关联维数在旋转机械支承系统状态监测中的应用

Application of Correlation Dimension in Supporting System Fault Diagnosis for Rotating Machinery

提出用关联维数来定量描述旋转机械支承系统的工作状态 ,进而对系统进行故障诊断 .文中首先采用迭代奇异值降噪算法对转子支承系统不同状态下的信号进行降噪处理 ,然后计算了信号的关联维数 .分析结果表明 ,系统不同运行状况对应的关联维数明显不同 ,因此可以用关联维数作为支承系统工作状态的特征参数 。

This paper prensented a study of correlation dimension in the supporting system of rotating machinery condition monitoring. First, the noise is reduced by an iterative singular value decomposition method. Then, the correlation dimension of the supporting system under different conditions is obtained. The result shows that the correlation dimension can be used to identify clearly a supporting system operating condition with larger bearing clearance and pedestal looseness compared with that of the normal condition.