Fault Early Warning and Judgment System of Low-Voltage Substation Based on Deep Learning and Knowledge Map
Zhe Liu, Haisheng Hong, Yan Deng, Qianying Li, Mingyuan Shang, Xian Tang · 2024
This article proposes a low-voltage substation fault warning and analysis system based on deep learning and knowledge graph, aiming to improve the safe operation level and fault handling efficiency of low-voltage substation power grids. Firstly, a prediction model was constructed using deep learning techniques, and features were learned and extracted from the data using Convolutional Neural Networks (CNN) and Long Short Term Memory Networks (LSTM) models. Taking the time series data of current, voltage and other parameters as input, the temporal and spatial features of the data are extracted through multi-layer convolution and pooling operation, and then the time correlation of the data is captured through the LSTM model, and finally the high-dimensional feature representation is obtained, so as to make full use of the temporal and spatial information and time series information. Secondly, the equipment relationship model based on knowledge map is established, and the correlation between equipment is used to assist fault diagnosis and analysis. In the experiment, the system is verified by using real low-voltage data, and compared with the traditional method. The experimental results show that the proposed system can accurately predict the possible faults in the low-voltage area, and diagnose the faults with the help of knowledge map, with high accuracy and efficiency. Therefore, this system has great significance and application value in improving the operation safety and fault handling efficiency of low-voltage power grid.