Merge pull request #260 from Breakthrough-Energy/daniel/hifld_wind_pr…

…ofiles

feat: generate wind profiles during HIFLD grid-building

prereise/gather/griddata/hifld/data_process/profiles.py

@@ -1,6 +1,10 @@
+import datetime as dt
+
 from prereise.gather.griddata.hifld import const
 from prereise.gather.griddata.hifld.data_access.load import get_eia_form_860
 from prereise.gather.solardata.nsrdb.sam import retrieve_data_individual
+from prereise.gather.winddata.hrrr.calculations import calculate_pout_individual
+from prereise.gather.winddata.hrrr.hrrr import retrieve_data
 
 
 def floatify(x):
@@ -57,3 +61,42 @@ def build_solar(nrel_email, nrel_api_key, solar_plants, **solar_kwargs):
         **solar_kwargs,
     )
     return profiles
+
+
+def build_wind(wind_plants, download_directory, year):
+    """Use plant-level data to build wind profiles.
+
+    :param pandas.DataFrame wind_plants: data frame of wind farms.
+    :param str download_directory: location to download wind speed data to.
+    :param int/str year: year of weather data to use for generating profiles.
+    :return: (*pandas.DataFrame*) -- data frame of normalized power profiles. The index
+        is hourly timestamps for the profile year, the columns are plant IDs, the values
+        are floats.
+    """
+    default_hub_height = 262.467  # (262.467 ft = 80m)
+    # Load raw 'extra' table data, join on plant & generating unit, re-establish index
+    extra_wind_data = get_eia_form_860(const.blob_paths["eia_form860_2019_wind"])
+    full_data = wind_plants.merge(
+        extra_wind_data, on=["Plant Code", "Generator ID"], suffixes=(None, "_extra")
+    )
+    full_data.index = wind_plants.index
+    # Process data to expected types for profile generation
+    full_data["Turbine Hub Height (Feet)"] = (
+        full_data["Turbine Hub Height (Feet)"].map(floatify).fillna(default_hub_height)
+    )
+    full_data["Predominant Turbine Manufacturer"] = (
+        full_data["Predominant Turbine Manufacturer"].astype("string").fillna("")
+    )
+    full_data["Predominant Turbine Model Number"] = (
+        full_data["Predominant Turbine Model Number"].astype("string").fillna("")
+    )
+    start_dt = dt.datetime.fromisoformat(f"{year}-01-01-00")
+    end_dt = dt.datetime.fromisoformat(f"{year}-12-31-23")
+    retrieve_data(start_dt=start_dt, end_dt=end_dt, directory=download_directory)
+    profiles = calculate_pout_individual(
+        wind_farms=full_data,
+        start_dt=start_dt,
+        end_dt=end_dt,
+        directory=download_directory,
+    )
+    return profiles

prereise/gather/griddata/hifld/orchestration.py

@@ -6,22 +6,37 @@
 from prereise.gather.griddata.hifld.const import powersimdata_column_defaults
 from prereise.gather.griddata.hifld.data_process.demand import assign_demand_to_buses
 from prereise.gather.griddata.hifld.data_process.generators import build_plant
-from prereise.gather.griddata.hifld.data_process.profiles import build_solar
+from prereise.gather.griddata.hifld.data_process.profiles import build_solar, build_wind
 from prereise.gather.griddata.hifld.data_process.transmission import build_transmission
 
 
-def create_csvs(output_folder, nrel_email, nrel_api_key, solar_kwargs={}):
+def create_csvs(
+    output_folder,
+    wind_directory,
+    year,
+    nrel_email,
+    nrel_api_key,
+    solar_kwargs={},
+):
     """Process HIFLD source data to CSVs compatible with PowerSimData.
 
     :param str output_folder: directory to write CSVs to.
+    :param str wind_directory: directory to save wind speed data to.
+    :param int/str year: weather year to use to generate profiles.
     :param str nrel_email: email used to`sign up <https://developer.nrel.gov/signup/>`_.
     :param str nrel_api_key: API key.
     :param dict solar_kwargs: keyword arguments to pass to
         :func:`prereise.gather.solardata.nsrdb.sam.retrieve_data_individual`.
     """
     full_tables = create_grid(output_folder)
     create_profiles(
-        full_tables["plant"], nrel_email, nrel_api_key, output_folder, solar_kwargs
+        full_tables["plant"],
+        year,
+        nrel_email,
+        nrel_api_key,
+        wind_directory,
+        output_folder,
+        solar_kwargs,
     )
 
 
@@ -86,14 +101,22 @@ def create_grid(output_folder=None):
 
 
 def create_profiles(
-    plants, nrel_email, nrel_api_key, output_folder=None, solar_kwargs={}
+    plants,
+    year,
+    nrel_email,
+    nrel_api_key,
+    wind_directory,
+    output_folder=None,
+    solar_kwargs={},
 ):
     """Process a table of plant data to produce profile CSVs compatible with
         PowerSimData.
 
     :param pandas.DataFrame plants: table of plant data.
+    :param int/str year: weather year to use to generate profiles.
     :param str nrel_email: email used to`sign up <https://developer.nrel.gov/signup/>`_.
     :param str nrel_api_key: API key.
+    :param str wind_directory: directory to save wind speed data to.
     :param str output_folder: directory to write CSVs to. If None, CSVs will not be
         written (just returned).
     :param dict solar_kwargs: keyword arguments to pass to
@@ -102,12 +125,18 @@ def create_profiles(
         indexed by timestamp, with plant IDs as columns.
     """
     # Use plant data to build profiles
+    full_solar_kwargs = {**solar_kwargs, **{"year": year}}
     profiles = {
         "solar": build_solar(
             nrel_email,
             nrel_api_key,
             plants.query("type == 'solar'"),
-            **solar_kwargs,
+            **full_solar_kwargs,
+        ),
+        "wind": build_wind(
+            plants.query("type == 'wind'"),
+            wind_directory,
+            year,
         ),
     }
     if output_folder is not None:

prereise/gather/winddata/hrrr/hrrr_api.py

@@ -1,3 +1,5 @@
+import os
+
 import requests
 from pandas import date_range
 from tqdm import tqdm
@@ -83,7 +85,7 @@ def download_meteorological_data(
                     )
                 )
 
-            with open(directory + filename, "ab") as f:
+            with open(os.path.join(directory, filename), "ab") as f:
                 for grib_record_information in grib_record_information_list:
                     try:
                         self.downloader.download(