Gridtopo-GAN for Distribution System Topology Identification

Wu Huayi, Zhao Xu, Jian Zhao, Songjian Chai

Research output: Journal article publicationJournal articleAcademic researchpeer-review

1 Citation (Scopus)

Abstract

Due to the limited presence of monitoring and measurement devices, timely identification of distribution grid topology has been challenging. Therefore, this paper proposes a power grid topological generative adversarial network (Gridtopo-GAN) model to identify the distribution grid topology of either meshed or radial structure with limited measurements. By leveraging the topology preserved node embedding architecture, this model can efficiently handle large-scale systems with different topological configurations. Because of the generative capability of GAN, the model is robust enough when fed with bad measurement data including missing data, commonly encountered in practical applications. Numerical simulations are carried out on the IEEE 33-node system, 118-node, 415-node, and real 76-node distribution systems to demonstrate the effectiveness and efficiency of the proposed topology identification model.

Original languageEnglish
JournalIEEE Transactions on Industrial Informatics
DOIs
Publication statusAccepted/In press - Mar 2022

Keywords

  • Adaptation models
  • Data models
  • distribution grid topology identification
  • Generative adversarial network
  • Generative adversarial networks
  • Markov processes
  • missing data
  • Monitoring
  • Network topology
  • node embedding
  • Topology

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Information Systems
  • Computer Science Applications
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Gridtopo-GAN for Distribution System Topology Identification'. Together they form a unique fingerprint.

Cite this