CHP with two electricity outputs

[janis]

I would like describe in the TIMES/VEDA combined heat and power plants (CHP) with two electricity outputs. CHP is modelled as a flexible pass-out turbine system by specifiing NCAP_CHPR and -1<NCAP_CEH<0. If it is possible then what steps should be followed to specify two elc outputs?

[Antti-L]

> I would like describe in the TIMES/VEDA combined heat and power plants (CHP) with two electricity outputs.

To understand the nature of the modeling issue, could you explain the purpose of having two electricity outputs?

[Sorry, I had to edit your post to fix the html issue in it with the < characters not being "escaped".]

[janis]

This technology represents multiple small-scale CHP stations, some of which receive subsidies. I would like to allocate the subsidy to one of the electricity outputs. The first output is fed into the low-voltage grid, while the second is directed to a dummy subsidized grid, which is also connected to the same low-voltage grid.
Just to tell that NCAP_CHPR is negativ.

[Antti-L]

> I would like describe in the TIMES/VEDA combined heat and power plants (CHP) with two electricity outputs.
> If it is possible then what steps should be followed to specify two elc outputs?

For making use of the built-in CHP attributes, it looks like it would only be possible if the process has a single heat output, and then by utilizing the reduction mechanism for this purpose. In that case it looks like it can be done pretty cleanly, simply by adding a dummy output flow (of type NRG, and LimType N).  Let's call this commodity DUMNRG. You could then just define PRC_FOFF(reg,prc,'DUMNRG','ANNUAL','BOH','EOH'), to disable this dummy flow. If it is not disabled, it would allow the reduction operation. But as far as I can see, when disabling the dummy flow, the desired two-electricity CHP would be represented quite cleanly this way (without reduction).

[janis]

Thank you, Anti, for your response!!! I have to admit that I didn’t fully understand it, especially where the second electricity output comes into play. So, I’d like to explain my approach what I did before I asked Q in the forum.

By default, we have a CHP plant with the following specifications: a heat output (LTHRES) and an electricity output (LVGREL), with the calibrated years 2020–2022.

Now, we’d like to add a second electricity output, which would include both LVGREL and the new output, LVGBIOSLD. To calibrate the new electricity output, we created a new group, ELC_EBIOSLCETS. However, this approach leads to different results. Specifically, in the years when LVGBIOSLD is produced, we observe a different heat output. Please see bellow.

Anti, could you please elaborate feather on your proposed approach by imlementing reduction mechanism?

[Antti-L]

I am sorry but I could not see the process topology in your pictures, and so I don't know what you have in the topology and what is the PCG. And I cannot see whether LVGREL and LVGBIOSLD are indeed ELC.

I was suggesting to do what I said in my post: Add a new dummy output flow (of type NRG, and LimType N).  It should be automatically in the PCG as well, if your PCG is the default PCG or NRGO. And then just define PRC_FOFF(reg,prc,'DUMNRG','ANNUAL','BOH','EOH'), to disable this dummy flow.  The two electricity outputs must, of course, also be defined of type ELC.

> Now, we’d like to add a second electricity output, which would include both LVGREL and the new output, LVGBIOSLD. To calibrate the new electricity output, we created a new group, ELC_EBIOSLCETS. However, this approach leads to different results.

Yes, I already explained that you cannot make a CHP process with two electricity outputs work correctly with the TIMES built-in CHP attributes, unless you use the reduction mechanism, as I described (and as described in the documentation).  Also, I am not fully able to see the purpose of the new group ELC_EBIOSLCETS, with FLO_SHAR(UP)=1 for the members.

[janis]

I am sorry but I could not see the process topology in your pictures, and so I don't know what you have in the topology and what is the PCG. And I cannot see whether LVGREL and LVGBIOSLD are indeed ELC.

JR. Please find attached xls file where extracted information abot CHP plant is compiled. I hope that the process topology will be seen.


I was suggesting to do what I said in my post: Add a new dummy output flow (of type NRG, and LimType N).  It should be automatically in the PCG as well, if your PCG is the default PCG or NRGO. And then just define PRC_FOFF(reg,prc,'DUMNRG','ANNUAL','BOH','EOH'), to disable this dummy flow.  The two electricity outputs must, of course, also be defined of type ELC.

JR. In xls a new dummy output flow is added too.

> Now, we’d like to add a second electricity output, which would include both LVGREL and the new output, LVGBIOSLD. To calibrate the new electricity output, we created a new group, ELC_EBIOSLCETS. However, this approach leads to different results.

Yes, I already explained that you cannot make a CHP process with two electricity outputs work correctly with the TIMES built-in CHP attributes, unless you use the reduction mechanism, as I described (and as described in the documentation).  Also, I am not fully able to see the purpose of the new group ELC_EBIOSLCETS, with FLO_SHAR(UP)=1 for the members.

JR. Probablly it is wrong, but ELC_EBIOSLCETS group was added in order to calibrate electricity production for years 2020-2022 as well as bound UP in future and use FLO_SHAR(UP)=1 in subsidy scenario to activate it.

Maybe we can use attach file to clarify the missing information and discard the redundant information?


Thank you very much for your help!!!

[Antti-L]

Ok, thanks for showing some of the process characteristics.  I see you are still using MARKAL notation, which caused some confusion to me.

Anyway, you had now even four electricity commodities in the process, which was unexpected.  So, I removed two of them, because you earlier talked about having two electricity outputs.  Specifically, the dummy DUMNRG should, of course, not be an electricity commodity. 

You also had very tight calibration parameters (fixed capacity, fixed flows), which may easily cause problems, and so I removed some of them (you are free to add them back later, but I suggest to at least test first without too many fixed bounds).

There is no need to define the group as a commodity, and so I removed that from FI_Comm.  The process should, however be on the DAYNITE level so I added DAYNITE as the Tslvl for it.

[janis]

Thank you, Anti, for your response! I ran some tests yesterday, and it worked perfectly—I was able to get consistent results for the CHP system with both two electricity outputs and a single output. Fantastic!

The main issue was that I had incorrectly specified DUMNRG as an electricity commodity. Also, it turns out there’s no need to define the group (ELC_EBIOSLCETS) as a commodity.

Could you please let me know if there are any other potential challenges (or 'underwater stones') in this setup that I might face in the future? For example, I’ve just realized I need to check which technologies are involved in the Peak equations.

[Antti-L]

Hmm... with respect to using the reduction mechanism, no, I am not able to see any challenges.  But concerning peak equations, you may of course have a problem due to having those two electricity commodities, unless you have combined the peaking equations to include both of them (and maybe others as well), or you are combining the two into another commodity that has the peaking equations. In the latter case your CHP processes would not be able to contribute to the peak by capacity, but only by their production level.