Hi all,
I'm having a conceptual difficulty here - you might be able to straighten it out...
I have created a model in a big fat spreatsheet to work out the efficiency when sharing a load between power stations (it's a simplified and hypothetical student project - so don't worry I'm not trying to wire into the grid or anything).
I have a desired end load (at 220kv line to line) and I know the impedances of the lines. So I have worked back using KVL and KCL etc to get my required voltage and current at the station end about 120km away.
Then I introduce the second station, about 80km away and balance the total required P & Q (real and reactive power) values between them - so if one station provides 20% of the P, the other provides 80%. Similar for Q. So long as in the end I get my total P and total Q at the end I have a working system.
Then I look at my total power loss as I tweak the P & Q values - I'm trying to find the sweet spot where power losses are the lowest.
My model appears to work - I have a macro that runs through about 1000 possibilities then I graph those as a surface and get a nice curved sheet with the sweet spot at the bottom.
However I am a bit conceptually challenged interpreting the results.
Basically changing the P value affects the losses greatly (as expected) but changing the Q values affects it to a far lesser degree. This is I suppose expected - the P is dissipated in the lines, the Q is not (much). However I can't work out whether my model has a flaw, because the change in loss based on balancing the Q values between the stations is very very small - pretty much inconsequetial in fact.
Should I be looking for something wrong in the calculations, or is altering the split of reactive power load share expected to have very little effect on overall efficiency?
(pf = 0.82 lagging, Load P = 350MW, R=0.2 Ohms /km, X=0.3 Ohms/km, C=7E-9 F/km).
Thanks...