中文摘要：

提出了基于核主元分析（KPCA）和支持向量机（SVM）相结合的齿轮箱齿轮故障识别方法。采用振动信号初始特征空间的内积核函数,将初始特征空间中的非线性问题转换成高维特征空间中的线性问题。通过主元分析对映射到高维空间中的数据信息进行处理,得到初始特征的非线性主元,实现对高维特征参数进行降维。再结合SVM良好的分类能力,结果表明,KPCA和SVM相结合的分类性能在齿轮箱故障诊断方面有更好的效果。

英文摘要：

A diagnosis method for fault of gears in a gearbox is proposed using kernel principal component analysis（KPCA）and support vector machine（SVM）.By computing the inner product kernel function of vibration signal of the original feature space,the nonlinear problem in the original feature space is converted into a linear problem in a high dimensional feature space.Then the mapped data in high dimensional space are analyzed to get the original features of the nonlinear principal component,achieving dimensionality reduction of the high-dimensional feature parameters.The SVM＇s good classification ability is combined.The results show that combining KPCA and SVM has a better effect in fault diagnosis of the gear box.