迭代能量算子机械故障诊断研究

Research on mechanical fault diagnosis based on iterated energy operator

旋转机械元件的故障振动信号通常为多分量调幅调频信号,针对此类信号的故障特征提取问题,介绍一种基于迭代能量算子的机械故障诊断方法。首先阐述迭代能量算子对多分量信号的各分量信号进行幅值包络提取和平滑瞬时频率估计的基本原理;然后通过仿真试验给出与Hilbert-Huang变换的比较,表明该方法解调精度很高且速度较快;最后给出滚动轴承故障诊断实例,证明基于迭代能量算子的多分量调幅调频信号解调方法能有效地提取机械故障振动信号的故障特征。

The fault vibration signals of rotary mechanical components are generally multi-component AM-FM signals. A new approach for mechanical fault diagnosis based on Iterated Energy Operator is introduced to extract the fault features of those signals. First, the basic principle of this operator is set forth in extracting the amplitude envelope and estimating the smooth instantaneous frequency of each component signal. Second, a simulation test is conducted and the test results are compared with Hilbert-Huang transformation. This new method has been proved to be more accurate and faster. Third, an example namely the fault diagnosis of rolling element bearings is given to demonstrate that this new approach can better extract the fault features of mechanical fault signals.