Getting the Reward for Hurist versions-from-01.06.2021.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#  THis is the new version of the "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# THIS IS THE FIRST WORKING VERSION WHERE YOU ARE ALSO ABLE TO HAVE REAL REWARDS: \n",
    "#PROBLEMS:\n",
    "\n",
    "# 2. the path is not nicely made\n",
    "# 3 the aspen doc breaks every once in while\n",
    "# 4 the agent makes many many actions which dont exist..\n",
    "# 5 make a logging system test timing.\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "#import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import math\n",
    "import numpy as np\n",
    "import time \n",
    "\n",
    "#from gym import Env\n",
    "#from gym.spaces import Discrete, Box\n",
    "#import random\n",
    "#from keras.models import Sequential\n",
    "#from keras.layers import Dense,Flatten\n",
    "#from keras.optimizers import Adam\n",
    "#from rl.agents.dqn import DQNAgent\n",
    "#from rl.policy import EpsGreedyQPolicy, LinearAnnealedPolicy\n",
    "#from rl.memory import SequentialMemory\n",
    "\n",
    "\n",
    "def DCoperationcost(N, R, totflow):\n",
    "    # N = number of stages\n",
    "    # R = reflux ratio\n",
    "    # totflow is total flowrate into DC column in kmol/h\n",
    "    # see operational cost paper (Massimiliano Erricoa) for breakdown of\n",
    "    if totflow ==0: \n",
    "        stable_env.TotalflowerrorcounterGlobal = stable_env.TotalflowerrorcounterGlobal +1\n",
    "        print(f' \\n \\n \\n PROBLEM: the total flow is zero.. with Bcounter being: {stable_env.Bcounter} \\n \\n \\n ')\n",
    "    C = 296.1  # $/m2yr\n",
    "    Cdash = 17.76  # $/m2yr\n",
    "    Cddash = 20.61 * 10 ** -3  # $/kmol\n",
    "    Y = 8000  # h/yr\n",
    "    E = 90  # % (efficiency assumed)\n",
    "    G = 219.71  # kmol/hm2\n",
    "    Gdash = 0.49  # kmol/hm2\n",
    "    CEPCI2009 = 521.9\n",
    "    CEPCI2019 = 613.16\n",
    "    #ZAR = 15.23  # rands per dollar for 01/10/2019\n",
    "\n",
    "    TAC = (C / (E * Y * G) * N * (1 + R) + Cdash / (Y * Gdash) * (1 + R) + Cddash * (1 + R)) * CEPCI2019 / CEPCI2009\n",
    "    #TAC = TAC * ZAR  # TAC in rands\n",
    "    #This is just in case the value is zero...\n",
    "\n",
    "    TAC = TAC * totflow\n",
    "    \n",
    "    return TAC\n",
    "\n",
    "\n",
    "\n",
    "def columnsize(N, Temp, P, totflow):    # # C bar kmol/sec\n",
    "    #V=nRT/P\n",
    "    totflow = (totflow*1000)/3600 #now its mol/sec\n",
    "    Temp = Temp+273 #from C to K\n",
    "    Rgas = 8.31445\n",
    "    P = P*100000 #from bar to Pascal\n",
    "    volumetric_flow=(totflow*Temp*Rgas)/P #should be m3/sec\n",
    "    \n",
    "    # size in feet unless stated otherwise\n",
    "    # heuristics taken from Ludwig Rules of Thumb\n",
    "\n",
    "    # NB UNIT CONVERSIONS\n",
    "\n",
    "    # height\n",
    "    space = (20 / 12 + 2) / 2  # tray spacing\n",
    "    trayheight = (N - 1) * space  # height due to tray spacing\n",
    "    vapallowance = 4  # ft\n",
    "    liqallowance = 6  # ft\n",
    "    L = trayheight + vapallowance + liqallowance\n",
    "\n",
    "    # diameter\n",
    "    # classed according to heuristic ranges in order to maintain satisfactory diameters\n",
    "    n_intervals = 10\n",
    "    interval = np.linspace(27, 170, n_intervals-1)\n",
    "    interval = list(interval)\n",
    "\n",
    "    velocity = 0.6  # m/s using Ludwig\n",
    "\n",
    "    Area = volumetric_flow/velocity\n",
    "    Di = np.sqrt(Area*4/np.pi)\n",
    "    Di *= 3.28\n",
    "    #print(f'for Di needed: area{Area},voluflow{volumetric_flow},totflow{totflow},T{Temp},P{P}')\n",
    "\n",
    "    \"\"\"\n",
    "    LDratio = [1 / 9, 1 / 10, 1 / 11, 1 / 12, 1 / 13, 1 / 14, 1 / 15, 1 / 16, 1 / 17]\n",
    "    for i in range(n_intervals-2):\n",
    "        maximum = interval[i + 1]\n",
    "        minimum = interval[i]\n",
    "        if L > minimum and L < maximum:\n",
    "            Di = LDratio[i] * L\n",
    "            break\n",
    "    if L < interval[0]:\n",
    "        Di = LDratio[0] * L\n",
    "    if L > interval[-1]:\n",
    "        Di = LDratio[-1] * L\n",
    "    if N < 11:\n",
    "        print(\"Too few stages- resulting column height too low\")\n",
    "    \"\"\"\n",
    "\n",
    "    return L, Di\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "def expensiveDC(Di, L, P, T, N):    # ft ft bar C \n",
    "\n",
    "    # VESSEL COSTING\n",
    "    # Di: feet\n",
    "    # L: feet\n",
    "    Po = P*100  #from bar to kpa\n",
    "    # Po: kpa\n",
    "    # To: deg c\n",
    "\n",
    "    To = (T * 9 / 5) + 32  # convert to Celcius to farenheit\n",
    "    Po = (Po - 101.325) * 0.145038  # kPa to Psig\n",
    "    # calculating operational pressure\n",
    "    if Po < 10:  # all pressures in psig\n",
    "        Pd = 10\n",
    "    elif Po < 1000:\n",
    "        Pd = math.exp(0.60608 + 0.91615 * (math.log(Po)) + 0.0015655 * (math.log(Po)) ** 2)\n",
    "    else:\n",
    "        Pd = 1.1 * Po\n",
    "    #print(Pd)\n",
    "\n",
    "    # calculating S\n",
    "    Td = To + 50  # degrees F\n",
    "    interval = np.linspace(650, 900, 6)\n",
    "    interval = list(interval)\n",
    "    interval.insert(0, -20)\n",
    "    Sreturn = [15000, 15000, 15000, 14750, 14200, 13100]\n",
    "    for i in range(6):\n",
    "        maximum = interval[i + 1]\n",
    "        minimum = interval[i]\n",
    "\n",
    "        if Td > minimum and Td < maximum:\n",
    "            S = Sreturn[i]\n",
    "            break\n",
    "\n",
    "    if Td > interval[-1]:\n",
    "        S = Sreturn[-1]\n",
    "        print(\"Smax\")\n",
    "\n",
    "    if Td < interval[0]:\n",
    "        S = Sreturn[0]\n",
    "        print(\"Smin\")\n",
    "    S = Sreturn[0]\n",
    "\n",
    "\n",
    "    # calculating E\n",
    "    E = 0.85  # NB can change dependent on tp\n",
    "    #print(f'for tp needed: di:{Di},pd{Pd},S{S},E{E},')\n",
    "    # calculating vessel wall thickness (tp)\n",
    "    tp = (Pd * Di * 12) / (2 * S * E - 1.2 * Pd)  # converts Di to inches\n",
    "    #print(tp)\n",
    "\n",
    "    #print(f'for W needed: Di{Di},tp{tp},L{L},')\n",
    "    # weight of vessel and heads\n",
    "    W = math.pi * (Di * 12 + tp) * (L + 0.8 * Di) * 12 * 0.284  # assumes no wind, earthquakes or corrosion, else tp changes\n",
    "    #print(W)  # pounds\n",
    "    #print(f'HEy the W is {W}')\n",
    "    # calculating Cv, Cpl and Cp\n",
    "    # assumes carbon steel\n",
    "    CEPCI2003 = 394\n",
    "    CEPCI2019 = 613.16\n",
    "    Fm = 1  # carbon steel, can change\n",
    "    #ZAR = 15.23  # rands per dollar for 01/10/2019\n",
    "    if W ==0: W= 5      #This just makes it run if there is no actual flow\n",
    "    Cv2003 = math.exp(7.0374 + 0.18255 * (math.log(W)) + 0.02297 * (math.log(W)) ** 2)\n",
    "    Cv = Cv2003 * CEPCI2019 / CEPCI2003\n",
    "\n",
    "    Cpl2003 = 237.1 * Di ** 0.63316 * L ** 0.80161\n",
    "    Cpl = Cpl2003 * CEPCI2003 / CEPCI2019\n",
    "\n",
    "    Cp = Fm * Cv + Cpl  # USD\n",
    "    #Cp = Cp * ZAR\n",
    "    #print(Cp)\n",
    "\n",
    "    # calculating tray cost\n",
    "    # assumes carbon steel sieve trays - else see Seader pg. 532\n",
    "    Fmt = 1  # material factor for carbon steel\n",
    "    Ftt = 1  # tray factor for sieve trays\n",
    "    if N >= 20:\n",
    "        Fnt = 1\n",
    "    else:\n",
    "        Fnt = 2.25 / (1.0414 ** N)\n",
    "\n",
    "    CBT = 369 * math.exp(0.1739 * Di)\n",
    "    Ctray2003 = N * Fmt * Ftt * Fnt * CBT\n",
    "    Ctray = Ctray2003 * CEPCI2019 / CEPCI2003  # USD\n",
    "    #Ctray = Ctray * ZAR\n",
    "\n",
    "    # calculating total capital cost\n",
    "    Ctot = Cp + Ctray\n",
    "    if W == 0: Ctot = 0\n",
    "    \"\"\"\n",
    "    # throwing error if operating outside of cost correlation limits\n",
    "    if W < 4200:\n",
    "        raise ValueError(\"Vessel weight is too low\")\n",
    "    elif W > 1000000:\n",
    "        raise ValueError(\"Vessel weight is too high\")\n",
    "    elif Di < 3:\n",
    "        raise ValueError(\"Vessel internal diameter is too low\")\n",
    "    elif Di > 17:\n",
    "        raise ValueError(\"Vessel internal diameter is too high\")\n",
    "    elif L < 27:\n",
    "        raise ValueError(\"Vessel height is too low\")\n",
    "    elif L > 170:\n",
    "        raise ValueError(\"Vessel height is too large\")\n",
    "    elif Td < -20:\n",
    "        raise ValueError(\"Vessel design temp is too low\")\n",
    "    elif Td > 900:\n",
    "        raise ValueError(\"Vessel design temp is too high\")\n",
    "    else:\n",
    "        return Ctot\n",
    "    \"\"\"\n",
    "\n",
    "    # throwing error if operating outside of cost correlation limits\n",
    "    \"\"\"\n",
    "    if W < 4200:\n",
    "        print(\"Vessel weight is too low\")\n",
    "    elif W > 1000000:\n",
    "        print(\"Vessel weight is too high\")\n",
    "    elif Di < 3:\n",
    "        print(\"Vessel internal diameter is too low\")\n",
    "    elif Di > 17:\n",
    "        print(\"Vessel internal diameter is too high\")\n",
    "    elif L < 27:\n",
    "        print(\"Vessel height is too low\")\n",
    "    elif L > 170:\n",
    "        print(\"Vessel height is too large\")\n",
    "    elif Td < -20:\n",
    "        print(\"Vessel design temp is too low\")\n",
    "    elif Td > 900:\n",
    "        print(\"Vessel design temp is too high\")\n",
    "    \"\"\"\n",
    "    return Ctot\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "bla = 1\n",
    "if bla == 1:\n",
    "    if bla == 1:\n",
    "\n",
    "        import xlwings as xw\n",
    "        wb = xw.Book(r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\test1.xlsm')\n",
    "\n",
    "        import random\n",
    "\n",
    "        path = r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC'      #Folder where you place all our stuff\n",
    "        book = 'test1.xlsm'     # Name of the excel document\n",
    "        module = 'Modul1'     #VBA module in the excel dokument \n",
    "        wb = xw.Book(r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\test1.xlsm')\n",
    "        \n",
    "        \n",
    "        \n",
    "        import os\n",
    "        from win32com import client\n",
    "        EXCEL_CLS_NAME = \"Excel.Application\"\n",
    "        class XlMacro:\n",
    "            def __init__(self, path, book, module, name, *args):\n",
    "                self._path = path  # path containing workbook\n",
    "                self._book = book  # workbook name like Book1.xlsm\n",
    "                self._module = module  # module name, e.g., Module1\n",
    "                self._name = name  # procedure or function name\n",
    "                self._params = args  # argument(s)\n",
    "                self._wb = None\n",
    "            @property\n",
    "            def workbook(self):\n",
    "                return self._wb\n",
    "            @property\n",
    "            def wb_path(self):\n",
    "                return os.path.join(self._path, self._book)\n",
    "            @property\n",
    "            def name(self):\n",
    "                return f'{self._book}!{self._module}.{self._name}'\n",
    "            @property\n",
    "            def params(self):\n",
    "                return self._params\n",
    "            def get_workbook(self):\n",
    "                wb_name = os.path.basename(self.wb_path)\n",
    "                try:\n",
    "                    xl = client.GetActiveObject(EXCEL_CLS_NAME)\n",
    "                except:\n",
    "                    # Excel is not running, so we need to handle it.\n",
    "                    xl = client.Dispatch(EXCEL_CLS_NAME)\n",
    "                if wb_name in [wb.Name for wb in xl.Workbooks]:\n",
    "                    return xl.Workbooks[wb_name]\n",
    "                else:\n",
    "                    return xl.Workbooks.Open(self.wb_path)\n",
    "            def Run(self, *args, **kwargs):\n",
    "                \"\"\" \n",
    "                Runs an Excel Macro or evaluates a UDF \n",
    "                returns None for macro, or the return value of Function\n",
    "                NB: there is no error-handling here, but there should be!\n",
    "                \"\"\"\n",
    "                keep_open = kwargs.get('keep_open', True)\n",
    "                save_changes = kwargs.get('save_changes', False)\n",
    "                self._wb = self.get_workbook()\n",
    "                xl_app = self._wb.Application\n",
    "                xl_app.Visible = True\n",
    "                ret = None\n",
    "                if args is None:\n",
    "                    ret = xl_app.Run(self.name)\n",
    "                elif not args:\n",
    "                    # run with some default parameters\n",
    "                    ret = xl_app.Run(self.name, *self.params)\n",
    "                else:\n",
    "                    ret = xl_app.Run(self.name, *args)\n",
    "                if not keep_open:\n",
    "                    self.workbook.Close(save_changes)\n",
    "                    self._wb = None\n",
    "                    xl_app.Quit()\n",
    "                return ret\n",
    "            \n",
    "\n",
    "    \n",
    "    \n",
    "#wb = xw.Book(r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\test1.xlsm') \n",
    "#RemoveErrormessages = wb.macro('RemoveErrormessages')\n",
    "#RemoveErrormessages()    # This sets the suppressed errrormesssages to true\n",
    "\n",
    "\n",
    "import os\n",
    "import win32com.client as win32\n",
    "from gym import Env, spaces\n",
    "import numpy as np\n",
    "\n",
    "\n",
    "class Environment(Env):\n",
    "    def __init__(self, Bcounter, Feed_temp_init, Feed_pressure_init, Feed_flow_Compound1_init, Feed_flow_Compound2_init, Feed_flow_Compound3_init, Feed_flow_Compound4_init, Feed_flow_Compound5_init, Feed_flow_Compound6_init, COMPOUNDNAME1, COMPOUNDNAME2 , COMPOUNDNAME3, COMPOUNDNAME4, COMPOUNDNAME5, COMPOUNDNAME6, COMPOUNDSinlighttoheavystring, Nmax, Refluxratio, Pressure_condens, Pressure_reboiler, Lightkey_recovery ,Heavykey_recovery, OPT_NTRR, BlockpropertiesErrorWeight, DestillationErrorWeight, LifetimeInYears, ValueCompound1, ValueCompound2, ValueCompound3, ValueCompound4, ValueCompound5, ValueCompound6, ConcentrationNeededToGetPaidForStream, FeedValue, PunishmentStageNumber,Punishment_for_invalid_action, HardMaxNegReward):\n",
    "        self.Bcounter = Bcounter\n",
    "        self.Feed_temp_init = Feed_temp_init\n",
    "        self.Feed_pressure_init = Feed_pressure_init\n",
    "        self.Feed_flow_Compound1_init = Feed_flow_Compound1_init\n",
    "        self.Feed_flow_Compound2_init = Feed_flow_Compound2_init\n",
    "        self.Feed_flow_Compound3_init = Feed_flow_Compound3_init\n",
    "        self.Feed_flow_Compound4_init = Feed_flow_Compound4_init\n",
    "        self.Feed_flow_Compound5_init = Feed_flow_Compound5_init\n",
    "        self.Feed_flow_Compound6_init = Feed_flow_Compound6_init\n",
    "        self.COMPOUNDNAME1 = COMPOUNDNAME1\n",
    "        self.COMPOUNDNAME2 = COMPOUNDNAME2\n",
    "        self.COMPOUNDNAME3 = COMPOUNDNAME3\n",
    "        self.COMPOUNDNAME4 = COMPOUNDNAME4\n",
    "        self.COMPOUNDNAME5 = COMPOUNDNAME5\n",
    "        self.COMPOUNDNAME6 = COMPOUNDNAME6\n",
    "        self.COMPOUNDSinlighttoheavystring = COMPOUNDSinlighttoheavystring\n",
    "        self.Nmax = Nmax\n",
    "        self.Refluxratio = Refluxratio\n",
    "        self.Pressure_condens = Pressure_condens\n",
    "        self.Pressure_reboiler = Pressure_reboiler\n",
    "        self.Lightkey_recovery = Lightkey_recovery\n",
    "        self.Heavykey_recovery = Heavykey_recovery\n",
    "        self.OPT_NTRR = OPT_NTRR\n",
    "#REWARD DATA\n",
    " \n",
    "        self.BlockpropertiesErrorWeight = BlockpropertiesErrorWeight #in $ perError\n",
    "        self.DestillationErrorWeight = DestillationErrorWeight   #in $ perError\n",
    "        self.LifetimeInYears = LifetimeInYears         #years\n",
    "        self.ValueCompound1 = ValueCompound1          # $ per kmol/hour\n",
    "        self.ValueCompound2 = ValueCompound2\n",
    "        self.ValueCompound3 = ValueCompound3\n",
    "        self.ValueCompound4 = ValueCompound4\n",
    "        self.ValueCompound5 = ValueCompound5\n",
    "        self.ValueCompound6 = ValueCompound6\n",
    "        self.ValueIfPure = [ValueCompound1, ValueCompound2, ValueCompound3, ValueCompound4, ValueCompound5, ValueCompound6]\n",
    "        self.ConcentrationNeededToGetPaidForStream = ConcentrationNeededToGetPaidForStream    #needed concentration to get paid\n",
    "        self.FeedValue = FeedValue       #  $ per kmol/hour\n",
    "        self.TotalFeedFlow = Feed_flow_Compound1_init + Feed_flow_Compound2_init + Feed_flow_Compound3_init + Feed_flow_Compound4_init + Feed_flow_Compound5_init + Feed_flow_Compound6_init\n",
    "        self.restarteverything = False\n",
    "        self.HardMaxNegReward = HardMaxNegReward\n",
    "\n",
    "\n",
    "\n",
    "        import numpy as np\n",
    "        import xlwings as xw\n",
    "        #wb = xw.Book(r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\test1.xlsm')\n",
    "\n",
    "        \n",
    "## THIS IS JUST MAKING IT POSSIBLE TO EXTRACT STUFF FROM VBA TO PYTHON\n",
    "        \n",
    "        from win32com import client\n",
    "        EXCEL_CLS_NAME = \"Excel.Application\"\n",
    "        class XlMacro:\n",
    "            def __init__(self, path, book, module, name, *args):\n",
    "                self._path = path  # path containing workbook\n",
    "                self._book = book  # workbook name like Book1.xlsm\n",
    "                self._module = module  # module name, e.g., Module1\n",
    "                self._name = name  # procedure or function name\n",
    "                self._params = args  # argument(s)\n",
    "                self._wb = None\n",
    "            @property\n",
    "            def workbook(self):\n",
    "                return self._wb\n",
    "            @property\n",
    "            def wb_path(self):\n",
    "                return os.path.join(self._path, self._book)\n",
    "            @property\n",
    "            def name(self):\n",
    "                return f'{self._book}!{self._module}.{self._name}'\n",
    "            @property\n",
    "            def params(self):\n",
    "                return self._params\n",
    "            def get_workbook(self):\n",
    "                wb_name = os.path.basename(self.wb_path)\n",
    "                try:\n",
    "                    xl = client.GetActiveObject(EXCEL_CLS_NAME)\n",
    "                except:\n",
    "                    # Excel is not running, so we need to handle it.\n",
    "                    xl = client.Dispatch(EXCEL_CLS_NAME)\n",
    "                if wb_name in [wb.Name for wb in xl.Workbooks]:\n",
    "                    return xl.Workbooks[wb_name]\n",
    "                else:\n",
    "                    return xl.Workbooks.Open(self.wb_path)\n",
    "            def Run(self, *args, **kwargs):\n",
    "                \"\"\" \n",
    "                Runs an Excel Macro or evaluates a UDF \n",
    "                returns None for macro, or the return value of Function\n",
    "                NB: there is no error-handling here, but there should be!\n",
    "                \"\"\"\n",
    "                keep_open = kwargs.get('keep_open', True)\n",
    "                save_changes = kwargs.get('save_changes', False)\n",
    "                self._wb = self.get_workbook()\n",
    "                xl_app = self._wb.Application\n",
    "                xl_app.Visible = True\n",
    "                ret = None\n",
    "                if args is None:\n",
    "                    ret = xl_app.Run(self.name)\n",
    "                elif not args:\n",
    "                    # run with some default parameters\n",
    "                    ret = xl_app.Run(self.name, *self.params)\n",
    "                else:\n",
    "                    ret = xl_app.Run(self.name, *args)\n",
    "                if not keep_open:\n",
    "                    self.workbook.Close(save_changes)\n",
    "                    self._wb = None\n",
    "                    xl_app.Quit()\n",
    "                return ret\n",
    "\n",
    "## THIS IS NO LONGER USER FRIENDLY STUFF NO TOUCHING!!!!!\n",
    "        self.Errorcounter = 0\n",
    "        \n",
    "        self.NumOfPortpossibilities = 0\n",
    "        self.DoesStreamExistnames = []\n",
    "        self.DoesStreamExist = []     #No Stream currently exists\n",
    "        self.IsStreamconnected = []\n",
    "        self.PossiblePortsforNextdestillation = []\n",
    "        self.Molflow = np.zeros( (2*Nmax, len(COMPOUNDSinlighttoheavystring)), float)   #define the Table with the flows\n",
    "        self.Blockproperties = np.zeros( (6, Nmax), float)   #define the Table with the Blockproperites\n",
    "        self.BlockpropertiesError = np.zeros( ( Nmax), float)\n",
    "        self.DestillationError = np.zeros( ( Nmax), float)           \n",
    "\n",
    "# cleaning the Memory of which previous actions where taken\n",
    "        self.Feedchoices = [ ]\n",
    "        self.FeedchoicesIndex = []\n",
    "        self.Lightkeychoices = []\n",
    "        self.LightkeychoicesIndex = []\n",
    "        self.RewardmemoryV1 = []\n",
    "        self.RewardmemoryV2 = []\n",
    "        self.countingWrongChoices = 0\n",
    "        self.countingRightChoices = 0\n",
    "        self.RememberactionFeed = []\n",
    "        self.RememberactionLightkey = []\n",
    "        self.RememberRewardV1 = []\n",
    "        self.RememberRewardV2 = []\n",
    "\n",
    "        # self.state_for_maximum_rewards = np.zeros( ( self.Nmax), float)\n",
    "        self.CurrentMaxrewardfromMe  = -1000000\n",
    "        self.GlobalcounterofError = 0\n",
    "   #cleaning the entire sheet\n",
    "        wb = xw.Book(r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\test1.xlsm') \n",
    "        #xREMOVEALLFROMASPEN = wb.macro('REMOVEALLFROMASPEN')\n",
    "        #xREMOVEALLFROMASPEN()\n",
    "    #Make action space\n",
    "        self.action_space = spaces.Discrete((len(self.COMPOUNDSinlighttoheavystring)-1)*self.Nmax*2)\n",
    "        self.observation_space = spaces.Box(low=0, high=self.TotalFeedFlow, shape=(len(self.Molflow.flatten()),))   #np.shape((self.Molflow)))                           \n",
    "        self.actions_for_maximum_rewards = []\n",
    "        #self.state_for_maximum_rewards = []\n",
    "        self.OperationalcostsForCurrentSequence = 0\n",
    "        self.molarflowofeachcompoundIncurrentblock = []\n",
    "        self.totalmolarflowinCurrentBlock = []\n",
    "        #self.action = 0\n",
    "        #self.state = self.Molflow\n",
    "        self.Outputreruncounter = 0\n",
    "        self.OutputreruncounterGlobal = 0\n",
    "        self.TotalflowerrorcounterGlobal = 0\n",
    "        self.NumberofBlockerrors = []\n",
    "## Making list of possible connections and initalize it\n",
    "\n",
    "#DoesStreamExistname = [B1T, B1B, B2T, B2B, B3T, B3B, B4T, B4B]\n",
    "#DoesStreamExist = [0 0 0 0 0 0 0 0 ]\n",
    "#IsStreamconnected = [0 0 0 0 0 0 0 0 ]\n",
    "                                            \n",
    "        for q in range(1,(2*self.Nmax+1)):    # Make string which has name of streams and Boolian wether it exists\n",
    "            if q & 1 == 1:\n",
    "                TorB = \"T\"\n",
    "            else:\n",
    "                TorB = \"B\"\n",
    "            Bcounternumber = round(q/2 + 0.001)  #Generate the string names of the different Streams\n",
    "            string = 'B'+ str(Bcounternumber) + str(TorB)  \n",
    "            self.DoesStreamExistnames.append(string)\n",
    "            self.DoesStreamExist.append(0)     #No Stream currently exists\n",
    "            self.IsStreamconnected.append(0)   #No connection has been made yet (only counting  if input amd Output are connected)    \n",
    "            self.PossiblePortsforNextdestillation.append(0)  # IF 1 then it can be used\n",
    "            \n",
    "\n",
    "        \n",
    "    def findchoices(self):\n",
    "## 3.1 CHOICE OF NEXT FEED\n",
    "    # 3.1.1  Find the Possibilities\n",
    "        import numpy as np\n",
    "        \n",
    "        if self.Bcounter > 0:    # If Bcounter = 1 then it is not necessary\n",
    "            self.NumOfPortpossibilities = 0    #each time a new block gets added reset and recalculate the possibilites\n",
    "            for i in range(len(self.DoesStreamExist)):\n",
    "                if self.DoesStreamExist[i] == 1 and self.IsStreamconnected[i] == 0:   #figure out what Ports streams can be connected to  \n",
    "                    self.PossiblePortsforNextdestillation[i] = 1\n",
    "                    self.NumOfPortpossibilities = self.NumOfPortpossibilities +1    # total number of available streams\n",
    "                else:\n",
    "                    self.PossiblePortsforNextdestillation[i] = 0    # Mark stream as not usable  \n",
    "        self.actionstringForFeed = np.ones( (self.NumOfPortpossibilities), float)\n",
    "        self.actionstringForLightkey = np.ones( (len(self.COMPOUNDSinlighttoheavystring)-1), float)\n",
    "        print(self.PossiblePortsforNextdestillation)\n",
    "\n",
    "\n",
    "\n",
    "    def Randomechoice(self):\n",
    "        # magic AI Stuff happens now.... muhuhuuuuhhu, booom, chakalaka\n",
    "        import random\n",
    "    # 4.1 Choiceofnextfeed    \n",
    "        if self.Bcounter == 1:\n",
    "            self.Nextfeedname = \"Initial Feed\"\n",
    "            self.choiceofstreamnumber = None\n",
    "        else:\n",
    "            self.choiceofstreamnumber = random.randint(0, 2*self.Nmax-1) #number of stream which was chosen\n",
    "            while self.PossiblePortsforNextdestillation[self.choiceofstreamnumber] == 0:   # will not run as long choice is not possible\n",
    "                self.choiceofstreamnumber = random.randint(0, 2*Nmax-1)\n",
    "            self.Nextfeedname = self.DoesStreamExistnames[self.choiceofstreamnumber]\n",
    "    \n",
    "    # 4.2 Choice of lightkey\n",
    "        self.lightnumber = random.randint(0,len(self.COMPOUNDSinlighttoheavystring)-2)\n",
    "        self.Lightkey_component = self.COMPOUNDSinlighttoheavystring[self.lightnumber]\n",
    "        self.Heavykey_component = self.COMPOUNDSinlighttoheavystring[self.lightnumber + 1]\n",
    "    \n",
    "\n",
    "    #4.3. Choice to go HOme\n",
    "# just go home and stop doing stuff. \n",
    "\n",
    "        #done = True\n",
    "\n",
    "    def InputChoice(self):\n",
    "\n",
    "        #action goes from zero to 2Nmax*(numberofcompounds -1) - 1\n",
    "        if self.Bcounter == 1:\n",
    "            self.Nextfeedname = \"Initial Feed\"\n",
    "            self.choiceofstreamnumber = None\n",
    "            self.actionValidBecauseStreamAvailable = True\n",
    "        else:  #for all other cases except Bcounter = 1\n",
    "            Streamchoice = self.action % (2*Nmax)   #the first one is a zero\n",
    "            self.Nextfeedname = self.DoesStreamExistnames[Streamchoice]       #In this case this stream is the next \n",
    "            self.choiceofstreamnumber = Streamchoice\n",
    "    ## NOW I NEED TO CHECK IF THE ACTION IS VALID....\n",
    "            if self.PossiblePortsforNextdestillation[Streamchoice] == 1:\n",
    "                self.actionValidBecauseStreamAvailable = True\n",
    "            else:\n",
    "                self.actionValidBecauseStreamAvailable = False   \n",
    "\n",
    "        lightkeychoice = self.action // (2*self.Nmax)   #first one is a zero\n",
    "        self.Lightkey_component = self.COMPOUNDSinlighttoheavystring[lightkeychoice]\n",
    "        self.Heavykey_component = self.COMPOUNDSinlighttoheavystring[lightkeychoice + 1]\n",
    "\n",
    "\n",
    "       \n",
    "        \n",
    "        \n",
    "    def makeProcessandRun(self):\n",
    "        import xlwings as xw\n",
    "        wb = xw.Book(r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\test1.xlsm') \n",
    "        ## Create Destillation column in Aspen\n",
    "        add_DSTWU = wb.macro('add_DSTWU')\n",
    "        add_DSTWU(self.Bcounter) # B_ without any data\n",
    "\n",
    "        if self.Bcounter >1:   #the first one cant be seen as \"taken\" since its not in the index list \n",
    "            self.IsStreamconnected[self.choiceofstreamnumber] = 1      # Take Choice and note it as \"taken\"\n",
    "\n",
    "# Create STREAM B_T, B_B and also if Bcounter=1 then Feed(creat+connect) but thats coded in VBAexcel \n",
    "        add_streams = wb.macro('add_streams')\n",
    "        add_streams(self.Bcounter, self.Feed_temp_init, self.Feed_pressure_init, self.Feed_flow_Compound1_init, self.Feed_flow_Compound2_init, self.Feed_flow_Compound3_init, self.Feed_flow_Compound4_init, self.Feed_flow_Compound5_init, self.Feed_flow_Compound6_init, self.COMPOUNDNAME1, self.COMPOUNDNAME2, self.COMPOUNDNAME3, self.COMPOUNDNAME4, self.COMPOUNDNAME5, self.COMPOUNDNAME6)\n",
    "\n",
    "# Note the Streams existance\n",
    "        self.DoesStreamExist[self.Bcounter*2 -2] = 1   #saves the creation of B_T\n",
    "        self.DoesStreamExist[self.Bcounter*2-1] = 1   #saves the creation of B_B\n",
    "        \n",
    "# Connect STREAM B_T, B_B     \n",
    "        connect_streams = wb.macro('connect_streams')\n",
    "        connect_streams(self.Bcounter, self.Nextfeedname)\n",
    "\n",
    "# Destill propertiesd           \n",
    "        Destillationinput_RUN = wb.macro('Destillationinput_RUN')\n",
    "        Destillationinput_RUN(self.Bcounter, self.Refluxratio, self.Pressure_condens, self.Pressure_reboiler, self.Lightkey_component, self.Heavykey_component, self.Lightkey_recovery, self.Heavykey_recovery, self.OPT_NTRR)\n",
    "        #CheckforErrorInput = XlMacro(path , book, module, 'CheckforErrorInput')\n",
    "        #self.CheckforErrorInput =  CheckforErrorInput.Run()\n",
    "\n",
    "        \n",
    "     \n",
    "        \n",
    "    def GetOutputs(self):\n",
    "    # Destillation properties\n",
    "        listoffunctions = ['GetDistil_Temp', 'GetBottom_Temp', 'GetREB_Duty', 'GetCond_Duty', 'GetAct_Stages', 'GetAct_Reflux']  \n",
    "#GO through all Block propertie outputs\n",
    "        for i in range(0, len(listoffunctions)):    \n",
    "            functionqq = listoffunctions[i]\n",
    "            f = XlMacro(path , book, module, functionqq )\n",
    "            Blockdata = f.Run(self.Bcounter)\n",
    "            #print(f'The {listoffunctions[i]} is {Blockdata} for B{self.Bcounter}')\n",
    "            \n",
    "            if Blockdata == 123456789:   #This means that the extraction is invalid...\n",
    "                Blockdata = 0\n",
    "                print(f'FUCK FUCK FUCK FUCK... the extraction of the {functionqq} in B{self.Bcounter} didnt work.. I had to set it to zero')\n",
    "            #If there doesnt seem to be a problem save it in the matrix\n",
    "            self.Blockproperties[self.Bcounter-1][i] = Blockdata    # place data into a table\n",
    "    #now check for errors\n",
    "            #if self.CheckforErrorInput == None:\n",
    "                  #Check if the Input worked and remember if it didnt + where \n",
    "                #self.Errorcounter = self.Errorcounter +1\n",
    "                #self.BlockpropertiesError[self.Bcounter] = 1\n",
    "                #print(f'The Blockproperties have a error: {self.BlockpropertiesError}')\n",
    "        \n",
    "        #Get Stream properties\n",
    "        Compoundcounter = 0\n",
    "        StreamcounterT = 2*self.Bcounter - 2   #Stream index finding with the Bcounter known\n",
    "        StreamcounterB = 2*self.Bcounter - 1\n",
    "        StreamnamesT = f'B{self.Bcounter}T'   #And its string version\n",
    "        StreamnamesB = f'B{self.Bcounter}B'\n",
    "\n",
    "# extract data and put it in nice Matrix\n",
    "        self.molarflowofeachcompoundIncurrentblock = []\n",
    "        for Compoundnames in self.COMPOUNDSinlighttoheavystring:\n",
    "            f = XlMacro(path , book, module, 'GetMolflow' )\n",
    "            self.Molflow[StreamcounterT][Compoundcounter] = f.Run(StreamnamesT, Compoundnames)\n",
    "            self.Molflow[StreamcounterB][Compoundcounter] = f.Run(StreamnamesB, Compoundnames)\n",
    "            if self.Molflow[StreamcounterT][Compoundcounter] == 123 or self.Molflow[StreamcounterB][Compoundcounter] == 123:\n",
    "                print(' \\n \\n something crashed. MOLFLOW DIDNT WORK \\n \\n ')\n",
    "                print(f'before killing it we will now test if the molflow extraction code works... the current values are: {self.Molflow[StreamcounterT][Compoundcounter]} and {self.Molflow[StreamcounterB][Compoundcounter]} if its 123 this means VBA gave a error')\n",
    "                print(f'the Streamname and Compoundname is: {StreamnamesT} {StreamnamesB} for {Compoundnames}')\n",
    "                \n",
    "                testiffileisavailable = 123\n",
    "                while testiffileisavailable == 123:\n",
    "                    #time.sleep(10)\n",
    "                    RunEngine = XlMacro(path , book, module, 'RunEngine')\n",
    "                    RunEngine = RunEngine.Run()\n",
    "                    print('run engine and see if you can get the molflow to get molflow')\n",
    "                    testiffileisavailable =XlMacro(path , book, module, 'GetMolflow')\n",
    "                    testiffileisavailable = testiffileisavailable.Run('FEED', 'ETHAN-01')\n",
    "                    print(f'test gave back {testiffileisavailable} this should be 3 and NOT 123 ')\n",
    "                    #time.sleep(10)\n",
    "                    self.Outputreruncounter =self.Outputreruncounter+1\n",
    "                    if self.Outputreruncounter >= 4:\n",
    "                        print(f'OK now we are seriously fucked.... I will now 1)Kill Aspen 2)sleep for 2min 3)Reopen Aspen 4)RunEngine 5) test Molflow 6) If failed:Set to zero')\n",
    "                        #kill all the aspen programs\n",
    "                        dead = os.system(\"TASKKILL /F /IM Aspenplus.exe\")\n",
    "                        print(f'killed aspen: dead = {dead}')\n",
    "                        time.sleep(120)\n",
    "                        #reopen aspen\n",
    "                        print('not opening aspen to see if that would change something')\n",
    "                        #revive = os.startfile(r\"C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\Aspenfresh2.bkp\")\n",
    "                        \n",
    "                        RunEngine = XlMacro(path , book, module, 'RunEngine')\n",
    "                        RunEngine = RunEngine.Run()\n",
    "                        time.sleep(120)\n",
    "                        print('engine ran')\n",
    "                        testiffileisavailable =XlMacro(path , book, module, 'GetMolflow')\n",
    "                        testiffileisavailable = testiffileisavailable.Run('FEED', 'ETHAN-01')\n",
    "                        f = XlMacro(path , book, module, 'GetMolflow' )\n",
    "                        self.Molflow[StreamcounterT][Compoundcounter] = f.Run(StreamnamesT, Compoundnames)\n",
    "                        self.Molflow[StreamcounterB][Compoundcounter] = f.Run(StreamnamesB, Compoundnames)\n",
    "                        print(f'test gave back {testiffileisavailable} and the actual MOLFLOW values are{self.Molflow[StreamcounterT][Compoundcounter]} and {self.Molflow[StreamcounterB][Compoundcounter]}')\n",
    "                        if self.Molflow[StreamcounterT][Compoundcounter] ==123 or self.Molflow[StreamcounterB][Compoundcounter] == 123:\n",
    "                            print('\\n \\n \\n \\n \\n \\n \\n \\n OK MAJOR MAJOR MAJOR FUCK UP!!!!!! WE KILLED IT, DID THE JESUS MOVE.... STILL BROKEN... \\n \\n \\n \\n \\n \\n \\n \\n ')\n",
    "                            print('I will now just set the value to 0.000001')\n",
    "                            self.Molflow[StreamcounterT][Compoundcounter] =0.000001\n",
    "                            self.Molflow[StreamcounterB][Compoundcounter]= 0.000001\n",
    "                        self.Outputreruncounter =0\n",
    "                print(f'The Aspen is open now. this is the Molflow in Feed: {testiffileisavailable}')\n",
    "                #rerun the simulation\n",
    "                \n",
    "\n",
    "                #Get the molflow back which previously didnt work\n",
    "                f = XlMacro(path , book, module, 'GetMolflow' )\n",
    "                self.Molflow[StreamcounterT][Compoundcounter] = f.Run(StreamnamesT, Compoundnames)\n",
    "                self.Molflow[StreamcounterB][Compoundcounter] = f.Run(StreamnamesB, Compoundnames)\n",
    "                print(f'after trying to fix it the new values are: {self.Molflow[StreamcounterT][Compoundcounter]} and {self.Molflow[StreamcounterB][Compoundcounter]}')\n",
    "\n",
    "                #os.startfile(r\"C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\Aspenfresh2.bkp\")\n",
    "                                #There is a stupid error... lets just rerun the input and the engine and see what happens\n",
    "                #Destillationinput_RUN = wb.macro('Destillationinput_RUN')\n",
    "                #Destillationinput_RUN(self.Bcounter, self.Refluxratio, self.Pressure_condens, self.Pressure_reboiler, self.Lightkey_component, self.Heavykey_component, self.Lightkey_recovery, self.Heavykey_recovery, self.OPT_NTRR)\n",
    "                #Now lets rerun it... it might have changed it\n",
    "                #f = XlMacro(path , book, module, 'GetMolflow' )\n",
    "                #self.Molflow[StreamcounterT][Compoundcounter] = f.Run(StreamnamesT, Compoundnames)\n",
    "                #self.Molflow[StreamcounterB][Compoundcounter] = f.Run(StreamnamesB, Compoundnames)\n",
    "            blablabla = self.Molflow[StreamcounterT][Compoundcounter]\n",
    "            blublublu = self.Molflow[StreamcounterB][Compoundcounter]\n",
    "            Compoundcounter +=1\n",
    "            self.molarflowofeachcompoundIncurrentblock.append(blablabla) # makes a list of the molflow of each compound in the current block and lists them\n",
    "            self.molarflowofeachcompoundIncurrentblock.append(blublublu)\n",
    "        self.totalmolarflowinCurrentBlock.append(sum(self.molarflowofeachcompoundIncurrentblock))\n",
    "        print(f'This is the sum molarflow of the different blocks {sum(self.molarflowofeachcompoundIncurrentblock)}')\n",
    "\n",
    "\n",
    "        \n",
    "    def CheckIfValidaction(self):\n",
    "#Errorchecker\n",
    "#Error Destillation collumn input error\n",
    "        CheckforErrorDestill = XlMacro(path , book, module, 'CheckforErrorDestill')\n",
    "        self.Destillerrorvalue = CheckforErrorDestill.Run(self.Bcounter)\n",
    "        if self.Destillerrorvalue == 0:\n",
    "            self.Errorcounter = self.Errorcounter + 0    #No error\n",
    "        elif self.Destillerrorvalue == \"None\": \n",
    "            self.Errorcounter = self.Errorcounter +1    #counting errors in the code\n",
    "        elif self.Destillerrorvalue == 1: \n",
    "            self.Errorcounter = self.Errorcounter +1    #counting errors in the code\n",
    "            self.DestillationError[self.Bcounter-1] = 1\n",
    "\n",
    "#Check how many errors happend in this episode\n",
    "        if self.Errorcounter == 0:\n",
    "            #print('     Episode was sucessful without error')\n",
    "            self.actionValidity = True\n",
    "        else:\n",
    "            #print('     Episode was ended with ' + str(self.Errorcounter)  + ' Errors')\n",
    "            self.actionValidity = False\n",
    "            \n",
    "        #print('')\n",
    "\n",
    "        \n",
    "        \n",
    "    def GetReward(self):\n",
    "        #initalization\n",
    "        self.StreamvalueV1 = np.zeros( (2*self.Nmax), float)\n",
    "        self.StreamvalueV2 = np.zeros( (2*self.Nmax), float)\n",
    "        self.Molfrac = np.zeros( (2*self.Nmax, len(self.COMPOUNDSinlighttoheavystring)), float)\n",
    "        self.ValuePerFlowV1 = np.zeros( (2*self.Nmax, len(self.COMPOUNDSinlighttoheavystring)), float)\n",
    "        self.ValuePerFlowV2 = np.zeros( (2*self.Nmax, len(self.COMPOUNDSinlighttoheavystring)), float)\n",
    "#Need to be made zero after each run...\n",
    "        self.Punishment =0\n",
    "        self.MonetaryGainV1 = 0\n",
    "        self.MonetaryGainV2 = 0\n",
    "        self.MonetaryLoss = 0\n",
    "        self.TOTALREWARDVersion1 = 0\n",
    "        self.TOTALREWARDVersion2 = 0\n",
    "        self.CapitalCostNeeded = 0\n",
    "        self.TotalOperationCost = 0\n",
    "        self.Ctot = 0\n",
    "\n",
    "\n",
    "## MONEY GAINED VIA SEPARATION\n",
    "\n",
    "        #VERSION 1: ValuePerFlow = ValueIfPure * MolFrac   #If you have 99% then you get 99% of value... so 1kg of 50/50 A,B is worth: 0.5*0.5*AValuePerKg + 0.5*0.5*BValuePerKg\n",
    "#TotalValueOfStream = AmountofA*FractionofA*ValueofPureA\n",
    "        for Streamnumber in range(0,self.Nmax*2):\n",
    "            for Compoundnumber in range(0,len(self.COMPOUNDSinlighttoheavystring)):\n",
    "                if (sum(self.Molflow[:][Streamnumber])) == 0:\n",
    "                    self.Molfrac[Streamnumber][Compoundnumber] = 0\n",
    "                else:\n",
    "                    self.Molfrac[Streamnumber][Compoundnumber] = ((self.Molflow[Streamnumber][Compoundnumber]))/(sum(self.Molflow[:][Streamnumber]))\n",
    "                \n",
    "                self.ValuePerFlowV1[Streamnumber][Compoundnumber] = self.ValueIfPure[Compoundnumber] * self.Molfrac[Streamnumber][Compoundnumber]\n",
    "                self.StreamvalueV1[Streamnumber] = self.StreamvalueV1[Streamnumber] + self.Molflow[Streamnumber][Compoundnumber] * self.ValuePerFlowV1[Streamnumber][Compoundnumber]\n",
    "        #print(f'VERSION 1: Molfrac is: \\n {self.Molfrac}')            \n",
    "        #print(f'VERSION 1: ValuePerFlow aka Value concentration = {self.ValuePerFlowV1}')\n",
    "        #print(f'VERSION 1: Streamvalue is {self.StreamvalueV1}')\n",
    "\n",
    "    \n",
    "        #VERSION 2: if MolFrac < WantedPurity: ValuePerFlow = ValueIfPure\n",
    "        for Streamnumber in range(0,Nmax*2):\n",
    "            for Compoundnumber in range(0,len(COMPOUNDSinlighttoheavystring)):\n",
    "                if (sum(self.Molflow[:][Streamnumber])) == 0:\n",
    "                    self.Molfrac[Streamnumber][Compoundnumber]= 0\n",
    "                else: \n",
    "                    self.Molfrac[Streamnumber][Compoundnumber] = ((self.Molflow[Streamnumber][Compoundnumber]))/(sum(self.Molflow[:][Streamnumber]))\n",
    "                \n",
    "                if self.Molfrac[Streamnumber][Compoundnumber]>= self.ConcentrationNeededToGetPaidForStream:\n",
    "                    self.ValuePerFlowV2[Streamnumber][Compoundnumber] = self.ValueIfPure[Compoundnumber]\n",
    "                else:\n",
    "                    self.ValuePerFlowV2[Streamnumber][Compoundnumber] = 0\n",
    "                self.StreamvalueV2[Streamnumber] = self.StreamvalueV2[Streamnumber] + self.Molflow[Streamnumber][Compoundnumber] * self.ValuePerFlowV2[Streamnumber][Compoundnumber]\n",
    "        #print(f'VERSION 2: Molfrac is: \\n {self.Molfrac}')            \n",
    "        #print(f'VERSION 2: ValuePerFlow aka Value concentration = {self.ValuePerFlowV2}')\n",
    "        #print(f'VERSION 2: Streamvalue is {self.StreamvalueV2}')        \n",
    "        \n",
    "#Find the ones which are outlets and get the value for those aka MonetaryGain = sum(Streamvalue which are also a OUTPUT)  \n",
    "        for Streamcounter in range(0,self.Nmax*2):\n",
    "            if self.PossiblePortsforNextdestillation[Streamcounter] == 1: \n",
    "                self.MonetaryGainV1 = self.MonetaryGainV1 + self.StreamvalueV1[Streamcounter]\n",
    "\n",
    "        for Streamcounter in range(0,self.Nmax*2):\n",
    "            if self.PossiblePortsforNextdestillation[Streamcounter] == 1:    \n",
    "                self.MonetaryGainV2 = self.MonetaryGainV2 + self.StreamvalueV2[Streamcounter]      #if stream is currently a outlet then add value\n",
    "            else:\n",
    "                self.MonetaryGainV2 = self.MonetaryGainV2 + 0\n",
    "        \n",
    "     \n",
    "\n",
    "        self.MonetaryLoss = self.FeedValue * self.TotalFeedFlow\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "        \n",
    "        self.TotalNumStages = 0\n",
    "        for Bcounterer in range(0, Nmax):\n",
    "            #Find Number of Stages, Refluxratio and Bottom temp in each one Block\n",
    "            NumberOfActualStages = self.Blockproperties[Bcounterer][4]    # place data into a table\n",
    "            self.TotalNumStages = self.TotalNumStages + NumberOfActualStages\n",
    "            Reluxratio = self.Blockproperties[Bcounterer][5]\n",
    "            Bottom_Temp =  self.Blockproperties[Bcounterer][1]\n",
    "            \n",
    "            #If there is a block error and the value is bad change it to these generic values\n",
    "            if np.isinf(Reluxratio) or np.isnan(Reluxratio):\n",
    "                Reluxratio = 4\n",
    "                self.Errorcounter =self.Errorcounter+ 1\n",
    "                #print(f'Refluxratio had to be set to 4 since there was no data found')\n",
    "                self.BlockpropertiesError[self.Bcounter] = 1\n",
    "            if np.isinf(Bottom_Temp) or np.isnan(Bottom_Temp):\n",
    "                Bottom_Temp = 50\n",
    "                self.Errorcounter =self.Errorcounter+ 1\n",
    "                #print(f'The Bottom temp had to be set to 50 since there was no data found')\n",
    "                self.BlockpropertiesError[self.Bcounter] = 1\n",
    "            if np.isinf(NumberOfActualStages) or np.isnan(NumberOfActualStages):\n",
    "                NumberOfActualStages = PunishmentStageNumber\n",
    "                #print(f'The punishemnt stage number had to be used since there was no data about it')\n",
    "                self.BlockpropertiesError[self.Bcounter] = 1\n",
    "                self.Errorcounter =self.Errorcounter+ 1\n",
    "\n",
    "           #Calculate the Operational costs, capital costs and sizes\n",
    "            if (len(self.totalmolarflowinCurrentBlock)-1)>=Bcounterer: #if there is such a block then do the economics + sizing\n",
    "                totflow = self.totalmolarflowinCurrentBlock[Bcounterer]            \n",
    "                TAC = DCoperationcost(NumberOfActualStages, Reluxratio, totflow)\n",
    "                self.OperationalcostsForCurrentSequence = self.OperationalcostsForCurrentSequence + TAC\n",
    "                L, Di = columnsize(NumberOfActualStages, Bottom_Temp, self.Pressure_condens, totflow)    # # C bar kmol/sec\n",
    "\n",
    "                Ctot = expensiveDC(Di, L, self.Pressure_condens, Bottom_Temp, NumberOfActualStages)    # ft ft bar C \n",
    "          #Add the capitalcosts and operational costs of each block\n",
    "            self.CapitalCostNeeded  =  self.CapitalCostNeeded + Ctot\n",
    "            self.TotalOperationCost = TAC + self.TotalOperationCost\n",
    "            \n",
    "## PUNISHMENTS:\n",
    "\n",
    "# POSSIBILITES:\n",
    "    #Punish for making error\n",
    "        #Bad action choice\n",
    "            #1 chose the heaviest as the light key\n",
    "            #2 try to chose a stream which was already chosen?\n",
    "            #3 try to chose\n",
    "\n",
    "        #Destillation output error\n",
    "#            BlockpropertiesError[Bcounter] =1   # This is stuff like Cond duty which might be not defined..\n",
    "        #Simulation error (not enough Trays possible)\n",
    "#             DestillationError[Bcounter] = 1     # Does simulation run?\n",
    "            \n",
    "        self.Punishment = sum(self.BlockpropertiesError)*self.BlockpropertiesErrorWeight + sum(self.DestillationError)*self.DestillationErrorWeight\n",
    "\n",
    "            \n",
    "     \n",
    "\n",
    "    # to turn the Capital costs into running costs it will be assumed that it is paid of over the entire lifetime every single hour\n",
    "        self.HoursInLifetimeOfPlant = self.LifetimeInYears*8760\n",
    "        self.MoneyForBankers = self.CapitalCostNeeded/self.HoursInLifetimeOfPlant\n",
    "\n",
    "\n",
    "        #print(f'Punishment for being a bad boy is around {self.Punishment}')\n",
    "        #print( f'MONETARY LOSS FROM BUYING THE FEED = {self.MonetaryLoss}')\n",
    "        #print(f'MonetaryGain V1  = { self.MonetaryGainV1} and for V2 = {self.MonetaryGainV2}')\n",
    "\n",
    "##ccccccccc           DO NOT DOUBLE COUNT PUNISHMENTS!!!!!!!!!!!!!!!!!!!!!!\n",
    "\n",
    "\n",
    "\n",
    "### Get the overall final Reward:  (Measured in $ per hour)\n",
    "        self.TOTALREWARDVersion1 = self.MonetaryGainV1 - self.Punishment - self.MonetaryLoss - self.MoneyForBankers - self.OperationalcostsForCurrentSequence\n",
    "        self.TOTALREWARDVersion2 = self.MonetaryGainV2 - self.Punishment - self.MonetaryLoss - self.MoneyForBankers - self.OperationalcostsForCurrentSequence\n",
    "\n",
    "        \n",
    "        \n",
    "    def GimmeAllMyPrintOut(self):\n",
    "        print('\\n ***************************** MAKING AND RUNNNING PROCESS ****************************** \\n \\n')\n",
    "        #print(f'IsStreamconnected = {Env2.IsStreamconnected} ')\n",
    "        #print(f'PossiblePortsforNextdestillation = {Env2.PossiblePortsforNextdestillation} ')\n",
    "\n",
    "\n",
    "## FINDING POSSIBLE CHOICES\n",
    "        print(f'\\n ****************************** FIND ACTIONSPACE after running Bcounter = {self.Bcounter} *****************************')\n",
    "        print(f'The Possile ports from all streams are: {self.PossiblePortsforNextdestillation}')     \n",
    "        print(f'actual choices for Feed: {self.actionstringForFeed}')\n",
    "        print(f'actual choices for Lightkey: {self.actionstringForLightkey}')\n",
    "\n",
    "\n",
    "## Make Randome Choice\n",
    "        print('\\n ****************************** CHOICE WHICH WAS MADE ******************************')\n",
    "        print(f'the lightkey is {self.Lightkey_component} and the heavykey is {self.Heavykey_component}')\n",
    "        #print(f'its numerical string value is: {self.lightnumber}')\n",
    "        print(f'The Nextfeedname is: {self.Nextfeedname}')\n",
    "        print(f'its numerical string value is: {self.choiceofstreamnumber}')\n",
    "\n",
    "\n",
    "## Set the Agents choice as a choice:\n",
    "#print('\\n ****************************** SET AGENT CHOICE ******************************')\n",
    "\n",
    "# Env2.lightnumber = action1\n",
    "# Env2.choiceofstreamnumber = action2\n",
    "\n",
    "\n",
    "## makeProcessandRun\n",
    "        print('\\n ****************************** Made Process + Ran ******************************')\n",
    "        print(f'After connecting feed IsStreamconnected is: {self.IsStreamconnected}')\n",
    "        print(f' DoesStreamExist is: {self.DoesStreamExist}')\n",
    "        #print('   Blockinput Error ' + str(self.CheckforErrorInput))\n",
    "\n",
    "## Get Outputs\n",
    "        print('\\n ************************ Get Outputs ******************************')\n",
    "        print(f'The molarflowrate is: \\n {self.Molflow}')\n",
    "        print(f'The Blockproperties are: \\n {self.Blockproperties}')\n",
    "\n",
    "## CHECK for ERROR\n",
    "        print('\\n ************************ CheckIfValid ******************************')\n",
    "        print(f'The problem in Block is in: {self.BlockpropertiesError}')\n",
    "        print(f'The problem in Destillation is in: {self.DestillationError}')\n",
    "        print(f'actionValidity is = {self.actionValidity}')\n",
    "\n",
    "# GET REWARD\n",
    "        print('\\n ************************ GET REWARD ******************************')\n",
    "\n",
    "        print(f'VERSION 1: ValuePerFlow aka Value concentration = {self.ValuePerFlowV1}')\n",
    "        print(f'VERSION 1: Streamvalue is {self.StreamvalueV1}')\n",
    "        print(f'VERSION 2: ValuePerFlow aka Value concentration = {self.ValuePerFlowV2}')\n",
    "        print(f'VERSION 2: Streamvalue is {self.StreamvalueV2}')     \n",
    "        print(f'Punishment for being a bad boy is around {self.Punishment}')\n",
    "        print( f'MONETARY LOSS FROM BUYING THE FEED = {self.MonetaryLoss}')\n",
    "        print(f'MonetaryGain V1  = { self.MonetaryGainV1} and for V2 = {self.MonetaryGainV2}')\n",
    "        print(f'The total Capitalcosts are around {Ctot}')\n",
    "        print(f'the money for the bankers is: {self.MoneyForBankers} ')\n",
    "        print(f'TOTALREWARDVersion1 = {self.TOTALREWARDVersion1}')\n",
    "        print(f'TOTALREWARDVersion2 = {self.TOTALREWARDVersion2}')\n",
    "        print('\\n \\n \\n')\n",
    "        print(f'the Operational costs in $/yearis : {TAC}')  #in $/year\n",
    "        print(f'the length of the column is {L}')\n",
    "        print(f'the inner diameter is: {Di}')\n",
    "        print(f'the total capital costs are {Ctot}')  \n",
    "## REMEMBERED ACTIONS AND REWARDS\\\n",
    "\n",
    "        print(f'For V1 the rewards were = {self.RememberRewardV1}')\n",
    "        print(f'For V2 the rewards were = {self.RememberRewardV2}')\n",
    "        print(f'The choices for Feed were: {self.RememberactionFeed}')\n",
    "        print(f'The choices for Lightkey were: {self.RememberactionLightkey}')\n",
    "        print(f'The Choice of the actual action: {self.action}  ')\n",
    "        \n",
    "        \n",
    "    def reset(self):\n",
    "        self.Bcounter = 1\n",
    "        import numpy as np\n",
    "        import xlwings as xw\n",
    "        wb = xw.Book(r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\test1.xlsm')\n",
    "## THIS IS JUST MAKING IT POSSIBLE TO EXTRACT STUFF FROM VBA TO PYTHON\n",
    "        import os\n",
    "        from win32com import client\n",
    "        EXCEL_CLS_NAME = \"Excel.Application\"\n",
    "        class XlMacro:\n",
    "            def __init__(self, path, book, module, name, *args):\n",
    "                self._path = path  # path containing workbook\n",
    "                self._book = book  # workbook name like Book1.xlsm\n",
    "                self._module = module  # module name, e.g., Module1\n",
    "                self._name = name  # procedure or function name\n",
    "                self._params = args  # argument(s)\n",
    "                self._wb = None\n",
    "            @property\n",
    "            def workbook(self):\n",
    "                return self._wb\n",
    "            @property\n",
    "            def wb_path(self):\n",
    "                return os.path.join(self._path, self._book)\n",
    "            @property\n",
    "            def name(self):\n",
    "                return f'{self._book}!{self._module}.{self._name}'\n",
    "            @property\n",
    "            def params(self):\n",
    "                return self._params\n",
    "            def get_workbook(self):\n",
    "                wb_name = os.path.basename(self.wb_path)\n",
    "                try:\n",
    "                    xl = client.GetActiveObject(EXCEL_CLS_NAME)\n",
    "                except:\n",
    "                    # Excel is not running, so we need to handle it.\n",
    "                    xl = client.Dispatch(EXCEL_CLS_NAME)\n",
    "                if wb_name in [wb.Name for wb in xl.Workbooks]:\n",
    "                    return xl.Workbooks[wb_name]\n",
    "                else:\n",
    "                    return xl.Workbooks.Open(self.wb_path)\n",
    "            def Run(self, *args, **kwargs):\n",
    "                \"\"\" \n",
    "                Runs an Excel Macro or evaluates a UDF \n",
    "                returns None for macro, or the return value of Function\n",
    "                NB: there is no error-handling here, but there should be!\n",
    "                \"\"\"\n",
    "                keep_open = kwargs.get('keep_open', True)\n",
    "                save_changes = kwargs.get('save_changes', False)\n",
    "                self._wb = self.get_workbook()\n",
    "                xl_app = self._wb.Application\n",
    "                xl_app.Visible = True\n",
    "                ret = None\n",
    "                if args is None:\n",
    "                    ret = xl_app.Run(self.name)\n",
    "                elif not args:\n",
    "                    # run with some default parameters\n",
    "                    ret = xl_app.Run(self.name, *self.params)\n",
    "                else:\n",
    "                    ret = xl_app.Run(self.name, *args)\n",
    "                if not keep_open:\n",
    "                    self.workbook.Close(save_changes)\n",
    "                    self._wb = None\n",
    "                    xl_app.Quit()\n",
    "                return ret\n",
    "\n",
    "## THIS IS NO LONGER USER FRIENDLY STUFF NO TOUCHING!!!!!\n",
    "        self.Errorcounter = 0 \n",
    "        self.NumOfPortpossibilities = 0\n",
    "        self.DoesStreamExistnames = []\n",
    "        self.DoesStreamExist = []     #No Stream currently exists\n",
    "        self.IsStreamconnected = []\n",
    "        self.PossiblePortsforNextdestillation = []\n",
    "        self.Molflow = np.zeros( (2*self.Nmax, len(self.COMPOUNDSinlighttoheavystring)), float)   #define the Table with the flows\n",
    "        self.Blockproperties = np.zeros( (6, self.Nmax), float)   #define the Table with the Blockproperites\n",
    "        self.BlockpropertiesError = np.zeros( ( self.Nmax), float)\n",
    "        self.DestillationError = np.zeros( ( self.Nmax), float)           \n",
    "\n",
    "# cleaning the Memory of which previous actions where taken\n",
    "        self.Feedchoices = [ ]\n",
    "        self.FeedchoicesIndex = []\n",
    "        self.Lightkeychoices = []\n",
    "        self.LightkeychoicesIndex = []\n",
    "        self.RewardmemoryV1 = []\n",
    "        self.RewardmemoryV2 = []\n",
    "        \n",
    "        #self.actions1_for_maximum_rewards = []\n",
    "        #self.actions2_for_maximum_rewards = []\n",
    "        #self.state_for_maximum_rewards = []\n",
    "        #self.CurrentMaxrewardfromMe  = 0\n",
    "        #self.RememberactionFeed = []\n",
    "        #self.RememberactionLightkey = []\n",
    "        #self.RememberRewardV1 = []\n",
    "        #self.RememberRewardV2 = []\n",
    "        \n",
    "   #cleaning the entire sheet\n",
    "        wb = xw.Book(r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\test1.xlsm') \n",
    "        xREMOVEALLFROMASPEN = wb.macro('REMOVEALLFROMASPEN')\n",
    "        xREMOVEALLFROMASPEN()\n",
    "    #Make action space\n",
    "        self.action_space = spaces.Discrete((len(self.COMPOUNDSinlighttoheavystring)-1)*self.Nmax*2)\n",
    "        self.observation_space = spaces.Box(low=0, high=self.TotalFeedFlow, shape=(0,len(self.Molflow.flatten())))     #np.shape((self.Molflow)))                          \n",
    "        self.action = 0\n",
    "        self.reward = 0\n",
    "        self.state = np.squeeze(self.Molflow.flatten())\n",
    "        self.OperationalcostsForCurrentSequence = 0\n",
    "        self.molarflowofeachcompoundIncurrentblock = []\n",
    "        self.totalmolarflowinCurrentBlock = []\n",
    "        self.Outputreruncounter = 0\n",
    "## Making list of possible connections and initalize it\n",
    "\n",
    "#DoesStreamExistname = [B1T, B1B, B2T, B2B, B3T, B3B, B4T, B4B]\n",
    "#DoesStreamExist = [0 0 0 0 0 0 0 0 ]\n",
    "#IsStreamconnected = [0 0 0 0 0 0 0 0 ]\n",
    "                                            \n",
    "        for q in range(1,(2*self.Nmax+1)):    # Make string which has name of streams and Boolian wether it exists\n",
    "            if q & 1 == 1:\n",
    "                TorB = \"T\"\n",
    "            else:\n",
    "                TorB = \"B\"\n",
    "            Bcounternumber = round(q/2 + 0.001)  #Generate the string names of the different Streams\n",
    "            string = 'B'+ str(Bcounternumber) + str(TorB)  \n",
    "            self.DoesStreamExistnames.append(string)\n",
    "            self.DoesStreamExist.append(0)     #No Stream currently exists\n",
    "            self.IsStreamconnected.append(0)   #No connection has been made yet (only counting  if input amd Output are connected)    \n",
    "            self.PossiblePortsforNextdestillation.append(0)  # IF 1 then it can be used\n",
    "                                            \n",
    "        return self.state\n",
    "        \n",
    "        \n",
    "        \n",
    "    def rememberAction(self):\n",
    "        self.RememberactionFeed.append(self.Nextfeedname)\n",
    "        self.RememberactionLightkey.append(self.Lightkey_component)\n",
    "        self.RememberRewardV1.append(self.TOTALREWARDVersion1)\n",
    "        self.RememberRewardV2.append(self.TOTALREWARDVersion2)\n",
    "        \n",
    "    def step(self,action):\n",
    "        self.action = action\n",
    "        self.oldstate = self.state\n",
    "        done = False\n",
    "        self.InputChoice() #insert action, convert that into internal index version  + check if valid.\n",
    "        if self.actionValidBecauseStreamAvailable == False:    #this means that the action is not valid. old state is returned, agent needs to try again\n",
    "            self.state = self.oldstate\n",
    "            self.countingWrongChoices = self.countingWrongChoices+1\n",
    "            self.Bcounter = self.Bcounter + 0\n",
    "            self.reward = self.reward  - Punishment_for_invalid_action\n",
    "            info = {\"sucessful action\": \"no\"}\n",
    "        elif self.actionValidBecauseStreamAvailable == True:\n",
    "            self.makeProcessandRun()\n",
    "            self.GetOutputs()\n",
    "            self.CheckIfValidaction()\n",
    "            self.GetReward()\n",
    "            self.findchoices()\n",
    "            self.rememberAction()\n",
    "            self.countingRightChoices = self.countingRightChoices +1\n",
    "            \n",
    "            self.reward = self.TOTALREWARDVersion1\n",
    "            #self.reward = self.TOTALREWARDVersion2\n",
    "            \n",
    "            self.Bcounter = self.Bcounter + 1   #increase Bcounter\n",
    "            self.state = np.squeeze(self.Molflow.flatten())      #define new state\n",
    "            self.GlobalcounterofError  = self.GlobalcounterofError + self.Errorcounter\n",
    "            if self.Bcounter == self.Nmax:\n",
    "                done = True\n",
    "                if self.CurrentMaxrewardfromMe <= self.reward:\n",
    "                    self.CurrentMaxrewardfromMe = self.reward\n",
    "                    for i in range((-Nmax+1),0):\n",
    "                        self.actions1_for_maximum_rewards.append(self.RememberactionFeed[i])\n",
    "                        self.actions2_for_maximum_rewards.append(self.RememberactionLightkey[i])\n",
    "                    self.state_for_maximum_rewards.append(self.state)\n",
    "                    #print(self.state_for_maximum_rewards)\n",
    "                self.NumberofBlockerrors.append(sum(self.BlockpropertiesError))\n",
    "                #print(f' \\n the Global number of errors for Blockproperties is {self.NumberofBlockerrors[-1]}')\n",
    "                #print(f'then Global number fo errors in total is: {self.GlobalcounterofError} only gets reset by initalizing the class')\n",
    "                \n",
    "              \n",
    "            info = {\"sucessful action\": \"yes\"}\n",
    "            #self.GimmeAllMyPrintOut()\n",
    "            if self.reward <= self.HardMaxNegReward:     #THis means the reward is extremly highly negaitve and we will set it to the maximum\n",
    "                self.reward = self.HardMaxNegReward\n",
    "                print('Hey it seems like the reward was too negative... ')\n",
    "                \n",
    "        return self.state, self.reward, done, info\n",
    "    \n",
    "    \n",
    "    def render(self):\n",
    "        pass\n",
    "\n",
    "\n",
    "    def RestartEverythingNow(self):\n",
    "        self.restarteverything = True\n",
    "        #close aspen\n",
    "        #restart aspen\n",
    "        #load old model\n",
    "        #resetaspen aka reset\n",
    "        #Get out of the stepfunction\n",
    "        print(' \\n \\n \\n \\n \\n \\n \\n \\n \\n HAD TO RESTART EVERYTHING \\n \\n \\n \\n \\n \\n \\n \\n \\n \\n')\n",
    "    \n",
    "class StableEnv(Environment):\n",
    "    def __init__(self, *args, **kwargs):\n",
    "        super(StableEnv, self).__init__(*args, **kwargs)\n",
    "        \n",
    "    def step(self, action):\n",
    "        state, reward, done, info = super(StableEnv, self).step(action)\n",
    "        if np.isinf(reward) or np.isnan(reward):\n",
    "            reward = -1e3\n",
    "        return state, reward, done, info\n",
    "    \n",
    "    def legal_actions(self):\n",
    "        # ones for legal actions, zeros for illgal actions\n",
    "        row_width = 6\n",
    "        mask = np.zeros(self.action_space.n)\n",
    "        for i in range(0, self.Bcounter):\n",
    "            slicer_legal = (np.arange(self.action_space.n) % i) == 0\n",
    "        mask[slicer_legal] = 1\n",
    "        mask[-1] = 0 # this is pretend so we have one illegal action\n",
    "        return mask"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# The class from 11.06.2021"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "path = r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC'      #Folder where you place all our stuff\n",
    "book = 'test1.xlsm'     # Name of the excel document\n",
    "module = 'Modul1'     #VBA module in the excel dokument \n",
    "\n",
    "f = XlMacro(path , book, module, 'GetDistil_Temp')\n",
    "Blockdata = f.Run(1)\n",
    "print(Blockdata)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# NEW WAY TO GET REWARD"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 38.0000000303608\n",
      "[1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 30.87062124964762\n",
      "[1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 22.299430546310106\n",
      "[1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 3.2579745224927157\n",
      "[1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 0.16260797572315847\n",
      "Monetary gain V1 2167.084165712852\n",
      "Monetary gain V2 1047.8893087518002\n",
      "Punishment 0.0\n",
      "Mooney for Bankers 3.396003816792059\n",
      "Operational costs: 20.684738089229565\n",
      "[[2.99998492e+00 4.99963406e+00 1.19971891e+01 1.97381316e+00\n",
      "  8.55000000e+00 3.50000000e-01]\n",
      " [1.50836668e-05 3.65943004e-04 2.81092269e-03 2.61868410e-02\n",
      "  4.50000000e-01 6.65000000e+00]\n",
      " [2.99998492e+00 4.99963406e+00 1.19971891e+01 1.87512250e+00\n",
      "  4.27500000e-01 0.00000000e+00]\n",
      " [1.44939986e-16 2.44199522e-11 1.16232002e-08 9.86906580e-02\n",
      "  8.12250000e+00 3.50000000e-01]\n",
      " [2.84998567e+00 2.49981703e-01 1.57965175e-01 3.93126074e-05\n",
      "  2.66070271e-06 0.00000000e+00]\n",
      " [1.49999246e-01 4.74965235e+00 1.18392239e+01 1.87508319e+00\n",
      "  4.27497339e-01 0.00000000e+00]\n",
      " [2.84998567e+00 2.37482618e-01 7.89825877e-03 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00]\n",
      " [3.12605092e-10 1.24990851e-02 1.50066917e-01 3.93126074e-05\n",
      "  2.66070271e-06 0.00000000e+00]\n",
      " [3.12605092e-10 1.24990851e-02 1.50066917e-01 3.73469771e-05\n",
      "  1.33035135e-07 0.00000000e+00]\n",
      " [6.80846408e-32 1.52556235e-16 3.78221957e-13 1.96563037e-06\n",
      "  2.52766757e-06 0.00000000e+00]\n",
      " [0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00]\n",
      " [0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00]]\n",
      "reward Version 1 in the AspennotusingthelastBlockandonlytakinga_uselessoperation___NO_ERROR_V2.bkp: 2143.0033858068305\n",
      "reward Version 2 in the AspennotusingthelastBlockandonlytakinga_uselessoperation___NO_ERROR_V2.bkp: 1023.8085288457787\n",
      "[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 37.99999999461971\n",
      "[1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 18.461160093921016\n",
      "[1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 19.538839900120163\n",
      "[1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 8.2612829893\n",
      "[1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 11.342293260277218\n",
      "Monetary gain V1 2401.1023400714876\n",
      "Monetary gain V2 920.3127897279999\n",
      "Punishment 0.0\n",
      "Mooney for Bankers 2.701068511731854\n",
      "Operational costs: 32.737915222922965\n",
      "[[2.99957300e+00 4.93864209e+00 1.14000000e+01 1.00000000e-01\n",
      "  1.00575110e-01 4.97059197e-05]\n",
      " [4.26997400e-04 6.13579113e-02 6.00000000e-01 1.90000000e+00\n",
      "  8.89942489e+00 6.99995029e+00]\n",
      " [4.26994502e-04 6.13535874e-02 5.99862132e-01 1.87619938e+00\n",
      "  8.45445365e+00 3.49997515e-01]\n",
      " [2.89734546e-09 4.32388367e-06 1.37868138e-04 2.38006151e-02\n",
      "  4.44971245e-01 6.64995278e+00]\n",
      " [2.99957300e+00 4.69170998e+00 5.70000000e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00]\n",
      " [2.00460506e-10 2.46932104e-01 1.08300000e+01 1.00000000e-01\n",
      "  1.00575110e-01 4.97059197e-05]\n",
      " [2.84959435e+00 2.34585499e-01 1.18334313e-02 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00]\n",
      " [1.49978650e-01 4.45712449e+00 5.58166569e-01 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00]\n",
      " [4.26994502e-04 6.13535874e-02 5.99862132e-01 1.78238942e+00\n",
      "  4.22722682e-01 0.00000000e+00]\n",
      " [2.91965998e-21 9.97719280e-14 1.75117772e-10 9.38099692e-02\n",
      "  8.03173096e+00 3.49997515e-01]\n",
      " [0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00]\n",
      " [0.00000000e+00 0.00000000e+00 0.00000000e+00 0.00000000e+00\n",
      "  0.00000000e+00 0.00000000e+00]]\n",
      "reward Version 1 in the ASPENequimolarseparation.bkp: 2365.6633183368326\n",
      "reward Version 2 in the ASPENequimolarseparation.bkp: 884.873767993345\n",
      "[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-14-918993c09b34>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     81\u001b[0m         \u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mDoesStreamExist\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mBcounter\u001b[0m\u001b[1;33m*\u001b[0m\u001b[1;36m2\u001b[0m \u001b[1;33m-\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;36m1\u001b[0m   \u001b[1;31m#saves the creation of B_T\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     82\u001b[0m         \u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mDoesStreamExist\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mBcounter\u001b[0m\u001b[1;33m*\u001b[0m\u001b[1;36m2\u001b[0m\u001b[1;33m-\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m]\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;36m1\u001b[0m   \u001b[1;31m#saves the creation of B_B\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 83\u001b[1;33m         \u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mGetOutputs\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     84\u001b[0m         \u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mCheckIfValidaction\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     85\u001b[0m         \u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mGetReward\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<ipython-input-3-81d06c5c3fe4>\u001b[0m in \u001b[0;36mGetOutputs\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    626\u001b[0m             \u001b[0mfunctionqq\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mlistoffunctions\u001b[0m\u001b[1;33m[\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    627\u001b[0m             \u001b[0mf\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mXlMacro\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mpath\u001b[0m \u001b[1;33m,\u001b[0m \u001b[0mbook\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mmodule\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mfunctionqq\u001b[0m \u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 628\u001b[1;33m             \u001b[0mBlockdata\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mf\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mRun\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mBcounter\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    629\u001b[0m             \u001b[1;31m#print(f'The {listoffunctions[i]} is {Blockdata} for B{self.Bcounter}')\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    630\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m<ipython-input-3-81d06c5c3fe4>\u001b[0m in \u001b[0;36mRun\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m    321\u001b[0m                     \u001b[0mret\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mxl_app\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mRun\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mname\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m*\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mparams\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    322\u001b[0m                 \u001b[1;32melse\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 323\u001b[1;33m                     \u001b[0mret\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mxl_app\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mRun\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mname\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    324\u001b[0m                 \u001b[1;32mif\u001b[0m \u001b[1;32mnot\u001b[0m \u001b[0mkeep_open\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    325\u001b[0m                     \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mworkbook\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mClose\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0msave_changes\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32m~\\AppData\\Local\\Temp\\7\\gen_py\\3.8\\00020813-0000-0000-C000-000000000046x0x1x9.py\u001b[0m in \u001b[0;36mRun\u001b[1;34m(self, Macro, Arg1, Arg2, Arg3, Arg4, Arg5, Arg6, Arg7, Arg8, Arg9, Arg10, Arg11, Arg12, Arg13, Arg14, Arg15, Arg16, Arg17, Arg18, Arg19, Arg20, Arg21, Arg22, Arg23, Arg24, Arg25, Arg26, Arg27, Arg28, Arg29, Arg30)\u001b[0m\n\u001b[0;32m  44401\u001b[0m                         \u001b[1;33m,\u001b[0m \u001b[0mArg24\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mdefaultNamedOptArg\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg25\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mdefaultNamedOptArg\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg26\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mdefaultNamedOptArg\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg27\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mdefaultNamedOptArg\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg28\u001b[0m\u001b[1;33m=\u001b[0m\u001b[0mdefaultNamedOptArg\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m  44402\u001b[0m \t\t\t, Arg29=defaultNamedOptArg, Arg30=defaultNamedOptArg):\n\u001b[1;32m> 44403\u001b[1;33m \t\treturn self._ApplyTypes_(259, 1, (12, 0), ((12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17), (12, 17)), 'Run', None,Macro\n\u001b[0m\u001b[0;32m  44404\u001b[0m                         \u001b[1;33m,\u001b[0m \u001b[0mArg1\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg2\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg3\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg4\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg5\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m  44405\u001b[0m                         \u001b[1;33m,\u001b[0m \u001b[0mArg6\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg7\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg8\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg9\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mArg10\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\win32com\\client\\__init__.py\u001b[0m in \u001b[0;36m_ApplyTypes_\u001b[1;34m(self, dispid, wFlags, retType, argTypes, user, resultCLSID, *args)\u001b[0m\n\u001b[0;32m    465\u001b[0m         \u001b[1;32mdef\u001b[0m \u001b[0m_ApplyTypes_\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mself\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mdispid\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mwFlags\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mretType\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margTypes\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0muser\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mresultCLSID\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    466\u001b[0m \t\treturn self._get_good_object_(\n\u001b[1;32m--> 467\u001b[1;33m                         \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_oleobj_\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mInvokeTypes\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mdispid\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;36m0\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mwFlags\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mretType\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0margTypes\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;33m*\u001b[0m\u001b[0margs\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    468\u001b[0m \t\t\tuser, resultCLSID)\n\u001b[0;32m    469\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "Feed_temp_init = 20               #degree C\n",
    "Feed_pressure_init = 5            # Bar\n",
    "Feed_flow_Compound1_init = 3      # KiloMOL/Hour\n",
    "Feed_flow_Compound2_init = 5\n",
    "Feed_flow_Compound3_init = 12\n",
    "Feed_flow_Compound4_init = 2\n",
    "Feed_flow_Compound5_init = 9\n",
    "Feed_flow_Compound6_init = 7\n",
    "COMPOUNDNAME1 = \"ETHAN-01\"\n",
    "COMPOUNDNAME2 = \"PROPY-01\"\n",
    "COMPOUNDNAME3 = \"PROPA-01\"\n",
    "COMPOUNDNAME4 = \"1-BUT-01\"\n",
    "COMPOUNDNAME5 = \"N-BUT-01\"\n",
    "COMPOUNDNAME6 = \"N-PEN-01\"\n",
    "COMPOUNDSinlighttoheavystring = [\"ETHAN-01\", \"PROPY-01\", \"PROPA-01\", \"1-BUT-01\", \"N-BUT-01\", \"N-PEN-01\"]\n",
    "Nmax = 6  #Number of Reactors\n",
    "Refluxratio = -1.1                    \n",
    "Pressure_condens = 5                 \n",
    "Pressure_reboiler = 5        \n",
    "Lightkey_recovery = 0.95\n",
    "Heavykey_recovery = 1 - Lightkey_recovery\n",
    "OPT_NTRR = \"RR\"       #'Define your selection of reflux and not by the num of stages'\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "LifetimeInYears = 20         #years\n",
    "ValueCompound1 = 0.000016 *1000000       # $ per kmol/hour\n",
    "ValueCompound2 = 0.0000285*1000000\n",
    "ValueCompound3 = 0.0000454*1000000\n",
    "ValueCompound4 = 0.0000381*1000000\n",
    "ValueCompound5 = 0.0000754*1000000\n",
    "ValueCompound6 = 0.0000576*1000000\n",
    "ConcentrationNeededToGetPaidForStream = 0.94    #needed concentration to get paid\n",
    "FeedValue = 0.000001        #  $ per kmol/hour\n",
    "Bcounter = 1\n",
    "\n",
    "Date = '14_6'\n",
    "CloseAspen = True\n",
    "LoadOldModelplease = False\n",
    "Loadfilename = 'This_is_just_the_most_recent_file.h5'\n",
    "LEARNINGRATE = 0.03\n",
    "NumberOfSteps = 1000\n",
    "NumberofLoops = 100\n",
    "Version = 'AAA'       #This represents how many times there was a error and it had to be retrained\n",
    "PunishmentStageNumber = 150\n",
    "Punishment_for_invalid_action = 200\n",
    "HardMaxNegReward = -7000\n",
    "BlockpropertiesErrorWeight = 1000 #in $ perError\n",
    "DestillationErrorWeight = 1000   #in $ perError\n",
    "ConcentrationNeededToGetPaidForStream = 0.90    #needed concentration to get paid\n",
    "maximumTAC = 2030\n",
    "RewardVersion = 'V2'\n",
    "\n",
    "\"\"\"\n",
    "CHANGES TO CLASS TO MAKE IT WORK\n",
    "    - stop it from cleaning the sheet aka make \"REMOVEALLFROMASPEN\" a comment\n",
    "    - make a copy to the aspen in a different folder (just for safety)\n",
    "    - change the excel name \n",
    "    - open aspen file and run simulation\n",
    "    \n",
    "    - TRY IT\n",
    "    ...\n",
    "    ...\n",
    "    - if fix bugs by commenting variables which dont exist\n",
    "\n",
    "\"\"\"\n",
    "\n",
    "list_of_AspenFiles = ['AspennotusingthelastBlockandonlytakinga_uselessoperation___NO_ERROR_V2.bkp', 'ASPENequimolarseparation.bkp', 'ASPENlightkeyfirstseparation.bkp', 'ASPENheavykeyfirstseparation.bkp', 'ASPENlargestcomponentfirst.bkp' ]\n",
    "#if 1<2:\n",
    "stable_env = StableEnv(Bcounter, Feed_temp_init, Feed_pressure_init, Feed_flow_Compound1_init, Feed_flow_Compound2_init, Feed_flow_Compound3_init, Feed_flow_Compound4_init, Feed_flow_Compound5_init, Feed_flow_Compound6_init, COMPOUNDNAME1, COMPOUNDNAME2 , COMPOUNDNAME3, COMPOUNDNAME4, COMPOUNDNAME5, COMPOUNDNAME6, COMPOUNDSinlighttoheavystring, Nmax, Refluxratio, Pressure_condens, Pressure_reboiler, Lightkey_recovery ,Heavykey_recovery, OPT_NTRR, BlockpropertiesErrorWeight, DestillationErrorWeight, LifetimeInYears, ValueCompound1, ValueCompound2, ValueCompound3, ValueCompound4, ValueCompound5, ValueCompound6, ConcentrationNeededToGetPaidForStream, FeedValue,PunishmentStageNumber, Punishment_for_invalid_action, HardMaxNegReward) #Initialization of the Simulator \n",
    "for i in list_of_AspenFiles:\n",
    "    \n",
    "    wb = xw.Book(r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\test1.xlsm') \n",
    "    xchange_Aspen_name = wb.macro('change_Aspen_name')\n",
    "    xchange_Aspen_name(i)\n",
    "    #xchange_Aspen_name('ASPENlargestcomponentfirst.bkp')\n",
    "    \n",
    "    for stable_env.Bcounter in range(1,Nmax,1):\n",
    "        stable_env.findchoices()\n",
    "        stable_env.DoesStreamExist[stable_env.Bcounter*2 -2] = 1   #saves the creation of B_T\n",
    "        stable_env.DoesStreamExist[stable_env.Bcounter*2-1] = 1   #saves the creation of B_B\n",
    "        stable_env.GetOutputs()\n",
    "        stable_env.CheckIfValidaction()\n",
    "        stable_env.GetReward()\n",
    "        if RewardVersion == 'V2':\n",
    "            stable_env.reward = stable_env.TOTALREWARDVersion2\n",
    "        if RewardVersion == 'V1':\n",
    "            stable_env.reward = stable_env.TOTALREWARDVersion1\n",
    "        \n",
    "        stable_env.state = np.squeeze(stable_env.Molflow.flatten())      #define new state\n",
    "    print(f'Monetary gain V1 {stable_env.MonetaryGainV1}')\n",
    "    print(f'Monetary gain V2 {stable_env.MonetaryGainV2}')\n",
    "    print(f'Punishment {stable_env.Punishment}')\n",
    "    #print(f' Monetary loss {stable_env.MonetaryLoss}') \n",
    "    print(f'Mooney for Bankers {stable_env.MoneyForBankers}')\n",
    "    print(f'Operational costs: {stable_env.OperationalcostsForCurrentSequence}')\n",
    "    print(stable_env.Molflow)\n",
    "    print(f'reward Version 1 in the {i}: {stable_env.TOTALREWARDVersion1}')\n",
    "    print(f'reward Version 2 in the {i}: {stable_env.TOTALREWARDVersion2}')\n",
    "    \n",
    "    stable_env.reset()\n",
    "#print(f'the reward after completion: {stable_env.reward}')\n",
    "### ON the 01.06.2021 teh values were:\n",
    "#lightest key first= -129.21, no error\n",
    "#heavykey = -43.88, no error\n",
    "#equimolar = -74.58, no error\n",
    "#largestcompound = -74.68, no error\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#  PROGRAM to make a aspen file from actions, find reward, get 5 best action list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "##INPUT\n",
    "## found from short training\n",
    "#feed_for_max_reward_original= ['Initial Feed', 'B1T', 'B2B', 'B2T', 'B3B', 'Initial Feed', 'B1B', 'B1T', 'B2B', 'B4B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B1T', 'B3B', 'B4B', 'Initial Feed', 'B1B', 'B1T', 'B3B', 'B4B', 'Initial Feed', 'B1T', 'B2B', 'B2T', 'B1B', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B2B', 'Initial Feed', 'B1T', 'B2B', 'B1B', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B1B', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T', 'Initial Feed', 'B1T', 'B2T', 'B2B', 'B3T', 'Initial Feed', 'B1T', 'B1B', 'B3B', 'B2B', 'Initial Feed', 'B1T', 'B2T', 'B1B', 'B4B', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B3B', 'B4T', 'Initial Feed', 'B1B', 'B2B', 'B1T', 'B4B', 'Initial Feed', 'B1T', 'B2B', 'B3T', 'B4T', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B1B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B2B', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T', 'Initial Feed', 'B1B', 'B2T', 'B1T', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B1B', 'Initial Feed', 'B1B', 'B1T', 'B2B', 'B4T', 'Initial Feed', 'B1B', 'B1T', 'B3B', 'B4T', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B3T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B4B', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4B', 'Initial Feed', 'B1T', 'B1B', 'B2B', 'B4B', 'Initial Feed', 'B1T', 'B2T', 'B2B', 'B4B', 'Initial Feed', 'B1T', 'B2B', 'B2T', 'B3T', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B3T']\n",
    "#lightkey_for_max_reward_original= ['1-BUT-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', 'PROPA-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPA-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', 'ETHAN-01', 'PROPY-01', '1-BUT-01', '1-BUT-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'ETHAN-01', 'ETHAN-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'ETHAN-01', 'PROPA-01', 'ETHAN-01', 'ETHAN-01', '1-BUT-01', 'PROPA-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', 'PROPY-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', '1-BUT-01', 'PROPY-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', 'ETHAN-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'N-BUT-01', 'PROPA-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', 'ETHAN-01', 'PROPY-01', 'ETHAN-01', 'PROPA-01', '1-BUT-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'PROPY-01', 'N-BUT-01', 'PROPY-01', 'PROPY-01', 'ETHAN-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'PROPA-01', '1-BUT-01', 'PROPY-01', '1-BUT-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', 'N-BUT-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', '1-BUT-01', 'ETHAN-01', '1-BUT-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', 'PROPY-01', 'N-BUT-01', 'PROPA-01', 'N-BUT-01', 'ETHAN-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01']\n",
    "    #This yielded  the max reward of: -38.94774942448791 ['PROPY-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01'] ['Initial Feed', 'B1T', 'B2T', 'B1B', 'B4B']\n",
    "    # -42.97795394309483  ['ETHAN-01', 'PROPA-01', 'ETHAN-01', 'ETHAN-01', '1-BUT-01']  ['Initial Feed', 'B1T', 'B2T', 'B3B', 'B1B']\n",
    "    # -42.79940444831174 ['1-BUT-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01'] ['Initial Feed', 'B1T', 'B1B', 'B2B', 'B4B']\n",
    "\n",
    "#These are the ones from the 70 000 steps... oddly short..\n",
    "#feed_for_max_reward_original= ['Initial Feed', 'B1T', 'B2T', 'B1B', 'B3B', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B1T', 'B3B', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B3B', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B4T']\n",
    "#lightkey_for_max_reward_original= ['PROPY-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', '1-BUT-01']\n",
    "maxrewardwhichhasbeenfoundV1 = []\n",
    "maxrewardwhichhasbeenfoundV2 = []\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "## 13.06. 25000 steps. very nice. started to get improvement even in the average\n",
    "\n",
    "\n",
    "#feed_for_max_reward_original= ['Initial Feed', 'B1T', 'B2T', 'B3T', 'B2B', 'Initial Feed', 'B1B', 'B2B', 'B1T', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B1B', 'B3T', 'Initial Feed', 'B1B', 'B2T', 'B2B', 'B3T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4B', 'Initial Feed', 'B1B', 'B1T', 'B2B', 'B3T', 'Initial Feed', 'B1B', 'B2B', 'B2T', 'B3B', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B2B']\n",
    "#lightkey_for_max_reward_original= ['1-BUT-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01', 'PROPY-01', 'PROPA-01']\n",
    "#lightkey_for_max_reward_original= ['PROPY-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01']\n",
    "#feed_for_max_reward_original= ['Initial Feed', 'B1T', 'B2T', 'B1B', 'B3B']\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "##15.06.21 37000 steps\n",
    "\n",
    "#feed_for_max_reward_original = ['Initial Feed', 'B1T', 'B2T', 'B3T', 'B2B', 'Initial Feed', 'B1B', 'B2T', 'B2B', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B1T', 'Initial Feed', 'B1T', 'B1B', 'B2B', 'B3T', 'Initial Feed', 'B1B', 'B2B', 'B2T', 'B3T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B1B', 'Initial Feed', 'B1B', 'B1T', 'B3T', 'B3B', 'Initial Feed', 'B1T', 'B2B', 'B3T', 'B1B', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B3B', 'Initial Feed', 'B1B', 'B2T', 'B1T', 'B4T']\n",
    "#lightkey_for_max_reward_original = ['PROPY-01', 'N-BUT-01', 'PROPY-01', 'PROPA-01', '1-BUT-01', 'PROPY-01', '1-BUT-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', 'ETHAN-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', '1-BUT-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', 'PROPY-01', 'N-BUT-01', 'PROPY-01', '1-BUT-01', 'PROPA-01', 'PROPA-01', 'PROPY-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', 'N-BUT-01']\n",
    "#the reward for the following action sequence was: 2239.2996539039946the reward for the following action sequence was: 2285.5875023696717the reward for the following action sequence was: 2345.3024866350156 the reward for the following action sequence was: 2442.796002389813\n",
    "#['N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', 'PROPY-01']['N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', 'PROPA-01'] ['PROPY-01', 'N-BUT-01', 'PROPY-01', 'PROPA-01', '1-BUT-01']['PROPY-01', '1-BUT-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01']\n",
    "#['Initial Feed', 'B1B', 'B1T', 'B3T', 'B3B']['Initial Feed', 'B1T', 'B2T', 'B3T', 'B1B']['Initial Feed', 'B1T', 'B2T', 'B3T', 'B2B'] ['Initial Feed', 'B1B', 'B2T', 'B2B', 'B4T']\n",
    "# the reward for the following action sequence was: 2489.3878430869368\n",
    "#lightkey_for_max_reward_original = ['ETHAN-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01']\n",
    "#feed_for_max_reward_original = ['Initial Feed', 'B1B', 'B2T', 'B3T', 'B1T']\n",
    "\n",
    "\n",
    "\n",
    "##16.06.2021 16:13\n",
    "#43 000\n",
    "feed_for_max_reward_original = ['Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B3B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B4B', 'Initial Feed', 'B1T', 'B2B', 'B2T', 'B1B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B2B', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B2B', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B1T', 'B3B', 'B4T', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B2B', 'Initial Feed', 'B1B', 'B1T', 'B3T', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4B', 'Initial Feed', 'B1B', 'B2B', 'B3T', 'B1T']\n",
    "lightkey_for_max_reward_original = ['PROPY-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'N-BUT-01', 'ETHAN-01', '1-BUT-01', '1-BUT-01', 'PROPY-01', 'ETHAN-01', 'ETHAN-01', 'ETHAN-01', '1-BUT-01', 'ETHAN-01', '1-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'PROPY-01', 'N-BUT-01', 'N-BUT-01', 'PROPY-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', '1-BUT-01', 'N-BUT-01', '1-BUT-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', 'ETHAN-01', 'ETHAN-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01', '1-BUT-01', 'N-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', '1-BUT-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01']\n",
    "#the reward for the following action sequence was: 1988.7047702613993, 1849.2184256205637, 2110.7650220180417, 2272.52780068834, 2461.683529653148\n",
    "#the reward for the following action sequence was V1: 1783.573614295369 and 640.7685288457786,  the reward for the following action sequence was V1: 1653.3704660506442 and 971.2208653176898,  the reward for the following action sequence was V1: 1822.5932655456447 and 994.6946480000748,  the reward for the following action sequence was V1: 1716.68291566146 and 1019.104396353503,  the reward for the following action sequence was V1: 2461.6835135590313 and 1401.5352768273406\n",
    "#lightkey_for_max_reward_original =['N-BUT-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'N-BUT-01','N-BUT-01', 'PROPA-01', 'PROPY-01', '1-BUT-01', '1-BUT-01','ETHAN-01', 'ETHAN-01', '1-BUT-01', 'ETHAN-01', '1-BUT-01', 'ETHAN-01', '1-BUT-01', '1-BUT-01', 'PROPY-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01']\n",
    "#feed_for_max_reward_original = ['Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1T', 'B2B', 'B2T', 'B1B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B4B', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B3B', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T']\n",
    "#lightkey_for_max_reward_original = ['PROPY-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01']\n",
    "#feed_for_max_reward_original = ['Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T']\n",
    "#lightkey_for_max_reward_original = ['N-BUT-01', '1-BUT-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01','N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01','N-BUT-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'N-BUT-01','PROPY-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01']\n",
    "#feed_for_max_reward_original = ['Initial Feed', 'B1T', 'B2T', 'B3T', 'B4T','Initial Feed', 'B1T', 'B2T', 'B3T', 'B4T','Initial Feed', 'B1B', 'B2T', 'B2B', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T','Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T']\n",
    "\n",
    "\n",
    "\n",
    "#22.6.2021\n",
    "#20 000 steps\n",
    "\n",
    "#feed_for_max_reward_original = ['Initial Feed', 'B1B', 'B1T', 'B3T', 'B4B', 'Initial Feed', 'B1T', 'B1B', 'B2T', 'B3T', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B4B', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B3T', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B3T', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B2T', 'Initial Feed', 'B1B', 'B2T', 'B2B', 'B1T', 'Initial Feed', 'B1T', 'B2B', 'B1B', 'B4T']\n",
    "#lightkey_for_max_reward_original = ['PROPA-01', 'PROPA-01', '1-BUT-01', 'N-BUT-01', 'PROPY-01', 'PROPY-01', 'N-BUT-01', 'PROPA-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'N-BUT-01', 'ETHAN-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01', 'PROPY-01', '1-BUT-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', '1-BUT-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01']\n",
    "#NOT TRUSTWORTHY for V2 #Totalreward V1 2296.409769181432  Totalreward V2 1443.1050145971808.....'Initial Feed', 'B1T', 'B2B', 'B1B', 'B4T'....'1-BUT-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01'\n",
    "#NOT TRUSTWORTHY v2#Totalreward V1 2458.556593453575 Totalreward V2 1410.3529631626495 ......'Initial Feed', 'B1B', 'B2T', 'B2B', 'B1T' ..... 'PROPY-01', '1-BUT-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01'\n",
    "#NOT TRUSTWORTHY v2#Totalreward V1 2139.508911762888Totalreward V2 959.0744342599136 ...... 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B2T' ..... 'PROPA-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01'\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "wb = xw.Book(r'C:\\Users\\s2191105\\Desktop\\DATA from the actual PC\\test1.xlsm') \n",
    "xchange_Aspen_name = wb.macro('change_Aspen_name')\n",
    "xchange_Aspen_name('Aspenfresh8.bkp')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This is the sum molarflow of the different blocks 38.000000004277304\n",
      "[1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 29.649955790081304\n",
      "[1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 18.39617996988529\n",
      "[1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 11.276123474266859\n",
      "[1, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 8.350044208046091\n",
      "[1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0]\n",
      "\n",
      "\n",
      "\n",
      " Monetary gain V1 2542.1221152620387\n",
      "Monetary gain V2 1481.9738623120002\n",
      " Monetary loss 3.7999999999999995e-05\n",
      "Mooney for Bankers 2.9750513161098575\n",
      "Operational costs: 77.46349629278083\n",
      "Totalreward V1 2461.683529653148\n",
      "Totalreward V2 1401.5352767031097\n",
      "\n",
      "\n",
      "\n",
      "k\n",
      "15\n",
      "This is the sum molarflow of the different blocks 38.0000000303608\n",
      "[1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 30.87062124964762\n",
      "[1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 22.299430546310106\n",
      "[1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 3.2579745224927157\n",
      "[1, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 0.16260797572315847\n",
      "[1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0]\n",
      "\n",
      "\n",
      "\n",
      " Monetary gain V1 1807.6543942013905\n",
      "Monetary gain V2 664.8493087518002\n",
      " Monetary loss 3.7999999999999995e-05\n",
      "Mooney for Bankers 3.396003816792059\n",
      "Operational costs: 20.684738089229565\n",
      "Totalreward V1 1783.573614295369\n",
      "Totalreward V2 640.7685288457786\n",
      "\n",
      "\n",
      "\n",
      "k\n",
      "15\n",
      "This is the sum molarflow of the different blocks 38.00000000409579\n",
      "[1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 15.649990320237407\n",
      "[1, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 22.350009704646737\n",
      "[1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 3.258053655550141\n",
      "[1, 0, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0]\n",
      "This is the sum molarflow of the different blocks 3.0954005514599996\n",
      "[1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0]\n",
      "\n",
      " Punishment 1000.0 \n",
      "\n",
      "\n",
      "\n",
      "\n",
      " Monetary gain V1 1745.4695050891567\n",
      "Monetary gain V2 1047.8909857812002\n",
      "\n",
      " Punishment 1000.0 \n",
      "\n",
      " Monetary loss 3.7999999999999995e-05\n",
      "Mooney for Bankers 4.037005843643851\n",
      "Operational costs: 24.749545584053486\n",
      "Totalreward V1 716.6829156614592\n",
      "Totalreward V2 19.1043963535029\n",
      "\n",
      "\n",
      "\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "Interrupted by user",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-20-78d691624b8f>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m    118\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34mf'Totalreward V1 {stable_env.TOTALREWARDVersion1}'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    119\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34mf'Totalreward V2 {stable_env.TOTALREWARDVersion2}\\n\\n\\n'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 120\u001b[1;33m     \u001b[0mqqqqqqq\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0minput\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    121\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mqqqqq\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    122\u001b[0m     \u001b[1;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m5\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel\\kernelbase.py\u001b[0m in \u001b[0;36mraw_input\u001b[1;34m(self, prompt)\u001b[0m\n\u001b[0;32m    858\u001b[0m                 \u001b[1;34m\"raw_input was called, but this frontend does not support input requests.\"\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    859\u001b[0m             )\n\u001b[1;32m--> 860\u001b[1;33m         return self._input_request(str(prompt),\n\u001b[0m\u001b[0;32m    861\u001b[0m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_parent_ident\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    862\u001b[0m             \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0m_parent_header\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;32mC:\\ProgramData\\Anaconda3\\lib\\site-packages\\ipykernel\\kernelbase.py\u001b[0m in \u001b[0;36m_input_request\u001b[1;34m(self, prompt, ident, parent, password)\u001b[0m\n\u001b[0;32m    902\u001b[0m             \u001b[1;32mexcept\u001b[0m \u001b[0mKeyboardInterrupt\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    903\u001b[0m                 \u001b[1;31m# re-raise KeyboardInterrupt, to truncate traceback\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m--> 904\u001b[1;33m                 \u001b[1;32mraise\u001b[0m \u001b[0mKeyboardInterrupt\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"Interrupted by user\"\u001b[0m\u001b[1;33m)\u001b[0m \u001b[1;32mfrom\u001b[0m \u001b[1;32mNone\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m    905\u001b[0m             \u001b[1;32mexcept\u001b[0m \u001b[0mException\u001b[0m \u001b[1;32mas\u001b[0m \u001b[0me\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m    906\u001b[0m                 \u001b[0mself\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mlog\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mwarning\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34m\"Invalid Message:\"\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mexc_info\u001b[0m\u001b[1;33m=\u001b[0m\u001b[1;32mTrue\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: Interrupted by user"
     ]
    }
   ],
   "source": [
    "\n",
    "feed_for_max_reward = 0\n",
    "lightkey_for_max_reward = 0\n",
    "feed_for_max_reward = feed_for_max_reward_original\n",
    "lightkey_for_max_reward = lightkey_for_max_reward_original\n",
    "Feed_temp_init = 20               #degree C\n",
    "Feed_pressure_init = 5            # Bar\n",
    "Feed_flow_Compound1_init = 3      # KiloMOL/Hour\n",
    "Feed_flow_Compound2_init = 5\n",
    "Feed_flow_Compound3_init = 12\n",
    "Feed_flow_Compound4_init = 2\n",
    "Feed_flow_Compound5_init = 9\n",
    "Feed_flow_Compound6_init = 7\n",
    "COMPOUNDNAME1 = \"ETHAN-01\"\n",
    "COMPOUNDNAME2 = \"PROPY-01\"\n",
    "COMPOUNDNAME3 = \"PROPA-01\"\n",
    "COMPOUNDNAME4 = \"1-BUT-01\"\n",
    "COMPOUNDNAME5 = \"N-BUT-01\"\n",
    "COMPOUNDNAME6 = \"N-PEN-01\"\n",
    "COMPOUNDSinlighttoheavystring = [\"ETHAN-01\", \"PROPY-01\", \"PROPA-01\", \"1-BUT-01\", \"N-BUT-01\", \"N-PEN-01\"]\n",
    "Nmax = 6  #Number of Reactors\n",
    "Refluxratio = -1.1                    \n",
    "Pressure_condens = 5                 \n",
    "Pressure_reboiler = 5        \n",
    "Lightkey_recovery = 0.95\n",
    "Heavykey_recovery = 1 - Lightkey_recovery\n",
    "OPT_NTRR = \"RR\"       #'Define your selection of reflux and not by the num of stages'\n",
    "\n",
    "\n",
    "\n",
    "LifetimeInYears = 20         #years\n",
    "ValueCompound1 = 0.000016 *1000000       # $ per kmol/hour\n",
    "ValueCompound2 = 0.0000285*1000000\n",
    "ValueCompound3 = 0.0000454*1000000\n",
    "ValueCompound4 = 0.0000381*1000000\n",
    "ValueCompound5 = 0.0000754*1000000\n",
    "ValueCompound6 = 0.0000576*1000000\n",
    "ConcentrationNeededToGetPaidForStream = 0.90    #needed concentration to get paid\n",
    "FeedValue = 0.000001        #  $ per kmol/hour\n",
    "Bcounter = 1\n",
    "\n",
    "Date = '14_6'\n",
    "CloseAspen = True\n",
    "LoadOldModelplease = False\n",
    "Loadfilename = 'This_is_just_the_most_recent_file.h5'\n",
    "LEARNINGRATE = 0.03\n",
    "NumberOfSteps = 1000\n",
    "NumberofLoops = 100\n",
    "Version = 'AAA'       #This represents how many times there was a error and it had to be retrained\n",
    "PunishmentStageNumber = 150\n",
    "Punishment_for_invalid_action = 200\n",
    "HardMaxNegReward = -15000\n",
    "BlockpropertiesErrorWeight = 1000 #in $ perError\n",
    "DestillationErrorWeight = 1000   #in $ perError\n",
    "ConcentrationNeededToGetPaidForStream = 0.90    #needed concentration to get paid\n",
    "maximumTAC = 2030\n",
    "RewardVersion = 'V2'\n",
    "\n",
    "\n",
    "#how many different maxima are there?\n",
    "qqqqq = int(len(feed_for_max_reward)/5)\n",
    "\n",
    "for iii in range(0, qqqqq):\n",
    "    stable_env = StableEnv(Bcounter, Feed_temp_init, Feed_pressure_init, Feed_flow_Compound1_init, Feed_flow_Compound2_init, Feed_flow_Compound3_init, Feed_flow_Compound4_init, Feed_flow_Compound5_init, Feed_flow_Compound6_init, COMPOUNDNAME1, COMPOUNDNAME2 , COMPOUNDNAME3, COMPOUNDNAME4, COMPOUNDNAME5, COMPOUNDNAME6, COMPOUNDSinlighttoheavystring, Nmax, Refluxratio, Pressure_condens, Pressure_reboiler, Lightkey_recovery ,Heavykey_recovery, OPT_NTRR, BlockpropertiesErrorWeight, DestillationErrorWeight, LifetimeInYears, ValueCompound1, ValueCompound2, ValueCompound3, ValueCompound4, ValueCompound5, ValueCompound6, ConcentrationNeededToGetPaidForStream, FeedValue,PunishmentStageNumber, Punishment_for_invalid_action, HardMaxNegReward) #Initialization of the Simulator \n",
    "    stable_env.reset()\n",
    "    \n",
    "    for bliblablub in range(-5,0,1):\n",
    "        stable_env.action = 1\n",
    "        stable_env.oldstate = stable_env.state\n",
    "        done = False\n",
    "        stable_env.InputChoice() #insert action, convert that into internal index version  + check if valid.\n",
    "        stable_env.Nextfeedname = feed_for_max_reward[bliblablub]      #In this case this stream is the next \n",
    "        stable_env.Lightkey_component = lightkey_for_max_reward[bliblablub]          #stable_env.COMPOUNDSinlighttoheavystring[lightkeychoice]\n",
    "        indexlightkey = stable_env.COMPOUNDSinlighttoheavystring.index(lightkey_for_max_reward[bliblablub] )\n",
    "\n",
    "\n",
    "        stable_env.Heavykey_component = stable_env.COMPOUNDSinlighttoheavystring[indexlightkey +1]      #stable_env.COMPOUNDSinlighttoheavystring[lightkeychoice + 1]\n",
    "    #print(stable_env.actionValidBecauseStreamAvailable)\n",
    "\n",
    "        stable_env.makeProcessandRun()\n",
    "        stable_env.GetOutputs()\n",
    "        stable_env.CheckIfValidaction()\n",
    "        stable_env.GetReward()\n",
    "        stable_env.findchoices()\n",
    "        stable_env.rememberAction()\n",
    "        stable_env.countingRightChoices = stable_env.countingRightChoices +1\n",
    "        if RewardVersion == 'V1':\n",
    "            stable_env.reward = stable_env.TOTALREWARDVersion1\n",
    "        if RewardVersion == 'V2':\n",
    "            stable_env.reward = stable_env.TOTALREWARDVersion2\n",
    "            \n",
    "        stable_env.Bcounter = stable_env.Bcounter + 1   #increase Bcounter\n",
    "        stable_env.state = np.squeeze(stable_env.Molflow.flatten())      #define new state\n",
    "        stable_env.GlobalcounterofError  = stable_env.GlobalcounterofError + stable_env.Errorcounter\n",
    "        #print(f'Monetary gain V1 {stable_env.MonetaryGainV1}')\n",
    "        #print(f'Monetary gain V2 {stable_env.MonetaryGainV2}')\n",
    "        if stable_env.Punishment !=0.0: \n",
    "            print(f'\\n Punishment {stable_env.Punishment} \\n')\n",
    "        #print(f' Monetary loss {stable_env.MonetaryLoss}') \n",
    "        #print(f'Mooney for Bankers {stable_env.MoneyForBankers}')\n",
    "        #print(f'Operational costs: {stable_env.OperationalcostsForCurrentSequence}')\n",
    "        if stable_env.Bcounter == stable_env.Nmax:\n",
    "            done = True\n",
    "            maxrewardwhichhasbeenfoundV1.append(stable_env.TOTALREWARDVersion1)\n",
    "            maxrewardwhichhasbeenfoundV2.append(stable_env.TOTALREWARDVersion2)\n",
    "        info = {\"sucessful action\": \"yes\"}\n",
    "        #self.GimmeAllMyPrintOut()\n",
    "        if stable_env.reward <= stable_env.HardMaxNegReward:     #THis means the reward is extremly highly negaitve and we will set it to the maximum\n",
    "            stable_env.reward = stable_env.HardMaxNegReward\n",
    "            print('Hey it seems like the reward was too negative... ')\n",
    "    \n",
    "    print(f'\\n\\n\\n Monetary gain V1 {stable_env.MonetaryGainV1}')\n",
    "    print(f'Monetary gain V2 {stable_env.MonetaryGainV2}')\n",
    "    if stable_env.Punishment !=0.0: \n",
    "        print(f'\\n Punishment {stable_env.Punishment} \\n')\n",
    "    print(f' Monetary loss {stable_env.MonetaryLoss}') \n",
    "    print(f'Mooney for Bankers {stable_env.MoneyForBankers}')\n",
    "    print(f'Operational costs: {stable_env.OperationalcostsForCurrentSequence}')\n",
    "    print(f'Totalreward V1 {stable_env.TOTALREWARDVersion1}')\n",
    "    print(f'Totalreward V2 {stable_env.TOTALREWARDVersion2}\\n\\n\\n')\n",
    "    qqqqqqq = input()\n",
    "    print(qqqqq)\n",
    "    for i in range(0,5):\n",
    "    #now we remove these data points    \n",
    "        feed_for_max_reward.pop(-1)\n",
    "        lightkey_for_max_reward.pop(-1)\n",
    "\n",
    "x = np.arange(0, len(maxrewardwhichhasbeenfoundV1), 1)\n",
    "plt.plot(x, maxrewardwhichhasbeenfoundV1, 'red')\n",
    "plt.plot(x, maxrewardwhichhasbeenfoundV2, 'blue')\n",
    "plt.ylabel(\"reward\")\n",
    "plt.xlabel(\"steps\")\n",
    "plt.title('the maximum reward in V1 and V2')\n",
    "plt.show()\n",
    "\n",
    "# x = np.arange(0, len(maxrewardwhichhasbeenfoundV2), 1)\n",
    "# plt.plot(x, maxrewardwhichhasbeenfoundV2)\n",
    "# plt.ylabel(\"reward\")\n",
    "# plt.xlabel(\"steps\")\n",
    "# plt.title('the maximum reward in V2 with randome actions')\n",
    "# plt.show()\n",
    "    \n",
    "    \n",
    "print(stable_env.RememberactionFeed) \n",
    "print(stable_env.RememberactionLightkey)\n",
    "print(maxrewardwhichhasbeenfoundV1)\n",
    "print(maxrewardwhichhasbeenfoundV2)\n",
    "\n",
    "\n",
    "# feed_for_max_reward_original = ['Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B3B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B4B', 'Initial Feed', 'B1T', 'B2B', 'B2T', 'B1B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B2B', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B2B', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B1T', 'B3B', 'B4T', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B2B', 'Initial Feed', 'B1B', 'B1T', 'B3T', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4B', 'Initial Feed', 'B1B', 'B2B', 'B3T', 'B1T']\n",
    "# lightkey_for_max_reward_original = ['PROPY-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'N-BUT-01', 'ETHAN-01', '1-BUT-01', '1-BUT-01', 'PROPY-01', 'ETHAN-01', 'ETHAN-01', 'ETHAN-01', '1-BUT-01', 'ETHAN-01', '1-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'PROPY-01', 'N-BUT-01', 'N-BUT-01', 'PROPY-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', '1-BUT-01', 'N-BUT-01', '1-BUT-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', 'ETHAN-01', 'ETHAN-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01', '1-BUT-01', 'N-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', '1-BUT-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01']\n",
    "#lightkey_for_max_reward_original = ['1-BUT-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', 'ETHAN-01','1-BUT-01', 'N-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPA-01','N-BUT-01', 'PROPY-01', 'N-BUT-01', 'N-BUT-01', 'PROPY-01','N-BUT-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01','PROPY-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01']\n",
    "#feed_for_max_reward_original = ['Initial Feed', 'B1B', 'B1T', 'B3B', 'B4T','Initial Feed', 'B1B', 'B1T', 'B3T', 'B4T','Initial Feed', 'B1T', 'B1B', 'B3T', 'B2B','Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T','Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T']\n",
    "#lightkey_for_max_reward_original =['N-BUT-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'N-BUT-01','N-BUT-01', 'PROPA-01', 'PROPY-01', '1-BUT-01', '1-BUT-01','ETHAN-01', 'ETHAN-01', '1-BUT-01', 'ETHAN-01', '1-BUT-01', 'ETHAN-01', '1-BUT-01', '1-BUT-01', 'PROPY-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01']\n",
    "#feed_for_max_reward_original = ['Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1T', 'B2B', 'B2T', 'B1B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B4B', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B3B', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T']\n",
    "#the reward for the following action sequence was V1: 1384.0021201151153 and 300.372204836642, the reward for the following action sequence was V1: 868.8464978040901 and 395.7569987510417,  the reward for the following action sequence was V1: 1636.7652981724839 and 492.94921878003015,  the reward for the following action sequence was V1: 1783.573614295369 and 640.7685288457786,  the reward for the following action sequence was V1: 2461.683529653148 and 1401.5352767031097\n",
    "#lightkey_for_max_reward_original = ['N-BUT-01', '1-BUT-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01','N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01','N-BUT-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'N-BUT-01','PROPY-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01']\n",
    "#feed_for_max_reward_original = ['Initial Feed', 'B1T', 'B2T', 'B3T', 'B4T','Initial Feed', 'B1T', 'B2T', 'B3T', 'B4T','Initial Feed', 'B1B', 'B2T', 'B2B', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T','Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T']\n",
    "\n",
    "#lightkey_for_max_reward_original = ['PROPY-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01']\n",
    "#feed_for_max_reward_original = ['Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T']\n",
    "# print(len(lightkey_for_max_reward_original))\n",
    "# print(len(feed_for_max_reward_original))\n",
    "\n",
    "\n",
    "feed_for_max_reward_original = ['Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B3B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B4B', 'Initial Feed', 'B1T', 'B2B', 'B2T', 'B1B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B2B', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B2B', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B1T', 'B3B', 'B4T', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B2B', 'Initial Feed', 'B1B', 'B1T', 'B3T', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4B', 'Initial Feed', 'B1B', 'B2B', 'B3T', 'B1T']\n",
    "lightkey_for_max_reward_original = ['PROPY-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'N-BUT-01', 'ETHAN-01', '1-BUT-01', '1-BUT-01', 'PROPY-01', 'ETHAN-01', 'ETHAN-01', 'ETHAN-01', '1-BUT-01', 'ETHAN-01', '1-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', '1-BUT-01', '1-BUT-01', 'N-BUT-01', 'PROPY-01', 'N-BUT-01', 'N-BUT-01', 'PROPY-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPY-01', '1-BUT-01', 'N-BUT-01', '1-BUT-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', 'ETHAN-01', 'ETHAN-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01', '1-BUT-01', 'N-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', '1-BUT-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01']\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "sortedmaxrewardwhichhasbeenfoundV2 = []\n",
    "sortedmaxrewardwhichhasbeenfoundV1 = []\n",
    "for value in maxrewardwhichhasbeenfoundV1:\n",
    "    sortedmaxrewardwhichhasbeenfoundV1.append(value)\n",
    "sortedmaxrewardwhichhasbeenfoundV1.sort()\n",
    "print(f' the sorted rewards are: {sortedmaxrewardwhichhasbeenfoundV1}')\n",
    "for value in maxrewardwhichhasbeenfoundV2:\n",
    "    sortedmaxrewardwhichhasbeenfoundV2.append(value)\n",
    "sortedmaxrewardwhichhasbeenfoundV2.sort()\n",
    "print(f' the sorted rewards are: {sortedmaxrewardwhichhasbeenfoundV2}')\n",
    "\n",
    "\n",
    "\n",
    "#reward_for_which_actionsshould_be_found = max(maxrewardwhichhasbeenfoundV1)\n",
    "\n",
    "for qaqaqa in range(-5,0,1):\n",
    "    if RewardVersion == 'V1':\n",
    "        reward_for_which_actionsshould_be_found = sortedmaxrewardwhichhasbeenfoundV1[qaqaqa]\n",
    "    if RewardVersion == 'V2':\n",
    "        reward_for_which_actionsshould_be_found = sortedmaxrewardwhichhasbeenfoundV2[qaqaqa]\n",
    "    lightkeylistaction = []\n",
    "    feedlistaction = []\n",
    "    difference = []\n",
    "    for i in range(0,len(maxrewardwhichhasbeenfoundV1)):\n",
    "        difference.append((maxrewardwhichhasbeenfoundV1[i]-maxrewardwhichhasbeenfoundV2[i])**2)\n",
    "    if RewardVersion == 'V1':\n",
    "        indexreward = maxrewardwhichhasbeenfoundV1.index(reward_for_which_actionsshould_be_found)\n",
    "    if RewardVersion == 'V2':\n",
    "        indexreward = maxrewardwhichhasbeenfoundV2.index(reward_for_which_actionsshould_be_found)\n",
    "        \n",
    "    for i in range(0,5,1):\n",
    "        feedlistaction.append(feed_for_max_reward_original[(indexreward*5+i)])\n",
    "        lightkeylistaction.append(lightkey_for_max_reward_original[(indexreward*5 +i)])\n",
    "    print(f'\\n the reward for the following action sequence was V1: {maxrewardwhichhasbeenfoundV1[indexreward]} and {maxrewardwhichhasbeenfoundV2[indexreward]}')\n",
    "    print(lightkeylistaction)\n",
    "    print(feedlistaction)\n",
    "\n",
    "print('\\n\\n\\n\\n\\n\\n\\n')\n",
    "print(f'the square average difference between the two Stream value versions is: {sum(difference)/len(difference)} ')\n",
    "\n",
    "\n",
    " \n",
    "\n",
    "\n",
    "\n",
    "\"\"\"\n",
    "    - Make the aspen name \"Testforthething\" in excel\n",
    "    - open the aspen\n",
    "\n",
    "\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0. 0. 0. 0. 1. 1.]\n",
      "1000.0\n"
     ]
    }
   ],
   "source": [
    "print(stable_env.BlockpropertiesError)\n",
    "print(stable_env.Punishment)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      " the sorted rewards are: [-434.1255507271343, -434.1255507271343, -434.1255507271343, -422.99037078140543, -92.72872719322088, -92.72872719322088, -92.72872719322086, -90.38856036193533, -64.76556677098529, -64.76556677098529, -64.76556677098527, -49.75375264933868, -49.75375264933868, -49.75375264933868, -48.55800632346675, -48.55800632346675, -48.55800632346675, -42.25894220960085, -41.299551865184256, -38.18948516560627]\n",
      "\n",
      " the reward for the following action sequence was: -48.55800632346675\n",
      "['PROPY-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01']\n",
      "['Initial Feed', 'B1T', 'B2T', 'B1B', 'B3B']\n",
      "\n",
      " the reward for the following action sequence was: -48.55800632346675\n",
      "['PROPY-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01']\n",
      "['Initial Feed', 'B1T', 'B2T', 'B1B', 'B3B']\n"
     ]
    },
    {
     "ename": "IndexError",
     "evalue": "list index out of range",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mIndexError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-47-4baeee577f35>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m     21\u001b[0m     \u001b[0mindexreward\u001b[0m \u001b[1;33m=\u001b[0m \u001b[0mmaxrewardwhichhasbeenfoundV1\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mindex\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mreward_for_which_actionsshould_be_found\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     22\u001b[0m     \u001b[1;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mrange\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;36m0\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m5\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;36m1\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m---> 23\u001b[1;33m         \u001b[0mfeedlistaction\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mfeed_for_max_reward_original\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mindexreward\u001b[0m\u001b[1;33m*\u001b[0m\u001b[1;36m5\u001b[0m\u001b[1;33m+\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m     24\u001b[0m         \u001b[0mlightkeylistaction\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mlightkey_for_max_reward_original\u001b[0m\u001b[1;33m[\u001b[0m\u001b[1;33m(\u001b[0m\u001b[0mindexreward\u001b[0m\u001b[1;33m*\u001b[0m\u001b[1;36m5\u001b[0m \u001b[1;33m+\u001b[0m\u001b[0mi\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m     25\u001b[0m     \u001b[0mprint\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;34mf'\\n the reward for the following action sequence was: {maxrewardwhichhasbeenfoundV1[indexreward]}'\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mIndexError\u001b[0m: list index out of range"
     ]
    }
   ],
   "source": [
    "#Reward = [-46.333285993461146, -71.73157866904748, -77.73822694813605, -83.95066056071371, -166.79254137633285, -307.50118821544834, -313.8478503691101, -327.433180245373, -694.1115929033054, -42.97795394309483, -83.91968848372144, -154.7199383790515, -227.8319264501214, -38.94774942448791, -59.418380156262856, -86.66806411102014, -100.22234289668502, -170.0495987324897, -580.3143666041694, -864.6530243471057, -50.06471238710684, -59.03874613071028, -59.35917420882296, -82.28946761308428, -97.3710712117489, -185.6544858357778, -228.60713717747254, -255.3632227309166, -42.79940444831174, -66.39262591771715, -85.55795237002846, -408.099979746019]\n",
    "feed_for_max_reward_original= ['Initial Feed', 'B1T', 'B2T', 'B1B', 'B3B', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B1T', 'B3B', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B3B', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B4T']\n",
    "lightkey_for_max_reward_original= ['PROPY-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', '1-BUT-01']\n",
    "\n",
    "sortedmaxrewardwhichhasbeenfoundV1 = []\n",
    "for value in maxrewardwhichhasbeenfoundV1:\n",
    "    sortedmaxrewardwhichhasbeenfoundV1.append(value)\n",
    "sortedmaxrewardwhichhasbeenfoundV1.sort()\n",
    "print(f' the sorted rewards are: {sortedmaxrewardwhichhasbeenfoundV1}')\n",
    "\n",
    "#reward_for_which_actionsshould_be_found = max(maxrewardwhichhasbeenfoundV1)\n",
    "\n",
    "for qaqaqa in range(-5,0,1):\n",
    "    reward_for_which_actionsshould_be_found = sortedmaxrewardwhichhasbeenfoundV1[qaqaqa]\n",
    "    lightkeylistaction = []\n",
    "    feedlistaction = []\n",
    "    difference = []\n",
    "    for i in range(0,len(maxrewardwhichhasbeenfoundV1)):\n",
    "        difference.append((maxrewardwhichhasbeenfoundV1[i]-maxrewardwhichhasbeenfoundV2[i])**2)\n",
    "    #print(f'the square average difference between the two Stream value versions is: {sum(difference)/len(difference)} ')\n",
    "    indexreward = maxrewardwhichhasbeenfoundV1.index(reward_for_which_actionsshould_be_found)\n",
    "    for i in range(0,5,1):\n",
    "        feedlistaction.append(feed_for_max_reward_original[(indexreward*5+i)])\n",
    "        lightkeylistaction.append(lightkey_for_max_reward_original[(indexreward*5 +i)])\n",
    "    print(f'\\n the reward for the following action sequence was: {maxrewardwhichhasbeenfoundV1[indexreward]}')\n",
    "    print(lightkeylistaction)\n",
    "    print(feedlistaction)\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the square average difference between the two Stream value versions is: 1.4099756012564337e-06 \n",
      "\n",
      " the reward for the following action sequence was: -38.94774942448791\n",
      "['PROPY-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01']\n",
      "['Initial Feed', 'B1T', 'B2T', 'B1B', 'B4B']\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "feed_for_max_reward_original= ['Initial Feed', 'B1T', 'B2B', 'B2T', 'B3B', 'Initial Feed', 'B1B', 'B1T', 'B2B', 'B4B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B4T', 'Initial Feed', 'B1B', 'B1T', 'B3B', 'B4B', 'Initial Feed', 'B1B', 'B1T', 'B3B', 'B4B', 'Initial Feed', 'B1T', 'B2B', 'B2T', 'B1B', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B2B', 'Initial Feed', 'B1T', 'B2B', 'B1B', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B1B', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T', 'Initial Feed', 'B1T', 'B2T', 'B2B', 'B3T', 'Initial Feed', 'B1T', 'B1B', 'B3B', 'B2B', 'Initial Feed', 'B1T', 'B2T', 'B1B', 'B4B', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B4T', 'Initial Feed', 'B1B', 'B2T', 'B3B', 'B4T', 'Initial Feed', 'B1B', 'B2B', 'B1T', 'B4B', 'Initial Feed', 'B1T', 'B2B', 'B3T', 'B4T', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B1B', 'Initial Feed', 'B1T', 'B1B', 'B3T', 'B2B', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B2T', 'Initial Feed', 'B1B', 'B2T', 'B1T', 'B4T', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B1B', 'Initial Feed', 'B1B', 'B1T', 'B2B', 'B4T', 'Initial Feed', 'B1B', 'B1T', 'B3B', 'B4T', 'Initial Feed', 'B1T', 'B2B', 'B3B', 'B3T', 'Initial Feed', 'B1B', 'B2T', 'B3T', 'B4B', 'Initial Feed', 'B1T', 'B2T', 'B3T', 'B4B', 'Initial Feed', 'B1T', 'B1B', 'B2B', 'B4B', 'Initial Feed', 'B1T', 'B2T', 'B2B', 'B4B', 'Initial Feed', 'B1T', 'B2B', 'B2T', 'B3T', 'Initial Feed', 'B1T', 'B2T', 'B3B', 'B3T']\n",
    "lightkey_for_max_reward_original= ['1-BUT-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', 'PROPY-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', 'PROPA-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'ETHAN-01', 'PROPY-01', 'PROPA-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', 'ETHAN-01', 'PROPY-01', '1-BUT-01', '1-BUT-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'ETHAN-01', 'ETHAN-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'ETHAN-01', 'PROPA-01', 'ETHAN-01', 'ETHAN-01', '1-BUT-01', 'PROPA-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', 'PROPY-01', 'ETHAN-01', 'PROPY-01', 'PROPY-01', '1-BUT-01', 'PROPY-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'PROPY-01', 'ETHAN-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'N-BUT-01', 'PROPA-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', 'ETHAN-01', 'PROPY-01', 'ETHAN-01', 'PROPA-01', '1-BUT-01', 'PROPA-01', 'N-BUT-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'PROPY-01', 'N-BUT-01', 'PROPY-01', 'PROPY-01', 'ETHAN-01', 'N-BUT-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'N-BUT-01', 'PROPA-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', '1-BUT-01', 'PROPA-01', '1-BUT-01', 'PROPY-01', '1-BUT-01', 'PROPY-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', 'N-BUT-01', 'ETHAN-01', 'N-BUT-01', '1-BUT-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', '1-BUT-01', 'ETHAN-01', '1-BUT-01', 'PROPA-01', 'N-BUT-01', 'N-BUT-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01', 'ETHAN-01', 'N-BUT-01', 'PROPY-01', 'N-BUT-01', 'PROPA-01', 'N-BUT-01', 'ETHAN-01', 'PROPA-01', 'PROPA-01', 'ETHAN-01', 'PROPY-01', '1-BUT-01', 'ETHAN-01']\n",
    "\n",
    "\n",
    "lightkeylistaction = []\n",
    "feedlistaction = []\n",
    "difference = []\n",
    "for i in range(0,len(maxrewardwhichhasbeenfoundV1)):\n",
    "    difference.append((maxrewardwhichhasbeenfoundV1[i]-maxrewardwhichhasbeenfoundV2[i])**2)\n",
    "print(f'the square average difference between the two Stream value versions is: {sum(difference)/len(difference)} ')\n",
    "indexreward = maxrewardwhichhasbeenfoundV1.index(max(maxrewardwhichhasbeenfoundV1))\n",
    "for i in range(0,5,1):\n",
    "    feedlistaction.append(feed_for_max_reward_original[(indexreward*5+i)])\n",
    "    lightkeylistaction.append(lightkey_for_max_reward_original[(indexreward*5 +i)])\n",
    "print(f'\\n the reward for the following action sequence was: {maxrewardwhichhasbeenfoundV1[indexreward]}')\n",
    "print(lightkeylistaction)\n",
    "print(feedlistaction)\n",
    "x = np.arange(0, len(maxrewardwhichhasbeenfoundV1), 1)\n",
    "plt.plot(x, maxrewardwhichhasbeenfoundV1, 'red')\n",
    "plt.plot(x, maxrewardwhichhasbeenfoundV2, 'blue')\n",
    "plt.ylabel(\"reward\")\n",
    "plt.xlabel(\"steps\")\n",
    "plt.title('the maximum reward in V1 with randome actions')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# This code takes the max rewards which were found from the best simulations and extracts the actions. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the square average difference between the two Stream value versions is: 1.4099756012508566e-06 \n",
      "\n",
      " the reward for the following action sequence was: -38.94774942448791\n",
      "['PROPY-01', 'ETHAN-01', 'PROPA-01', 'N-BUT-01', '1-BUT-01']\n",
      "['Initial Feed', 'B1T', 'B2T', 'B1B', 'B4B']\n"
     ]
    }
   ],
   "source": [
    "\n",
    "q = [-46.333285993461146, -71.73157866904748, -77.73822694813605, -83.95066056071373, -166.79254137633285, -307.50118821544834, -313.8478503691101, -327.4331760468841, -826.0966513208647, -42.97795394309483, -83.91968848372144, -154.7199482936461, -227.8319264501214, -38.94774942448791, -59.418380156262856, -86.66806393046784, -100.22234335884512, -170.0495987324897, -580.3143666041694, -864.6530243471057, -50.06471238717572, -59.03874613076937, -59.35917420882296, -82.2894675550954, -97.3710712117489, -185.65448583584666, -228.60713717747254, -255.3632227309166, -42.79940444831174, -66.39262591771715, -311.1913947879428, -408.099979746019]\n",
    "qq = [-46.334117372433816, -71.73376726155473, -77.73939026683342, -83.95178167193998, -166.7936908961861, -307.50252014384307, -313.8480427970136, -327.43438562799673, -826.0983515531202, -42.979001054412706, -83.92101414645153, -154.72014072614834, -227.83297945707204, -38.94847395981446, -59.41846301092828, -86.66983618582665, -100.2232325574851, -170.05128604805256, -580.3155343955129, -864.6542867497205, -50.065750696621286, -59.03993665450352, -59.35926933976362, -82.29067675037165, -97.37176117093944, -185.65554495369264, -228.60744694828037, -255.36392234704982, -42.80096566432073, -66.39440872548208, -311.19295930926586, -408.10100965431286]\n",
    "maxrewardwhichhasbeenfoundV1 = q\n",
    "maxrewardwhichhasbeenfoundV2 = qq\n",
    "lightkeylistaction = []\n",
    "feedlistaction = []\n",
    "difference = []\n",
    "for i in range(0,len(q)):\n",
    "    difference.append((q[i]-qq[i])**2)\n",
    "print(f'the square average difference between the two Stream value versions is: {sum(difference)/len(difference)} ')\n",
    "indexreward = maxrewardwhichhasbeenfoundV1.index(max(maxrewardwhichhasbeenfoundV1))\n",
    "for i in range(0,5,1):\n",
    "    feedlistaction.append(feed_for_max_reward[(indexreward*5+i)])\n",
    "    lightkeylistaction.append(lightkey_for_max_reward[(indexreward*5 +i)])\n",
    "print(f'\\n the reward for the following action sequence was: {maxrewardwhichhasbeenfoundV1[indexreward]}')\n",
    "print(lightkeylistaction)\n",
    "print(feedlistaction)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 191,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'hist' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-191-932b110bcfbb>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[1;32m----> 1\u001b[1;33m \u001b[0mhist\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mhistory\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mkeys\u001b[0m\u001b[1;33m(\u001b[0m\u001b[1;33m)\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m: name 'hist' is not defined"
     ]
    }
   ],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the square average difference between the two Stream value versions is: 8.551885149271755e-07 \n"
     ]
    },
    {
     "ename": "AttributeError",
     "evalue": "'StableEnv' object has no attribute 'MonetaryGainV1remembered'",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mAttributeError\u001b[0m                            Traceback (most recent call last)",
      "\u001b[1;32m<ipython-input-38-943b3bf3030c>\u001b[0m in \u001b[0;36m<module>\u001b[1;34m\u001b[0m\n\u001b[0;32m      5\u001b[0m \u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      6\u001b[0m \u001b[0mcounter\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;36m0\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[1;32m----> 7\u001b[1;33m \u001b[0mfunctionlist\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mMonetaryGainV1remembered\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mMonetaryGainV2remembered\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mPunishmentremembered\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mMonetaryLossremembered\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mMoneyForBankersremembered\u001b[0m\u001b[1;33m,\u001b[0m \u001b[0mstable_env\u001b[0m\u001b[1;33m.\u001b[0m\u001b[0mOperationalcostsForCurrentSequenceremembered\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0m\u001b[0;32m      8\u001b[0m \u001b[0mtitelList\u001b[0m \u001b[1;33m=\u001b[0m \u001b[1;33m[\u001b[0m\u001b[1;34m'Monetary Gain Version 1'\u001b[0m \u001b[1;33m,\u001b[0m \u001b[1;34m'Monetary Gain Version 2'\u001b[0m\u001b[1;33m,\u001b[0m \u001b[1;34m'Punishment'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;34m'Monetary Loss'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;34m'Fixed Capital costs'\u001b[0m\u001b[1;33m,\u001b[0m\u001b[1;34m'Operational costs'\u001b[0m\u001b[1;33m]\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n\u001b[0;32m      9\u001b[0m \u001b[1;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[1;32min\u001b[0m \u001b[0mfunctionlist\u001b[0m\u001b[1;33m:\u001b[0m\u001b[1;33m\u001b[0m\u001b[1;33m\u001b[0m\u001b[0m\n",
      "\u001b[1;31mAttributeError\u001b[0m: 'StableEnv' object has no attribute 'MonetaryGainV1remembered'"
     ]
    }
   ],
   "source": [
    "print(f'the square average difference between the two Stream value versions is: {sum(difference)/len(difference)} ')\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "counter = 0\n",
    "functionlist = [stable_env.MonetaryGainV1remembered, stable_env.MonetaryGainV2remembered, stable_env.Punishmentremembered, stable_env.MonetaryLossremembered, stable_env.MoneyForBankersremembered, stable_env.OperationalcostsForCurrentSequenceremembered]\n",
    "titelList = ['Monetary Gain Version 1' , 'Monetary Gain Version 2', 'Punishment','Monetary Loss','Fixed Capital costs','Operational costs']\n",
    "for i in functionlist:\n",
    "    plt.plot(i)\n",
    "    plt.title(f'{titelList[counter]}')\n",
    "    plt.xlabel(\"steps\")\n",
    "    plt.ylabel(\"money $/h\")\n",
    "    plt.plot((np.ones(len(i))*sum(i)/len(i)), color = 'red')\n",
    "    plt.plot((statistics.median(i)*np.ones(len(i))) ,color = 'green')\n",
    "    plt.show()\n",
    "    print(f'the average value is: {sum(i)/len(i)}')\n",
    "    print(f'the median value is: {statistics.median(i)}')\n",
    "    print(f'the minimum value is: {min(i)}')\n",
    "    print(f'the maximum value is: {max(i)}')\n",
    "    counter = counter +1    \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.0011136293575086452\n",
      "0.0\n",
      "300.0\n",
      "3.7999999999999995e-05\n",
      "8.91090272937165\n",
      "125.21572362712018\n"
     ]
    }
   ],
   "source": [
    "print(stable_env.MonetaryGainV1)\n",
    "print(stable_env.MonetaryGainV2)\n",
    "print(stable_env.Punishment)\n",
    "print(stable_env.MonetaryLoss) \n",
    "print(stable_env.MoneyForBankers)\n",
    "print(stable_env.OperationalcostsForCurrentSequence)"
   ]
  },
  {
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}