Source code for reegis.openego

# -*- coding: utf-8 -*-

"""Processing the openego map for the electricity demand.

SPDX-FileCopyrightText: 2016-2021 Uwe Krien <>

SPDX-License-Identifier: MIT
__copyright__ = "Uwe Krien <>"
__license__ = "MIT"

# Python libraries
import os
import logging
from shapely import wkb
import warnings

# External libraries
import pandas as pd

# Internal modules
from reegis import config as cfg
from reegis import geometries
from reegis import bmwi as bmwi_data
from reegis import oedb
from reegis import tools

[docs]def wkb2wkt(x): """Loads geometry from wkb.""" return wkb.loads(x, hex=True)
[docs]def download_oedb(oep_url, schema, table, query, fn, overwrite=False): """Download map from oedb in WGS84 and store as csv file.""" if not os.path.isfile(fn) or overwrite: gdf = oedb.oedb(oep_url, schema, table, query, "geom_centre", 3035) gdf = gdf.to_crs(crs="epsg:4326")"Write data to {0}".format(fn)) gdf.to_csv(fn) else: logging.debug("File {0} exists. Nothing to download.".format(fn)) return fn
[docs]def get_ego_data(osf=True, sectors=False, query="?where=version=v0.4.5"): """ Parameters ---------- osf : bool If True the file will be downloaded from the osf page instead of of selected from the database. You may not get the latest version. (default: False) sectors : bool By default (False) only the total comsumption is returned. If "True" the consumption devided by sectors will be returned. query : str Database query to filter the data set. (default: '?where=version=v0.4.5') Returns ------- Examples -------- >>> from reegis import openego >>> # download from file (faster) >>> openego.get_ego_data() # doctest: +SKIP >>> # download from oedb database (get latest updates, very slow) >>> openego.get_ego_data(osf=False) # doctest: +SKIP """ oep_url = "" local_path = cfg.get("paths", "ego") fn_large_consumer = os.path.join( local_path, cfg.get("open_ego", "ego_large_consumers") ) fn_load_areas = os.path.join( local_path, cfg.get("open_ego", "ego_load_areas") ) # Large scale consumer schema = "model_draft" table = "ego_demand_hv_largescaleconsumer" query_lsc = "" download_oedb(oep_url, schema, table, query_lsc, fn_large_consumer) large_consumer = pd.read_csv(fn_large_consumer, index_col=[0]) msg = ( "\nYou are going to download the load areas from file created " "2019-10-09.\nThis is much faster and useful for most users but you" " may find more actual data on the oedb database.\n" "Please check:" "/ego_dp_loadarea \n" "Use 'openego.get_ego_data(osf=False)' to fetch data from oedb.\n" ) # Load areas if osf is True: warnings.warn(msg) url = cfg.get("open_ego", "osf_url") tools.download_file(fn_load_areas, url) load_areas = pd.DataFrame(pd.read_csv(fn_load_areas, index_col=[0])) else: schema = "demand" table = "ego_dp_loadarea" download_oedb(oep_url, schema, table, query, fn_load_areas) load_areas = pd.DataFrame(pd.read_csv(fn_load_areas, index_col=[0])) load_areas.rename( columns={"sector_consumption_sum": "consumption"}, inplace=True ) if sectors: large_consumer["sector_consumption_large_consumers"] = large_consumer[ "consumption" ] cols_lc = [c for c in large_consumer.columns if "consumption" in c] + [ "geom_centre" ] cols_la = [c for c in load_areas.columns if "consumption" in c] + [ "geom_centre" ] else: cols_la = ["consumption", "geom_centre"] cols_lc = ["consumption", "geom_centre"] load = pd.concat([load_areas[cols_la], large_consumer[cols_lc]]) load = load.rename(columns={"geom_centre": "geom"}) return load.reset_index(drop=True)
[docs]def get_ego_demand(filename=None, sectors=False, overwrite=False): """ Parameters ---------- filename : str sectors : bool overwrite : bool Returns ------- pandas.DataFrame """ if filename is None: path = cfg.get("paths", "demand") filename = os.path.join(path, cfg.get("open_ego", "ego_file")) if sectors is True: filename = filename.replace(".", "_sectors.") if os.path.isfile(filename) and not overwrite: return pd.DataFrame(pd.read_hdf(filename, "demand")) else: load = get_ego_data(osf=True, sectors=sectors) load.to_hdf(filename, "demand") return load
[docs]def get_ego_demand_by_region( regions, name, outfile=None, infile=None, dump=False, grouped=False, sectors=False, overwrite=False, ): """ Add the region id from a given region set to the openego demand table. This can be used to calculate the demand or the share of each region. Parameters ---------- regions : GeoDataFrame A region set. name : str The name of the region set will be used as the name of the column in the openego GeoDataFrame and to distinguish result files. outfile : str (optional) It is possible to pass a filename (with path) where the results should be stored. Only valid if `dump` is True. infile : str (optional) It is possible to use a specific infile (with path) where the openego map is stored. dump : bool If dump is True the result will be returned and stored into a file. Otherwise the result is just returned. (default: False) grouped : bool If grouped is False the openego table with a region column is returned. Otherwise the map is grouped by the region column and the consumption column is summed up. (default: False) sectors : bool Still missing. overwrite : bool Returns ------- pandas.DataFrame or pandas.Series : A Series is returned if grouped is True. Notes ----- The openego map may not be updated in the future so it might be necessary to scale the results to an overall demand. Examples -------- >>> federal_states=geometries.get_federal_states_polygon() >>> bmwi_annual=bmwi_data.get_annual_electricity_demand_bmwi( ... 2015) # doctest: +SKIP >>> egodemand=get_ego_demand_by_region( ... federal_states, 'federal_states', grouped=True) # doctest: +SKIP >>> egodemand.div(ego_demand.sum()).mul(bmwi_annual) # doctest: +SKIP """ if outfile is None: path = cfg.get("paths", "demand") outfile = os.path.join(path, "open_ego_demand_{0}.h5") if sectors: outfile = outfile.format(name + "_sectors") else: outfile = outfile.format(name) if not os.path.isfile(outfile) or overwrite: ego_data = get_ego_demand(filename=infile, sectors=sectors) ego_demand = geometries.create_geo_df(ego_data) # Add column with regions logging.debug( "OpenEgo spatial join: Demand polygon centroids with " "{0}".format(name) ) ego_demand = geometries.spatial_join_with_buffer( ego_demand, regions, name ) # Overwrite Geometry object with its DataFrame, because it is not # needed anymore. ego_demand = pd.DataFrame(ego_demand) ego_demand["geometry"] = ego_demand["geometry"].astype(str) # Write out file (hdf-format). if dump is True: ego_demand.to_hdf(outfile, "demand") else: ego_demand = pd.DataFrame(pd.read_hdf(outfile, "demand")) if grouped is True: return ego_demand.groupby(name)["consumption"].sum() else: return ego_demand