ENGINEERING

What Keeps the Power On?

+ See all authors and affiliations

Science  29 Oct 2010:
Vol. 330, Issue 6004, pp. 561
DOI: 10.1126/science.330.6004.561-a
CREDIT: THINKSTOCK.COM

Topological models use tools from graph theory to explore connections among elements of complex systems. Recently their application to electricity distribution has stoked fears, including in the U.S. Congress, that massive grids could be crippled by seemingly minor initial disruptions. Targeted attacks on nodes with low loads but high connectivity, some argued, could inflict more damage than attacks on the highest-loaded nodes. Yet such systemwide failures are dictated not only by the nodes and connectivity of the system but also by the laws of Ohm and Kirchhoff that describe the physics of electrical flow. In a systematic comparison of topological and current-flow models, Hines et al. show that topological models, which do not fully capture the effects of electrical flow, can lead to some misleading conclusions. Though all models showed that different types of targeted disruption would inflict more damage than would random failures, the physics-based measure of blackout size—the amount of electrical load curtailed—did not show the same susceptibility to disruption of low-traffic nodes as did the topological measures of connectivity that so alarmed Congress. Allocation of infrastructure protection resources informed by physics-based models would focus on nodes that transport the largest amounts of power.

Chaos 20, 33122 (2010).

Navigate This Article