def test_baseline_with_temp(self):
b = Loadshape(self.get_kw_data_filepath(),
self.get_temp_data_filepath(),
timezone='America/Los_Angeles',
log_level=30)
prediction = b.baseline()
assert len(prediction.data()) > 0
def test_baseline_with_temp(self):
b = Loadshape(self.get_kw_data_filepath(),
self.get_temp_data_filepath(),
timezone='America/Los_Angeles',
log_level=30)
prediction = b.baseline()
assert len(prediction.data()) > 0
def process_request(self, arg_set):
inst_args = ["load_data", "temp_data", "timezone", "temp_units", "sq_ft"]
inst_args = {a_name: arg_set[a_name] for a_name in inst_args if arg_set.get(a_name)}
baseline_args = ["start_at", "end_at", "weighting_days", "modeling_interval", "step_size"]
baseline_args = {a_name: arg_set[a_name] for a_name in baseline_args if arg_set.get(a_name)}
ls = Loadshape(**inst_args)
baseline_series = ls.baseline(**baseline_args)
response = {"baseline": baseline_series.data(), "error_stats": ls.error_stats}
return response
def process_request(self, arg_set):
inst_args = ["load_data", "temp_data", "timezone", "temp_units", "sq_ft"]
inst_args = { a_name: arg_set[a_name] for a_name in inst_args if arg_set.get(a_name)}
baseline_args = ["start_at", "end_at", "weighting_days", "modeling_interval", "step_size"]
baseline_args = { a_name: arg_set[a_name] for a_name in baseline_args if arg_set.get(a_name)}
ls = Loadshape(**inst_args)
baseline_series = ls.baseline(**baseline_args)
response = {
"baseline": baseline_series.data(),
"error_stats": ls.error_stats
}
return response
load_data=LOAD_DATA,
temp_data=TEMP_DATA,
# tariff_schedule=tariff_schedule
timezone='America/Los_Angeles',
temp_units="F",
sq_ft=BUILDING_SQ_FT)
# ----- add exclusions as necessary ----- #
my_load_shape.add_exclusion("2013-09-23 00:00:00", "2013-09-24 00:00:00")
my_load_shape.add_exclusion("2013-09-27 00:00:00", "2013-09-28 00:00:00")
my_load_shape.add_named_exclusion("US_HOLIDAYS")
# ----- generate a 7 day baseline ----- #
seven_day_baseline = my_load_shape.baseline(start_at=WEEK_START,
end_at=WEEK_END,
weighting_days=14,
modeling_interval=900,
step_size=900)
# ----- assemble a payload summarizng the seven day baseline ----- #
out = {"power_data": {}}
out["building"] = BUILDING_NAME
out["baseline_start_at"] = WEEK_START
out["baseline_end_at"] = WEEK_END
out["error_stats"] = my_load_shape.error_stats
out["power_data"]["actual"] = my_load_shape.actual_data(WEEK_START, WEEK_END)
out["power_data"]["baseline"] = my_load_shape.baseline_data(
WEEK_START, WEEK_END)
# ----- write output to file ----- #
file_name = path.join(EXAMPLES_DIR, "output",
# ----- add exclusions as necessary ----- #
my_load_shape.add_exclusion("2013-09-23 00:00:00", "2013-09-24 00:00:00")
my_load_shape.add_exclusion("2013-09-27 00:00:00", "2013-09-28 00:00:00")
my_load_shape.add_named_exclusion("US_HOLIDAYS")
# ----- add tariff to enable cost calculations ----- #
tariff = Tariff(tariff_file=TARIFF, timezone='America/Los_Angeles')
tariff.add_dr_period("2013-09-23 14:00:00", "2013-09-23 16:00:00")
tariff.add_dr_period("2013-09-27 14:00:00", "2013-09-27 16:15:00")
my_load_shape.set_tariff(tariff)
# ----- build the baseline to use as a reference for performance ----- #
event_baseline = my_load_shape.baseline(weighting_days=14,
modeling_interval=900,
step_size=900)
# ----- calculate the performance summary for the event period ----- #
event_performance = my_load_shape.event_performance(DR_EVENT_START, DR_EVENT_END)
# ----- calculate the performance summary for the whole day ----- #
event_day_performance = my_load_shape.event_performance(DR_EVENT_DAY_START, DR_EVENT_DAY_END)
out = { "power_data": {} }
out["name"] = "DR Event - %s" % DR_EVENT_DAY_START
out["building"] = BUILDING_NAME
out["event_start_at"] = DR_EVENT_START
out["event_end_at"] = DR_EVENT_END
out["dr_event_stats"] = event_performance
out["dr_event_day_stats"] = event_day_performance
outfile.close()
# ----- build loadshape object ----- #
my_load_shape = Loadshape(load_data=LOAD_DATA, temp_data=TEMP_DATA,
# tariff_schedule=tariff_schedule
timezone='America/Los_Angeles',
temp_units="F", sq_ft=BUILDING_SQ_FT)
# ----- add exclusions as necessary ----- #
my_load_shape.add_exclusion("2013-09-23 00:00:00", "2013-09-24 00:00:00")
my_load_shape.add_exclusion("2013-09-27 00:00:00", "2013-09-28 00:00:00")
my_load_shape.add_named_exclusion("US_HOLIDAYS")
# ----- generate the appropriate baseline ----- #
baseline = my_load_shape.baseline(weighting_days=CUMULATIVE_SUM_WEIGHTING_DAYS,
modeling_interval=900,
step_size=900)
# ----- set up the cumulative sum dates ----- #
tz = utils.get_timezone('America/Los_Angeles')
start_at = utils.str_to_datetime(CUMULATIVE_SUM_START, tz)
end_at = utils.int_to_datetime(my_load_shape.baseline_series.end_at(), tz)
# ----- calculate long term event performance and cumulative sum series ----- #
event_performance = my_load_shape.event_performance(start_at, end_at)
cumulative_sum_series = my_load_shape.cumulative_sum(start_at, end_at)
# ----- assemble a payload summarizng the cumulative sum ----- #
out = {}
out["building"] = BUILDING_NAME
out["event_name"] = CUMULATIVE_SUM_NAME
my_load_shape = Loadshape(
load_data=LOAD_DATA,
temp_data=TEMP_DATA,
# tariff_schedule=tariff_schedule
timezone='America/Los_Angeles',
temp_units="F",
sq_ft=BUILDING_SQ_FT)
# ----- add exclusions as necessary ----- #
my_load_shape.add_exclusion("2013-09-23 00:00:00", "2013-09-24 00:00:00")
my_load_shape.add_exclusion("2013-09-27 00:00:00", "2013-09-28 00:00:00")
my_load_shape.add_named_exclusion("US_HOLIDAYS")
# ----- generate the appropriate baseline ----- #
baseline = my_load_shape.baseline(weighting_days=CUMULATIVE_SUM_WEIGHTING_DAYS,
modeling_interval=900,
step_size=900)
# ----- set up the cumulative sum dates ----- #
tz = utils.get_timezone('America/Los_Angeles')
start_at = utils.str_to_datetime(CUMULATIVE_SUM_START, tz)
end_at = utils.int_to_datetime(my_load_shape.baseline_series.end_at(), tz)
# ----- calculate long term event performance and cumulative sum series ----- #
event_performance = my_load_shape.event_performance(start_at, end_at)
cumulative_sum_series = my_load_shape.cumulative_sum(start_at, end_at)
# ----- assemble a payload summarizng the cumulative sum ----- #
out = {}
out["building"] = BUILDING_NAME
out["event_name"] = CUMULATIVE_SUM_NAME
load_data=LOAD_DATA,
temp_data=TEMP_DATA,
# tariff_schedule=tariff_schedule
timezone="America/Los_Angeles",
temp_units="F",
sq_ft=BUILDING_SQ_FT,
)
# ----- add exclusions as necessary ----- #
my_load_shape.add_exclusion("2013-09-23 00:00:00", "2013-09-24 00:00:00")
my_load_shape.add_exclusion("2013-09-27 00:00:00", "2013-09-28 00:00:00")
my_load_shape.add_named_exclusion("US_HOLIDAYS")
# ----- generate a 7 day baseline ----- #
seven_day_baseline = my_load_shape.baseline(
start_at=WEEK_START, end_at=WEEK_END, weighting_days=14, modeling_interval=900, step_size=900
)
# ----- assemble a payload summarizng the seven day baseline ----- #
out = {"power_data": {}}
out["building"] = BUILDING_NAME
out["baseline_start_at"] = WEEK_START
out["baseline_end_at"] = WEEK_END
out["error_stats"] = my_load_shape.error_stats
out["power_data"]["actual"] = my_load_shape.actual_data(WEEK_START, WEEK_END)
out["power_data"]["baseline"] = my_load_shape.baseline_data(WEEK_START, WEEK_END)
# ----- write output to file ----- #
file_name = path.join(EXAMPLES_DIR, "output", "seven-day-baseline-example.json")
write_json(data=out, file_name=file_name)
