Affiliations: School of Mechanical and Electrical Engineering,
Jiangsu Normal University, Xuzhou, Jiangsu, China
Note: [] Corresponding author: W.Y. Liu, School of Mechanical and
Electrical Engineering, Jiangsu Normal University, Xuzhou 221116, Jiangsu,
China. E-mail: liuwenyi1984@yahoo.com.cn
Abstract: A rolling element bearing fault recognition approach is proposed in
this paper. This method combines the basic Higher-order spectrum (HOS) theory
and fuzzy clustering method in data mining area. In the first step, all the
bispectrum estimation results of the training samples and test samples are
turned into binary feature images. Secondly, the binary feature images of the
training samples are used to construct object templates including kernel images
and domain images. Every fault category has one object templates. At last, by
calculating the distances between test samples' binary feature images and the
different object templates, the object classification and pattern recognition
can be effectively accomplished. Bearing is the most important and much easier
to be damaged component in rotating machinery. Furthermore, there exist large
amounts of noise jamming and nonlinear coupling components in bearing vibration
signals. The Higher Order Cumulants (HOC), which can quantitatively describe
the nonlinear characteristic signals with close relationship between the
mechanical faults, is introduced in this paper to de-noise the raw bearing
vibration signals and obtain the bispectrum estimation pictures. In the
experimental part, the rolling bearing fault diagnosis experiment results
proved that the classification was completely correct.
