Convolutional Recurrent Neural Network for Fault Diagnosis of High-Speed Train Bogie

Table 6

Comparison of the time-consuming conditions and accuracy between CRNN, 1D-CNN, and LSTM.

	Method
	CRNN	1D-CNN	LSTM
highest precision iterations	800	1600	1500
time-consuming situations	24m35s	6m40s	4h53m15s
accuracy	0.974	0.952	0.952