Title :
Fast and robust neural network based wheel bearing fault detection with optimal wavelet features
Author :
Xu, Peng ; Chan, Andrew K.
Author_Institution :
Dept. of Electr. Eng., Texas A&M Univ., College Station, TX, USA
fDate :
6/24/1905 12:00:00 AM
Abstract :
We propose a new design of a neural network based system to detect faulty bearings using acoustic signals in a noisy wayside environment. Statistical features are generated from discrete wavelet transform coefficients, and a genetic algorithm is used to select the optimal features. The false negative rate for detecting a condemnable bearing is as low as 0.1% regardless of the speed, load condition, and bearing type
Keywords :
condition monitoring; discrete wavelet transforms; fault diagnosis; genetic algorithms; machine bearings; neural nets; railways; condition monitoring; discrete wavelet transform; fault detection; genetic algorithm; neural network; optimal features selection; railways; wheel bearings; Acoustic noise; Acoustic signal detection; Discrete wavelet transforms; Fault detection; Genetic algorithms; Neural networks; Robustness; Signal design; Wheels; Working environment noise;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1007461
