phase2update.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Academic license - for non-commercial use only - expires 2022-09-07\n",
      "Using license file C:\\Users\\uqrbean1\\gurobi.lic\n",
      "Warning: linear constraint 91251 and linear constraint 91252 have the same name \"prec\"\n",
      "Warning: quadratic constraint 0 and quadratic constraint 1 have the same name \"buso1\"\n",
      "Warning: default constraint names used to write mps file\n",
      "Warning: default Q constraint names used to write mps file\n",
      "Warning: linear constraint 91251 and linear constraint 91252 have the same name \"prec\"\n",
      "Warning: quadratic constraint 0 and quadratic constraint 1 have the same name \"buso1\"\n",
      "Warning: default constraint names used to write mps file\n",
      "Warning: default Q constraint names used to write mps file\n",
      "Warning: linear constraint 91251 and linear constraint 91252 have the same name \"prec\"\n",
      "Warning: quadratic constraint 0 and quadratic constraint 1 have the same name \"buso1\"\n",
      "Warning: default constraint names used to write mps file\n",
      "Warning: default Q constraint names used to write mps file\n",
      "Warning: linear constraint 91251 and linear constraint 91252 have the same name \"prec\"\n",
      "Warning: quadratic constraint 0 and quadratic constraint 1 have the same name \"buso1\"\n",
      "Warning: default constraint names used to write mps file\n",
      "Warning: default Q constraint names used to write mps file\n",
      "Warning: linear constraint 91251 and linear constraint 91252 have the same name \"prec\"\n",
      "Warning: quadratic constraint 0 and quadratic constraint 1 have the same name \"buso1\"\n",
      "Warning: default constraint names used to write mps file\n",
      "Warning: default Q constraint names used to write mps file\n",
      "Warning: linear constraint 252201 and linear constraint 252202 have the same name \"prec\"\n",
      "Warning: quadratic constraint 0 and quadratic constraint 1 have the same name \"buso1\"\n",
      "Warning: default constraint names used to write mps file\n",
      "Warning: default Q constraint names used to write mps file\n",
      "Warning: linear constraint 252201 and linear constraint 252202 have the same name \"prec\"\n",
      "Warning: quadratic constraint 0 and quadratic constraint 1 have the same name \"buso1\"\n",
      "Warning: default constraint names used to write mps file\n",
      "Warning: default Q constraint names used to write mps file\n",
      "Warning: linear constraint 252201 and linear constraint 252202 have the same name \"prec\"\n",
      "Warning: quadratic constraint 0 and quadratic constraint 1 have the same name \"buso1\"\n",
      "Warning: default constraint names used to write mps file\n",
      "Warning: default Q constraint names used to write mps file\n",
      "Warning: linear constraint 252201 and linear constraint 252202 have the same name \"prec\"\n",
      "Warning: quadratic constraint 0 and quadratic constraint 1 have the same name \"buso1\"\n",
      "Warning: default constraint names used to write mps file\n",
      "Warning: default Q constraint names used to write mps file\n",
      "Warning: linear constraint 252201 and linear constraint 252202 have the same name \"prec\"\n",
      "Warning: quadratic constraint 0 and quadratic constraint 1 have the same name \"buso1\"\n",
      "Warning: default constraint names used to write mps file\n",
      "Warning: default Q constraint names used to write mps file\n"
     ]
    }
   ],
   "source": [
    "#!/usr/bin/env python\n",
    "# coding: utf-8\n",
    "\n",
    "from gurobipy import *\n",
    "import numpy as np\n",
    "import pandas as pd \n",
    "import numpy as np\n",
    "os.chdir(\"C:\\\\Users\\\\uqrbean1\\\\OneDrive - The University of Queensland\\\\Documents\\\\ieee\")\n",
    "\n",
    "price = np.loadtxt(\"price.flat\")\n",
    "netload = np.loadtxt(\"ranger2.flat\") \n",
    "        \n",
    "## only do small -- 5 hours\n",
    "\n",
    "#flist = glob.glob(\"phase2_instance_small_0.txt\")\n",
    "\n",
    "flist = [    \n",
    "    \"phase2_instance_small_0.txt\",\n",
    "    \"phase2_instance_small_1.txt\",\n",
    "    \"phase2_instance_small_2.txt\",\n",
    "    \"phase2_instance_small_3.txt\",\n",
    "    \"phase2_instance_small_4.txt\",\n",
    "    \"phase2_instance_large_0.txt\",\n",
    "    \"phase2_instance_large_1.txt\",\n",
    "    \"phase2_instance_large_2.txt\",\n",
    "    \"phase2_instance_large_3.txt\",\n",
    "    \"phase2_instance_large_4.txt\"    \n",
    "        ]\n",
    "\n",
    "for file in flist:\n",
    "    count_data = np.genfromtxt(file,delimiter=' ',max_rows=1,usecols=(1,2,3,4,5),dtype=int)\n",
    "    building_count = count_data[0]\n",
    "    solar_count = count_data[1]\n",
    "    battery_count = count_data[2]\n",
    "    recurring_count = count_data[3]\n",
    "    once_count = count_data[4]\n",
    "    building_data = np.genfromtxt(file,delimiter=' ',skip_header=1,max_rows=building_count,usecols=(1,2,3),dtype=int)\n",
    "    building_num = building_data[:,0]\n",
    "    building_small = building_data[:,1]\n",
    "    building_large = building_data[:,2]\n",
    "    battery_data = np.genfromtxt(file,delimiter=' ',skip_header=1+solar_count+building_count,max_rows=battery_count,\n",
    "                                 usecols=(1,3,4,5),dtype=float)\n",
    "\n",
    "    #array([[  0.  , 150.  ,  75.  ,   0.85],\n",
    "    #       [  1.  , 420.  ,  60.  ,   0.6 ]])\n",
    "\n",
    "    # now variable columns\n",
    "\n",
    "    with open(file, 'r') as f:\n",
    "        file_lines = [line.rstrip() for line in f]\n",
    "\n",
    "    df =  pd.DataFrame([string.split(' ') for string in file_lines])\n",
    "    alldata = np.array(df)\n",
    "\n",
    "    recur_data = alldata[alldata[:,0]=='r',]\n",
    "    once_data = alldata[alldata[:,0]=='a',]\n",
    "\n",
    "    once_data = np.array(df)\n",
    "\n",
    "    recur_small = recur_data[recur_data[:,3]=='S']\n",
    "    recur_large = recur_data[recur_data[:,3]=='L']\n",
    "\n",
    "    T               = 2880                                      # number of time periods\n",
    "    timeperiods     = range(T)     \n",
    "    TX = 2881\n",
    "    timeperiodsx = range(TX)\n",
    "    \n",
    "    T1 = 532\n",
    "    timeperiods1 = range(T1)\n",
    "    \n",
    "    nBuildings = building_count \n",
    "    buildings = range(nBuildings)\n",
    "    \n",
    "    B = 2\n",
    "    batteries = range(B)\n",
    "\n",
    "    smallActIdx = pd.to_numeric(recur_small[:,1])\n",
    "    largeActIdx = pd.to_numeric(recur_large[:,1])\n",
    "\n",
    "    smallAct = range(smallActIdx.shape[0])\n",
    "    largeAct = range(largeActIdx.shape[0])\n",
    "\n",
    "    ActEnergy = pd.to_numeric(recur_data[:,4])\n",
    "\n",
    "    smallActPeriods = recur_small[:,5]\n",
    "    largeActPeriods = recur_large[:,5]\n",
    "    ActPeriods = pd.to_numeric(recur_data[:,5])\n",
    "\n",
    "    ActPrec = pd.to_numeric(recur_data[:,6])\n",
    "\n",
    "    ActRooms = pd.to_numeric(recur_data[:,2])\n",
    "\n",
    "    Act = range(recurring_count)\n",
    "    \n",
    "    bat_cap = [150, 420]   \n",
    "\n",
    "    week1 = list(range(88,120)) + list(range(184,216)) + list(range(280,312)) + list(range(376,408)) + list(range(472,504))\n",
    "    week2 = list(range(88+672,120+672)) + list(range(184+672,216+672)) + list(range(280+672,312+672)) + list(range(376+672,408+672)) + list(range(472+672,504+672))\n",
    "    week3 = list(range(88+672*2,120+672*2)) + list(range(184+672*2,216+672*2)) + list(range(280+672*2,312+672*2)) + list(range(376+672*2,408+672*2)) + list(range(472+672*2,504+672*2))\n",
    "    week4 = list(range(88+672*3,120+672*3)) + list(range(184+672*3,216+672*3)) + list(range(280+672*3,312+672*3)) + list(range(376+672*3,408+672*3)) + list(range(472+672*3,504+672*3))\n",
    "    #week5 = list(range(88+672*4,120+672*4)) + list(range(184+672*4,2880)) # Monday 30 Nov and 8 periods of Tuesday 1 Dec\n",
    "   \n",
    "    #weeks = week1 + week2 + week3 + week4 + week5\n",
    "    weeks = week1 + week2 + week3 + week4\n",
    "    \n",
    "    notweek = list(set(timeperiods) - set(weeks))\n",
    "    notweek1 = list(set(timeperiods) - set(week1)) \n",
    "        \n",
    "    m = Model(\"monash_schedule\")\n",
    "\n",
    "    ## for every Activity ... 0 to 49 or 0 to 199 ...\n",
    "    ## the \"bu\" matrix reflects activities * buildings -- can't mix small and large buildings\n",
    "\n",
    "    w            = m.addVars(Act, timeperiods, lb=-1, ub=1, name=\"w\") # intermediate variable for differences\n",
    "    v            = m.addVars(Act, timeperiods, name=\"v\") # max v \n",
    "    s            = m.addVars(Act, timeperiods, vtype=GRB.BINARY, name=\"s\") # binary allocation variable\n",
    "\n",
    "    bu           = m.addVars(Act, buildings, lb=0, ub=3, vtype=GRB.INTEGER, name=\"roomlimit\") # small_ac_buildings - sum to activity room count\n",
    "    small_use       = m.addVars(timeperiods, buildings, lb=0, name=\"small_use\")\n",
    "    large_use       = m.addVars(timeperiods, buildings, lb=0, name=\"large_use\")\n",
    "\n",
    "    actLoad      = m.addVars(Act, timeperiods, name=\"load\")\n",
    "    load         = m.addVars(timeperiods, name=\"load1\") # a variable load -- add net load and activity load and batteries \n",
    "    maxload      = m.addVar(name=\"maxload\")\n",
    "    maxrecurload = m.addVar(name=\"maxrecurload\")\n",
    "    obj_lin      = m.addVar(name=\"obj_lin\")\n",
    "    recurload = m.addVars(timeperiods, name=\"recurload\")\n",
    "    \n",
    "    disch  =   m.addVars(batteries, timeperiods, name=\"batd\", vtype=GRB.BINARY) # binary decision - ch or disch at time t, 0 to T-1\n",
    "    ch    =    m.addVars(batteries, timeperiods, name=\"batc\", vtype=GRB.BINARY)\n",
    "    batlevel = m.addVars(batteries,timeperiodsx, name=\"batlevel\", lb=0) ## 0 to T-1 and T itself \n",
    "    charge_effect = m.addVars(timeperiods, name=\"ch_effect\", lb=-GRB.INFINITY ) ## the net charge effect to calculate load1 then maxload \n",
    "    \n",
    "    #guess_max_netload = 744.33 ## guess max load for October 2020\n",
    "            \n",
    "    ## get the start hours - in the derived variable \"z\"\n",
    "\n",
    "    z  = m.addVars(Act, lb=0,ub=T1-1, name=\"z\")\n",
    "    z0 = m.addVars(Act, timeperiods1, name=\"z0\")\n",
    "    \n",
    "    for a in Act:\n",
    "        m.addConstrs((z0[(a,t)] == t*v[(a,t)]) for t in timeperiods1)\n",
    "\n",
    "    for a in Act:\n",
    "        m.addConstr(z[a] == max_(z0[a,t] for t in timeperiods1))\n",
    "\n",
    "    constr = m.addConstr((obj_lin == quicksum([(load[t] * price[t]) for t in timeperiods])/4000) , \"linear_objective\")\n",
    "    m.setObjective(obj_lin + maxload*maxload/200, GRB.MINIMIZE)\n",
    "        \n",
    "    constr = m.addConstrs((load[t] == (netload[t] + charge_effect[t] + recurload[t]) for t in week1), name=\"netload1\")\n",
    "    constr = m.addConstrs((load[t] == (netload[t] + charge_effect[t] + recurload[t-672]) for t in week2), name=\"netload2\")\n",
    "    constr = m.addConstrs((load[t] == (netload[t] + charge_effect[t] + recurload[t-672*2]) for t in week3), name=\"netload3\")\n",
    "    constr = m.addConstrs((load[t] == (netload[t] + charge_effect[t] + recurload[t-672*3]) for t in week4), name=\"netload4\")\n",
    "    #constr = m.addConstrs((load[t] == (netload[t] + charge_effect[t] + recurload[t-672*4]) for t in week5), name=\"netloadx\")\n",
    "    constr = m.addConstrs((load[t] == (netload[t] + charge_effect[t]) for t in notweek), name=\"netload5\")\n",
    "\n",
    "    ## constraint s, the binary allocation variable -- activities must occur in blocks, and z calculates the start hour \n",
    "\n",
    "    constr = m.addConstrs((w[(a,t)] == (s[(a,t)] - s[(a,t-1)]) for a in Act for t in range(1,T1)), name=\"shift_starts_hat\")\n",
    "    constr = m.addConstrs((v[(a,t)] == max_(w[(a,t)], 0.0) for a in Act for t in range(1,T1)), name=\"shift_starts\")\n",
    "    constr = m.addConstrs((v[(a,0)] == s[(a,0)] for a in Act), name=\"shift_starts_init\")\n",
    "    constr = m.addConstrs((quicksum([v[(a,t)] for t in timeperiods1]) <= 1 for a in Act), name=\"shift_start_max\")\n",
    "\n",
    "    ## maxload is the load over all time periods \n",
    "    constr = m.addConstrs((recurload[t] == (quicksum([actLoad[(a,t)] for a in Act])) for t in week1), name=\"netloada\")\n",
    "    \n",
    "    ## should have been for t in weeks in this line\n",
    "    \n",
    "    constr = m.addConstr(maxload == max_([load[t] for t in weeks]),name=\"get_max_load\") ## for the objective function\n",
    "    constr = m.addConstr(maxrecurload == max_([recurload[t] for t in week1]),name=\"get_maxr_load1\")\n",
    "\n",
    "    ## the sum down timeperiods of s must equal the number of periods required for that activity\n",
    "\n",
    "    constr = m.addConstrs((quicksum([s[(a,t)] for t in week1]) == ActPeriods[a] for a in Act), name=\"col\")\n",
    "\n",
    "    #constr = m.addConstrs((small_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t)] for a in smallActIdx]) for b in buildings for t in timeperiods ),name=\"bus\")\n",
    "\n",
    "    #constr = m.addConstrs((large_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t)] for a in largeActIdx]) for b in buildings for t in timeperiods ),name=\"bul\")\n",
    "    \n",
    "    for b in buildings:\n",
    "        for t in week1:\n",
    "            constr = m.addConstr((small_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t)] for a in smallActIdx])),name=\"buso1\")\n",
    "            \n",
    "    for b in buildings:\n",
    "        for t in week2:\n",
    "            constr = m.addConstr((small_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t-672)] for a in smallActIdx])),name=\"buso2\")\n",
    "\n",
    "    for b in buildings:\n",
    "        for t in week3:\n",
    "            constr = m.addConstr((small_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t-672*2)] for a in smallActIdx])),name=\"buso3\")\n",
    "\n",
    "    for b in buildings:\n",
    "        for t in week4:\n",
    "            constr = m.addConstr((small_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t-672*3)] for a in smallActIdx])),name=\"buso4\")\n",
    "    \n",
    "    #for b in buildings:\n",
    "    #    for t in week5:\n",
    "    #        constr = m.addConstr((small_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t-672*4)] for a in smallActIdx])),name=\"buso5\")\n",
    "\n",
    "    for b in buildings:\n",
    "        for t in week1:\n",
    "            constr = m.addConstr((large_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t)] for a in largeActIdx])),name=\"bulo1\")\n",
    "            \n",
    "    for b in buildings:\n",
    "        for t in week2:\n",
    "            constr = m.addConstr((large_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t-672)] for a in largeActIdx])),name=\"bulo2\")\n",
    "\n",
    "    for b in buildings:\n",
    "        for t in week3:\n",
    "            constr = m.addConstr((large_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t-672*2)] for a in largeActIdx])),name=\"bulo3\")\n",
    "\n",
    "    for b in buildings:\n",
    "        for t in week4:\n",
    "            constr = m.addConstr((large_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t-672*3)] for a in largeActIdx])),name=\"bulo4\")\n",
    "        \n",
    "    #for b in buildings:\n",
    "    #    for t in week5:\n",
    "    #        constr = m.addConstr((large_use[(t,b)] == quicksum([bu[(a,b)]*s[(a,t-672*4)] for a in largeActIdx])),name=\"bulo5\") \n",
    "\n",
    "    constr = m.addConstrs((small_use[(t,b)] <= building_small[b] for t in timeperiods for b in buildings), name=\"small_lim\")\n",
    "    constr = m.addConstrs((large_use[(t,b)] <= building_large[b] for t in timeperiods for b in buildings), name=\"large_lim\")\n",
    "\n",
    "    const = m.addConstrs((quicksum([bu[(a,b)] for b in buildings]) == ActRooms[a] for a in smallActIdx ),name=\"smallrooms\")\n",
    "    const = m.addConstrs((quicksum([bu[(a,b)] for b in buildings]) == ActRooms[a] for a in largeActIdx ),name=\"largerooms\")\n",
    "\n",
    "    constr = m.addConstrs((quicksum([s[(a,t)] for t in range(0,88)]) == 0 for a in Act),name=\"zero1\")\n",
    "    constr = m.addConstrs((quicksum([s[(a,t)] for t in range(120,184)]) == 0 for a in Act),name=\"zero2\")\n",
    "    constr = m.addConstrs((quicksum([s[(a,t)] for t in range(216,280)]) == 0 for a in Act),name=\"zero3\")\n",
    "    constr = m.addConstrs((quicksum([s[(a,t)] for t in range(312,376)]) == 0 for a in Act),name=\"zero4\")\n",
    "    constr = m.addConstrs((quicksum([s[(a,t)] for t in range(408,472)]) == 0 for a in Act),name=\"zero5\")\n",
    "    constr = m.addConstrs((quicksum([s[(a,t)] for t in range(504,2880)]) == 0 for a in Act),name=\"zero6\")\n",
    "    for a in Act:\n",
    "        ro = pd.to_numeric(recur_data[a,6:])\n",
    "        co = int(ro[0])\n",
    "        for i in range(co):\n",
    "            m.addConstr((z[a] - z[int(ro[i+1])] >= 65),name=\"prec\") ## has to be on a previous day -- let's say at least 16 hours ago\n",
    "\n",
    "    constr = m.addConstrs((actLoad[(a,t)] == ActEnergy[a]*ActRooms[a]*s[(a,t)]) for a in Act for t in week1)\n",
    "    \n",
    "    #########################\n",
    "    \n",
    "    constr = m.addConstrs((disch[b,t] + ch[b,t] <= 1 for b in batteries for t in timeperiods), name=\"batres\")\n",
    "\n",
    "    ## do what cha like, but make sure for all t NETLOAD[t] + CHARGE_EFFECT[t] <= MAX(NETLOAD(T))\n",
    "    ## perhaps experiment with removing this if I like living on the edge?????\n",
    "\n",
    "    #constr = m.addConstrs((recurload[t] + charge_effect[t] <= maxrecurload for t in week1), name=\"batmaxlimit1\")\n",
    "    #constr = m.addConstrs((recurload[t-672] + charge_effect[t] <= maxrecurload for t in week2), name=\"batmaxlimit2\")\n",
    "    #constr = m.addConstrs((recurload[t-672*2] + charge_effect[t] <= maxrecurload for t in week3), name=\"batmaxlimit3\")\n",
    "    #constr = m.addConstrs((recurload[t-672*3] + charge_effect[t] <= maxrecurload for t in week4), name=\"batmaxlimit4\")\n",
    "    #constr = m.addConstrs((recurload[t-672*4] + charge_effect[t] <= maxrecurload for t in week5), name=\"batmaxlimit5\")\n",
    "    \n",
    "    constr = m.addConstrs((ch[b,t] == 0 for b in batteries for t in weeks), name=\"batpeak1\")\n",
    "    #constr = m.addConstrs((disch[0,t] + disch[1,t] >= 1 for t in weeks), name=\"batpeak2\") # one battery is disch\n",
    "\n",
    "    constr = m.addConstrs((batlevel[b,0] == bat_cap[b] for b in batteries), name=\"batstart\")\n",
    "    #constr = m.addConstr((mdis == max_([disch[0,t] for t in timeperiods])), name=\"batend\") ## must discharge battery 0 at some point \n",
    "    #constr = m.addConstr(mdis > 0.0,name=\"md0\")\n",
    "\n",
    "    bat_rate = [ 75, 60 ] \n",
    "    bat_ch    = [ 75/math.sqrt(0.85), 60/math.sqrt(0.60) ] # charging battery by 75 actually takes 81 kW, disch by 75 sends 69 \n",
    "    bat_disch = [ 75*math.sqrt(0.85), 60*math.sqrt(0.60) ] \n",
    "\n",
    "    constr = m.addConstrs((batlevel[b,t+1] == (batlevel[b,t] - bat_rate[b]*disch[b,t]/4 + bat_rate[b]*ch[b,t]/4 ) ##\n",
    "                          for t in timeperiods for b in batteries ), name=\"batlev\")\n",
    "\n",
    "    constr = m.addConstrs((batlevel[b,t] <= bat_cap[b] for t in timeperiodsx for b in batteries),name=\"batcap\")\n",
    "\n",
    "    constr = m.addConstrs((charge_effect[t] == quicksum([bat_ch[b] * ch[b,t] - bat_disch[b] * disch[b,t] for b in batteries]) \n",
    "                          for t in timeperiods), name=\"charge_effect\")\n",
    "    #################################     \n",
    "        \n",
    "    sfile = file.replace(\"phase2\",\"p2sol\\\\\") # the given start point\n",
    "    sfile2 = sfile.replace(\"_instance_\",\"\") # the given start point\n",
    "    sfile3 = sfile2.replace(\"_\",\"\") # the given start point\n",
    "    sfile4 = sfile3.replace(\"txt\",\"mps\") # the given start point\n",
    "    m.write(sfile4)\n",
    "        \n",
    "    #constr = m.addConstr(((maxrecurload+583)*(maxrecurload+583)-1257*1257) / 200 <= obj_lin + maxload*maxload/200 - 26985)\n",
    "    \n",
    "    #m.read(\"s45.sol\")"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Academic license - for non-commercial use only - expires 2022-09-07\n",
      "Using license file C:\\Users\\uqrbean1\\gurobi.lic\n"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "m.update()\n",
    "m.read(\"s45.sol\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Changed value of parameter timelimit to 14400.0\n",
      "   Prev: inf  Min: 0.0  Max: inf  Default: inf\n"
     ]
    }
   ],
   "source": [
    "m.Params.timelimit = 4*3600.0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Gurobi Optimizer version 9.1.2 build v9.1.2rc0 (win64)\n",
      "Thread count: 6 physical cores, 12 logical processors, using up to 12 threads\n",
      "Optimize a model with 120874 rows, 663435 columns and 428123 nonzeros\n",
      "Model fingerprint: 0xca922252\n",
      "Model has 1 quadratic objective term\n",
      "Model has 8161 quadratic constraints\n",
      "Model has 26602 general constraints\n",
      "Variable types: 507615 continuous, 155820 integer (155520 binary)\n",
      "Coefficient statistics:\n",
      "  Matrix range     [2e-05, 6e+02]\n",
      "  QMatrix range    [5e-03, 1e+00]\n",
      "  QLMatrix range   [1e+00, 6e+00]\n",
      "  Objective range  [1e+00, 1e+00]\n",
      "  QObjective range [1e-02, 1e-02]\n",
      "  Bounds range     [1e+00, 5e+02]\n",
      "  RHS range        [1e+00, 7e+02]\n",
      "  QRHS range       [3e+04, 3e+04]\n",
      "\n",
      "Warning: Completing partial solution with 144000 unfixed non-continuous variables out of 155820\n",
      "User MIP start produced solution with objective 26993.1 (1.18s)\n",
      "Loaded user MIP start with objective 26993.1\n",
      "Processed MIP start in 1.20 seconds\n",
      "\n",
      "Presolve removed 79734 rows and 617473 columns\n",
      "Presolve time: 3.99s\n",
      "Presolved: 123620 rows, 72977 columns, 356577 nonzeros\n",
      "Presolved model has 720 SOS constraint(s)\n",
      "Presolved model has 1 quadratic objective terms\n",
      "Presolved model has 2 quadratic constraint(s)\n",
      "Variable types: 14738 continuous, 58239 integer (34770 binary)\n",
      "\n",
      "Root simplex log...\n",
      "\n",
      "Iteration    Objective       Primal Inf.    Dual Inf.      Time\n",
      "       0    1.4094635e+04   0.000000e+00   0.000000e+00      6s\n",
      "   86849    2.4072807e+04   1.352579e+03   0.000000e+00     10s\n",
      "  129480    2.4073031e+04   1.073619e+03   0.000000e+00     15s\n",
      "  170289    2.4075088e+04   1.158148e+03   0.000000e+00     20s\n",
      "  206310    2.4076026e+04   1.168663e+03   0.000000e+00     25s\n",
      "  226318    2.4075893e+04   6.234963e+02   0.000000e+00     30s\n",
      "  245011    2.4298744e+04   3.398194e+02   0.000000e+00     35s\n",
      "  259137    2.4298818e+04   1.939436e+03   0.000000e+00     40s\n",
      "  272958    2.4298534e+04   3.635686e+02   0.000000e+00     45s\n",
      "  287646    2.4298647e+04   1.815492e+03   0.000000e+00     50s\n",
      "  301389    2.4296795e+04   1.104381e+03   0.000000e+00     55s\n",
      "  315383    2.4297010e+04   3.763500e+02   0.000000e+00     60s\n",
      "  328477    2.4297165e+04   2.954104e+02   0.000000e+00     65s\n",
      "  342365    2.4297355e+04   2.980530e+02   0.000000e+00     70s\n",
      "  355015    2.4297489e+04   4.808066e+02   0.000000e+00     75s\n",
      "  367635    2.4297605e+04   3.963439e+02   0.000000e+00     80s\n",
      "  379245    2.4297696e+04   3.048827e+02   0.000000e+00     85s\n",
      "  390579    2.4297767e+04   2.424076e+02   0.000000e+00     90s\n",
      "  403981    2.4535808e+04   1.119394e+02   0.000000e+00     95s\n",
      "  416324    2.4535901e+04   3.385907e+02   0.000000e+00    100s\n",
      "  427623    2.4535961e+04   1.389787e+03   0.000000e+00    105s\n",
      "  437276    2.4536029e+04   3.577278e+02   0.000000e+00    110s\n",
      "  448036    2.4536059e+04   1.067952e+02   0.000000e+00    115s\n",
      "  457769    2.4530222e+04   8.630232e+01   0.000000e+00    120s\n",
      "  468276    2.4530383e+04   1.690168e+03   0.000000e+00    125s\n",
      "  478582    2.4530503e+04   3.080294e+02   0.000000e+00    130s\n",
      "  488534    2.4530623e+04   1.169770e+02   0.000000e+00    135s\n",
      "  498315    2.4530752e+04   8.967334e+02   0.000000e+00    140s\n",
      "  508737    2.4530840e+04   9.938093e+01   0.000000e+00    145s\n",
      "  519454    2.4530972e+04   8.884652e+01   0.000000e+00    150s\n",
      "  529921    2.4624938e+04   2.012603e+02   0.000000e+00    155s\n",
      "  540047    2.4625040e+04   1.202223e+03   0.000000e+00    160s\n",
      "  549902    2.4625133e+04   7.488385e+01   0.000000e+00    165s\n",
      "  559971    2.4625191e+04   1.793201e+01   0.000000e+00    170s\n",
      "  569035    2.4625258e+04   3.409683e+01   0.000000e+00    175s\n",
      "  578739    2.4625319e+04   2.443173e+01   0.000000e+00    180s\n",
      "  588272    2.4625367e+04   2.122070e+01   0.000000e+00    185s\n",
      "  597513    2.4625406e+04   1.172942e+01   0.000000e+00    190s\n",
      "  607012    2.4625421e+04   6.922670e+01   0.000000e+00    195s\n",
      "  616503    2.4625454e+04   7.410139e+01   0.000000e+00    200s\n",
      "  625992    2.4625479e+04   1.334069e+01   0.000000e+00    205s\n",
      "  635708    2.4625492e+04   3.628464e+03   0.000000e+00    210s\n",
      "  650359    2.4628118e+04   0.000000e+00   0.000000e+00    214s\n",
      "\n",
      "Root relaxation: objective 2.462812e+04, 650359 iterations, 208.16 seconds\n",
      "\n",
      "    Nodes    |    Current Node    |     Objective Bounds      |     Work\n",
      " Expl Unexpl |  Obj  Depth IntInf | Incumbent    BestBd   Gap | It/Node Time\n",
      "\n",
      "     0     0 24628.1184    0 3182 26993.1134 24628.1184  8.76%     - 5414s\n",
      "H    0     0                    26950.552523 24628.1184  8.62%     - 5415s\n",
      "     0     0 24632.7519    0 3159 26950.5525 24632.7519  8.60%     - 8029s\n",
      "H    0     0                    26794.309877 24632.7519  8.07%     - 8030s\n",
      "     0     0 24633.0440    0 3016 26794.3099 24633.0440  8.07%     - 11140s\n",
      "     0     0          -    0      26794.3099 24633.0440  8.07%     - 14400s\n",
      "\n",
      "Cutting planes:\n",
      "  Gomory: 149\n",
      "  Cover: 98\n",
      "  Implied bound: 737\n",
      "  Clique: 489\n",
      "  MIR: 174\n",
      "  StrongCG: 1\n",
      "  Flow cover: 148\n",
      "  Flow path: 1\n",
      "  GUB cover: 1\n",
      "  Zero half: 16\n",
      "  Network: 393\n",
      "  RLT: 616\n",
      "  Relax-and-lift: 274\n",
      "\n",
      "Explored 1 nodes (24478316 simplex iterations) in 14400.47 seconds\n",
      "Thread count was 12 (of 12 available processors)\n",
      "\n",
      "Solution count 3: 26794.3 26950.6 26993.1 \n",
      "\n",
      "Time limit reached\n",
      "Best objective 2.679430987691e+04, best bound 2.463304403334e+04, gap 8.0661%\n"
     ]
    }
   ],
   "source": [
    "m.optimize()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "m.write(\"s45-testing.sol\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "m.optimize()"
   ]
  },
  {
   "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": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}