An Integrated Method of Rolling Bearing Fault Diagnosis Based on Convolutional Neural Network Optimized by Sparrow Optimization Algorithm

Figure 8

Dimension-reduced maps of different running state: (a) normal, (b–d) inner race fault of different fault diameters, (e–g) ball race fault with different fault diameters, and (h–j) outer race fault of different fault diameters (from 0.1778 mm to 0.5334 mm).

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)