Two Algorithms Analysing Discharge Parameters Based on Neural-network and Wavelet Transformer
- Resource Type
- Conference
- Authors
- Ruan, Fangming; Guan, Sheng; Meng, Yang; Yin, Lan; Chen, Yanli; Zhou, Kui
- Source
- 2021 IEEE International Joint EMC/SI/PI and EMC Europe Symposium EMC/SI/PI and EMC Europe Symposium, 2021 IEEE International Joint. :794-797 Jul, 2021
- Subject
- Components, Circuits, Devices and Systems
Engineered Materials, Dielectrics and Plasmas
Fields, Waves and Electromagnetics
General Topics for Engineers
Electrodes
Electric potential
Current measurement
Neural networks
Electrostatic measurements
Electrostatic discharges
Prediction algorithms
electrostatic discharge
characteristic parameters
neural network
nonlinear fitting
prediction
- Language
Special relationship exists between environmental conditions and discharge characteristic parameters in electrostatic discharge (ESD) events. The neural network can explore the potential law between input and output if taken discharge condition parameters as a neural network input. Characteristics of discharge results are affected by environmental conditions, and hence discharge parameters can be described with the output of a neural network. Two algorithms of artificial intelligence were used to analyzing discharge results in electrostatic discharge. The nonlinear relationship between discharge conditions and discharge effect may be a new potential discharge feature. Noise in discharge current can be suppressed with wavelet and Kalman filter method. The characteristics measured in the real experiment were compared with the prediction parameters from the neuro-network calculation result. According to the prediction data, the discussion was conducted on correctness accuracy and the discharge process trend analysis.