A Novel Fault Identification Technique for Transmission Lines based on Spectral Entropy and One- Dimensional CNN

This work proposes a fault detection and classification method using spectral entropy and IDCNN. The proposed method uses one end current signals for analyzing the fault and no fault scenarios. The current signals are recorded with 1kHz sampling frequency and processed using spectral entropy. The input current features are then given to 1D-CNN network for fault detection and classification. The proposed method has been validated varying fault and no fault situations. The proposed method has specificity and sensitivity of 100
^%
.