Rotating machine fault detection based on HOS and artificial neural networks

Chih Chung Wang, Gee-Pinn James Too

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

In order to identify the faults of rotating machinery, classification process can be divided into two stages: one is the signal preprocessing and the feature extraction; the other is the recognition process. In the preprocessing and feature extraction stage, the higher-order statistics (HOS) is used to extract features from the vibration signals. In the recognition process, two kinds of neural network classifier are used to evaluate the classification results. These two classifiers are self-organizing feature mapping (SOM) network for collecting data at the initial stage and learning vector quantization (LVQ) network at the identification stage. The experimental results obtained from HOS as preprocessor to extract the features of fault are clearer than those obtained from the power spectrum. In addition, the recognizable rate by using either SOM or LVQ as classifiers is 100%.

Original languageEnglish
Article number5089563
Pages (from-to)283-293
Number of pages11
JournalJournal of Intelligent Manufacturing
Volume13
Issue number4
DOIs
Publication statusPublished - 2002 Jan 1

Fingerprint

Higher order statistics
Fault detection
Classifiers
Vector quantization
Neural networks
Feature extraction
Rotating machinery
Power spectrum

All Science Journal Classification (ASJC) codes

  • Software
  • Industrial and Manufacturing Engineering
  • Artificial Intelligence

Cite this

@article{e5fd6fbc7b724126a2480bfd36eec1cb,
title = "Rotating machine fault detection based on HOS and artificial neural networks",
abstract = "In order to identify the faults of rotating machinery, classification process can be divided into two stages: one is the signal preprocessing and the feature extraction; the other is the recognition process. In the preprocessing and feature extraction stage, the higher-order statistics (HOS) is used to extract features from the vibration signals. In the recognition process, two kinds of neural network classifier are used to evaluate the classification results. These two classifiers are self-organizing feature mapping (SOM) network for collecting data at the initial stage and learning vector quantization (LVQ) network at the identification stage. The experimental results obtained from HOS as preprocessor to extract the features of fault are clearer than those obtained from the power spectrum. In addition, the recognizable rate by using either SOM or LVQ as classifiers is 100{\%}.",
author = "Wang, {Chih Chung} and Too, {Gee-Pinn James}",
year = "2002",
month = "1",
day = "1",
doi = "10.1023/A:1016024428793",
language = "English",
volume = "13",
pages = "283--293",
journal = "Journal of Intelligent Manufacturing",
issn = "0956-5515",
publisher = "Springer Netherlands",
number = "4",

}

Rotating machine fault detection based on HOS and artificial neural networks. / Wang, Chih Chung; Too, Gee-Pinn James.

In: Journal of Intelligent Manufacturing, Vol. 13, No. 4, 5089563, 01.01.2002, p. 283-293.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Rotating machine fault detection based on HOS and artificial neural networks

AU - Wang, Chih Chung

AU - Too, Gee-Pinn James

PY - 2002/1/1

Y1 - 2002/1/1

N2 - In order to identify the faults of rotating machinery, classification process can be divided into two stages: one is the signal preprocessing and the feature extraction; the other is the recognition process. In the preprocessing and feature extraction stage, the higher-order statistics (HOS) is used to extract features from the vibration signals. In the recognition process, two kinds of neural network classifier are used to evaluate the classification results. These two classifiers are self-organizing feature mapping (SOM) network for collecting data at the initial stage and learning vector quantization (LVQ) network at the identification stage. The experimental results obtained from HOS as preprocessor to extract the features of fault are clearer than those obtained from the power spectrum. In addition, the recognizable rate by using either SOM or LVQ as classifiers is 100%.

AB - In order to identify the faults of rotating machinery, classification process can be divided into two stages: one is the signal preprocessing and the feature extraction; the other is the recognition process. In the preprocessing and feature extraction stage, the higher-order statistics (HOS) is used to extract features from the vibration signals. In the recognition process, two kinds of neural network classifier are used to evaluate the classification results. These two classifiers are self-organizing feature mapping (SOM) network for collecting data at the initial stage and learning vector quantization (LVQ) network at the identification stage. The experimental results obtained from HOS as preprocessor to extract the features of fault are clearer than those obtained from the power spectrum. In addition, the recognizable rate by using either SOM or LVQ as classifiers is 100%.

UR - http://www.scopus.com/inward/record.url?scp=0036681731&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036681731&partnerID=8YFLogxK

U2 - 10.1023/A:1016024428793

DO - 10.1023/A:1016024428793

M3 - Article

VL - 13

SP - 283

EP - 293

JO - Journal of Intelligent Manufacturing

JF - Journal of Intelligent Manufacturing

SN - 0956-5515

IS - 4

M1 - 5089563

ER -