Nuclear Capital Cost is set to zero

[trevorb1]

Checklist

• I am using the current master branch
• I am running on an up-to-date pypsa-usa environment. Update via conda env update -f envs/environment.yaml

The Issue

I think with the switch to getting ATB data from PUDL, something happened with capital costs of nuclear and its not being brought in correctly. This is for a 2020 single period model run with technical retirement (which should not have nuclear as extendable, but in any case it should have correct costs). I have confirmed this is an issue for both electricity only and sector studies

Steps To Reproduce

No response

Expected Behavior

No response

Error Message

No response

Anything else?

No response

[ktehranchi]

Nuclear in the 2024 ATB doesn't start until 2030. We could fix this by setting the minimum technology year to 2030 but that would unnecessarily restrict earlier planning horizon years:

    tech_year = snakemake.wildcards.year
    if int(tech_year) < 2030:
        logger.warning(
            "Minimum cost year supported is 2030, using 2030 expansion costs.",
        )
    years = range(2030, 2051)
    tech_year = min(years, key=lambda x: abs(x - int(tech_year)))

Other solutions could be:

• query the tech year requested, and if it doesn't exist we search a year higher
• use an earlier ATB for planning horizons pre 2024

[trevorb1]

I guess your solution of query the tech year requested, and if it doesn't exist we search a year higher matches what we currently do for offshore wind? Can we just do this for nuclear as well?

pypsa-usa/workflow/scripts/build_cost_data.py

Lines 274 to 284 in 1dadca4

	if tech_year < 2030:
	logger.warning(
	"Using 2030 ATB data for offwind_floating; earlier data not available.",
	)
	pudl_atb_offwind_floating = pudl_atb[
	(pudl_atb["pypsa-name"] == "offwind_floating") & (pudl_atb.projection_year == 2030)
	]
	pudl_atb = pd.concat(
	[pudl_atb_filt, pudl_atb_offwind_floating],
	ignore_index=True,
	)