ACT_EFF and CHP configuration

[AdvisorDK]

Dear Forum,

I configured a CHP process (electricity-driven) using ACT_EFF for the PG commodity - as suggested in the VEDA documentation-, but after running the model I get the warning:

Found CHP processes with PG commodity efficiencies – unsupported

When I switch to the traditional EFF attribute, the warning disappears and results (annual production, system cost) look unchanged.

Besides, EFF does not seem to be an alias or abbreviation of ACT_EFF as referred in the forum, at least according to the TIMES and VEDA attribute lists.

Should I stick with ACT_EFF and ignore the warning (to avoid future inconsistencies when increasing model dimensions like time-slices, milestones, storages, or investments), or is it safe to just use EFF?

Any guidance would be appreciated.

[Antti-L]

Could you point me to the VEDA documentation suggesting that "using ACT_EFF for the PG commodity" would make sense or would even be recommended?  As far as I can tell, the warning states correctly that such is basically not supported.  The TIMES documentation describes how BP or pass-out turbine type of CHP processes should be modeled, and that does not involve using ACT_EFF for any PG commodity.

Anyway EFF is indeed a shortcut for ACT_EFF(ACT), i.e. an efficiency defined for the default shadow group, as described in the TIMES documentation.

You should definitely not stick to using ACT_EFF for any PG commodity, because the process would then hardly work as intended. Could you explain what is your intended purpose of using ACT_EFF for a PG commodity?

[AdvisorDK]

The core parameters for CHP modelling (including ACT_EFF) are listed in:
PART II Reference Manual
4. Usage Notes on Special types of processes
Table 17, Page 120

As you said, EFF might be fine, so I am using it.

But I am still having Dummy issues (IMPNRGZ) for heat, as the CHP - extraction by-pass - produces very little heat.
I have already fixed "FX" commodity LimType, and have "reasonable" import prices

2222 IMP*Z
8888 IMPDEMZ

What do you think could be the root of this issue?

I could not find a way to attach my VT-file to this conversation

[Antti-L]

Sorry, I only now looked at your Excel attachment, but it did not show much more, other than that it looks like your ACT_EFF was actually meant to be defined for the input commodities. And so, yes, EFF and ACT_EFF(ACT) are both fine, and so would be ACT_EFF(NRG), and ACT_EFF(DHECOA) & ACT_EFF(DHEOIL).

But I am not able to see why you have dummy issues with heat, sorry.
Maybe you could show what the flows of that CHP are in the results?

[AdvisorDK]

I've attached the image of CHP flows in the results.
You also find my VT file and Sys settings, in case you want to have a look at them

[Antti-L]

Hmm... that certainly looks a bit strange, and the process is different now from what was in your earlier file.

Could you explain why do you here also have a DUMBP output? And why is the DUMBP flow equal to the ELCDK1 flow?  Anyway, it does look like you have the DUMBP included in the PG, which explains the smallness of the heat output (as far as I could check, the capacity appears to be fully used for producing the outputs).  Producing the DUMBP flow is "consuming" about half of the available capacity.

So, I think it otherwise working as expected, but I am puzzled about your DUMBP.  What is the purpose/meaning of that flow?

Ahh... maybe you meant DUMBP to be for reduction operation? But if so, note that such a dummy flow should not be defined to be an electricity output (i.e. of type ELC, like your DUMBP is defined).

[AdvisorDK]

Thank you Antti for your prompt responses.

Right, this file is different from the previous one. The first one was a CHP extraction plant, and this one adds the by-pass reduction or flexibility operation.

The DUMBP is a dummy commodity to activate that bypass feature (flexible operation).
This dummy is of type NRG and has a limit type 'N', and is also a member of the PCG.
With this feature, the CHP production can follow better demand profiles and electricity prices.
However, it ends up with high cost, and low CHP heat.

[Antti-L]

Do you mean the reduction operation?  If so, note that such a dummy flow should not be defined to be an electricity output (i.e. of type ELC, like your DUMBP is defined). However, the operation of pass-out turbines is flexible already without including the reduction option. So, where did you get the idea that you need a dummy electricity output to get flexible operation? I think the documentation is clear about it: No such dummy flows are needed (or even supported) for flexible CHP operation, other than for the optional reduction option.

[AdvisorDK]

Thank you, Antti, for getting back to me so quickly.

Yes, I did mean reduction operation.
You were right about the DUMBP configuration. I deleted the ELC type from the dummy, and now the model has much lower artificial-dummy imports of heat and electricity, and total system costs have gone down a lot.
Then, I made other changes to allow electricity exports, and the dummy imports are gone.

I agree that the documentation explains how to use pass-out turbines without reduction operation, and those are more flexible than backpressure with reduction operation (bypass).
However, in reality, we do have pass-out turbines with reduction operation, which we call extraction with bypass. Those are even more flexible than the single pass-out turbine without reduction operation.
In theory, this is how one would expect the behaviour of those four plant types to be.
Because in Denmark we do have all four types of plants, we need to investigate these special configurations.

[Antti-L]

Thanks very much for the insightful follow-up. 

[Antti-L]

Small additional remark:
Your CHP parameters were the following:
   ACT_EFF(ACT) = 0.38
   NCAP_CHPR(UP) = 1.259173
   NCAP_CEH = 0.15;

With these parameters, the maximum CHP total efficiency would be only 0.722, i.e. pretty low. Due to this low-looking efficiency, I am not quite sure whether this indeed reflects the reality for the modeled existing plant(s), or is it possible that you had not specified the parameters fully correctly (for example, the CHPR seems rather low compared to the levels going even over 2.5 in your figure, and also recall that when 0<CEH≤1 , the efficiency specified (ACT_EFF) should represent the electrical efficiency in full condensing mode).

But anyway, the additional heat gain in the reduction operation is designed to maintain the same efficiency level as in the full CHP mode, i.e. the total energy efficiency would stay at 0.722 for the reduction operation as well. And to me, that makes the efficiency look even more quite low. Consequently, it occurred to me that perhaps it might be useful to allow the user to adjust also the marginal efficiency of the reduction operation?  Such could be easily implemented in TIMES, but maybe you could let me know whether you think that would be useful?

[AdvisorDK]

Yes, it would definitely be interesting to see how the marginal efficiency of the reduction operation could be adjusted in TIMES, thank you!

Just to clarify, the values I posted were only for a test run based on a pass-out plant without reduction (our extraction without bypass). The idea was simply to see the effect of the dummy while keeping all other parameters unchanged.

For `ACT_EFF` we indeed use the nameplate electrical efficiency in condensing mode from the technology catalogue. But for the CHPR I calculated it as:

CHPR = heat efficiency / electricity efficiency;   using observed operational values.

Now I’m wondering if that’s consistent — it feels a bit odd to use observed heat values together with nameplate electricity values. Do you think both should come from the same reference? (either observed or nameplate catalogue), even though nameplate values for heat efficiencies are typically not available.

[AdvisorDK]

Now I have a question about the CHPR~LO values.
For the test experiment purpose, I have used the same hypothetical values in all four types of plants.
For the backpressure reduction operation, I have set this value as the CHPR~LO boundary, allowing bypassing of the turbine to produce more heat, as suggested on page 121 of the VEDA documentation, Part IV.



However, for the pass-out (extraction) turbines, I wonder whether this CHPR~LO value corresponds to the heat and electricity efficiencies at minimum low-pressure condensation.

Our Technology Cathalogue suggests a way to calculate such efficiencies. In case you are curious about it, you can find more information in page 34

[KristofferEML]

Dear Forum,

I configured a CHP process (electricity-driven) using ACT_EFF for the PG commodity - as suggested in the VEDA documentation-, but after running the model I get the warning:

Found CHP processes with PG commodity efficiencies – unsupported

When I switch to the traditional EFF attribute, the warning disappears and results (annual production, system cost) look unchanged.

Besides, EFF does not seem to be an alias or abbreviation of ACT_EFF as referred in the forum, at least according to the TIMES and VEDA attribute lists.

Should I stick with ACT_EFF and ignore the warning (to avoid future inconsistencies when increasing model dimensions like time-slices, milestones, storages, or investments), or is it safe to just use EFF?

Any guidance would be appreciated.

I am right in my understanding that if you use ACT_EFF to model CHP and want to avoid the warning "Found CHP processes with PG commodity efficiencies – unsupported"  in the QA_CHECK.LOG then you can add "other_indexes" in the ~FI_T table? I.e.:

[Antti-L]

> I am right in my understanding that if you use ACT_EFF to model CHP and want to avoid the warning "Found CHP processes with PG commodity efficiencies – unsupported"  in the QA_CHECK.LOG then you can add "other_indexes" in the ~FI_T table? I.e.:

Yes, you can specify the input commodity group for ACT_EFF in several ways:
   In the value column header:  e.g.: ACT_EFF~ACT, ACT_EFF~NRG, ACT_EFF~NRGI, ACT_EFF~ELCWPE
   In a Comm_IN column (on a row where Comm_Out is empty), e.g.: ELCWPE
   In a CommName column, e.g.:  ELCWPE
   In a CommGrp column, e.g.: ACT, NRG, NRGI, ELCWPE
   In an Other_Indexes column, e.g.:  ACT, NRG, NRGI, ELCWPE

If the commodity group is a commodity in the topology, it defines a commodity-specific efficiency.
If the commodity group is 'ACT', it defines a group efficiency for the default shadow group (which would consist of the NRG inputs of a CHP process, excluding any auxiliary flows).
If the commodity group is 'NRG', it defines a group efficiency for the NRG inputs.
If the commodity group is 'NRGI', it likewise defines a group efficiency for the NRG inputs.
If the commodity group is a user-defined CG, it defines a group efficiency for the input flows in that group.

You can use both commodity-specific efficiencies and a group efficiency (the resulting efficiency will be the group efficiency multiplied by the commodity-specific efficiency (default for both is 1).  So, for a multi-fuel CHP you could define a group efficiency, say ACT_EFF(NRG)=0.38, and e.g. a commodity-specific efficiency for one of the inputs, say ACT_EFF(ELCWPE)= 0.9, and then all the other input fuels would have efficiency=0.38, but that ELCWPE would have efficiency=0.38×0.9.