Intelligent Fault Diagnosis of Bearing Based on Convolutional Neural Network and Bidirectional Long Short-Term Memory

<div>Model training iterative process. (a) CNN-BLSTM highest accuracy = 99.55%. (b) CNN-LSTM highest accuracy = 99.45%. (c) CNN highest accuracy = 97.9%. (d) DNN minimum accuracy = 77.15%.</div>

Shock and Vibration

fig10

Figure 10

Figure 10: Intelligent Fault Diagnosis of Bearing Based on Convolutional Neural Network and Bidirectional Long Short-Term Memory 