Flow-Aware GNN for Transmission Network Reconfiguration via Substation Breaker Optimization

Dekang Meng, Rabab Haider, Pascal van Hentenryck · 2025

Read Flow-Aware GNN for Transmission Network Reconfiguration via Substation Breaker Optimization by Dekang Meng, Rabab Haider, Pascal van Hentenryck (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-2508-01951.txt.

Description

Dekang Meng, Rabab Haider, and Pascal van Hentenryck teach neural networks to reconfigure power grids far faster than classic solvers. Their 2025 paper deals with choosing substation breaker settings to maximize power exports between regions. Framed as a mixed integer program, the task is NP hard and slow on large networks. Their framework, OptiGridML, replaces repeated solves with two linked networks.

One network works on a line graph form of the grid to estimate power flows for a given layout. The other, a mixed type graph network, predicts breaker states under structural and physical limits. A physics based loss ties the two by enforcing Kirchhoff's law on predicted flows. On synthetic grids with up to one thousand breakers, exports rose by up to eighteen percent. Answers came in milliseconds instead of hours.

Catalog record

CREATORDekang Meng, Rabab Haider, Pascal van Hentenryck
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE37 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2508-01951

Browse the full library · About the Tower of Babel