Research Article
Gearbox Fault Identification and Classification with Convolutional Neural Networks
Table 6
Parameters tuning of the architecture of CNN.
| Number | Architecture of CNN | Classification rate | Time (s/epoch) | | | | | | |
| #1 | | 6 | 2 | 12 | 2 | 5 | 86.73% | 11.6 s | #2 | | 8 | 2 | 8 | 2 | 5 | 88.48% | 12.8 s | #3 | | 12 | 2 | 12 | 2 | 5 | 92.50% | 21.7 s | #4 | | 8 | 4 | — | — | 5 | 86.71% | 8.00 s | #5 | | 6 | 2 | 12 | 2 | 5 | 90.23% | 3.90 s | #6 | | 8 | 2 | 8 | 2 | 5 | 89.50% | 3.80 s | #7 | | 6 | 2 | 6 | 1 | 5 | 95.71% | 2.40 s | #8 | | 6 | 1 | 6 | 1 | 5 | 98.77% | 4.50 s | #9 | | 6 | 2 | — | — | 5 | 96.71% | 1.04 s | #10 | | 8 | 2 | — | — | 5 | 98.35% | 1.30 s | #11 | | 12 | 2 | — | — | 5 | 98.20% | 2.02 s |
|
|