Operating Power Grids with Few Flow Control Buses

TL;DR

This paper investigates the minimal number of flow control buses needed in power grids to achieve optimal and feasible power flows, demonstrating that few controllers can suffice for high controllability and cost reduction.

Contribution

It provides an experimental analysis showing that a small number of flow control buses can achieve global optimality and improve feasibility under increased loads, supported by graph-theoretic insights.

Findings

Few controllers suffice for optimal power flow.

Adding controllers reduces flow costs.

Controllers extend feasible load range.

Abstract

Future power grids will offer enhanced controllability due to the increased availability of power flow control units (FACTS). As the installation of control units in the grid is an expensive investment, we are interested in using few controllers to achieve high controllability. In particular, two questions arise: How many flow control buses are necessary to obtain globally optimal power flows? And if fewer flow control buses are available, what can we achieve with them? Using steady state IEEE benchmark data sets, we explore experimentally that already a small number of controllers placed at certain grid buses suffices to achieve globally optimal power flows. We present a graph-theoretic explanation for this behavior. To answer the second question we perform a set of experiments that explore the existence and costs of feasible power flow solutions at increased loads with respect to the number of flow control buses in the grid. We observe that adding a small number of flow control buses reduces the flow costs and extends the existence of feasible solutions at increased load.