In order to minimize the losses, you need to use equal, but non-zero, marginal
costs. In your example, the total cost is non-zero, but the marginal cost is
zero, so the objective function value is 20, independent of the solution. In
this case, the solver simple terminates as soon as it finds a feasible
solution. For example, try it with ...
mpc.gencost = [
2 0 0 2 10 0;
2 0 0 2 10 0;
];
... and I think you'll see the results you expect.
--
Ray Zimmerman
Senior Research Associate
211 Warren Hall, Cornell University, Ithaca, NY 14853
phone: (607) 255-9645
On Mar 21, 2011, at 8:13 AM, Guilherme Lira wrote:
> Simon,
>
> Thanks for the input.
>
> Yes, Pmax for the generator at bus 1 is 35 MW, which is 15 MW short of
> being able to supply the load alone, but (leaving both online) in this
> substation emulation it is desirable to fully use its capacity as its
> connection to the load has the lowest impedance. I'm thinking that,
> even though they are being modeled as substation busbars, the
> generators' operating points might be limited by their capability
> curves, is that correct?
>
> Regards,
> Guilherme Lira
>
> On Mon, Mar 21, 2011 at 6:36 AM, Simon Schneider <[email protected]>
> wrote:
>> Hi Guilherme,
>>
>> if you turn off the second gen (mpc.gen(2,8) = -1) you see, that several
>> other constraints are violated and the opf doesn´t converge.
>> The Generator at 1 isn´t powerfull enough to satisfy the load alone.
>>
>> Greetings
>> Simon
>> Am 3/20/2011 10:33 PM, schrieb Guilherme Lira:
>>>
>>> Ray/others,
>>>
>>> In order to test the power loss minimization, I have tried modeling
>>> the generators with equal linear costs simulating substation busbars.
>>>
>>> Please have a look at the attached m-file, it contains a simple test
>>> case consisting of two generators connected to a load, with different
>>> branch impedance values. Theoretically, minimizing losses would mean
>>> that the generator (substation busbar) connected to the load through
>>> the branch of lowest impedance would be serving the load at full power
>>> (Pmax), right? Can you please explain why it is not?
>>>
>>> Thanks,
>>> Guilherme
>>
>>
>
>
