科研之友
微信
新浪微博
Facebook
分享链接
摘要
The microgrid controller needs to obtain the correct network topology in time to adjust its control strategy and ensure the stable and safe operation of the microgrid. Aiming at the problems of low identification accuracy and poor real-time performance in the existing power grid topology identification methods, this paper proposes a microgrid topology identification method based on graph convolutional network. First, the microgrid topology is abstracted into a line graph. Then, the edge features of the microgrid lines are extracted according to the node feature of the microgrid topology as the node features of the line graph. Furthermore, the adjacency matrix and the bus feature of the abstracted line graph are used to train graph convolutional network to obtain the microgrid topology identification model. Finally, the effectiveness of the proposed topology identification method is verified by experimental data. ? 2022 Automation of Electric Power Systems Press.
- 单位
