噪声与振动控制2019，Vol.39Issue(1)：192-196,5.DOI:10.3969/j.issn.1006-1355.2019.01.036
基于KICA-GDA和LSSVM的齿轮箱轴承故障诊断
Fault Diagnosis of Gearbox Bearings based on KICA-GDA and LSSVM
摘要
Abstract
In order to achieve higher fault recognition rate of gearbox rolling bearings, a fault recognition method is proposed based on KICA (Kernel Independent Component Analysis)-GDA (Generalized Discriminant Analysis) and LSSVM (Least Squares Support Vector Machine). Firstly, fault features of rolling bearing vibration signals such as kurtosis and information entropy are computed and recognized as initial feature vectors, which are mapped into a kernel feature space with KICA to omit the redundancy and eliminate the correlation among the fault features. Then, the GDA method is used to implement the feature fusion and the new features are input to LSSVM classifier for fault classification. The experimental results of the fault diagnosis of the gearbox bearings show that the KICA-GDA and LSSVM methods can identify more fault signals of gearbox bearings and improve the LSSVM's classification performance. This method has better classifying performance than the plain LSSVM method for gearbox bearing fault identification.关键词
振动与波/滚动轴承/齿轮箱/故障诊断/KICA/GDA/LSSVMKey words
vibration and wave/rolling bearing/gearbox/fault diagnosis/KICA/GDA/LSSVM分类
信息技术与安全科学引用本文复制引用
杨伟新,王平,李舜酩..基于KICA-GDA和LSSVM的齿轮箱轴承故障诊断[J].噪声与振动控制,2019,39(1):192-196,5.基金项目
中国航发技术创新基金资助项目(2014B60815R) （2014B60815R）
