Open AccessApplication of Projection Pursuit Analysis Method Based on Kernel Function in Fault Diagnosis for Rolling Bearing
Author(s) -
Jing Cheng,
Le Su,
Weiqing Wang,
Shan He
Publication year - 2021
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/770/1/012001
Subject(s) - bearing (navigation) , artificial intelligence , projection (relational algebra) , fault (geology) , computer science , kernel (algebra) , matlab , gaussian function , nonlinear system , projection pursuit , feature (linguistics) , pattern recognition (psychology) , gaussian , computer vision , algorithm , mathematics , geology , combinatorics , quantum mechanics , linguistics , philosophy , physics , seismology , operating system
In view of nonlinear and non-Gaussian characteristics of fault feature for rolling bearing of wind turbine, the projection pursuit analysis method based on kernel function is put forward to make pattern recognition for all bearing running states. It elaborates the principle of projection pursuit method, and plan and steps for pattern recognition, then has a simulation and test by MATLAB software. The test results show that the projection pursuit analysis method based on kernel function is an effective method for fault diagnosis, and it can quickly and efficiently identify rolling bearing running states.