Advanced Rolling Bearing Fault Diagnosis Using Ensemble Empirical Mode Decomposition, Principal Component Analysis and Probabilistic Neural Network

Authors
Caixia Gao*, [email protected], Tong [email protected], Ziyi [email protected]
School of Electrical Engineering and Automation, Henan Polytechnic University, Jiaozuo 454000, China
http://www.hpu.edu.cn/www/index.html

*
Corresponding author: Caixia Gao (1981- ), female, associate professor, master tutor, fault diagnosis research. Tel (Tel.): 0391-3987580. E-mail: [email protected].
Corresponding Author
Caixia [email protected]
Available Online 30 June 2018.
DOI
https://doi.org/10.2991/jrnal.2018.5.1.3
Keywords
Rolling bearing; fault recognition; ensemble empirical modal decomposition; principal component analysis; probabilistic neural network
Abstract
Aiming at the problem that the vibration signal of the incipient fault is weak, an automatic and intelligent fault diagnosis algorithm combined with ensemble empirical mode decomposition (EEMD), principal component analysis (PCA) and probabilistic neural network (PNN) is proposed for rolling bearing in this paper. EEMD is applied to decompose the vibration signal into a sum of several intrinsic mode function components (IMFs), which represents the signal characteristics of different scales. The energy, kurtosis and skewness of first few IMFs are extracted as fault feature index. PCA is employed to the fault features as the linear transform for dimension reduction and elimination of linear dependence between the fault features. PNN is applied to detect rolling bearing occurrence and recognize its type. The simulation shows that this method has higher fault diagnosis accuracy.

Copyright
Copyright © 2018, the Authors. Published by ALife Robotics Corp. Ltd.
Open Access
This is an open access article under the CC BY-NC license (http://creativecommons.org/licences/by-nc/4.0/).

Download article (PDF)