Globally Optimal AC Power System Upgrade Planning under Operational Policy Constraints

Abstract

In order to accommodate the increasing amounts of renewable generation in power distribution systems, system operators are facing the problem of how to upgrade transmission capacities. Since line and transformer upgrades are costly, optimization procedures are used to find the minimal number of upgrades required. The resulting design optimization formulations are generally mixed-integer non-convex problems. Traditional approaches to solving them are usually of a heuristic nature, yielding no bounds on suboptimality or even termination. In contrast, this work combines heuristics, lower-bounding procedures and practical operational policy constraints. The resulting algorithm finds both suboptimal solutions quickly and the global solution deterministically by a Branch-and-Bound procedure augmented with lazy cuts.