US Electrical Grid On the Edge of Failure 293
ananyo writes "Facebook can lose a few users and remain a perfectly stable network, but where the national grid is concerned, simple geography dictates that it is always just a few transmission lines from collapse, according to a mathematical study of spatial networks. The upshot of the study is that spatial networks are necessarily dependent on any number of critical nodes whose failure can lead to abrupt — and unpredictable — collapse. The warning comes ten years after a blackout that crippled parts of the midwest and northeastern United States and parts of Canada. In that case, a series of errors resulted in the loss of three transmission lines in Ohio over the course of about an hour. Once the third line went down, the outage cascaded towards the coast, cutting power to some 50 million people. The authors say that this outage is an example of the inherent instability the study describes. But others question whether the team's conclusions can really be extrapolated to the real world. 'The problem is that this doesn't reflect the physics of how the power grid operates,' says Jeff Dagle, an electrical engineer at the Pacific Northwest National Laboratory in Richland, Washington, who served on the government task force that investigated the 2003 outage."
The story of the 2003 blackout (Score:5, Interesting)
Basically, the problem can be almost entirely blamed on FirstEnergy of Ohio. They had, in a matter of hours:
- A software bug in the monitoring tool.
- No backup monitoring, so when the first one wasn't started properly there was no way of knowing there was a problem.
- A plant shutdown due to poor maintenance.
- Multiple power lines failures due to not cutting back trees as they were supposed to.
- Alarm systems breaking, that were simply ignored.
- Utterly failing to notify nearby states that there was a problem so they could prevent it from spreading.
You'll notice that almost all of these problems would not have happened had they not cut corners wherever they thought they could get away with it. And if the US electric grid is in trouble, I'd have every reason to expect that it was other electric companies doing the same sort of thing.
Can we get Morgan Freeman [wikipedia.org] on the case?
Re:Wrong analogy (Score:5, Interesting)
I was thinking the same thing. I felt a little shafted, to be honest, during the 2003 blackout. In my area (Southern Ontario) power was restored quite early, before 11:00pm IIRC. I wish that it had lasted a bit longer so that I could appreciate the beautiful night sky a little longer. You don't often see the milky way within city limits... I almost wish they regularly scheduled these sorts of blackouts. It wouldn't hurt us to be reminded once in a while that the centre of the universe is somewhere above our heads, and not in the middle of the city where we live...
Re:'extrapolated to the real world' (Score:4, Interesting)
Re:Inherently unstable (Score:5, Interesting)
As every electrical engineer knows, an AC transmission system is a quadratic-complex system. And in the sense of both the inherent complexity and the complex numbers involved. There is no energy storage in the system (no inertia), has noticeable delays, and it is tightly coupled. Only high redundancy and decoupling can make the system more reliable. But that is costly. Who wants to pay more?
The challenge is balancing the system's ability to self-heal with the system's ability to self-destruct. There is no reason that losing 3 transmission lines (out of a dozen running through the corridor) should have done anything more than taken three lines worth of subscriber capacity offline. If the system "let them go dark" there wouldn't have been a cascading failure. Instead, in an attempt to self heal (something that works great for just one or two lines going down) the system self destructed instead. Identifying where the tipping point is and acting before it is reached is the only real barrier to preventing such a large problem from happening again. Shame it's taken 10 years to really understand the problem.
Re:The story of the 2003 blackout (Score:5, Interesting)
Basically, the problem can be almost entirely blamed on FirstEnergy of Ohio. They had, in a matter of hours: - A software bug in the monitoring tool. - No backup monitoring, so when the first one wasn't started properly there was no way of knowing there was a problem. - A plant shutdown due to poor maintenance. - Multiple power lines failures due to not cutting back trees as they were supposed to. - Alarm systems breaking, that were simply ignored. - Utterly failing to notify nearby states that there was a problem so they could prevent it from spreading.
You'll notice that almost all of these problems would not have happened had they not cut corners wherever they thought they could get away with it. And if the US electric grid is in trouble, I'd have every reason to expect that it was other electric companies doing the same sort of thing.
Can we get Morgan Freeman [wikipedia.org] on the case?
I can tell you that the industry has really taken this event to heart and learned from it. The linked articles are based on some awfully shoddy conclusions- the scientific article is about interconnected networks in a theoretical sense, and not one of the references has anything to do with the electrical grid. The other link is from "somebody" making conclusions about the power grid based on the scientific article. The grid today is not the same grid we had in 2003. For the last 10 years, NERC [nerc.com] has been throwing down standards and requirements for electrical production and distribution based on the lessons learned in 2003. NERC's website may make them seem like "recommendations", but for many parts of the country, an power station or transmission company must follow their standards if they wish to do business.
A failure of the type experienced in 2003 is unlikely to happen. Even if a company such as FirstEnergy makes colossal screwups, rules are in place which make the other parts of the grid more robust to that kind of problem. The chance of a large-scale blackout is reduced in the last 10 years (as opposed to the articles arguments that it is the same, or greater than ever before).
Think about it. Unless you live on the end of a low-population road, your electricity is probably more reliable than any other service you have. The average electric customer in the US loses service for about 8 hours a year. That is 99.9% reliability. The average Japanese electric customer has 5 minutes of outage per year. That 99.999% reliability sounds great, but those extra 9's cost them dearly. The average TEPCO customer pays about 26-32 cents per KWH. My cost in Connecticut is about 8 cents per KWH. I don't want to pay 3-4 times as much for electricity just to have five 9 reliability. Do you?
Re:Coincidentally... (Score:1, Interesting)
German Efficiency, even more of a myth than ever before.
Re:Coincidentally... (Score:5, Interesting)
Delivering electricity to a socket isn't hard.
Delivering electricity at constant voltage and frequency is already hard.
Delivering electricity at constant voltage and frequency in a grid where a few wandering clouds and a gust of wind create production spikes is definitly hard.
You missed two other factors...
Delivering electricity at constant voltage and frequency in a grid where a few wandering clouds and a gust of wind create *unpredictable* production spikes and drops, and where the source of some of the generation assets is hundreds of miles from the distribution points it needs to get to, is hard.
Lots of people like to talk about how much sun the US gets, and how much space there is to put up wind farms. But they don't realize a few things. One, the best places for PV farms and wind farms are far, far from population centers...and that means that utilities have to figure out how to manage VARS over those distances which is still not a problem that's entirely been solved. T. Boone Pickens had to bail on his whole wind farm venture in the Southeast because of this. And two, while the cost of PV panels (as would be put on the roof of a home of business) has dropped significantly, the majority of the cost of an on-premise solar installation is the anti-islanding gear that ensures the safety of any linemen who show up to deal with a power outage, assuming that only the end of the break in a line that leads back to the rest of the larger grid is live. And the cost of that gear has not changed much at all.
Re:Coincidentally... (Score:5, Interesting)
Hm, american electricity prices: http://www.eia.gov/electricity/monthly/epm_table_grapher.cfm?t=epmt_5_6_a [eia.gov] obviously in $ cents.
A link that compares several countries (in german, but the countries should be easy to read) prices in dollar cents:
http://de.statista.com/statistik/daten/studie/13020/umfrage/strompreise-in-ausgewaehlten-laendern/ [statista.com]
German electricity prices according to wikipedia however are 25 EURO cent.
Pretty strange, as far as I recall I pay 17 EURO cent per kWh.
So you are right: you pay less per kWh in the USA: However you use between 4 to 10 times the electricity a German household or person does. So bottom line you pay far more than we do.
You know efficiency can be defined arbitrarily. You seem to define it on "cost per kWh" we define it on "consumed kWh".
Re:Wrong analogy (Score:4, Interesting)
You did it wrong. I was living in Toronto at the time. Apartment building or not, we just didn't go home. We all banded together to fight a much bigger problem than darkness and stairs. Bars everywhere couldn't keep the beer cold.
Seriously, local bars and pubs were giving away free beer. You've never seen a more instantly-friendly megalopolis.