基于峭度-小波包分析的滚动轴承故障诊断

Fault Diagnosis of Rolling Bearing Based on Kurtosis-Wavelet Packet Analysis

理论分析和实例证明,峭度指标等滚动轴承振动时域统计参量可以判定轴承工作是否正常,但不能给出更多信息。通过小波包分析对振动信号进行分解,并有针对性地对包含有故障特征频率的频段信号进行重构,能有效地滤去各种干扰信号,显示故障特征信息,为滚动轴承的故障诊断提供了一种快速有效的途径。

Theoretical analysis and examination were made and it indicates that present statistical parameters of time domain of rolling bearing vibration like kurtosis index,can only provide the type information about a rolling bearing in normal instead of more information.The feature spectrum of fault rolling bearing can be obtained by decomposing the vibration signal through wavelet packet analysis and then reconstructing the signal which including fault feature spectrum,Thus,the information effectively helps the fault diagnosis of rolling bearing.