def create_generation_process_df():
"""
Reads emissions and generation data from different sources to provide
facility-level emissions. Most important inputs to this process come
from the model configuration file.
Parameters
----------
None
Returns
----------
dataframe
Datafrane includes all facility-level emissions
"""
from electricitylci.eia923_generation import (build_generation_data,
eia923_primary_fuel)
from electricitylci.egrid_filter import (
egrid_facilities_to_include,
emissions_and_waste_for_selected_egrid_facilities,
)
from electricitylci.generation import (
egrid_facilities_w_fuel_region,
add_technological_correlation_score,
add_temporal_correlation_score,
)
import electricitylci.emissions_other_sources as em_other
import electricitylci.ampd_plant_emissions as ampd
from electricitylci.combinator import ba_codes
import electricitylci.manual_edits as edits
COMPARTMENT_DICT = {
"emission/air": "air",
"emission/water": "water",
"emission/ground": "ground",
"input": "input",
"output": "output",
"waste": "waste",
"air": "air",
"water": "water",
"ground": "ground",
}
if model_specs.replace_egrid:
generation_data = build_generation_data().drop_duplicates()
cems_df = ampd.generate_plant_emissions(model_specs.eia_gen_year)
cems_df.drop(columns=["FlowUUID"], inplace=True)
emissions_and_waste_for_selected_egrid_facilities = em_other.integrate_replace_emissions(
cems_df, emissions_and_waste_for_selected_egrid_facilities)
else:
from electricitylci.egrid_filter import electricity_for_selected_egrid_facilities
generation_data = electricity_for_selected_egrid_facilities
generation_data["Year"] = model_specs.egrid_year
generation_data["FacilityID"] = generation_data["FacilityID"].astype(
int)
# generation_data = build_generation_data(
# egrid_facilities_to_include=egrid_facilities_to_include
# )
emissions_and_waste_for_selected_egrid_facilities.drop(
columns=["FacilityID"])
emissions_and_waste_for_selected_egrid_facilities[
"eGRID_ID"] = emissions_and_waste_for_selected_egrid_facilities[
"eGRID_ID"].astype(int)
final_database = pd.merge(
left=emissions_and_waste_for_selected_egrid_facilities,
right=generation_data,
right_on=["FacilityID", "Year"],
left_on=["eGRID_ID", "Year"],
how="left",
)
egrid_facilities_w_fuel_region[
"FacilityID"] = egrid_facilities_w_fuel_region["FacilityID"].astype(
int)
final_database = pd.merge(
left=final_database,
right=egrid_facilities_w_fuel_region,
left_on="eGRID_ID",
right_on="FacilityID",
how="left",
suffixes=["", "_right"],
)
if model_specs.replace_egrid:
primary_fuel_df = eia923_primary_fuel(year=model_specs.eia_gen_year)
primary_fuel_df.rename(columns={'Plant Id': "eGRID_ID"}, inplace=True)
primary_fuel_df["eGRID_ID"] = primary_fuel_df["eGRID_ID"].astype(int)
key_df = (primary_fuel_df[[
"eGRID_ID", "FuelCategory"
]].dropna().drop_duplicates(subset="eGRID_ID").set_index("eGRID_ID"))
final_database["FuelCategory"] = final_database["eGRID_ID"].map(
key_df["FuelCategory"])
else:
key_df = (final_database[[
"eGRID_ID", "FuelCategory"
]].dropna().drop_duplicates(subset="eGRID_ID").set_index("eGRID_ID"))
final_database.loc[final_database["FuelCategory"].isnull(),
"FuelCategory"] = final_database.loc[
final_database["FuelCategory"].isnull(),
"eGRID_ID"].map(key_df["FuelCategory"])
# if replace_egrid:
# final_database["FuelCategory"].fillna(
# final_database["FuelCategory_right"], inplace=True
# )
final_database["Final_fuel_agg"] = final_database["FuelCategory"]
# if model_specs.use_primaryfuel_for_coal:
# final_database.loc[
# final_database["FuelCategory"] == "COAL", ["Final_fuel_agg"]
# ] = final_database.loc[
# final_database["FuelCategory"] == "COAL", "PrimaryFuel"
# ]
try:
year_filter = final_database["Year_x"] == final_database["Year_y"]
final_database = final_database.loc[year_filter, :]
final_database.drop(columns="Year_y", inplace=True)
except KeyError:
pass
final_database.rename(columns={"Year_x": "Year"}, inplace=True)
final_database = map_emissions_to_fedelemflows(final_database)
dup_cols_check = [
"FacilityID",
"FuelCategory",
"FlowName",
"FlowAmount",
"Compartment",
]
final_database = final_database.loc[:,
~final_database.columns.duplicated()]
final_database = final_database.drop_duplicates(subset=dup_cols_check)
final_database.drop(
columns=["FuelCategory", "FacilityID_x", "FacilityID_y"], inplace=True)
final_database.rename(
columns={
"Final_fuel_agg": "FuelCategory",
"TargetFlowUUID": "FlowUUID",
},
inplace=True,
)
final_database = add_temporal_correlation_score(
final_database, model_specs.electricity_lci_target_year)
final_database = add_technological_correlation_score(final_database)
final_database["DataCollection"] = 5
final_database["GeographicalCorrelation"] = 1
final_database["eGRID_ID"] = final_database["eGRID_ID"].astype(int)
final_database.sort_values(by=["eGRID_ID", "Compartment", "FlowName"],
inplace=True)
final_database["stage_code"] = "Power plant"
final_database["Compartment_path"] = final_database["Compartment"]
final_database["Compartment"] = final_database["Compartment_path"].map(
COMPARTMENT_DICT)
final_database["Balancing Authority Name"] = final_database[
"Balancing Authority Code"].map(ba_codes["BA_Name"])
final_database["EIA_Region"] = final_database[
"Balancing Authority Code"].map(ba_codes["EIA_Region"])
final_database["FERC_Region"] = final_database[
"Balancing Authority Code"].map(ba_codes["FERC_Region"])
final_database = edits.check_for_edits(final_database, "generation.py",
"create_generation_process_df")
return final_database
def add_flow_representativeness_data_quality_scores(db, total_gen):
db = add_technological_correlation_score(db)
db = add_temporal_correlation_score(db)
db = add_data_collection_score(db, total_gen)
return db
def add_fuel_inputs(gen_df, upstream_df, upstream_dict):
"""
Converts the upstream emissions database to fuel inputs and adds them
to the generator dataframe. This is in preparation of generating unit
processes for openLCA.
Parameters
----------
gen_df : dataframe
The generator df containing power plant emissions.
upstream_df : dataframe
The combined upstream dataframe.
upstream_dict : dictionary
This is the dictionary of upstream "unit processes" as generated by
electricitylci.upstream_dict after the upstream_dict has been written
to json-ld. This is important because the uuids for the upstream
"unit processes" are only generated when written to json-ld.
Returns
-------
dataframe
"""
from electricitylci.generation import (
add_technological_correlation_score,
add_temporal_correlation_score,
)
upstream_reduced = upstream_df.drop_duplicates(
subset=["plant_id", "stage_code", "quantity"]
)
fuel_df = pd.DataFrame(columns=gen_df.columns)
# The upstream reduced should only have one instance of each plant/stage code
# combination. We'll first map the upstream dictionary to each plant
# and then expand that dictionary into columns we can use. The goal is
# to generate the fuels and associated metadata with each plant. That will
# then be merged with the generation database.
fuel_df["flowdict"] = upstream_reduced["stage_code"].map(upstream_dict)
expand_fuel_df = fuel_df["flowdict"].apply(pd.Series)
fuel_df.drop(columns=["flowdict"], inplace=True)
fuel_df["Compartment"] = "input"
fuel_df["FlowName"] = expand_fuel_df["q_reference_name"]
fuel_df["stage_code"] = upstream_reduced["stage_code"]
fuel_df["FlowAmount"] = upstream_reduced["quantity"]
fuel_df["FlowUUID"] = expand_fuel_df["q_reference_id"]
fuel_df["Unit"] = expand_fuel_df["q_reference_unit"]
fuel_df["eGRID_ID"] = upstream_df["plant_id"]
fuel_df["FacilityID"] = upstream_df["plant_id"]
fuel_df["FuelCategory"] = upstream_df["FuelCategory"]
fuel_df["Year"] = upstream_df["Year"]
merge_cols = [
"Age",
"Balancing Authority Code",
"Balancing Authority Name",
"Electricity",
# "FRS_ID",
"NERC",
"Subregion",
]
fuel_df.drop(columns=merge_cols, inplace=True)
gen_df_reduced = gen_df[merge_cols + ["eGRID_ID"]].drop_duplicates(
subset=["eGRID_ID"]
)
fuel_df = fuel_df.merge(
right=gen_df_reduced,
left_on="eGRID_ID",
right_on="eGRID_ID",
how="left",
)
fuel_df.dropna(subset=["Electricity"], inplace=True)
fuel_df["Source"] = "eia"
fuel_df = add_temporal_correlation_score(fuel_df)
fuel_df["DataCollection"] = 5
fuel_df["GeographicalCorrelation"] = 1
fuel_df["TechnologicalCorrelation"] = 1
fuel_df["ReliabilityScore"] = 1
fuel_df["ElementaryFlowPrimeContext"] = "input"
fuel_cat_key = (
gen_df[["FacilityID", "FuelCategory"]]
.drop_duplicates(subset="FacilityID")
.set_index("FacilityID")
)
fuel_df["FuelCategory"] = fuel_df["FacilityID"].map(
fuel_cat_key["FuelCategory"]
)
gen_plus_up_df = pd.concat([gen_df, fuel_df], ignore_index=True)
gen_plus_up_df = fill_nans(gen_plus_up_df)
return gen_plus_up_df