Feature Extraction Method of Rolling Bearing Fault Signal Based on EEMD and Cloud Model Characteristic Entropy

The randomness and fuzziness that exist in rolling bearings when faults occur result in uncertainty in acquisition signals and reduce the accuracy of signal feature extraction. To solve this problem, this study proposes a new method in which cloud model characteristic entropy (CMCE) is set as the si...

Full description

Bibliographic Details
Main Authors: Long Han, Chengwei Li, Hongchen Liu
Format: Article
Language:English
Published: MDPI AG 2015-09-01
Series:Entropy
Subjects:
Online Access:http://www.mdpi.com/1099-4300/17/10/6683
id doaj-63bdb4562ad445c7af1c7469470c459f
record_format Article
spelling doaj-63bdb4562ad445c7af1c7469470c459f2020-11-25T01:23:03ZengMDPI AGEntropy1099-43002015-09-0117106683669710.3390/e17106683e17106683Feature Extraction Method of Rolling Bearing Fault Signal Based on EEMD and Cloud Model Characteristic EntropyLong Han0Chengwei Li1Hongchen Liu2School of Electrical Engineering and Automation, Harbin Institute of Technology, Harbin 150001, ChinaSchool of Electrical Engineering and Automation, Harbin Institute of Technology, Harbin 150001, ChinaSchool of Electrical Engineering and Automation, Harbin Institute of Technology, Harbin 150001, ChinaThe randomness and fuzziness that exist in rolling bearings when faults occur result in uncertainty in acquisition signals and reduce the accuracy of signal feature extraction. To solve this problem, this study proposes a new method in which cloud model characteristic entropy (CMCE) is set as the signal characteristic eigenvalue. This approach can overcome the disadvantages of traditional entropy complexity in parameter selection when solving uncertainty problems. First, the acoustic emission signals under normal and damage rolling bearing states collected from the experiments are decomposed via ensemble empirical mode decomposition. The mutual information method is then used to select the sensitive intrinsic mode functions that can reflect signal characteristics to reconstruct the signal and eliminate noise interference. Subsequently, CMCE is set as the eigenvalue of the reconstructed signal. Finally, through the comparison of experiments between sample entropy, root mean square and CMCE, the results show that CMCE can better represent the characteristic information of the fault signal.http://www.mdpi.com/1099-4300/17/10/6683rolling bearingfeature extractionEEMDcloud model characteristic entropy
collection DOAJ
language English
format Article
sources DOAJ
author Long Han
Chengwei Li
Hongchen Liu
spellingShingle Long Han
Chengwei Li
Hongchen Liu
Feature Extraction Method of Rolling Bearing Fault Signal Based on EEMD and Cloud Model Characteristic Entropy
Entropy
rolling bearing
feature extraction
EEMD
cloud model characteristic entropy
author_facet Long Han
Chengwei Li
Hongchen Liu
author_sort Long Han
title Feature Extraction Method of Rolling Bearing Fault Signal Based on EEMD and Cloud Model Characteristic Entropy
title_short Feature Extraction Method of Rolling Bearing Fault Signal Based on EEMD and Cloud Model Characteristic Entropy
title_full Feature Extraction Method of Rolling Bearing Fault Signal Based on EEMD and Cloud Model Characteristic Entropy
title_fullStr Feature Extraction Method of Rolling Bearing Fault Signal Based on EEMD and Cloud Model Characteristic Entropy
title_full_unstemmed Feature Extraction Method of Rolling Bearing Fault Signal Based on EEMD and Cloud Model Characteristic Entropy
title_sort feature extraction method of rolling bearing fault signal based on eemd and cloud model characteristic entropy
publisher MDPI AG
series Entropy
issn 1099-4300
publishDate 2015-09-01
description The randomness and fuzziness that exist in rolling bearings when faults occur result in uncertainty in acquisition signals and reduce the accuracy of signal feature extraction. To solve this problem, this study proposes a new method in which cloud model characteristic entropy (CMCE) is set as the signal characteristic eigenvalue. This approach can overcome the disadvantages of traditional entropy complexity in parameter selection when solving uncertainty problems. First, the acoustic emission signals under normal and damage rolling bearing states collected from the experiments are decomposed via ensemble empirical mode decomposition. The mutual information method is then used to select the sensitive intrinsic mode functions that can reflect signal characteristics to reconstruct the signal and eliminate noise interference. Subsequently, CMCE is set as the eigenvalue of the reconstructed signal. Finally, through the comparison of experiments between sample entropy, root mean square and CMCE, the results show that CMCE can better represent the characteristic information of the fault signal.
topic rolling bearing
feature extraction
EEMD
cloud model characteristic entropy
url http://www.mdpi.com/1099-4300/17/10/6683
work_keys_str_mv AT longhan featureextractionmethodofrollingbearingfaultsignalbasedoneemdandcloudmodelcharacteristicentropy
AT chengweili featureextractionmethodofrollingbearingfaultsignalbasedoneemdandcloudmodelcharacteristicentropy
AT hongchenliu featureextractionmethodofrollingbearingfaultsignalbasedoneemdandcloudmodelcharacteristicentropy
_version_ 1725123915737464832