基于深度学习的辊道电机故障诊断分析方法

为了避免电机频繁无预警停机,机电设备诊断分析是一项非常重要的任务。针对辊道电机中单个电机故障,根据所有电机正常运行状况和故障状况下收集的电机电流数据,提出一种基于深度学习的智能化电机故障诊断分析策略。首次提出了考虑用协同工作的其他电机的电流来对被监测电机进行故障诊断。该方案选择采用Lenet-5模型进行分类预测训练,以诊断电机中的故障,在生产现场实际应用此方案并进行数据结果验证。结果表明,所提出的方法在电机故障诊断中可行且有效。

Fault diagnosis and analysis method of roller motor based on deep learning

In order to avoid frequent unwarned shutdowns of motors, diagnosis and analysis of electromechanical equipment is a very important task. Aiming at a single motor fault in the roller motor, an intelligent motor fault diagnosis and analysis strategy based on deep learning was proposed according to the motor current data collected under normal operating conditions and fault conditions of all motors. Considering the current of other motors working in coordination to diagnose the fault of the monitored motor was proposed for the first time. In this scheme, the Lenet-5 model was used for classification prediction training to diagnose faults of the motor. In addition, the actual deployment and implementation of this program and the collection of experimental data showed that the proposed method was feasible and effective in motor fault diagnosis.