应用经验模态分解下的AR模型提取旋转机械故障特征

Research on fault feature extraction of rotating machine based on empirical mode decomposition and AR model

将时间序列的AR模型引入到旋转机械故障诊断中,采用了经验模态分解与AR模型相结合的方法提取旋转机械的故障特征。通过选取含有故障信息的固有模态函数进行功率谱分析,提取故障特征,分析故障原因。仿真和试验结果表明,此法能够有效地提取故障特征参数,为旋转机械的故障诊断提供了方法保障。

The auto regressive model for time series prediction is introduced into rotating machinery fault diagnosis.The empirical mode decomposition and auto regressive model forecast parameters are put in use to extract the characteristics of rotating mechanical failure.To extract the fault feature and analyze the fault cause,the IMFs relating to fault information are applied to AR spectrum analysis.It is verified that the method yields effectively the characteristic parameters of fault.This work is helpful to diagnose the rotating machinery fault.