基于工业互联网和多传感器数据的电机故障诊断方法

提出了一种基于工业互联网和多传感器数据的电机故障诊断方法。通过各类传感器在线实时得到电机的电压、电流、振动、温度等信号的瞬时值,并转化为表征电机状态的各个特征参数。根据各个特征参数在各个故障模式下的变动情况,得到各个故障模式下故障特征及其隶属度。把故障特征与故障模式之间的关系分为充分条件和必要条件关系。按照充分条件和必要条件分类后,对每个故障模式对应的2类条件下的故障特征的隶属度进行融合,最后得出每个故障模式的隶属度,为远程运维系统决策服务。该方法既可以部署在电机远程运维工业互联网的边缘设备中,也可以部署在云平台服务程序中,实现快速而可靠的电机故障诊断。

Fault Diagnosis Method for Motor Based on Industrial Internet andMultiSensor Data

A fault diagnosis method for motor based on industrial internet and multisensor data was presented. The instantaneous values of voltage, current, vibration and temperature of the motor could be obtained online in real time through a number of sensors, such as voltage, current, vibration and temperature sensor. The instantaneous values could be transformed into various characteristic parameters to characterize the state of the motor. According to the fluctuation of each characteristic parameter in each fault mode, the fault features and their membership functions in each fault mode could be obtained. The relationship between fault features and fault modes could be classified into sufficient and necessary conditions. The membership function of the fault features under the two conditions corresponding to each fault mode were fused. The membership function of each fault mode was obtained to serve the decisionmaking of the remote operation and maintenance system. This method could be deployed not only in the edge equipment of remote operation and maintenance industrial internet, but also in the service program of cloud platform to realize fast and reliable motor fault diagnosis.