谐波小波包自适应分解在故障诊断中的应用

将谐波小波包与匹配追踪方法相结合,提出了谐波小波包自适应分解的新方法。根据信号特征自适应选择谐波小波包字典内的时频原子,可以将非平稳振动信号既不交叠又无遗漏地分解到相互独立的频带上去,算法实现简单,频率分辨率好。通过仿真算例将该法与小波包变换、小波包追踪结果相比较,验证了该方法时频定位性好的优越性。将该方法应用于轴承和转子的故障诊断,结果表明,故障特征提取是有效的。

Adaptive Harmonic Wavelet Packet Transform and Its Application to Machine Fault Diagnosis

Combined harmonic wavelet packet with matching pursuit, an adaptive decomposing method was presented. In the proposed method, the time-frequency atoms could be adaptively selected from the harmonic wavelet packet dictionary. Non-stationary signal could be decomposed into a number of independent frequency sub-bands and the frequency overlapping and leakage was avoided. The proposed algorithm is simple and has good resolution. The simulation results showed that the proposed approach is superior to the wavelet packet transform and matching pursuit based on wavelet packet. At last, the proposed method has been applied to the fault diagnosis of bearing and rotor system, the experiment result showed the proposed method is effective.