energy_analysis.chile_fleet.pipeline#

Leakage-safe per-site fitting, hourly screens and fleet accounting.

class energy_analysis.chile_fleet.pipeline.PredictiveModel(*args, **kwargs)[source]#

Bases: Protocol

property training_scale_mwh: float#
predict(poa_wm2, temperature_c)[source]#
Parameters:
  • poa_wm2 (ndarray[tuple[Any, ...], dtype[float64]])

  • temperature_c (ndarray[tuple[Any, ...], dtype[float64]])

Return type:

ndarray[tuple[Any, …], dtype[float64]]

weather_support(poa_wm2, temperature_c)[source]#
Parameters:
  • poa_wm2 (ndarray[tuple[Any, ...], dtype[float64]])

  • temperature_c (ndarray[tuple[Any, ...], dtype[float64]])

Return type:

ndarray[tuple[Any, …], dtype[bool]]

to_dict()[source]#
Return type:

dict[str, object]

class energy_analysis.chile_fleet.pipeline.StageResult(expected: 'FloatArray', poa: 'FloatArray', elevation: 'FloatArray', temperature: 'FloatArray', accepted: 'NDArray[np.bool_]', supported: 'NDArray[np.bool_]', model: 'PredictiveModel | None', band: 'FittedAcceptanceBand | None', provenance: 'dict[str, object]')[source]#

Bases: object

Parameters:
  • expected (ndarray[tuple[Any, ...], dtype[float64]])

  • poa (ndarray[tuple[Any, ...], dtype[float64]])

  • elevation (ndarray[tuple[Any, ...], dtype[float64]])

  • temperature (ndarray[tuple[Any, ...], dtype[float64]])

  • accepted (ndarray[tuple[Any, ...], dtype[bool]])

  • supported (ndarray[tuple[Any, ...], dtype[bool]])

  • model (PredictiveModel | None)

  • band (FittedAcceptanceBand | None)

  • provenance (dict[str, object])

expected: ndarray[tuple[Any, ...], dtype[float64]]#
poa: ndarray[tuple[Any, ...], dtype[float64]]#
elevation: ndarray[tuple[Any, ...], dtype[float64]]#
temperature: ndarray[tuple[Any, ...], dtype[float64]]#
accepted: ndarray[tuple[Any, ...], dtype[bool]]#
supported: ndarray[tuple[Any, ...], dtype[bool]]#
model: PredictiveModel | None#
band: FittedAcceptanceBand | None#
provenance: dict[str, object]#
energy_analysis.chile_fleet.pipeline.fit_stage(frame, metadata, config, *, cutoff)[source]#

Fit geometry, acceptance band and generation model strictly before cutoff.

Parameters:
  • frame (DataFrame)

  • metadata (dict[str, object])

  • config (FleetConfig)

  • cutoff (Timestamp)

Return type:

StageResult

energy_analysis.chile_fleet.pipeline.analyze_site(frame, metadata, config)[source]#
Parameters:
  • frame (DataFrame)

  • metadata (dict[str, object])

  • config (FleetConfig)

Return type:

tuple[DataFrame, dict[str, object], dict[str, object]]

energy_analysis.chile_fleet.pipeline.run_fleet(config, *, progress=None)[source]#

Process the entire supplied registry with one site’s arrays resident at a time.

Parameters:
  • config (FleetConfig)

  • progress (Callable[[str], None] | None)

Return type:

dict[str, object]