中文摘要：

变模式分解（Variational Mode Decomposition,VMD）是近年出现一种新的自适应分解方法,对其最小化可以把一个信号分解为固有模式。基于此,提出了变模式分解（VMD）降噪。首先通过VMD将故障信号分解为一系列模式,计算各模式与原信号的相关系数,再根据相关系数选择相应的分量进行重构以达到对原信号降噪的目地,之后通过差值形态滤波对重构信号进行解调以提取故障特征。为验证降噪的必要性和有效性,对比了直接差值形态滤波提取特征。仿真信号和对滚动轴承的实验证明了方法的有效性。

英文摘要：

Variational mode decomposition is a new,fully intrinsic and adaptive,variational method in recent years,the minimization of which leads to a decomposition of a signal into its principal modes. So it proposed a new diagnosis method based on VMD denoising. Firstly,it decomposed the fault signal into a series of principal modes,and then computed the correlation coefficients between modes and the original signal. After that,filtering the modes according to the correlation coefficients,then selected modes were used to reconstruct the signal to denoise. After that it used morphological difference filter to extract the fault characteristic frequency. To verify the necessity and validity of the VMD denoising,it compared it with the method which extracted the fault characteristic frequency by morphological difference filter directly. The simulating signal and experiment results show the validity of the method.