本发明公开了一种多源信息融合的电机故障诊断方法及系统，涉及电机故障诊断领域，所述方法包括：获取电机各变量当前的运行数据，对电机各变量当前的运行数据进行预处理和数据标准化，得到处理后的数据，对处理后的数据进行数据融合，得到总变量的统计量，判断所述总变量的统计量是否超出了控制限，若是，则确定电机当前运行状态下各个变量对总变量的统计量的贡献率，针对任一变量，判断电机当前运行状态下所述变量的贡献率是否大于电机正常运行时所述变量的贡献率，若是，则确定所述变量所对应的部位发生故障。通过本发明有效提高了故障识别准确率。

The invention discloses a motor fault diagnosis method and system of multi-source information fusion, which relates to the field of motor fault diagnosis. The method includes: Obtain the current operating data of each variable of the motor, preprocess and standardize the current operating data of each variable of the motor, obtain the data after processing, and carry out data fusion on the data after processing to obtain the statistics of the total variable, and judge whether the statistics of the total variable exceeds the control limit. If so, Determine the contribution rate of each variable to the statistics of total variables under the current running state of the motor. For any variable, determine whether the contribution rate of the variable under the current running state of the motor is greater than the contribution rate of the variable when the motor is in normal operation. If so, determine the fault of the corresponding part of the variable. The invention effectively improves the accuracy of fault identification.