A graph convolution network model for multi area distribution system fast state estimation
Author(s)
Ranjbar, Mohammad Amin
Azad, Sasan
Ameli, Mohammad T
Ameli, Hossein
Strbac, goran
Type
Conference Paper
Abstract
State estimation (SE) plays a crucial role in the
operation and control of power systems. While traditional
approaches are promising, they face challenges such as
increased computation time in large networks and reduced efficiency in specific scenarios due to limited knowledge of network topology. The network is divided into areas, and only local measuring devices are used for state estimation within each area. Distance data that may lead to model misinterpretation is not included in this approach. By performing calculations in parallel for different areas, state estimation time is reduced, and
the accuracy of the results is improved. This paper also develops a graph convolution network (GCN)-based model to enhance SE accuracy, which considers the topological information of the power system in the form of an adjacency matrix. The model is evaluated on an 118-bus network, with results confirming its effectiveness based on various evaluation metrics.
operation and control of power systems. While traditional
approaches are promising, they face challenges such as
increased computation time in large networks and reduced efficiency in specific scenarios due to limited knowledge of network topology. The network is divided into areas, and only local measuring devices are used for state estimation within each area. Distance data that may lead to model misinterpretation is not included in this approach. By performing calculations in parallel for different areas, state estimation time is reduced, and
the accuracy of the results is improved. This paper also develops a graph convolution network (GCN)-based model to enhance SE accuracy, which considers the topological information of the power system in the form of an adjacency matrix. The model is evaluated on an 118-bus network, with results confirming its effectiveness based on various evaluation metrics.
Date Acceptance
2025-08-01
Publisher
IEEE
Copyright Statement
Subject to copyright. This paper is embargoed until publication. Once published the author’s accepted manuscript will be made available under a CC-BY License in accordance with Imperial’s Research Publications Open Access policy (www.imperial.ac.uk/oa-policy).
Source
IEEE PES ISGT Middle East 2025
Publication Status
Accepted
Start Date
2025-11-23
Finish Date
2025-11-26
Coverage Spatial
Dubai, UAE