Abstract:

In order to determine the type and degree of damage of rolling bearing faults effectively, a classification method of rolling bearing fault modes combining MTS and SVM is proposed. The original vibration signal is decomposed using the EEMD method to obtain a series of IMF. After selecting the IMF through the fault-sensitive IMF selection method, the time and frequency domain characteristic parameters and the energy entropy parameters of the original signal are calculated to construct the initial multi-dimensional feature space. The orthogonal table and the signal-to-noise ratio in the MTS are used to reduce the feature dimensions to obtain a reduced feature space. The partial binary tree method is used to build a SVM multi-classification model. The model is verified by experimental data, and the results show that the method can classify rolling bearing failure modes.

Key words: failure modes classification, Mahalanobis-Taguchi System（MTS）, Support Vector Machine（SVM）, Ensemble Empirical Mode Decomposition（EEMD）

摘要：

为了有效地确定滚动轴承的故障类型和受损程度，提出了结合马田系统和SVM的滚动轴承故障模式分类方法。利用EEMD方法对原始振动信号进行分解，得到一系列IMF。经过故障敏感IMF选取方法筛选IMF后计算其时域和频域特征参数以及原始信号的能量熵参数，构造初始的多维特征空间。运用马田系统中的正交表和信噪比进行特征降维，得到精简特征空间。接下来使用偏二叉树方法构建支持向量机多分类模型。通过实验数据进行模型验证，结果表明该方法可以实现滚动轴承故障模式分类。

关键词: 故障模式分类, 马田系统（MTS）, 支持向量机（SVM）, 集合经验模态分解（EEMD）