det_LP_comp_cost.ipynb

{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Test computational cost of Stochastic Programming (Scenario Optimisation)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Investigate computational complexity and runtime of scnearios optimisation for system design problem, where each scenario uses a realisation of uncertain battery efficiencies."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import csv\n",
    "import json\n",
    "import time\n",
    "import random\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import inspect\n",
    "from tqdm import tqdm\n",
    "from multiprocessing import Pool\n",
    "\n",
    "from citylearn.citylearn import CityLearnEnv\n",
    "from linmodel import LinProgModel\n",
    "from schema_builder import build_schema\n",
    "from sys_eval import evaluate_system"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# set up test parameters\n",
    "dataset_dir = os.path.join('data', 'A37_analysis_test') # dataset directory\n",
    "schema_path = os.path.join(dataset_dir, 'schema.json')\n",
    "opex_factor = 10\n",
    "pricing_dict = {'carbon':5e-1,'battery':1e3,'solar':2e3}\n",
    "\n",
    "with open(os.path.join(dataset_dir,'metadata_ext.json'),'r') as json_file:\n",
    "    annex_defaults = json.load(json_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set up base parameters of system.\n",
    "ids = [0,3,9,11,12,15,16,25,26,32,38,44,45,48,49]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2023-08-09 16:37:25\n",
      "# Buildings: 1\n",
      "[1751.25346574] [1056.4815295]\n",
      "Solve time: 17.1s\n",
      "2023-08-09 16:37:42\n",
      "# Buildings: 2\n",
      "[1754.68712609  126.01910736] [1057.63923157   91.93252796]\n",
      "Solve time: 97.4s\n",
      "2023-08-09 16:39:20\n",
      "# Buildings: 3\n",
      "[1754.68712609  126.01910736   26.65207726] [1057.63923157   91.93252796   23.16979028]\n",
      "Solve time: 233.3s\n",
      "2023-08-09 16:43:14\n",
      "# Buildings: 4\n",
      "[1751.25346574  124.62654802   26.64722756  927.73728492] [1056.4815295    91.59964675   23.16979028  661.26134874]\n",
      "Solve time: 472.6s\n",
      "2023-08-09 16:51:07\n",
      "# Buildings: 5\n",
      "[1751.25346574  125.39802019   26.65193204  928.2464427   239.71230186] [1056.4815295    91.73059042   23.17231638  658.87850467  155.0091287 ]\n",
      "Solve time: 779.5s\n",
      "2023-08-09 17:04:08\n",
      "# Buildings: 6\n",
      "[1746.81935754  124.55786909   26.62825556  924.25170666  238.21895761\n",
      "  375.55207824] [1054.98651377   91.45742741   23.16979028  660.15037594  155.40540541\n",
      "  231.99374383]\n",
      "Solve time: 1353.3s\n",
      "2023-08-09 17:26:42\n",
      "# Buildings: 7\n",
      "[1747.59176819  125.81966553   26.64771841  928.05808795  239.42925953\n",
      "  377.01521923  690.58365044] [1068.84422111   92.37175287   23.24561404  661.36359962  157.03201345\n",
      "  235.12781807  576.66666667]\n",
      "Solve time: 2023.7s\n",
      "2023-08-09 18:00:27\n",
      "# Buildings: 8\n",
      "[1745.62537315  125.78579513   26.64666785  928.05808795  238.4874301\n",
      "  376.20786242  690.38118201  572.86699823] [1068.71257485   92.35855215   23.26388889  661.36359962  156.67457557\n",
      "  234.85213483  576.27336699  363.28749034]\n",
      "Solve time: 2846.4s\n",
      "2023-08-09 18:47:55\n",
      "# Buildings: 9\n",
      "[1749.1348087   125.81966553   26.68244631  928.05808795  239.42925953\n",
      "  377.06965015  691.1716913   572.98975524 3513.98111314] [1068.84422111   92.37175287   23.2613614   661.36359962  157.03201345\n",
      "  235.29624526  578.22685789  363.28749034 1953.19665636]\n",
      "Solve time: 4220.9s\n",
      "2023-08-09 19:58:18\n",
      "# Buildings: 10\n",
      "[1748.84814549  125.81966553   26.64771841  927.90576959  239.05804634\n",
      "  376.4995614   690.02969975  572.86699823 3514.4824096   392.84769169] [1068.           92.37175287   23.24561404  661.31505056  156.89113268\n",
      "  235.11111111  576.58227848  363.28749034 1949.54117837  326.33087441]\n",
      "Solve time: 5558.9s\n",
      "2023-08-09 21:30:58\n",
      "# Buildings: 11\n",
      "[1749.47487322  125.81966553   26.64771841  928.05808795  239.12448791\n",
      "  376.4995614   690.02969975  572.86699823 3514.4824096   392.84769169\n",
      "  122.58797979] [1066.85989206   92.37175287   23.24561404  661.36359962  156.91634821\n",
      "  235.11111111  576.58227848  363.28749034 1949.54117837  326.33087441\n",
      "   82.1732452 ]\n",
      "Solve time: 7365.7s\n",
      "2023-08-09 23:33:46\n",
      "# Buildings: 12\n",
      "[1749.92286168  125.81966553   26.64771841  928.05808795  239.48037891\n",
      "  377.08250174  690.38118201  572.86699823 3515.10383793  393.19416235\n",
      "  123.02126263 3273.83574425] [1064.24581006   92.37175287   23.24561404  661.36359962  156.66666667\n",
      "  235.11111111  574.25218721  363.28749034 1949.39113757  326.2267373\n",
      "   81.70212766 1983.55734876]\n",
      "Solve time: 8531.1s\n",
      "2023-08-10 01:55:59\n",
      "# Buildings: 13\n",
      "[1745.83442289  125.44779528   26.64771841  927.73728492  238.32150117\n",
      "  374.93732209  686.02337547  572.5746788  3514.55942481  392.6594454\n",
      "  122.61320259 3273.83574425 3372.70819683] [1060.09615385   92.22681937   23.24561404  661.26134874  156.57680203\n",
      "  234.60377984  576.10619469  363.28749034 1949.18967086  326.25385375\n",
      "   81.70212766 1983.55734876 2257.08789928]\n",
      "Solve time: 9816.1s\n",
      "2023-08-10 04:39:38\n",
      "# Buildings: 14\n",
      "[1750.70932887  125.655299     26.64771841  927.73728492  238.32141449\n",
      "  374.93732209  686.94215982  572.5746788  3514.55942481  392.43562353\n",
      "  122.85601917 3272.40979976 3374.77599241 1373.5098467 ] [1059.65665236   92.30769231   23.24561404  661.26134874  156.61157025\n",
      "  234.60377984  576.10619469  363.28749034 1949.18967086  326.1622774\n",
      "   81.63265306 1982.79881322 2257.08789928 1010.62825106]\n",
      "Solve time: 11815.2s\n",
      "2023-08-10 07:56:36\n",
      "# Buildings: 15\n",
      "[1745.83623699  125.32077498   26.64771841  927.73728492  238.09036275\n",
      "  374.93732209  685.95676175  572.5746788  3514.55942481  392.22382235\n",
      "  122.66674667 3272.40979976 3373.58517241 1372.85783877  145.82448211] [1059.25989722   92.17731421   23.24561404  661.26134874  156.52388278\n",
      "  234.60377984  576.58227848  363.28749034 1949.18967086  326.0756193\n",
      "   81.62982948 1982.79881322 2257.08789928 1010.44393991  110.55058779]\n",
      "Solve time: 14432.8s\n"
     ]
    }
   ],
   "source": [
    "num_buildings = range(1,len(ids)+1)\n",
    "runtimes = {}\n",
    "\n",
    "for B in num_buildings:\n",
    "    print(time.strftime(\"%Y-%m-%d %H:%M:%S\", time.localtime()))\n",
    "    print(f\"# Buildings: {B}\")\n",
    "\n",
    "    base_kwargs = {\n",
    "            'output_dir_path':dataset_dir,\n",
    "            'building_names': ['UCam_Building_%s'%id for id in ids][:B],\n",
    "            'battery_energy_capacities': [annex_defaults[\"building_attributes\"][\"battery_energy_capacities (kWh)\"][str(id)] for id in ids][:B],\n",
    "            'battery_power_capacities': [annex_defaults[\"building_attributes\"][\"battery_power_capacities (kW)\"][str(id)] for id in ids][:B],\n",
    "            'battery_efficiencies': [0.85]*B,\n",
    "            'pv_power_capacities': [annex_defaults[\"building_attributes\"][\"pv_power_capacities (kW)\"][str(id)] for id in ids][:B],\n",
    "            'load_data_paths': ['UCam_Building_%s.csv'%id for id in ids][:B],\n",
    "            'weather_data_path': 'weather.csv',\n",
    "            'carbon_intensity_data_path': 'carbon_intensity.csv',\n",
    "            'pricing_data_path': 'pricing.csv',\n",
    "            'schema_name': 'schema_temp'\n",
    "        }\n",
    "\n",
    "    schema_path = build_schema(**base_kwargs)\n",
    "\n",
    "    # Initialise CityLearn environment object.\n",
    "    env = CityLearnEnv(schema=schema_path)\n",
    "    lp = LinProgModel(env=env)\n",
    "\n",
    "    lp.set_time_data_from_envs()\n",
    "    lp.generate_LP(clip_level='m',design=True,pricing_dict=pricing_dict,opex_factor=opex_factor)\n",
    "    lp.set_LP_parameters()\n",
    "\n",
    "    start = time.time()\n",
    "    lp_results = lp.solve_LP(verbose=False,ignore_dpp=True)\n",
    "    end = time.time()\n",
    "\n",
    "    print(lp_results['battery_capacities'],lp_results['solar_capacities'])\n",
    "\n",
    "    runtimes.update({B: end-start})\n",
    "    print(f\"Solve time: {round(runtimes[B],1)}s\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{1: 17.130981922149658, 2: 97.37760543823242, 3: 233.32676362991333, 4: 472.6319968700409, 5: 779.5362808704376, 6: 1353.336327791214, 7: 2023.7432000637054, 8: 2846.3628962039948, 9: 4220.939216852188, 10: 5558.859155893326, 11: 7365.732852458954, 12: 8531.051749706268, 13: 9816.090405464172, 14: 11815.228953361511, 15: 14432.782981157303}\n"
     ]
    }
   ],
   "source": [
    "print(runtimes)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[2.51541373 2.75738424]\n"
     ]
    }
   ],
   "source": [
    "# Fit linear trend to log(runtime)\n",
    "trend = np.polyfit(np.log(list(runtimes.keys())),np.log(list(runtimes.values())),deg=1)\n",
    "print(trend)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "58.11777295433248\n"
     ]
    }
   ],
   "source": [
    "# Find max allowable samples\n",
    "runtime_limit = 60*60*24*5\n",
    "scenarios_limit = np.exp((np.log(runtime_limit) - trend[1])/trend[0])\n",
    "print(scenarios_limit)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "from labellines import labelLines"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot computational time results\n",
    "max_scenarios = 30\n",
    "\n",
    "fig,ax = plt.subplots()\n",
    "ax.plot(runtimes.keys(),runtimes.values(),'ok')\n",
    "ax.plot(np.arange(1,max_scenarios+1),np.exp(np.log(np.arange(1,max_scenarios+1))*trend[0] + trend[1]),'k--',label=round(trend[0],2))\n",
    "plt.xlabel(\"Number of buildings\")\n",
    "plt.xscale('log')\n",
    "plt.ylabel('LP Solve Time (s)')\n",
    "plt.yscale('log')\n",
    "plt.xlim(0.9,30)\n",
    "plt.ylim(10,1e5)\n",
    "plt.grid(True,'major',alpha=0.5,linestyle='--')\n",
    "labelLines(ax.get_lines(),xvals=[20], fontsize=10, backgroundcolor='w')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "VOI-CAS",
   "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": "3.8.16"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}