def follow_generated_consumption_profile(
start: datetime,
end: datetime,
resolution: timedelta,
max_capacity: float,
dispatch_factor: float = None,
profile: Series = None,
) -> DataFrame:
"""Can be used to model a device with the artificially generatred consumption or production profiles."""
if profile is None:
full_year_monthly_profile = (
cos(2 * pi / 12 * linspace(0, 11, 12)) * 50 + 100) * 0.75
full_year_monthly_load_profile = el.make_timeseries(
full_year_monthly_profile)
dummy_index = DatetimeIndex(
start=datetime(year=2018, month=1, day=1),
end=datetime(year=2019, month=1, day=1, hour=0),
freq="H",
)
dummy_index = dummy_index.drop(dummy_index[-1])
weight = uniform(0.5, 0.8)
daily_load_working = el.gen_daily_stoch_el()
daily_load_non_working = el.gen_daily_stoch_el()
profile = el.gen_load_from_daily_monthly(
full_year_monthly_load_profile,
daily_load_working,
daily_load_non_working,
weight,
)
profile_noized = el.add_noise(profile, 3, 0.25)
profile_noized.index = dummy_index
if int(resolution.seconds / 3600) == 1.0:
pass
else:
profile_noized = (profile_noized.drop(
index=profile_noized.index[-1]).resample(rule="15T").mean())
profile_noized = profile_noized.interpolate(method="linear").drop(
index=profile_noized.index[-1])
profile_noized = profile_noized + abs(
normal(0, 0.05, [len(profile_noized.index)]))
profile = profile_noized * max_capacity
else:
pass
df = limited_consumption_profile(start=start,
end=end,
resolution=resolution,
capacity=max_capacity)
df["derivative equals"] = profile.loc[start:end - resolution].values
if dispatch_factor is not None:
if dispatch_factor == 0:
pass
if dispatch_factor == 1:
df["derivative max"] = df["derivative equals"]
df["derivative equals"] = nan
else:
df["derivative max"] = df["derivative equals"]
df["derivative equals"] = df["derivative equals"] * (
1 - dispatch_factor)
# print(df)
# print("consumption")
return df
def follow_generated_production_profile(
start: datetime,
end: datetime,
resolution: timedelta,
max_capacity: float,
dispatch_factor: float = None,
profile: Series = None,
) -> DataFrame:
if profile is None:
dummy_index = DatetimeIndex(
start=datetime(year=2018, month=1, day=1),
end=datetime(year=2019, month=1, day=1, hour=0),
freq="15T",
)
dummy_index = dummy_index.drop(dummy_index[-96])
mu, sigma = 1, 0.01
s = normal(mu, sigma, 5000)
count, bins, ignored = hist(s, 72, normed=False)
daily_profile = count / count.max()
yearly_curve = (
-cos(2 * pi / 35040 * linspace(0, 35039, 35040) + 0.2) * 50 +
100) * 0.75
stacks = []
for i in range(1, 366):
s = Series(index=RangeIndex(start=1, stop=97, step=1), data=0)
s[12:84] = el.add_noise(daily_profile,
mode=3,
st=0.05,
Lmin=0,
r=0.01).values
stacks.append(list(s.values))
stacks = [item for sublist in stacks for item in sublist]
full_year_profile = stacks * yearly_curve
full_year_profile = full_year_profile / full_year_profile.max()
profile = Series(data=full_year_profile, index=dummy_index)
profile *= max_capacity
else:
pass
df = limited_production_profile(start=start,
end=end,
resolution=resolution,
capacity=max_capacity)
df["derivative equals"] = -profile.loc[start:end - resolution].values
if dispatch_factor is not None:
if dispatch_factor == 0:
pass
if dispatch_factor == 1:
df["derivative min"] = df["derivative equals"]
df["derivative equals"] = nan
else:
df["derivative min"] = df["derivative equals"]
df["derivative equals"] = df["derivative equals"] * (
1 - dispatch_factor)
# print("generation")
# print(df)
return df
