中文摘要：

针对滚动轴承出现故障时故障信号常常被强的背景噪声所淹没,故障特征难以准确提取的问题,提出了一种基于总体平均经验模态分解（EEMD）降噪与二次相位耦合的故障诊断方法.对原始信号进行EEMD分解,通过相关系数-峭度值的大小选取EEMD分解后得到的固有模态函数（IMF）.对筛选出来的IMF进行了重构从而达到降噪的目的.对重构后的包络信号进行1.5维谱分析,提取二次相位耦合产生的非线性特征,从而得到滚动轴承故障特征频率信息.通过实测滚动轴承数据验证了该方法可以有效地提取轴承内圈和外圈的故障特征,从而识别轴承的故障.

英文摘要：

Aimed at the problem that the fault signals always is masked by strong background noise,so that it is difficult to extract accurately the fault feature of roller bearing,a method for fault diagnosis is proposed based on ensemble empirical mode decomposition（EEMD）denoising and quadratic phase coupling.EEMD method is used to decompose the original signal and instrinsic mode functions（IMFs）are selected according to the magnitude of correlation coefficient-kurtosis.These IMFs selected are reconstructed so that the goal of denoising is achieved.1.5dimensional spectrum analysis of reconstructed envelope signal is conducted to extract non-linear feature produced by the quadratic phase coupling,so that the fault characteristic frequencies of the roller bearing are obtained.It is verified by actual signals measurement that this method is effective to extract the fault feature of inner and outer races of the bearing,so that the fault of the bearing is identified.