detquantlib 5.1.2__tar.gz → 5.3.0__tar.gz

{detquantlib-5.1.2 → detquantlib-5.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: detquantlib
-Version: 5.1.2
+Version: 5.3.0
 Summary: An internal library containing functions and classes that can be used across Quant models.
 License-File: LICENSE.txt
 Author: Dynamic Energy Trading

{detquantlib-5.1.2 → detquantlib-5.3.0}/detquantlib/assets/wind_turbine.py

@@ -5,6 +5,7 @@ from scipy.interpolate import CubicSpline
 
 # Internal modules
 from detquantlib.converters import power_to_energy
+from detquantlib.utils import list_to_str
 
 
 class WindTurbine:
@@ -35,6 +36,73 @@ class WindTurbine:
         self.nr_turbines = nr_turbines
         self.power_curve = power_curve
 
+    @staticmethod
+    def list_models() -> list[str]:
+        """Gets the list of available wind turbine models."""
+        return [d for d in WindTurbineModels.__dict__ if not d.startswith("__")]
+
+    @staticmethod
+    def get_model(model_name: str) -> dict:
+        """Gets the model information of a given wind turbine."""
+        try:
+            return getattr(WindTurbineModels, model_name)
+        except AttributeError as e:
+            raise ValueError(
+                f"Invalid model name '{model_name}'. Available models: "
+                f"{list_to_str(WindTurbine.list_models())}."
+            ) from e
+
+    @staticmethod
+    def build_from_model(model_name: str = None, model_info: dict = None, nr_turbines: int = 1):
+        """
+        Builds a WindTurbine object from a known wind turbine model.
+
+        Args:
+            model_name: Wind turbine model name, as given in the WindTurbineModels class. If not
+                provided, argument 'model_info' should be provided.
+            model_info: Wind turbine model information. The dictionary can be taken from the
+                list of models in the WindTurbineModels class. Otherwise, the dictionary should
+                have the same structure as those in the WindTurbineModels class. If not provided,
+                argument 'model_name' should be provided.
+            nr_turbines: Number of turbines in the wind farm
+
+        Returns:
+            WindTurbine object
+
+        Raises:
+            ValueError: Raises an error if both 'model_name' and 'model_info' are missing.
+            ValueError: Raises an error if both 'model_name' and 'model_info' are provided.
+        """
+        # Inputs validation
+        if model_name is None and model_info is None:
+            raise ValueError(
+                "Cannot determine turbine model. Please provide either argument 'model_name' "
+                "or 'model_info'."
+            )
+        elif model_name is not None and model_info is not None:
+            raise ValueError(
+                "Cannot determine turbine model. Please provide either argument 'model_name' "
+                "or 'model_info', but not both."
+            )
+        elif model_name is not None:
+            model_info = WindTurbine.get_model(model_name)
+
+        # Create object
+        obj = WindTurbine(
+            rated_power_mw=model_info["rated_power_mw"],
+            cut_in=model_info["cut_in"],
+            cut_out=model_info["cut_out"],
+            nr_turbines=nr_turbines,
+        )
+
+        # Set power_curve attribute
+        obj.fit_power_curve(
+            wind_speed=[d["wind_speed_(m/s)"] for d in model_info["power_curve"]],
+            power_mw=[d["power(mw)"] for d in model_info["power_curve"]],
+        )
+
+        return obj
+
     def fit_power_curve(
         self, wind_speed: list | np.ndarray | pd.Series, power_mw: list | np.ndarray | pd.Series
     ):
@@ -92,8 +160,8 @@ class WindTurbine:
         # Cubic spline interpolation
         power_mw = self.power_curve(wind_speed)
 
-        # Cap to rated power
-        power_mw = np.minimum(power_mw, self.rated_power_mw)
+        # Bound between 0 and rated power
+        power_mw = np.clip(a=power_mw, a_min=0.0, a_max=self.rated_power_mw)
 
         # Adjust values outside of cut-in - cut-out bounds
         idx = (wind_speed < self.cut_in) | (wind_speed > self.cut_out)
@@ -121,36 +189,6 @@ class WindTurbine:
         )
         return energy_mwh
 
-    @staticmethod
-    def build_from_model(model: dict, nr_turbines: int):
-        """
-        Builds a WindTurbine object from a known wind turbine model.
-
-        Args:
-            model: Wind turbine model. The dictionary can be taken from the list of models in
-                the WindTurbineModels class. Otherwise, the dictionary should have the same
-                structure as those in the WindTurbineModels class.
-            nr_turbines: Number of wind turbines in the wind farm
-
-        Returns:
-            WindTurbine object
-        """
-        # Create object
-        obj = WindTurbine(
-            rated_power_mw=model["rated_power_mw"],
-            cut_in=model["cut_in"],
-            cut_out=model["cut_out"],
-            nr_turbines=nr_turbines,
-        )
-
-        # Set power_curve attribute
-        obj.fit_power_curve(
-            wind_speed=[d["wind_speed_(m/s)"] for d in model["power_curve"]],
-            power_mw=[d["power(mw)"] for d in model["power_curve"]],
-        )
-
-        return obj
-
 
 class WindTurbineModels:
     """

{detquantlib-5.1.2 → detquantlib-5.3.0}/detquantlib/dates/dates.py

@@ -3,10 +3,45 @@ from datetime import datetime, timezone
 from typing import Literal
 
 # Third-party packages
-from dateutil.relativedelta import *
+import pandas as pd
+from dateutil.relativedelta import relativedelta
 
 # Specify functions to expose
-__all__ = ["count_nr_hours", "calc_months_diff", "datetime_to_month_code"]
+__all__ = ["ts_ceil", "ts_floor", "count_nr_hours", "calc_months_diff", "datetime_to_month_code"]
+
+
+def ts_ceil(ts: pd.Timestamp, freq: str) -> pd.Timestamp:
+    """
+    Short wrapper function around the pd.Timestamp object.ceil() method that also supports
+    October DST timestamps (i.e. between 02:00:00+0200 and 03:00:00+0100), because the built-in
+    method raises an error in those cases.
+
+    Args:
+        ts: Timestamp
+        freq: Frequency
+
+    Returns:
+        Ceiled timestamp
+    """
+    local_tz = ts.tz
+    return ts.tz_convert("UTC").ceil(freq).tz_convert(local_tz)
+
+
+def ts_floor(ts: pd.Timestamp, freq: str) -> pd.Timestamp:
+    """
+    Short wrapper function around the pd.Timestamp object.floor() method that also supports
+    October DST timestamps (i.e. between 02:00:00+0200 and 03:00:00+0100), because the built-in
+    method raises an error in those cases.
+
+    Args:
+        ts: Timestamp
+        freq: Frequency
+
+    Returns:
+        Floored timestamp
+    """
+    local_tz = ts.tz
+    return ts.tz_convert("UTC").floor(freq).tz_convert(local_tz)
 
 
 def count_nr_hours(start_date: datetime, end_date: datetime) -> float:

{detquantlib-5.1.2 → detquantlib-5.3.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "detquantlib"
-version = "5.1.2"
+version = "5.3.0"
 description = "An internal library containing functions and classes that can be used across Quant models."
 authors = ["Dynamic Energy Trading"]
 readme = "README.md"