reVX.hybrid_stats.hybrid_stats.HybridCrossCorrelation

class HybridCrossCorrelation(solar_h5, wind_h5, res_cls=<class 'rex.resource.Resource'>, year=None)

Bases: HybridStats

Compute the temporal cross correlations for co-located wind and solar generation

Parameters:

• solar_h5 (str) – Path to solar h5 file(s)

• wind_h5 (str) – Path to wind h5 file(s)

• res_cls (Class, optional) – Resource class to use to access res_h5, by default Resource

• year (str | int, optional) – Year to extract time-index for if running on a multi-year file, by default None

Methods

cf_profile(solar_h5, wind_h5[, lag_range, ...])	Compute cross correlations on cf_profile dataset
compute_stats(dataset[, lag_range, ...])	Compute correlations
cross_correlation(solar_data, wind_data, m)	Compute the cross-correlation between solar and wind time-series data with time-lag m
run(solar_h5, wind_h5, dataset[, lag_range, ...])	Compute cross correlations between solar and wind time-series
save_stats(out_stats, out_fpath)	Save correlations to disk

Attributes

STATS
lat_lon	Resource (lat, lon) coordinates
meta	Resource meta-data table
res_cls	Resource class to use to access res_h5
solar_h5	Path to solar h5 file(s)
statistics	Dictionary of statistic functions/kwargs to run
time_index	Resource Datetimes
wind_h5	Path to wind h5 file(s)

static cross_correlation(solar_data, wind_data, m)

Compute the cross-correlation between solar and wind time-series data with time-lag m

Parameters:

• solar_data (ndarray) – Time-series solar data

• wind_data (ndarray) – Time-series wind data

• m (int) – Integer shift between solar and wind time-series, is pass directly to np.roll, so it will be the number of time-steps that are shifted. The lag time will be m * dt where dt is the time-step size.

Returns:

corr (ndarray) – Cross-correlation coefficient for each solar, wind site pair

compute_stats(dataset, lag_range=(-50, 51, 1), max_workers=None, sites_per_worker=1000, lat_lon_only=True)

Compute correlations

Parameters:

• dataset (tuple | str) – Dataset to compare, if a string, extract the same dataset for both with and solar, other wise a tuple of the form: (solar_dataset, wind_dataset)

• lag_range (tuple, optional) – The range of lag (m) values to compute the cross-correlation for (start, stop, step). Cross-correlation will be run for all lags in range(start, stop, step), each value in the range is the number of timesteps by which the time-series will be shifted to compute the cross-correlation. by default (-50, 51, 1)

• max_workers (None | int, optional) – Number of workers to use, if 1 run in serial, if None use all available cores, by default None

• sites_per_worker (int, optional) – Number of sites to extract on each worker, by default 1000

• lat_lon_only (bool, optional) – Only append lat, lon coordinates to stats, by default True

Returns:

res_stats (pandas.DataFrame) – DataFrame of desired statistics at desired time intervals

classmethod run(solar_h5, wind_h5, dataset, lag_range=(-50, 51, 1), res_cls=<class 'rex.resource.Resource'>, year=None, max_workers=None, sites_per_worker=1000, lat_lon_only=True, out_path=None)

Compute cross correlations between solar and wind time-series

Parameters:

• solar_h5 (str) – Path to solar h5 file(s)

• wind_h5 (str) – Path to wind h5 file(s)

• dataset (tuple | str) – Dataset to compare, if a string, extract the same dataset for both with and solar, other wise a tuple of the form: (solar_dataset, wind_dataset)

• year (str | int, optional) – Year to extract time-index for if running on a multi-year file, by default None

• lag_range (tuple, optional) – The range of lag (m) values to compute the cross-correlation for (start, stop, step). Cross-correlation will be run for all lags in range(start, stop, step), each value in the range is the number of timesteps by which the time-series will be shifted to compute the cross-correlation. by default (-50, 51, 1)

• res_cls (Class, optional) – Resource class to use to access res_h5, by default Resource

• max_workers (None | int, optional) – Number of workers to use, if 1 run in serial, if None use all available cores, by default None

• sites_per_worker (int, optional) – Number of sites to extract on each worker, by default 1000

• lat_lon_only (bool, optional) – Only append lat, lon coordinates to stats, by default True

• out_path (str, optional) – .csv, or .json path to save statistics too, by default None

Returns:

out_stats (pandas.DataFrame) – DataFrame of resource statistics

classmethod cf_profile(solar_h5, wind_h5, lag_range=(-50, 51, 1), res_cls=<class 'rex.resource.Resource'>, max_workers=None, sites_per_worker=1000, lat_lon_only=True, out_path=None)

Compute cross correlations on cf_profile dataset

Parameters:

• solar_h5 (str) – Path to solar h5 file(s)

• wind_h5 (str) – Path to wind h5 file(s)

• dataset (str) – Dataset to extract stats for

• lag_range (tuple, optional) – The range of lag (m) values to compute the cross-correlation for (start, stop, step). Cross-correlation will be run for all lags in range(start, stop, step), each value in the range is the number of timesteps by which the time-series will be shifted to compute the cross-correlation. by default (-50, 51, 1)

• res_cls (Class, optional) – Resource class to use to access res_h5, by default Resource

• max_workers (None | int, optional) – Number of workers to use, if 1 run in serial, if None use all available cores, by default None

• sites_per_worker (int, optional) – Number of sites to extract on each worker, by default 1000

• lat_lon_only (bool, optional) – Only append lat, lon coordinates to stats, by default True

• out_path (str, optional) – .csv, or .json path to save statistics too, by default None

Returns:

out_stats (pandas.DataFrame) – DataFrame of resource statistics

property lat_lon

Resource (lat, lon) coordinates

Returns:

pandas.DataFrame

property meta

Resource meta-data table

Returns:

pandas.DataFrame

property res_cls

Resource class to use to access res_h5

Returns:

Class

static save_stats(out_stats, out_fpath)

Save correlations to disk

Parameters:

• out_stats (pandas.DataFrame) – Table of correlations to save

• out_path (str) – .csv, or .json path to save statistics too

property solar_h5

Path to solar h5 file(s)

Returns:

str

property statistics

Dictionary of statistic functions/kwargs to run

Returns:

dict

property time_index

Resource Datetimes

Returns:

pandas.DatetimeIndex

property wind_h5

Path to wind h5 file(s)

Returns:

str