Interesting People mailing list archives

Experts Retrace a String of Mishaps Before the Blackout

From: Dave Farber <dave () farber net>
Date: Sat, 23 Aug 2003 03:31:06 -0400

-------------------------------------------------------/

August 23, 2003
By JAMES GLANZ and ANDREW C. REVKIN

When an overheating electrical transmission line sagged
into a tree just outside Cleveland at 3:32 p.m. on Aug. 14,
the events that would lead to the greatest power failure in
North American history began their furious avalanche,
according to the most extensive analysis of the blackout
yet.

The failure of that transmission line was crucial, because
it put enormous strain on other lines in Ohio. Soon a
utility that serves southern Ohio, with its overloaded
lines close to burning up, sealed itself off, creating in
very real terms an electrical barrier between the southern
part of the state and the northern.

What happened next, by this account, was almost
inevitable:To the north, Cleveland, starving for
electricity, began to drain huge, unsustainable amounts of
power from Michigan and then Ontario, knocking out more
lines and power plants and pushing the crisis to the
borders of northwestern New York.

First the New York system, acting to protect itself, sealed
the state's border with Canada, the analysis found. But
that only created a different, devastating problem: New
York power plants, without anywhere to quickly send
electricity not needed within the state, overloaded their
own system. That in turn quickly led to a general shutdown
- the last stage in the largest blackout in the nation's
history.

That picture, based on large amounts of data from the
utilities involved, was presented yesterday by Cambridge
Energy Research Associates, a private energy consulting
firm based in Cambridge, Mass. Two researchers at Cambridge
Energy, Hoff Stauffer and Lawrence Makovich, presented the
analysis at a conference in which federal energy officials,
industry representatives and reporters participated by
telephone and on the Web.

The Federal Energy Regulatory Commission was represented by
its chairman, Pat Wood III, who would not comment on
whether he thought the specifics of the theory were
correct.

Both Mr. Wood and the Cambridge Energy experts were
unequivocal, though, on one emerging aspect of what went
wrong last week: the system for communication among the
people and organizations that operate that part of the
nation's electrical grid was inadequate.

When problems soar around the grid in seconds, as they did
that day, Mr. Wood said, "you need to make sure that the
communication between the different regions can move
similarly fast." That cannot happen in the current setup,
in which the various oversight organizations are fragmented
and often connected only by phone lines, he said.

Ellen P. Vancko, a spokeswoman for the North American
Electric Reliability Council, the industry group charged
with preventing grid problems, said the organization had
not seen the Cambridge Energy analysis and could not
comment.

Michael Holstein, vice president and chief financial
officer at the Midwest Independent Transmission System
Operator, which helps manage electrical flows on the grid
for certain companies in the region where the blackout
began, called the analysis "an interesting hypothesis." But
he rejected any suggestion that a lack of communication had
contributed to the problems.

"At a certain point in time, things happened so fast that
human intervention was not possible," Mr. Holstein said.

That assessment was echoed by the organization that carries
out day-to-day management of the grid in New York. Ken
Klapp, a spokesman for that organization, the New York
Independent System Operator, said events had moved too
quickly for humans to react and keep the state from going
dark.

Mr. Klapp said a federal investigation just begun by the
Department of Energy would give definitive answers about
the blackout's cause. "Our position is that the facts will
come out," he said. "Eventually it will all be there."

In the days since power went off in eight states and
southern Ontario and left tens of millions of people
without lights and other necessities, some of the companies
involved have made public their partial timelines of events
and sometimes their theories about what led to what.

But Cambridge Energy said yesterday that it had had access
to a more complete set of information. Mr. Stauffer, the
Cambridge Energy researcher, said of the firm's theory: "I
think this one is supported by the facts. This attempts to
say what really happened, where it started and why it
progressed."

And several officials from industry groups charged with
preventing blackouts in various regions around the grid
said the analysis did seem consistent with information they
had collected so far.

The analysis also meshed with data released yesterday by
American Electric Power, the southern Ohio utility that
sealed itself off from the developing trouble to protect
its line and customers. As elsewhere around the system,
sturdy circuit-protecting sensors and relays detected
enormous flows of electricity in American Electric's region
of Ohio.

"It is likely that the automated controls tripped some
transmission lines moments before they would have burned
down because of extremely high power flows out of our
system," said Henry Fayne, American Electric's executive
vice president for energy delivery.

Beyond that, American Electric officials said they would
not discuss theories about the cause of the blackout. But
they did say they had been in extensive contact with
officials of FirstEnergy, the utility whose lines were
failing to the north, during roughly an hour of escalating
trouble. They did not describe the substance of that
communication.

The prologue to the blackout occurred with problems that on
a normal day would have been handled as part of the routine
of moving electricity on the grid. At 2 in the afternoon,
one generator shut down at a power plant east of Cleveland
that is operated by FirstEnergy. About a half-hour later,
American Electric says, it started detecting unusual
activity in shared lines, events that led it to contact
control rooms in the region.

But according to Cambridge Energy, the downward slide to
darkness truly started when the loss of a single
transmission line near Cleveland at 3:06 p.m., in itself an
unremarkable event, was followed at 3:32 by the shutdown of
a second line after it overheated and sagged toward the
tree.

At that point, Mr. Stauffer said, a blackout was
inevitable. From then on, it was just a matter of where it
would be.

Normally, procedures would have delayed sales of power or
diverted them through other lines. But those actions can
require a half-hour or more to undertake, according to
various utility officials, and there was only about a
half-hour to go before the troubles hurtled through the
system.

American Electric Power's safety systems isolated the
trouble spot, as designed. But the severing of the two
crucial Ohio lines was the equivalent of suddenly damming
an onrushing stream: the flow had to divert to find a way
to reach the Cleveland area. The power did that by spilling
around the barrier into Indiana, then into Michigan and
finally back to Ohio.

Within a few seconds, the modest amount of power, around
200 megawatts, that was flowing from a utility in lower
Michigan called the International Transmission Company, or
I.T.C., soared to 10 times that amount.

About the same time, the power that had been flowing from
I.T.C. to Ontario suddenly reversed, pouring still more
electricity onto that part of the grid, while another
Michigan utility began supplying power to the very same
area in an effort to meet the demand. The chain of events
overwhelmed lines and power plants on the I.T.C. system,
and they all began shutting down. By a little after 4:10,
most of the system was out of service.

Shortly thereafter, the rapidly changing flows in Ontario
created alerts or shutdowns of some kind in New York,
though the details remain unclear, and the state sealed its
border with Canada, according to the Cambridge Energy
analysis.

One oddity in Cambridge Energy's reconstruction lies in its
theory that it was the relatively low demand for power in
New York State on Aug. 14 that contributed to the last
episode of the blackout. In the past, of course, it has
been the great demand for power in New York that has
produced problems.

As it happened that day, because the state had relatively
light internal demands, it was exporting lots of power to
Ontario. But when the border was sealed, Cambridge Energy
says, that large flow of power had nowhere to go, and it
sparked the abrupt New York shutdown.

http://www.nytimes.com/2003/08/23/national/23POWE.html?ex=1062622934&ei=1&en=64aa01ac7f456b9e

---------------------------------

Get Home Delivery of The New York Times Newspaper. Imagine
reading The New York Times any time & anywhere you like!
Leisurely catch up on events & expand your horizons. Enjoy
now for 50% off Home Delivery! Click here:

http://www.nytimes.com/ads/nytcirc/index.html

HOW TO ADVERTISE
---------------------------------
For information on advertising in e-mail newsletters
or other creative advertising opportunities with The
New York Times on the Web, please contact
onlinesales () nytimes com or visit our online media
kit at http://www.nytimes.com/adinfo

For general information about NYTimes.com, write to
help () nytimes com.


-------------------------------------
You are subscribed as interesting-people () lists elistx com
To manage your subscription, go to
 http://v2.listbox.com/member/?listname=ip

Archives at: http://www.interesting-people.org/archives/interesting-people/

Current thread:

Experts Retrace a String of Mishaps Before the Blackout Dave Farber (Aug 23)