investment reduction in group of technologies

[shivika]

Hi All,

I m trying to reduce investment in the group of technologies compared to previous run. But in some regions, this constraint is not working. I m not sure why this constraint is not working.

[Antti-L]

> I m trying to reduce investment in the group of technologies compared to previous run.

What exactly do you mean by "investment"?  You are using the INVCOST modifier, which applies the annuity multiplier, and so you are now limiting the annualized investment costs. That will make the constraint to be working with cumulative sums up to each period (the annual costs are thus summed over all NCAP vintages still available in t, giving the total annualized costs for them in t, over all vintages).

> But in some regions, this constraint is not working.

What exactly do you mean by "not working"?  How did you verify it is not working for some, but working for some others?  As far as I can see, the constraint is identical for all regions, and therefore I think it can only work for all regions or not work for any.

Anyway, I see your process filter is different for 2070 that for other years, which looks strange.

And the Sheet title says "Max lump-sum investment recovery subsidies by period", which also sounds strange, because you are neither constraining lump-sum investments nor subsidies on that sheet. Does that imply that "lump-sum investment recovery subsidies" in fact means "annualized investment costs"?

[shivika]

Sorry sheet is label wrongly. I want to reduce the annualized investment in the low carbon technologies to showcase the climate change impact using damage curve. In some regions maximum annualized investment is higher that is set in this sheet for these technologies.

[Antti-L]

> In some regions maximum annualized investment is higher that is set in this sheet for these technologies.

Ok, I assume that you compare the sum of the reported annual costs to the bounds you have defined, and you see some bounds being exceeded, is that correct?

There are several possible reasons why the sum of the reported annual costs might not be fully consistent with the bounds defined by INVCOST:
  ● As mentioned, your process filter for the bound in 2070 appears to be inconsistent
  ● The INVCOST modifier only sums over new capacity vintages, while reported costs include also past investments
  ● If using ANNCOST LEV, reported annualized costs are inconsistent with the INVCOST modifier, which uses standard cost annualizing;
  ● There may also be some other small, cosmetic differences between INVCOST and Cost_Inv reporting

Maybe your differences are related to some of the above?  It might be best that you would calculate the bounds from the non-binding equation levels instead of reported annual costs.

For curiosity, I just tested with the INVCOST modifier myself, and I got a perfect match between the UC equation levels and the reported Cost_Inv. But I did not have any past investments for the processes, nor was I using ANNCOST LEV.

I guess it would be possible to make the INVCOST modifier optionally consistent also with ANNCOST LEV, and so I will look if that option can be implemented.