Centrality Fingerprints for Power Grid Network Growth Models

TL;DR

This paper introduces the 'centrality fingerprint' tool to compare power grid growth models by analyzing current flow and centrality measures, aiding in understanding and modeling power grid development.

Contribution

The paper presents a novel method, the 'centrality fingerprint,' for evaluating and comparing different power grid growth models based on network centrality and current flow analysis.

Findings

Centrality fingerprints effectively differentiate growth models.

Application to Maryland grid demonstrates practical utility.

Method reproduces properties of real power grids.

Abstract

In our previous work, we have shown that many of the properties of the Florida power grid are reproduced by deterministic network growth models based on the minimization of energy dissipation $E_\mathrm{diss}$. As there is no $a~ priori$ best $E_\mathrm{diss}$ minimizing growth model, we here present a tool, called the "centrality fingerprint," for probing the behavior of different growth models. The centrality fingerprints are comparisons of the current flow into/out of the network with the values of various centrality measures calculated at every step of the growth process. Finally, we discuss applications to the Maryland power grid.