Research on Diagnosis Method of Brushless DC Motor Based on CNN-A-BiLSTM
Brushless DC motors are one of important power devices for large equipment,and the operating status of motors is highly consistent with the operating status of equipment.However,current motor fault diagnosis methods are difficult to make accu-rate state judgments on motors in environments with multiple motors or electromagnetic interference.In order to achieve state diagno-sis of brushless DC motors in complex environments,this paper presets the fusion of convolutional neural network(CNN)algorithm and long short-term memory network algorithm.The bidirectional long short-term memory(BiLSTM)propagation network algorithm is adopted to capture the impact of bidirectional propagation environments on the characteristics of the motors,thereby improving the diagnostic accuracy of the model.Experimental results show that the average convergence time of the proposed model on the mechani-cal and electrical equipment fault diagnosis dataset is 8.91 minutes,and the average convergence time on the motor fault dataset is 12.66 minutes,both of which are lower than that of the control model in the same group.Secondly,the F1 value of the proposed model is 94.17％,which is 4.87％and 7.46％higher than that of the control model,respectively.In addition,in the comparison of voltage detection before and after motor faults,the proposed model provides more detailed detection results when motor faults occur.According to experimental results,the proposed motor diagnosis model has excellent performance and meets the requirements of mo-tor fault diagnosis.
CNNbrushless DC motorlong short-term memory networkactivation functionfault diagnosis
万方数据
维普
