patio.data.entity_ids

Attributes

BA_REPLACE
RESPS_TO_KEEP

Functions

add_ba_code(input_df[, new_ba_col, ba_rollup_only, ...])	Add respondent and final ba_codes.
adjust_ba_codes(df[, new_ba_col])	Process for assigning plants to EIA BA codes and FERC 1 Respondent IDs.

Module Contents

patio.data.entity_ids.BA_REPLACE

patio.data.entity_ids.RESPS_TO_KEEP = (130, 177, 186, 195, 210, 22, 531, 552, 556, 58, 656, 658)

patio.data.entity_ids.add_ba_code(input_df, new_ba_col='final_ba_code', *, ba_rollup_only=False, drop_interim=False, apply_purchaser=True)

Add respondent and final ba_codes.

Steps if ba_rollup_only is False. If it is True, we simply apply the BA_REPLACE map to balancing_authority_code_eia. 1. First we read in UTH utility information which associates utility_id_eia

to FERC respondent_id. In that dataset we grab those two columns, drop rows where either is null and remove duplicates. For the next step it is critical that a given utility_id_eia map to one and only one respondent_id, though multiple utility_id_eia can map to single a respondent_id.

2. We then merge the result of (1) onto the incoming pandas.DataFrame

3. [Optional if applying purchaser data] Read in UTH purchased power data, select only purchase data from 2020 (the last manually tagged year) where plant_id_eia is not null, i.e. cases where the purchased power tagging found a plant-level match. This data will allow us to associate plant_id_eia with the respondent_id that purchases its power. To make this unique we require that the respondent_id purchases ≥90% of power that the plant sells (NOTE: this is not as a % of the total plant generation, just that plant’s sales that we have identified in the purchase power tagging process). We then merge this onto the result of (2) on plant_id_eia. The resulting final_respondent_id for a plant_id_eia only uses the respondent_id_purchaser if that plant_id_eia did not have a respondent_id assigned in step (2).

4. The final step is to adjust the final ba codes using adjust_ba_codes(). This process attempts to group the rows of input_df into planning areas by doing one of the following:

   1. Using the final_respondent_id irrespective of a given row’s balancing_authority_code_eia.

   2. Using the balancing_authority_code_eia irrespective of a given row’s final_respondent_id.

   2. Combining one or more balancing_authority_code_eia into a single new region.

   3. Doing something more complicated (see adjust_ba_codes() body for the 8 special cases).

Parameters:

• input_df (pandas.DataFrame) – frame to add BA / respondent columns to

• new_ba_col (str) – name of ultimate BA code column

• ba_rollup_only (bool) – if True, we only rollup EIA BA codes using BA_REPLACE

• drop_interim (bool) – if True, drop all the intermediate respondent / BA code columns

• apply_purchaser (bool) – if True, change respondent to respondent of purchaser, requires plant_id_eia

Returns:

patio.data.entity_ids.adjust_ba_codes(df, new_ba_col='final_ba_code')

Process for assigning plants to EIA BA codes and FERC 1 Respondent IDs.

Parameters:

df (pandas.DataFrame)

Return type:

pandas.DataFrame