def test_tcell_filter(self):
filtered = tcell_filter(self.tcell,
temperature_cell_low=-50,
temperature_cell_high=110)
# Expected high and low tcell cutoffs to be non-inclusive.
expected_result = np.array([False, True, True, True, False])
self.assertListEqual(filtered.tolist(), expected_result.tolist())
def test_tcell_filter():
''' Unit tests for cell temperature filter.'''
tcell = np.array([-50, -49, 0, 109, 110])
filtered = tcell_filter(tcell,
temperature_cell_low=-50,
temperature_cell_high=110)
# Expected high and low tcell cutoffs to be non-inclusive.
expected_result = np.array([False, True, True, True, False])
assert filtered.tolist() == expected_result.tolist()
# Plot the normalized power time series
fig, ax = plt.subplots()
ax.plot(normalized.index, normalized, 'o', alpha=0.05)
ax.set_ylim(0, 2)
fig.autofmt_xdate()
ax.set_ylabel('Normalized energy')
# 2: Filter
# Data filtering is used to exclude data points that represent invalid data, create bias in the analysis, or introduce significant noise.
# It can also be useful to remove outages and outliers. Sometimes outages appear as low but non-zero yield. Automatic functions for this are not yet included in `rdtools`. Such filters should be implimented by the analyst if needed.
# Calculate a collection of boolean masks that can be used
# to filter the time series
nz_mask = (df['normalized'] > 0)
poa_mask = rdtools.poa_filter(df['poa'])
tcell_mask = rdtools.tcell_filter(df['Tcell'])
clip_mask = rdtools.clip_filter(df['power'])
# filter the time series and keep only the columns needed for the
# remaining steps
filtered = df[nz_mask & poa_mask & tcell_mask & clip_mask]
filtered = filtered[['insolation', 'normalized']]
fig, ax = plt.subplots()
ax.plot(filtered.index, filtered.normalized, 'o', alpha=0.05)
ax.set_ylim(0, 2)
fig.autofmt_xdate()
ax.set_ylabel('Normalized energy')
# 3: Aggregate
# Data is aggregated with an irradiance weighted average. This can be useful, for example with daily aggregation, to reduce the impact of high-error data points in the morning and evening.
def average_degradation():
file_name = config['real_time_data']
df = pd.read_csv(file_name)
df = df.rename(
columns={
u'12 BP Solar - Active Power (kW)': 'power',
u'12 BP Solar - Wind Speed (m/s)': 'wind',
u'12 BP Solar - Weather Temperature Celsius (\xb0C)': 'Tamb',
u'12 BP Solar - Global Horizontal Radiation (W/m\xb2)': 'ghi',
u'12 BP Solar - Diffuse Horizontal Radiation (W/m\xb2)': 'dhi'
})
# Specify the Metadata
meta = config
df.index = pd.to_datetime(df.Timestamp)
# TZ is required for irradiance transposition
df.index = df.index.tz_localize(meta['timezone'], ambiguous='infer')
# Explicitly trim the dates so that runs of this example notebook
# are comparable when the source dataset has been downloaded at different times
df = df[config['start_date']:config['end_date']]
# Change power from kilowatts to watts
df['power'] = df.power * 1000.0
# There is some missing data, but we can infer the frequency from the first several data points
freq = pd.infer_freq(df.index[:10])
# And then set the frequency of the dataframe
df = df.resample(freq).median()
# Calculate energy yield in Wh
df['energy'] = df.power * pd.to_timedelta(
df.power.index.freq).total_seconds() / (3600.0)
# Calculate POA irradiance from DHI, GHI inputs
loc = pvlib.location.Location(meta['latitude'],
meta['longitude'],
tz=meta['timezone'])
sun = loc.get_solarposition(df.index)
# calculate the POA irradiance
sky = pvlib.irradiance.isotropic(meta['tilt'], df.dhi)
df['dni'] = (df.ghi - df.dhi) / np.cos(np.deg2rad(sun.zenith))
beam = pvlib.irradiance.beam_component(meta['tilt'], meta['azimuth'],
sun.zenith, sun.azimuth, df.dni)
df['poa'] = beam + sky
# Calculate cell temperature
df_temp = pvlib.pvsystem.sapm_celltemp(df.poa,
df.wind,
df.Tamb,
model=meta['temp_model'])
df['Tcell'] = df_temp.temp_cell
# Specify the keywords for the pvwatts model
pvwatts_kws = {
"poa_global": df.poa,
"P_ref": meta['pdc'],
"T_cell": df.Tcell,
"G_ref": 1000, # reference irradiance
"T_ref": 25, # reference temperature
"gamma_pdc": meta['tempco']
}
# Calculate the normaliztion, the function also returns the relevant insolation for
# each point in the normalized PV energy timeseries
normalized, insolation = rdtools.normalize_with_pvwatts(
df.energy, pvwatts_kws)
df['normalized'] = normalized
df['insolation'] = insolation
# Calculate a collection of boolean masks that can be used
# to filter the time series
nz_mask = (df['normalized'] > 0)
poa_mask = rdtools.poa_filter(df['poa'])
tcell_mask = rdtools.tcell_filter(df['Tcell'])
clip_mask = rdtools.clip_filter(df['power'])
# filter the time series and keep only the columns needed for the
# remaining steps
filtered = df[nz_mask & poa_mask & tcell_mask & clip_mask]
filtered = filtered[['insolation', 'normalized']]
daily = rdtools.aggregation_insol(filtered.normalized,
filtered.insolation,
frequency='D')
# Calculate the degradation rate using the YoY method
yoy_rd, yoy_ci, yoy_info = rdtools.degradation_year_on_year(
daily, confidence_level=68.2)
# yoy_rd is mean degradation rate per year.
return (df, 0.01 * yoy_rd)