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

Table 2

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

The size of the convolution kernel	48	64	96	128 (%)	192 (%)	256 (%)	384 (%)	512 (%)
Accuracy	71.44	85.17	83.37	89.84	91.51	91.64	91.64	91.37