基于VMD相关系数峭度提取行星齿轮箱故障特征

Fault Feature Extraction of Planetary Gearbox Based on VMD Correlation Coefficient Kurtosis

针对行星齿轮箱振动信号源多,信号间耦合调制强烈,信号成分复杂,信噪比低导致其故障特征提取困难等问题,提出了一种新的基于VMD相关系数峭度的行星齿轮箱故障特征提取方法。即通过行星齿轮箱齿轮故障模拟试验采集振动信号,并对其进行变分模态分解(VMD),得到本征模态函数分量(IMF),计算IMF分量的相关系数,选择最优IMF分量作为特征分量,提取其峭度作为齿轮故障特征。实例分析结果表明,采用本文提出的方法较EEMD相关系数峭度方法、VMD相关系数样本熵方法更能有效地提取行星齿轮箱齿轮故障特征。

Aiming at the problems of extracting fault features for planetary gearboxes due to multiple vibration signal sources,strong coupling modulation between signal sources,and low signal-to-noise ratio,complex signal components,a new method based on variational mode decomposition(VMD),correlation coefficient and Kurtosis theory was proposed.The vibration signal was collected by the gear fault simulation test of planetary gearbox,which was decomposed to obtain the Intrinsic Mode Function(IMF)component by VMD.The correlation coefficient of the IMF component was calculated,the optimal IMF component was selected as feature component,and its kurtosis was extracted as fault feature.The results show that the proposed method is more effective than EEMD correlation coefficient kurtosis method and VMD correlation coefficient sample entropy method.