The Phase Selection of Three-phase Arc Fault Based on Improved TCN-LSTM

Abstract According to the increasingly severe situation of electricity safety. Aiming at difficulty in extracting arc fault features low-voltage three-phase systems and inconspicuous phase distinction, a selection method based on improved LSTM-TCN neural network is proposed. Firstly, structure activation function Long Short-Term Memory Network (LSTM) Temporal Convolutional (TCN) are according experimental data. Finally, feature vectors obtained by LSTM TCN fused determine difference. The results show that model proposed this paper can effectively extract characteristics faults phases. It great significance industrial household electrical