PNN) for training and output the diagnosisresults. The method is verified using Case Western bearingdatasets. Through simulation comparison of this method and BP neural network method, the experimental results show that the method proposed in this paper is more accurate in bearing fault diagnosis.