Underground coal mining machine has complex structure and poor working environment,which is prone to various kinds of faults and difficult to dispose of.To meet the demand for fault diagnosis of the cut-off gear of MG 1000/2500-WD coal mining machine,a DCNN fault prediction model is designed by combining the fault feature extraction principle of convolutional neural network,and the data training and model testing are implemented.The test results show that the fault screening accuracy of the prediction model reaches 98.17％,and the accuracy of the fault training value and the standard value reaches 99.13％,and the fault prediction has a high reliability,which achieves the expected research purpose.
NETL
NSTL
维普
万方数据
