Results of UC_Growth are not matching with given constarint

[Anjali]

Hello,
I am using the Scen_UC Growth user constraint for the capacity growth of bio-mass plants in my model.

Image of input file is:


And the result image is :



Can anyone please help me to troubleshoot the problem why capacity growth is higher than the given 1% limit in constraint file.
New capacity is given to be added after yr. 2019 in this case.

Thank You

[Antti-L]

Could you post a screenshot from the VEDA-FE UC Master for UC_GrowthOfBioTechs, Data Tab, with all dimensions expanded?
And to see what's inside T_BIO, could you also post the VAR_CAP results with all dimensions expanded?

[Anjali]

Hi,

Thank You for the response.

Please find the image of UC master and results of capacity.

Veda FE UC master  

VAR_CAP result 

Actually this happens for other technology also if I give UC_growth constraints. Maybe I am not able to understand the constraint correctly. Actually, in my model I wanted to limit the yearly growth of technologies according to their historical growth and tried to use this constraint. Please help me to correct where I am wrong or if there is any other way to do this.

[Antti-L]

The UC Master screenshot shows that you have defined no UC_CAP attributes.  Without them, the constraint of course cannot work, as they have only the RHS defined.  You should thus correct your constraint, by defining also the UC_CAP coefficients for the "T_BIO" technologies. Maybe you could try putting T_BIO into the Pset_Set column, assuming you have such a process set T_BIO defined?

[Anjali]

I hope I did the same as you said. I added T_BIO as Pset_set.
But still, the result is not correct.

Attached are updated constraints, UC master and result.

[Anjali]

Actually, T_BIO is the process set defined in VEDA BE that includes both existing and new technology of BIO plant. And I have linked the VEDA BE database with VEDA FE as defined in TIMES documentation IV.

[Antti-L]

I cannot see any clear evidence of it not being correct.  Could you explain what is in your view incorrect about it?

You have five regions, and the constraint is defined separately for each region.  To see better what is going on, please show the results with the all dimensions expanded, at least the region.  

Note that you are allowing a growth of 1% per year plus additionally 1 unit per period. For five regions that would allow 5 units plus 1% per year increase in the capacity in each period.  As far as I can see, the numbers in your results fit quite well with the constraints you have defined. In all periods the growth appears to be 5 units or less (after 2020, which is the first period T in your (T,T+1) constraints).

[Anjali]

Hi,
Thanks a lot for this explanation. It has almost resolved my problem.
Attached it the expanded result image.
I am sorry that I misunderstood the constraint and result both. I didn't know about the extra 1 unit per period addition.

It means this constraint will increase the total capacity (existing+new) in each region with 1% per year with an additional 1 unit per period increment. Value 1.01 under column UC_CAP defines 1% increament and value -1 under column UC_RHSRTS defines 1 unit per period addition. I hope this is right.


Only doubt that what is the meaning of value 1 under coloumn UC_CAP~RHS?

Also, a high capacity is added in year 2020. Won't this constraint be applicable to year 2020 also? Do I need to define the additional CAP limit for T_BIO in the year 2020 to restrict this high capacity addition or it can be controlled by this constraint only.

Thank you again.

[Antti-L]

Yes, you have understood quite correctly. You can of course define any values (any trajectory) for the additional capacity per period instead of the constant -1 (UC_RHSRTS).

Denoting by Growth(t) the growth multiplier (defined by UC_CAP(LHS)), you are defining a constraint with the following formulation (constraint type = LO):

  VAR_CAP(t) × Growth(t) ≥ VAR_CAP(t+1)×UC_CAP(RHS) + UC_RHSRTS(t)

As you can see, the value 1 under column UC_CAP~RHS is applied as the multiplier of VAR_CAP(t+1), and for a growth constraint (T,T+1) the value 1 is normally used, because you want to constraint the growth from VAR_CAP(t) to VAR_CAP(t+1).

Swapping the sides (and substituting a few values) we can write the constraint equivalently as follows:

  VAR_CAP(t+1) ≤ VAR_CAP(t)×Growth(t) + 1

As you can see, the constraint is as intended, where the +1 on the right hand side is now the opposite value of UC_RHSRTS.

The capacity in 2020 is not constrained by your equation, because you start it from 2020. If you want the constraint to be limiting also the 2020 capacity, you could just define the UC_RHSRTS already for the preceding period 2017, or even for 2015.

[Anjali]

Thanks a lot

[anshfr]

Hi, I'm having the same problem as above and I was wondering, can this constraint growth be made as an aggregate for all regions, so that the constant 1 used to initialize the capacity does not apply to every region?

Thank you.

[Antti-L]

Yes, of course. You get an aggregate for all regions by using UC_RHST, and the constant then applies to the aggregate.
(in VEDA, you may also need to make sure that R_E: is empty and R_S: is AllRegions)

[anshfr]

It works! Thank you so much