Fixed and flexible inputs combined

[Antti-L]

Dear Antti,

After updating my VEDA, I indeed have the correct equation when using INPUT and the model is working fine.

I am very glad that you find it working correctly now.  I have just learned that in the TIAM model the same approach has been used as you have: The TIAM model defines some user-defined groups in the topology of refinery processes, and defines an Input attribute for those groups!  And when running the TIAM model you get domain violation errors for those nonsensical topology entries, just like in your example.

I guess the TIAM users should be made aware of that with respect to these groups, the Input attribute has been used in a non-standard way in the model, and that the model will behave differently with the latest TIMES version (now consistently for all Inputs).

[wnijs]

I face a new problem when trying to let all ELE processes consume two new materials.

It has to be flexible so that material 1 or material 2 can be used.

I defined the materials (with also a  mining process), I used TOPINS to include the new topology and I defined a User Defined CommodityGroup (FloMat) and used ACT_EFF as below. However, no EQE_Acteff is constructed ?

The only difference I see is that I defined ACT_EFF in an INS tabel (but I guess that is not a problem) and that the topology of two new input materials are being set by TOPINS.

Attribute	Year	Cset_gSet	Cset_CN	Cset_CD	Pset_Set	Pset_PN
ACT_EFF			FloMat			EU*

Is there an alternative way ?

Thanks,
Wouter

[AKanudia]

Lets do a GTM in about 4 hours from now.

[wnijs]

The problem is solved. The commodity group definition was not read because of Excel and so it did not have two commodities. ACT_EFF is working fine now.

Thanks to GTM with Amit, I (re)discovered the "CommGrp Master" where you can see what commodities are included in a Commodity Group.

[STM]

Hi,

I follow up on this post with a related question.

I would like to model multiple inputs - multiple outputs processes. The flow shares of each output and the efficiency of the process are different with regard to the input. For example, if I have inputs i1 and i2, and products p1 and p2:

1. Efficiencies are different, so that producing on unit of (p1+p2) requires more of e.g. the input i2 than input i1.
2. Flo_Shares are different, so that consuming i1 will produce up to e.g. 70% of p1 and 40% of p2, while consuming i2 will produce up to e.g. 40% of p1 and 70% of p2.

I do not know if there is any workaround, other than making two different processes consuming only i1 or i2. In that case, economic characteristics and constraints should be set globally for the two processes, since in the real life, those are one single process.

I started a google spreadsheet behind Wouter's: http://spreadsheets.google.com/ccc?key=0Agw7MNXBx500dFd1Qjc4YWkwUEZhS0JoUXhKZFhUOUE&hl=en&authkey=CNPFnPYL#gid=2

Thanks,

Stephane

[Antti-L]

This should be rather easy, as you have different commodity names for the outputs. See if my suggestion might be of help.

If desired, you can then easily combine the OILGSL1+OILGSL2 and OILDSL1+OILDSL2 commodities together, by using COM_AGG.

[STM]

Your suggestion is good, thanks Antti.

[AKanudia]

Actually, it is a little better than good. More on the brilliant side

I was not aware that c did not have to be a part of the cg in FLO_SHAR!! I guess they still have to be on the same side. But do they have to be of the same type (NRG, MAT etc.)?

[Antti-L]

Thanks Stephane and Amit.

Most of the FLO_SHAR generalizations have been implemented only in the VDA extension, and are briefly described in the documentation:

See: http://www.iea-etsap.org/docs/TIMES-3.1-Information-Note.pdf, page 6.

The c and cg thus need not be on the same side, and neither do they need to be of same type.

But without the VDA extension (i.e. when using standard TIMES), the restrictions of the original design apply:  c and cg have to be on the same side and c must be a member of cg.