基于EMD和AR模型的轴承故障诊断

构建集外圈、内圈和滚动于一体的轴承故障动力学模型,获取轴承运行时的轴承振动信号,采用基于EMD和AR模型的轴承故障诊断方法,将轴承振动信号分解成IMF分量后构建AR模型,再采用该模型自回归参数与残差的方差构建轴承综合判定距离,依据最小综合判别距离对应的状态完成轴承故障诊断。分析研究表明:诊断轴承正常、外圈故障以及内圈故障时,该方法诊断结果误差小、复杂度低,达到了高效诊断轴承故障的目标。

Research on bearing fault diagnosis method based on EMD and AR model

Bearing is a vulnerable component in food processing machinery and packaging machinery. Bearing status is related to the safe operation of machinery. In this study, bearing fault diagnosis method based on EMD and AR models was adopted to solve the problem of bearing fault diagnosis, ensuring the maximum quality and efficiency of food production. A bearing fault dynamics model integrating outer ring, inner ring, and rolling were built to obtain the bearing vibration signal during the operation of the bearing. Therefore, the bearing fault diagnosis method based on EMD and AR model was adapted to decompose the bearing vibration signal into IMF components. After constructing AR model, it was used to construct the bearing comprehensive determination distance between the autoregressive parameters and the variance of the residuals, and the fault diagnosis of the bearing was completed according to the state corresponding to the minimum comprehensive discrimination distance. The analysis and research showed that the outer ring fault and the inner ring fault were diagnosed with the normal bearing. Moreover, the error of the diagnosis result was small with low complexity, and this helped to achieve the goal of high efficiency diagnosis of bearing fault.