Last active
January 23, 2018 01:54
-
-
Save jdherman/3aeb9e90558e68176052fa1cd3892084 to your computer and use it in GitHub Desktop.
Download CAISO hourly renewables data
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# this runs in python 3 as of Jan. 2018 | |
import pandas as pd | |
import time | |
import urllib | |
from bs4 import BeautifulSoup | |
import numpy as np | |
from io import StringIO | |
import matplotlib.pyplot as plt | |
dates = pd.date_range(start='4/20/2010', end='1/20/2018') | |
dates = dates.strftime('%Y%m%d') | |
hours = pd.date_range(start='4/20/2010', end='1/21/2018', freq='H') | |
cols = ['geothermal', 'biomass', 'biogas', 'small hydro', 'wind total', 'solar', | |
'renewables', 'nuclear', 'thermal', 'imports', 'hydro'] | |
# this will be the dataframe we fill in | |
total_gen = pd.DataFrame(index=hours, columns=cols) | |
for d in dates: | |
if d in ['20150308', '20160313', '20170312']: # some bad bad days | |
continue | |
print(d) | |
url = 'http://content.caiso.com/green/renewrpt/%s_DailyRenewablesWatch.txt' % d | |
try: | |
page = urllib.request.urlopen(url).read() | |
except: | |
print('Day %s failed, continuing...' % d) | |
soup = BeautifulSoup(page, 'lxml') | |
t = str(soup.find('p').text).replace('\t\t', ',').replace('\t',',') | |
# grab the two tables and combine them into one | |
df1 = pd.read_csv(StringIO(t), header=1, nrows=24) | |
df2 = pd.read_csv(StringIO(t), header=29, nrows=24) | |
df = pd.concat([df1, df2.drop('Hour', axis=1)], axis=1) | |
df.set_index('Hour', inplace=True) | |
df = df[df.columns[~df.columns.str.contains('Unnamed:')]] | |
df.columns = [c.lower() for c in df.columns] # uppercase | |
# starting in July 2012, they separate solar pv and solar thermal | |
# add them back together and fix missing values | |
if df.values.shape[1] > 11: | |
if df['solar thermal'].dtype == np.object_: # string | |
ix = df['solar thermal'].str.contains('No Good Data') | |
df['solar thermal'][ix] = np.nan | |
df['solar thermal'] = pd.to_numeric(df['solar thermal']) | |
df['solar'] = df['solar pv'] + df['solar thermal'] | |
df = df[list(total_gen.columns)] # re-order and only keep cols in output | |
total_gen[d] = df.values | |
# renew_gen.to_csv('CAISO-renewables-hourly.csv') | |
total_gen.to_csv('CAISO-all-hourly.csv') | |
# even after this, beware of bad data values like #NAME, #REF, #VALUE etc | |
# I did some manual cleaning in the final CSV files |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment