数学形态学在故障信号提取中的应用

数学形态学滤波是一种非线性数字滤波方式，它可以有效地提取信号的边缘轮廓及信号几何特征。针对旋转机械信号的故障特征，对传统形态学滤波进行改进。根据故障特征多为脉冲信号的特点，采用指数衰减信号作为形态学滤波中的结构元素。仿真证明该方法可以有效提取噪声中的故障特征。 

Application of Mathematical Morphology in Fault Signal Extraction

Mathematical morphology filter is a nonlinear digital filtering algorithm.It can effectively extract edge contour and geometric characteristic of signal.According to the characteristic of fault signal in rotating machinery，the morphology filter was improved.As most fault features were pulse signal，damped exponential signals were used as the structure element in morphology filter.Simulation showed that this method was efficient to extract the fault features of noise.