Source code for energy_analysis_toolbox.timeseries.profiles.thresholds.relative
"""Module for computing threshold profiles based on relative deviations.
This module provides tools to calculate profiles that deviate from the mean profile
by a user-defined relative offset. The profiles are designed to scale proportionally
with the baseline consumption pattern derived from historical data.
"""
import pandas as pd
from energy_analysis_toolbox.timeseries.profiles.mean_profile import MeanProfile
[docs]
class RelativeThreshold(MeanProfile):
"""A class which implements a relative deviation from the mean profile."""
def __init__(
self,
offset_relative: float = 0.5,
**kwargs,
) -> None:
"""Return a threshold profile.
The threshold profile is obtained using a user-defined relative variation
from the mean profile built from history.
Parameters
----------
offset_rel : float, optional
Relative difference VS the computed reference to obtain
the threshold profile. Default is 0.5 (profile is 150%
of reference)
"""
self.offset_rel = offset_relative
super().__init__(**kwargs)
[docs]
def compute(
self,
history: pd.Series,
time: pd.Timestamp,
**kwargs,
) -> pd.Series:
"""Return a threshold profile.
The threshold profile is obtained using a user-defined relative variation
from the mean profile built from history.
Parameters
----------
history : pd.Series
Consumption history used to computed the reference
profile.
time : pd.Timestamp
The time at which the profile is of interest. Only
the information about the date is used in the passed
timestamp.
Returns
-------
profile : pd.Series
Profile threshold with same resolution as the history data.
Notes
-----
The profile threshold is obtained as ``(1 + tshd) *`` the mean
profile obtained from history.
"""
reference = super().compute(history, time, **kwargs)
offset = self.offset_rel * reference
offset.index = reference.index
return reference + offset
