start year violation and domain violation

[cyang]

Hi,
We are having a couple of different issues that I’m hoping someone can help with:

1. We are seeing start year violations where we specify a start year for new technologies that is after our base year (2011 vs base year 2010) and the model still invests in these new technologies in 2010.  We have enough specified base year stock to produce the required demand, so that is not the issue.  We have had to resort to putting NCAP_BND constraints =0 for 2010 for some new technologies to squash this, but I was wondering if anyone has experienced this issue before and how we can address it.

2. The model is able to run but when we look at the LST file, there are a bunch of “domain violations” $170.  We are not sure what this means and are not sure if it is some indication that something we specified is not being correctly read into the model.

One potentially important thing to note is that we are not using VEDA for our full model.  We’ve expanded the model pretty significantly so VEDA tends to choke on some versions of the model but we are using the .dd files that VEDA generates for a more simplified version of the model and then adding analogous technologies and specifications for our expanded model.  So some of the problems could be related to our work in this area.

thanks!

Chris

[Antti-L]

Thanks for the report about start year violations. That may indicate a possible bug somewhere. However, I am afraid that without seeing a reproducible case, it is difficult to guess what might be the cause for such an issue.
 
Therefore, I would also be interested to know whether anyone has experienced this issue, in a reproducible way. If anyone can reproduce the problem with some test model (for example, by testing the same kind of periods and start years with the DEMO model), please provide the input files for the model (*.DD files + the RUN file). By having a reproducible case, and I am sure I can immediately identify the reason for the problem, as well as suggest a way around it and/or fix the bug if the problem is in TIMES.
 
Concerning the second issue (domain violations), they are caused by some processes, commodities or commodity groups not being defined in the model, even though they are used in some attributes. Those attributes will then be ignored by TIMES. That can be in some cases serious (if phantom topology entries are getting created), but often it is quite harmless, if they are just related to some scenario which is not used in the current run. Again, it is difficult to say about your case without seeing the model. But please check the QA_CHECK.log for any error messages about phantom entries in the topology. If such are reported, you should try and fix the errors.

[Antti-L]

After reading the recent post by Saleh, it occurred to me that maybe you are using an old TIMES version (e.g. v3.4.3).

That would be able to explain the start year violations, because NCAP_START was introduced in TIMES v3.4.5 (in year 2013).

If this is the case, could you please inform me of the reason for not using the latest TIMES version, but such an old version?