Fault Diagnosis of Rotating Machinery Based on One-Dimensional Deep Residual Shrinkage Network with a Wide Convolution Layer

Table 3

Influence of the number of convolution kernels in the wide convolution layer on the 1D-WDRSN model.

The number of the convolution kernels	4	8	16	24	32	48	64	96 (%)
Accuracy	31.37	53.54	74.51	84.84	91.64%	94.44	92.17	94.37