Generalizable Graph Neural Networks for Robust Power Grid Topology Control Journal: Applied Energy
Matthijs de Jong, Jan Viebahn, Yuliya Shapovalova · 2025
Read Generalizable Graph Neural Networks for Robust Power Grid Topology Control Journal: Applied Energy by Matthijs de Jong, Jan Viebahn, Yuliya Shapovalova (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-07186.txt.
Description
Matthijs de Jong, Jan Viebahn, and Yuliya Shapovalova use graph neural networks to steer power grid layouts before lines overload. Switching connections at substations can relieve congestion, but searching the options by classic means is too slow. How the grid is drawn for the network turns out to matter: the popular uniform drawing hides which switches sit at which busbar, a flaw the team names busbar information asymmetry.
Their varied drawing, which gives different part types their own handling, fixes the flaw. Trained by copying expert decisions, the varied version ran the grid best on familiar layouts. Both network types adapted to new layouts better than a plain fully connected model, which matters for real control rooms.
Catalog record
| CREATOR | Matthijs de Jong, Jan Viebahn, Yuliya Shapovalova |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 45 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-07186