def get_hourly_model(data):
logger.info('get_hourly_model: ...')
# create a design matrix for occupancy and segmentation
logger.info('get_hourly_model: creating baseline_design_matrix ...')
preliminary_design_matrix = (
eemeter.create_caltrack_hourly_preliminary_design_matrix(
data['baseline_meter_data'],
data['temperature_data'],
))
# build 12 monthly models - each step from now on operates on each segment
logger.info('get_hourly_model: creating segment_time_series ...')
segmentation = eemeter.segment_time_series(preliminary_design_matrix.index,
'three_month_weighted')
# assign an occupancy status to each hour of the week (0-167)
logger.info('get_hourly_model: creating occupancy_lookup ...')
occupancy_lookup = eemeter.estimate_hour_of_week_occupancy(
preliminary_design_matrix,
segmentation=segmentation,
)
# assign temperatures to bins
logger.info('get_hourly_model: creating temperature_bins ...')
occupied_temperature_bins, unoccupied_temperature_bins = eemeter.fit_temperature_bins(
preliminary_design_matrix,
segmentation=segmentation,
occupancy_lookup=occupancy_lookup,
)
# build a design matrix for each monthly segment
logger.info('get_hourly_model: creating segmented_design_matrices ...')
segmented_design_matrices = (
eemeter.create_caltrack_hourly_segmented_design_matrices(
preliminary_design_matrix,
segmentation,
occupancy_lookup,
occupied_temperature_bins,
unoccupied_temperature_bins,
))
# build a CalTRACK hourly model
logger.info('get_hourly_model: building CalTRACK model ...')
baseline_model = eemeter.fit_caltrack_hourly_model(
segmented_design_matrices,
occupancy_lookup,
occupied_temperature_bins,
unoccupied_temperature_bins,
)
logger.info('get_hourly_model: DONE')
return baseline_model
def test_json_hourly():
meter_data, temperature_data, sample_metadata = (
eemeter.load_sample("il-electricity-cdd-hdd-hourly"))
blackout_start_date = sample_metadata["blackout_start_date"]
blackout_end_date = sample_metadata["blackout_end_date"]
# get meter data suitable for fitting a baseline model
baseline_meter_data, warnings = eemeter.get_baseline_data(
meter_data, end=blackout_start_date, max_days=365)
# create a design matrix for occupancy and segmentation
preliminary_design_matrix = (
eemeter.create_caltrack_hourly_preliminary_design_matrix(
baseline_meter_data,
temperature_data,
))
# build 12 monthly models - each step from now on operates on each segment
segmentation = eemeter.segment_time_series(preliminary_design_matrix.index,
'three_month_weighted')
# assign an occupancy status to each hour of the week (0-167)
occupancy_lookup = eemeter.estimate_hour_of_week_occupancy(
preliminary_design_matrix,
segmentation=segmentation,
)
# assign temperatures to bins
temperature_bins = eemeter.fit_temperature_bins(
preliminary_design_matrix,
segmentation=segmentation,
)
# build a design matrix for each monthly segment
segmented_design_matrices = (
eemeter.create_caltrack_hourly_segmented_design_matrices(
preliminary_design_matrix,
segmentation,
occupancy_lookup,
temperature_bins,
))
# build a CalTRACK hourly model
baseline_model = eemeter.fit_caltrack_hourly_model(
segmented_design_matrices,
occupancy_lookup,
temperature_bins,
)
# get a year of reporting period data
reporting_meter_data, warnings = eemeter.get_reporting_data(
meter_data, start=blackout_end_date, max_days=365)
# compute metered savings
metered_savings_dataframe, error_bands = eemeter.metered_savings(
baseline_model,
reporting_meter_data,
temperature_data,
with_disaggregated=True)
# total metered savings
total_metered_savings = metered_savings_dataframe.metered_savings.sum()
# test JSON
json_str = json.dumps(baseline_model.json())
m = eemeter.CalTRACKHourlyModelResults.from_json(json.loads(json_str))
# compute metered savings from the loaded model
metered_savings_dataframe, error_bands = eemeter.metered_savings(
m, reporting_meter_data, temperature_data, with_disaggregated=True)
# total metered savings
total_metered_savings_2 = metered_savings_dataframe.metered_savings.sum()
assert total_metered_savings == total_metered_savings_2
def eemeter_baseline_ami(file, temperature, install_start, install_end):
"""
This method uses linear regression to create hourly load profile to
serve as the counterfactual for the period where the new measure has
been installed. The two key parts of this model are the temperature
and occupancy binning.
CalTRACK refers to a standardized model used in california to measure
energy efficiency savings from various measures (i.e. LED lightbulbs,
efficient appliances, weatherization, etc.). In this case it will be
used to model energy growth instead of savings.
"""
ami_data = eemeter.meter_data_from_csv(file, freq='hourly')
# get meter data suitable for fitting a baseline model
baseline_meter_data, warnings = eemeter.get_baseline_data(
ami_data, end=install_start, max_days=365)
# create design matrix for occupancy and segmentation
preliminary_design_matrix = (
eemeter.create_caltrack_hourly_preliminary_design_matrix(
baseline_meter_data,
temperature,
))
# build matrix with weights for monthly models of:
# 0.5 = prior month
# 1.0 = current month
# 0.5 = post month
segmentation = eemeter.segment_time_series(
preliminary_design_matrix.index,
'three_month_weighted' # using 3 month weighted approach
)
# assign an occupancy status to each hour of the week (0-167)
occupancy_lookup = eemeter.estimate_hour_of_week_occupancy(
preliminary_design_matrix,
segmentation=segmentation,
)
# assign temperatures to bins
temperature_bins = eemeter.fit_temperature_bins(
preliminary_design_matrix,
segmentation=segmentation,
)
# build a desgin matrix for each monthly segment
segmented_design_matrices = (
eemeter.create_caltrack_hourly_segmented_design_matrices(
preliminary_design_matrix,
segmentation,
occupancy_lookup,
temperature_bins,
))
# build a CalTRACK hourly model
baseline_model = eemeter.fit_caltrack_hourly_model(
segmented_design_matrices, occupancy_lookup, temperature_bins)
# get a year of post installation reporting data
reporting_meter_data, warnings = eemeter.get_reporting_data(
ami_data, start=install_end, max_days=365)
# compute metered load growth for the year of reporitng period
metered_growth_dataframe, error_bands = eemeter.metered_savings(
baseline_model,
reporting_meter_data,
temperature,
with_disaggregated=True)
metered_growth_dataframe['temp'] = temperature # append temperature
# change signs for load growth
metered_growth_dataframe['metered_savings'] = metered_growth_dataframe[
'metered_savings'].apply(lambda x: x * -1)
# totaled load growth
additional_load = metered_growth_dataframe.metered_savings.sum()
# metrics
r_squared_adj_list = []
cvrmse_adj_list = []
# results in a dict
model_results = list(baseline_model.json().values())
results = model_results[6]
for segment, measures in results.items():
for measure, value in measures.items():
if measure == 'r_squared_adj':
r_squared_adj_list.append(value)
if measure == 'cvrmse_adj':
cvrmse_adj_list.append(value)
r_squared_adj = mean(r_squared_adj_list)
cvrmse_adj = mean(cvrmse_adj_list)
metrics = [r_squared_adj, cvrmse_adj, additional_load]
# Return Section
return metered_growth_dataframe, metrics, baseline_model