Root Cause Reasoning and Visualization Analysis of Abnormal Line Loss in the Substation Based on Knowledge Graph

The analysis of abnormal line loss in the substation is an important issue in improving operating efficiency and reducing energy consumption in the power system. Traditional line loss analysis methods face the problems of large data volume, complex abnormal causes and weak correlation, making it difficult to locate the root cause of abnormalities efficiently and accurately. Therefore, this paper presents a method of analyzing the root cause and visualization of the loss of power line in substation. This paper analyzes the relationship between devices, lines and other multi-dimension data in power system based on the knowledge graph technique. This paper builds the knowledge graph of substation line loss, and combine the data of the power equipment and the sensor with the line loss information to form the system knowledge network. Then, by deeply mining the association rules in the graph, the root cause of abnormal line loss is accurately located. The experiment results show that the precision of this method is 15% higher than that of the conventional methods, and the efficiency and precision of the anomaly detection are greatly increased. Furthermore, the system is also able to display the results of the abnormal root cause analysis in combination with the visualization technique, which helps the decision-makers to make an accurate judgement.