user constraint

[nukelucas]

Fellow Friends of Veda, 

I am having an issue setting a user constraint for the Demo03. As an exercise, I am trying to set a user constraint on the penetration of fossil fuel plants in 2020 (last milestone year, start year is 2005). This is the table I created, as I want the sum of the total capacity of these plants to be less or equal than 1 GW: 


I will explain some of the Table's features to give some context: 

• Powerplant nomenclature: ELCTNCOA00 and ELCTECOA00 are respectively new and existing coal powerplants (same is valid for gas and oil). 
  


• In Demo03, there is only one region (REG1);
  

Nevertheless, as you can see from the screenshot below (Var-Cap in the Results Window), it does not seem to working correctly as the only difference between Demos_003 (normal Demo003) without and Demos_003_ucc (Demo003 with the user constraint) is in the installed capacity (can't explain why). 

Here is also a view of the UC from the Item List window: 


Can anyone point out what am I doing wrong? 

Thanks in advance for the help!
Lucas

[nukelucas]

Please note that I also tried changing the interpolation rule, without success.

[Antti-L]

Welcome to the Forum.

As you have marked your post with the “Bug” indicator, which gives the impression that you are reporting a bug somewhere, as a long-time TIMES maintainer I felt that I need to look at your issue with respect to TIMES, although you provided only limited information about what you actually have in your model.

However, I was not able to reproduce your results in either case:  Neither with the original Demos_003 model, nor by adding the constraint that you showed.  So, it seems that you have modified even the original Demos_003 model database, but did not provide the modified test model here.  Anyway, concerning your constraint, I have the following comments:

  • The constraint attempts to limit the capacities of both existing and new coal, gas, and oil fuelled power plants to 1 GW in 2020, while the existing capacities of those plants amount to 100.4 GW still remaining in 2020 (according to the original DemoS_003 data).  Therefore, your constraint would not make any sense, and would lead to an infeasible model, unless you would also enable early retirements for those existing capacities. Without the explicit retirement option, existing power plant installations would of course not be wiped out just by adding a user constraint. But you did not clarify in your post whether you had enabled early retirements or not. 

  • If you have enabled dummy imports for user constraints, the infeasibility of your constraint would not be directly visible to you, but would be hidden behind the dummy import levels.  But you did not clarify in your post whether you had dummy imports enabled for user constraints or not.

Anyway, I think I was able to fully verify that there is no bug related to that TIMES constraint: The constraint worked correctly and fully as expected both with and without enabling early retirements for the existing capacities. When enabled, the capacities were limited to 1 GW in 2020, as modeled, and when not enabled, the model was infeasible, just as expected. Therefore, my conclusion is that the “Bug” label is not related to TIMES.  It remains unclear to me what that “Bug” label was meant to be referring to, if anything.

[MakRacz]

Hello Lucas,



In case you don't know how to disable dummy imports in Veda, you can do this under Tools -> User Options. Uptade the options and then synchronize the scenario with your user constraint again.









Then in Items Detail those entries with IMPNRGZ should disappear.


If the model run after these changes returns an infeasible solution, now you know why.

[nukelucas]

Dear Antti-L and Maciej,

Thank you both for your replies. I apologize for mistakenly marking my thread as a "bug"; I'm new to the Veda Forum and did so unintentionally.
I will try your suggestions regarding the Early Retirements and UC Dummy Imports and get back to you if further questions arise.

Thanks again for your help.

Best regards,
Lucas