# -*- coding: utf-8 -*-
# pylint: disable=no-member
"""
Classes to handle renewable resource data
"""
import re
from abc import abstractmethod
import numpy as np
import os
import pandas as pd
import warnings
import logging
from rex.resource import BaseResource
from rex.sam_resource import SAMResource
from rex.utilities.exceptions import (ResourceValueError, ExtrapolationWarning,
ResourceWarning, ResourceRuntimeError,
ResourceKeyError,
MoninObukhovExtrapolationError)
from rex.utilities.parse_keys import parse_keys
logger = logging.getLogger(__name__)
[docs]class WaveResource(BaseResource):
"""
Class to handle Wave BaseResource .h5 files
See Also
--------
resource.BaseResource : Parent class
"""
[docs] def get_SAM_df(self, site):
"""
Get SAM wave resource DataFrame for given site
Parameters
----------
site : int
Site to extract SAM DataFrame for
Returns
-------
res_df : pandas.DataFrame
time-series DataFrame of resource variables needed to run SAM
"""
if not self._unscale:
raise ResourceValueError("SAM requires unscaled values")
res_df = pd.DataFrame({'Year': self.time_index.year,
'Month': self.time_index.month,
'Day': self.time_index.day,
'Hour': self.time_index.hour})
if len(self) > 8784:
res_df['Minute'] = self.time_index.minute
time_zone = self.meta.loc[site, 'timezone']
time_interval = len(self.time_index) // 8760
for var in ['significant_wave_height', 'energy_period']:
ds_slice = (slice(None), site)
var_array = self._get_ds(var, ds_slice)
var_array = SAMResource.roll_timeseries(var_array, time_zone,
time_interval)
res_df[var] = var_array
col_map = {'significant_wave_height': 'wave_height',
'energy_period': 'wave_period'}
res_df = res_df.rename(columns=col_map)
res_df.name = "SAM_-{}".format(site)
return res_df
@staticmethod
def _preload_SAM(res, sites, means=False, time_index_step=None):
"""
Pre-load project_points for SAM
Parameters
----------
res : rex.Resource
rex Resource handler or similar (NSRDB, WindResource,
MultiFileResource, etc...)
sites : list
List of sites to be provided to SAM
means : bool
Boolean flag to compute mean resource when res_array is set
time_index_step: int, optional
Step size for time_index, used to reduce temporal resolution,
by default None
Returns
-------
SAM_res : SAMResource
Instance of SAMResource pre-loaded with Wave resource for sites
in project_points
"""
SAM_res = super()._preload_SAM(res, sites, 'wave', means=means,
time_index_step=time_index_step)
return SAM_res
[docs] @classmethod
def preload_SAM(cls, h5_file, sites, unscale=True, str_decode=True,
group=None, hsds=False, hsds_kwargs=None, means=False,
time_index_step=None):
"""
Pre-load project_points for SAM
Parameters
----------
h5_file : str
h5_file to extract resource from
sites : list
List of sites to be provided to SAM
unscale : bool
Boolean flag to automatically unscale variables on extraction
str_decode : bool
Boolean flag to decode the bytestring meta data into normal
strings. Setting this to False will speed up the meta data read.
group : str
Group within .h5 resource file to open
hsds : bool, optional
Boolean flag to use h5pyd to handle .h5 'files' hosted on AWS
behind HSDS, by default False
hsds_kwargs : dict, optional
Dictionary of optional kwargs for h5pyd, e.g., bucket, username,
password, by default None
means : bool
Boolean flag to compute mean resource when res_array is set
time_index_step: int, optional
Step size for time_index, used to reduce temporal resolution,
by default None
Returns
-------
SAM_res : SAMResource
Instance of SAMResource pre-loaded with Wave resource for sites
in project_points
"""
kwargs = {"unscale": unscale, "hsds": hsds, 'hsds_kwargs': hsds_kwargs,
"str_decode": str_decode, "group": group}
with cls(h5_file, **kwargs) as res:
SAM_res = res._preload_SAM(res, sites, means=means,
time_index_step=time_index_step)
return SAM_res
[docs]class AbstractInterpolatedResource(BaseResource):
"""Class to handle resource dataset interpolation.
Default type of interpolation is linear.
Pressure and Temperature lapse rates are used if p and t are only given at
a single hub height. The lapse rates are from the International Standard
Atmosphere (ISA) or ICAO Standard Atmosphere:
(https://www.faa.gov/regulations_policies/handbooks_manuals/aviation/
phak/media/06_phak_ch4.pdf)
"""
LAPSE_RATES = {'temperature': 6.56, 'pressure': 11_109}
"""Air Temperature and Pressure lapse rate in C/km and Pa/km"""
def __init__(self, h5_file, mode='r', unscale=True, str_decode=True,
group=None, use_lapse_rate=True, hsds=False,
hsds_kwargs=None):
"""
Parameters
----------
h5_file : str
Path to .h5 resource file
mode : str, optional
Mode to instantiate h5py.File instance, by default 'r'
unscale : bool
Boolean flag to automatically unscale variables on extraction
str_decode : bool
Boolean flag to decode the bytestring meta data into normal
strings. Setting this to False will speed up the meta data read.
group : str
Group within .h5 resource file to open
use_lapse_rate : bool
If a dataset is only available at a single hub-height and this flag
value is set to `True`, pressure / temperature values will be
calculated using linear lapse rate adjustment from the available
hub height to the requested one. If the flag value is set to
`False`, the value of these variables at the single available
hub-height will be returned for *all* requested heights. This
option has no effect if data is available at multiple hub-heights.
hsds : bool, optional
Boolean flag to use h5pyd to handle .h5 'files' hosted on AWS
behind HSDS, by default False
hsds_kwargs : dict, optional
Dictionary of optional kwargs for h5pyd, e.g., bucket, username,
password, by default None
"""
self._interp_var = None
self._use_lapse = use_lapse_rate
super().__init__(h5_file, unscale=unscale, str_decode=str_decode,
group=group, hsds=hsds, mode=mode,
hsds_kwargs=hsds_kwargs)
# this is where self.heights or self.depths gets set
self._interpolation_variable = self._parse_interp_var(self.datasets)
prop_name = "{}s".format(self.VARIABLE_NAME)
setattr(self, prop_name, self._interpolation_variable)
@classmethod
def _parse_interp_var(cls, datasets):
"""Extract available interpolation variable values for the
interpolable datasets. Used for interpolation/extrapolation.
Parameters
----------
datasets : list
List of dataset names that will be parsed for interpolation value
suffixes like "windspeed_100m" -> windspeed at 100 meters
Returns
-------
dict
Dictionary of available interpolation variable values for
the interpolable datasets. For example, this could be:
{'windspeed': [10, 100, 200]}
"""
interp_var = {dset: [] for dset in cls.INTERPOLABLE_DSETS}
ignore = ['meta', 'time_index', 'coordinates']
for ds in datasets:
if ds not in ignore:
ds_name, val = cls._parse_name(ds)
if ds_name in interp_var and val is not None:
interp_var[ds_name].append(val)
return interp_var
@classmethod
def _parse_name(cls, ds_name):
"""Extract dataset name and interpolation variable value from
dataset name.
Parameters
----------
ds_name : str
Dataset name
Returns
-------
name : str
Dataset name without interpolation value.
val : int | float
Variable value.
"""
regex = "_-?[0-9]+(.[0-9]+)?{}$".format(cls.VARIABLE_UNIT)
regex = re.search(regex, ds_name)
if regex:
val = regex.group()
name = ds_name.rstrip(val)
val = val.lstrip('_').rstrip(cls.VARIABLE_UNIT)
try:
val = int(val)
except ValueError:
val = float(val)
return name, val
return ds_name, None
@classmethod
def _get_nearest_val(cls, val, vals):
"""
Get two nearest two values in `vals`.
Determine if val is inside or outside the range of vals
(requiring extrapolation instead of interpolation)
Parameters
----------
val : int | float
Value of interest.
vals : list
List of available values.
Returns
-------
nearest_val : list
List of 1st and 2nd nearest val in vals.
extrapolate : bool
Flag as to whether val is inside or outside vals range
"""
vals_arr = np.array(vals, dtype='float32')
dist = np.abs(vals_arr - val)
pos = dist.argsort()[:2]
nearest_d = sorted([vals[p] for p in pos])
extrapolate = np.all(val < vals_arr) or np.all(val > vals_arr)
if extrapolate:
v_min, v_max = np.sort(vals)[[0, -1]]
msg = ('{} is outside the {} range'.format(val, cls.VARIABLE_NAME),
'({}, {}).'.format(v_min, v_max),
'Extrapolation to be used.')
warnings.warn(' '.join(msg), ExtrapolationWarning)
return nearest_d, extrapolate
def _get_closest_existing_dset_name(self, dset):
"""Get the name of an existing dataset closest to interp value. """
var_name, val = self._parse_name(dset)
if val is not None and var_name in self._interpolation_variable:
available_vals = self._interpolation_variable[var_name]
(val, _), _ = self._get_nearest_val(val, available_vals)
dset = '{}_{}{}'.format(var_name, val, self.VARIABLE_UNIT)
return dset
[docs] def get_dset_properties(self, dset):
"""
Get dataset properties (shape, dtype, chunks)
Parameters
----------
dset : str
Dataset to get scale factor for
Returns
-------
shape : tuple
Dataset array shape
dtype : str
Dataset array dtype
chunks : tuple
Dataset chunk size
"""
dset = self._get_closest_existing_dset_name(dset)
return super().get_dset_properties(dset)
[docs] def get_attrs(self, dset=None):
"""
Get h5 attributes either from file or dataset
Parameters
----------
dset : str
Dataset to get attributes for, if None get file (global) attributes
Returns
-------
attrs : dict
Dataset or file attributes
"""
if dset is None:
attrs = dict(self.h5.attrs)
else:
dset = self._get_closest_existing_dset_name(dset)
attrs = super().get_attrs(dset=dset)
return attrs
def _get_ds(self, ds_name, ds_slice):
"""
Extract data from given dataset
Parameters
----------
ds_name : str
Variable dataset to be extracted
ds_slice : tuple
Tuple of (int, slice, list, ndarray) of what to extract from
ds, each arg is for a sequential axis
Returns
-------
out : ndarray
ndarray of variable timeseries data
If unscale, returned in native units else in scaled units
"""
var_name, val = self._parse_name(ds_name)
if val is not None and var_name in self._interpolation_variable:
out = self._get_ds_interpolated(ds_name, ds_slice)
else:
out = super()._get_ds(ds_name, ds_slice)
return out
def _get_ds_interpolated(self, ds_name, ds_slice):
"""Extract data from given dataset at desired interpolation value.
Data is interpolated or extrapolated as needed.
Parameters
----------
ds_name : str
Variable dataset to be extracted
ds_slice : tuple
Tuple of (int, slice, list, ndarray) of what to extract
from ds, each arg is for a sequential axis
Returns
-------
out : ndarray
ndarray of variable timeseries data
If unscale, returned in native units else in scaled units
"""
var_name, val = self._parse_name(ds_name)
interpolation_values = self._interpolation_variable[var_name]
if not interpolation_values:
warnings.warn('No {0} info available for {1!r}, returning '
'single {2!r} value for requested {0} {3}{4}'
.format(self.VARIABLE_NAME, ds_name, var_name, val,
self.VARIABLE_UNIT), ResourceWarning)
out = super()._get_ds(var_name, ds_slice)
elif val in interpolation_values:
ds_name = '{}_{}{}'.format(var_name, int(val), self.VARIABLE_UNIT)
out = super()._get_ds(ds_name, ds_slice)
elif (len(interpolation_values) == 1
and self._use_lapse
and var_name in self.LAPSE_RATES):
out = self._get_ds_lapse(ds_name, ds_slice)
elif len(interpolation_values) == 1:
val = interpolation_values[0]
ds_name = '{}_{}{}'.format(var_name, int(val), self.VARIABLE_UNIT)
warnings.warn('Only one {} available, returning {!r}'
.format(self.VARIABLE_NAME, ds_name),
ResourceWarning)
out = super()._get_ds(ds_name, ds_slice)
else:
out = self._get_calculated_ds(val, ds_name, var_name, ds_slice)
return out
def _get_ds_lapse(self, ds_name, ds_slice,
valid_units=('pa', 'pascals',
'c', 'celsius',
'k', 'kelvin')):
"""Extract data from given dataset where there is only temperature or
pressure data at a single elevation and a lapse rate must be used to
adjust to a new elevation.
Parameters
----------
ds_name : str
Variable dataset to be extracted
ds_slice : tuple
Tuple of (int, slice, list, ndarray) of what to extract
from ds, each arg is for a sequential axis
valid_units : tuple
Tuple of valid lower-case units that can be lapse-rate adjusted. If
the dataset doesnt have units in this list, a warning will be
raised.
Returns
-------
out : ndarray
ndarray of variable timeseries data
If unscale, returned in native units else in scaled units
"""
var_name, val = self._parse_name(ds_name)
interpolation_values = self._interpolation_variable[var_name]
ds_name = '{}_{}{}'.format(var_name, int(interpolation_values[0]),
self.VARIABLE_UNIT)
lapse_rate = self.LAPSE_RATES[var_name]
if self.VARIABLE_UNIT != 'm':
msg = ('Cannot use temperature/pressure lapse rate when vertical '
'interpolation unit is not "m": {}'
.format(self.VARIABLE_UNIT))
logger.error(msg)
raise RuntimeError(msg)
logger.info('Only one {} available for {} at {}, using '
'lapse rate of {} to get to {}'
.format(self.VARIABLE_NAME, var_name,
interpolation_values[0], lapse_rate, val))
ds_units = str(self.units.get(ds_name, None))
if ds_units.lower() not in valid_units:
msg = ('Dataset "{}" is being adjusted to elevation {} via lapse '
'rate but valid units not found: {} (must be Pa or C). '
'Proceeding, but if this is not the desired behavior, set '
'use_lapse_rate=False'.format(ds_name, val, ds_units))
logger.warning(msg)
warnings.warn(msg, ResourceWarning)
out = super()._get_ds(ds_name, ds_slice)
delta_h = (interpolation_values[0] - val) / 1000 # to kilometers
lapse = delta_h * lapse_rate
return out + lapse
def _get_calculated_ds(self, val, ds_name, var_name, ds_slice):
"""Get interpolated/extrapolated values for the dataset. """
vals = self._interpolation_variable[var_name]
(v1, v2), extrapolate = self._get_nearest_val(val, vals)
if extrapolate:
msg = ('Extrapolating {} using linear extrapolation'
.format(ds_name))
warnings.warn(msg, ExtrapolationWarning)
dset_name_1 = '{}_{}{}'.format(var_name, v1, self.VARIABLE_UNIT)
ts1 = super()._get_ds(dset_name_1, ds_slice)
dset_name_2 = '{}_{}{}'.format(var_name, v2, self.VARIABLE_UNIT)
ts2 = super()._get_ds(dset_name_2, ds_slice)
out = linear_interp(ts1, v1, ts2, v2, val)
return out
@staticmethod
def _set_sam_res(res, values, dsets, SAM_res, time_slice, sites):
"""
Set the resource for individual sites at various values
(i.e. hub-heights, depths, etc).
Parameters
----------
res : rex.Resource
rex Resource handler or similar (NSRDB, WindResource,
MultiFileResource, etc...)
values : list | int
List of interpolation values e.g. hub heights or geothermal depths
dsets : list
List of dataset names to set
SAM_res : SAMResource
SAMResource object to load resource data into
time_slice : slice
Slice object representing any temporal subsampling
sites : list | slice
Spatial indices to load.
"""
if isinstance(values, (int, float)):
SAM_res.load_rex_resource(res, dsets, time_slice, sites,
hh=values, hh_unit=res.VARIABLE_UNIT)
else:
_, unique_index = np.unique(values, return_inverse=True)
unique_values = sorted(list(set(values)))
for dset in dsets:
for index, value in enumerate(unique_values):
pos = np.where(unique_index == index)[0]
sites = np.array(SAM_res.sites)[pos]
ds_name = '{}_{}{}'.format(dset, value, res.VARIABLE_UNIT)
SAM_res[dset, :, pos] = res[ds_name, time_slice, sites]
@property
@abstractmethod
def INTERPOLABLE_DSETS(self):
"""
list: Names of the datasets allowed to be interpolated/extrapolated.
"""
raise NotImplementedError
@property
@abstractmethod
def VARIABLE_NAME(self):
"""
str: Name of the variable to interpolate over (e.g. "height").
"""
raise NotImplementedError
@property
@abstractmethod
def VARIABLE_UNIT(self):
"""
str: Abbreviation used in dataset names for the interpolation
variable unit (e.g. "m").
"""
raise NotImplementedError
[docs]class SolarResource(AbstractInterpolatedResource):
"""
Class to handle Solar BaseResource .h5 files
See Also
--------
resource.BaseResource : Parent class
"""
INTERPOLABLE_DSETS = ["temperature", "windspeed"]
VARIABLE_NAME = "height"
VARIABLE_UNIT = "m"
[docs] def get_SAM_df(self, site, extra_cols=None):
"""
Get SAM solar resource DataFrame for given site
Parameters
----------
site : int
Site to extract SAM DataFrame for.
extra_cols : dict, optional
A dictionary where they keys are extra columns
to extract from the SAM solar resource DataFrame
and the values are the names the new columns should
have (e.g. extra_cols={'surface_albedo': 'Surface
Albedo'} will extract the 'surface_albedo' from the
resource file and call it 'Surface Albedo' in the output).
Returns
-------
res_df : pandas.DataFrame
time-series DataFrame of resource variables needed to run SAM
"""
if not self._unscale:
raise ResourceValueError("SAM requires unscaled values")
res_df = pd.DataFrame({'Year': self.time_index.year,
'Month': self.time_index.month,
'Day': self.time_index.day,
'Hour': self.time_index.hour})
if len(self) > 8784 or (self.time_index.minute != 0).any():
res_df['Minute'] = self.time_index.minute
time_zone = self.meta.loc[site, 'timezone']
time_interval = len(self.time_index) // 8760
main_cols = ['dni', 'dhi', 'wind_speed', 'air_temperature']
extra_cols = extra_cols or {}
for var in main_cols + list(extra_cols):
ds_slice = (slice(None), site)
var_array = self._get_ds(var, ds_slice)
var_array = SAMResource.roll_timeseries(var_array, time_zone,
time_interval)
res_df[var] = SAMResource.check_units(var, var_array,
tech='pvwattsv8')
col_map = {'dni': 'DNI', 'dhi': 'DHI', 'wind_speed': 'Wind Speed',
'air_temperature': 'Temperature'}
col_map.update(extra_cols)
res_df = res_df.rename(columns=col_map)
res_df.name = "SAM_-{}".format(site)
return res_df
@staticmethod
def _preload_SAM(res, sites, tech='pvwattsv8', time_index_step=None,
means=False, clearsky=False, bifacial=False):
"""
Pre-load project_points for SAM
Parameters
----------
res : rex.Resource
rex Resource handler or similar (NSRDB, WindResource,
MultiFileResource, etc...)
sites : list
List of sites to be provided to SAM
tech : str, optional
SAM technology string, by default 'pvwattsv8'
time_index_step: int, optional
Step size for time_index, used to reduce temporal resolution,
by default None
means : bool, optional
Boolean flag to compute mean resource when res_array is set,
by default False
clearsky : bool
Boolean flag to pull clearsky instead of real irradiance
bifacial : bool
Boolean flag to pull surface albedo for bifacial modeling.
Returns
-------
SAM_res : SAMResource
Instance of SAMResource pre-loaded with Solar resource for sites
in project_points
"""
time_slice = slice(None, None, time_index_step)
SAM_res = SAMResource(sites, tech, res['time_index', time_slice],
means=means)
sites = SAM_res.sites_slice
SAM_res['meta'] = res['meta', sites]
if clearsky:
SAM_res.set_clearsky()
if bifacial and 'surface_albedo' not in SAM_res.var_list:
SAM_res._var_list.append('surface_albedo')
SAM_res.check_irradiance_datasets(res.datasets, clearsky=clearsky)
SAM_res.load_rex_resource(res, SAM_res.var_list, time_slice, sites,
hh=2)
SAM_res.compute_irradiance(clearsky=clearsky)
return SAM_res
[docs] @classmethod
def preload_SAM(cls, h5_file, sites, unscale=True, str_decode=True,
group=None, hsds=False, hsds_kwargs=None,
tech='pvwattsv8', time_index_step=None, means=False,
clearsky=False, bifacial=False):
"""
Pre-load project_points for SAM
Parameters
----------
h5_file : str
h5_file to extract resource from
sites : list
List of sites to be provided to SAM
unscale : bool
Boolean flag to automatically unscale variables on extraction
str_decode : bool
Boolean flag to decode the bytestring meta data into normal
strings. Setting this to False will speed up the meta data read.
group : str
Group within .h5 resource file to open
hsds : bool, optional
Boolean flag to use h5pyd to handle .h5 'files' hosted on AWS
behind HSDS, by default False
hsds_kwargs : dict, optional
Dictionary of optional kwargs for h5pyd, e.g., bucket, username,
password, by default None
tech : str, optional
SAM technology string, by default 'pvwattsv8'
time_index_step: int, optional
Step size for time_index, used to reduce temporal resolution,
by default None
means : bool, optional
Boolean flag to compute mean resource when res_array is set,
by default False
clearsky : bool
Boolean flag to pull clearsky instead of real irradiance
bifacial : bool
Boolean flag to pull surface albedo for bifacial modeling.
Returns
-------
SAM_res : SAMResource
Instance of SAMResource pre-loaded with Solar resource for sites
in project_points
"""
kwargs = {"unscale": unscale, "hsds": hsds, 'hsds_kwargs': hsds_kwargs,
"str_decode": str_decode, "group": group}
with cls(h5_file, **kwargs) as res:
SAM_res = res._preload_SAM(res, sites, tech=tech,
time_index_step=time_index_step,
means=means, clearsky=clearsky,
bifacial=bifacial)
return SAM_res
[docs]class NSRDB(SolarResource):
"""
Class to handle NSRDB .h5 files
See Also
--------
resource.BaseResource : Parent class
"""
ADD_ATTR = ['add_offset', 'psm_add_offset']
SCALE_ATTR = ['scale_factor', 'psm_scale_factor']
UNIT_ATTR = ['units', 'psm_units']
@staticmethod
def _preload_SAM(res, sites, tech='pvwattsv8', time_index_step=None,
means=False, clearsky=False, bifacial=False,
downscale=None):
"""
Pre-load project_points for SAM
Parameters
----------
res : rex.Resource
rex Resource handler or similar (NSRDB, WindResource,
MultiFileResource, etc...)
sites : list
List of sites to be provided to SAM
tech : str, optional
SAM technology string, by default 'pvwattsv8'
time_index_step: int, optional
Step size for time_index, used to reduce temporal resolution,
by default None
means : bool, optional
Boolean flag to compute mean resource when res_array is set,
by default False
clearsky : bool
Boolean flag to pull clearsky instead of real irradiance
bifacial : bool
Boolean flag to pull surface albedo for bifacial modeling.
downscale : NoneType | dict
Option for NSRDB resource downscaling to higher temporal
resolution. Expects a dict of downscaling kwargs with a minimum
requirement of the desired frequency e.g. 'frequency': '5min'
and an option to add "variability_kwargs".
Returns
-------
SAM_res : SAMResource
Instance of SAMResource pre-loaded with Solar resource for sites
in project_points
"""
time_slice = slice(None, None, time_index_step)
SAM_res = SAMResource(sites, tech, res['time_index', time_slice],
means=means)
sites = SAM_res.sites_slice
SAM_res['meta'] = res['meta', sites]
if clearsky:
SAM_res.set_clearsky()
if bifacial and 'surface_albedo' not in SAM_res.var_list:
SAM_res._var_list.append('surface_albedo')
SAM_res.check_irradiance_datasets(res.datasets, clearsky=clearsky)
if not downscale:
SAM_res.load_rex_resource(res, SAM_res.var_list, time_slice,
sites, hh=2)
SAM_res.compute_irradiance(clearsky=clearsky)
else:
# contingent import to avoid dependencies
from rex.utilities.downscale import downscale_nsrdb
frequency = downscale.pop('frequency')
SAM_res = downscale_nsrdb(SAM_res, res, sam_vars=SAM_res.var_list,
frequency=frequency,
variability_kwargs=downscale)
return SAM_res
[docs] @classmethod
def preload_SAM(cls, h5_file, sites, unscale=True, str_decode=True,
group=None, hsds=False, hsds_kwargs=None,
tech='pvwattsv8', time_index_step=None, means=False,
clearsky=False, bifacial=False, downscale=None):
"""
Pre-load project_points for SAM
Parameters
----------
h5_file : str
h5_file to extract resource from
sites : list
List of sites to be provided to SAM
unscale : bool
Boolean flag to automatically unscale variables on extraction
str_decode : bool
Boolean flag to decode the bytestring meta data into normal
strings. Setting this to False will speed up the meta data read.
group : str
Group within .h5 resource file to open
hsds : bool, optional
Boolean flag to use h5pyd to handle .h5 'files' hosted on AWS
behind HSDS, by default False
hsds_kwargs : dict, optional
Dictionary of optional kwargs for h5pyd, e.g., bucket, username,
password, by default None
tech : str, optional
SAM technology string, by default 'pvwattsv8'
time_index_step: int, optional
Step size for time_index, used to reduce temporal resolution,
by default None
means : bool, optional
Boolean flag to compute mean resource when res_array is set,
by default False
clearsky : bool
Boolean flag to pull clearsky instead of real irradiance
bifacial : bool
Boolean flag to pull surface albedo for bifacial modeling.
downscale : NoneType | dict
Option for NSRDB resource downscaling to higher temporal
resolution. Expects a dict of downscaling kwargs with a minimum
requirement of the desired frequency e.g. 'frequency': '5min'
and an option to add "variability_kwargs".
Returns
-------
SAM_res : SAMResource
Instance of SAMResource pre-loaded with Solar resource for sites
in project_points
"""
kwargs = {"unscale": unscale, "hsds": hsds, 'hsds_kwargs': hsds_kwargs,
"str_decode": str_decode, "group": group}
with cls(h5_file, **kwargs) as res:
SAM_res = res._preload_SAM(res, sites, tech=tech,
time_index_step=time_index_step,
means=means, clearsky=clearsky,
bifacial=bifacial, downscale=downscale)
return SAM_res
[docs]class WindResource(AbstractInterpolatedResource):
"""
Class to handle Wind BaseResource .h5 files
See Also
--------
resource.AbstractInterpolatedResource : Parent class
Examples
--------
>>> import os
>>> from rex import TESTDATADIR, WindResource
>>> file = os.path.join(TESTDATADIR, 'wtk/ri_100_wtk_2012.h5')
>>> with WindResource(file) as res:
>>> print(res.datasets)
['meta', 'pressure_0m', 'pressure_100m', 'pressure_200m',
'temperature_100m', 'temperature_80m', 'time_index', 'winddirection_100m',
'winddirection_80m', 'windspeed_100m', 'windspeed_80m']
WindResource can interpolate between available hub-heights (80 & 100)
>>> with WindResource(file) as res:
>>> wspd_90m = res['windspeed_90m']
>>>
>>> wspd_90m
[[ 6.865 6.77 6.565 ... 8.65 8.62 8.415 ]
[ 7.56 7.245 7.685 ... 5.9649997 5.8 6.2 ]
[ 9.775 9.21 9.225 ... 7.12 7.495 7.675 ]
...
[ 8.38 8.440001 8.85 ... 11.934999 12.139999 12.4 ]
[ 9.900001 9.895 9.93 ... 12.825 12.86 12.965 ]
[ 9.895 10.01 10.305 ... 14.71 14.79 14.764999 ]]
WindResource can also extrapolate beyond available hub-heights
>>> with WindResource(file) as res:
>>> wspd_150m = res['windspeed_150m']
>>>
>>> wspd_150m
ExtrapolationWarning: 150 is outside the height range (80, 100).
Extrapolation to be used.
[[ 7.336291 7.2570405 7.0532546 ... 9.736436 9.713792 9.487364 ]
[ 8.038219 7.687255 8.208041 ... 6.6909685 6.362647 6.668326 ]
[10.5515785 9.804363 9.770399 ... 8.026898 8.468434 8.67222 ]
...
[ 9.079792 9.170363 9.634542 ... 13.472508 13.7102585 14.004617 ]
[10.710078 10.710078 10.698757 ... 14.468795 14.514081 14.6386175]
[10.698757 10.857258 11.174257 ... 16.585903 16.676476 16.653833 ]]
"""
INTERPOLABLE_DSETS = ["temperature", "pressure", "windspeed",
"winddirection"]
VARIABLE_NAME = "height"
VARIABLE_UNIT = "m"
def __getitem__(self, keys):
ds, ds_slice = parse_keys(keys)
_, ds_name = os.path.split(ds)
if 'SAM' in ds_name:
site = ds_slice[0]
if isinstance(site, (int, np.integer)):
_, height = self._parse_name(ds_name)
out = self.get_SAM_df(site, height)
else:
msg = "Can only extract SAM DataFrame for a single site"
raise ResourceRuntimeError(msg)
else:
out = super().__getitem__(keys)
return out
[docs] @staticmethod
def stability_function(zeta):
"""
Calculate stability function depending on sign of L
(negative is unstable, positive is stable)
Parameters
----------
zeta : ndarray
Normalized length
Returns
-------
numpy.ndarray
stability measurements.
"""
stab_fun = np.zeros(len(zeta))
zeta = zeta.astype(float)
# Unstable conditions
x = (np.power(1 - 16 * zeta[zeta < 0], 0.25))
paulson_func = (np.pi / 2 - 2 * np.arctan(x)
+ np.log(np.power(1 + x, 2)
* (1 + np.power(x, 2)) / 8))
y = np.power(1 - 10 * zeta[zeta < 0], 1. / 3)
conv_func = (3 / 2 * np.log(np.power(y, 2) + y + 1. / 3) - np.sqrt(3)
* np.arctan(2 * y + 1 / np.sqrt(3)) + np.pi / np.sqrt(3))
o = ((paulson_func + np.power(zeta[zeta < 0], 2) * conv_func)
/ (1 + np.power(zeta[zeta < 0], 2)))
stab_fun[np.where(zeta < 0)] = o
# Stable conditions
a = 6.1
b = 2.5
o = np.log(zeta[zeta >= 0]
+ (1 + np.power(zeta[zeta >= 0], b))**(1 / b))
o *= -a
stab_fun[np.where(zeta >= 0)] = o
return stab_fun
[docs] @staticmethod
def power_law_interp(ts_1, h_1, ts_2, h_2, h, mean=True):
"""
Power-law interpolate/extrapolate time-series data to height h
Parameters
----------
ts_1 : ndarray
Time-series array at height h_1
h_1 : int | float
Height corresponding to time-seris ts_1
ts_2 : ndarray
Time-series array at height h_2
h_2 : int | float
Height corresponding to time-seris ts_2
h : int | float
Height of desired time-series
mean : bool
Calculate average alpha versus point by point alpha
Returns
-------
out : ndarray
Time-series array at height h
"""
if h_1 > h_2:
h_1, h_2 = h_2, h_1
ts_1, ts_2 = ts_2, ts_1
if mean:
alpha = (np.log(ts_2.mean() / ts_1.mean())
/ np.log(h_2 / h_1))
if alpha < 0.06:
warnings.warn('Alpha is < 0.06', RuntimeWarning)
elif alpha > 0.6:
warnings.warn('Alpha is > 0.6', RuntimeWarning)
else:
# Replace zero values for alpha calculation
ts_1[ts_1 == 0] = 0.001
ts_2[ts_2 == 0] = 0.001
alpha = np.log(ts_2 / ts_1) / np.log(h_2 / h_1)
# The Hellmann exponent varies from 0.06 to 0.6
alpha[alpha < 0.06] = 0.06
alpha[alpha > 0.6] = 0.6
out = ts_1 * (h / h_1)**alpha
return out
[docs] @staticmethod
def shortest_angle(a0, a1):
"""
Calculate the shortest angle distance between a0 and a1
Parameters
----------
a0 : int | float
angle 0 in degrees
a1 : int | float
angle 1 in degrees
Returns
-------
da : int | float
shortest angle distance between a0 and a1
"""
da = (a1 - a0) % 360
return 2 * da % 360 - da
[docs] @classmethod
def circular_interp(cls, ts_1, h_1, ts_2, h_2, h):
"""
Circular interpolate/extrapolate time-series data to height h
Parameters
----------
ts_1 : ndarray
Time-series array at height h_1
h_1 : int | float
Height corresponding to time-seris ts_1
ts_2 : ndarray
Time-series array at height h_2
h_2 : int | float
Height corresponding to time-seris ts_2
h : int | float
Height of desired time-series
Returns
-------
out : ndarray
Time-series array at height h
"""
h_f = (h - h_1) / (h_2 - h_1)
da = cls.shortest_angle(ts_1, ts_2) * h_f
da = np.sign(da) * (np.abs(da) % 360)
out = (ts_2 + da) % 360
return out
@staticmethod
def _check_hub_height(heights, h):
"""
Check requested hub-height against available windspeed hub-heights
If only one hub-height is available change request to match available
hub-height
Parameters
----------
heights : dict
Dictionary of available interpolation variable values for
the interpolable datasets. For example, this could be:
{'windspeed': [10, 100, 200]}
h : int | float
Requested hub-height
Returns
-------
h : int | float
Hub-height to extract
"""
heights = heights['windspeed']
if len(heights) == 1:
h = heights[0]
warnings.warn('Wind speed is only available at {h}m, '
'all variables will be extracted at {h}m'
.format(h=h), ResourceWarning)
return h
def _try_monin_obukhov_extrapolation(self, ts_1, ds_slice, h_1, h):
rmol = 'inversemoninobukhovlength_2m'
if rmol not in self:
msg = ("{} is needed to run monin obukhov extrapolation"
.format(rmol))
raise MoninObukhovExtrapolationError(msg)
if 'roughness_length' in self:
z0 = self._get_ds('roughness_length', ds_slice)
elif 'z0' in self.meta:
z0 = self.meta['z0']
else:
msg = ("roughness length ('z0') is needed to run monin obukhov"
"extrapolation")
raise MoninObukhovExtrapolationError(msg)
L = 1 / self._get_ds(rmol, ds_slice)
out = self.monin_obukhov_extrapolation(ts_1, h_1, z0, L, h)
return out
def _get_ds_interpolated(self, ds_name, ds_slice):
"""Extract data from given dataset at desired interpolation value.
Data is interpolated or extrapolated as needed.
Parameters
----------
ds_name : str
Variable dataset to be extracted
ds_slice : tuple
Tuple of (int, slice, list, ndarray) of what to extract
from ds, each arg is for a sequential axis
Returns
-------
out : ndarray
ndarray of variable timeseries data
If unscale, returned in native units else in scaled units
"""
var_name, __ = self._parse_name(ds_name)
heights = self.heights[var_name]
if not heights:
msg = ("Missing height info for dataset '{}' in {}"
.format(var_name, self.h5_file))
logger.error(msg)
raise ResourceKeyError(msg)
return super()._get_ds_interpolated(ds_name, ds_slice)
def _get_calculated_ds(self, val, ds_name, var_name, ds_slice):
"""Get interpolated/extrapolated values for the dataset. """
heights = self._interpolation_variable[var_name]
(h1, h2), extrapolate = self._get_nearest_val(val, heights)
if extrapolate:
msg = 'Extrapolating {}'.format(ds_name)
dset_name_1 = '{}_{}{}'.format(var_name, h1, self.VARIABLE_UNIT)
ts1 = super()._get_ds(dset_name_1, ds_slice)
dset_name_2 = '{}_{}{}'.format(var_name, h2, self.VARIABLE_UNIT)
ts2 = super()._get_ds(dset_name_2, ds_slice)
if (var_name == 'windspeed') and extrapolate:
if val < h1:
try:
out = self._try_monin_obukhov_extrapolation(ts1,
ds_slice,
h1, val)
msg += ' using Monin Obukhov Extrapolation'
warnings.warn(msg, ExtrapolationWarning)
except MoninObukhovExtrapolationError:
out = self.power_law_interp(ts1, h1, ts2, h2, val)
msg += ' using Power Law Extrapolation'
warnings.warn(msg, ExtrapolationWarning)
else:
out = self.power_law_interp(ts1, h1, ts2, h2, val)
msg += ' using Power Law Extrapolation'
warnings.warn(msg, ExtrapolationWarning)
elif var_name == 'winddirection':
out = self.circular_interp(ts1, h1, ts2, h2, val)
else:
out = linear_interp(ts1, h1, ts2, h2, val)
return out
[docs] def get_SAM_df(self, site, height, require_wind_dir=False, icing=False,
add_header=False):
"""
Get SAM wind resource DataFrame for given site
Parameters
----------
site : int
Site to extract SAM DataFrame for
height : int
Hub height to extract SAM variables at
require_wind_dir : bool, optional
Boolean flag as to whether wind direction will be loaded,
by default False
icing : bool, optional
Boolean flag to include relativehumitidy for icing calculation,
by default False
add_header : bool, optional
Add units and hub_height below variable names, needed for SAM .csv,
by default False
Returns
-------
res_df : pandas.DataFrame
time-series DataFrame of resource variables needed to run SAM
"""
if not self._unscale:
raise ResourceValueError("SAM requires unscaled values")
height = self._check_hub_height(self.heights, height)
units = ['year', 'month', 'day', 'hour']
res_df = pd.DataFrame({'Year': self.time_index.year,
'Month': self.time_index.month,
'Day': self.time_index.day,
'Hour': self.time_index.hour})
if len(self) > 8784:
res_df['Minute'] = self.time_index.minute
time_zone = self.meta.loc[site, 'timezone']
time_interval = len(self.time_index) // 8760
variables = ['pressure', 'temperature', 'winddirection', 'windspeed']
if not require_wind_dir:
variables.remove('winddirection')
if icing:
variables.append('relativehumidity_2m')
for var in variables:
var_name = "{}_{}m".format(var, height)
ds_slice = (slice(None), site)
var_array = self._get_ds(var_name, ds_slice)
var_array = SAMResource.roll_timeseries(var_array, time_zone,
time_interval)
res_df[var] = SAMResource.check_units(var, var_array,
tech='windpower')
res_df[var] = SAMResource.enforce_arr_range(
var, res_df[var],
SAMResource.WIND_DATA_RANGES[var], [site])
col_map = {'pressure': 'Pressure', 'temperature': 'Temperature',
'windspeed': 'Speed', 'winddirection': 'Direction',
'relativehumidity_2m': 'Relative Humidity'}
res_df = res_df.rename(columns=col_map)
res_df.name = "SAM_{}m-{}".format(height, site)
if add_header:
header = pd.DataFrame(columns=res_df.columns)
header_units = units + [self.get_units(v) for v in variables]
header_heights = [height] * len(header_units)
header = pd.DataFrame([header_units, header_heights],
columns=res_df.columns)
res_df = pd.concat((header, res_df)).reset_index(drop=True)
return res_df
@staticmethod
def _preload_SAM(res, sites, hub_heights, time_index_step=None,
means=False, require_wind_dir=False,
precip_rate=False, icing=False):
"""
Pre-load project_points for SAM
Parameters
----------
res : rex.Resource
rex Resource handler or similar (NSRDB, WindResource,
MultiFileResource, etc...)
sites : list
List of sites to be provided to SAM
hub_heights : int | float | list
Hub heights to extract for SAM
time_index_step: int, optional
Step size for time_index, used to reduce temporal resolution,
by default None
means : bool, optional
Boolean flag to compute mean resource when res_array is set,
by default False
require_wind_dir : bool, optional
Boolean flag as to whether wind direction will be loaded,
by default False
precip_rate : bool, optional
Boolean flag as to whether precipitationrate_0m will be preloaded,
by default False
icing : bool, optional
Boolean flag as to whether icing is analyzed.
This will preload relative humidity, by default False
Returns
-------
SAM_res : SAMResource
Instance of SAMResource pre-loaded with Solar resource for sites
in project_points
"""
time_slice = slice(None, None, time_index_step)
SAM_res = SAMResource(sites, 'windpower',
res['time_index', time_slice],
hub_heights=hub_heights,
require_wind_dir=require_wind_dir,
means=means)
sites = SAM_res.sites_slice
SAM_res['meta'] = res['meta', sites]
var_list = SAM_res.var_list
if not require_wind_dir:
var_list.remove('winddirection')
h = res._check_hub_height(res.heights, SAM_res.h)
res._set_sam_res(res, h, var_list, SAM_res, time_slice, sites)
if precip_rate:
var = 'precipitationrate'
ds_name = '{}_0m'.format(var)
SAM_res.append_var_list(var)
SAM_res[var] = res[ds_name, time_slice, sites]
if icing:
var = 'rh'
ds_name = 'relativehumidity_2m'
SAM_res.append_var_list(var)
SAM_res[var] = res[ds_name, time_slice, sites]
return SAM_res
[docs] @classmethod
def preload_SAM(cls, h5_file, sites, hub_heights, unscale=True,
str_decode=True, group=None, hsds=False, hsds_kwargs=None,
time_index_step=None, means=False,
require_wind_dir=False, precip_rate=False, icing=False):
"""
Placeholder for classmethod that will pre-load project_points for SAM
Parameters
----------
h5_file : str
h5_file to extract resource from
sites : list
List of sites to be provided to SAM
hub_heights : int | float | list
Hub heights to extract for SAM
unscale : bool
Boolean flag to automatically unscale variables on extraction
str_decode : bool
Boolean flag to decode the bytestring meta data into normal
strings. Setting this to False will speed up the meta data read.
group : str
Group within .h5 resource file to open
hsds : bool, optional
Boolean flag to use h5pyd to handle .h5 'files' hosted on AWS
behind HSDS, by default False
hsds_kwargs : dict, optional
Dictionary of optional kwargs for h5pyd, e.g., bucket, username,
password, by default None
time_index_step: int, optional
Step size for time_index, used to reduce temporal resolution,
by default None
means : bool, optional
Boolean flag to compute mean resource when res_array is set,
by default False
require_wind_dir : bool, optional
Boolean flag as to whether wind direction will be loaded,
by default False
precip_rate : bool, optional
Boolean flag as to whether precipitationrate_0m will be preloaded,
by default False
icing : bool, optional
Boolean flag as to whether icing is analyzed.
This will preload relative humidity, by default False
Returns
-------
SAM_res : SAMResource
Instance of SAMResource pre-loaded with Solar resource for sites
in project_points
"""
kwargs = {"unscale": unscale, "hsds": hsds, 'hsds_kwargs': hsds_kwargs,
"str_decode": str_decode, "group": group}
with cls(h5_file, **kwargs) as res:
SAM_res = res._preload_SAM(res, sites, hub_heights,
require_wind_dir=require_wind_dir,
precip_rate=precip_rate, icing=icing,
means=means,
time_index_step=time_index_step)
return SAM_res
[docs]class GeothermalResource(AbstractInterpolatedResource):
"""
Class to handle Geothermal Resource .h5 files
See Also
--------
resource.AbstractInterpolatedResource : Parent class
Examples
--------
>>> file = '$TESTDATADIR/geo/template_geo_data.h5'
>>> with GeothermalResource(file) as res:
>>> print(res.datasets)
['meta', 'temperature_3500m', 'temperature_4500m']
GeothermalResource can linearly interpolate between available depths
(3.5km & 4.5km).
>>> with GeothermalResource(file) as res:
>>> temp_4km = res['temperature_4000m']
>>>
>>> temp_4km
[450.5, 434. , 383.5, 422. , 387. , 316.5, 438. , 419. , 424. ,
438.5]
GeothermalResource can also linearly extrapolate beyond available depths
>>> with GeothermalResource(file) as res:
>>> temp_5km = res['temperature_5000m']
>>>
>>> temp_5km
ExtrapolationWarning: 5000 is outside the depth range (3500, 4500).
Extrapolation to be used.
[501.5, 338. , 428.5, 548. , 405. , 301.5, 440. , 565. , 446. ,
341.5]
"""
LAPSE_RATES = {}
INTERPOLABLE_DSETS = ["temperature", "potential_MW"]
VARIABLE_NAME = "depth"
VARIABLE_UNIT = "m"
@staticmethod
def _preload_SAM(res, sites, depths, time_index_step=None,
means=False):
"""
Pre-load project_points for SAM
Parameters
----------
res : rex.Resource
rex Resource handler or similar (NSRDB, WindResource,
MultiFileResource, etc...)
sites : list
List of sites to be provided to SAM
depths : int | float | list
Depths to extract for SAM
time_index_step: int, optional
Step size for time_index, used to reduce temporal resolution,
by default None
means : bool, optional
Boolean flag to compute mean resource when res_array is set,
by default False
Returns
-------
SAM_res : SAMResource
Instance of SAMResource pre-loaded with Solar resource for sites
in project_points
"""
time_slice = slice(None, None, time_index_step)
SAM_res = SAMResource(sites, "geothermal",
res['time_index', time_slice],
depths=depths, means=means)
sites = SAM_res.sites_slice
SAM_res['meta'] = res['meta', sites]
res._set_sam_res(res, SAM_res.d, SAM_res.var_list, SAM_res, time_slice,
sites)
return SAM_res
[docs] @classmethod
def preload_SAM(cls, h5_file, sites, depths, unscale=True, str_decode=True,
group=None, hsds=False, hsds_kwargs=None,
time_index_step=None, means=False):
"""
Pre-load project_points for SAM
Parameters
----------
h5_file : str
h5_file to extract resource from
sites : list
List of sites to be provided to SAM
depths : int | float | list
Depths to extract for SAM
unscale : bool
Boolean flag to automatically unscale variables on extraction
str_decode : bool
Boolean flag to decode the bytestring meta data into normal
strings. Setting this to False will speed up the meta data read.
group : str
Group within .h5 resource file to open
hsds : bool, optional
Boolean flag to use h5pyd to handle .h5 'files' hosted on AWS
behind HSDS, by default False
hsds_kwargs : dict, optional
Dictionary of optional kwargs for h5pyd, e.g., bucket, username,
password, by default None
tech : str, optional
SAM technology string, by default 'geothermal'
time_index_step: int, optional
Step size for time_index, used to reduce temporal resolution,
by default None
means : bool, optional
Boolean flag to compute mean resource when res_array is set,
by default False
Returns
-------
SAM_res : SAMResource
Instance of SAMResource pre-loaded with Solar resource for sites
in project_points
"""
kwargs = {"unscale": unscale, "hsds": hsds, 'hsds_kwargs': hsds_kwargs,
"str_decode": str_decode, "group": group}
with cls(h5_file, **kwargs) as res:
SAM_res = res._preload_SAM(res, sites, depths,
time_index_step=time_index_step,
means=means)
return SAM_res
[docs]def linear_interp(ts_1, h_1, ts_2, h_2, h):
"""
Linear interpolate/extrapolate time-series data to height h
Parameters
----------
ts_1 : ndarray
Time-series array at height h_1
h_1 : int | float
Height corresponding to time-seris ts_1
ts_2 : ndarray
Time-series array at height h_2
h_2 : int | float
Height corresponding to time-seris ts_2
h : int | float
Height of desired time-series
Returns
-------
out : ndarray
Time-series array at height h
"""
if h_1 > h_2:
h_1, h_2 = h_2, h_1
ts_1, ts_2 = ts_2, ts_1
# Calculate slope for every position in variable arrays
m = (ts_2 - ts_1) / (h_2 - h_1)
# Calculate intercept for every position in variable arrays
b = ts_2 - m * h_2
out = m * h + b
return out