## Background

This is level 2 of 5 and continues the work of the earlier levels. The new files for this level are in `workspace/t2/`. The records from the start are in `workspace/t1/`, and the earlier results are in `workspace/t1/results/`.

## Task

Subject: PJM: the load PJM files itself

Thanks for the benchmark note. Next question: two prospects' risk teams work from PJM's own FERC Form 714 filings, or from Catalyst's PUDL copy of them, rather than from EIA-930, and they will hold our PJM numbers against those. Tomas pulled the FERC 714 archive into ferc714/ and the PUDL tables and docs into pudl/; his note is in mail/.

Please put PJM's filed hourly load for 2022 to 2025 on the same clock as our EIA history, set it against EIA's demand series and the zone loads, and score PJM's day-ahead forecast against PJM's own filed load the way you scored it for the benchmark. I also need to know whether PUDL's processed table can stand in for the raw filings for us.

In workspace/t2/results/:
- the aligned hourly series for 2022 to 2025 as a CSV the modelling side can load;
- a workbook with the comparison and the scores.

Maren

## Input

- `workspace/t1/`: the records supplied with the work (`company/`, `eia861/`, `eia930/`, `forecasting/`, `github/`, `mail/`, `meetings/`, `projects/`, `reading/`, `weather/`).
- `workspace/t2/`: the files that came with level t2 (`ferc714/`, `mail/`, `pudl/`).
- `workspace/t1/results/`: the results of level t1.

## Output

- `workspace/t2/results/pjm_load_2022_2025_utc.csv`
- `workspace/t2/results/pjm_load_reconciliation.xlsx`