Bispectrum Texture Feature Manifold for Feature Extraction in Rolling Bear Fault Diagnosis

Fei Wang and Liqing Fang

Mathematical Problems in Engineering, 2019, vol. 2019, 1-11

Abstract:

Effectively classify the fault types and the degradation degree of a rolling bearing is an important basis for accurate malfunction detection. A novel feature extract method - bispectrum image texture features manifold (BTM) of the rolling bearing vibration signal is proposed in this paper. The BTM method is realized by three main steps: bispectrum image analysis, texture feature construction and manifold feature dimensionality reduction. In this method, bispectrum analysis is employed to convert the mass vibration signals into bispectrum contour map, the typical texture features were extracted from the contour map by gray level co-occurrence matrix (GLCM), then the manifold dimensionality reduction method liner local tangent space alignment (LLTSA) is used to remove redundant information and reduce the dimension from the extracted texture features and obtain more meaningful low-dimensional information. Furthermore, the low-dimensional texture features were identified by support vector machine (SVM) which was optimized by genetic optimization algorithm (GA). The validity of BTM is confirmed by rolling bear experiments, the result show that the proposed feature extraction method can accurately distinguish different fault types and have a good performance to classify the degradation degree of inner race fault, outer race fault and rolling ball fault.

Date: 2019
References: Add references at CitEc
Citations:

Downloads: (external link)
http://downloads.hindawi.com/journals/MPE/2019/3805729.pdf (application/pdf)
http://downloads.hindawi.com/journals/MPE/2019/3805729.xml (text/xml)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hin:jnlmpe:3805729

DOI: 10.1155/2019/3805729

Access Statistics for this article

More articles in Mathematical Problems in Engineering from Hindawi
Bibliographic data for series maintained by Mohamed Abdelhakeem ().