一种基于EEMD-SVD和FCM的轴承故障诊断方法

A Bearing Fault Diagnosis Method Based on EEMD-SVD and FCM Clustering

提出了一种基于总体平均经验模式分解（EEMD）和奇异值分解（SVD）的模糊C均值聚类（FCM）相结合的轴承故障诊断方法。首先对轴承信号进行EEMD分解，得到若干个平稳的本征模函数（IMF），再通过相关性分析筛选包含主要信息的前几个分量进行奇异值分解，然后将得到的奇异值矩阵作为特征向量，通过FCM模糊聚类进行识别。实验结果表明，此方法可有效地对轴承故障类型进行识别。

A new method for diagnosis based on ensemble empirical mode of decomposition（EEMD） , singularity value decomposition（SVD） and fuzzy C-means clustering（FCM） is proposed. First of all, the mechanical vibration signals were decomposed by EEMD into a certain number of intrinsic mode functions （IMFs）. Secondly, IMF components were chosen by the criteria of mutual correlation coefficient and got several component containing the main information of signals, then , with the SVD method , singular value sequences were obtained. At last, the constructed eigenvector were put into the FCM fuzzy clustering classifier to recognize different fault types. The results of experiment and engineering analysis demonstrate that the method proposed is able to diagnose mechanical fauhs accurately and effectively.