基于经验模态分解法和Hilbert谱的齿轮箱故障诊断

Fault Diagnosis in Gearbox Based on Empirical Mode Decomposition and Hilbert Spectrum

将经验模态分解法(Empirical Mode Decomposition,简称EMD)和Hilbert谱引入到齿轮箱故障诊断,提出了一种新的齿轮箱故障诊断方法。通过运用该方法和连续小波变换分别对某齿轮箱齿轮齿根裂纹故障振动信号进行分析,结果表明,该方法能更有效地提取齿轮故障信息,提高了齿轮故障诊断的准确性。这种自适应的信号处理方法非常适合分析非线性、非平稳过程。

The empirical mode decomposition （EMD） and Hilbert spectrum were introduced into vibration signal analysis for local gearbox fault diagnosis as a new method for adaptive analysis of non-linear and non-stationary signals. This method was applied to analyzed the vibration signals collected from a gearbox with an incipient tooth crack. The results show that the EMD algorithms and the Hilbert spectrum perform excellently in extracting the fault information. They are found to be more effective than the often used continuoas wavelet transform in detection of the vibration signatures.