Research on Distribution Network Topology Identification Algorithm Based on Distribution Variable Measurement Data

The distribution network is located at the end of the power system and is an important link connecting the transmission system and users. The observability and controllability of the distribution system are the foundation for its safe and stable operation, and also the guarantee for providing good services to users. During the actual operation of the distribution network, the switch state of the network and the physical parameters of the circuit will dynamically change due to network reconstruction, fault isolation, and other reasons. Nowadays, most of these dynamic topology parameters are obtained through manual calculation and verification, which is costly. This paper takes the medium voltage distribution network as the research object, relying on the distribution network operation measurement data provided by smart meter, constructs the dynamic verification of distribution network line transformer topology based on the topology identification model of deep neural network, and it is verified that the accuracy of the calculated line transformer verification is more than 99.85%.