def test_compute_gross_energy(self):
# Set some test values
net = np.array([[1, 1, 1],
[0, 1, 1],
[1, 1, 1]])
avail = np.array([[0.05, 0.08, 0.2],
[0.05, .9, 0.2],
[0.05, .9, 0]])
curt = np.array([[0.05, 0.05, 0.05],
[0.05, 0.2, 0.05],
[0.05, -0.1, 0.05]])
# Make sure different combinations of 'frac' and 'energy' based measurements work out
nptest.assert_almost_equal(
unit_conversion.compute_gross_energy(net[0, :], avail[0, :], curt[0, :], 'frac', 'frac'),
[1.1111, 1.1494, 1.3333],
decimal=4)
nptest.assert_almost_equal(
unit_conversion.compute_gross_energy(net[0, :], avail[0, :], curt[0, :], 'energy', 'frac'),
[1.1026, 1.1326, 1.2526],
decimal=4)
nptest.assert_almost_equal(
unit_conversion.compute_gross_energy(net[0, :], avail[0, :], curt[0, :], 'frac', 'energy'),
[1.1026, 1.1370, 1.3000],
decimal=4)
nptest.assert_almost_equal(
unit_conversion.compute_gross_energy(net[0, :], avail[0, :], curt[0, :], 'energy', 'energy'),
[1.1000, 1.1300, 1.2500],
decimal=4)
# Make sure exceptions are thrown when bad input data is identified
def func(): # Function to return exception
unit_conversion.compute_gross_energy(net[1, :], avail[1, :], curt[1, :], 'frac', 'frac')
with self.assertRaises(Exception):
func()
def func(): # Function to return exception
unit_conversion.compute_gross_energy(net[2, :], avail[2, :], curt[2, :], 'frac', 'frac')
with self.assertRaises(Exception):
func()
def process_loss_estimates(self):
"""
Append availability and curtailment losses to monthly data frame
Args:
(None)
Returns:
(None)
"""
df = getattr(self._plant, 'curtail').df
curt_monthly = np.divide(df.resample('MS')[['availability_kwh', 'curtailment_kwh']].sum(),
1e6) # Get sum of avail and curt losses in GWh
curt_monthly.rename(columns={'availability_kwh': 'availability_gwh', 'curtailment_kwh': 'curtailment_gwh'},
inplace=True)
# Merge with revenue meter monthly data
self._monthly.df = self._monthly.df.join(curt_monthly)
# Add gross energy field
self._monthly.df['gross_energy_gwh'] = un.compute_gross_energy(self._monthly.df['energy_gwh'],
self._monthly.df['availability_gwh'],
self._monthly.df['curtailment_gwh'], 'energy',
'energy')
# Calculate percentage-based losses
self._monthly.df['availability_pct'] = np.divide(self._monthly.df['availability_gwh'],
self._monthly.df['gross_energy_gwh'])
self._monthly.df['curtailment_pct'] = np.divide(self._monthly.df['curtailment_gwh'],
self._monthly.df['gross_energy_gwh'])
self._monthly.df['avail_nan_perc'] = df.resample('MS')['availability_kwh'].apply(
tm.percent_nan) # Get percentage of 10-min meter data that were NaN when summing to monthly
self._monthly.df['curt_nan_perc'] = df.resample('MS')['curtailment_kwh'].apply(
tm.percent_nan) # Get percentage of 10-min meter data that were NaN when summing to monthly
self._monthly.df['nan_flag'] = False # Set flag to false by default
self._monthly.df.loc[(self._monthly.df['energy_nan_perc'] > self.uncertainty_nan_energy) |
(self._monthly.df['avail_nan_perc'] > self.uncertainty_nan_energy) |
(self._monthly.df['curt_nan_perc'] > self.uncertainty_nan_energy), 'nan_flag'] \
= True # If more than 1% of data are NaN, set flag to True
# By default, assume all reported losses are representative of long-term operational
self._monthly.df['availability_typical'] = True
self._monthly.df['curtailment_typical'] = True
# By default, assume combined availability and curtailment losses are below the threshold to be considered valid
self._monthly.df['combined_loss_valid'] = True
def func(): # Function to return exception
unit_conversion.compute_gross_energy(net[2, :], avail[2, :], curt[2, :], 'frac', 'frac')
def prepare(self):
"""
Do all loading and preparation of the data for this plant.
"""
# Set time frequencies of data
self._meter_freq = '1MS' # Monthly meter data
self._curtail_freq = '1MS' # Monthly curtailment data
######################
# MONTHLY METER DATA #
######################
self._meter.load(self._path, 'plant_data', 'csv') # Load monthly data
# Get 'time' field in datetime format
self._meter.df['time'] = pd.to_datetime(self._meter.df['year_month'], format='%Y %m ')
self._meter.df.set_index('time', inplace=True, drop=False) # Set datetime as index
# Rename variables
self._meter.df['energy_kwh'] = self._meter.df['net_energy_mwh']*1000. # convert MWh to kWh
# Remove the fields we are not yet interested in
self._meter.df.drop(['year_month', 'net_energy_mwh', 'availability_pct', 'curtailment_pct'], axis=1, inplace=True)
#############################################
# MONTHLY CURTAILMENT AND AVAILABILITY DATA #
#############################################
self._curtail.load(self._path, 'plant_data', 'csv') # Load monthly data
# Get 'time' field in datetime format
self._curtail.df['time'] = pd.to_datetime(self._curtail.df['year_month'], format='%Y %m')
self._curtail.df.set_index('time', inplace=True, drop=False) # Set datetime as index
# Get losses in energy units
gross_energy = un.compute_gross_energy(self._curtail.df['net_energy_mwh'],
self._curtail.df['availability_pct'],
self._curtail.df['curtailment_pct'], 'frac', 'frac')
self._curtail.df['curtailment_kwh'] = self._curtail.df['curtailment_pct']*gross_energy*1000
self._curtail.df['availability_kwh'] = self._curtail.df['availability_pct']*gross_energy*1000
# Remove the fields we are not yet interested in
self._curtail.df.drop(['net_energy_mwh', 'year_month'], axis=1, inplace=True)
###################
# REANALYSIS DATA #
###################
# merra2
self._reanalysis._product['merra2'].load(self._path,"merra2_data","csv")
self._reanalysis._product['merra2'].rename_columns({"time":"datetime",
"windspeed_ms": "ws_50m",
"rho_kgm-3": "dens_50m",
"winddirection_deg":"wd_50m"})
self._reanalysis._product['merra2'].normalize_time_to_datetime("%Y-%m-%d %H:%M:%S")
self._reanalysis._product['merra2'].df.set_index('time',inplace=True,drop=False)
# ncep2
self._reanalysis._product['ncep2'].load(self._path,"ncep2_data","csv")
self._reanalysis._product['ncep2'].rename_columns({"time":"datetime",
"windspeed_ms": "ws_10m",
"rho_kgm-3": "dens_10m",
"winddirection_deg":"wd_10m"})
self._reanalysis._product['ncep2'].normalize_time_to_datetime("%Y%m%d %H%M")
self._reanalysis._product['ncep2'].df.set_index('time',inplace=True,drop=False)
# erai
self._reanalysis._product['erai'].load(self._path,"erai_data","csv")
self._reanalysis._product['erai'].rename_columns({"time":"datetime",
"windspeed_ms": "ws_58",
"rho_kgm-3": "dens_58",
"winddirection_deg":"wd_58"})
self._reanalysis._product['erai'].normalize_time_to_datetime("%Y-%m-%d %H:%M:%S")
self._reanalysis._product['erai'].df.set_index('time',inplace=True,drop=False)
