Veda2.0 Released!

Modeling transmission lines in multi-regional models

I am developing a multi-regional electricity dispatch model. The commodity that can be traded between different regions is electricity. I defined transmission lines as bi-directional trade links between regions. I think I can define each link (transmission line) capacity and cost using ScenTrade_Trade_PARAM.xls, however, I am wondering how I can model capacity addition for each transmission line in the future i.e. finding a way that model can invest in additional capacity if it is necessary. 

(30-11-2016, 01:16 PM)saleh Wrote: I am wondering how I can model capacity addition for each transmission line in the future i.e. finding a way that model can invest in additional capacity if it is necessary. 
Capacity additions for inter-regional transmission line processes can be modelled in the same way as for other processes, i.e. the model can invest in additional capacity if it is necessary, unless capacity additions are prevented by the user (e.g. by NCAP_BND).

Does it not work like that for you?
Thanks Antti, but I am not sure where I can define new processes for trade.
You mentioned that you have "defined transmission lines as bi-directional trade links between regions". Then you have defined the processes for the lines. You should see those processes in the VEDA Browser. You can define the existing capacities, the investment costs for new capacity, and other process parameters for these trade link processes in any scenario file.  But there is some ambiguity caused by there being a process on both sides of each trade link, so you need to choose whether to model the capacities and new investments on both sides or just either side.
It seems possible that you have assumed that the processes you have defined for the trade links would only represent existing capacities. Is that perhaps the problem?

Anyway, that is not the case. You can, of course, define any number of parallel trade links for the same commodity and between the same regions, but each of the trade links will have an associated process on both sides of the link. And all trade processes in TIMES can have both existing capacities and new investments, just like any other processes in TIMES. The fact that you are using VEDA as your user interface does not change this basic feature of TIMES.
Thank you Antti. I found out that is the case and I can have both existing capacity and new investments.

I have another question regarding multi-regional model. I have a share constraint scenario and I would like this scenario to be active only in one region (CA). I use ~UC_Sets: R_E: CA and ~UC_Sets: T_E: CA, however, these constraints are defined for all regions (please see attached files). I would appreciate your help.


Attached Files Thumbnail(s)
Sorry, for not being a VEDA expert.

But I did try to reproduce your problem in the DEMO model, even with two versions of VEDA-FE, but had no success.
I got the UC parameters just for the ROW region (I used R_E: ROW, while the model has the WEU region as well).
I am also puzzled by your use of "T_E: CA", but my test with T_E: ROW did not make any difference.
Moreover, you have strange values for the UC_RHSRTS IE options (140), maybe that is somehow 28*5, but I am getting just 5 in the DEMO model, just as it should be.

If you can yourself reproduce the problem in the DEMO model, then it could be easy to resolve.
Thanks. I reload the model and it solved. I also want to impose an import constraint, i.e. imported electricity should not contribute to more that 35% of total electricity in a specif region/all regions. I use the attached share constraint.
That does not work. I also changed UC_FLO with UC_IRE but that does not help. One option is putting lower limit on generation in each region. But, I would like to know how share constraint works on trade processes. 


Attached Files Thumbnail(s)
Yes, the UC_FLO parameter is meant for standard processes (and auxiliary flows of storage), which have the VAR_FLO flow variables. If you specify UC_FLO for a trade process, it will usually have no effect at all, because the trade processes do not have any VAR_FLO variables. And the UC_FLO parameter does not even have the IE dimension required for referring to trade flows.

For referring to the import/export flows of IRE processes, which have the VAR_IRE variables, you must use UC_IRE, and with that you must also specify the IE dimension (IE=IMP/EXP) to define that the coefficient is to be applied to the import or to the export flow. If I remember correctly, you should use the Other_Indexes column for the IE dimension.

In VEDA-FE, you can also use UC_IRE-I and UC_IRE-E for imports and exports, respectively, which may be easier. I have used these attributes in many constraints, for example for limiting the net imports in proportion to total electricity supply, and they have worked well for me. Maybe you could give them a try?
I have a simple question regarding the "CAP2ACT" parameter in a trade process. For an IRE process created by Veda, for instance a "TB_Comm_Reg1_Reg2" process, will the CAP2ACT by default be equal to 1? So, if I had defined a CAP2ACT for all power plants equal to 8.76 (Act=GWh/Cap=MW), shall I also define (via INS?) a CAP2ACT attribute for the TB process, that would be equal to 8.76?

Nevertheless, as far as I know, the CAP2ACT attribute always refers to the ANNUAL timeslice, even though all power plants processes had been defined in a DAYNITE timeslice, right? It would be the same regarding the TB process?

CAP2ACT is a VEDA attribute, but I think it is directly converted into the TIMES attribute PRC_CAPACT (which I use in VEDA, too).

The modeler is responsible for defining the PRC_CAPACT value appropriately, whenever the process capacity is to be modeled. 
The default value is 1. If there is no need for a process capacity, or the default value is fine, then no action is required. 

PRC_CAPACT defines an unit conversion from the capacity unit to the activity unit.  For example, the conversion from GW to PJ would be 8760*3.6/1000 = 31.536. The capacity and activity units are assumed to be the same for all timeslices, an therefore you cannot use a capacity unit that would be e.g. PJ/a for one timeslice and GW for another timeslice in the model.
Thanks for your response, Antti. Yes, I understand about the "CAP2ACT" units, that must be the same for all timslices. Therefoe, CAP2ACT will always be defined in an ANNUAL basis (in your exemple, 8760*3.6/1000 = 31.536). So, it will never be defined in a DAYNITE basis, even though for a process that is tracked in a DAYNITE (something like  24*3.6/1000 = 0.0864 - for a hourly timeslice in a day).

Well, regarding the need or not of defining a timeslice for the TBs (Trade links) processes, I have followed your advice and tried to define a new value, since I want a CAP2ACT different from the default value (that is equal to 1). I have added it in the trade parameters file. I don´t know the reason, but Veda is not reading it (It can't be reached in the Process Master table or in "Veda View"). During this process, I also found out that Veda is not reading other trade parameters, such as "INVCOST", "VAROM" and "EFF". The file is enclosed (INS sheet).


Attached Files
.xlsx   ScenTrade_Trade_Parameters.xlsx (Size: 725.88 KB / Downloads: 29)
Sorry, I am not able to confirm your findings.
I just tested it in the DEMO model, scenario TRADEAttribs, with the TFM_INS table below:

[Image: attachment.php?aid=222]

As you can see from the Browser snippet, it gets correctly in the database.  And other trade parameters are also read correctly.

Maybe you could also try it first in the DEMO model, and see if it works there?  Rolleyes

Attached Files Thumbnail(s)
Ahh..., just looked at your scenario file again, and I can see that you don't have any Pset_PNcolumn in the table!
You need to add such into the TFM_INS table in order to insert process parameters. See example in my table above.  Blush
I have just tried to add the attribute again , but now with the "PRC_CAPACT" attribute, rather CAP2ACT, but Veda is still not reading it. The PSET_PN is there (excel column AO) . Is this not the correct way of doing so?

It is really strange, the only attribute that Veda is actually reading is the "CAP_BND", defined in the "BI_2010" sheet. Maybe, should I use the "~TRADE_Param:" tag?

Forum Jump:

Users browsing this thread: 1 Guest(s)