Abstract

Proceedings Abstracts of the Twenty-Third International Joint Conference on Artificial Intelligence

Planning with MIP for Supply Restoration in Power Distribution Systems / 2900
Sylvie Thiébaux, Carleton Coffrin, Hassan Hijazi, John Slaney

The next generation of power systems faces significant challenges, both in coping with increased loading of an aging infrastructure and incorporating renewable energy sources. Meeting these challenges requires a fundamental change in the operation of power systems by replacing human-in-the-loop operations with autonomous systems. This is especially acute in distribution systems, where renewable integration often occurs. This paper investigates the automation of power supply restoration (PSR), that is, the process of optimally reconfiguring a faulty distribution grid to resupply customers. The key contributions of the paper are (1) a flexible mixed-integer programming framework for solving PSR, (2) a model decomposition to obtain high-quality solutions within the required time constraints, and (3) an experimental validation of the potential benefits of the proposed PSR operations.