苏 运,赵 琦,瞿海妮.基于机器学习的中压配电网断线不接地故障检测[J].电测与仪表,2019,56(1):96-101. Su Yun,Zhao Qi,Qu Haini.Disconnected Unground Fault Detection in Medium Voltage Distribution Network Based on Machine Learning[J].Electrical Measurement & Instrumentation,2019,56(1):96-101.
基于机器学习的中压配电网断线不接地故障检测
Disconnected Unground Fault Detection in Medium Voltage Distribution Network Based on Machine Learning
As medium-voltage distribution network substations often use the neutral point is ungrounded, so it will not form some characteristics when a single break line both sides of the wire is ungrounded or non-power side of the wire landing. At present, it is still not possible to determine the fault by the relay protection device in the substation. In order to solve this problem, we propose a system that intelligently detects the disconnection of the distribution network based on data and improved random forest algorithm with electrical information systems and pure data-driven methods,which can be applied to real-time fault detection. This is the major innovation in this article. Besides, based on the historical data from an electrical information system in an area in East China, a practical example is analyzed. The accuracy of classification of this method in the test set is as high as 91.44%, and the voltage is judged as the main criterion in the disconnection of the distribution network, which provides a scientific basis for further research on disconnected unground fault detection in distribution networks.
