Skip to content
R2X framework logoR2X framework logor2x-reeds

Add Purchaser Load Transform

For installation, follow the published-package tutorial. For the complete modifier sequence and alternative load representations, see prepare a PCM handoff.

Use the add-purchaser-load transform to attach purchaser-load consuming technologies after parsing a base ReEDS system.

This transform adds and/or updates:

  • ReEDSElectrolyzerDemand components and hourly profiles
  • ReEDSSteamMethaneReformingDemand components and hourly profiles
  • ReEDSDataCenterDemand components and hourly profiles

Why This Is A Transform

Section titled “Why This Is A Transform”

Purchaser load is intentionally handled outside the parser build flow so it can be enabled only when needed.

For the ReEDS electricity-consuming technologies handled here (electrolyzer, smr for steam methane reforming and smr_ccs for steam methane reforming with CCS), this transform also avoids duplicate creation by skipping cap.csv-based component creation when the corresponding demand component already exists in the system.

Required Inputs

Section titled “Required Inputs”
  • hour_map_myr_fpath: path to hmap_myr.csv

Optional Inputs

Section titled “Optional Inputs”
  • hydrogen_production_capacity_fpath: path to cap.csv with hydrogen-production capacity
  • consume_characteristics_fpath: path to consume_char.csv with electricity-consuming technology characteristics
  • hydrogen_production_load_fpath: path to prod_load.csv with hydrogen-production demand profiles
  • hydrogen_production_annual_load_fpath: path to prod_load_ann.csv with annual hydrogen-production demand
  • loadsite_op_fpath: path to loadsite_op.csv
  • solve_year
  • weather_year

Check a small data-center profile

Section titled “Check a small data-center profile”

Continue from the fixture tutorial in the same Python session. This three-hour synthetic example creates a separate data-center component in region p1. It demonstrates the input contract and is not a complete annual demand case. For hydrogen-production inputs, use the configuration reference and PCM handoff schemas above.

>>> from pathlib import Path
>>> from tempfile import TemporaryDirectory
>>> import polars as pl
>>> from numpy.testing import assert_allclose
>>> from r2x_reeds import ReEDSDataCenterDemand
>>> from r2x_reeds.sysmod.purchaser_load import PurchaserLoadConfig, add_purchaser_load
>>> with TemporaryDirectory() as folder:
...     policy = Path(folder)
...     pl.DataFrame({"yearhour": [1, 2, 3], "h": ["h1", "h2", "h3"]}).write_csv(policy / "hours.csv")
...     pl.DataFrame({"r": ["p1", "p1", "p1"], "allh": ["h1", "h2", "h3"], "t": [2032, 2032, 2032], "Value": [5.0, 6.0, 7.0]}).write_csv(policy / "sites.csv")
...     purchaser_cfg = PurchaserLoadConfig(solve_year=2032, weather_year=2012, hour_map_myr_fpath=policy / "hours.csv", loadsite_op_fpath=policy / "sites.csv")
...     result = add_purchaser_load(system, purchaser_cfg)
...     assert result.is_ok(), result.unwrap_err()
...     demand = system.get_component(ReEDSDataCenterDemand, "p1_data_center_demand")
...     assert demand.capacity == 7.0
...     series = system.get_time_series(demand, name="max_active_power")
...     assert_allclose(series.data, [5.0, 6.0, 7.0])
...

The assertions check both component capacity and actual profile values. The original electric-demand component remains separate. For a production run, reconcile its complete year with the selected weather calendar and avoid adding the same site increments through add-optimal-siting.