Power Grid Experiment

[Bob Bruhns]

Evidently there are four separate power regions in the North American 
grid.  Inside any one of these four grids, the generators have to match 
the system frequency exactly, and align with the system phase, and stay 
with it.  The generators have to spin at the right speed, but system 
loading changes try to bump their speed around.  Certainly the frequency 
of a generator and even its phase at that frequency need to match the 
system, and the amplitude of the output must also match - although the 
generator must produce a slightly higher amplitude than exists where it 
is on the grid, in order to drive power into the system.

Generator speed is controlled by a combination of the driving force from 
whatever engine it uses (nuclear, fossil, hydro), and the load resulting 
from its output amplitude (voltage) setting.  More driving force 
produces increased rotational speed and therefore higher frequency 
(advancing phase), and higher output voltage setting produces more 
output power and slows down the rotation.  So, constant dynamic 
adjustments of the output amplitude setting and the applied engine power 
are necessary to keep the voltage and frequency right.  Surely the 
generators have a phase error indicator that controls the engine drive, 
just as they adjust their field magnets to maintain the output amplitude 
as necessary.  Output amplitude is not locked to the rotational speed.  
They interact, but they are adjusted separately, and indeed they have to 
be adjusted all the time.

But why anybody would want to stop doing long-term synchronization is 
beyond me - it is just a background adjustment of a process that they 
already have to do 24 hours a day.  They already adjust their output 
amplitude, and they already vary the engine power to keep the rotational 
speed and therefore the frequency and phase matched to the system.  I 
have to believe that the precision with which they do that is very 
important to their operating efficiency.  And I have to think that 
today's control systems must be able to do this with essentially no 
effort at all.

Maybe this is about the unpredictable wind and solar power that is 
starting to enter the grid - maybe it's just harder to control the speed 
in this situation.  Probably the answer is to relax the accumulated time 
error standards a little - but 20 minutes?  Maybe the peak rate would be 
20 minutes in a year, but I see no need for the total accumulated error 
to be anywhere near that much.  Maybe they are really talking about more 
plus and minus time errors caused by long-term error.

    Bob, WA3WDR



On 6/25/2011 11:32 AM, Richard O'Neill wrote:
>  It looks as though we're about to simulate (for real) a third world 
> type of power grid.
> Great, another snafu fiasco in the making. I wonder who dreams up this 
> stuff?
>
> http://hosted2.ap.org/APDEFAULT/495d344a0d10421e9baa8ee77029cfbd/Article_2011-06-24-US-SCI-Power-Clocks/id-f494b2ea48ff4f4b93618f07414af639 
>
> _______________________________________________
> Tacos mailing list
> Tacos at amrad.org
> https://amrad.org/mailman/listinfo/tacos
>