支持向量机和基于IMF的特征能量法在汽车变速器轴承故障诊断中的应用

Application of Support Vector Machine and IMF-based Feature Energy Method to Automotive Transmission Bearing Fault Diagnosis

针对汽车变速器轴承振动信号的非平稳特征和现实中难以获得大量典型故障样本的实际情况,提出了一种基于内禀模态函数(IMF)的特征能量法和支持向量机的变速器轴承故障诊断方法。对变速器轴承内圈、外圈故障振动信号的分析结果表明,该方法在小样本情况下仍可有效提取变速器轴承的故障特征,并能成功地对其工作状态和故障类型进行分类。

In view of the non-stationary features of vibration signals of automotive transmission bearing and the difficulty to obtain a large number of fault samples in practice, a fault diagnosis scheme for automotive transmission beating is proposed by using support vector machine and feature energy method based on intrinsic mode func- tion. The results of analysis on fault vibration signals of inner and outer races of transmission bearing show that the proposed scheme can extract beating fault features effectively and classify the working conditions and fault patterns of transmission bearings successfully even in the case of smaller number of samples.