True, except for the corrections made below …

On Aug 6, 2013, at 11:02 AM, spyros gian <[email protected]> wrote:

> Dear Dr Zimmerman,
> 
> In our class, the followings were mentioned for the ACOPF, and I would like 
> to verify with you:
> 
> -> in the mpc.bus matrix, the columns:
>     'area', and 'zone' : not taken into account. Thus we can let them have a 
> value of '1'.

Area can be used for output summaries, but it does not affect the computations.

>     'basekV'               : not taken into account. The results are shown in 
> per unit after all.  
>     'Vm' and 'Va'       : not taken into account. Thus we can let them have a 
> value of '1' and '0' respectively.

These provide starting voltages. Many of the solvers ignore this, but some do 
not. For example, minopf does use it as a starting point.

> ->in the mpc.gen matrix, the column:
>     'Vg'                      : not taken into account. Thus we can let it 
> have a value of '1'.
>      
> -> in the mpc.branch matrix, the columns: 
>     'ratio', and 'angle': not taken into account. Thus we can let them both 
> have a value of '0'.

These are *not* ignored. These are the parameters that determine the 
off-nominal tap ratio of a transformer and the phase shift angle of a phase 
shifting transformer. Zero means that it is a transmission line and not a 
transformer.

>     'rateA'                   : if we want no line constraints, we can give 
> it a large value e.g. '99999'

Better to give it a value of 0 if you want no line constraints. If you use a 
large number the constraints are still included in the problem (increasing the 
size) though they will never bind. Setting them to 0 actually removes them from 
the problem.

>     'rateB', and 'rateC' : not taken into account. We can let them '0' 
> (normal operation).
> 
> Do you agree?
> 
> Thank you
> Syros


-- 
Ray Zimmerman
Senior Research Associate
B30 Warren Hall, Cornell University, Ithaca, NY 14853
phone: (607) 255-9645


Reply via email to