Fault Line Selection of Low Current Grounding Based on LSTM Neural Network
Weirong Yang, Zhimin Na, Changlin Geng, Anbin Zhang, Panpan Li · 2023
This paper introduces the basic principles of LSTM neural networks and applies LSTM neural networks to low-current ground fault selection. The small current ground fault model is constructed by Simulink simulation software, and the relevant data of the fundamental feature component, the fifth harmonic feature component and the wavelet energy feature component are obtained, and the LSTM neural network is trained, which verifies the accuracy and feasibility of the method.