I am struggling to make one of UCs in my model work. I am trying to restrict the model import of electricity by level of 10 % of all the demands for electricity together. Please see the UC in Excel attached.
Following the notation in the file, as I understand it, I get sum of all the imports of ELCHIG from all the import processes minus 0,1 times all the electricity flows (ELCHIGG + ELCLOW + ELCMED) that go in the corresponding processes should be max zero. What is interesting, in the results it seems that the model successfully computes that negative part meaning that I see 0,1 times the demand for electicity as level of UC in the results, but it does not account the positive part of the import...
Could you please suggest me, what could be possibly wrong with the construction of the UC?
18-05-2023, 09:07 PM (This post was last modified: 18-05-2023, 09:58 PM by Antti-L.)
UC_FLO is ignored by TIMES for the trade flows of IRE processes. That is because 1) IRE processes, in general, may have IMP and EXP flows and so UC_FLO would be ambiguous, and 2) UC_FLO can still be used for auxiliary capacity-related flows of IRE processes. Therefore, even if in principle one might try and convert UC_FLO into UC_IRE in cases where only an IMP or EXP flow is present and there is no capacity-related flow, doing so would nonetheless become prone to error.
I suspect that the IMPELCHIG* processes are IRE processes? If so, the UC_FLO defined for them will not work. You would need to use UC_IRE instead.
I attached the screenshot of the UC. But I am afraid that no valuable information could be seen from it because there are too many processes affected by the UC and my screen simply is not big enough. But it shows one thing, these IMPELCHIG* processes seem to be read properly by the model... Is there any specific part of the UC that I should show?
Thank you for the idea, Antti, but I am afraid that it is not the case. These processes are of the NRG type. Does this match with the UC_FLO fine?
I tried to switch for the UC_IRE constraint for the IMPELCHIG* processes and it did not work. So there might be a flaw in the definition of these processes or something. But then, I realized I can restrict just the activity of the processes by UC_ACT and now it seems to work fine. Sometimes the easiest way forward is also the finest.
Yes, I know, UC_IRE is somewhat more difficult to use, as it requires the IMP/EXP indicator as well, which you probably missed. Using UC_ACT for the same purpose is fine in the special case where the process has only a single trade flow. Good that you found it fitting your need.
(18-05-2023, 09:07 PM)Antti-L Wrote: UC_FLO is ignored by TIMES for the trade flows of IRE processes. That is because 1) IRE processes, in general, may have IMP and EXP flows and so UC_FLO would be ambiguous, and 2) UC_FLO can still be used for auxiliary capacity-related flows of IRE processes. Therefore, even if in principle one might try and convert UC_FLO into UC_IRE in cases where only an IMP or EXP flow is present and there is no capacity-related flow, doing so would nonetheless become prone to error.
I suspect that the IMPELCHIG* processes are IRE processes? If so, the UC_FLO defined for them will not work. You would need to use UC_IRE instead.
(18-05-2023, 09:07 PM)Antti-L Wrote: UC_FLO is ignored by TIMES for the trade flows of IRE processes. That is because 1) IRE processes, in general, may have IMP and EXP flows and so UC_FLO would be ambiguous, and 2) UC_FLO can still be used for auxiliary capacity-related flows of IRE processes. Therefore, even if in principle one might try and convert UC_FLO into UC_IRE in cases where only an IMP or EXP flow is present and there is no capacity-related flow, doing so would nonetheless become prone to error.
I suspect that the IMPELCHIG* processes are IRE processes? If so, the UC_FLO defined for them will not work. You would need to use UC_IRE instead.
I am working on the EPA US 9 Region VEDA2.0 and struggling to make one of UCs in my model work: Scen_UC_TRNLDV20 for Net Zero by 2050 scenario. I am trying to restrict the model for low bound for electrical LDV by level of 10 %. I tried the following: replaced UC_FLO on UC_IRE and UC_ACT – there are no difference and IMPNRGZ are identical (UC_S_TLELC values are the same in 2020-2055). Please see the UC.
07-06-2023, 04:48 PM (This post was last modified: 07-06-2023, 04:51 PM by Antti-L.)
>I am trying to restrict the model for low bound for electrical LDV by level of 10 %. I tried the following: replaced UC_FLO on UC_IRE and UC_ACT – there are no difference and IMPNRGZ are identical (UC_S_TLELC values are the same in 2020-2055).
You cannot use UC_IRE for normal processes, only for IRE processes (trade processes). I suppose the TLE* processes are not IRE processes, and if so, using UC_FLO is fine (and UC_IRE cannot be used), unless they happen to be storage processes?
I am not able to see any clear issues in the six UCs shown in your picture, except that you seem to be defining them even for historical years, which I think should usually not be needed (and might easily cause infeasibilities) if you have calibrated your model to statistics. However, I can see that by 2030 the constraints will have no effect (the market share is only bounded to be ≥ 0). Does that mean that these constraints are actually meant for calibrating the model to statistics?
Can you describe what the problem is that you are seeing?
• Do these constraints have marginals reported (User_conFXM)?
• Do these constraints have any levels reported (User_con)?
• Do these constraints have dummy import flows?
