IEEE: About Wind Turbines
mo at ccr.org
Sat Feb 6 18:45:40 CST 2010
the problem that happened in California was blatant,
malicious market manipulation, pure and simple.
doing DC interties between smaller networks has
huge advantages - it's easy to make the power flow
the direction you want across a DC intertie, and
it doesn't involve messing with reactive power.
btw - the last great blackout happened because of
insufficient imaginary (ie, reactive) power,
not insufficient real power. essentially the power
factor got sufficiently out of whack that real voltage
couldn't be maintained and that started the cascade.
while speed-of-light-delay on data networks does indeed
matter, the real governing value is actually much slower -
the rate of routing update notifications, but it amounts
to the same thing as synchrony in an AC network. ie,
every part of the network has to know the state of every
other part of the network. in an AC network, that's
speed-of-light-minus-velocity-factor, and in a packet
network, that's certainly no faster than update flooding
rate for link-state protocols. if the state of the network
is changing faster than the news can propagate across the
network, "grave disorder will result". (quote from Dennis
Richie on the malloc()/free() man page) in particular,
it will oscillate, very possibly chaotically as multiple
feedback paths create a mesh of coupled oscillators
On 2/6/10 2:53 PM, Alex Fraser wrote:
> I did some studying of the grid(s) a couple of years ago. I wondered
> about the speed of light problem and how they could ever balance such a
> large complex network(s). It is interesting comparing it to data
> networks. BTW As a network guy in the 90's I was intrigued by the
> competition between Token Ring and Ethernet, the later of course winning
> out. People had to take a leap to believe in a statistical approach and
> ever faster hardware made the leap easy.
> I'm not an engineer (though I play one on Face Book), so some of the
> math eludes me. I'm not an economist either (I do use money often) and
> once again some of the math eludes me. I did some casual reading on
> complexity and chaos. Turbulence is so exciting to me. The bit about two
> non linear systems interacting could go into a chaotic state in a very
> short time really impressed me. I like the term "strange attractor" a lot.
> So it being a snowy day and having drank cups and cups of Kenyan "AA"
> coffee let's consider two interacting non linear networks, one being a
> very large power grid and the other being a very large economic system.
> I might mention that I looked into the grid system back when they were
> having problems in California. It turned out that the cause of the
> problems wasn't the grid, but the money system used to control it.
> Certain controls on the trading of power had been removed and it went
> quickly wild and failed. Well it didn't fail from the point of view of
> those pocketing the money, but the electrical delivery sure failed. How
> can we design a system for power delivery based on two complex systems
> interacting where there are times when they can and do drive each other
> into failure mode? And they both are too big to fail.
> There are always scaling issues. Could we do better with many smaller
> grids? Independent linked is different than interdependent. Have any
> studies been funded on NOT turning this over to mega corporations? Who
> would do these studies? How would we ever know?
> No electrons were harmed in the creation of this message
> ~~~********************Alex Fraser********************~~~
> Tacos mailing list
> Tacos at amrad.org
"Of course it's hard!
If it was easy, we'd be buying it from somebody else!"
More information about the Tacos