CAISO - Load, Fuel Mix, and LMP Data

This notebook walk through how to use gridstatus to access to the data availabe on OASIS.

While we will be using CAISO in this example, but most of this API will also work with all other ISOs.

import gridstatus
import pandas as pd
import plotly.express as px

caiso = gridstatus.CAISO()

let’s also define a method to make it easier to name our files

folder = "../../../archived_iso_data/caiso/"
def make_filename(name, start, end):
    return f"{name}_{start.strftime('%Y-%m-%d')}_{end.strftime('%Y-%m-%d')}.csv" 

Historical Fuel Mix

start = pd.Timestamp("April 10, 2018").normalize()
end = pd.Timestamp.now().normalize()

mix_df = caiso.get_fuel_mix(start, end=end, verbose=False)
mix_df.to_csv(make_filename(folder+"fuel_mix", start, end), index=None)

100%|██████████| 1662/1662 [14:11<00:00,  1.95it/s]

monthly_mix = mix_df.set_index("Time").resample("MS").sum().reset_index()[1:-1]
fig = px.bar(monthly_mix, x="Time", y=monthly_mix.columns[1:], title="Fuel Mix by Month - CAISO")
fig.show("svg")

Historical Load

start = pd.Timestamp("April 10, 2018").normalize()
end = pd.Timestamp.now().normalize()

load_df = caiso.get_load(start, end=end, verbose=False)

load_df.to_csv(make_filename(folder+"load", start, end), index=None)

100%|██████████| 1662/1662 [11:33<00:00,  2.40it/s]

daily_load = load_df.set_index("Time").resample("W").sum().reset_index()[1:-1]
fig = px.line(daily_load, x="Time", y="Load", title="Weekly Load - CAISO")
fig.show("svg")

Historical Locational Marginal Pricing (LMP)

You can supply whatever nodes or market you’d like, but for now let’s download data for 3 trading hubs in the Day Head Hourly Market

start = pd.Timestamp("Jan 01, 2022").normalize()
end = pd.Timestamp.now().normalize()

locations = ['TH_NP15_GEN-APND', 'TH_SP15_GEN-APND', 'TH_ZP26_GEN-APND']

lmp_df = caiso.get_lmp(start=start,  
                                  end=end,
                                  market='DAY_AHEAD_HOURLY', 
                                  locations=locations, 
                                  sleep=5)

lmp_df.to_csv(make_filename(folder+"dam_lmp", start, end), index=None)

100%|██████████| 10/10 [03:56<00:00, 23.68s/it]

daily_lmp = lmp_df.set_index("Time").groupby("Location").resample("1D").mean().reset_index()[1:-1]
fig = px.line(daily_lmp, x="Time", y="LMP", title="DAILY LMP - CAISO", color="Location")
fig.show("svg")

Gas Prices

CAISO also publish information about gas prices and greenhouse gas allownces that we will download

start = start = pd.Timestamp("Jan 01, 2022").normalize()
end = pd.Timestamp.now().normalize()

gas_price_df = caiso.get_gas_prices(start=start, end=end, fuel_region_id="FRPGE2GHG")

gas_price_df.to_csv(make_filename(folder+"gas_prices", start, end), index=None)

100%|██████████| 10/10 [00:58<00:00,  5.81s/it]

start = start = pd.Timestamp("Jan 01, 2022").normalize()
end = pd.Timestamp.now().normalize()

ghg_df = caiso.get_ghg_allowance(start=start, end=end)

ghg_df.to_csv(make_filename(folder+"ghg_allowance", start, end), index=None)

100%|██████████| 10/10 [00:55<00:00,  5.52s/it]