Research Article
Bearing Fault Diagnosis Based on Deep Belief Network and Multisensor Information Fusion
Table 4
The parameters of DBN in sensors 1–3 and multisensors.
| Parameters | Sensor 1 | Sensor 2 | Sensor 3 | Multisensors |
| Input | 14 | 14 | 14 | 42 | Output | 4 | 4 | 4 | 4 | Number of hidden layers | 2 | 2 | 2 | 2 | Number of neurons in a layer | 8 | 8 | 8 | 12 | Iteration | 20 | 20 | 20 | 20 | Minibatch size | 10 | 10 | 10 | 10 | Momentum | 0.02 | 0.02 | 0.02 | 0.02 | Learning rate | 0.01 | 0.01 | 0.01 | 0.01 |
|
|