Application of Wavelet Analysis and Neural Network in Fault Diagnosis of Rolling Bearing

DOI

10.2991/jimet-15.2015.1How to use a DOI?

Keywords

rolling bearings; fault diagnosis; wavelet packet analysis; neural network

Abstract

In this paper, a fault-diagnosis method is proposed for generator rolling bearings based on wavelet packet analysis and neural network. Acquisition of wind farm rolling bearings real-time signal under different conditions.Firstly, decomposes vibration acceleration signals use wavelet packets analysis, make the original vibration signal decomposed into different frequency bands, then calculate the energy values, so extracts energy values of various vibration signal to construct fault eigenvector; which use as the input of the neural network. Then, by the parameter setting created a BP neural network ; in order to make the network has memory classification function we need training the network.Finally, the test sample put into the already trained BP get the fault pattern recognition. Using the wind farm real-time data for simulation experimental, the results show that the fault diagnosis model of high precision, can make a fast and effective fault diagnosis for rolling bearings.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Li Xinli
AU  - Yao Wanye
AU  - Yang Xiao
AU  - Zhou Qingjie
PY  - 2015/12
DA  - 2015/12
TI  - Application of Wavelet Analysis and Neural Network in Fault Diagnosis of Rolling Bearing
BT  - Proceedings of the 2015 Joint International Mechanical, Electronic and Information Technology Conference
PB  - Atlantis Press
SP  - 1
EP  - 6
SN  - 2352-538X
UR  - https://doi.org/10.2991/jimet-15.2015.1
DO  - 10.2991/jimet-15.2015.1
ID  - Xinli2015/12
ER  -