Among other goals, the Smart Grid will reduce or eliminate the possibility of “cascading dominoes” blackouts (see the figure) like the one that darkened parts of the northeastern U.S. on August 14, 2003. The blackout affected an estimated 10 million people in Ontario and 45 million people in eight U.S. states.
According to the final report on the event, at 1:31 p.m., local time, a generating plant near Cleveland, Ohio, went offline because of excess electrical demand. (It was a hot day.) The plant transferred that demand onto some high-voltage lines in a remote area, which then got too hot and sagged into some trees that should have been trimmed. There was a short circuit and the first domino fell.
From that point on, the problem cascaded as multiple alarm systems intended to isolate the fault to a small geographical area failed one after another. At 4:13 p.m., after a series of cascading failures, 256 power plants had gone offline, most of them unnecessarily, because automatic controls had already separated sections of the grid into localized islands.
For the 2003 blackout, that’s where the grid should have been “smarter.” But the underlying problem has been diagnosed as something much broader, and that’s what’s driving the Smart Grid in the U.S.
Fundamentally, the grid needs to be redesigned, rather than patched, and upgraded from the existing 20th century pastiche to something that anticipates and has a well thought-out response to 21st century demands, energy sources and costs, and risks.