Comparison of denoising schemes and dimensionality reduction techniques for fault diagnosis of rolling element bearing using wavelet transform

H. S. Kumar, P. Srinivasa Pai, N. S. Sriram, G. S. Vijay, M. Vijay Patil

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

This paper presents the evaluation of five wavelets-based denoising schemes in order to select the best possible scheme for denoising bearing vibration signals and dimensionality reduction techniques using artificial neural network (ANN). Vibration signals from four conditions of rolling element bearing (REB) namely normal (N), defect on inner race (IR), defect on ball (B) and defect on outer race (OR) have been denoised using interval-dependent denoising scheme, which is the best possible scheme. The denoised signal is subjected to discrete wavelet transform (DWT) to extract 17 statistical features. Principal component analysis (PCA)-based dimensionality reduction technique (DRT) namely PCA alone, Kernel-PCA (KPCA) alone, PCA using SVD and KPCA using SVD have been used for reducing the dimension of the features. It is found that KPCA using SVD resulted in highest prediction accuracy using ANN, making it suitable for effective REB fault diagnosis.

Original languageEnglish
Pages (from-to)238-258
Number of pages21
JournalInternational Journal of Manufacturing Research
Volume11
Issue number3
DOIs
Publication statusPublished - 01-01-2016

Fingerprint

Bearings (structural)
Dimensionality Reduction
Denoising
Fault Diagnosis
Principal component analysis
Wavelet transforms
Wavelet Transform
Failure analysis
Singular value decomposition
Kernel PCA
Principal Component Analysis
Vibration Signal
Defects
Artificial Neural Network
Neural networks
Kernel Principal Component Analysis
Discrete wavelet transforms
Wavelets
Ball
Interval

All Science Journal Classification (ASJC) codes

  • Control and Systems Engineering
  • Modelling and Simulation
  • Computer Science Applications
  • Industrial and Manufacturing Engineering

Cite this

@article{b67e7ebc8d6448aa8df8111677ed75c7,
title = "Comparison of denoising schemes and dimensionality reduction techniques for fault diagnosis of rolling element bearing using wavelet transform",
abstract = "This paper presents the evaluation of five wavelets-based denoising schemes in order to select the best possible scheme for denoising bearing vibration signals and dimensionality reduction techniques using artificial neural network (ANN). Vibration signals from four conditions of rolling element bearing (REB) namely normal (N), defect on inner race (IR), defect on ball (B) and defect on outer race (OR) have been denoised using interval-dependent denoising scheme, which is the best possible scheme. The denoised signal is subjected to discrete wavelet transform (DWT) to extract 17 statistical features. Principal component analysis (PCA)-based dimensionality reduction technique (DRT) namely PCA alone, Kernel-PCA (KPCA) alone, PCA using SVD and KPCA using SVD have been used for reducing the dimension of the features. It is found that KPCA using SVD resulted in highest prediction accuracy using ANN, making it suitable for effective REB fault diagnosis.",
author = "Kumar, {H. S.} and Pai, {P. Srinivasa} and Sriram, {N. S.} and Vijay, {G. S.} and Patil, {M. Vijay}",
year = "2016",
month = "1",
day = "1",
doi = "10.1504/IJMR.2016.079461",
language = "English",
volume = "11",
pages = "238--258",
journal = "International Journal of Manufacturing Research",
issn = "1750-0591",
publisher = "Inderscience Enterprises Ltd",
number = "3",

}

Comparison of denoising schemes and dimensionality reduction techniques for fault diagnosis of rolling element bearing using wavelet transform. / Kumar, H. S.; Pai, P. Srinivasa; Sriram, N. S.; Vijay, G. S.; Patil, M. Vijay.

In: International Journal of Manufacturing Research, Vol. 11, No. 3, 01.01.2016, p. 238-258.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Comparison of denoising schemes and dimensionality reduction techniques for fault diagnosis of rolling element bearing using wavelet transform

AU - Kumar, H. S.

AU - Pai, P. Srinivasa

AU - Sriram, N. S.

AU - Vijay, G. S.

AU - Patil, M. Vijay

PY - 2016/1/1

Y1 - 2016/1/1

N2 - This paper presents the evaluation of five wavelets-based denoising schemes in order to select the best possible scheme for denoising bearing vibration signals and dimensionality reduction techniques using artificial neural network (ANN). Vibration signals from four conditions of rolling element bearing (REB) namely normal (N), defect on inner race (IR), defect on ball (B) and defect on outer race (OR) have been denoised using interval-dependent denoising scheme, which is the best possible scheme. The denoised signal is subjected to discrete wavelet transform (DWT) to extract 17 statistical features. Principal component analysis (PCA)-based dimensionality reduction technique (DRT) namely PCA alone, Kernel-PCA (KPCA) alone, PCA using SVD and KPCA using SVD have been used for reducing the dimension of the features. It is found that KPCA using SVD resulted in highest prediction accuracy using ANN, making it suitable for effective REB fault diagnosis.

AB - This paper presents the evaluation of five wavelets-based denoising schemes in order to select the best possible scheme for denoising bearing vibration signals and dimensionality reduction techniques using artificial neural network (ANN). Vibration signals from four conditions of rolling element bearing (REB) namely normal (N), defect on inner race (IR), defect on ball (B) and defect on outer race (OR) have been denoised using interval-dependent denoising scheme, which is the best possible scheme. The denoised signal is subjected to discrete wavelet transform (DWT) to extract 17 statistical features. Principal component analysis (PCA)-based dimensionality reduction technique (DRT) namely PCA alone, Kernel-PCA (KPCA) alone, PCA using SVD and KPCA using SVD have been used for reducing the dimension of the features. It is found that KPCA using SVD resulted in highest prediction accuracy using ANN, making it suitable for effective REB fault diagnosis.

UR - http://www.scopus.com/inward/record.url?scp=84990193033&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84990193033&partnerID=8YFLogxK

U2 - 10.1504/IJMR.2016.079461

DO - 10.1504/IJMR.2016.079461

M3 - Article

VL - 11

SP - 238

EP - 258

JO - International Journal of Manufacturing Research

JF - International Journal of Manufacturing Research

SN - 1750-0591

IS - 3

ER -