Sparse Fast Fourier Transform and its application in intelligent diagnosis system of train rolling bearing

Healthy status monitoring of train bearing online is very meaningful work. But as traditional diagnosis system does, performing Fourier spectrum with the datum from more than 200 bearings in a marshalling train is an enormous challenge. Here a healthy status monitoring system of train rolling bearing based on Sparse Fast Fourier Transform (SFFT) is proposed. The monitoring system consists two sequential parts. First, extract fault features based on SFFT spectrum and other time-domain parameters. According to train bearing working environment, altogether 7 fault features are extracted in this paper. Another part is constructing a classifier based on BP neural network. Experimental results show that the system proposed here achieves gratifying results comparing with traditional fault diagnosis system

[1]  Piotr Indyk,et al.  Simple and practical algorithm for sparse Fourier transform , 2012, SODA.

[2]  Christopher J. Dodge.  A fast Fourier Transform accelerator for a transputer system , 1993 .

[3]  Jörn Schumacher.  High performance Sparse Fast Fourier Transform , 2013 .

[4]  Si Zhen-zhen.  Application of Fourier transform and wavelet transform in signal de-noising , 2011 .

[5]  Piotr Indyk,et al.  Nearly optimal sparse fourier transform , 2012, STOC '12.

[6]  Zhi-Hua Zhou,et al.  Hybrid decision tree , 2002, Knowl. Based Syst..

[7]  P. Swain,et al.  Neural Network Approaches Versus Statistical Methods In Classification Of Multisource Remote Sensing Data , 1990 .

[8]  David G. Kleinbaum,et al.  Introduction to Logistic Regression , 2010 .

[9]  S. K. Pathan,et al.  Pattern Association for character recognition by Back-Propagation algorithm using Neural Network approach , 2012 .

[10]  Aurobinda Routray,et al.  Porosity and Permeability Estimation using Neural Network Approach from Well Log Data , 2012 .

[11]  Piotr Indyk,et al.  Faster GPS via the sparse fourier transform , 2012, Mobicom '12.

[12]  Limin Jia,et al.  Fault Diagnosis of Train Axle Box Bearing Based on Multifeature Parameters , 2015 .

[13]  Ely Porat,et al.  Approximate Sparse Recovery: Optimizing Time and Measurements , 2012, SIAM J. Comput..

[14]  Jae-Seob Kwak,et al.  Trouble diagnosis of the grinding process by using acoustic emission signals , 2001 .

[15]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[16]  Heekuck Oh,et al.  Neural Networks for Pattern Recognition , 1993, Adv. Comput..