def load_GFS_data(latitude=33.8688,
longitude=151.2093,
tz='Australia/Sydney',
days=7):
# latitude, longitude, tz = 32.2, -110.9, 'US/Arizona'
# latitude, longitude, tz = 32.2, -110.9, 'US/Arizona'
# latitude = 33.8688
# longitude=151.2093
# tz='Australia/Sydney'
start = pd.Timestamp(datetime.date.today(), tz=tz)
end = start + pd.Timedelta(days=7)
irrad_vars = ['ghi', 'dni', 'dhi']
model = GFS()
raw_data = model.get_data(latitude, longitude, start, end)
print(raw_data.head())
data = raw_data
data = model.rename(data)
data['temp_air'] = model.kelvin_to_celsius(data['temp_air'])
data['wind_speed'] = model.uv_to_speed(data)
irrad_data = model.cloud_cover_to_irradiance(data['total_clouds'])
data = data.join(irrad_data, how='outer')
data = data[model.output_variables]
print(data.head())
data = model.process_data(raw_data)
print(data.head())
data = model.get_processed_data(latitude, longitude, start, end)
print(data.head())
return (data)
def _get_forecast_gfs_day(lat: float, lon: float, date: datetime.date) -> pd.DataFrame:
"""
Suspected that multithreading this does not work, because of
some internal logic within pvlib.forecast.ForecastModel.get_data
"""
logger.debug(f"GFS day {lat} {lon} {date}")
today = datetime.datetime.today()
date_after = date + datetime.timedelta(days=1)
one_week_from_today = today + datetime.timedelta(weeks=1)
assert lat % 0.25 == 0, "Latitude must be multiple of 0.25"
assert lon % 0.25 == 0, "Longitude must be multiple of 0.25"
# assert one_week_from_today.date() - date >= datetime.timedelta(), "Cannot be more than 1 week in the future"
model = GFS(resolution="quarter")
start = datetime.datetime.combine(date, datetime.time())
end = datetime.datetime.combine(date_after, datetime.time())
try:
data: pd.DataFrame = model.get_data(lat, lon, start, end)
except requests.exceptions.ConnectionError:
raise ConnectionAbortedError("Connection Error while fetching forecast data. Check network connection.")
return model.process_data(data)
Created on Mon Mar 10 05:00:26 2019
@author: tstone
"""
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import pvlib
from pvlib.forecast import GFS
location_path = "streetlight_locations_datasd_ids.csv"
sll = pd.read_csv(location_path)
# GFS model, defaults to 0.5 degree resolution
model_gfs = GFS()
#Create times to retrieve archive data
tz = 'US/Pacific'
end = pd.Timestamp.today(tz=tz)
start = end - pd.Timedelta(days=7)
for index, row in sll.iterrows():
#print(index, row.longitude, row.latitude, row.ID)
latitude = row.latitude
longitude = row.longitude
raw_data = model_gfs.get_data(latitude, longitude, start, end)
data = model_gfs.process_data(raw_data)
data.to_csv("../Hackathon Datasets/Cloud/cloud_ID_" + str(row.ID) + ".csv")
