Intelligent Fault Diagnosis of Bearing Based on Convolutional Neural Network and Bidirectional Long Short-Term Memory

Table 2

CNN-BLSTM model parameters applied.

Name	Filters	Kernel size/stride	Padding	Units	Activation function
Conv_1	16	8/2	Same		Elu
Pooling_1		3/-	Valid
Conv_2	32	3/1	Same		Elu
Pooling_2		3/-	Valid
Conv_3	64	3/1	Same		Elu
Pooling_3		3/-	Valid
BLSTM_1				64
Dense				32	Elu
Dense				4	Softmax