Anomaly Detection Method of Distribution Network Line Loss Based on Hybrid Clustering and LSTM

The line loss rate of the distribution network is one of the important evaluation indicators for power companies. It has a significant impact on the economic benefits of the power companies. Therefore, accurate diagnosis of abnormal line loss is an urgent problem for power companies to solve. A method for detecting abnormal line loss of distribution network based on hybrid clustering and long-short term memory is proposed in this paper. Due to the long time required for data prediction, hybrid clustering is proposed to quickly detect feeders with abnormal line loss. And then use long-short term memory to predict the line loss and load to realize the detection of the transformer stations under the jurisdiction of the abnormal feeders. The distribution network operating data under the jurisdiction of a certain municipal power company is used in this article to carry out the simulation test. And designed multiple sets of comparative experiments to verify the effectiveness of the proposed methods. The test results showed that the method proposed in this article can quickly and accurately diagnose abnormal line loss and has practical application value.