2008 9th International Conference on Signal Processing 2008
DOI: 10.1109/icosp.2008.4697740
|View full text |Cite
|
Sign up to set email alerts
|

Feature extraction and selection for fault diagnosis of gear using wavelet entropy and mutual information

Abstract: This paper aims to develop an complete system including signal processing, feature extraction, feature selection and classification approaches for fault diagnosis of gear by using the wavelet transform, the entropy, the mutual information and the least-square support vector machine (LS-SVM). Firstly, the vibration signals are decomposed to several wavelet coefficients. The energy of every coefficient and the singularity values (SV) of the coefficient matrix are extracted. Two type entropies means the Shannon e… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
12
0

Year Published

2014
2014
2022
2022

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 19 publications
(12 citation statements)
references
References 9 publications
0
12
0
Order By: Relevance
“…where S denotes the number of features contained by S . 23,31 However, it is possible that the features selected based on max-relevance may have rich redundancy. 32 Therefore, the min-redundancy criterion should be added to select mutually exclusive features to a certain extent, which is represented by   (12) Finally, combining with the advantages of DET and mRMR, the features that are sensitive to the faults and contain less irrelevant or redundant information can be selected.…”
Section: Max-relevance and Min-redundancymentioning
confidence: 99%
“…where S denotes the number of features contained by S . 23,31 However, it is possible that the features selected based on max-relevance may have rich redundancy. 32 Therefore, the min-redundancy criterion should be added to select mutually exclusive features to a certain extent, which is represented by   (12) Finally, combining with the advantages of DET and mRMR, the features that are sensitive to the faults and contain less irrelevant or redundant information can be selected.…”
Section: Max-relevance and Min-redundancymentioning
confidence: 99%
“…Identification of circuit breaker defect types often depends on various accompanying signals in operation, in which vibration is the most easily obtained lossless signal during the energy transfer process of the operating mechanism components. Using vibration signals to diagnose circuit breaker mechanical faults has always been the focus of research at home and abroad . Runde et al .…”
Section: Introductionmentioning
confidence: 99%
“…However, DTW is prone to distortion and affects the accuracy of diagnosis when planning the optimal path. Wavelet transform (WT) is used to deal with circuit breaker vibration signals to extract features, such as the singularity index , energy entropy , node maximum coefficient , feature entropy, and other features, and based on the difference in characteristics of the different operating states of the circuit breaker, a good diagnostic effect in the mechanical fault diagnosis was observed. However, WT must firstly determine the wavelet basis function and decomposition scale, and cannot decompose adaptively.…”
Section: Introductionmentioning
confidence: 99%
“…New methods have been introduced to analyze and detect anomalous events such as epileptic seizure, coronary artery disease, and Alzheimer's disease from biomedical signals using different entropy types [21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39]. The application of the entropy approach in mechanical engineering is given in [40][41][42][43][44][45]. The application of entropy in the power systems was started because the system under investigation will have different entropy values under different states.…”
Section: Introductionmentioning
confidence: 99%