A rolling bearing fault diagnosis method using kernel-domain spectrum
Aimed at the non-Gaussian and nonlinear characteristics of wind turbine vibration signal,a fault diagnosis method based on the kernel-domain spectrum is proposed in this paper.At first,the raw vibration signal monitored from the wind turbine is divided into groups for the pre-process.Then the bispectrum method is applied on the group data analysis and some two-dimension data matrixes are obtained.The object kernel templates,which are used for comparison and classification,are then built by the binarization processing.The fault classes can be identified by the comparison between object kernel templates and test kernel templates,which are built by the test data using the same method with the former.This kernel-domain spectrum method can not only distinguish different faults,but also overcome the disadvantages brought by small sample.The rolling bear fault diagnosis experiment proved the effectiveness and availability of the proposed method.
Wind turbine Kemel-do1llaill spectrum Fault diagnosis Feature extraction
GAO Qinwu LlU Wenyi
School of Mechanical and Electrical Engineering,Jiangsu Normal University Xuzhou,China
国际会议
第十七届国际制造会议(The 17th International Manufacturing Conference in China)(IMCC 2017))
深圳
英文
1-4
2017-11-23（万方平台首次上网日期，不代表论文的发表时间）