The bus 1 has constant voltage and it is considered as an infinite bus. The
negative active power means that PVs send power to the grid.

Στις 24 Νοε 2016 13:33, ο χρήστης "Saranya A" <[email protected]> έγραψε:

> I dont see in the case file where you have connected the Bus 1 to the grid.
> Can you clarify on this?
>
> On Thu, Nov 24, 2016 at 1:31 PM, Alexandros Gesios <
> [email protected]> wrote:
>
>> Bus 1 represents the grid connection which has negative active power
>> because the active power from PV is larger that the load active power.
>>
>> Στις 24 Νοε 2016 07:38, ο χρήστης "Saranya A" <[email protected]> έγραψε:
>>
>>> I ran the "casePV3" and observed that bus 1 which is slack bus has Pg as
>>> -3MW. Can a bus which has photovoltaic unit as generator observe real
>>> power? I am not pretty aware of the power electronics of it. I think if you
>>> can reason that out you may get some direction.
>>>
>>> Regards,
>>> Saranya A
>>>
>>> On Thu, Nov 24, 2016 at 4:08 AM, Alexandros Gesios <
>>> [email protected]> wrote:
>>>
>>>> Ι made some changes in the matrices and OPF completes but I do not
>>>> think that it allocates the reactive power among the PVs in an optimal way
>>>> regarding the losses.
>>>> The last three PVs absorb Qm reactive power while the other operate in
>>>> a unity power factor.
>>>> I attach the case in order to be more specific. Also the network is
>>>> radial.
>>>>
>>>>
>>>> 2016-11-23 20:46 GMT+02:00 Alexandros Gesios <
>>>> [email protected]>:
>>>>
>>>>> Dr. Zimmerman
>>>>>
>>>>>
>>>>> I applied the change you have mentioned and the result is the following:
>>>>>
>>>>>
>>>>> MATPOWER Version 6.0b1, 01-Jun-2016 -- AC Optimal Power Flow
>>>>> MATLAB Interior Point Solver -- MIPS, Version 1.2.1, 01-Jun-2016
>>>>>  (using built-in linear solver)
>>>>> Numerically Failed
>>>>>
>>>>> Did not converge in 9 iterations.
>>>>>
>>>>> >>>>>  Did NOT converge (0.11 seconds)  <<<<<
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> 2016-11-23 19:54 GMT+02:00 Ray Zimmerman <[email protected]>:
>>>>>
>>>>>> I think you want to change the first column of your gencost entries
>>>>>> to 2 (polynomial) instead of 1 (piecewise linear). Hopefully, that fixes
>>>>>> your problem and it will solve.
>>>>>>
>>>>>>    Ray
>>>>>>
>>>>>> > On Nov 23, 2016, at 10:30 AM, Alexandros Gesios <
>>>>>> [email protected]> wrote:
>>>>>> >
>>>>>> > Hello to everyone
>>>>>> >
>>>>>> > I am new to MATPOWER and I am trying to operate an OPF analysis in
>>>>>> a MV grid constisted of two feeders.
>>>>>> > The first feeder has 10 inductive loads and the second feeder has
>>>>>> 10 photovoltaic units and their are connected to the grid by the slack 
>>>>>> bus.
>>>>>> > What I want to achieve is loss minimization by allocating the
>>>>>> reactive power among the photovoltaic units in a optimal way. The nominal
>>>>>> active power of the PV units is 10.6 MW and the load is 6 MW so the extra
>>>>>> active power is being absorbed by the HV/MV Grid.
>>>>>> > Also I want the slack bus to operate at constant 1.05 pu voltage so
>>>>>> in the mpc.bus slack bus has voltage limits Vmax=Vmin=1.05.
>>>>>> > Also I want the PV units to operate in nominal power so in mpc.gen
>>>>>> their active power limits are Pmax=Pmin=Pnom.
>>>>>> >
>>>>>> > Finally the mpc.gencost matrix constists of 11 rows and every row
>>>>>> has these elements     1    0       0       2       2   4; because I read
>>>>>> in the archive that in order to achieve loss minimization you need the 
>>>>>> same
>>>>>> linear costs in the gencost matrix.
>>>>>> >
>>>>>> > When I run runopf(mycase) I take the following errors
>>>>>> >
>>>>>> > Index exceeds matrix dimensions.
>>>>>> >
>>>>>> > Error in opf_setup (line 84)
>>>>>> >   x1 = mpc.gencost(pwl1, COST+2);
>>>>>> >
>>>>>> > Error in opf (line 198)
>>>>>> > om = opf_setup(mpc, mpopt);
>>>>>> >
>>>>>> > Error in runopf (line 75)
>>>>>> > [r, success] = opf(casedata, mpopt);
>>>>>> >
>>>>>> > Also, I would like to ask whether my problem can be formulated in
>>>>>> MATPOWER.
>>>>>> >
>>>>>> >
>>>>>> > Thank you in advance
>>>>>> >
>>>>>> >
>>>>>> > Alexandros Gesios
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>

Reply via email to