Merge efcbc7d into 4b9c018

reegis/bmwi.py

@@ -106,6 +106,10 @@ def read_bmwi_sheet_7(sub):
 def bmwi_re_energy_capacity():
     """Prepare the energy production and capacity table from sheet 20.
 
+    capacity: [MW]
+    energy: [GWh]
+    fraction: [-]
+
     Examples
     --------
     >>> re = bmwi_re_energy_capacity()  # doctest: +SKIP
@@ -136,16 +140,21 @@ def get_annual_electricity_demand_bmwi(year):
     --------
     >>> get_annual_electricity_demand_bmwi(2014)  # doctest: +SKIP
     523.988
-    >>> get_annual_electricity_demand_bmwi(1900)  # doctest: +SKIP
-    None
     """
+    import math
     infile = get_bmwi_energiedaten_file()
 
     table = pd.read_excel(infile, '21', skiprows=7, index_col=[0])
     try:
-        return table.loc['   zusammen', year]
+        value = table.loc['   zusammen', year]
+        if math.isnan(value):
+            value = None
     except KeyError:
-        return None
+        value = None
+    if value is None:
+        msg = "No BMWi electricity demand found for {year}."
+        raise ValueError(msg.format(year=year))
+    return value
 
 
 if __name__ == "__main__":

reegis/coastdat.py

@@ -520,11 +520,11 @@ def store_average_weather(data_type, weather_path=None, years=None, keys=None,
                 ws, verify_integrity=True)
 
         # calculate the average wind speed for one grid item
-        coastdat_polygons.loc[key, 'v_wind_avg'] = (
+        coastdat_polygons.loc[key, '{0}_avg'.format(data_type)] = (
             data_type_avg.mean())
 
     # Close hdf files
-    for year in years:
+    for year in used_years:
         weather[year].close()
 
     if keys is not None:
@@ -993,8 +993,7 @@ def scenario_feedin_wind(year, name, regions=None, feedin_ts=None,
     # Rename columns and remove obsolete level
     wind.columns = wind.columns.droplevel(2)
     cols = wind.columns.get_level_values(1).unique()
-    rn = {c: c.replace('coastdat_{weather_year}_wind_'
-                       .format(weather_year=weather_year), '') for c in cols}
+    rn = {c: c.replace('coastdat_{0}_wind_'.format(year), '') for c in cols}
     wind.rename(columns=rn, level=1, inplace=True)
     wind.sort_index(1, inplace=True)
 
@@ -1006,7 +1005,7 @@ def scenario_feedin_wind(year, name, regions=None, feedin_ts=None,
         regions = wind.columns.get_level_values(0).unique()
 
     if feedin_ts is None or len(feedin_ts.index) == 0:
-        cols = pd.MultiIndex(levels=[[], []], labels=[[], []])
+        cols = pd.MultiIndex(levels=[[], []], codes=[[], []])
         feedin_ts = pd.DataFrame(index=wind.index, columns=cols)
 
     for region in regions:
@@ -1048,7 +1047,7 @@ def scenario_feedin_pv(year, name, regions=None, feedin_ts=None,
         regions = pv.columns.get_level_values(0).unique()
 
     if feedin_ts is None or len(feedin_ts.index) == 0:
-        cols = pd.MultiIndex(levels=[[], []], labels=[[], []])
+        cols = pd.MultiIndex(levels=[[], []], codes=[[], []])
         feedin_ts = pd.DataFrame(index=pv.index, columns=cols)
 
     orientation_fraction = pd.Series(pv_orientation)
@@ -1067,7 +1066,7 @@ def scenario_feedin_pv(year, name, regions=None, feedin_ts=None,
     return feedin_ts.sort_index(1)
 
 
-def get_feedin_per_region(year, region, name, weather_year=None,
+def get_feedin_per_region(year, region, name, weather_year=None, reset_pp=True,
                           windzones=True, subregion=False, pp=None):
     """
     Aggregate feed-in time series for the given geometry set.
@@ -1078,6 +1077,7 @@ def get_feedin_per_region(year, region, name, weather_year=None,
     region : geopandas.geoDataFrame
     name : str
     weather_year : int
+    reset_pp
     windzones : bool
     pp : pd.DataFrame or None
     subregion : bool
@@ -1111,7 +1111,7 @@ def get_feedin_per_region(year, region, name, weather_year=None,
     # Add a column named with the name parameter, adding the region id to
     # each power plant
     pp = powerplants.add_regions_to_powerplants(
-        region, name, filename=filename, path=path, dump=False, pp=pp,
+        region, name, filename=filename, path=path, dump=True, pp=pp,
         subregion=subregion)
 
     # Get only the power plants that are online in the given year.
@@ -1217,8 +1217,8 @@ def get_solar_time_series_for_one_location_all_years(latitude, longitude,
     """
     path = os.path.join(cfg.get('paths', 'feedin'), 'coastdat')
     years = os.listdir(path)
-    df = pd.DataFrame(columns=pd.MultiIndex(levels=[[], []], labels=[[], []]))
-    # years = [2012, 2013, 2014]
+    df = pd.DataFrame(columns=pd.MultiIndex(levels=[[], []], codes=[[], []]))
+
     for year in years:
         if os.path.isdir(os.path.join(path, str(year))):
             tmp = get_solar_time_series_for_one_location(
@@ -1230,15 +1230,19 @@ def get_solar_time_series_for_one_location_all_years(latitude, longitude,
 
 def federal_states_feedin_example():
     """Get fullload hours for renewable sources for a federal states."""
-    federal_states = geometries.load(
-        cfg.get('paths', 'geometry'),
-        cfg.get('geometry', 'federalstates_polygon'))
+    federal_states = geometries.get_federal_states_polygon()
     get_feedin_per_region(2014, federal_states, 'federal_states')
-
     return scenario_feedin(2014, 'federal_states')
 
 
 if __name__ == "__main__":
     logger.define_logging()
-    print(federal_states_feedin_example().sum())
-    print(federal_state_average_weather(2014, 'temp_air'))
+    powerplants.pp_opsd2reegis()
+    for my_year in [2014, 2013, 2012, 2011, 2010]:
+        my_federal_states = geometries.get_federal_states_polygon()
+        get_feedin_per_region(my_year, my_federal_states, 'federal_states',
+                              reset_pp=False)
+        my_path = os.path.join(cfg.get('paths', 'feedin'), 'federal_states')
+        os.makedirs(my_path, exist_ok=True)
+        my_fn = os.path.join(my_path, 'federal_states_{0}'.format(my_year))
+        scenario_feedin(my_year, 'federal_states').to_csv(my_fn)

reegis/demand.py

@@ -0,0 +1,80 @@
+# -*- coding: utf-8 -*-
+
+# -*- coding: utf-8 -*-
+
+""" Download and prepare entsoe load profile from opsd data portal.
+
+Copyright (c) 2016-2018 Uwe Krien <uwe.krien@rl-institut.de>
+
+SPDX-License-Identifier: GPL-3.0-or-later
+"""
+__copyright__ = "Uwe Krien <uwe.krien@rl-institut.de>"
+__license__ = "GPLv3"
+
+
+# Python libraries
+import logging
+
+# External packages
+import pandas as pd
+
+# internal modules
+import reegis.config as cfg
+from reegis import entsoe
+from reegis import geometries
+from reegis import openego
+from reegis import bmwi as bmwi_data
+
+
+def get_slp_profile_by_region(region, year, annual, profile='entsoe'):
+    pass
+
+
+def get_entsoe_profile_by_region(region, year, name, annual_demand=None):
+    """
+
+    Parameters
+    ----------
+    region
+    year
+    name
+    annual_demand
+
+    Returns
+    -------
+
+    """
+    logging.debug("Get entsoe profile {0} for {1}".format(name, year))
+    de_load_profile = entsoe.get_entsoe_load(2014).DE_load_
+
+    load_profile = pd.DataFrame()
+
+    annual_region = openego.get_ego_demand_by_region(year, region, name)
+
+    share = annual_region.div(annual_region.sum())
+
+    for region in region.index:
+        if region not in share:
+            share[region] = 0
+        load_profile[region] = de_load_profile.multiply(float(share[region]))
+
+    if annual_demand == 'bmwi':
+        annual_demand = bmwi_data.get_annual_electricity_demand_bmwi(year)
+
+    if annual_demand is not None:
+        load_profile = load_profile.div(load_profile.sum().sum()).multiply(
+            annual_demand)
+    return load_profile
+
+
+def get_electricity_profile_by_federal_states(year, profile=entsoe):
+    federal_states = geometries.load(
+        cfg.get('paths', 'geometry'),
+        cfg.get('geometry', 'federalstates_polygon'))
+    federal_states.set_index('iso', drop=True, inplace=True)
+    return get_entsoe_profile_by_region(federal_states, 'federal_states', year)
+
+
+if __name__ == "__main__":
+    logging.basicConfig(level=logging.DEBUG)
+    print(get_electricity_profile_by_federal_states(2014)['BE'])

reegis/entsoe.py

@@ -23,29 +23,30 @@
 import dateutil
 import pandas as pd
 
-# oemof packages
-from oemof.tools import logger
-
 # internal modules
 import reegis.config as cfg
 
 
 def read_original_timeseries_file(overwrite=False):
     """Read timeseries file if it exists. Otherwise download it from opsd.
     """
-
-    orig_csv_file = os.path.join(cfg.get('paths', 'entsoe'),
-                                 cfg.get('entsoe', 'original_file'))
-    readme = os.path.join(cfg.get('paths', 'entsoe'),
-                          cfg.get('entsoe', 'readme_file'))
-    json = os.path.join(cfg.get('paths', 'entsoe'),
-                        cfg.get('entsoe', 'json_file'))
+    version = cfg.get('entsoe', 'time_series_version')
+    orig_csv_file = os.path.join(
+        cfg.get('paths', 'entsoe'),
+        cfg.get('entsoe', 'original_file')).format(version=version)
+    readme = os.path.join(
+        cfg.get('paths', 'entsoe'),
+        cfg.get('entsoe', 'readme_file')).format(version=version)
+    json = os.path.join(
+        cfg.get('paths', 'entsoe'),
+        cfg.get('entsoe', 'json_file')).format(version=version)
 
     version = cfg.get('entsoe', 'timeseries_version')
 
     if not os.path.isfile(orig_csv_file) or overwrite:
-        req = requests.get(cfg.get('entsoe', 'timeseries_data').format(
-            version=version))
+        req = requests.get(
+            cfg.get('entsoe', 'timeseries_data').format(version=version))
+
         if not overwrite:
             logging.warning("File not found. Try to download it from server.")
         else:
@@ -57,26 +58,26 @@ def read_original_timeseries_file(overwrite=False):
         logging.warning("Downloaded from {0} and copied to '{1}'.".format(
             cfg.get('entsoe', 'timeseries_data').format(version=version),
             orig_csv_file))
-        req = requests.get(cfg.get('entsoe', 'timeseries_readme').format(
-            version=version))
+        req = requests.get(
+            cfg.get('entsoe', 'timeseries_readme').format(version=version))
         with open(readme, 'wb') as fout:
             fout.write(req.content)
-        req = requests.get(cfg.get('entsoe', 'timeseries_json').format(
-            version=version))
+        req = requests.get(
+            cfg.get('entsoe', 'timeseries_json').format(version=version))
         with open(json, 'wb') as fout:
             fout.write(req.content)
-
     logging.debug("Reading file: {0}".format(orig_csv_file))
-    orig = pd.read_csv(orig_csv_file, index_col=[0], parse_dates=True,
-                       date_parser=lambda col: pd.to_datetime(col, utc=True))
+    orig = pd.read_csv(orig_csv_file, index_col=[0], parse_dates=True)
     orig = orig.tz_convert('Europe/Berlin')
     return orig
 
 
 def prepare_de_file(overwrite=False):
     """Convert demand file. CET index and Germany's load only."""
-    de_file = os.path.join(cfg.get('paths', 'entsoe'),
-                           cfg.get('entsoe', 'de_file'))
+    version = cfg.get('entsoe', 'time_series_version')
+    de_file = os.path.join(
+        cfg.get('paths', 'entsoe'),
+        cfg.get('entsoe', 'de_file').format(version=version))
     if not os.path.isfile(de_file) or overwrite:
         ts = read_original_timeseries_file(overwrite)
         for col in ts.columns:
@@ -88,8 +89,10 @@ def prepare_de_file(overwrite=False):
 
 def split_timeseries_file(overwrite=False, csv=False):
     logging.info("Splitting time series.")
+    version = cfg.get('entsoe', 'time_series_version')
     path_pattern = os.path.join(cfg.get('paths', 'entsoe'), '{0}')
-    de_file = path_pattern.format(cfg.get('entsoe', 'de_file'))
+    de_file = path_pattern.format(cfg.get('entsoe', 'de_file').format(
+        version=version))
 
     if not os.path.isfile(de_file) or overwrite:
         prepare_de_file(overwrite)
@@ -123,19 +126,29 @@ def split_timeseries_file(overwrite=False, csv=False):
     renewables = de_ts.dropna(subset=re_subset, how='any')[re_columns]
 
     if csv:
-        load_file = path_pattern.format(cfg.get('entsoe', 'load_file_csv'))
+        load_file = path_pattern.format(
+            cfg.get('entsoe', 'load_file_csv').format(version=version))
     else:
-        load_file = path_pattern.format(cfg.get('entsoe', 'load_file'))
+        load_file = path_pattern.format(
+            cfg.get('entsoe', 'load_file').format(version=version))
 
     if not os.path.isfile(load_file) or overwrite:
         if csv:
             load.to_csv(load_file)
         else:
             load.to_hdf(load_file, 'entsoe')
 
-    re_file = path_pattern.format(cfg.get('entsoe', 'renewables_file'))
+    if csv:
+        re_file = path_pattern.format(
+            cfg.get('entsoe', 'renewables_file_csv').format(version=version))
+    else:
+        re_file = path_pattern.format(
+            cfg.get('entsoe', 'renewables_file').format(version=version))
     if not os.path.isfile(re_file) or overwrite:
-        renewables.to_csv(re_file)
+        if csv:
+            renewables.to_csv(re_file)
+        else:
+            renewables.to_hdf(re_file, 're')
 
 
 def prepare_entsoe_timeseries(overwrite=False):
@@ -152,11 +165,27 @@ def get_entsoe_load(year):
     f = pd.datetime(year, 1, 1, 0)
     t = pd.datetime(year, 12, 31, 23)
     logging.info("Read entsoe load series from {0} to {1}".format(f, t))
-    df = pd.read_hdf(filename, 'entsoe')
+    df = pd.DataFrame(pd.read_hdf(filename, 'entsoe'))
     return df.loc[f:t]
 
 
+def get_entsoe_renewable_data(csv=False, overwrite=False):
+    version = cfg.get('entsoe', 'time_series_version')
+    path_pattern = os.path.join(cfg.get('paths', 'entsoe'), '{0}')
+    if csv:
+        fn = path_pattern.format(
+            cfg.get('entsoe', 'renewables_file_csv').format(version=version))
+    else:
+        fn = path_pattern.format(
+            cfg.get('entsoe', 'renewables_file').format(version=version))
+    if not os.path.isfile(fn):
+        split_timeseries_file(csv=csv)
+    if csv:
+        re = pd.read_csv(fn, index_col=[0], parse_dates=True)
+    else:
+        re = pd.DataFrame(pd.read_hdf(fn, 're'))
+    return re
+
+
 if __name__ == "__main__":
-    logger.define_logging()
-    print(get_entsoe_load(2014))
-    logging.info("Done!")
+    pass

reegis/feedin.py

@@ -305,7 +305,7 @@ def feedin_windpowerlib(weather, turbine, installed_capacity=1):
     >>> turbine = {
     ...     'hub_height': 135,
     ...     'rotor_diameter': 127,
-    ...     'name': 'E-141/4200',
+    ...     'name': 'E-82/2300',
     ...     'nominal_power': 4200000,
     ...     'fetch_curve': 'power_coefficient_curve'}
     >>> data_height = cfg.get_dict('coastdat_data_height')