A multi-fault diagnosis method of gear-box running on edge equipment

Journal of Cloud Computing

Advances, Systems and Applications

Table 4 Network parameters of the proposed model

Layertype	Kernel size	Stride	Kernel channel size	Output Size	Batch Normalization
Batch Input	∖	∖	∖	1000*1	∖
Convolution C1	32*1	2	32	485*32	Yes
Pooling P1	2*1	2	32	242*32	∖
Convolution C2	16*1	1	64	227*64	Yes
Pooling P2	2*1	2	64	113*64	∖
Convolution C3_1	8*1	1	32	106*32	Yes
Pooling P3_1	2*1	2	32	53*32	∖
Convolution C3_2	8*1	1	32	106*32	Yes
Pooling P3_2	2*1	2	32	53*32	∖
Fully-connected F1	100	∖	1,Dropout = 0.8	100*1	Yes
Fully-connected F2	100	∖	1,Dropout = 0.8	100*1	Yes
Softmax S1	5	∖	∖	5*1	∖
Softmax S2	6	∖	∖	6*1	∖