OSeMOSYS_TS_calculations.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import datetime\n",
    "import numpy as np\n",
    "import seaborn as sns; sns.set()\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "%matplotlib auto"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "demand_df = pd.read_csv(r'hourly_demand_all_regions.csv', encoding='latin-1')\n",
    "demand_df = pd.DataFrame(demand_df)\n",
    "demand_df.rename({'Año':'Year', 'Mes':'Month', 'Día':'Day', 'Hora':'Hour'}, inplace=True, axis=1)\n",
    "demand_df.drop(['La Paz', 'SIN', 'SIN8', 'SIN8+LPZ'], axis=1, inplace=True)\n",
    "demand_df['09-Baja California Sur'] = demand_df['09-Baja California Sur'] - demand_df['10-Mulegé']\n",
    "#demand_df = pd.melt(demand_df, id_vars= ['Year', 'Month','Day','Hour'], value_vars=[x for x in demand_df.columns if x not in ['Year', 'Month','Day','Hour']])\n",
    "demand_df.rename({'variable':'Region'}, inplace=True, axis=1)\n",
    "#demand_df.drop(['Day','Hour'], axis=1, inplace=True)\n",
    "\n",
    "# Create column for weekday/weekend\n",
    "demand_df['Date'] = demand_df['Year'].astype(str) + '-' + demand_df['Month'].astype(str) + '-' + demand_df['Day'].astype(str)\n",
    "demand_df['Date'] = pd.to_datetime(demand_df['Date'], errors='coerce', format='%Y-%m-%d')\n",
    "demand_df['Day-of-week'] = demand_df['Date'].dt.dayofweek\n",
    "demand_df.drop('Date', axis=1, inplace=True)\n",
    "demand_df['Day-of-week'][demand_df['Day-of-week']<5] = 'WD'\n",
    "demand_df['Day-of-week'][demand_df['Day-of-week']!='WD'] = 'WE'\n",
    "\n",
    "\n",
    "print(demand_df['Day-of-week'].unique())\n",
    "print(demand_df['Year'].unique())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "seasons = {\n",
    "    1:'SEA1',\n",
    "    2:'SEA1',\n",
    "    3:'SEA2',\n",
    "    4:'SEA2',\n",
    "    5:'SEA2',\n",
    "    6:'SEA3',\n",
    "    7:'SEA3',\n",
    "    8:'SEA3',\n",
    "    9:'SEA4',\n",
    "    10:'SEA4',\n",
    "    11:'SEA4',\n",
    "    12:'SEA1',\n",
    "}\n",
    "\n",
    "day_parts = {\n",
    "    'DP1':[0,7],\n",
    "    'DP2':[7,14],\n",
    "    'DP3':[14,21],\n",
    "    'DP4':[21,24],\n",
    "    \n",
    "}\n",
    "\n",
    "timeslices = {}\n",
    "\n",
    "\n",
    "regions = [x for x in demand_df.columns if x not in ['Hour', 'Day', 'Month', 'Year', 'Day-of-week']]\n",
    "years = demand_df['Year'].unique()\n",
    "\n",
    "\n",
    "demand_df_hours = pd.DataFrame(columns=demand_df.columns)\n",
    "\n",
    "for each_year in years:\n",
    "    demand_df_year = demand_df[demand_df['Year']==each_year]\n",
    "    demand_df_year.insert(4,'Hour_continuous',range(1,1+len(demand_df_year)))\n",
    "    demand_df_hours = demand_df_hours.append(demand_df_year)\n",
    "\n",
    "demand_df_hours['Month'].replace(seasons, inplace=True)\n",
    "demand_df_hours['Total']  = demand_df_hours[regions].sum(axis=1)\n",
    "\n",
    "df_yearsplit = demand_df_hours[['Total', 'Month', 'Day-of-week', 'Hour', 'Year']]\n",
    "df_yearsplit = df_yearsplit[df_yearsplit['Year']==2018]\n",
    "\n",
    "for each_season in df_yearsplit['Month'].unique():\n",
    "    for each_day in df_yearsplit['Day-of-week'].unique():\n",
    "        df_temp = df_yearsplit[(df_yearsplit['Month']==each_season) \n",
    "                           & (df_yearsplit['Day-of-week'] == each_day)].sort_values(by='Total')\n",
    "        df_temp = df_temp.groupby(['Month', 'Day-of-week', 'Hour', 'Year']).mean().sort_values(by='Total')\n",
    "        \n",
    "        for each_DP in day_parts:\n",
    "            #print(each_DP, day_parts[each_DP][0], day_parts[each_DP][1])\n",
    "            df_temp_2 = df_temp[day_parts[each_DP][0]:day_parts[each_DP][1]]\n",
    "            df_temp_2['Day_part'] = each_DP\n",
    "            df_temp_2 = df_temp_2.reset_index()\n",
    "            df_temp_2['Timeslice_key'] = df_temp_2['Month'] + df_temp_2['Day-of-week'] + df_temp_2['Hour'].astype(str)\n",
    "            df_temp_2['Timeslice_value'] = df_temp_2['Month'] + df_temp_2['Day-of-week'] + df_temp_2['Day_part']\n",
    "            #print(df_temp_2)\n",
    "            timeslice_dict = dict(zip(df_temp_2['Timeslice_key'], \n",
    "                                      df_temp_2['Timeslice_value']))\n",
    "            timeslices.update(timeslice_dict)\n",
    "\n",
    "demand_df_hours['Timeslice'] = demand_df_hours['Month'] + demand_df_hours['Day-of-week'] + demand_df_hours['Hour'].astype(str)\n",
    "#print(demand_df_hours)\n",
    "\n",
    "demand_df_hours['Timeslice'].replace(timeslices, inplace=True)\n",
    "print(demand_df_hours)\n",
    "\n",
    "\n",
    "#demand_df = demand_df.groupby('Year').sum().mul(3.6*1e-6)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## YearSplit calculation\n",
    "\n",
    "demand_df_yearsplit = demand_df_hours.drop(['Total', 'Month', 'Day', 'Day-of-week', 'Hour', 'Hour_continuous'] \n",
    "                                           +regions, \n",
    "                                           axis=1)\n",
    "demand_df_yearsplit = demand_df_yearsplit.groupby('Year')['Timeslice'].value_counts(normalize=True).to_frame('value')\n",
    "demand_df_yearsplit = demand_df_yearsplit.reset_index()\n",
    "demand_df_yearsplit = pd.pivot_table(demand_df_yearsplit, \n",
    "                                   index='Timeslice', \n",
    "                                   values='value', \n",
    "                                   columns='Year', \n",
    "                                   aggfunc=np.sum)\n",
    "\n",
    "#print(demand_df_yearsplit)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## SpecifiedDemandProfile calculation\n",
    "\n",
    "demand_df_regions = demand_df_hours.drop(['Month', 'Day', 'Day-of-week', 'Hour', 'Hour_continuous'], axis=1)\n",
    "\n",
    "demand_df_regions = demand_df_regions.groupby(['Year', 'Timeslice']).sum()\n",
    "demand_df_total = demand_df_regions.groupby(['Year']).sum()\n",
    "\n",
    "demand_df_regions[regions] = demand_df_regions[regions]/demand_df_total[regions]\n",
    "demand_df_regions = demand_df_regions.reset_index()\n",
    "\n",
    "demand_df_regions = pd.melt(demand_df_regions, \n",
    "                            id_vars=['Year', 'Timeslice'], \n",
    "                            value_vars=regions)\n",
    "\n",
    "\n",
    "demand_df_regions = pd.pivot_table(demand_df_regions, \n",
    "                                   index=['variable','Timeslice'], \n",
    "                                   values='value', \n",
    "                                   columns='Year', \n",
    "                                   aggfunc=np.sum)\n",
    "\n",
    "demand_df_total = demand_df_total.mul(3.6*1e-6)\n",
    "demand_df_total = demand_df_total.reset_index()\n",
    "demand_df_total = pd.melt(demand_df_total, \n",
    "                            id_vars='Year', \n",
    "                            value_vars=regions)\n",
    "demand_df_total = pd.pivot_table(demand_df_total, \n",
    "                                   index='variable', \n",
    "                                   values='value', \n",
    "                                   columns='Year', \n",
    "                                   aggfunc=np.sum)\n",
    "#print(demand_df_regions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "demand_df_yearsplit.to_csv(r'yearsplit.csv')\n",
    "demand_df_regions.to_csv(r'specified_demand_profiles.csv')\n",
    "demand_df_total.to_csv(r'total_demand_PJ.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "demand_df_pvcf = demand_df_hours.drop(regions +\n",
    "                                     ['Month', 'Day', 'Hour', 'Day-of-week', 'Total'],\n",
    "                                     axis=1)\n",
    "demand_df_pvcf = demand_df_pvcf[demand_df_pvcf['Year']==2018].drop('Year', axis=1)\n",
    "\n",
    "df_pvcf = pd.read_csv(r'hourly_pvdata.csv')\n",
    "\n",
    "demand_df_pvcf = pd.merge(demand_df_pvcf, df_pvcf, on=['Hour_continuous']).drop('Hour_continuous', axis=1)\n",
    "demand_df_pvcf = demand_df_pvcf.groupby(['Timeslice']).mean()\n",
    "print(demand_df_pvcf)\n",
    "\n",
    "demand_df_pvcf.to_csv(r'pv_cf_timeslices.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}