Electricity variables

[enokandi]

The yaml file containing newly defined electricity grid variables has been added.

[danielhuppmann]

Here's an initial set of review comments for this PR.

[danielhuppmann]

fyi @JKNTNU

[JKNTNU]

[like] Jannis Kerl reacted to your message:

…

________________________________ From: Daniel Huppmann ***@***.***> Sent: Wednesday, January 8, 2025 1:24:50 PM To: iiasa/idesignres-workflow ***@***.***> Cc: Jannis Kerl ***@***.***>; Mention ***@***.***> Subject: Re: [iiasa/idesignres-workflow] Electricity variables (PR #7) fyi @JKNTNU<https://github.com/JKNTNU> — Reply to this email directly, view it on GitHub<#7 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/BNB6JGX7QXS2HMXR3VQWRED2JURKFAVCNFSM6AAAAABT5NVQD2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNZXGY3DKOBWGE>. You are receiving this because you were mentioned.Message ID: ***@***.***>

[enokandi]

Thanks, dear Daniel, for your comments.

Before updating my PR, I wanted to ask about the level of freedom we have in defining variables. For example, for generation unit types, we have more detailed data than what is currently reflected in the existing nomenclature. Taking gas-fired generation units as an example, the nomenclature currently provides only “Capacity|Electricity|Gas.” However, we need more specificity, such as “Capacity|Electricity|Gas|Combined Cycle” or “Capacity|Electricity|Gas|Steam Turbine.”

I’m not sure if we’re allowed to define new variables like these or if doing so might cause conflicts with other variables, such as “Capacity|Electricity|Gas|w/ CC” or “Capacity|Electricity|Gas|w/o CC.”

[danielhuppmann]

The newly added variables should be consistent with the variables in common-definitions, so specifically for your examples:

1. There are variables like "Capacity|Electricity|Gas", so adding "Capacity|Electricity|Gas|Combined Cycle" in this project repository makes sense.
2. There are variables like "Capital Cost|..." in common-definitions, so adding "Operational Cost|..." and "Fixed Cost|..." (or similar) makes sense.

[enokandi]

Thanks. I corrected the issues.
I used the 'Marginal Cost' since I believe it is more accurate than operational cost. However, if I have to use Operational Cost instead, please tell me.

[danielhuppmann]

Thanks! A few more suggestions plus an explanation for using tag-lists, which would make this file much readable and shorter.

[danielhuppmann]

Are these variables really necessary? A gas turbine will always be used with Natural Gas, not other Fossil Fuels... Also, it is confusing that the other types of gas power plants are not listed here. So better to remove.

[enokandi]

Hi Again Daniel. Thank you for your comments. The point is that the gas turbine units do not use natural gas as the fuel necessarily. The term "gas" in a gas turbine refers to the working fluid used to drive the turbine, similar to how steam is used in steam turbines.

[danielhuppmann]

Fair enough (I meant that oil-based gas turbines are probably not really used any more, right?).

But that still doesn't mean that we need a part-aggregate "Fossil". After all, you could also run it with Hydrogen, which could or could not be from fossil sources...

[enokandi]

OK. Sorry for the misunderstanding on my part.
However, in our dataset, we still have a considerable installed capacity of gas turbine units using various primary fuels, so I had to include all of them.

Regarding "Fossil," I used this term to classify the diesel generation units in my dataset because "Diesel" was not listed as a separate category in the common nomenclature.

[danielhuppmann]

Diesel-fired power plants should be reported as "Oil", or is there any reason against that?

[danielhuppmann]

You could simplify this file by using a tag-list, so creating a separate file tag_power_turbine_types.yaml (or similar) that has

- Turbine Type:
    - Combined Cycle:
        description: combined-cycle (CC)
    - Gas Turbine:
         description: gas turbines (GT)
    - Integrated Gasification Combined-Cycle:
        description: integrated gasification combined cycle plants (IGCC)
    - Steam Turbine:
        description: steam turbines

And then simplify this section

Suggested change

	- Capacity|Electricity|Gas|Combined Cycle:
	description: Installed (available) capacity to generate electricity from natural gas combined cycle plants
	unit: GW
	- Capacity|Electricity|Gas|Gas Turbine:
	description: Installed (available) capacity to generate electricity from natural gas gas turbines
	unit: GW
	- Capacity|Electricity|Gas|Integrated Gasification Combined Cycle:
	description: Installed (available) capacity to generate electricity from integrated gasification combined cycle plants
	unit: GW
	- Capacity|Electricity|Gas|Steam Turbine:
	description: Installed (available) capacity to generate electricity from natural gas steam turbines
	unit: GW
	- Capacity|Electricity|Gas|{Turbine Type}:
	description: Installed (available) capacity to generate electricity with natural gas
	using {Turbine Type} plants
	unit: GW

Same for all other variables that use all combinations of turbines.

This produces a much more concise and readable variable list file. The nomenclature package will automatically create all permutations during scenario validation.

[enokandi]

Nice. I will do it.

[enokandi]

It's done now.

[danielhuppmann]

Almost there, please remove "Capacity|Electricity|Fossil|{Turbine Type}:" (or split "Oil" into different components if this is necessary and relevant)

[enokandi]

Almost there, please remove "Capacity|Electricity|Fossil|{Turbine Type}:" (or split "Oil" into different components if this is necessary and relevant)

Hi Daniel. I removed both "Capacity|Fossil" and "Marginal Cost|Fossil" as you asked. As a result, we currently have no representation for diesel, so we need to combine it with oil. Does that seem correct to you?

Regarding the term “Marginal Cost,” here at NTNU, we discussed using “Variable Cost” and “Fixed Cost” instead. Therefore, the total generation cost would consist of fixed, variable, and fuel costs. We also agreed that “Variable Cost” is more accurate than “Operation Cost.” I’d like to get your opinion on this as well.

[danielhuppmann]

For the record, using ”Fossil“ only for diesel didn’t make any sense. The current description of “Oil” says “electricity from oil-based sources”, which does include diesel.

if you want to distinguish between different types of oil products, please add distinct variable for each fuel type.

[danielhuppmann]

About “Marginal Cost” vs. “Variable Cost” vs. “Operational Cost”, all are ok. But I find it confusing that you do not include fuel costs in that. From my understanding about power systems, linear costs excluding fuel costs are small compared to the other components, and variable costs usually include all linear costs in simple models that do not model input fuels endogenously.

Are you planning to also add variables for “Fuel Cost”?

if yes, I would stick with “Operational Cost” for the non-fuel linear component to avoid confusion.

[enokandi]

About “Marginal Cost” vs. “Variable Cost” vs. “Operational Cost”, all are ok. But I find it confusing that you do not include fuel costs in that. From my understanding about power systems, linear costs excluding fuel costs are small compared to the other components, and variable costs usually include all linear costs in simple models that do not model input fuels endogenously.

Are you planning to also add variables for “Fuel Cost”?

if yes, I would stick with “Operational Cost” for the non-fuel linear component to avoid confusion.

No actually. The idea here was to have both Variable Cost and Fuel Cost individually, similar to the Empire model. So I replaced Marginal cost with Variable cost.

[enokandi]

Moreover, I removed IGCC from turbine technologies and used abbreviations (CCGT and OCGT) to make it consistent with Jannis.

[danielhuppmann]

It seems to me that this PR has overlap with #9, so let's wait until that PR is merged and then revisit.