基于经验模式分解和主元分析的滚动轴承故障诊断方法研究

A fault diagnosis method for rolling bearings based on empirical mode decomposition and principal component analysis

提出了一种融合经验模式分解和多元统计的轴承故障诊断新方法,主要包括基于信号Hilbert-Huang变换的特征提取和对故障特征集的主成分分析:首先运用EMD将振动信号分解成不同特征时间尺度的单分量固有模态函数,采取Hilbert-Huang变换获取分解信号的瞬时频率,计算基本模式分量与瞬时频率的统计特征集;之后对统计特征集进行主成分分析,大幅降低特征向量的维数,获取主元特征集;最后利用支持向量机,完成了对于滚动轴承常见三类故障的分类,并分析了振动信号时域频域的统计特征值与故障模式之间的联系。

A new fault diagnosis method for rolling bearings based on empirical mode decomposition（ EMD） and multivariate statistics was proposed. Hilbert-Huang transformation and principal component analysis（ PCA） were used in this method. A vibration signal was decomposed into the basic mode components with EMD and the instantaneous frequency of each component was obtained with Hilbert-Huang transformation. Then,the statistical characteristics of the instantaneous frequency and the basic mode components were calculated. The statistical characteristics were analyzed with PCA in order to reduce the number of dimensions of the feature vector and get the principal component characteristics.Finally,the classification of three failure modes in rolling bearings was completed and the relationship between the statistical characteristics and failure modes was obtained.