CNN-based acoustic identification of gas–liquid jet: Evaluation of noise resistance and visual explanation using Grad-CAM

Keyphrases

• CNN-based 100%
• Noise Resistance 100%
• Gradient-weighted Class Activation Mapping (Grad-CAM) 100%
• Visual Explanation 100%
• Two-phase Jet 100%
• Acoustic Identification 100%
• Signal-to-noise Ratio 75%
• ResNet18 75%
• Water Flow 50%
• Noisy Environment 50%
• Sodium-cooled Fast Reactor 50%
• Anomaly Analysis 50%
• Background Noise 25%
• Spectral Peak 25%
• Identification Method 25%
• Signal Intensity 25%
• Intensity Pattern 25%
• Flow State 25%
• Steam Generator 25%
• Safe Operation 25%
• High-frequency Noise 25%
• VGG16 25%
• AlexNet 25%
• Time-frequency Representation 25%
• Low-level Noise 25%

Engineering

• Signal-to-Noise Ratio 100%
• Computer Aided Manufacturing 100%
• Convolutional Neural Network 100%
• Sodium-Cooled Fast Reactor 66%
• Flow State 33%
• Signal Intensity 33%
• Steam Generator 33%
• Noise Level 33%
• Safe Operation 33%

Physics

• Signal-to-Noise Ratio 100%
• Convolutional Neural Network 100%
• Water Flow 66%
• Fast Reactor 66%