Hi Thomas,

It's always nice to have such a clear description of the problem you are trying 
to solve for questions like this. Unfortunately, I can't think of a way that 
MATPOWER's runopf function can solve the second part of your problem exactly as 
stated. The issue is that the losses and reactive power exchanges are 
non-linear functions of the optimization variables, so the problem does not fit 
the generalized form of the extensible OPF, with it's user-defined variables, 
constraints and costs.

I'm not sure if there is a way to use some sort of linear approximations. I 
couldn't think of a way to do it.

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



On Feb 26, 2010, at 10:09 AM, 
<[email protected]<mailto:[email protected]>> 
<[email protected]<mailto:[email protected]>> wrote:

Dear all

I’m working on an optimization problem. The power grid contains about 520 
branches, 390 buses and 50 generators. Now here is the problem:

The following magnitudes are to be minimized:


 *   The active power losses in part of the grid (in about 300 branches). Ploss 
of each of these 300 branches contributes ($/MW) to the overall costs.
 *   The reactive power exchange at certain branches (at about 15 branches) is 
to be kept below a threshold (below 50% of the active power exchange Pexch). 
Qexch also contributes to the overall costs in $/MVar but only for the amount 
that exceeds the threshold (if Pexch = 10MW and Qexch = 7MVar, we pay only for 
2 MVar).


For this task, we have the following options:


 *   We can change the tap settings of transformers in order to distribute the 
reactive power among these 15 branches.
 *   We can use some of the generators to produce reactive power.



Now here is my approach:

I will use some sort of trial and error to find “good” transformer tap settings 
and afterwards do the “fine tuning” by changing the Qgen of the mentioned 
generators.


Here is my question:

Can I use the runopf-function to get an optimal solution for the second part of 
the above optimization problem, i.e. the triggering of Qgen?


Looking forward to some constructive answers.

Cheers in advance


Thomas


Reply via email to