Thank you very much for you help.
Dispatchable loads is exactly what i was looking for. I am having a bit of
trouble in extending the opf. I have read the manual and i think the Callback
functions method is the best for my case.
I am trying to understand how this works using the ‘case9’, simply adding a new
variable and a new cost to the case. I am trying to understand the “userfcn_”
method described in 7.3.2 but i don’t think i am getting this right.
case9 = loadcase(case9);
case9.gen(4,1) = 6; %
case9.gen(4,2) = 0; %Pg, real power output (MW)
case9.gen(4,3) = 0; %Qg, reactive power output (MVAr)
case9.gen(4,4) = 0;
case9.gen(4,5) = 0;
case9.gen(4,6)= 1.025;
case9.gen(4,7)= 100;
case9.gen(4,8) = 1;
case9.gen(4,9) = 0;
case9.gen(4,10) = -1;
case9 = add_userfcn(case9, 'formulation', @userfcn_formulation);
results = runopf(case9);
function om = userfcn_formulation(om)
%% variable bounds
L=-1;
U=1;
Cw = 10 * case9.baseMVA
om.add_var('S',1,case9.gen(4,2),case9.gen(4,10),case9.gen(4,9))
om.add_lin_constraint('limite1',L,1,U,{'S'})
om.add_quad_cost('costo1',[],Cw,0,{S});
end
i think my problem is that i can’t understand how to properly modify the opf
model. I hope someone can help me.
Best regards,
Davide Maglione
Da: Ray Daniel Zimmerman
Inviato: sabato 19 settembre 2020 17:39
A: MATPOWER-L
Oggetto: Re: Bus, Loads and Generators question
That sounds like a dispatchable load … see Section 6.4.2 in the User’s Manual.
Ray
On Sep 18, 2020, at 8:22 AM, Davide Maglione <davide.maglion...@gmail.com>
wrote:
Thank you for the answer,
i am working on a transmission network so i don’t think a dc network model is
adequate. I have read few articles that uses OPF extensible structure to
optimize storage size so i am leaning toward that.
Sorry i didn’t mean to ask you to show me how to use matpower to solve my
problem, i thought my question was way more trivial. I am just struggling a bit
with the User’s Manual.
Hope my question is not too specific:
Is there a way to assign a maximum demand for a load (in order to simulate the
size of the storage), and then, after creating my cost function and
constraints, solve the OPF to track the variable power absorbed by that
load/storage?
Hope i was clear, just asking this to understand if i am following the right
path or i have to walk in another direction to solve this.
Thank you again, best regards,
Davide Maglione
Da: Ray Daniel Zimmerman
Inviato: giovedì 10 settembre 2020 19:58
A: MATPOWER-L
Oggetto: Re: Bus, Loads and Generators question
It sounds like you want to solve a problem that MATPOWER does not implement
directly, so I’m afraid you’ll have to formulate the problem yourself and
determine whether or not it fits MATPOWER’s extensible OPF structure (see also
FAQ #2).
In general, when introducing storage, you need to solve a multi-period problem
that optimizes across multiple time steps simultaneously. If a DC network model
is adequate, you might consider whether MOST would be able to solve your
problem.
Ray
On Sep 10, 2020, at 11:00 AM, Davide Maglione <davide.maglion...@gmail.com>
wrote:
Hi Ray,
thank you for your support and your answers. I am new to power flows so i
struggled a bit, but i managed to succesfully run a pf for a 4 month period in
my network. Thank you for your help.
I want to add energy storage in my network, and, taking into account a surplus
of power occasionally in the 4 month period, i would like to create a new
objective function in order to minimize the cost of installation of the storage
system.
Is the Extended OPF feature the best way to do this? Should i think of the
storage simply as an additional load on the bus and the occasional surplus of
energy as an additional power from the generators?
Thank you for your help,
Best Regards,
Davide Maglione
Da: Ray Daniel Zimmerman
Inviato: venerdì 24 luglio 2020 16:10
A: MATPOWER-L
Oggetto: Re: Bus, Loads and Generators question
It sounds like your P and Q are branch flow values, not nodal loads, so no they
should not be used as loads in the bus matrix.
Ray
On Jul 23, 2020, at 12:42 PM, Davide Maglione <davide.maglion...@gmail.com>
wrote:
Hi Ray and thank you for your reply,
i will try to make my case more clear. I am studying a radial transmission
network (150 kV). There are 3 branches directly connected to a station. I am
thinking of using this station as a “sink” (a load for Matpower) that collects
the power generated by the wind power plants in the network.
My final goal is to do an extended optimal power flow, using some custom
constraints, to optimize size and location of energy storage in order to avoid
wind curtailment by the transmission operator.
I have data of P, Q and I of branches, and V at the nodes (intersection of
branch and power plant). Should i put those data (P,Q) in the 3rd and 4th
column of the bus matrix data despite they are not load buses? Or should i do
something different in my scenario?
I have put an image to semplify things.
Hope this time i have make it more clear, thank you again for your time,
Davide Maglione
Da: Ray Daniel Zimmerman
Inviato: venerdì 17 luglio 2020 22:40
A: MATPOWER-L
Oggetto: Re: Bus, Loads and Generators question
It isn’t completely clear what data you have to create your case, but you
definitely need both load and generation for the system to make any sense.
Total generation is equal to total load plus losses, so you need both.
And dispatchable load refers to loads whose quantities are not fixed, but can
vary depending on the price.
Ray
On Jul 16, 2020, at 10:06 AM, Davide Maglione <davide.maglion...@gmail.com>
wrote:
Hello,
For my thesys work i am trying to recreate a trasmission network with high
penetration of wind power. I have some doubts on how to consider my buses,
because in my network there are no loads but only generators and i am
considering the buses as the nodes of the system.
I have V,I and P datas of each n_i node, so can i put in the 3rd and 4th column
of the bus matrix those flow data despite they are not load buses?
Finally, what do you mean with “dispatchable load”?
Thank you very much,
best regards.
Davide Maglione
Mail priva di virus. www.avast.com
<network.PNG>
--
Questa e-mail è stata controllata per individuare virus con Avast antivirus.
https://www.avast.com/antivirus
