【作者】 李巍华； 刘雯；

【Author】 LI Weihua,LIU Wen (1.School of Mechanical & Automotive Engineering,South China University of Technology,Guangzhou,510640; 2.Guangdong Key Laboratory of Vehicle Engineering,Guangzhou,510640)

【机构】 华南理工大学机械与汽车工程学院； 广东省汽车工程重点实验室；

【摘要】 针对机械齿轮故障诊断中缺乏大量故障样本的问题,提出基于主元分析和直推式支持向量机的齿轮故障诊断方法。通过主元分析对数据进行重构和降维,减少数据之间的冗余,将主元提取后的结果输入直推式支持向量机进行协同训练,实现故障模式分类。结果表明,该方法能有效应用于齿轮早期故障诊断。更多还原

【Abstract】 In order to solve the problem of insufficient samples in mechanical fault diagnosis,a novel method using PCA(principal component analysis) and TSVM(transductive support vector machine) is presented for gear incipient fault diagnosis.The approach reconstruct and reduce dimension of original data by PCA.After that,the TSVM is trained and then applied in faults classification, experimental results demonstrate the effectiveness of the proposed method in gear incipient fault diagnosis.更多还原