Before I call it a night, let me throw something out there for folks to
consider and ponder upon.
Commercial grids are very, very large. A "nanosecond of wire" is 30 cm long. A
"microsecond of wire" is 300 m long. A "millisecond of wire" is 300 km long.
The period of an entire cycle for a 60 Hz system is only 16.7 ms. Some of
these trunks run at least 300 km. That amounts to 1/60 of a cycle or 6°.
(Gee, it's late; someone check my mental math, please.)
This is long enough for it to take an appreciable time for a perturbation in
one part of a large system to be sensed in another part of the system.
But here's the big, big fly in the ointment. What if we locate sensors all
over to keep all parts of the system aprised of what's happening elsewhere?
Uh, oh. The signals from those sensors travel at the same speed as the
"ripples" from the perturbations. The messanger cannot outrun the spear.
Sensing and control circuits thus work fine in terms of centralized control
for slow effects, such as load increases or decreases in each geographic area
with time of day and the movement of the sun across the meridians. But they
cannot outrun a line fault. Line faults must be sensed, diagnosed, and
properly protected against by local sensors only and the associated control
circuis have a fraction of a cyclie to do the right thing. Shucks, that
protective action is a perturbation itself and will eventually (fraction of a
cycle) be felt elsewhere in the system.
Massive load shedding (as happened this last week) can lead to a domino effect
that is hard to outrun. Vermont was fortunate to protect itself by virtue of
timely local sensing and isolation procedures. (Besides, they weren't drawing
all that much power from outside anyway, I venture --- but it saved them from
being "instantly" tasked to supply all of the Northeast with their nuclear
reactors.) TVA's control programs recognized the rapid and abnormal shift in
power transmission and took early action, thus protecting their grid area.
Others waited a fraction of a cycle too long.
A couple of times I have seen an unstable "sloshing" of reactive loads in a
submarine electical distribution system and it was a scary sight to behold as
current meters bounced between low (zero) and high (overload) pegs like some
sort of obscene SHM animal. It was like an imaginary beast suddenly grabbing
up all of our loads plus an unimaginable "imaginary" (reactive) load and
running with it under its arm from one side of the plant to the other side.
In each of those cases, we saved the system by opening all tie breakers as
fast as we can and when we estimated that the associated generator's observed
monmentary loading was "about right". Imagine what that's like with gigawatts
of power sloshing around in a system whose dimensions are measured in
kilometers!
It's more a wonder that it works than it is that it "felled down and goed
boom".
Jim
--
James R. Frysinger
Lifetime Certified Advanced Metrication Specialist
Senior Member, IEEE
Office:
Physics Lab Manager, Lecturer
Dept. of Physics and Astronomy
University/College of Charleston
66 George Street
Charleston, SC 29424
843.953.7644 (phone)
843.953.4824 (FAX)