中文摘要：

针对航空发动机滚动轴承微弱损伤识别问题,设计了一种识别滚动轴承早期损伤的第二代小波包方法.构造了第二代小波包分解重构算法和第二代小波包算子计算方法;将滚动轴承振动信号进行第二代小波包分解,然后各个频带分解信号分别进行重构;利用希尔伯特变换解调分析各频带信号,得到对应频带信号的包络谱;计算各个频带中滚动轴承故障特征频率对应包络谱幅值的分贝值,提取各频带中故障特征频率对应分贝值的最大值,对滚动轴承早期损伤情况做出定量识别.仿真信号和试验信号分析表明,该识别方法结果可靠,能准确识别出滚动轴承早期损伤.

英文摘要：

To identify weak damage of aircraft engine rolling bearing, a new method to construct second generation wavelet packet was presented. In this method, the decomposition and reconstruction procedures of second generation packet were designed, and the second generation packet operators which lock on to the dominant signal structure were calculated. The vibration signal of rolling bearing was decomposed with second generation wavelet packet, and the signal at each frequency band was reconstructed. The reconstruction signals were demodulated with Hilbert transform, and envelope spectrum at each band was obtained. The dB values of fault feature frequency components corresponding envelope spectrum at each frequency band were calculated, and the maximal dB value of fault feature frequency components was extracted to identify damage of rolling bearing. The proposed damage identification method was used to analyze simulation signal and experimental signal from a rolling bearing with weak outer race damage, and results confirm the advantages of second generation wavelet packet for rolling bearing damage identification.