Posts: 26
Threads: 7
Likes Received: 0 in 0 posts
Likes Given: 0
Joined: Nov 2014
Hi,
I am looking for some help in troubleshooting an issue I am having with the gams-solution gdx-file generated after a TIMES solution. When I look at the VAR_CAP parameter within the gdx-file, it does not include capacity-information for all active TIMES processes. This is strange since the information is available in the VEDA output (VD-file) from the same scenario. Any help regarding this issue would be much appreciated?
I would also be delighted if someone could point me to the file in the source code that generates information on VAR_CAP and writes it to the VD-file.
An observation: It seems that the processes that are not included in the gdx (VAR_CAP) are processes that have been specified in the VT-files using STOCK~region. In contrast, the VT-file-processes where the region is specified in an independent column appears to be present in the GDX -file with var_cap-information.
Best regards,
Kristoffer
Kristoffer S. Andersen
<br />Advisor, Danish Energy Agency
Posts: 1,947
Threads: 26
Likes Received: 54 in 46 posts
Likes Given: 15
Joined: Jun 2010
18-03-2021, 08:03 PM
(This post was last modified: 18-03-2021, 08:46 PM by Antti-L.)
Are you sure?
1) There is no such parameter as "STOCK" in TIMES; the parameter is called PRC_RESID.
2) The existing capacities defined by using PRC_RESID are reported in the PAR_PASTI parameter, and are subsequently imported into VEDA-BE or VEDA2 results in the VAR_Cap parameter. So they are absolutely both in the VD file and the GDX file and therefore I suspect that you are mistaken. At least I can always see them correctly included in VEDA-BE and VEDA2 VAR_Cap results.
See the vdd file entries:
* VEDA Attr GAMS - indexes -
VAR_Cap par_capl r t p
VAR_Cap par_pasti r t p v
As to the model variables VAR_CAP(r,t,p), they are only present in the model when needed. But most users want all capacities reported, whether they are existing or new, or having a VAR_CAP variable or not, and therefore the full capacity is always reported into VEDA in VAR_Cap.
Posts: 26
Threads: 7
Likes Received: 0 in 0 posts
Likes Given: 0
Joined: Nov 2014
(18-03-2021, 08:03 PM)Dear Antti, Wrote: Thank you a ton for taking the time to help me :-D
1) Thanks for the clarification on the STOCK (attribute used in VEDA) and PRC_RESID (the parameter in the GDX-file).
2) I was able to extract the information I need from par_capl(r,t,p):"new capacity" and par_pasti(r,t,p,"0"):"residual capacity".
Best regards,
Kristoffer
N.B. The quest to extract data directly from the GDX-file reflects that I am updating the tool we use at the Danish Energy Agency to: 1) exchange information between a TIMES model and a CGE model; and 2) report results from a TIMES solution. The ultimate goal is to collect all relevant output from a TIME model in one parameter with 12ish dimensions. I will be happy to share and discuss this work with anyone interested, and at some point, provide the gams-code here on the VEDA-forum.
Are you sure?
1) There is no such parameter as "STOCK" in TIMES; the parameter is called PRC_RESID.
2) The existing capacities defined by using PRC_RESID are reported in the PAR_PASTI parameter, and are subsequently imported into VEDA-BE or VEDA2 results in the VAR_Cap parameter. So they are absolutely both in the VD file and the GDX file and therefore I suspect that you are mistaken. At least I can always see them correctly included in VEDA-BE and VEDA2 VAR_Cap results.
See the vdd file entries:
* VEDA Attr GAMS - indexes -
VAR_Cap par_capl r t p
VAR_Cap par_pasti r t p v
As to the model variables VAR_CAP(r,t,p), they are only present in the model when needed. But most users want all capacities reported, whether they are existing or new, or having a VAR_CAP variable or not, and therefore the full capacity is always reported into VEDA in VAR_Cap.
Kristoffer S. Andersen
<br />Advisor, Danish Energy Agency
Posts: 1,947
Threads: 26
Likes Received: 54 in 46 posts
Likes Given: 15
Joined: Jun 2010
19-03-2021, 02:32 PM
(This post was last modified: 20-03-2021, 06:13 AM by Antti-L.
Edit Reason: improved wording
)
Thank you for the additional info.
If you want to collect all relevant output from a TIMES model, then I think the vdd file (times2veda.vdd) can best give you the overview of all the TIMES results parameters (stored in the GDX file), from which the VD file results are taken. The VD file is likewise an attempt to present all relevant output from a TIMES model.
And the documentation, Part II, gives some more info on the report parameters in Section 3.3 Report parameters.