Veda2.0 Released!


UC_Growth seed value violated
#16
I apologize, I think I posted an older file that I was messing around with, trying to get it to work, shame on me. I do want to implement the t,t-1 type constraint. I have actually been working with UC_CAP~RHS, and seeing the same results as seen in the previous screenshots. I deleted the UC_CAP~0 column based on what you said. Hopefully, LIMTYPE=UP makes sense in this case...based on this post http://forum.kanors-emr.org/showthread.php?tid=801
Reply
#17
The new file does not make much difference. You are still choosing the (T, T+1) variant for the dynamic growth constraints, as I have tried to explain.  As mentioned several times earlier, you should define GROWTH on the RHS side if you would like to use the (T, T−1) variant, but it seems you don't want that. You are defining it on the LHS side (=default), and so your UC_CAP~LHS thus represents the growth value.  Now you are just setting all the growth coefficients to 1.  Confused  

So, your LimType=UP does not constrain growth (earlier you used LO, which made sense), and no violation of the constraints occurs with your results. You could again take a look at the earlier explanation:  http://forum.kanors-emr.org/showthread.php?tid=801
Reply
#18
I'm sorry, I'm totally lost.
So that post says when I have Growth on the right, I'll have the t,t-1 formulation, and the LimType=UP.
This is with the Growth on the left.
  VAR_CAP(t) × Growth(t) ≥ VAR_CAP(t+1) − 1
Is this is what you mean by Growth on the right?
  VAR_CAP(t) ≥ VAR_CAP(t+1) / Growth(t) − 1/Growth(t)?
Or do you mean
-VAR_CAP(t+1) ≥ -VAR_CAP(t) x Growth(t) − 1
VAR_CAP(t+1) <= VAR_CAP(t) x Growth(t) + 1

?
Reply
#19
Could you please edit a row in the UC_Growth file to help me understand the correct way to do this?
Reply
#20
The last file that I uploaded seems to match what's here http://forum.kanors-emr.org/showthread.php?tid=645&highlight=t-1+growth
Reply
#21
Quote:The last file that I uploaded seems to match what's here http://forum.kanors-emr.org/showthread.p...t-1+growth

No, that is not at all true, it does not match at all what is there! This is what I posted there, to illustrate using the (t, t−1) variant:



You are doing something quite different, and using (t, t+1). In your latest file, you are defining constraints like this:

   VAR_CAP(t) × UC_CAP(t,LHS)   ≤   VAR_CAP(t+1) × UC_CAP(t+1,RHS) + UC_RHSRTS(t,'UP')

You should immediately see that these kind of constraints will not be working as intended (they are NOT constraining the capacity growth).

In summary, you should use LimType=LO when you use (t, t+1) variant (and positive coefficients). For example, in the following LimType should be LO and the Growth rate is on the LHS side (where normally Growth(t) > 1):

   VAR_CAP(t) × Growth(t)   ≥   VAR_CAP(t+1) − 1

You can use LimType=UP if you use the (t, t−1) variant and the Growth rate is on the RHS side, but thus far you have been rejecting that suggestion:

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


Forum Jump:


Users browsing this thread: 1 Guest(s)