回转齿轮箱故障特征提取的CMF-EEMD方法分析

Analysis of the rotary gearbox fault diagnosis extraction based on CMF-EEMD method

针对经验模态分解在回转齿轮箱故障中容易出现分解精度低的问题,提出了一种CMF-EEMD信号处理方法。组合模式函数(CMF)将集合经验模态(EEMD)分解得到本征模函数(IMF),并通过把高频和低频信号叠加的方式来形成包含高低频的模态函数,然后通过EEMD对循环自相关函数解调分析。将CMF-EEMD信号处理方法应用于仿真信号和齿轮箱故障诊断测试平台,成功提取了多故障特征频率,验证了该方法的有效性。该方法还可以清晰地反馈故障振动信号冲击幅值改变情况,从高频分量中可清楚地看到振动信号冲击幅值发生改变的明显特征。

Aiming at the low decomposition accuracy affected by the empirical mode decomposition in rotating gear box fault,it proposes a CMF-EEMD signal processing method.Combining the mode function(CMF)decomposed the collective empirical mode(EEMD)into IMF,it superposes the high and low frequency signals to form a mode function containing high and low frequency,demodulates and analyzes the cyclic autocorrelation function through EEMD.The CMF-EEMD signal processing method is applied to the simulation signal and the wind gear box test table,and the multi-fault characteristic frequency is successfully extracted.The method can also give a clear feedback of the change of the shock amplitude of the fault vibration signal.