基于谐波小波分析的故障诊断方法研究

FAULT DIAGNOSIS METHOD BASED ON HARMONIC WAVELET ANALYSIS

谐波小波分析可有效提取非平稳信号中的奇异成分。但当信号中存在噪声时,谐波小波分解的时频等高线图无法凸显其奇异成分。本文采用谐波小波时频剖面图,对仿真信号和齿轮故障信号进行分析,成功提取出信号中的奇异成分。诊断实例证明,该方法可有效用于设备故障诊断。

Harmonic wavelet analysis can extract the singular components in the non-stationary signal effectively. But when the signal involves noise, the singular signal does not appear sharply in the time-frequency contour of harmonic wavelet. In the paper, using the time-frequency profile of harmonic wavelet to analyze the simulation signal and the signal including gear fault, the singular signals are extracted successfully. The diagnosis example indicates the method is an effective tool for device fault diagnosis.