Identification of corona discharges based on wavelet scalogram images with deep convolutional neural networks

Keyphrases

• Deep Convolutional Neural Network (deep CNN) 100%
• Corona Discharge 100%
• Wavelet Scalogram 100%
• Scalogram 80%
• Discharge Pulse 40%
• Discharge Current 40%
• Continuous Wavelet Transform 20%
• Wavelet Basis 20%
• Pulse Current 20%
• Time-varying Data 20%
• Current Pulse 20%
• Pre-trained Convolutional Neural Networks 20%
• Sampling Interval 20%
• Generalization Ability 20%
• Shunt Resistor 20%
• Gap Size 20%
• Time-frequency Representation 20%
• Noise Component 20%
• AlexNet 20%
• MobileNet 20%
• Convolutional Neural Network Model 20%
• Data Shifting 20%
• External Noise 20%
• Negative DC Corona 20%
• New Wavelets 20%
• Variable Gap 20%
• Ensemble Structure 20%
• Commercial Sensors 20%
• Positive DC Corona 20%
• ShuffleNet 20%
• Two-electrode System 20%
• Data Acquisition Time 20%
• DC Corona Discharge 20%
• High Frequency Current Transformer 20%

Engineering

• Corona Discharge 100%
• Scalogram 100%
• Convolutional Neural Network 100%
• Current Pulse 33%
• Transients 16%
• Network Model 16%
• Current Transformer 16%
• Frequency Representation 16%
• Interval Data 16%
• Sampling Interval 16%
• Acquisition Time 16%

Physics

• Electric Corona 100%
• Wavelet 100%
• Convolutional Neural Network 100%
• Wavelet Analysis 16%
• Data Acquisition 16%
• Transients 16%