Distributed Optimization in the Electric Power System

Xinliang Dai and Veit Hagenmeyer

Karlsruhe Institute of Technology (KIT)

Friday, April 28, 2023, 14:00

Room 01-012, Georges-Köhler-Allee 102, Freiburg 79110, Germany

Abstract: As one of the most practically important optimization problems in the electric power system, nonconvex AC Optimal Power Flow (OPF) is extensively studied. During the energy transition, centralized energy management is challenged due to privacy and/or sovereignty concerns. However, as an alternative, solving the AC OPF problem in a distributed fashion empowers different grid operators to compute the overall grid state without having to share grid models—this is a practical problem to which industry does not have off-the-shelf answers. Thereby, we propose to handle the AC optimal power flow problem by the Augmented Lagrangian based Alternating Direction Inexact Newton method (ALADIN), which comes with convergence guaranteed. Towards large-scale power systems, several methods are introduced to improve numerical robustness of the ALADIN algorithm. And a fully Matpower-compatible distributed optimization toolbox will be available on GitHub soon.