Research on Defect Identification Algorithm of Power Network Telesignaling and Telemetry Data Based on Dynamic programming

Distribution networks manage numerous devices, however, their level of automation is relatively low, often resulting in false signals and relatively low data quality. Therefore, it has become extremely challenging to directly use telecontrol and telemeasurement data for defect identification. To address this issue, this paper proposes a defect identification algorithm for power grid telecontrol and telemeasurement data based on dynamic programming. Firstly, we employ data cleansing techniques to filter out erroneous data. Then, we introduce a dynamic density clustering-based method for detecting anomalies in power quality data, which can extract feature indicators of abnormal data and identify the defects in telecontrol and telemeasurement data. Next, we utilize the GRU-LightGBM network to identify defects in the telecontrol and telemeasurement data of the power grid. Finally, we compare this method with existing classification methods for defect identification in telecontrol and telemeasurement data through experiments. The experimental results demonstrate that our proposed method not only improves classification accuracy but also enhances generalization ability.