import numpy as np
import pandas as pd
from scipy.interpolate import interpn
from foxes.core import TurbineType
from foxes.utils import PandasFileHelper
from foxes.data import PCTCURVE, parse_Pct_two_files
import foxes.variables as FV
import foxes.constants as FC
[docs]
class WsTI2PCtFromTwo(TurbineType):
"""
Calculate turbulent intensity dependent power
and ct values, as given by two individual
files.
The structure of each file is:
ws,0.05,0.06,0.07,...
The first column represents wind speed in m/s
and the subsequent columns are TI values
(not neccessarily in order).
Attributes
----------
source_P: str or pandas.DataFrame
The file path for the power curve, static name, or data
source_ct: str or pandas.DataFrame
The file path for the ct curve, static name, or data
WSCT: str
The wind speed variable for ct lookup
WSP: str
The wind speed variable for power lookup
rpars_P: dict, optional
Parameters for pandas power file reading
rpars_ct: dict, optional
Parameters for pandas ct file reading
ipars_P: dict, optional
Parameters for scipy.interpolate.interpn()
ipars_ct: dict, optional
Parameters for scipy.interpolate.interpn()
rho: float
The air densitiy for which the data is valid
or None for no correction
:group: models.turbine_types
"""
[docs]
def __init__(
self,
data_source_P,
data_source_ct,
rho=None,
p_ct=1.0,
p_P=1.88,
var_ws_ct=FV.REWS2,
var_ws_P=FV.REWS3,
pd_file_read_pars_P={},
pd_file_read_pars_ct={},
interpn_pars_P=None,
interpn_pars_ct=None,
**parameters,
):
"""
Constructor.
Parameters
----------
data_source_P: str or pandas.DataFrame
The file path for the power curve, static name, or data
data_source_ct: str or pandas.DataFrame
The file path for the ct curve, static name, or data
rho: float, optional
The air densitiy for which the data is valid
or None for no correction
p_ct: float
The exponent for yaw dependency of ct
p_P: float
The exponent for yaw dependency of P
var_ws_ct: str
The wind speed variable for ct lookup
var_ws_P: str
The wind speed variable for power lookup
pd_file_read_pars_P: dict
Parameters for pandas power file reading
pd_file_read_pars_ct: dict
Parameters for pandas ct file reading
interpn_pars_P: dict, optional
Parameters for scipy.interpolate.interpn()
interpn_pars_ct: dict, optional
Parameters for scipy.interpolate.interpn()
parameters: dict, optional
Additional parameters for TurbineType class
"""
if not isinstance(data_source_P, pd.DataFrame) or not isinstance(
data_source_ct, pd.DataFrame
):
pars = parse_Pct_two_files(data_source_P, data_source_ct)
else:
pars = parameters
super().__init__(**pars)
self.source_P = data_source_P
self.source_ct = data_source_ct
self.rho = rho
self.p_ct = p_ct
self.p_P = p_P
self.WSCT = var_ws_ct
self.WSP = var_ws_P
self.rpars_P = pd_file_read_pars_P
self.rpars_ct = pd_file_read_pars_ct
self.ipars_P = interpn_pars_P
self.ipars_ct = interpn_pars_ct
if self.ipars_P is None:
self.ipars_P = dict(method="linear", bounds_error=True, fill_value=0.0)
if self.ipars_ct is None:
self.ipars_ct = dict(method="linear", bounds_error=True, fill_value=0.0)
self._P = None
self._ct = None
[docs]
def __repr__(self):
a = f"D={self.D}, H={self.H}, P_nominal={self.P_nominal}, P_unit={self.P_unit}, rho={self.rho}"
return f"{type(self).__name__}({a})"
[docs]
def output_farm_vars(self, algo):
"""
The variables which are being modified by the model.
Parameters
----------
algo: foxes.core.Algorithm
The calculation algorithm
Returns
-------
output_vars: list of str
The output variable names
"""
return [FV.P, FV.CT]
[docs]
def load_data(self, algo, verbosity=0):
"""
Load and/or create all model data that is subject to chunking.
Such data should not be stored under self, for memory reasons. The
data returned here will automatically be chunked and then provided
as part of the mdata object during calculations.
Parameters
----------
algo: foxes.core.Algorithm
The calculation algorithm
verbosity: int
The verbosity level, 0 = silent
Returns
-------
idata: dict
The dict has exactly two entries: `data_vars`,
a dict with entries `name_str -> (dim_tuple, data_ndarray)`;
and `coords`, a dict with entries `dim_name_str -> dim_array`
"""
# read power curve:
if isinstance(self.source_P, pd.DataFrame):
data = self.source_P
else:
fpath = algo.dbook.get_file_path(PCTCURVE, self.source_P, check_raw=True)
pars = {"index_col": 0}
pars.update(self.rpars_P)
data = PandasFileHelper.read_file(fpath, **pars)
data.sort_index(inplace=True)
data.columns = data.columns.astype(FC.DTYPE)
self._ws_P = data.index.to_numpy(FC.DTYPE)
self._ti_P = np.sort(data.columns.to_numpy())
self._P = data[self._ti_P].to_numpy(FC.DTYPE)
# read ct curve:
if isinstance(self.source_ct, pd.DataFrame):
data = self.source_ct
else:
fpath = algo.dbook.get_file_path(PCTCURVE, self.source_ct, check_raw=True)
pars = {"index_col": 0}
pars.update(self.rpars_ct)
data = PandasFileHelper.read_file(fpath, **pars)
data.sort_index(inplace=True)
data.columns = data.columns.astype(FC.DTYPE)
self._ws_ct = data.index.to_numpy(FC.DTYPE)
self._ti_ct = np.sort(data.columns.to_numpy())
self._ct = data[self._ti_ct].to_numpy(FC.DTYPE)
return super().load_data(algo, verbosity)
def _bounds_info(self, target, qts):
"""Helper function for printing bounds info"""
print(f"\nBOUNDS INFO FOR TARGET {target}")
WS = self.WSP if target == FV.P else self.WSCT
ws = self._ws_P if target == FV.P else self._ws_ct
ti = self._ti_P if target == FV.P else self._ti_ct
print(f" {WS}: min = {np.min(ws):.4f}, max = {np.max(ws):.4f}")
print(f" {FV.TI}: min = {np.min(ti):.4f}, max = {np.max(ti):.4f}")
print(f"DATA INFO FOR TARGET {target}")
ws = qts[:, 0]
ti = qts[:, 1]
print(f" {WS}: min = {np.min(ws):.4f}, max = {np.max(ws):.4f}")
print(f" {FV.TI}: min = {np.min(ti):.4f}, max = {np.max(ti):.4f}")
print()
[docs]
def calculate(self, algo, mdata, fdata, st_sel):
""" "
The main model calculation.
This function is executed on a single chunk of data,
all computations should be based on numpy arrays.
Parameters
----------
algo: foxes.core.Algorithm
The calculation algorithm
mdata: foxes.core.MData
The model data
fdata: foxes.core.FData
The farm data
st_sel: numpy.ndarray of bool
The state-turbine selection,
shape: (n_states, n_turbines)
Returns
-------
results: dict
The resulting data, keys: output variable str.
Values: numpy.ndarray with shape (n_states, n_turbines)
"""
# calculate P:
st_sel_P = (
st_sel
& (fdata[self.WSP] >= self._ws_P[0])
& (fdata[self.WSP] <= self._ws_P[-1])
)
st_sel_P0 = st_sel & ~st_sel_P
if np.any(st_sel_P0):
fdata[FV.P][st_sel_P0] = 0
if np.any(st_sel_P):
# prepare interpolation:
n_sel = np.sum(st_sel_P)
qts = np.zeros((n_sel, 2), dtype=FC.DTYPE) # ws, ti
qts[:, 0] = fdata[self.WSP][st_sel_P]
qts[:, 1] = fdata[FV.TI][st_sel_P]
# apply air density correction:
if self.rho is not None:
# correct wind speed by air density, such
# that in the partial load region the
# correct value is reconstructed:
rho = fdata[FV.RHO][st_sel]
qts[:, 0] *= (self.rho / rho) ** (1.0 / 3.0)
del rho
# apply yaw corrections:
if FV.YAWM in fdata and self.p_P is not None:
# calculate corrected wind speed wsc,
# gives ws**3 * cos**p_P in partial load region
# and smoothly deals with full load region:
yawm = fdata[FV.YAWM][st_sel_P]
if np.any(np.isnan(yawm)):
raise ValueError(
f"{self.name}: Found NaN values for variable '{FV.YAWM}'. Maybe change order in turbine_models?"
)
cosm = np.cos(yawm / 180 * np.pi)
qts[:, 0] *= (cosm**self.p_P) ** (1.0 / 3.0)
del yawm, cosm
# run interpolation:
try:
fdata[FV.P][st_sel_P] = interpn(
(self._ws_P, self._ti_P), self._P, qts, **self.ipars_P
)
except ValueError as e:
self._bounds_info(FV.P, qts)
raise e
del st_sel_P, st_sel_P0
# calculate ct:
st_sel_ct = (
st_sel
& (fdata[self.WSCT] >= self._ws_P[0])
& (fdata[self.WSCT] <= self._ws_P[-1])
)
st_sel_ct0 = st_sel & ~st_sel_ct
if np.any(st_sel_ct0):
fdata[FV.CT][st_sel_ct0] = 0
if np.any(st_sel_ct):
# prepare interpolation:
n_sel = np.sum(st_sel_ct)
qts = np.zeros((n_sel, 2), dtype=FC.DTYPE) # ws, ti
qts[:, 0] = fdata[self.WSP][st_sel_ct]
qts[:, 1] = fdata[FV.TI][st_sel_ct]
# apply air density correction:
if self.rho is not None:
# correct wind speed by air density, such
# that in the partial load region the
# correct value is reconstructed:
rho = fdata[FV.RHO][st_sel]
qts[:, 0] *= (self.rho / rho) ** 0.5
del rho
# apply yaw corrections:
if FV.YAWM in fdata and self.p_ct is not None:
# calculate corrected wind speed wsc,
# gives ws**3 * cos**p_P in partial load region
# and smoothly deals with full load region:
yawm = fdata[FV.YAWM][st_sel_ct]
if np.any(np.isnan(yawm)):
raise ValueError(
f"{self.name}: Found NaN values for variable '{FV.YAWM}'. Maybe change order in turbine_models?"
)
cosm = np.cos(yawm / 180 * np.pi)
qts[:, 0] *= (cosm**self.p_ct) ** 0.5
del yawm, cosm
# run interpolation:
try:
fdata[FV.CT][st_sel_ct] = interpn(
(self._ws_ct, self._ti_ct), self._ct, qts, **self.ipars_ct
)
except ValueError as e:
self._bounds_info(FV.CT, qts)
raise e
return {v: fdata[v] for v in self.output_farm_vars(algo)}
[docs]
def finalize(self, algo, verbosity=0):
"""
Finalizes the model.
Parameters
----------
algo: foxes.core.Algorithm
The calculation algorithm
verbosity: int
The verbosity level
"""
del self._ws_P, self._ti_P, self._ws_ct, self._ti_ct
self._P = None
self._ct = None
super().finalize(algo, verbosity)