Unnormal results

[[email protected]]

> INSTCAP is same as the VAR_NCAP and LUMPINV is the product of VAR_NCAP and Investment Cost (10, which was supplied by us)

Close, but not quite.  INSTCAP takes into account repeated investments within periods, if such occur. That might possibly explain some of the confusion (about VAR_Ncap being smaller than INSTCAP; I noticed above some INSTCAPs being mentioned as exactly 11×VAR_NCAP, which might suggest such taking place; but just guessing without seeing the model). And LUMPINV includes also the impacts of IDC and hurdle rate, if defined.

Ahh, but maybe the user just summed up INSTCAP and 10 × INSTCAP (LUMPINV)!  That would also explain it, but summing them up makes no sense.

Hi Antti,

Thank you so much for explaining that, so I will use the VAR_NCAP (rather than Cap_new) as the added capacity. 

Best,
Xiao

[[email protected]]

To help you understand the relationship between Cap_New, VAR_Ncap, and Cost_Inv better, I have attached a result table of tech AFM-AuxiliaryEquip_ELE. 

You can see that CAP_NEW contains two pieces of information - INSTCAP and LUMPINV, where INSTCAP is same as the VAR_NCAP and LUMPINV is the product of VAR_NCAP and Investment Cost (10, which was supplied by us).

Hi!

Thank you for addressing that!

However, where did you find this table (downloaded results in local veda?). Because when I tried to export the results about AFM-AuxiliaryEquip_ELE from VEDA-ONLINE, the results are different (P.S. the only difference with original version is that I changed the start-year as 2020 rather than 2005, because we abstract 2020's data as the base-scenario), as follows.

Region Attribute Commodity Process - 2020 2021 2025 2030 2035 2040 2045 2050
AL Cap_New - AFM-AuxiliaryEquip_ELE 0 0 2.673853308 2.673853308 2.673853308 2.673853308 2.673853308 2.673853308

Region Attribute Commodity Process - 2020 2021 2025 2030 2035 2040 2045 2050
AL Cost_Inv - AFM-AuxiliaryEquip_ELE 0.566388226 0.528629011 0.606134768 0.64588131 0.685627852 0.685627852 0.685627852 0.685627852

Well, I mean, why is there no Cap_New OR VAR_NCAP in 2020 or 2021 even with Cost_Inv?

Best,
Xiao

[Antti-L]

If you have defined investment costs for vintage v, then there are investment costs reported for vintage v in period t, whenever there is capacity of vintage v in period t. Therefore, I think you must have some capacity of vintage v in period t in your results.  That capacity does not have to be new (VAR_NCAP), it can be old capacity as well (endogenous or exogenous). So, it is perfectly normal to have investment costs reported for vintage v in period t, even if you have no VAR_NCAP in period t.

[[email protected]]

If you have defined investment costs for vintage v, then there are investment costs reported for vintage v in period t, whenever there is capacity of vintage v in period t. Therefore, I think you must have some capacity of vintage v in period t in your results.  That capacity does not have to be new (VAR_NCAP), it can be old capacity as well (endogenous or exogenous). So, it is perfectly normal to have investment costs reported for vintage v in period t, even if you have no VAR_NCAP in period t.

OMG.. Thank you for that..

So if I wanna extract the Inv_cost for VAR_NCAP (the new capacity), I just need to locate to the specific year with VAR_NCAP ?

Besides, is that the carbon trajectory not strictly meet the requirement we defined here normal? (2021's carbon emissions even higher than 2020, and both years higher than the limit we set).

Total CO2 emissions 1168447.196 1190536.842 521318.2658 419711.6573 291521.7024 276494.1503 156786.308 0
CO2 constraint target 645400 623886.6667 537833.3333 430266.6667 322700 215133.3333 107566.6667 0

Best,
Xiao

[Antti-L]

> So if I wanna extract the Inv_cost for VAR_NCAP (the new capacity), I just need to locate to the specific year with VAR_NCAP

As you know by reading the documentation, TIMES reports the investment costs in several report attributes:

  ● Cost_Inv(r,v,t,p) : Annualized undiscounted investment costs (caused by NCAP_COST) in period (t) spread over the economic lifetime (NCAP_ELIFE) of a process (p) with vintage period (v).
  ● Cap_New(r,v,t,p,'LUMPINV'): Lump-sum investment costs related to the new capacity of process (p) of vintage (v) commissioned in period (t).
  ● Cost_NPV(r,p,'INV') : Total present value of the investment costs in the base year, by process (optional, activate by setting RPT_OPT('OBJ','1')=1):

I don't know what you mean by "if I wanna extract the Inv_cost for VAR_NCAP ".  All of those three attributes report the investment costs for VAR_NCAP(r,v,p) (where for convenience, v=t), and both Cost_Inv(r,v,t,p) and Cap_New(r,v,t,p,'LUMPINV') have the vintage index v, which directly gives you the costs attributable VAR_NCAP(r,v,p).

Concerning your user constraint for limiting carbon emissions, I cannot see any ~UC_Sets: tags specified in your post:
https://forum.kanors-emr.org/showthread.php?tid=1345
So, I find it somewhat difficult to comment on it, as I don't know what exactly you have defined, and I don't know what the results numbers you showed represent. For example, I don't know whether the constraint is actually meant to be by region or sum over regions, or which commodities (*CO2) are included in the constraint, and I don't know whether the result numbers you show for SUM-CO2 are for some region or a sum over all regions, or which commodities (*CO2) are included in those results, or from which result attribute they have been taken, and so on. It seems that the results numbers you have posted are not copied directly from a VEDA Results table, or at least I cannot see the attribute name and the attribute dimensions appropriately shown. But if you can post the *.VD file for the case, it might provide sufficient information to comment.

[[email protected]]

> So if I wanna extract the Inv_cost for VAR_NCAP (the new capacity), I just need to locate to the specific year with VAR_NCAP

As you know by reading the documentation, TIMES reports the investment costs in several report attributes:

  ● Cost_Inv(r,v,t,p) : Annualized undiscounted investment costs (caused by NCAP_COST) in period (t) spread over the economic lifetime (NCAP_ELIFE) of a process (p) with vintage period (v).
  ● Cap_New(r,v,t,p,'LUMPINV'): Lump-sum investment costs related to the new capacity of process (p) of vintage (v) commissioned in period (t).
  ● Cost_NPV(r,p,'INV') : Total present value of the investment costs in the base year, by process (optional, activate by setting RPT_OPT('OBJ','1')=1):

I don't know what you mean by "if I wanna extract the Inv_cost for VAR_NCAP ".  All of those three attributes report the investment costs for VAR_NCAP(r,v,p) (where for convenience, v=t), and both Cost_Inv(r,v,t,p) and Cap_New(r,v,t,p,'LUMPINV') have the vintage index v, which directly gives you the costs attributable VAR_NCAP(r,v,p).

Concerning you user constraint for limiting carbon emissions, I cannot see any ~UC_Sets: tags specified in your post:
https://forum.kanors-emr.org/showthread.php?tid=1345
So, I find it somewhat difficult to comment on it, as I don't know what exactly you have defined, and I don't know what the results numbers you showed represent. For example, I don't know whether the constraint is actually meant to be by region or sum over regions, or which commodities (*CO2) are included in the constraint, and I don't know whether the result numbers you show for SUM-CO2 are for some region or a sum over all regions, or which commodities (*CO2) are included in those results, or from which result attribute they have been taken, and so on. It seems that the results numbers you have posted are not copied directly from a VEDA Results table, or at least I cannot see the attribute name and the attribute dimensions appropriately shown. But if you can post the *.VD file for the case, it might provide sufficient information to comment.

Hi Antti!

Thank you! So I have attached it here.

I only found .vdt rather than .vd in VEDAONLINE. Would you like to give some advice?


Best,
Xiao

[Antti-L]

Yes, I meant the *.VD file, not .vdt. TIMES always produces that file, also under VO.

[[email protected]]

Yes, I meant the *.VD file, not .vdt.  TIMES always produces that file, also under VO.

Oops! So I have re-checked and found no .vd definitely, does that mean this model may have other potential problem for that?

I have attahched .lst and quick-chek log here.

Best,
Xiao

[[email protected]]

Yes, I meant the *.VD file, not .vdt.  TIMES always produces that file, also under VO.

Oops! So I have re-checked and found no .vd definitely, does that mean this model may have other potential problem for that?

I have attahched .lst and quick-chek log here.

Best,
Xiao

[Antti-L]

Ok, maybe VO does not give you access to the *.VD file, but the GDX file is at least provided (I just checked on VO, and yes, it is provided).  The VD file can be produced from the full GDX file.  So, can you provide the full GDX file instead?  A download link for it would be fine, if it is too big to attach here.

[Antti-L]

The only user constraint in the model is AU_CO2_BND, which limits the VAR_COMNET of TOTCO2 by region.  For each and every region, the limit appears to be 6000 in 2050, and the results show that the constraint is working as expected. So, for example in the AL, AT, BC and SA regions the TOTCO2 emissions are indeed 6000 in 2050, and in other regions it is even below the limit.

If these regional AU_CO2_BND constraints are not the ones you mean, it looks like you have failed to define the additional constraint(s) you have been talking about.  As I said, there are no other user constraints defined in that model.

[[email protected]]

The only user constraint in the model is AU_CO2_BND, which limits the VAR_COMNET of TOTCO2 by region.  For each and every region, the limit appears to be 6000 in 2050, and the results show that the constraint is working as expected. So, for example in the AL, AT, BC and SA regions the TOTCO2 emissions are indeed 6000 in 2050, and in other regions it is even below the limit.

If these regional AU_CO2_BND constraints are not the ones you mean, it looks like you have failed to define the additional constraint(s) you have been talking about.  As I said, there are no other user constraints defined in that model.

Hi Antti,

Yes and plus, I also constrained the 2020's emission as 645400 and the other years emission by interpolation rules ("5"), as follows. 
 
Cset_CN Attribute Year LimType UC_COMNET AllRegions   UC_RHSRTS~0       UC_Desc
*CO2                  2020 UP            1               645400           5                  CO2 Bound Constraint
*CO2                  2050 UP            1               6000

So, you mean, these constraints were limited for each region respectively rather than their sum ?? How should I limit the sum of all regions to 6000 then ?

Best,
Xiao

[Antti-L]

For a constraint over all regions you would need to use ~UC_Sets: R_S: AllRegions and UC_RHST instead of UC_RHSRTS.

[[email protected]]

For a constraint over all regions you would need to use ~UC_Sets: R_S: AllRegions and UC_RHST instead of UC_RHSRTS.

Hi Antti,

Thank you for the help! P.s. Do you know what it ("Delayed Process but PASTInvestment") stands for ? Do I need to adjust it?
              ******          TIMES -- VERSION 4.7.3          ******
              **************  QUALITY ASSURANCE LOG  **************

*** Delayed Process but PASTInvestment
*01 WARNING      - Delay is ignored:  R=AL          P=AFM-AuxiliaryEquip_COA
*01 WARNING      - Delay is ignored:  R=AL          P=AFM-AuxiliaryEquip_ELE
*01 WARNING      - Delay is ignored:  R=AL          P=AFM-AuxiliaryEquip_GAS
*01 WARNING      - Delay is ignored:  R=AL          P=ART-AuxiliaryEquip_ELE
*01 WARNING      - Delay is ignored:  R=AL          P=EDU-AuxiliaryEquip_COA
*01 WARNING      - Delay is ignored:  R=AL          P=EDU-AuxiliaryEquip_ELE
*01 WARNING      - Delay is ignored:  R=AL          P=EDU-AuxiliaryEquip_GAS
*01 WARNING      - Delay is ignored:  R=AL          P=HSS-AuxiliaryEquip_COA
*01 WARNING      - Delay is ignored:  R=AL          P=HSS-AuxiliaryEquip_ELE
*01 WARNING      - Delay is ignored:  R=AL          P=HSS-AuxiliaryEquip_GAS
*01 WARNING      - Delay is ignored:  R=AL          P=ICS-AuxiliaryEquip_ELE
*01 WARNING      - Delay is ignored:  R=AL          P=OS-AuxiliaryEquip_COA
*01 WARNING      - Delay is ignored:  R=AL          P=OS-AuxiliaryEquip_ELE
*01 WARNING      - Delay is ignored:  R=AL          P=OS-AuxiliaryEquip_GAS
*01 WARNING      - Delay is ignored:  R=AL          P=OTH-AuxiliaryEquip_ELE
*01 WARNING      - Delay is ignored:  R=AL          P=RTS-AuxiliaryEquip_COA
*01 WARNING      - Delay is ignored:  R=AL          P=RTS-AuxiliaryEquip_ELE
*01 WARNING      - Delay is ignored:  R=AL          P=RTS-AuxiliaryEquip_GAS
*01 WARNING      - Delay is ignored:  R=AL          P=WST-AuxiliaryEquip_ELE
*01 WARNING      - Delay is ignored:  R=AL          P=WST-AuxiliaryEquip_GAS

Best,
Xiao

[Antti-L]

Such a warning is given when you define NCAP_START(r,p) = year (meaning that process p becomes available in that year), but you nonetheless have some existing capacities defined.  In such a case, the delay of the process is ignored for the existing capacity, which is thus assumed available despite NCAP_START. The GDX file shows that you had defined NCAP_START=2035 for these processes, and therefore the warning.