提出基于核函数主元分析的机械故障诊断方法, 它保留主元分析的优点并具有处理非线性的能力。该方法通过核函数映射将非线性问题转换成高维的线性特征空间, 然后对高维空间中的映射数据作主元分析,提取其非线性特征, 对机械故障模式进行识别。并与主元分析方法进行对比分析, 实验结果表明核函数主元分析法非常有效。-Proposed mechanical fault diagnosis method based on Kernel Principal Component Analysis, it retains the advantages of principal component analysis and has the ability to handle nonlinear. The kernel function mapping the nonlinear problem is converted into a high-dimensional linear feature space, the primary element analysis and mapping data in high-dimensional space, extract the nonlinear characteristics of mechanical failure mode identification. Comparative analysis and principal component analysis method, the experimental results show that Kernel Principal Component Analysis method is very effective.