经验模式分解在滚动轴承故障诊断中的应用

Application of Empirical Mode Decomposition in Rolling Bearing Fault Diagnosis

经验模式分解作为一种自适应的非线性、非平稳性信号分析方法,广泛应用于滚动轴承故障诊断中,但存在模态混淆、本征模式函数的判据、边界效应等问题。首先系统阐述了经验模式分解方法原理及特性,分别对经验模式分解的改进算法固有时间尺度分解、局部特征尺度分解和完备总体经验模式分解作了分析,研究了其基本原理、应用和特点,提出经验模式分解与其他诊断方法相融合是滚动轴承故障诊断的研究方向。

Empirical mode decomposition（ EMD）,as a self-adaptive nonlinear and non-stationary signal analysis method,is widely used in rolling bearing fault diagnosis. However,there are still some problems for further research such as mode confusion,criterion of intrinsic mode function and endpoint effect. EMD is systematically studied in this paper,and the improved calculating methods,including intrinsic time-scale decomposition（ ITD）,local characteristic-scale decomposition（ LCD） and complete ensemble empirical mode decomposition（ CEEMD） are analyzed with an introduction to their basic principles,applications and features. It is pointed out that the combination of EMD with other diagnosis methods will be a new research direction in the rolling bearing fault diagnosis.