02_solar_power_forecasts.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Solar Power Forecasts\n",
    "\n",
    "This notebook picks up where we left off in [01_intro_to_solar_actuals_and_forecasts.ipynb](01_intro_to_solar_actuals_and_forecasts.ipynb), with a forecast dataset covering the Southeast including irradiance, temperature, and wind speed. In this notebook we will turn that into power forecasts for a fleet of solar plants. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from herbie import Herbie\n",
    "from herbie import FastHerbie\n",
    "import pandas as pd\n",
    "import pickle\n",
    "\n",
    "import cartopy.crs as ccrs\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from paint.standard2 import cm_tmp, cm_wind, cm_wave_height, cm_pcp\n",
    "from toolbox import EasyMap, pc, ccrs\n",
    "import xarray as xr\n",
    "from dateutil.relativedelta import relativedelta\n",
    "from dateutil.parser import parse\n",
    "import pvlib\n",
    "\n",
    "import cartopy\n",
    "import cartopy.crs as crs\n",
    "import cartopy.feature as cfeature"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading forecast data\n",
    "Let's load the netCDF file we created in the last notebook, and then coarsen it to get the mean, min, and max of larger areas:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "ds = xr.open_dataset('nc_files/sample_hrrr_dataset_20210408.nc')\n",
    "\n",
    "window_size = 10\n",
    "ds_coarse = ds.coarsen(x=window_size, y=window_size, boundary='trim').mean() \n",
    "ds_coarse_max = ds.coarsen(x=window_size, y=window_size, boundary='trim').max()\n",
    "ds_coarse_min = ds.coarsen(x=window_size, y=window_size, boundary='trim').min() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Power for multiple PV plants\n",
    "### Loading plant specifications\n",
    "Let's load a list of plants that represents the fleet of interest.\n",
    "\n",
    "This list is based on EIA Form 860 data [1]. I pulled list of solar plants (`3_3_Solar_Y2022_Early_Release.xlsx`), used `xlookup()` to merge latitude, longitude from full plant list (`2___Plant_Y2022_Early_Release.xlsx`), filtered on Balancing Area = SOCO, sorted by AC capacity. \n",
    "\n",
    "I made the following metadata assumptions:\n",
    " - `temperature_coefficient`: if crystalline, -0.35 %/deg C, if thin film, -0.25 %/deg C\n",
    " - `axis_tilt`, `axis_azimuth`, `rotation_angle`: if `tracking_type` is `single_axis`, 0, 180, 60, respectively\n",
    " - `backtracking`: if `single_axis` and crystalline, `True`, else, `False`\n",
    " - `ground_coverage_ratio`: `fixed` - 0.5; `single_axis` with thin film - 0.45; `single_axis` with crystalline - 0.35\n",
    "\n",
    " Merging, filtering, and filling in metadata was done in a seperate spreadsheet, then saved as a .csv as `pv_fleet_metadata3.csv`.\n",
    "\n",
    " [1] https://www.eia.gov/electricity/data/eia860/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "number of plants: 51\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>index</th>\n",
       "      <th>plant_name</th>\n",
       "      <th>plant_code</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>ac_capacity</th>\n",
       "      <th>dc_capacity</th>\n",
       "      <th>dc_loss_factor</th>\n",
       "      <th>temperature_coefficient</th>\n",
       "      <th>tracking</th>\n",
       "      <th>...</th>\n",
       "      <th>fixed_azimuth</th>\n",
       "      <th>axis_tilt</th>\n",
       "      <th>axis_azimuth</th>\n",
       "      <th>backtrack</th>\n",
       "      <th>max_rotation_angle</th>\n",
       "      <th>ground_coverage_ratio</th>\n",
       "      <th>Crystalline Silicon?</th>\n",
       "      <th>Thin-Film (CdTe)?</th>\n",
       "      <th>Operating_Month</th>\n",
       "      <th>Operating_Year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>Cool Springs Solar (Hybrid)</td>\n",
       "      <td>63721.0</td>\n",
       "      <td>30.747916</td>\n",
       "      <td>-84.680492</td>\n",
       "      <td>213.0</td>\n",
       "      <td>298.0</td>\n",
       "      <td>0.15</td>\n",
       "      <td>-0.35</td>\n",
       "      <td>Y</td>\n",
       "      <td>...</td>\n",
       "      <td>180</td>\n",
       "      <td>0.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>True</td>\n",
       "      <td>60.0</td>\n",
       "      <td>0.35</td>\n",
       "      <td>Y</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12.0</td>\n",
       "      <td>2021.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Twiggs Solar</td>\n",
       "      <td>61696.0</td>\n",
       "      <td>32.604997</td>\n",
       "      <td>-83.491239</td>\n",
       "      <td>204.0</td>\n",
       "      <td>261.0</td>\n",
       "      <td>0.15</td>\n",
       "      <td>-0.25</td>\n",
       "      <td>Y</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>False</td>\n",
       "      <td>60.0</td>\n",
       "      <td>0.45</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Y</td>\n",
       "      <td>9.0</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>Hickory Park Solar Hybrid</td>\n",
       "      <td>63522.0</td>\n",
       "      <td>31.269676</td>\n",
       "      <td>-84.176321</td>\n",
       "      <td>195.5</td>\n",
       "      <td>268.8</td>\n",
       "      <td>0.15</td>\n",
       "      <td>-0.35</td>\n",
       "      <td>Y</td>\n",
       "      <td>...</td>\n",
       "      <td></td>\n",
       "      <td>0.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>True</td>\n",
       "      <td>60.0</td>\n",
       "      <td>0.35</td>\n",
       "      <td>Y</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2022.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Camilla Solar Energy Project</td>\n",
       "      <td>61785.0</td>\n",
       "      <td>31.269550</td>\n",
       "      <td>-84.228721</td>\n",
       "      <td>171.4</td>\n",
       "      <td>224.0</td>\n",
       "      <td>0.15</td>\n",
       "      <td>-0.35</td>\n",
       "      <td>Y</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>True</td>\n",
       "      <td>60.0</td>\n",
       "      <td>0.35</td>\n",
       "      <td>Y</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2020.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>SR DeSoto I, LLC</td>\n",
       "      <td>64841.0</td>\n",
       "      <td>31.886486</td>\n",
       "      <td>-84.021586</td>\n",
       "      <td>165.0</td>\n",
       "      <td>177.4</td>\n",
       "      <td>0.15</td>\n",
       "      <td>-0.25</td>\n",
       "      <td>Y</td>\n",
       "      <td>...</td>\n",
       "      <td>180</td>\n",
       "      <td>0.0</td>\n",
       "      <td>180.0</td>\n",
       "      <td>False</td>\n",
       "      <td>60.0</td>\n",
       "      <td>0.45</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Y</td>\n",
       "      <td>12.0</td>\n",
       "      <td>2022.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 22 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   index                    plant_name  plant_code   latitude  longitude  \\\n",
       "0      0   Cool Springs Solar (Hybrid)     63721.0  30.747916 -84.680492   \n",
       "1      1                  Twiggs Solar     61696.0  32.604997 -83.491239   \n",
       "2      2     Hickory Park Solar Hybrid     63522.0  31.269676 -84.176321   \n",
       "3      3  Camilla Solar Energy Project     61785.0  31.269550 -84.228721   \n",
       "4      4              SR DeSoto I, LLC     64841.0  31.886486 -84.021586   \n",
       "\n",
       "   ac_capacity  dc_capacity  dc_loss_factor  temperature_coefficient tracking  \\\n",
       "0        213.0        298.0            0.15                    -0.35        Y   \n",
       "1        204.0        261.0            0.15                    -0.25        Y   \n",
       "2        195.5        268.8            0.15                    -0.35        Y   \n",
       "3        171.4        224.0            0.15                    -0.35        Y   \n",
       "4        165.0        177.4            0.15                    -0.25        Y   \n",
       "\n",
       "   ... fixed_azimuth axis_tilt axis_azimuth  backtrack  max_rotation_angle  \\\n",
       "0  ...           180       0.0        180.0       True                60.0   \n",
       "1  ...             0       0.0        180.0      False                60.0   \n",
       "2  ...                     0.0        180.0       True                60.0   \n",
       "3  ...             0       0.0        180.0       True                60.0   \n",
       "4  ...           180       0.0        180.0      False                60.0   \n",
       "\n",
       "  ground_coverage_ratio  Crystalline Silicon?  Thin-Film (CdTe)?  \\\n",
       "0                  0.35                     Y                NaN   \n",
       "1                  0.45                   NaN                  Y   \n",
       "2                  0.35                     Y                NaN   \n",
       "3                  0.35                     Y                NaN   \n",
       "4                  0.45                   NaN                  Y   \n",
       "\n",
       "  Operating_Month Operating_Year  \n",
       "0            12.0         2021.0  \n",
       "1             9.0         2020.0  \n",
       "2             6.0         2022.0  \n",
       "3             2.0         2020.0  \n",
       "4            12.0         2022.0  \n",
       "\n",
       "[5 rows x 22 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plants_df = pd.read_csv('pv_fleet_metadata_rev3.csv') # read in as a Pandas DataFrame\n",
    "plants_df = plants_df[plants_df['ac_capacity']>=10].copy().reset_index()\n",
    "num_plants = len(plants_df.index) # number of plants\n",
    "print('number of plants: ' + str(num_plants))\n",
    "plants_df.head() # preview the DataFrame"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here's a plot of the plants using `cartopy`, with the size of the marker corresponding to the AC capacity of each plant."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.figure(figsize=(6, 6))\n",
    "    \n",
    "ax = fig.add_subplot(1,1,1, projection=ccrs.PlateCarree())\n",
    "\n",
    "resolution = '50m'\n",
    "\n",
    "ax.add_feature(cfeature.LAND.with_scale(resolution), facecolor='lightgrey', zorder=0)\n",
    "ax.add_feature(cfeature.STATES.with_scale(resolution), facecolor='none', edgecolor='black', lw=0.1, zorder=0)\n",
    "ax.coastlines(resolution=resolution, edgecolor='black', lw=0.4)\n",
    "\n",
    "# ax.set_extent([-95, -79, 24, 38])\n",
    "ax.set_extent([-91, -80.5, 27, 35.5])\n",
    "\n",
    "marker_scaling = 2\n",
    "\n",
    "params = {'color': 'red', 'edgecolors': 'black', 'alpha': 0.6}\n",
    "\n",
    "plt.scatter(x=plants_df.longitude, y=plants_df.latitude,\n",
    "            s=plants_df.ac_capacity*marker_scaling,\n",
    "            transform=ccrs.PlateCarree(),\n",
    "            **params\n",
    ")\n",
    "\n",
    "for size in [20, 100, 200]:\n",
    "    plt.scatter(0, 0, s=size*marker_scaling, transform=ccrs.PlateCarree(), **params, label=size)\n",
    "\n",
    "ax.legend(loc='lower left', title='Plant size [MW]', frameon=False, ncol=3,\n",
    "            borderaxespad=1, labelspacing=1, columnspacing=0.5, handletextpad=0.3)\n",
    "\n",
    "# plt.savefig(\"site_map.png\", dpi=150, bbox_inches='tight') # <-- uncomment this to save the image\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Irradiance data\n",
    "\n",
    "Now let's get some irradiance data from the NREL NSRDB [1] using `pvlib` [2]. We will use this to as a proxy for measured data from a real fleet of plants. The NSRDB requires an API key from the NREL Developer Network [3]. We will use the `pvlib.iotools.get_psm3` function - details are here https://pvlib-python.readthedocs.io/en/stable/reference/generated/pvlib.iotools.get_psm3.html.\n",
    "\n",
    "We will download the NSRDB data and metadata once and then store it in pickle files, so the next time we run it we can simply open the pickle file. If the list of plants changes after running the first time, you will need to delete the pickle files and run this again. \n",
    "\n",
    "Since data in the hourly HRRR model represent values on the hour (not an average across the hour), we want to make sure our NSRDB-based calculations represent a similar time interval. In the 30-minute interval version of the NSRDB PSM3, every other value represents the average of seven 5-min interval measurements centered at the top of the hour:\n",
    "\n",
    "```\n",
    "Timestamps: 00 05 10 15 20 25 30 35 40 45 50 55 00 05 10 15 20 25\n",
    "Interval 1:          --------------------\n",
    "Interval 2:                            --------------------\n",
    "```\n",
    "\n",
    "(from [4])\n",
    "\n",
    "We will pull the 30-min data and just save measurements with `Minute` of 0 (centered at the top of the hour).\n",
    "\n",
    "*Note: For about 50 plants, this takes about 5-10 minutes to run. I'm not sure how much this is limited by my internet connection (~300 Mb/s) or NREL servers.* \n",
    "\n",
    "[1] https://nsrdb.nrel.gov/.\n",
    "\n",
    "[2] William F. Holmgren, Clifford W. Hansen, and Mark A. Mikofski. “pvlib python: a python package for modeling solar energy systems.” Journal of Open Source Software, 3(29), 884, (2018). https://doi.org/10.21105/joss.00884\n",
    "\n",
    "[3] https://developer.nrel.gov/signup/\n",
    "\n",
    "[4] https://assessingsolar.github.io/unofficial-psm3-userguide/pages/data-aggregation.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "loaded dictionaries from existing pickle files\n"
     ]
    }
   ],
   "source": [
    "year = 2021\n",
    "interval = 30\n",
    "\n",
    "## API key credentials are stored in a separate text file\n",
    "with open('nrel_api_key_details.txt') as f: # full path and filename of text file with email on first line and api key on second line\n",
    "    email = f.readline().rstrip()  # rstrip removes trailing line breaks and spaces\n",
    "    nrel_api_key = f.readline().rstrip()\n",
    "\n",
    "nsrdb_data_dict = {}\n",
    "nsrdb_metadata_dict = {}\n",
    "\n",
    "data_filename = 'nsrdb_data_dictionary_pickle_' + str(year) + '_60min_top_of_hour_from_30min.pickle'\n",
    "metadata_filename = 'nsrdb_metadata_dictionary_pickle_' + str(year) + '_60min_top_of_hour_from_30min.pickle'\n",
    "\n",
    "try: \n",
    "    # try to load dictionaries from existing pickle files\n",
    "    pickle_in = open('example_data/'+data_filename, 'rb')\n",
    "    nsrdb_data_dict = pickle.load(pickle_in)\n",
    "    pickle_in = open('example_data/'+metadata_filename, 'rb')\n",
    "    nsrdb_metadata_dict = pickle.load(pickle_in)\n",
    "    print('loaded dictionaries from existing pickle files')\n",
    "except: # if the pickle files don't exist, get NSRDB data, store in dictionaries, save as pickle files\n",
    "    print('downloading new data from the NSRDB')\n",
    "    for x in range(0,num_plants):\n",
    "        lat = plants_df.iloc[x]['latitude'] \n",
    "        lon = plants_df.iloc[x]['longitude']\n",
    "        nsrdb_data_dict[x], nsrdb_metadata_dict[x] = pvlib.iotools.get_psm3(latitude=lat,\n",
    "                                                longitude=lon,\n",
    "                                                api_key=nrel_api_key,\n",
    "                                                email=email,\n",
    "                                                names=year,\n",
    "                                                interval=interval,\n",
    "                                                leap_day=True,\n",
    "                                                map_variables=True\n",
    "                                                )\n",
    "\n",
    "        # convert to UTC\n",
    "        nsrdb_data_dict[x].index = nsrdb_data_dict[x].index.tz_convert('UTC')\n",
    "\n",
    "        # only keep values that are centered on top of the hour \n",
    "        # to match HRRR hourly. See \n",
    "        # https://assessingsolar.github.io/unofficial-psm3-userguide/pages/data-aggregation.html\n",
    "        nsrdb_data_dict[x] = nsrdb_data_dict[x].loc[nsrdb_data_dict[x]['Minute'] == 0]\n",
    "\n",
    "        # adjust metadata to match\n",
    "        nsrdb_metadata_dict[x]['Time Zone'] = 0\n",
    "\n",
    "        # save dictionaries as pickle files\n",
    "        pickle_out = open('example_data/'+data_filename, 'wb')\n",
    "        pickle.dump(nsrdb_data_dict, pickle_out)\n",
    "        pickle_out.close()\n",
    "\n",
    "        pickle_out = open('example_data/'+metadata_filename, 'wb')\n",
    "        pickle.dump(nsrdb_metadata_dict, pickle_out)\n",
    "        pickle_out.close()\n",
    "    print('done')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Modeling \"observed\" power\n",
    "\n",
    "Let's use the NSRDB data to calculate ac power for each plant using the PVWatts model in `pvlib`. Remember, this is a proxy for measured ac power from real plants. \n",
    "\n",
    "We will also model clear sky power profiles using clear sky irradiance calculated with `pvlib`. See https://pvlib-python.readthedocs.io/en/stable/reference/clearsky.html. \n",
    "\n",
    "*Note: Clear sky irradiance is also available from the NSRDB - adding the option `attributes=[]` to `get_psm3()` above would retrieve those values. We are going to create out own clear sky profiles to more closely match how this might work in an operational environment with real plant data instead of NSRDB-based data.* \n",
    "\n",
    "Areas for improvement:\n",
    " - Add row-to-row self-shade for fixed tilt and single axis tracking *without backtracking* (i.e., thin-film truetracking). See:\n",
    "   - https://github.com/pvlib/pvlib-python/discussions/1432#discussioncomment-7209858\n",
    "   - https://github.com/pvlib/pvlib-python/pull/1725\n",
    "   - https://github.com/pvlib/pvlib-python/issues/1689\n",
    "   - https://github.com/pvlib/pvlib-python/issues/1690"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We will create a function to run our power model that can be used later with forecasted weather. Inputs will be a _resource_data_ dataframe, a _plant_data_ dictionary (with a single plant's specifications pulled from _plants_df_), and then a boolean value for whether or not we want to calculate clear sky power."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def pvlib_power_model(resource_data, plant_data, calc_clear_sky = False):\n",
    "    transposition_model = 'perez'\n",
    "    eta_inv_nom = 0.98\n",
    "\n",
    "    # solar position\n",
    "    lat = plant_data['latitude'] \n",
    "    lon = plant_data['longitude']\n",
    "    times = resource_data.index \n",
    "    loc= pvlib.location.Location(latitude=lat, longitude=lon, tz=times.tz)\n",
    "    solar_position = loc.get_solarposition(times)\n",
    "\n",
    "    # dni and airmass\n",
    "    dni_extra = pvlib.irradiance.get_extra_radiation(resource_data.index)\n",
    "    airmass = pvlib.atmosphere.get_relative_airmass(\n",
    "        solar_position.apparent_zenith,\n",
    "        model='gueymard2003'\n",
    "    )\n",
    "\n",
    "    if calc_clear_sky == True:\n",
    "        # simlified solis with aod700=0.05 and precipitable_water=0.5 appears to give \n",
    "        # good results; goal is to set an upper-bound, not necessarily to be as accurate \n",
    "        # as possible\n",
    "        clear_sky = loc.get_clearsky(times,\n",
    "                                     model='simplified_solis',\n",
    "                                     dni_extra=dni_extra,\n",
    "                                     aod700 = 0.05,\n",
    "                                     precipitable_water = 0.5)\n",
    "    \n",
    "    # surface tilt and azimuth\n",
    "    if plant_data['tracking_type'] == 'single_axis':  \n",
    "        # tracker orientation angles\n",
    "        singleaxis_kwargs = dict(apparent_zenith=solar_position.apparent_zenith,\n",
    "                                 apparent_azimuth=solar_position.azimuth,\n",
    "                                 axis_tilt=plant_data['axis_tilt'],\n",
    "                                 axis_azimuth=plant_data['axis_azimuth'],\n",
    "                                 backtrack=plant_data['backtrack'],\n",
    "                                 gcr=plant_data['ground_coverage_ratio'],\n",
    "                                 )\n",
    "        orientation = pvlib.tracking.singleaxis(max_angle=plant_data['max_rotation_angle'],\n",
    "                                                **singleaxis_kwargs)\n",
    "        surface_tilt = orientation.surface_tilt.fillna(0)\n",
    "        surface_azimuth = orientation.surface_azimuth.fillna(0)\n",
    "    elif plant_data['tracking_type'] == 'fixed':\n",
    "        surface_tilt = float(plant_data['fixed_tilt'])\n",
    "        surface_azimuth = float(plant_data['fixed_azimuth'])\n",
    "\n",
    "    # Transposed components of POA\n",
    "    total_irradiance = pvlib.irradiance.get_total_irradiance(\n",
    "        surface_tilt, \n",
    "        surface_azimuth,\n",
    "        solar_position.apparent_zenith, \n",
    "        solar_position.azimuth, \n",
    "        resource_data.dni, \n",
    "        resource_data.ghi, \n",
    "        resource_data.dhi, \n",
    "        airmass=airmass,\n",
    "        albedo=resource_data.albedo, \n",
    "        dni_extra=dni_extra, \n",
    "        model=transposition_model\n",
    "    )\n",
    "\n",
    "    # steady state cell temperature - faiman is much faster than fuentes, simpler than sapm\n",
    "    t_cell = pvlib.temperature.faiman(total_irradiance.poa_global,\n",
    "                                      resource_data.temp_air,\n",
    "                                      resource_data.wind_speed,\n",
    "    )\n",
    "\n",
    "    # PVWatts dc power\n",
    "    pdc = pvlib.pvsystem.pvwatts_dc(total_irradiance.poa_global,\n",
    "                                    t_cell,\n",
    "                                    plant_data['dc_capacity'],\n",
    "                                    plant_data['temperature_coefficient']/100)\n",
    "    pdc0 = plant_data['ac_capacity']/eta_inv_nom # inverter dc input is ac nameplate divided by nominal inverter efficiency\n",
    "    pdc_inv = pdc*(1-plant_data['dc_loss_factor']) # dc power into the inverter is modeled pdc after losses\n",
    "\n",
    "    # PVWatts ac power\n",
    "    power_ac = pvlib.inverter.pvwatts(pdc_inv, pdc0, eta_inv_nom)\n",
    "\n",
    "    if calc_clear_sky==True:\n",
    "        # Transposed components of POA\n",
    "        total_irradiance = pvlib.irradiance.get_total_irradiance(\n",
    "            surface_tilt, \n",
    "            surface_azimuth,\n",
    "            solar_position.apparent_zenith, \n",
    "            solar_position.azimuth, \n",
    "            clear_sky.dni, \n",
    "            clear_sky.ghi, \n",
    "            clear_sky.dhi, \n",
    "            airmass=airmass,\n",
    "            albedo=resource_data.albedo, \n",
    "            dni_extra=dni_extra, \n",
    "            model=transposition_model\n",
    "        )\n",
    "\n",
    "        # steady state cell temperature - faiman is much faster than fuentes, simpler than sapm\n",
    "        t_cell = pvlib.temperature.faiman(total_irradiance.poa_global,\n",
    "                                        resource_data.temp_air,\n",
    "                                        resource_data.wind_speed,\n",
    "        )\n",
    "\n",
    "        # PVWatts dc power\n",
    "        pdc = pvlib.pvsystem.pvwatts_dc(total_irradiance.poa_global,\n",
    "                                        t_cell,\n",
    "                                        plant_data['dc_capacity'],\n",
    "                                        plant_data['temperature_coefficient']/100)\n",
    "        pdc0 = plant_data['ac_capacity']/eta_inv_nom # inverter dc input is ac nameplate divided by nominal inverter efficiency\n",
    "        pdc_inv = pdc*(1-plant_data['dc_loss_factor']) # dc power into the inverter is modeled pdc after losses\n",
    "\n",
    "        # PVWatts ac power\n",
    "        power_ac_clear_sky = pvlib.inverter.pvwatts(pdc_inv, pdc0, eta_inv_nom)\n",
    "\n",
    "    if calc_clear_sky == True:\n",
    "        return power_ac, power_ac_clear_sky\n",
    "    else:\n",
    "        return power_ac"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here's a quick test of the function, manually specifying a single plant from the list:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAigAAAGjCAYAAAAGku4DAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8fJSN1AAAACXBIWXMAAA9hAAAPYQGoP6dpAABPY0lEQVR4nO3deVxU1f8/8NdlGxgERBSQAkXFFROX3BXMcl/KytI+ftwyt1Ryza+puIGoKaXpx0rF8lNqn8w2d1PUTEOUUlFRQ8OFD7+MAB0cBub8/uDDzZFd5zJ3htfz8ZjHgzn3zL3nHg533pxz7zmSEEKAiIiISEXsLF0AIiIioocxQCEiIiLVYYBCREREqsMAhYiIiFSHAQoRERGpDgMUIiIiUh0GKERERKQ6DFCIiIhIdRwsXYBHYTQacevWLbi5uUGSJEsXh4iIiMpBCIHs7Gz4+fnBzq70PhKrDFBu3boFf39/SxeDiIiIHkFqaiqefPLJUvNYZYDi5uYGoOAE3d3dLVwaIiIiKo+srCz4+/vL3+OlscoApXBYx93dnQEKERGRlSnP7RkVvkn2yJEj6N+/P/z8/CBJEnbu3Fli3rFjx0KSJMTExJik6/V6TJo0CTVr1oSrqysGDBiAGzduVLQoREREZKMqHKDcu3cPLVq0wJo1a0rNt3PnTpw8eRJ+fn5FtoWHh+Orr77C1q1bcezYMdy9exf9+vVDfn5+RYtDRERENqjCQzy9e/dG7969S81z8+ZNvPnmm9i7dy/69u1rsi0zMxMbNmzAp59+imeffRYAsGXLFvj7++PAgQPo2bNnRYtERERENsbs86AYjUYMGzYMM2bMQLNmzYpsT0hIgMFgQI8ePeQ0Pz8/BAcH4/jx48XuU6/XIysry+RFREREtsvsAUp0dDQcHBwwefLkYrenpaXByckJnp6eJuk+Pj5IS0sr9jNRUVHw8PCQX3zEmIiIyLaZNUBJSEjAe++9h9jY2ApPoCaEKPEzs2fPRmZmpvxKTU01R3GJiIhIpcwaoBw9ehTp6ekICAiAg4MDHBwccP36dUybNg1169YFAPj6+iI3NxcZGRkmn01PT4ePj0+x+9VoNPIjxXy0mIiIyPaZNUAZNmwYfv31VyQmJsovPz8/zJgxA3v37gUAtG7dGo6Ojti/f7/8udu3b+PcuXPo2LGjOYtDREREVqrCT/HcvXsXV65ckd+npKQgMTERNWrUQEBAALy8vEzyOzo6wtfXF40aNQIAeHh4YPTo0Zg2bRq8vLxQo0YNTJ8+Hc2bN5ef6iEiIqKqrcIByqlTp9CtWzf5/dSpUwEAw4cPR2xsbLn2sWrVKjg4OGDw4MHIyclB9+7dERsbC3t7+4oWh4iIiGyQJIQQli5ERWVlZcHDwwOZmZm8H4WIiB6ZEAI5hpInCXVxtK/wQx9Usop8f1vlWjxERESPSwiBl/71ExKuZ5SYp00dT3wxrgODFAsw+zwoRERE1iDHkF9qcAIAp65nlNrDQsphDwoREVV5p955Flqnv++D1OXmo83iAxYsETFAISKiKk/rZA+tE78S1YRDPERERKQ6DFCIiIhIdRigEBERkepwwI2IiEhFODdLAQYoREREKsG5Wf7GAMVCGCETEdHDKjI3i60/dWTbZ6dSjJCJiKgsVX1uFgYoFsAImYiISibgAj20uA+tydd0HlxwHznQQJdbfA+8LfW+89vPwqp6hExERA8QAv9xWoA2dsnACtNNWgAXnIF4Y0O0WQwARQMRW+p9Z4BiYZy9kIiIZAZdQXBSiqftkuGFLOigKbLt1PU/bab33frPgMyGN+5SRZTVXgC2GbINlhpO0U25CK2r+98JuTpgRQMAQILz+GI/E29sCIieipWpMjFAIQC8cZcqpjztBWCbIdtQ0rC74u3bUQs4uZq+928PpJ4o8SNP2yVDZ9ABGg9lylSJGKAQAN64SxVTnvYCsM2Q9XJxtEebOp44VUo7r/T2LUnAqD2AQVdkk+5eFrTvNa6cclQSXjWoCN64SxXxcHsBLNdmOOxE5iJJEr4Y16HY9mTRa6IkmfaqFMrNq/yyKIwBChXBG3epItTSXjjsROYmSZIq2nZVxZqnCinpZjGA/5mSZXHYici28C+UKqS0bk3+Z0pqoaZhJyJ6NHaWLgCpX+HNYmUp/M+UyNIKh51MX/Zlf5CIVIM9KFSm0m4WA/ifKVmfqjBNOJG1Y4CiYmq634M3i5Etsdi8FkQo/WkzXW4+tJVcHrXiN46K8X4PIvNR5bwWVOWU9bSZC+7jgnMlF0ql+FeoMuW5iAK8kBJVlGrntaAqpeBpsz/hAn2x27UPpLs4Vu37pvjtpjK834NsjZru9+BQJVncg6sVl6Gq95DzL1WFeBElW8L7PYgeUI7VigEUrLnjWLXvRuG3IBGZHe/3ICpbkdWKH+SoLZjWvgrjlYGIzI73exCVw8OrFZMJBihVSFmPthGZE4cqqSpQ03QQtoZXjyqivAupERFR+XE6COVUeKr7I0eOoH///vDz84MkSdi5c6e8zWAwYNasWWjevDlcXV3h5+eHf/7zn7h165bJPvR6PSZNmoSaNWvC1dUVAwYMwI0bNx77ZKhk5V1IrU0dzyr/aBupmxACuty8Yl7sBaTKweU/KkeFe1Du3buHFi1aYOTIkXjxxRdNtul0Opw+fRpz585FixYtkJGRgfDwcAwYMACnTp2S84WHh+Pbb7/F1q1b4eXlhWnTpqFfv35ISEiAvT2/HJVW3EJqhdglSWrGnkBSA04HUTkqHKD07t0bvXv3Lnabh4cH9u/fb5K2evVqtG3bFr///jsCAgKQmZmJDRs24NNPP8Wzzz4LANiyZQv8/f1x4MAB9OzZ8xFOgyqicCE1ImtTnp5A9gJSZSjvPVYP9+xxKvvyU/xbKjMzE5IkoXr16gCAhIQEGAwG9OjRQ87j5+eH4OBgHD9+nAEKEZVLST2B7AUkNXm4J4VT2ZefogHK/fv38fbbb2Po0KFwdy941jstLQ1OTk7w9DQdv/Px8UFaWlqx+9Hr9dDr/57+NysrS7lCE5FVYE8gqVV5lyxhT1/pFPvrNhgMePXVV2E0GrF27doy8wshSvyvJyoqCgsWLDB3EYmIiMxOkiR8MbY9cnTZRTcadMB7f+ejkikSoBgMBgwePBgpKSn44Ycf5N4TAPD19UVubi4yMjJMelHS09PRsWPHYvc3e/ZsTJ06VX6flZUFf39/JYpuVdS0xgkREf2PEJA29YI29aSlS2LVzB6gFAYnly9fxqFDh+Dl5WWyvXXr1nB0dMT+/fsxePBgAMDt27dx7tw5LFu2rNh9ajQaaDQacxfV6nGNEyIiFTLogLKCE661U6YKByh3797FlStX5PcpKSlITExEjRo14Ofnh5deegmnT5/Gd999h/z8fPm+kho1asDJyQkeHh4YPXo0pk2bBi8vL9SoUQPTp09H8+bN5ad6qGRc44SIyIpMvwI4FROIcK2dMlX4G+zUqVPo1q2b/L5w6GX48OGIiIjAN998AwAICQkx+dyhQ4cQFhYGAFi1ahUcHBwwePBg5OTkoHv37oiNjeUcKOXANU7IlnCYkmyeE9fbeVQVDlDCwsIghChxe2nbCjk7O2P16tVYvXp1RQ9PKN/z98U9e0+kNhymJKVxDTLrxTEAG8WeFFIrDlNSZVF05mEhCu41KU5uCelUIfzrtyHlufBzlk2yNA5TUmVRbA0yIYCNPcu+EZYeCwMUG1LW+hAAx/ZJHco7TTiRuZh1DbLyPKUD8Emdx8QrhI3hhZ+IqCjFZh4u6SkdgE/qPCZ+kxGRKhV3AyNvaiTV4VM6imGAQkSqxHtRiKo2O0sXgIioUOGN3mXhzd5Eto89KArhs/dEFVeeG70B3uxNVBUwQFGAos/eE9k43uhNRACHeBSh2LP3REREVQT/TVGYWZ+9J6pkJQ1VcpiSiJTGAEVhij17T6QwDlUSkSVxiIeIilWeoUoOUxKRUvivPRGVqaShSg5TEpFSGKAQUZk4VElElY1DPERERKQ6DFCIiIhIdRigEBERkeowQCEiIiLVYYBCREREqsMAhYiIiFSHAQoRERGpDgMUIiIiUh0GKERERKQ6DFCIiIhIdRigEBERkeowQCEiIiLVYYBCREREqsMAhYiIiFSHAQoRERGpDgMUIiIiUh0GKERERKQ6DFCIiIhIdSocoBw5cgT9+/eHn58fJEnCzp07TbYLIRAREQE/Pz+4uLggLCwM58+fN8mj1+sxadIk1KxZE66urhgwYABu3LjxWCdCREREtqPCAcq9e/fQokULrFmzptjty5Ytw8qVK7FmzRrEx8fD19cXzz33HLKzs+U84eHh+Oqrr7B161YcO3YMd+/eRb9+/ZCfn//oZ0JEREQ2w6GiH+jduzd69+5d7DYhBGJiYjBnzhwMGjQIALB582b4+Pjgs88+w9ixY5GZmYkNGzbg008/xbPPPgsA2LJlC/z9/XHgwAH07NnzMU6HiIiIbIFZ70FJSUlBWloaevToIadpNBqEhobi+PHjAICEhAQYDAaTPH5+fggODpbzPEyv1yMrK8vkRURERLbLrAFKWloaAMDHx8ck3cfHR96WlpYGJycneHp6lpjnYVFRUfDw8JBf/v7+5iw2ERERqYwiT/FIkmTyXghRJO1hpeWZPXs2MjMz5VdqaqrZykpERETqY9YAxdfXFwCK9ISkp6fLvSq+vr7Izc1FRkZGiXkeptFo4O7ubvIiIiIi22XWACUwMBC+vr7Yv3+/nJabm4u4uDh07NgRANC6dWs4Ojqa5Ll9+zbOnTsn5yEiIqKqrcJP8dy9exdXrlyR36ekpCAxMRE1atRAQEAAwsPDERkZiaCgIAQFBSEyMhJarRZDhw4FAHh4eGD06NGYNm0avLy8UKNGDUyfPh3NmzeXn+ohIiKiqq3CAcqpU6fQrVs3+f3UqVMBAMOHD0dsbCxmzpyJnJwcTJgwARkZGWjXrh327dsHNzc3+TOrVq2Cg4MDBg8ejJycHHTv3h2xsbGwt7c3wykRERGRtatwgBIWFgYhRInbJUlCREQEIiIiSszj7OyM1atXY/Xq1RU9PBEREVUBXIuHiIiIVIcBChEREakOAxQiIiJSHQYoREREpDoMUIiIiEh1GKAQERGR6jBAISIiItVhgEJERESqwwCFiIiIVIcBChEREakOAxQiIiJSHQYoREREpDoMUIiIiEh1GKAQERGR6jBAISIiItVhgEJERESqwwCFiIiIVIcBChEREakOAxQiIiJSHQYoREREpDoMUIiIiEh1GKAQERGR6jBAISIiItVhgEJERESqwwCFiIiIVIcBChEREakOAxQiIiJSHQdLF4CIiMhihAAMupK3O2oBSaq88pCMAQoREVVNQgAbewKpJ0vO498eGLWHQYoFcIiHiIiqJoOu9OAEAFJPlN7DQophDwoREdH0K4CT9u/3uTpgRQPLlYcYoBAREcFJCzi5WroU9ACzD/Hk5eXhnXfeQWBgIFxcXFCvXj0sXLgQRqNRziOEQEREBPz8/ODi4oKwsDCcP3/e3EUhIiIiK2X2ACU6Ohr/+te/sGbNGly4cAHLli3D8uXLsXr1ajnPsmXLsHLlSqxZswbx8fHw9fXFc889h+zsbHMXh4iIiKyQ2QOUn376CQMHDkTfvn1Rt25dvPTSS+jRowdOnToFoKD3JCYmBnPmzMGgQYMQHByMzZs3Q6fT4bPPPjN3cYiIiMgKmT1A6dy5Mw4ePIjk5GQAwC+//IJjx46hT58+AICUlBSkpaWhR48e8mc0Gg1CQ0Nx/PhxcxeHiIiIrJDZb5KdNWsWMjMz0bhxY9jb2yM/Px9LlizBkCFDAABpaWkAAB8fH5PP+fj44Pr168XuU6/XQ6/Xy++zsrLMXWwiIiJSEbP3oGzbtg1btmzBZ599htOnT2Pz5s1YsWIFNm/ebJJPemjSGyFEkbRCUVFR8PDwkF/+/v7mLjYRERGpiNl7UGbMmIG3334br776KgCgefPmuH79OqKiojB8+HD4+voCKOhJqV27tvy59PT0Ir0qhWbPno2pU6fK77OyshikkNXIz8+HwWCwdDEqTJ+bhyfc7At+vn8fdkbbnZXAVs7V0dER9vb2li4GkVmY/a9Qp9PBzs60Y8be3l5+zDgwMBC+vr7Yv38/WrZsCQDIzc1FXFwcoqOji92nRqOBRqMxd1GJFCWEQFpaGv766y9LF+WRGIVARDdvAMCtG7/Dzoan+ralc61evTp8fX1L7JEmshZmD1D69++PJUuWICAgAM2aNcOZM2ewcuVKjBo1CkDB0E54eDgiIyMRFBSEoKAgREZGQqvVYujQoeYuDpHFFAYn3t7e0Gq1VveFkW8UyEsvePS/rrcb7O2sq/wVYQvnKoSATqdDeno6AJj0UBNZI7MHKKtXr8bcuXMxYcIEpKenw8/PD2PHjsW8efPkPDNnzkROTg4mTJiAjIwMtGvXDvv27YObm5u5i0NkEfn5+XJw4uXlZeniPJJ8o4DkUHBzurOzs1V+aZeXrZyri4sLgIIhc29vbw73kFUze4Di5uaGmJgYxMTElJhHkiREREQgIiLC3IcnUoXCe060Wm0ZOYnMq7DNGQwGBihk1biaMZGCrG1Yh6wf2xzZCgYoREREpDoMUIioXK5duwZJkpCYmGjpohRRt27dUoeVicj6MEAhIiIi1WGAQkSqYo2T2llCfn6+PL8UkS1igEJEMqPRiOjoaDRo0ABaF2f0bBeMj95fUWL+pKQk9OnTB9WqVYOPjw+GDRuGP/74Q96+Z88edO7cGdWrV4eXlxf69euHq1evytsLh422b9+OsLAwODs7Y8uWLcUeKyIiAgEBAdBoNPDz88PkyZNLLNemTZvg4eGB/fv3o0GDBlixwvQczp07Bzs7O5OyPGjEiBF4/vnnsWDBAnh7e8Pd3R1jx45Fbm6unEev12Py5Mnw9vaGs7MzOnfujPj4eHl769at8e6778rvn3/+eTg4OMhriaWlpUGSJFy6dAlAwYSVM2fOxBNPPAFXV1e0a9cOhw8flj8fGxuL6tWr47vvvkPTpk2h0WhKXL+MyBYwQCGqBEII6HLzLPISQpS7nLNnz0Z0dDTmzp2Ls+fOI2r1R/Cq5V1s3tu3byM0NBQhISE4deoU9uzZg//+978YPHiwnOfevXuYOnUq4uPjcfDgQdjZ2eGFF14o8p//rFmzMHnyZFy4cAE9e/Yscqz//Oc/WLVqFdavX4/Lly9j586daN68ebHlWrFiBaZPn469e/fiueeew6hRo7Bp0yaTPBs3bkSXLl1Qv379Euvi4MGDuHDhAg4dOoTPP/8cX331FRYsWCBvnzlzJr788kts3rwZp0+fRoMGDdCzZ0/8+eefAICwsDA5wBBC4OjRo/D09MSxY8cAAIcOHYKvry8aNWoEABg5ciR+/PFHbN26Fb/++itefvll9OrVC5cvX5aPqdPpEBUVhY8//hjnz5+Ht3fxvxt6kIAL7gO594p56SxdOCqFdS44QWRlcgz5aDpvr0WOnbSwJ7ROZf+pZ2dn47333sOaNWswfPhw5BsF7rvURKu2HYrNv27dOrRq1QqRkZFy2saNG+Hv74/k5GQ0bNgQL774oslnNmzYAG9vbyQlJSE4OFhODw8Px6BBg0os2++//w5fX188++yzcHR0REBAANq2bVsk3+zZs7F582YcPnxYDmBGjhyJefPm4eeff0bbtm1hMBiwZcsWLF++vNT6cHJywsaNG6HVatGsWTMsXLgQM2bMwKJFi5CTk4N169YhNjYWvXv3BgB89NFH2L9/PzZs2IAZM2YgLCwMGzZsgNFoxNmzZ2Fvb49//OMfOHz4MPr06YPDhw8jNDQUAHD16lV8/vnnuHHjBvz8/AAA06dPx549e7Bp0ya5jg0GA9auXYsWLVqUWnb6HyHwH6cFaGOXDJTcEUgqxR4UIgIAXLhwAXq9Ht27dy9X/oSEBBw6dAjVqlWTX40bNwYAeejk6tWrGDp0KOrVqwd3d3cEBgYCKAg4HtSmTZtSj/Xyyy8jJycH9erVw5gxY/DVV18hLy/PJM+7776L9evX49ixYya9K7Vr10bfvn2xceNGAMB3332H+/fv4+WXXy71mC1atDCZaK9Dhw64e/cuUlNTcfXqVRgMBnTq1Ene7ujoiLZt2+LChQsAgK5duyI7OxtnzpxBXFwcQkND0a1bN8TFxQGASYBy+vRpCCHQsGFDk/qMi4szGYZycnLCU089VWq56QEGXUFwUhb/9oAjJ1VUG/agEFUCF0d7JC0sOnRRWccuV77/TZNeXkajEf379y92kc/CdWD69+8Pf39/fPTRR/Dz84PRaERwcLDJvRwA4OrqWuqx/P39cenSJezfvx8HDhzAhAkTsHz5csTFxcHR0REA0KVLF3z//ffYvn073n77bZPPv/766xg2bBhWrVqFTZs24ZVXXnnkWX4lSZKHzR6eFE0IIad5eHggJCQEhw8fxvHjx/HMM8+gS5cuSExMxOXLl5GcnIywsDAABXVpb2+PhISEIrO/VqtWTf7ZxcWFE7E9It2Ui9C6uhe/0VELsF5VhwEKUSWQJKlcwyyWFBQUBBcXFxw8eBCvv/56mflbtWqFL7/8EnXr1oWDQ9Fzu3PnDi5cuID169ejS5cuACDff/EoXFxcMGDAAAwYMAATJ05E48aNcfbsWbRq1QoA0LZtW0yaNAk9e/aEvb09ZsyYIX+2T58+cHV1xbp167B7924cOXKkzOP98ssvyMnJkQO3EydOoFq1anjyySfh5eUFJycnHDt2TF7k1GAw4NSpUwgPD5f3ERYWhkOHDuHkyZNYuHAhqlevjqZNm2Lx4sXw9vZGkyZNAAAtW7ZEfn4+0tPT5boiM3PUAk6lB8KkLhziISIABYvkzZo1CzNnzsQnn3yCq1ev4tfT8dix9dNi80+cOBF//vknhgwZgp9//hm//fYb9u3bh1GjRiE/Px+enp7w8vLChx9+iCtXruCHH37A1KlTH6lssbGx2LBhA86dO4fffvsNn376KVxcXFCnTh2TfB06dMDu3buxcOFCrFq1Sk63t7fHiBEjMHv2bDRo0AAdOhR/X82DcnNzMXr0aCQlJWH37t2YP38+3nzzTdjZ2cHV1RXjx4/HjBkzsGfPHiQlJWHMmDHQ6XQYPXq0vI+wsDDs2bMHkiShadOmctq///1veXgHABo2bIjXXnsN//znP7Fjxw6kpKQgPj4e0dHR2LVr1yPVGZG1U/e/dERUqebOnQsHBwfMmzcPt27dQk1vH7z8j5HF5vXz88OPP/6IWbNmoWfPntDr9ahTpw569eoFOzs7SJKErVu3YvLkyQgODkajRo3w/vvvy8MaFVG9enUsXboUU6dORX5+Ppo3b45vv/222JWiO3XqhO+//x59+vSBvb29/Djy6NGjERkZiVGjRpXrmN27d0dQUBC6du0KvV6PV1991WSB06VLl8JoNGLYsGHIzs5GmzZtsHfvXnh6esp5unbtCgAIDQ2Vh2ZCQ0MRExNjEqAABY9GL168GNOmTcPNmzfh5eWFDh06oE+fPhWqKyJbIYmKPIOoEllZWfDw8EBmZibc3UsYU7QgXW6e/MRGeZ+gsGZV7XzL4/79+0hJSUFgYCCcnZ0tXZxHkm8UOH8rEwDQzM8D9nbWPUb/448/IiwsDDdu3ICPj4/JtofPdfSokfjrr7+wc+dOC5T08dhC2zMX3d1MaFcEFPw8/Xdoq3mU/8O594DIgieq8H+3TIeHSttmIY91rpWoIt/f/CYhIpum1+uRmpqKuXPnYvDgwUWCEyJSJ96DQkQ27fPPP0ejRo2QmZmJZcuWWbo4RFRO7EEhIps2YsQIjBgxokKfiY2NVaQsRFR+7EEhIiIi1WGAQkRERKrDAIWIiIhUh/egKEbABfqCx9FKqmZOr0xEpH4Pr3rMVZArBQMUJZR3BU3/9sCoPQxSiIjUbEUDS5egSuIQjxLKu4Jm6gnAwEiciEh1HLUF/0SWhqsgK4o9KAordgXNXB0jciIiNZOkgh7u0v6J5DC9otiDorTCFTRNXoy4SZ1GjBgBSZKwdOlSk/Svd+6U15IptH79erRo0QKurq6oXr06WrZsiejoaHl7REQEJEmCJEmws7ODn58fXnvtNaSmpprsJywsTM6n0WjwxBNPoH///tixY0eR8hXmkyQJDg4OCAgIwNSpU6HX681YC0T/I0nFXL8feDE4URQDFCIy4ezsjOjoaGRkZJSYZ8OGDZg6dSomT56MX375BT/++CNmzpyJu3fvmuRr1qwZbt++jRs3bmDbtm04e/YsBg8eXGR/Y8aMwe3bt3HlyhV8+eWXaNq0KV599VW88cYbRfJu2rQJt2/fRkpKCtauXYtPP/0UixcvfvwTt7Dc3FxLF4FIVRigEJGJZ599Fr6+vli6NKrEPN9++y0GDx6M0aNHo0GDBmjWrBmGDBmCRYsWmeRzcHCAr68v/Pz80KVLF4wZMwYnTpxAVlaWST6tVgtfX1/4+/ujffv2iI6Oxvr16/HRRx/hwIEDJnmrV68u5+3Xrx8GDBiA06dPl1jWa9euySsrd+zYEc7OzmjWrBkOHz5sku/UTz9iaL/u0Lo4o3bt2nj77beRl5cnn2/16tVhNBoBAImJiZAkCTNmzJA/P3bsWAwZMkR+f/z4cXTt2hUuLi7w9/fH5MmTce/ePXl73bp1sXjxYowYMQIeHh4YM2ZMiedAVBUxQCGqDEIUPHJuiVcFFyy3t7dHZGQkPlizBv+9fbPYPL6+vjhx4gSuX79e7v2mpaVhx44dsLe3h729fZn5hw8fDk9Pz2KHegolJyfj0KFDaNeuXZn7mzFjBqZNm4YzZ86gY8eOGDBgAO7cuQMAuHnzJiYOH4zgFi1x+kwi1q1bhw0bNsg9M127dkV2djbOnDkDAIiLi0PNmjURFxcn7//w4cMIDQ0FAJw9exY9e/bEoEGD8Ouvv2Lbtm04duwY3nzzTZMyLV++HMHBwUhISMDcuXPLPAeiqoQ3yRJVBoPu7+XZK9sjLAf/wgsvoEVICNa+uxQLVqwusn3+/PkYNGgQ6tati4YNG6JDhw7o06cPXnrpJdjZ/f1/z9mzZ1GtWjUYjUbk5OQAACZPngxX17LLY2dnh4YNG+LatWsm6UOGDIG9vT3y8vKg1+vRr18/zJ49u8z9vfnmm3jxxRcBAOvWrcOePXuwYcMGzJw5E+vWrYWv3xOYvXg5Gj9RHc2aNsGtW7cwa9YszJs3Dx4eHggJCcHhw4fRunVrHD58GG+99RYWLFiA7Oxs3Lt3D8nJyQgLCwNQEHgMHToU4eHhAICgoCC8//77CA0Nxbp16+Ds7AwAeOaZZzB9+vQyy05UFbEHhYiKFRW1FN/+53NcTb5YZFvt2rXx008/4ezZs5g8eTIMBgOGDx+OXr16ycMgANCoUSMkJiYiPj4eS5YsQUhICJYsWVLuMgghitycu2rVKiQmJuKXX37Bd999h+TkZAwbNqzMfXXo0EH+2cHBAW3atMGFCxcAABcvXMRTrZ42OVanTp1w9+5d3LhxA0DBzbyHDx+GEAJHjx7FwIEDERwcjGPHjuHQoUPw8fFB48aNAQAJCQmIjY1FtWrV5FfPnj1hNBqRkpIiH6NNmzblrguiqoY9KESVwVFb0JNhqWM/gq5du6Jj6DNYHb0Ib459vdg8wcHBCA4OxsSJE3Hs2DF06dIFcXFx6NatGwDAyckJDRoUPFLfrFkzXL58GePHj8enn35a5vHz8/Nx+fJlPP300ybpvr6+8j4bNWqE7OxsDBkyBIsXL5bTy6swICkuEBL/GxorTA8LC8OGDRvwyy+/wM7ODk2bNkVoaCji4uKQkZEhD+8AgNFoxNixYzF58uQixwwICJB/Lk9PElFVxQCFqDIUPq5oZaa8PR+De3VFq6ealpm3adOCPA/eCPqwuXPnomHDhnjrrbfQqlWrUve3efNmZGRkyMMyJSm8n6VwCKkkJ06cQNeuXQEAeXl5SEhIkO8JadK0CbZu/48clAAFN7m6ubnhiSeeAPD3fSgxMTEIDQ2FJEkIDQ1FVFQUMjIyMGXKFPmzrVq1wvnz5yscMBHR3zjEQ0QlCmrSDH1eeBkfrFljkj5+/HgsWrQIP/74I65fv44TJ07gn//8J2rVqmUylPKwevXqYeDAgZg3b55Juk6nQ1paGm7cuIGTJ09i1qxZGDduHMaPHy/3xhT666+/kJaWhlu3biEuLg4LFy5Ew4YN0aRJk1LP5YMPPsBXX32FixcvYuLEicjIyMCoUaP+dz4TkHbrJqLmzsTFixfx9ddfY/78+Zg6dap8T03hfShbtmyR7zXp2rUrTp8+bXL/CQDMmjULP/30EyZOnIjExERcvnwZ33zzDSZNmlRqGYnob4oEKDdv3sQ//vEPeHl5QavVIiQkBAkJCfJ2IQQiIiLg5+cHFxcXhIWF4fz580oUhYge08Tp/2fSswAUPIp84sQJvPzyy2jYsCFefPFFODs74+DBg/Dy8ip1f9OmTcP333+PkydPymkfffQRateujfr16+OFF15AUlIStm3bhrVr1xb5/MiRI1G7dm08+eSTGDJkCJo1a4bdu3fDwaH0DuGlS5ciOjoaLVq0wNGjR/H111+jZs2aAIAnnngCH2zejnOJp9GqZQjGjRuH0aNH45133jHZR7du3ZCfny8HI56enmjatClq1aplEiA99dRTiIuLw+XLl9GlSxe0bNkSc+fORe3atUstIxH9TRIPX3keU0ZGBlq2bIlu3bph/Pjx8Pb2xtWrV1G3bl3Ur18fABAdHY0lS5YgNjYWDRs2xOLFi3HkyBFcunQJbm5uZR4jKysLHh4eyMzMhLu7e5n5K5vubia0KwrGmXXTf4e2modphtx7fz/R8QhPWKiNLjcPTeftBQAkLewJrRNHDu/fv4+UlBQEBgbKT2xYm3yjwPlbmQCAZn4esLezzlkzr127hsDAQJw5cwYhISHF5rGVcwVso+2ZS5nXYhtiLedake9vs3+TREdHw9/fH5s2bZLT6tatK/8shEBMTAzmzJmDQYMGASgYa/bx8cFnn32GsWPHmrtIREREZGXMPsTzzTffoE2bNnj55Zfh7e2Nli1b4qOPPpK3p6SkIC0tDT169JDTNBoNQkNDcfz48WL3qdfrkZWVZfIiIiIi22X2AOW3337DunXrEBQUhL1792LcuHGYPHkyPvnkEwAFs0kCgI+Pj8nnfHx85G0Pi4qKgoeHh/zy9/c3d7GJyEbVrVsXQogSh3eISJ3MHqAYjUa0atUKkZGRaNmyJcaOHYsxY8Zg3bp1JvmKm3Pg4bRCs2fPRmZmpvx6eDVUIiIisi1mD1Bq164tz4dQqEmTJvj9998BFEyyBKBIb0l6enqRXpVCGo0G7u7uJi8ia2Dme9CJysQ2R7bC7AFKp06dcOnSJZO05ORk1KlTBwAQGBgIX19f7N+/X96em5uLuLg4dOzY0dzFIbIIR0dHAAXzexBVpsI2V9gGiayV2Z/ieeutt9CxY0dERkZi8ODB+Pnnn/Hhhx/iww8/BFAwtBMeHo7IyEgEBQUhKCgIkZGR0Gq1GDp0qLmLQ2QR9vb2qF69OtLT0wEAWq22xCFMtco3Coi8XAAFj65a86O3ZbGFcxVCQKfTIT09HdWrVy/XitFEamb2AOXpp5/GV199hdmzZ2PhwoUIDAxETEwMXnvtNTnPzJkzkZOTgwkTJiAjIwPt2rXDvn37yjUHCpG1KBzOLAxSrI1RCKT/dR8A4KBzhp2VBVgVYUvnWr16dbntEVkzRWbU6tevH/r161fidkmSEBERgYiICCUOT6QKkiShdu3a8Pb2hsFgsHRxKiwnNw9vfHUMAPDdpM5wseEJ+GzlXB0dHdlzQjbDOv8KiayIvb29VX5pGO3ycDM7HwCgcXaGs5V+aZdHVTpXsn263HwgN6/YbS6O9lYz3My/QiIiIhvSZdkh5KD4ZQ7a1PHEF+M6WEWQwtWMiYiIrJyLY/l6aU9dz0COIV/h0pgHe1CIiIis3IM9IgnvPFtkEVpdbj7aLD5Q2cV6LAxQiIiIbIgWuSj69Z4HF9xHDjSWKNIjYYBCRERkS1Y0KJKkBXDBGYg3NgREz8ov0yPgPShERA/Q5eZDl5tX5MUp5EnVHLWAf/sysz1tlwwYrGOGa/agEBE9oKRxemt6+oGqIEkCRu0pMfjQ3cuC9r3GlVyox8MeFCKq8lwc7dGmjmepeazp6QeqoiSp4ObY4l6OWkuXrsLYg0Jmpcst/gJuTZMDUdUjSRK+GNeh2ADEGp9+ILIFDFDIrNg9TtZKkiRoOYMskWpwiIceG7vHiYjI3PjvAj02do8TEZG5MUAhs2D3OBERmROHeIiIiEh1GKAQERGR6jBAISIiItVhgEJERESqw7sabY0Qpa+z4KgtmG2QiIhIxRig2BIhgI09gdSTJefxb1+wXgODFCIiUjEO8dgSg6704AQAUk9YzUqWRERUdbEHxVZNvwI4PbA4VK4OWNHAcuUhIiKqAAYotspJW7CCJRERkRXiEA8RERGpDgMUIiIiUh0GKERERKQ6DFCIiIhIdRigEBERkeowQCEiIiLVYYBCREREqsMAhYiIiFSHAQoRERGpjuIBSlRUFCRJQnh4uJwmhEBERAT8/Pzg4uKCsLAwnD9/XumiEBERkZVQNECJj4/Hhx9+iKeeesokfdmyZVi5ciXWrFmD+Ph4+Pr64rnnnkN2draSxSEiIiIroViAcvfuXbz22mv46KOP4OnpKacLIRATE4M5c+Zg0KBBCA4OxubNm6HT6fDZZ58pVRwiIiKyIooFKBMnTkTfvn3x7LPPmqSnpKQgLS0NPXr0kNM0Gg1CQ0Nx/PhxpYpDREREVkSR1Yy3bt2K06dPIz4+vsi2tLQ0AICPj49Juo+PD65fv17s/vR6PfR6vfw+KyvLjKUlIiIitTF7D0pqaiqmTJmCLVu2wNnZucR8kiSZvBdCFEkrFBUVBQ8PD/nl7+9v1jITERGRupg9QElISEB6ejpat24NBwcHODg4IC4uDu+//z4cHBzknpPCnpRC6enpRXpVCs2ePRuZmZnyKzU11dzFJiIiIhUx+xBP9+7dcfbsWZO0kSNHonHjxpg1axbq1asHX19f7N+/Hy1btgQA5ObmIi4uDtHR0cXuU6PRQKPRmLuoREREpFJmD1Dc3NwQHBxskubq6govLy85PTw8HJGRkQgKCkJQUBAiIyOh1WoxdOhQcxeHiIiIrJAiN8mWZebMmcjJycGECROQkZGBdu3aYd++fXBzc7NEcYiIiEhlKiVAOXz4sMl7SZIQERGBiIiIyjg8ERERWRmuxUNERESqwwCFiIiIVIcBChEREakOAxQiIiJSHQYoREREpDoMUIiIiEh1GKAQERGR6jBAISIiItVhgEJERESqwwCFiIiIVMcia/HQA3J1xac7agFJqtyyEBERqQQDFEtb0aD4dP/2wKg9DFKIiKhK4hCPJThqCwKQ0qSeAAwl9K4QERHZOPagWIIkFfSOFBeA5OpK7lUhIiKqIhigWIokAU6uli4FERGRKnGIh4iIiFSHAQoRERGpDgMUIiIiUh0GKERERKQ6DFCIiIhIdRigEBERkeowQCEiIiLVYYBCREREqsMAhYiIiFSHAQoRERGpDgMUIiIiUh0GKERERKQ6DFCIiIhIdRigEBERkeowQCEiIiLVYYBCREREqsMAhYiIiFSHAQoRERGpjtkDlKioKDz99NNwc3ODt7c3nn/+eVy6dMkkjxACERER8PPzg4uLC8LCwnD+/HlzF4WIiIislNkDlLi4OEycOBEnTpzA/v37kZeXhx49euDevXtynmXLlmHlypVYs2YN4uPj4evri+eeew7Z2dnmLg4RERFZIQdz73DPnj0m7zdt2gRvb28kJCSga9euEEIgJiYGc+bMwaBBgwAAmzdvho+PDz777DOMHTvW3EVSjBACOYb8Ium63HxolT0wYNAVTc8tJo2IyMZZ7FpMijJ7gPKwzMxMAECNGjUAACkpKUhLS0OPHj3kPBqNBqGhoTh+/HixAYper4der5ffZ2VlKVzqsgkh8NK/fkLC9Ywi21xwHxecFTswsLEnkHpSoQNQVVLShR0ouLgTqZ3FrsWkOEUDFCEEpk6dis6dOyM4OBgAkJaWBgDw8fExyevj44Pr168Xu5+oqCgsWLBAyaJWWI4hv9g/iIe5ONqb98AGXdnBiX97wJH/N1DpSruwE1kLi12LSXGKBihvvvkmfv31Vxw7dqzINkmSTN4LIYqkFZo9ezamTp0qv8/KyoK/v795C/sYTr3zLLRODzT+3HvAioIfSzons5h+BXAqJhBx1AJKHpdsQnkv7G3qePLiTion4AI9js7sZnotNuiA9wp+VPRaTIpQLECZNGkSvvnmGxw5cgRPPvmknO7r6wugoCeldu3acnp6enqRXpVCGo0GGo1GqaI+Nq2TPbROD1al4iNnBZy0gJNr5RyLbFqRIPsBLo72vLiTegmB/zgtQBu7ZOB9SxeGzMnsT/EIIfDmm29ix44d+OGHHxAYGGiyPTAwEL6+vti/f7+clpubi7i4OHTs2NHcxSGicigMsot7MTghVTPoCoKT0nDY2yqZ/V/9iRMn4rPPPsPXX38NNzc3+Z4TDw8PuLi4QJIkhIeHIzIyEkFBQQgKCkJkZCS0Wi2GDh1q7uIQEVEVoZtyEVpX96IbOOxtlcweoKxbtw4AEBYWZpK+adMmjBgxAgAwc+ZM5OTkYMKECcjIyEC7du2wb98+uLm5mbs4RERUVThy2NuWmD1AEUKUmUeSJERERCAiIsLchyciIiIbwLV4iIiISHUYoBAREZHqMEAhIiIi1WGAQkRERKrDAIWIiIhUhwEKERERqQ4DFCIiIlKdSlo0hojITIQoWASuJJw1lMgmMEChv/HCT+akRHsSAtjYE0g9WXIe//bAqD1sq0RWjgHKYylY4hu592BSlbmlXJTVihd+Miel2pNBV/o+ASD1REE+TnlOZNUYoDyqB5f4XmHpwpgBL/xkTpXRnqZfAZweWKE2VwesaPBo+yIi1WGA8qhseYlvXvjJnJRqT05cGI7IljFAMQOrW+K7uCGoB9N44SdzYnsiokfAAMUcrG2Jb/aGEBGRynEeFDXL1RXcgGvyesQbcB21BUNOZbHWYSkiIrIp7EFRM3P2dEhSwRMTpT32Cah3WIqIiKoUBihqU9jTkXqi9HyP0tMhSdY1FEVERFUWAxS1YU8HERERAxRVYk8HVRUl3VPFAJyoymOAQkSWU9J9Vpy1mKjK41M8RFS5yvNEWeEss0RUZbEHhYgqV2n3WXHWYiL6HwYoRFT5ynOfVVkzHhORTWOAQkTqxJ4UoiqN96AQkXpwxmMi+h/2oBDRoxGi5BtZH3UohvMAEdH/MEAh8yjpyyo3Dy64jxxoKr9MpBwhgI09gdST5t835wEiIjBAIXMo5ctKC+CCMxBvbAiInpVfNlKGQVe+4MRSQzGPs6gme2aIVIEBCj2+cnxZPW2XDJ1BB2g8KqlQVGmmXwGcSghCLPWF/6g32HKCOCLVYIBC5vXQl5XuXha07zW2YIFIcU5adQzJlHehzdIUThCnhvMhquIYoJB5PfxllZtnubJQ1VLeG2yLwwniiFSHAQoR2Q7eYEtkMyw6D8ratWsRGBgIZ2dntG7dGkePHrVkcYiIiEglLBagbNu2DeHh4ZgzZw7OnDmDLl26oHfv3vj9998tVSQqj1wdkHvvoRenH7dZQhTz++bvnIiUZ7EhnpUrV2L06NF4/fXXAQAxMTHYu3cv1q1bh6ioKEsVi8rCcXobI+ACfUHQ8fDlQAhgUy8g7axFSmYxDwdfnMuHyCIsEqDk5uYiISEBb7/9tkl6jx49cPz4cUsUCQAgjEbk6LLLlTfnXjaqzETb5X06oow5L3Lula9uqXLocvPwvdP/oZnddWDFI+7EFqecfygIL5zL57yxDnR3OwBOvHVPTarUtdhMKnItdtG6QbKzzGCLRf7S/vjjD+Tn58PHx8ck3cfHB2lpaUXy6/V66PV6+X1WVpYi5crRZUO7IqBceavUH4SZph/3WtfMzAWjx6EFULM81x3f5sDIEuYGsZWJzcoRhDezuw68X68SC0XlUaWuxWZSkWuxbvrv0FazzPxVFv1XQHrowiaEKJIGAFFRUViwYEFlFatCLjg2RWOtm6WLobxHfDrCReuGC45N0cSQpEChyBxSHOqh7owjkKQSohVbCUJKU0oQLoQR15Z3RWDebxYoGJVXlbkWPyJrvBZLQghR2QfNzc2FVqvFF198gRdeeEFOnzJlChITExEXF2eSv7geFH9/f2RmZsLd3d1s5arIEE8hS3Z/WYtHqVeqPGzDZWMbVj+247Kp4TsuKysLHh4e5fr+tkgPipOTE1q3bo39+/ebBCj79+/HwIEDi+TXaDTQaJS/QU2ys7NYV5YtY72StWMbJltgbe3YYkM8U6dOxbBhw9CmTRt06NABH374IX7//XeMGzfOUkUiIiIilbBYgPLKK6/gzp07WLhwIW7fvo3g4GDs2rULderUsVSRiIiISCUscg/K46rIGBYRERGpQ0W+v3lHEREREakOAxQiIiJSHQYoREREpDoMUIiIiEh1GKAQERGR6jBAISIiItVhgEJERESqY5XrhhdO3aLUqsZERERkfoXf2+WZgs0qA5Ts7ILFjvz9/S1cEiIiIqqo7OxseHiUvi6QVc4kazQacevWLbi5uUGqxGXgC1dRTk1N5Qy2CmEdK491rCzWr/JYx8pTqo6FEMjOzoafnx/sylgl2Sp7UOzs7PDkk09a7Pju7u78o1AY61h5rGNlsX6VxzpWnhJ1XFbPSSHeJEtERESqwwCFiIiIVIcBSgVoNBrMnz8fGo3G0kWxWaxj5bGOlcX6VR7rWHlqqGOrvEmWiIiIbBt7UIiIiEh1GKAQERGR6jBAISIiItVhgEJERESqwwDlAbxfmKwd2zBZO7ZhKsQA5X/y8/PlNX7I/AwGA27evCm/50XI/NiGlcd2rCy2YeVZUxtmgAJg1apV6NSpE55//nmEh4fj6tWrAArW/KHH9+677yIoKAh9+/ZFv3798NNPP1XqGkpVAduw8tiOlcU2rDxra8NVeh6Uy5cvY9y4cUhNTcWcOXNw8eJFHD58GG5ubti3b5+li2cTVq5cidWrV2PFihXIysrC119/jR9//BFffPEFwsLCLF08q8c2XDnYjpXDNlw5rLINiyoqPz9fvPvuu6JPnz7i5s2bcvoXX3whWrRoIS5cuGDB0lm//Px8YTAYRO/evcX48eNNtnXu3Fn06tVLnDlzxjKFsxFsw8pjO1YW27DyrLkNV9khHiEEgoKCMHHiRPj5+cndiNWqVUNaWhq8vLwsXELrZmdnB6PRiLNnz6JVq1YAgPv37wMAYmJikJycjD179kCv11uymFaNbVh5bMfKYhtWnjW34SoToOzZs0e+GUgIAXt7e/Tt2xd9+vQBAHkcLisrC35+fnBxcbFYWa3RJ598grfeeguffPIJ0tPTAQBOTk5o3749tmzZAgBwdnaG0WhE69at0aNHD3z55Zf4448/LFlsq8I2rDy2Y2WxDSvPptqwxfpuKklSUpIIDQ0VkiSJRYsWCSEKurweZjQahRBCTJkyRQwbNqzEfGQqLS1NdO/eXTzxxBPihRdeEAEBAaJRo0bixIkTQgghtmzZIp544gmxa9cuIYQQOTk5QgghUlNThSRJcj4qGduw8tiOlcU2rDxbbMM23YNy7do1REdHo1atWpg8eTKWLVuG9PR02NnZFXm0qjBy37t3r3zDkJ2dHa5duwZA3Y9iWdLRo0dx+/ZtnD59Gjt27EBycjLc3d2xaNEinDt3Ds899xyefvppREZGAiiI3IUQ0Gg08Pf3R1JSkoXPQN3YhisH27Fy2IYrhy22YZsOULy9vRESEoIZM2bg7bffRmBgIN56660S8//666/466+/0KtXL/z5558YPXo06tWrh+TkZFU/imUpQggcPXoUtWrVgpubG4xGIzQaDWJiYpCeno6PP/4YtWrVwpgxY3Dt2jVMnz4dBoMBkiTh7Nmz0Gg06r17XCXYhpXHdqwstmHl2WwbtkzHjfIKuwrv378vp23fvt2kK+vhrsOtW7eKtm3biqioKOHu7i66du0qkpKSKq/QVqSwfmfNmiWaNGkihBAiLy9P3v5///d/on379uLo0aNCiIK61Wq1olWrVmLEiBHC3d1dvPHGGyInJ0feF5liG1Ye27Gy2IaVZ8tt2GYDlAcVVnpWVpbo16+faNu2bbH5hg8fLiRJEkFBQeLLL7+szCKqVkkNtjA9ISFBODo6iv379wsh/r4QpaSkiMDAQPHBBx/Inzl48KBYsWKFGDlypPjuu+8ULrn1KM9FgW348bAdK4tt2HJsuQ1bdYBS+It5MFosy/Hjx4VGoxGffPKJ/NmsrCwhhBAbN24UH374ofkLaqVyc3NFbm5uqXn++OMPMWjQING8eXM5rfD30bdvX/HKK68oWkZrV546fhjbcMXcv39f/L//9//k98V9mbIdP7ry1O/D2IYrJisrSxw8eFAIUfJNw7bYhq02QFm1apUYPnx4hT+Xm5srwsPDhb+/v0hKShJDhw4Vc+bMEXq93vyFtGJRUVEiLCxM9O3bV8TExMgXoOL+OA4ePCg8PT3F4sWL5TS9Xi+6d+8uZsyYUWlltjYVqeMHsQ2X3+LFi0Xz5s1Fx44dxZAhQ8TVq1eFEMX/U8N2XHEVqd8HsQ2X33//+1+h0WiEJEnixo0bQoiSg0Bba8NWF6BcuHBBDBw4ULi6ugpvb2/x+eefCyEq1oty6tQpIUmSkCRJNG7cmOObDzhz5oxo27ataNq0qfj444/FkCFDRKtWrcSkSZNK/Ixerxdr164VkiSJadOmiUOHDonVq1cLPz8/Oeqnvz1KHT+Mbbh0x48fF61atRJNmjQR27dvF8uXLxcdO3YUL7/8comfYTsuv0ep34exDZfNaDSKS5cuiW7duomgoCDx4osvyunFsbU2bHUBSmxsrOjfv7/Ytm2bGDlypOjQoYPQ6XRCiLK7Fo1Gozh48KDw9fUVAQEBqh57s4R79+6JWbNmiSFDhoiMjAw5ff78+aJPnz7izp07pX5++fLlonPnzqJJkyYiICBAfPHFFwqX2Po8bh2zDZfPokWLxBtvvCEyMzPltAkTJoiRI0fK70u6XrAdl+1x6pdtuGL27t0rOnXqJA4dOiTs7OzEoUOHyvyMrbRhqwlQCru97969K+Li4oQQQnzzzTeiZcuWcndWWV3jBoNBREVFiYiICGULa6UyMzPF2rVrxZEjR4QQBfUlhBDvvfeeaNCggfz+YQ/Wu9Fo5H9CpXjUOi7ENly6wi/FtLQ0cfnyZTk9Ly9PPPvss2Lx4sXi4sWLxX6W7bhsj1O/hdiGy/ZgcPf555+LCRMmCKPRKPr16ydCQkKEEAV1/vDIga21YVWvZhwbG4vU1FS0bdsWnTt3hqurq8n2jIwMREVFYdeuXfj6669Rv3595Ofnw97evsi+hBCQJAlGoxF2djY9/Uu5PVi/nTp1QrVq1ZCXlwcHBwcAkOtq0aJFSExMxH/+859S5yEorGP6mznrmG24eGVdJ7Zu3Ypx48ahdu3aqFatGu7cuYMBAwYgJiam2P2xHZsyZ/2yDRevuDourKsPPvgAcXFx2L59O65fv45mzZqhR48e+OuvvxAdHY2nn366yP5spg1bLDQqxdWrV0Xr1q1FYGCgCAsLE76+vqJHjx7ijz/+kPMURopHjx4VXbt2Fa+//rqlimt1iqvf5557zqR+jUajHMUPHDhQ7qVS23PyasU6Vl55rhNCCLF+/XrxxRdfCL1eL/7880/x3XffCUmSxLlz54QQrO+SsH6VV546Hj58uPj444+FEAWjBjVq1BCSJIl169ZZqtiVRpUBSkxMjGjXrp3Q6XRCp9OJCxcuCC8vL/H666+L69evCyH+vinWYDCI5cuXi0aNGsljcydOnOAfRSlKq9/ff/9dCPF3AJiVlSVq1aplcoNVamqqEIIXntKwjpVXnuuEEEXr8LfffhNeXl5iw4YNlV1kq8L6VV5pdVz4RNSMGTPEhAkTxIABA4Szs7N44403hL+/v5g5c6YQwrbXKlJdH5vBYMC3336L5s2bw8XFBU5OTmjcuDH+9a9/4cCBA9i1axcAwN7eHkajEQ4ODujTpw+aN2+Od955B/3790eHDh1w/vx5C5+JOpVVv99//z0AyN2vhw8fRrVq1dCtWzfcvn0bQ4cORUhICNLS0myjC1EBrGPllfc6AaBIHR49ehQNGjRA3759K7vYVoP1q7yy6njfvn0AgN9++w3r1q2Do6Mjzpw5g/Xr12Pu3LlYvnw5rly5YtNDZao6MyEEHB0d4eHhgVu3bslpAPDSSy+hVatW2LFjB65evQrg7wt8rVq1kJ6ejuPHj8PBwQG//fYbgoODLXMSKlbR+gWAixcvokWLFli6dCkaNGiA9PR0JCQkwNfX1yLnoHasY+U9Sh3/8ccfuHXrFhYuXIjZs2fjhRdeQM2aNbn4XDFYv8orTx1v374dmZmZWLJkCQ4fPoxt27ahcePGAIBBgwZh3rx58PLystg5VApLdNuU5dNPPxVeXl7yGGbhlL0JCQlCo9GIY8eOyXnPnj0rGjVqJOrVqyc/3UOlq0j9durUSZ6nYPfu3RYprzViHSuvvHWclJQkFi9eLBo2bCiaNm0qDhw4YLEyWxPWr/JKq2MnJyeT60RVpKoeFIPBAABo2rQpnnrqKSxatAgAoNFoYDQa0apVKwQFBcldXwAQFBSEpUuX4urVq+jatatFym0tKlq/9+7dQ7du3RAbG4sLFy6gV69eFiu7tWAdK6+8dbx3714AQN26ddGhQwcsX74c58+fR/fu3S1WdmvA+lVeeeq4YcOGJt91VVGlBig7duzAX3/9VSQ9Pz8fAODo6AgAaNWqFV599VUcOXIE27ZtA1AwnJOWlgaDwQB/f3/5cxqNBs8//3yllF/tzF2/rq6uWLRoEf75z39WzglYAdax8sxVxwEBAQAAFxcXPPPMMxgwYEDlnIDKsX6VZ+7rRJVVGd00cXFxomnTpkKSJJMVFR+2bds24ejoKL788kvx559/irfeektotVqxdu1a8csvv4jo6GjRoEED8euvv1ZGsa0G61d5rGPlsY6VxfpVHuvYvBQPUC5duiSGDh0qxo8fLyZMmCD8/PzkBY+EKHhELTMzU7z66quiVq1aYsWKFSYLRo0bN040a9ZM1K9fXwQEBIjvv/9e6SJbFdav8ljHymMdK4v1qzzWsfkpHqCkp6eLjz/+WJw7d07cu3dPBAQEiClTppjk0ev1YuvWrSItLU1OK3y23mg0irt374pTp04pXVSrxPpVHutYeaxjZbF+lcc6Nj+zByhHjx41mcRHCGGyvsgnn3winJycREJCghCi7ImobHkSmkfB+lUe61h5rGNlsX6VxzpWntkClIMHD4rAwEBRp04dUbt2bTFs2DBx+vRpIUTRX0ynTp1Enz59RG5urrkOb/NYv8pjHSuPdaws1q/yWMeVxywBSmpqqujQoYOYM2eOuH79uvj2229FSEiI6N69u7zi5YOrLh4/flzY2dmJHTt2CCEKIseH13egv7F+lcc6Vh7rWFmsX+WxjiuXWQKUffv2CWdnZ5GcnCyn7d27V3Tr1k288sorxX5mxIgR4qmnnhIHDhwQvXr1ErNnz5YnqSFTrF/lsY6VxzpWFutXeazjymWWAGXr1q0iJCREXLx4UU7Lz88XH3zwgahbt67Yu3evEMI0sjx58qSQJElIkiSee+45cefOHXMUxSaxfpXHOlYe61hZrF/lsY4rl1kClLNnzwqNRiN27txpkn7x4kXx/PPPixEjRshpeXl54t///rfQaDSiTZs2Ij4+3hxFsGmsX+WxjpXHOlYW61d5rOPKZZaZZIODg/HMM89g1apVuHv3rpzeqFEjBAQEIC0tDTk5OQCA3Nxc3LlzBzExMYiPj0ebNm3MUQSbxvpVHutYeaxjZbF+lcc6rmTminQSExOFg4ODWLduncn42qJFi0RAQIDJ41dUcaxf5bGOlcc6VhbrV3ms48rjYK5Ap0WLFpg1axYWLlwIe3t7DBkyBEajET///DNee+01ODiY7VBVEutXeaxj5bGOlcX6VR7ruPJIQghhzh1OnDgRX375JQICApCeng6tVovt27cjODjYnIepsli/ymMdK491rCzWr/JYx8oze4Ci1+uRlJSExMREODk54bXXXjPn7qs81q/yWMfKYx0ri/WrPNax8sweoBARERE9LrM8xUNERERkTgxQiIiISHUYoBAREZHqMEAhIiIi1WGAQkRERKrDAIWIiIhUhwEKERERqQ4DFCIiIlIdBihEFhIREYGQkBBLF0M17ty5A29vb1y7ds3SRSlRbGwsqlevXmoeS/xe16xZgwEDBlTqMYmUxgCFSAGSJJX6GjFiBKZPn46DBw9auqgAgLp16yImJqZIekxMDOrWrSvnKe2cwsLCAABXrlzByJEj8eSTT0Kj0SAwMBBDhgzBqVOnSi1DVFQU+vfvLx/PVl27dg2SJCExMbHItrCwMISHh5f4vtDDgdKYMWMQHx+PY8eOmb/ARBbCZReJFHD79m35523btmHevHm4dOmSnObi4oJq1aqhWrVqlijeI4mPj0d+fj4A4Pjx43jxxRdx6dIluLu7AwCcnJxw6tQpdO/eHcHBwVi/fj0aN26M7OxsfP3115g2bRri4uKK3XdOTg42bNiAXbt2Vdr52BKNRoOhQ4di9erV6Ny5s6WLQ2QW7EEhUoCvr6/88vDwgCRJRdIeHgoYMWIEnn/+eURGRsLHxwfVq1fHggULkJeXhxkzZqBGjRp48sknsXHjRpNj3bx5E6+88go8PT3h5eWFgQMHKjJMUqtWLbn8NWrUAAB4e3vLaZ6enhgxYgSCgoJw9OhR9O3bF/Xr10dISAjmz5+Pr7/+usR97969Gw4ODujQoYOcdvjwYUiShL1796Jly5ZwcXHBM888g/T0dOzevRtNmjSBu7s7hgwZAp1OJ38uLCwMkyZNQnh4ODw9PeHj44MPP/wQ9+7dw8iRI+Hm5ob69etj9+7dZZ5zbGwsAgICoNVq8cILL+DOnTtF8ixduhQ+Pj5wc3PD6NGjcf/+/YpUq9kMGDAAO3fuRE5OjkWOT2RuDFCIVOSHH37ArVu3cOTIEaxcuRIRERHo168fPD09cfLkSYwbNw7jxo1DamoqAECn06Fbt26oVq0ajhw5gmPHjqFatWro1asXcnNzK7XsiYmJOH/+PKZNmwY7u6KXltLu3Thy5AjatGlT7LaIiAisWbMGx48fR2pqKgYPHoyYmBh89tln+P7777F//36sXr3a5DObN29GzZo18fPPP2PSpEkYP348Xn75ZXTs2BGnT59Gz549MWzYMJPA5mEnT57EqFGjMGHCBCQmJqJbt25YvHixSZ7t27dj/vz5WLJkCU6dOoXatWtj7dq1pdSSctq0aQODwYCff/7ZIscnMjtBRIratGmT8PDwKJI+f/580aJFC/n98OHDRZ06dUR+fr6c1qhRI9GlSxf5fV5ennB1dRWff/65EEKIDRs2iEaNGgmj0Sjn0ev1wsXFRezdu7fcZaxTp45YtWpVkfRVq1aJOnXqFEk/dOiQACAyMjLktG3btgkA4vTp0+U+bqGBAweKUaNGFXuMAwcOyGlRUVECgLh69aqcNnbsWNGzZ0/5fWhoqOjcubP8vrDOhg0bJqfdvn1bABA//fRTiWUaMmSI6NWrl0naK6+8YvK77NChgxg3bpxJnnbt2pn8Xh+WkpIiAAgXFxfh6upq8rKzsxNTpkwxORdHR8ci+TQaTbFtytPTU8TGxpZ4bCJrwh4UIhVp1qyZSe+Dj48PmjdvLr+3t7eHl5cX0tPTAQAJCQm4cuUK3Nzc5HtaatSogfv37+Pq1auVWnYhBICCG4QrKicnB87OzsVue+qpp+SffXx8oNVqUa9ePZO0wvoo7jOFdfZgPfr4+ACA/LlmzZrJ9de7d28AwIULF0yGnAAUeV+ePCXZtm0bEhMTTV7F9SK99tprRfItXLiw2H26uLiU2itEZE14kyyRijg6Opq8lySp2DSj0QgAMBqNaN26Nf79738X2VetWrXKfVx3d3dkZmYWSf/rr7/g4eFRrn00bNgQQMGXdkUfs61ZsyYyMjKK3fbg+ZdVH8V9prjPFQZRhZ/btWsXDAYDgIIveeDvgEsp/v7+aNCggUla4bEf5OHhUSSft7d3sfv8888/K/R7J1Iz9qAQWbFWrVrh8uXL8Pb2RoMGDUxe5Q0sAKBx48aIj48vkh4fH49GjRqVax8hISFo2rQp3n333SIBA1AQ7JSkZcuWSEpKKnd5za1OnTpyvT3xxBMAgKZNm+LEiRMm+R5+36RJkzLzVJarV6/i/v37aNmypUWOT2RuDFCIrNhrr72GmjVrYuDAgTh69ChSUlIQFxeHKVOm4MaNG+Xez9SpU7F7924sXLgQSUlJSEpKwqJFi7Bnzx5MmzatXPuQJAmbNm1CcnIyunbtil27duG3337Dr7/+iiVLlmDgwIElfrZnz544f/58ib0oljB58mTs2bMHy5YtQ3JyMtasWYM9e/aY5JkyZQo2btyIjRs3Ijk5GfPnz8f58+ctUt6jR4+iXr16qF+/vkWOT2RuDFCIrJhWq8WRI0cQEBCAQYMGoUmTJhg1ahRycnLk+UkKH9ct7dHj9u3bY+/evThw4AA6d+6Mzp07Y9++fdi7dy/atWtX7vK0bdsWp06dQv369TFmzBg0adIEAwYMwPnz54udCK5Q8+bN0aZNG2zfvr3cx1Ja+/bt8fHHH2P16tUICQnBvn378M4775jkeeWVVzBv3jzMmjULrVu3xvXr1zF+/HiLlPfzzz/HmDFjLHJsIiVIQumBViKyqNjYWCxZsgRJSUlF7s1Qk127dmH69Ok4d+5csY8pU8nOnTuH7t27Izk5uUJDe0RqxptkiWzcnj17EBkZqergBAD69OmDy5cv4+bNm/D397d0cazKrVu38MknnzA4IZvCHhQiIiJSHfajEhERkeowQCEiIiLVYYBCREREqsMAhYiIiFSHAQoRERGpDgMUIiIiUh0GKERERKQ6DFCIiIhIdRigEBERkeowQCEiIiLV+f8L6XDoqL1MRQAAAABJRU5ErkJggg==",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "x = 3\n",
    "\n",
    "# convert row of dataframe to a dictionary to be passed to power function\n",
    "plant_data = plants_df.iloc[x].dropna().to_dict()\n",
    "\n",
    "# get resource data\n",
    "resource_data = nsrdb_data_dict[x]\n",
    "\n",
    "# calculate power and plot\n",
    "power_ac, power_ac_clear_sky = pvlib_power_model(resource_data, plant_data, calc_clear_sky=True)\n",
    "plt.plot(power_ac_clear_sky[6:68], drawstyle='steps-mid',)\n",
    "plt.plot(power_ac[6:68], drawstyle='steps-mid',)\n",
    "plt.legend(['clear sky power', 'NSRDB power'])\n",
    "plt.gcf().autofmt_xdate()\n",
    "plt.xlabel('Time, UTC (mm-dd HH)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we can run that in a loop for all the plants, writing the output to a dictionary:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "power_ac_nsrdb = {}\n",
    "power_ac_cs = {}\n",
    "\n",
    "for x in range(0,num_plants):\n",
    "    plant_data = plants_df.iloc[x].dropna().to_dict()\n",
    "    resource_data = nsrdb_data_dict[x]\n",
    "    power_ac_nsrdb[x], power_ac_cs[x] = pvlib_power_model(\n",
    "        resource_data,\n",
    "        plant_data,\n",
    "        calc_clear_sky=True,\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And sum the NSRDB-modeled power for each plant into a single series:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max power: [3694.99506046]\n",
      "mean power: [1013.38275292]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>power_ac</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>time_center_labeled</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2021-01-01 05:00:00+00:00</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 06:00:00+00:00</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 07:00:00+00:00</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 08:00:00+00:00</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 09:00:00+00:00</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 10:00:00+00:00</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 11:00:00+00:00</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 12:00:00+00:00</th>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 13:00:00+00:00</th>\n",
       "      <td>187.827911</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 14:00:00+00:00</th>\n",
       "      <td>854.508945</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 15:00:00+00:00</th>\n",
       "      <td>445.290180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2021-01-01 16:00:00+00:00</th>\n",
       "      <td>457.895196</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             power_ac\n",
       "time_center_labeled                  \n",
       "2021-01-01 05:00:00+00:00    0.000000\n",
       "2021-01-01 06:00:00+00:00    0.000000\n",
       "2021-01-01 07:00:00+00:00    0.000000\n",
       "2021-01-01 08:00:00+00:00    0.000000\n",
       "2021-01-01 09:00:00+00:00    0.000000\n",
       "2021-01-01 10:00:00+00:00    0.000000\n",
       "2021-01-01 11:00:00+00:00    0.000000\n",
       "2021-01-01 12:00:00+00:00    0.000000\n",
       "2021-01-01 13:00:00+00:00  187.827911\n",
       "2021-01-01 14:00:00+00:00  854.508945\n",
       "2021-01-01 15:00:00+00:00  445.290180\n",
       "2021-01-01 16:00:00+00:00  457.895196"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "power_ac_nsrdb_all = pd.concat(power_ac_nsrdb, axis=1, sort=False).sum(axis=1)\n",
    "power_ac_nsrdb_all = power_ac_nsrdb_all.to_frame(name='power_ac')\n",
    "power_ac_nsrdb_all.index.names = ['time_center_labeled']\n",
    "print('max power: ' + str(power_ac_nsrdb_all.max().values))\n",
    "print('mean power: ' + str(power_ac_nsrdb_all.mean().values))\n",
    "power_ac_nsrdb_all.head(12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And sum the clear sky power:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max power: [3703.5051469]\n",
      "mean power: [1489.92586171]\n"
     ]
    }
   ],
   "source": [
    "power_ac_cs_all = pd.concat(power_ac_cs, axis=1, sort=False).sum(axis=1)\n",
    "power_ac_cs_all = power_ac_cs_all.to_frame(name='power_ac')\n",
    "power_ac_cs_all.index.names = ['time_center_labeled']\n",
    "print('max power: ' + str(power_ac_cs_all.max().values))\n",
    "print('mean power: ' + str(power_ac_cs_all.mean().values))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Forecasted power\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`herbie.accessors` includes a function, `nearest_points`, to get the nearest points from a Dataset [1]. You can pass it a tuple of a latitude, longitude pair, a list of pairs, or a DataFrame with `latitude` and `longitude` columns. It can also take in a point name, which we will use to assign EIA plant codes.\n",
    "\n",
    "Then, let's add each plant's ac capacity (using the plant code as a reference). We will also drop unneeded columns and rename time columns to include \"utc\".\n",
    "\n",
    "[1] https://herbie.readthedocs.io/en/latest/api_reference/_autosummary/herbie.accessors.HerbieAccessor.html#herbie.accessors.HerbieAccessor.nearest_points"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>time_utc</th>\n",
       "      <th>step</th>\n",
       "      <th>plant_code</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>valid_time_utc</th>\n",
       "      <th>dswrf</th>\n",
       "      <th>t2m</th>\n",
       "      <th>u10</th>\n",
       "      <th>v10</th>\n",
       "      <th>t2m_c</th>\n",
       "      <th>wspd</th>\n",
       "      <th>y</th>\n",
       "      <th>x</th>\n",
       "      <th>point_latitude</th>\n",
       "      <th>point_longitude</th>\n",
       "      <th>ac_capacity</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2021-04-08 06:00:00</td>\n",
       "      <td>1 days 03:00:00</td>\n",
       "      <td>63721.0</td>\n",
       "      <td>30.759312</td>\n",
       "      <td>275.325856</td>\n",
       "      <td>2021-04-09 09:00:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>293.006805</td>\n",
       "      <td>-0.522706</td>\n",
       "      <td>2.86133</td>\n",
       "      <td>19.856812</td>\n",
       "      <td>2.908682</td>\n",
       "      <td>-777306.152557</td>\n",
       "      <td>1.232480e+06</td>\n",
       "      <td>30.747916</td>\n",
       "      <td>-84.680492</td>\n",
       "      <td>213.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>2021-04-08 06:00:00</td>\n",
       "      <td>1 days 03:00:00</td>\n",
       "      <td>59876.0</td>\n",
       "      <td>31.553886</td>\n",
       "      <td>275.901393</td>\n",
       "      <td>2021-04-09 09:00:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>292.506805</td>\n",
       "      <td>0.102294</td>\n",
       "      <td>2.36133</td>\n",
       "      <td>19.356812</td>\n",
       "      <td>2.363545</td>\n",
       "      <td>-681306.152557</td>\n",
       "      <td>1.274480e+06</td>\n",
       "      <td>31.560729</td>\n",
       "      <td>-84.093261</td>\n",
       "      <td>31.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>2021-04-08 06:00:00</td>\n",
       "      <td>1 days 03:00:00</td>\n",
       "      <td>60306.0</td>\n",
       "      <td>31.390420</td>\n",
       "      <td>270.503694</td>\n",
       "      <td>2021-04-09 09:00:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>291.319305</td>\n",
       "      <td>0.664794</td>\n",
       "      <td>2.23633</td>\n",
       "      <td>18.169312</td>\n",
       "      <td>2.333050</td>\n",
       "      <td>-759306.152557</td>\n",
       "      <td>7.644799e+05</td>\n",
       "      <td>31.403531</td>\n",
       "      <td>-89.497640</td>\n",
       "      <td>52.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>2021-04-08 06:00:00</td>\n",
       "      <td>1 days 03:00:00</td>\n",
       "      <td>59894.0</td>\n",
       "      <td>32.574280</td>\n",
       "      <td>275.757254</td>\n",
       "      <td>2021-04-09 09:00:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>290.569305</td>\n",
       "      <td>0.727294</td>\n",
       "      <td>1.23633</td>\n",
       "      <td>17.419312</td>\n",
       "      <td>1.434388</td>\n",
       "      <td>-570306.152557</td>\n",
       "      <td>1.244480e+06</td>\n",
       "      <td>32.572875</td>\n",
       "      <td>-84.256778</td>\n",
       "      <td>30.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>2021-04-08 06:00:00</td>\n",
       "      <td>1 days 03:00:00</td>\n",
       "      <td>58628.0</td>\n",
       "      <td>33.673616</td>\n",
       "      <td>276.308339</td>\n",
       "      <td>2021-04-09 09:00:00</td>\n",
       "      <td>0.0</td>\n",
       "      <td>288.881805</td>\n",
       "      <td>2.164794</td>\n",
       "      <td>0.79883</td>\n",
       "      <td>15.731812</td>\n",
       "      <td>2.307480</td>\n",
       "      <td>-441306.152557</td>\n",
       "      <td>1.277480e+06</td>\n",
       "      <td>33.678056</td>\n",
       "      <td>-83.675833</td>\n",
       "      <td>30.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              time_utc            step  plant_code   latitude   longitude  \\\n",
       "0  2021-04-08 06:00:00 1 days 03:00:00     63721.0  30.759312  275.325856   \n",
       "28 2021-04-08 06:00:00 1 days 03:00:00     59876.0  31.553886  275.901393   \n",
       "29 2021-04-08 06:00:00 1 days 03:00:00     60306.0  31.390420  270.503694   \n",
       "30 2021-04-08 06:00:00 1 days 03:00:00     59894.0  32.574280  275.757254   \n",
       "31 2021-04-08 06:00:00 1 days 03:00:00     58628.0  33.673616  276.308339   \n",
       "\n",
       "        valid_time_utc  dswrf         t2m       u10      v10      t2m_c  \\\n",
       "0  2021-04-09 09:00:00    0.0  293.006805 -0.522706  2.86133  19.856812   \n",
       "28 2021-04-09 09:00:00    0.0  292.506805  0.102294  2.36133  19.356812   \n",
       "29 2021-04-09 09:00:00    0.0  291.319305  0.664794  2.23633  18.169312   \n",
       "30 2021-04-09 09:00:00    0.0  290.569305  0.727294  1.23633  17.419312   \n",
       "31 2021-04-09 09:00:00    0.0  288.881805  2.164794  0.79883  15.731812   \n",
       "\n",
       "        wspd              y             x  point_latitude  point_longitude  \\\n",
       "0   2.908682 -777306.152557  1.232480e+06       30.747916       -84.680492   \n",
       "28  2.363545 -681306.152557  1.274480e+06       31.560729       -84.093261   \n",
       "29  2.333050 -759306.152557  7.644799e+05       31.403531       -89.497640   \n",
       "30  1.434388 -570306.152557  1.244480e+06       32.572875       -84.256778   \n",
       "31  2.307480 -441306.152557  1.277480e+06       33.678056       -83.675833   \n",
       "\n",
       "    ac_capacity  \n",
       "0         213.0  \n",
       "28         31.0  \n",
       "29         52.0  \n",
       "30         30.5  \n",
       "31         30.0  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_hrrr_plants = ds.herbie.nearest_points(points=plants_df,\n",
    "                                           names=plants_df['plant_code']) \\\n",
    "                                            .to_dataframe().reset_index()\n",
    "df_hrrr_plants = df_hrrr_plants.merge(plants_df[['plant_code','ac_capacity']],\n",
    "                                       left_on='point', right_on='plant_code')\n",
    "df_hrrr_plants.drop(columns={'plant_code', 'surface', 'heightAboveGround', \n",
    "                             'metpy_crs', 'gribfile_projection'}, inplace=True)\n",
    "df_hrrr_plants.rename(columns={'time':'time_utc','valid_time':'valid_time_utc',\n",
    "                               'point':'plant_code'},\n",
    "                      inplace=True)\n",
    "\n",
    "# sort as times may end up out of order after saving and loading the .netCDF file?\n",
    "df_hrrr_plants.sort_values(by=['valid_time_utc'], inplace=True)\n",
    "\n",
    "df_hrrr_plants.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's create a function to create a _resource_data_ dataframe from the HRRR data. It needs to add DNI and DHI that are calculated with a decomposition model. This is setup for options of Erbs and DIRINT models, but other models like Erbs-Driesse might also work well. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def prep_hrrr_data(plant_data, forecast_df, decomposition_model = 'ers', albedo = 0.2):\n",
    "    # Resource data\n",
    "    plant_code = plant_data['plant_code']\n",
    "    # arrange weather data for plant\n",
    "    resource_data = forecast_df.loc[forecast_df['plant_code'] == plant_code,\n",
    "                                ['valid_time_utc','dswrf', 't2m_c','wspd']].copy()\n",
    "    resource_data = resource_data.set_index('valid_time_utc')\n",
    "    resource_data.index = resource_data.index.tz_localize(tz='UTC')\n",
    "    resource_data.rename(columns={'dswrf':'ghi', 't2m_c':'temp_air', 'wspd':'wind_speed'},\n",
    "                    inplace=True)\n",
    "    \n",
    "    # solar position\n",
    "    lat = plant_data['latitude'] \n",
    "    lon = plant_data['longitude']\n",
    "    times = resource_data.index \n",
    "    loc= pvlib.location.Location(latitude=lat, longitude=lon, tz=times.tz)\n",
    "    solar_position = loc.get_solarposition(times)\n",
    "\n",
    "    # decomposition\n",
    "    if decomposition_model=='erbs':\n",
    "        out_erbs = pvlib.irradiance.erbs(\n",
    "            resource_data.ghi,\n",
    "            solar_position.zenith,\n",
    "            resource_data.index,\n",
    "        )\n",
    "        resource_data['dni'] = out_erbs.dni\n",
    "        resource_data['dhi'] = out_erbs.dhi\n",
    "    elif decomposition_model=='dirint':\n",
    "        dni_dirint = pvlib.irradiance.dirint(\n",
    "            resource_data.ghi,\n",
    "            solar_position.zenith,\n",
    "            resource_data.index,\n",
    "        )\n",
    "        df_dirint = pvlib.irradiance.complete_irradiance(\n",
    "            solar_zenith=solar_position.apparent_zenith,\n",
    "            ghi=resource_data.ghi,\n",
    "            dni=dni_dirint, dhi=None,\n",
    "        )\n",
    "        resource_data['dni'] = dni_dirint\n",
    "        resource_data['dhi'] = df_dirint.dhi\n",
    "    \n",
    "    resource_data['albedo'] = albedo\n",
    "\n",
    "    return resource_data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then loop through each plant:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "decomposition_model = 'erbs' # dirint or erbs\n",
    "albedo = 0.2 # generic albedo value to use for the forecasts - something better could be added later\n",
    "\n",
    "power_ac_hrrr = {}\n",
    "\n",
    "for x in range(0,num_plants):\n",
    "    plant_data = plants_df.iloc[x].dropna().to_dict()\n",
    "\n",
    "    # resource data\n",
    "    resource_data = prep_hrrr_data(\n",
    "        plant_data,\n",
    "        forecast_df=df_hrrr_plants,\n",
    "        decomposition_model=decomposition_model,\n",
    "        albedo=albedo,\n",
    "    )\n",
    "\n",
    "    # calculate power\n",
    "    power_ac_hrrr[x] = pvlib_power_model(\n",
    "        resource_data,\n",
    "        plant_data,\n",
    "        calc_clear_sky=False,\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And sum up the forecasted power for all sites:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "max power: [3680.3837371]\n",
      "mean power: [1317.91613125]\n"
     ]
    }
   ],
   "source": [
    "power_ac_hrrr_all = pd.concat(power_ac_hrrr, axis=1, sort=False).sum(axis=1)\n",
    "power_ac_hrrr_all = power_ac_hrrr_all.to_frame(name='power_ac')\n",
    "power_ac_hrrr_all.index.names = ['time_center_labeled']\n",
    "print('max power: ' + str(power_ac_hrrr_all.max().values))\n",
    "print('mean power: ' + str(power_ac_hrrr_all.mean().values))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's plot \"observed\", forecasted, and clear sky power:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mask = ((power_ac_nsrdb_all.index >= power_ac_hrrr_all.index.min()) & \n",
    "        (power_ac_nsrdb_all.index <= power_ac_hrrr_all.index.max()))\n",
    "\n",
    "plt.plot(power_ac_hrrr_all.index,power_ac_hrrr_all.power_ac,\n",
    "         drawstyle='steps-mid', label='HRRR power')\n",
    "plt.plot(power_ac_nsrdb_all.loc[mask].index, \n",
    "         power_ac_nsrdb_all.loc[mask].power_ac,\n",
    "         drawstyle='steps-mid',\n",
    "         label='NSRDB power')\n",
    "plt.plot(power_ac_cs_all.loc[mask].index, \n",
    "         power_ac_cs_all.loc[mask].power_ac,\n",
    "         drawstyle='steps-mid',\n",
    "         label='clear sky power')\n",
    "plt.legend()\n",
    "plt.gcf().autofmt_xdate()\n",
    "plt.xlabel('Time, UTC (mm-dd HH)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Coarsened mean, max, and min forecasts\n",
    "Then we can do the same thing, but with the coarsened mean, max, and min HRRR datasets to look at a more interesting forecast.\n",
    "\n",
    "First, prep the HRRR datasets like we did previously:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# coarse mean\n",
    "df_hrrr_plants_coarse = ds_coarse.herbie.nearest_points(points=plants_df,\n",
    "                                           names=plants_df['plant_code']) \\\n",
    "                                            .to_dataframe().reset_index()\n",
    "df_hrrr_plants_coarse = df_hrrr_plants_coarse.merge(\n",
    "    plants_df[['plant_code','ac_capacity']],\n",
    "    left_on='point', right_on='plant_code'\n",
    "    )\n",
    "df_hrrr_plants_coarse.drop(\n",
    "    columns={'plant_code', 'surface', 'heightAboveGround', \n",
    "             'metpy_crs', 'gribfile_projection'},\n",
    "             inplace=True\n",
    "             )\n",
    "df_hrrr_plants_coarse.rename(\n",
    "    columns={'time':'time_utc','valid_time':'valid_time_utc',\n",
    "             'point':'plant_code'},\n",
    "             inplace=True\n",
    "             )\n",
    "\n",
    "# max\n",
    "df_hrrr_plants_coarse_max = ds_coarse_max.herbie.nearest_points(points=plants_df,\n",
    "                                           names=plants_df['plant_code']) \\\n",
    "                                            .to_dataframe().reset_index()\n",
    "df_hrrr_plants_coarse_max = df_hrrr_plants_coarse_max.merge(\n",
    "    plants_df[['plant_code','ac_capacity']],\n",
    "    left_on='point', right_on='plant_code'\n",
    "    )\n",
    "df_hrrr_plants_coarse_max.drop(\n",
    "    columns={'plant_code', 'surface', 'heightAboveGround', \n",
    "             'metpy_crs', 'gribfile_projection'},\n",
    "             inplace=True\n",
    "             )\n",
    "df_hrrr_plants_coarse_max.rename(\n",
    "    columns={'time':'time_utc','valid_time':'valid_time_utc',\n",
    "             'point':'plant_code'},\n",
    "             inplace=True\n",
    "             )\n",
    "\n",
    "# min\n",
    "df_hrrr_plants_coarse_min = ds_coarse_min.herbie.nearest_points(points=plants_df,\n",
    "                                           names=plants_df['plant_code']) \\\n",
    "                                            .to_dataframe().reset_index()\n",
    "df_hrrr_plants_coarse_min = df_hrrr_plants_coarse_min.merge(\n",
    "    plants_df[['plant_code','ac_capacity']],\n",
    "    left_on='point', right_on='plant_code'\n",
    "    )\n",
    "df_hrrr_plants_coarse_min.drop(\n",
    "    columns={'plant_code', 'surface', 'heightAboveGround', \n",
    "             'metpy_crs', 'gribfile_projection'},\n",
    "             inplace=True\n",
    "             )\n",
    "df_hrrr_plants_coarse_min.rename(\n",
    "    columns={'time':'time_utc','valid_time':'valid_time_utc',\n",
    "             'point':'plant_code'},\n",
    "             inplace=True\n",
    "             )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sort as times may end up out of order after saving and loading the .netCDF file?\n",
    "df_hrrr_plants_coarse.sort_values(by=['valid_time_utc'], inplace=True)\n",
    "df_hrrr_plants_coarse_min.sort_values(by=['valid_time_utc'], inplace=True)\n",
    "df_hrrr_plants_coarse_max.sort_values(by=['valid_time_utc'], inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then loop through each plant:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "decomposition_model = 'erbs' # dirint or erbs\n",
    "albedo = 0.2 # generic albedo value to use for the forecasts - something better could be added later\n",
    "\n",
    "p_ac_hrrr_coarse = {}\n",
    "p_ac_hrrr_coarse_max = {}\n",
    "p_ac_hrrr_coarse_min = {}\n",
    "\n",
    "for x in range(0,num_plants):\n",
    "    plant_data = plants_df.iloc[x].dropna().to_dict()\n",
    "\n",
    "    # coarse mean\n",
    "    resource_data_coarse = prep_hrrr_data(\n",
    "        plant_data,\n",
    "        forecast_df=df_hrrr_plants_coarse,\n",
    "        decomposition_model=decomposition_model,\n",
    "        albedo=albedo,\n",
    "    )\n",
    "    p_ac_hrrr_coarse[x] = pvlib_power_model(\n",
    "        resource_data_coarse,\n",
    "        plant_data,\n",
    "        calc_clear_sky=False,\n",
    "    )\n",
    "\n",
    "    # coarse max\n",
    "    resource_data_coarse_max = prep_hrrr_data(\n",
    "        plant_data,\n",
    "        forecast_df=df_hrrr_plants_coarse_max,\n",
    "        decomposition_model=decomposition_model,\n",
    "        albedo=albedo,\n",
    "    )\n",
    "    p_ac_hrrr_coarse_max[x] = pvlib_power_model(\n",
    "        resource_data_coarse_max,\n",
    "        plant_data,\n",
    "        calc_clear_sky=False,\n",
    "    )\n",
    "\n",
    "    # coarse min\n",
    "    resource_data_coarse_min = prep_hrrr_data(\n",
    "        plant_data,\n",
    "        forecast_df=df_hrrr_plants_coarse_min,\n",
    "        decomposition_model=decomposition_model,\n",
    "        albedo=albedo,\n",
    "    )\n",
    "    p_ac_hrrr_coarse_min[x] = pvlib_power_model(\n",
    "        resource_data_coarse_min,\n",
    "        plant_data,\n",
    "        calc_clear_sky=False,\n",
    "    )"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then sum all plants for each forecast:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "p_ac_hrrr_coarse_all = pd.concat(p_ac_hrrr_coarse, axis=1, sort=False).sum(axis=1)\n",
    "p_ac_hrrr_coarse_all = p_ac_hrrr_coarse_all.to_frame(name='power_ac')\n",
    "p_ac_hrrr_coarse_all.index.names = ['time_center_labeled']\n",
    "\n",
    "p_ac_hrrr_coarse_max_all = pd.concat(p_ac_hrrr_coarse_max, axis=1, sort=False).sum(axis=1)\n",
    "p_ac_hrrr_coarse_max_all = p_ac_hrrr_coarse_max_all.to_frame(name='power_ac')\n",
    "p_ac_hrrr_coarse_max_all.index.names = ['time_center_labeled']\n",
    "\n",
    "p_ac_hrrr_coarse_min_all = pd.concat(p_ac_hrrr_coarse_min, axis=1, sort=False).sum(axis=1)\n",
    "p_ac_hrrr_coarse_min_all = p_ac_hrrr_coarse_min_all.to_frame(name='power_ac')\n",
    "p_ac_hrrr_coarse_min_all.index.names = ['time_center_labeled']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Results\n",
    "Now create a plot of everything together:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(power_ac_hrrr_all.index, power_ac_hrrr_all.power_ac, drawstyle='steps-mid',\n",
    "         label='HRRR')\n",
    "plt.plot(p_ac_hrrr_coarse_all.index, p_ac_hrrr_coarse_all.power_ac, drawstyle='steps-mid',\n",
    "         label='HRRR coarsened')\n",
    "plt.plot(power_ac_cs_all.loc[mask].index, power_ac_cs_all.loc[mask].power_ac,\n",
    "         drawstyle='steps-mid', label='clear sky')\n",
    "plt.fill_between(p_ac_hrrr_coarse_min_all.index,\n",
    "                 p_ac_hrrr_coarse_min_all.power_ac,\n",
    "                 p_ac_hrrr_coarse_max_all.power_ac,\n",
    "                 step='mid', color=\"gray\", alpha=0.3, label='HRRR range')\n",
    "plt.plot(power_ac_nsrdb_all.loc[mask].index, \n",
    "         power_ac_nsrdb_all.loc[mask].power_ac,\n",
    "         drawstyle='steps-mid',\n",
    "         label='NSRDB power')\n",
    "plt.legend()\n",
    "plt.gcf().autofmt_xdate()\n",
    "plt.xlabel('Time, UTC (mm-dd HH)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can see that there are a number of interval where the coarsened version of the HRRR forecast is a little closer to NSRDB-based actual power. And, for almost every interval, the actual power falls in the range of the \"min\" and \"max\" forecasts from the HRRR. Moving forward, we will only use the coarsened mean versino of the HRRR for our deterministic forecasts. \n",
    "\n",
    "Now, let's save a few key DataFrames for use later:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "p_ac_hrrr_coarse_all.to_pickle('example_data/p_ac_hrrr_coarse_all.pkl')\n",
    "p_ac_hrrr_coarse_max_all.to_pickle('example_data/p_ac_hrrr_coarse_max_all.pkl')\n",
    "p_ac_hrrr_coarse_min_all.to_pickle('example_data/p_ac_hrrr_coarse_min_all.pkl')\n",
    "power_ac_cs_all.to_pickle('example_data/power_ac_cs_all.pkl')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That's the end of this notebook, please move on to the next one, [03_considerations_for_actual_weather.ipynb](03_considerations_for_actual_weather.ipynb), where we will look at some considerations around actual weather and the decision to use HRRR-based actuals instead of the NSRDB for this project. "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "herbie_conda_venv",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}