基于小波包分解-峭度值指标-希尔伯特包络解调融合方法处理声发射信号的滚动轴承故障诊断

为实现对航空发动机主轴承进行故障诊断，以复杂传递路径下声发射信号的波形分析为基础，提出一种基于小波包分解(wavelet packet decomposition, WPD)、峭度值指标(kurtosis index, KI)以及希尔伯特包络解调(Hilbert envelope demodulation, HED)相结合的滚动轴承故障特征信息提取方法。采用WPD方法对滚动轴承声发射信号分解获得节点分量，基于KI对节点分量排序筛选进行信号重构，进而对重构信号进行HED分析，提取出轴承故障特征频率用于对比诊断。开展简单以及复杂传递路径下滚动轴承故障模拟试验，采用建立的方法分别针对滚动轴承外圈、内圈典型故障试验数据进行分析和诊断。结果表明：该方法可有效提取滚动轴承故障特征频率及其倍频，且针对复杂传递路径下处于工作状态的滚动轴承，仍可实现精准的特征信息提取和有效的故障诊断。

Fault Diagnosis of Rolling Bearing based on Acoustic Emission Signal Analysis by WPD-KI-HED Combination method

In order to effectively diagnose the fault of main bearing in complicated path of Aircraft engine, based on acoustic emission signal waveform analysis, an extraction method of rolling bearing fault information based on wavelet packet decomposition (WPD), kurtosis index (KI) and Hilbert envelope demodulation (HED) was proposed. The rolling bearing ae signals were decomposed by WPD method to obtain node components, signal reconstruction is carried out by sorting and filtering node components based on kurtosis, HED analysis was performed on the reconstructed signal, bearing fault characteristic frequency were extracted to diagnosis. Rolling bearing fault simulation experiment under simple and complex transfer paths were carried out respectively, the method was used to analyze and diagnose the typical fault test data of outer ring and inner ring. Research results show that the method can effectively extract the bearing fault characteristic frequency and the frequency doubling. And even for rolling bearings with complex transmission paths, fault diagnosis can still be completed.