Fault Monitoring and Classification Method of Rolling Bearing Based on KICA and LSSVM

• DOI: 10.4028/www.scientific.net/amr.971-973.476
• Publication Date: 2014-06-25T12:47:44Z
• Authors: Bao Ma; Jian De Wu; Jun Ma; Xiao Dong Wang; Yu Gang Fan
• Keywords: Combing; Feature vector; Kernel (algebra)

The running process of rolling bearing is often nonlinear and abnormal. Therefore, this paper uses the method which combing KICA LS-SVM to achieve bearings' fault monitoring classification. Firstly, vibration signal mapped into high dimensional space by using kernel methods, constructing I2, Ie2 SPE indicator in monitor data. And then when occurred, extracting time domain wavelet energy features construct multi-domain mixed feature set, input vector for Experimental results show can complete classification effectively.