extract_data.py

import requests
import pandas as pd
import json

translation_dict = {
     'VindkraftAnleggId': 'WindPowerPlantId',
     'Navn': 'Name',
     'IdriftsettelseForsteByggetrinn': 'CommissioningFirstPhase',
     'AnleggsNr': 'FacilityNumber',
     'InstallertEffekt_MW': 'InstalledCapacity_MW',
     'HovedEierNavn': 'MainOwnerName',
     'HovedEierOrgNr': 'MainOwnerOrgNumber',
     'ElspotomraadeNummer': 'ElspotAreaNumber',
     'Fylke': 'County',
     'Kommune': 'Municipality',
     'NormalAArsproduksjon_GWh': 'NormalAnnualProduction_GWh',
     'GjsnittGeneratorytelse': 'AvgGeneratorOutput',
     'GjsnittNavhoeyde': 'AvgHubHeight',
     'GjsnittRotordiameter': 'AvgRotorDiameter',
     'EnergiPerSveiptAreal': 'EnergyPerSweptArea',
     'AntallOperativeTurbiner': 'NumberOfOperationalTurbines',
     'AnlKonsNr_Vind': 'FacilityPermitNumber_Wind',
     'AntallTurbiner': 'NumberOfTurbines',
     'DatoIdriftsatt': 'CommissioningDate',
     'DatoUtavdrift': 'DecommissioningDate',
     'ForventetProd_NormalAAr_GWh': 'ExpectedProduction_NormalYear_GWh',
     'KR_Saksid': 'NVE_CaseId',
     'TurbinID': 'TurbineID',
     'TurbinProdusent': 'TurbineManufacturer',
     'TurbinStorrelse_kW': 'TurbineSize_kW',
     'TurbinType': 'TurbineType',
     'TurbintypeID': 'TurbineTypeID',
     }

def translate_keys_recursive(obj, translation_dict):
    if isinstance(obj, dict):
        return {
            translation_dict.get(k, k): translate_keys_recursive(v, translation_dict)
            for k, v in obj.items()
        }
    elif isinstance(obj, list):
        return [translate_keys_recursive(item, translation_dict) for item in obj]
    else:
        return obj

def get_power():
        
    output_path = 'data/vindprod2002-2024_kraftverk_utcplus1.xlsx'
    url = 'https://www.nve.no/media/18018/vindprod2002-2024_kraftverk_utcplus1.xlsx'

    response = requests.get(url)
    with open(output_path, 'wb') as f:
        f.write(response.content)

    print("Power data saved to:", output_path)
    
def get_metadata():
        
    output_path = 'data/metadata.json'
    url = 'https://api.nve.no/web/WindPowerplant/GetWindPowerPlants'
    # url = "https://api.nve.no/web/WindPowerplant/GetWindPowerPlantsInOperation"

    response = requests.get(url)
    data = response.json()

    with open(output_path, 'w', encoding='utf-8') as f:
        json.dump(data, f, indent=4, ensure_ascii=False)

    print("Metadata saved to:", output_path)    

def get_geodata():
    
    output_path = 'data/geodata.json'
    latlon_wkid = 4326
    url = f'https://nve.geodataonline.no/arcgis/rest/services/Vindkraft2/MapServer/0/query?f=json&cacheHint=true&resultOffset=0&resultRecordCount=1000&where=1%3D1&orderByFields=OBJECTID&outFields=*&outSR={latlon_wkid}&spatialRel=esriSpatialRelIntersects'        

    response = requests.get(url)
    data = response.json()

    with open(output_path, 'w', encoding='utf-8') as f:
        json.dump(data, f, indent=4, ensure_ascii=False)

    print("Geodata saved to:", output_path) 

    
def extract_meta():
    
    output_path_1 = 'nve-windpower-metadata.csv'
    output_path_2 = 'nve-windpower-metadata-extended.csv'
    
    file_path_1 = 'data/metadata.json'
    file_path_2 = 'data/geodata.json'
    
    with open(file_path_1, 'r', encoding='utf-8') as f:
        metadata = json.load(f)
    with open(file_path_2, 'r', encoding='utf-8') as f:
        geodata = json.load(f)        
    
    metadata = translate_keys_recursive(metadata, translation_dict)
    
    # Convert to pandas dataframe
    metadata_df = pd.DataFrame(metadata)
    geodata_df = pd.DataFrame([{'name': park_feature['attributes']['anleggNavn'],
                         'code': park_feature['attributes']['anleggsNr'],
                         'capacity_MW': park_feature['attributes']['effekt_MW'],
                         'no_turbines': park_feature['attributes']['antallTurbiner'],
                         'start_date': pd.to_datetime(park_feature['attributes']['forsteIdriftDato'], unit='ms'),
                         'lat': park_feature['geometry']['y'],
                         'lon': park_feature['geometry']['x']
                       }
                      for park_feature in geodata['features']])

    metadata_df = metadata_df.set_index('FacilityNumber')
    geodata_df = geodata_df.set_index('code')
    
    # Add lat and lon from geodata_df
    metadata_df['lat'] = geodata_df['lat']
    metadata_df['lon'] = geodata_df['lon']
    
    # Reset index
    metadata_df = metadata_df.reset_index()
    
    # Set colums as int
    for c in ['WindPowerPlantId','FacilityNumber','MainOwnerOrgNumber','ElspotAreaNumber','NumberOfOperationalTurbines']:
        metadata_df[c] = pd.to_numeric(metadata_df[c], errors='coerce').astype('Int64')

    # Remove column with turbine meta
    metadata_df1 = metadata_df.copy()
    metadata_df1 = metadata_df1.drop('Turbiner', axis=1)
    metadata_df1 = metadata_df1.set_index('WindPowerPlantId').sort_index()

    # Explode turbine list
    df_exploded = metadata_df.explode('Turbiner').reset_index(drop=True)
    # Normalize turbine dictionaries into columns
    data_normalized = pd.json_normalize(df_exploded['Turbiner'])
    # Combine with original dataframe (without the old turbine column)
    metadata_df2 = pd.concat([df_exploded.drop(columns='Turbiner'), data_normalized], axis=1)
    metadata_df2 = metadata_df2.set_index('WindPowerPlantId').sort_index()

    # Save dataframe as cvs
    metadata_df1.to_csv(output_path_1, index=True)
    metadata_df2.to_csv(output_path_2, index=True)

def extract_power():
    
    output_path = 'nve-windpower-timeseries.csv'

    file_path = 'data/vindprod2002-2024_kraftverk_utcplus1.xlsx'
    
    power = pd.read_excel(file_path, header=1, skiprows=[2])

    power = power.rename(columns={'kraftverknr':'datetime'})
    power = power.set_index('datetime')
    power.index = pd.to_datetime(power.index, utc=True)
    
    # Sort by park id
    power = power[sorted(power.columns)]
    
    # Save dataframe as cvs
    power.to_csv(output_path, index=True)


if __name__ == '__main__':
    get_power()
    get_metadata()
    extract_meta()
    extract_power()