VMD和ICA联合降噪方法在轴承故障诊断中的应用

发布时间：2018-06-11 10:08

本文选题：变分模态分解 + 独立分量分析　；参考：《振动与冲击》2017年13期

【摘要】：针对振动信号易受噪声干扰的影响、故障特征提取困难的问题,提出一种基于变分模态分解(Variational Mode Decomposition,VMD)和独立分量分析(Independent Component Analysis,ICA)相结合的去噪方法。该方法首先利用VMD算法将振动信号分解成若干不同频率的本征模态分量(Intrinsic Mode Function,IMF),有效的抑制了LMD分解中存在的模态混叠现象和端点效应等问题,然后依据峭度准则选取相应分量进行重构,引入虚拟噪声通道;最后利用Fast ICA将重构后信号再次进行去噪处理,分离出有效的故障特征分量,从而识别故障类型。将该方法应用到滚动轴承故障数据中,并与LMD-ICA方法作对比,结果表明,提出方法不仅能够有效的解决去噪过程中丢失故障信息以及由于模态混叠导致噪声不能完全去除的问题,还能更清晰、准确地提取出故障特征频率。
[Abstract]:Aiming at the problem that vibration signals are easily affected by noise interference and it is difficult to extract fault features, a new denoising method based on variational Mode decomposition (VMDM) and Independent component Analysis (ICA) is proposed. Firstly, the vibration signal is decomposed into Intrinsic Mode Intrinsic Mode IMF (IMF) with different frequencies by using VMD algorithm, which can effectively suppress the problems of modal aliasing and endpoint effect in LMD decomposition. Then the corresponding components are selected according to the kurtosis criterion for reconstruction and the virtual noise channel is introduced. Finally the reconstructed signal is de-noised again using Fast ICA to separate out the effective fault feature components and identify the fault type. The method is applied to the rolling bearing fault data and compared with the LMD-ICA method. The results show that the proposed method can not only effectively solve the problem of missing fault information in the process of de-noising and the problem that the noise can not be completely removed due to modal aliasing. The frequency of fault feature can be extracted more clearly and accurately.
【作者单位】：石家庄铁道大学电气与电子工程学院;
【基金】：国家自然科学基金(11227201;11372199;11572206) 河北省自然科学基金(A2014210142)
【分类号】：TH133.33

【参考文献】