改进的EMD结合重复降噪在故障诊断中的应用

滚动轴承的故障信号采集中往往含有大量的噪声信号。对采集信号进行小波包降噪后，利用经验模态分解(empirical mode decomposition，EMD)得到若干个固有模态函数(intrinsic mode function，IMF)。计算各个IMF与去噪后信号的相关系数以此确定哪几个IMF是待分析信号的有效集，根据有效集中IMF的突变程度来选择不同消失矩的db系小波进行小波降噪。对IMF进行边际谱分析来判断滚动轴承哪个部位发生故障。该方法有效地去除了混杂在故障信号中的噪声，提高了信噪比，准确地判断出滚动轴承发生故障的部位。

Application of the Improved EMD Method Combined with Repeated Noise Reduction in Fault Diagnosis

In fact, the roller bearing vibration signal collected often contained a lot of noise.Through to the signal wavelet packet noise reduction,by use of empirical mode decomposition to get a group of intrinsic mode functions.Calculated the correlation coefficient of the IMF and the denoised signal so as to determine which IMFs were effective set of signals to be analyzed,according to the mutations degree of the IMFs to choose Daubechies wavelet of different vanishing moments for wavelet denoising.Finally,through to analysing the marginal spectrum of IMF to determine which part of rolling bearing had been fault.This method removed the noise mixed with the fault signal well,improved the signal to noise ratio,determined the site of the rolling bearing failure accurately.