If you have a power flow solution, you could relax the generator, voltage and
branch flow limits to include the power flow solution. Then gradually begin to
tighten the constraints by moving the limits gradually toward their original
values and resolving the OPF. This way you may be able to discover which
constraints are in conflict with one another.
Ray
> On Nov 8, 2016, at 11:05 AM, Angel Pastor <[email protected]> wrote:
>
> And ¿is there a way to "make" the OPF converge?... Which would be the main
> reasons why an OPF does not converge?
>
>
> De: [email protected]
> <mailto:[email protected]>
> <[email protected]
> <mailto:[email protected]>> en nombre de Ray
> Zimmerman <[email protected] <mailto:[email protected]>>
> Enviado: martes, 08 de noviembre de 2016 11:45 a.m.
> Para: MATPOWER discussion forum
> Asunto: Re: Optimal reactive compensation in MATPOWER
>
> I’m afraid I cannot speculate on the reasons for non-convergence of the OPF
> in your case. There are numerous ways in which an OPF problem can end up with
> an empty feasible region.
>
> Ray
>
>
>> On Nov 8, 2016, at 12:07 AM, Angel Pastor <[email protected]
>> <mailto:[email protected]>> wrote:
>>
>> Thank's for your answer mister Zimmerman, i tried with some of the cases in
>> the Matpower files like the case30 and case14 and it worked.
>>
>>
>> But in the system i'm working the OPF does not converge and it says it
>> numerically fails, when i run the normal power flow it converges but it
>> doesn't satisfy neither the voltage or the generation constraints, ¿that
>> could be the reasson why the OPF does not converge?
>>
>>
>> I was trying to apply an optimal reactive compensation to solve that problem
>> and get the OPF satisfying all the constraints.
>>
>>
>>
>> De: [email protected]
>> <mailto:[email protected]>
>> <[email protected]
>> <mailto:[email protected]>> en nombre de Ray
>> Zimmerman <[email protected] <mailto:[email protected]>>
>> Enviado: lunes, 07 de noviembre de 2016 11:06 a.m.
>> Para: MATPOWER discussion forum
>> Asunto: Re: Optimal reactive compensation in MATPOWER
>>
>> You could add a generator at each bus and then use an OPF to solve for the
>> power flows, with the real power dispatches constrained to their initial
>> values.
>>
>> Ray
>>
>>
>>> On Nov 6, 2016, at 9:40 AM, Angel Pastor <[email protected]
>>> <mailto:[email protected]>> wrote:
>>>
>>> Dear MATPOWER community.
>>>
>>> Hi to everyone, i'm trying to implement reactive power compensation in a
>>> power system, my goal is to make voltage in all buses have values within
>>> the constraints, but i don't know neither the ubication for the
>>> compensation or the MVAR's that would give me the optimal solution and
>>> satisfy all the constraints. So i was wondering if there is a way for
>>> matpower to get me that information.
>>>
>>> My system consist in 14 buses, 5 PV (one generator per bus) and 9 PQ. I was
>>> thinking in modeling the compensators as reactive power generators with no
>>> Pg, as in a OPF the reactive power generated Qg is an optimization
>>> variable, but i only have that variable in the PV buses, and i need to be
>>> able to compensate in any bus to have the optimal solution.
>>>
>>> I could make all the buses PV to put generator in all the buses, but then
>>> the voltage would be attached to de Vg values from the gen_matrix.
>>>
>>> Other idea i had was to add a new variable with the add_userfcn. function
>>> and change the OPF formulation with
>>>
>>> function om = userfcn_reserves_formulation(om, args)
>>>
>>> But i think that variable would still be limitated to to the PV buses and
>>> would only add a reserve to the current generators.
>>>
>>> If you could help me with this, i would really appreciate it.
>>>
>>> Hoping to get response. Thanks in advance.
>>>
>>> Angel Araque, Electrical Engineering student in University of Los Andes.
>>> Venezuela
