Esempio n. 1
0
def getCapacity(folders,
                dbs,
                switch='fuel',
                sector_name='electric',
                save_data='N',
                create_plots='N',
                run_name=''):
    #    inputs:
    #    1) folders         - paths containing dbs (list or single string if all in the same path)
    #    2) dbs             - names of databases (list)
    #    3) switch          - 'fuel' or 'tech', basis of categorization
    #    4) sectorName      - name of temoa sector to be analyzed
    #    5) saveData         - 'Y' or 'N', default is 'N'
    #    6) createPlot      - 'Y' or 'N', default is 'N'
    #    7) run_name         - Used for saving results in dedicated folder
    #
    #    outputs:
    #    1) capacity     - pandas DataFrame holding capacity for each model year
    # ==============================================================================
    print("Analyzing capacity")

    # Save original directory
    wrkdir = os.getcwd()

    # If only a single db and folder provided, change to a list
    if type(dbs) == str and type(folders) == str:
        dbs = [dbs]
        folders = [folders]
    # If a list of folders is provided with one database, only use first folder
    elif type(dbs) == str:
        dbs = [dbs]
        folders = [folders[0]]
    # If only a single folder provided, create a list of the same folder
    elif type(folders) == str:
        fldrs = []
        for db in dbs:
            fldrs.append(folders)
        folders = fldrs

    # Create dictionary to hold each capacity_single series
    capacity = pd.DataFrame(dtype='float64')

    # Iterate through each db
    for folder, db in zip(folders, dbs):
        capacity_single = SingleDB(folder,
                                   db,
                                   switch=switch,
                                   sector_name=sector_name)
        capacity = pd.concat([capacity, capacity_single])

    # Reset index (remove multi-level indexing, easier to use in Excel)
    capacity = capacity.reset_index()

    # Directory to hold results
    if save_data == 'Y' or create_plots == 'Y':
        tt.create_results_dir(wrkdir=wrkdir, run_name=run_name)

    # Save results to Excel
    if save_data == 'Y':
        # Create savename based on switch
        if switch == 'fuel':
            savename = 'capacity_by_fuel.csv'
        elif switch == 'tech':
            savename = 'capacity_by_tech.csv'
        # Save
        capacity.to_csv(savename)

    # Create plots
    if create_plots == 'Y':
        import matplotlib.pyplot as plt
        import seaborn as sns
        plt.rcParams.update({'figure.max_open_warning': 0})  # ignore warning

        # new figure
        plt.figure()
        # set aesthetics
        sns.set_style("white", {
            "font.family": "serif",
            "font.serif": ["Times", "Palatino", "serif"]
        })
        sns.set_context("paper")
        sns.set_style("ticks", {"xtick.major.size": 8, "ytick.major.size": 8})

        # wide to long
        df2 = pd.melt(capacity,
                      id_vars=['database', 'scenario', 'fuelOrTech'],
                      var_name='var',
                      value_name='value')
        # plot
        sns.relplot(x='var',
                    y='value',
                    hue='database',
                    data=df2,
                    kind='line',
                    col='fuelOrTech',
                    col_wrap=4)

        # save
        if switch == 'fuel':
            savename = 'capacity_by_fuel.png'
        elif switch == 'tech':
            savename = 'capacity_by_tech.png'
        plt.savefig(savename, dpi=resolution)

        # close figure
        plt.close()

    # Return to original directory
    os.chdir(wrkdir)

    # return capacity as a dictionary
    return capacity
Esempio n. 2
0
def getCosts(folders,
             dbs,
             elc_dmd='ELC_DMD',
             conversion=0.359971,
             save_data='N',
             create_plots='N',
             run_name=''):
    #    inputs:
    #    1) folders         - paths containing dbs (list or single string if all in the same path)
    #    2) dbs             - names of databases (list)
    #    3) elc_dmd         - quantity that represents electricity demand
    #    4) conversion      - converts from cost units per activity to cents/kWH
    #           default is conversion from M$/PJ to cents/KWh (1E6*100 / (2.778E8))
    #    5) save_data         - 'Y' or 'N', default is 'N'
    #    6) create_plots     - 'Y' or 'N', default is 'N'
    #    7) run_name         - Used for saving results in dedicated folder
    #
    #    outputs:
    #    1) yearlyCosts     - pandas DataFrame holding yearly_costs
    #    2) LCOE            - dictionary holding LCOE, calculated wrt first model year
    # ==============================================================================
    print("Analyzing costs")

    # Save original directory
    wrkdir = os.getcwd()

    # If only a single db and folder provided, change to a list
    if type(dbs) == str and type(folders) == str:
        dbs = [dbs]
        folders = [folders]
    # If a list of folders is provided with one database, only use first folder
    elif type(dbs) == str:
        dbs = [dbs]
        folders = [folders[0]]
    # If only a single folder provided, create a list of the same folder
    elif type(folders) == str:
        fldrs = []
        for db in dbs:
            fldrs.append(folders)
        folders = fldrs

    # Create a dataframe
    yearlyCosts = pd.DataFrame(dtype='float64')
    LCOE = pd.DataFrame(dtype='float64')

    # Iterate through each db
    for folder, db in zip(folders, dbs):
        # Access costs
        yearlyCosts_single, LCOE_single = SingleDB(folder,
                                                   db,
                                                   elc_dmd=elc_dmd,
                                                   conversion=conversion)

        yearlyCosts = pd.concat([yearlyCosts, yearlyCosts_single])
        LCOE = pd.concat([LCOE, LCOE_single])

    # Reset index (remove multi-level indexing, easier to use in Excel)
    yearlyCosts = yearlyCosts.reset_index()
    LCOE = LCOE.reset_index()

    # Directory to hold results
    if save_data == 'Y' or create_plots == 'Y':
        tt.create_results_dir(wrkdir=wrkdir, run_name=run_name)

    # Save results to CSV
    if save_data == 'Y':
        yearlyCosts.to_csv('costs_yearly.csv')
        LCOE.to_csv('LCOE.csv')

    # Plot Results
    if create_plots == 'Y':
        import matplotlib.pyplot as plt
        import seaborn as sns
        plt.rcParams.update({'figure.max_open_warning': 0})  # ignore warning

        # new figure
        plt.figure()
        # set aesthetics
        sns.set_style("white", {
            "font.family": "serif",
            "font.serif": ["Times", "Palatino", "serif"]
        })
        sns.set_context("paper")
        sns.set_style("ticks", {"xtick.major.size": 8, "ytick.major.size": 8})

        # wide to long
        df2 = pd.melt(yearlyCosts,
                      id_vars=['database', 'scenario'],
                      var_name='var',
                      value_name='value')
        # plot
        ax = sns.lineplot(x='var', y='value', hue='database', data=df2)
        ax.set_xlabel("Year [-]")
        ax.set_ylabel("Costs [cents/kWh]")
        fig = ax.get_figure()
        fig.savefig('costs_yearly.png', dpi=resolution)

        # close figure
        plt.close()

    # Return to original directory
    os.chdir(wrkdir)

    return yearlyCosts, LCOE
def SingleDB(folder,
             db,
             switch='fuel',
             sectorName='electric',
             saveData='N',
             createPlots='N',
             conversion=277.777778):
    #    inputs:
    #    1) folder          - path containing db
    #    2) db              - name of databas
    #    3) switch          - 'fuel' or 'tech', basis of categorization
    #    4) sectorName      - name of temoa sector to be analyzed
    #    5) saveData         - 'Y' or 'N', default is 'N'
    #    6) createPlot      - 'Y' or 'N', default is 'N'
    #    7) conversion      - conversion to GWh, default is 277.778 (from PJ)

    #    outputs:
    #    1) activity     - pandas DataFrame holding capacity for each model year
    # ==============================================================================

    if switch == 'fuel':
        savename = 'Results_ActivityTOD_byFuel_' + name(db)
    elif switch == 'tech':
        savename = 'Results_ActivityTOD_byTech_' + name(db)

    # save original folder
    origDir = os.getcwd()

    # move to folder
    os.chdir(folder)

    # Connect to Database
    con = sqlite3.connect(db)
    cur = con.cursor()

    # Read from database:
    #   Select All Efficiencies
    qry = "SELECT * FROM Efficiency"
    cur.execute(qry)
    db_efficiency = cur.fetchall()
    #   Select All time_of_day
    qry = "SELECT * FROM time_of_day"
    cur.execute(qry)
    db_time_of_day = cur.fetchall()
    #   Select All time_season
    qry = "SELECT * FROM time_season"
    cur.execute(qry)
    db_time_season = cur.fetchall()
    #   Select All time_periods
    qry = "SELECT * FROM time_periods"
    cur.execute(qry)
    db_time_periods = cur.fetchall()
    #   Select All technologies
    qry = "SELECT * FROM technologies"
    cur.execute(qry)
    db_technologies = cur.fetchall()
    #   Select All Flows
    qry = "SELECT * FROM Output_VFlow_Out"
    cur.execute(qry)
    db_Output_VFlow_Out = cur.fetchall()

    # Review db_time_of_day to select timesOfDay
    tods = []
    for tod in db_time_of_day:
        tods.append(str(tod[0]))

    # Review db_time_season to select seasons
    seasons = []
    for season in db_time_season:
        seasons.append(str(season[0]))
    n_seasons = len(seasons)

    # Review db_time_periods to select future periods
    years = []
    for year, flag in db_time_periods:
        if flag == 'f':
            years.append(str(year))
    n_years = len(years)

    # Review db_technologies to select related sector
    techs = []
    for tech, flag, sector, tech_desc, tech_category in db_technologies:
        if sector == sectorName:
            if tech not in techs:
                techs.append(tech)

    # Review db_efficiency to create a dictionary of fuels
    d = {}
    for input_comm, tech, vintage, output_comm, efficiency, ef_notes in db_efficiency:
        if tech in techs:
            if tech not in d.keys():
                d[tech] = input_comm

    # Sort data and assign as columns and rows
    if switch == 'fuel':
        cols = sorted(set(d.values()))
    elif switch == 'tech':
        cols = sorted(techs)
    rows = sorted(tods)

    # Directory to hold results
    if save_data == 'Y' or create_plots == 'Y':
        tt.create_results_dir(wrkdir=wrkdir, run_name=run_name)

    # Create plot
    if createPlots == 'Y':
        f, a = plt.subplots(n_years, n_seasons, sharex=True, sharey=True)

    # Create connection to excel
    if saveData == 'Y':
        writer = pd.ExcelWriter(savename + '.xls')

    # Dictionary to store results
    activity = {}

    # Fill-in each subplot
    for i, year in enumerate(years[:-1]):
        for j, season in enumerate(seasons):

            # Create appropriate title
            titlename = str(year) + ' ' + str(season)

            # Create dataframe initialized to zero
            df = pd.DataFrame(data=0.0, index=rows, columns=cols)

            ## Review db_Output_VFlow_Out to fill data frame
            for scenario, sector, t_periods, t_season, t_day, input_comm, tech, vintage, output_comm, vflow_out in db_Output_VFlow_Out:
                if sector == sectorName:
                    if str(t_periods) == year and t_season == season:
                        if switch == 'fuel':
                            df.loc[t_day, d[tech]] = df.loc[
                                t_day, d[tech]] + vflow_out * conversion
                        elif switch == 'tech':
                            df.loc[t_day, tech] = df.loc[
                                t_day, tech] + vflow_out * conversion

            # Store results
            activity[str(year) + '_' + str(season)] = df

            # Plot
            if createPlots == 'Y':
                # Access corresponding subplot
                ax = a[i, j]

                # Legend
                if i == 0 and j == len(seasons) - 1:
                    df.plot.bar(ax=ax,
                                stacked=True,
                                title=titlename,
                                legend=True)
                    a[i, j].legend(bbox_to_anchor=(1.7, -0.2), ncol=1)
                else:
                    df.plot.bar(ax=ax,
                                stacked=True,
                                title=titlename,
                                legend=False)

                if i == len(years) - 1:  # Only bottom of plot
                    ax.set_xlabel("Year [-]")

                if j == 0:  # Only leftside of plot
                    ax.set_ylabel("Activity [GWh]")

            # Store to excel
            if saveData == 'Y':
                sheetname = str(year) + '_' + str(season)
                df.to_excel(writer, sheetname)

    # Save plot
    if createPlots == 'Y':
        fig = ax.get_figure()
        fig.savefig(savename, dpi=resolution, bbox_inches="tight")

        # Save excel file
    if saveData == 'Y':
        writer.save()

    # Return to original directory
    os.chdir(origDir)

    # Return results
    return activity
def getActivityTOD(folders,
                   dbs,
                   switch='fuel',
                   sector_name='electric',
                   save_data='N',
                   create_plots='N',
                   conversion=277.777778,
                   run_name=''):
    #    inputs:
    #    1) folders         - paths containing dbs (list or single string if all in the same path)
    #    2) dbs             - names of databases (list)
    #    3) switch          - 'fuel' or 'tech', basis of categorization
    #    4) sectorName      - name of temoa sector to be analyzed
    #    5) saveData         - 'Y' or 'N', default is 'N'
    #    6) createPlots     - 'Y' or 'N', default is 'N'
    #    7) conversion      - conversion to GWh, default is 277.778 (from PJ)
    #    8) run_name         - Used for saving results in dedicated folder

    #    outputs:
    #    1) activity
    #    2) plots - optional
    #    3) Data  - optional
    # ==============================================================================
    print("Analyzing activity by time of day (TOD)")

    # Save original directory
    wrkdir = os.getcwd()

    # If only a single db and folder provided, change to a list
    if type(dbs) == str and type(folders) == str:
        dbs = [dbs]
        folders = [folders]
    # If a list of folders is provided with one database, only use first folder
    elif type(dbs) == str:
        dbs = [dbs]
        folders = [folders[0]]
    # If only a single folder provided, create a list of the same folder
    elif type(folders) == str:
        fldrs = []
        for db in dbs:
            fldrs.append(folders)
        folders = fldrs

    # Create dataframe to hold each capacity_single series
    activity = pd.DataFrame(dtype='float64')

    # Iterate through each db
    for folder, db in zip(folders, dbs):
        activity_single = SingleDB(folder,
                                   db,
                                   switch=switch,
                                   sector_name=sector_name,
                                   conversion=conversion)
        activity = pd.concat([activity, activity_single])

    # Reset index (remove multi-level indexing, easier to use in Excel)
    activity = activity.reset_index()

    # Directory to hold results
    if save_data == 'Y' or create_plots == 'Y':
        tt.create_results_dir(wrkdir=wrkdir, run_name=run_name)

    # Save results to CSV
    if save_data == 'Y':
        # Create savename based on switch
        if switch == 'fuel':
            savename = 'activityTOD_by_fuel.csv'
        else:
            savename = 'activityTOD_by_tech.csv'
        activity.to_csv(savename)

    if create_plots == 'Y':

        df = activity.reset_index()

        import matplotlib.pyplot as plt
        import seaborn as sns
        plt.rcParams.update({'figure.max_open_warning': 0})  # ignore warning

        for database in df.database.unique():
            # new figure
            plt.figure()
            # set aesthetics
            sns.set_style(
                "white", {
                    "font.family": "serif",
                    "font.serif": ["Times", "Palatino", "serif"]
                })
            sns.set_context("talk")

            # select relevant database
            df2 = df[(df.database == database)]
            # plot
            sns.relplot(x='tod',
                        y='value',
                        hue='fuelOrTech',
                        row='year',
                        col='season',
                        data=df2,
                        kind='line')

            # save
            if switch == 'fuel':
                savename = 'yearlyActivityTOD_byFuel' + tt.remove_ext(
                    database) + '.pdf'
            else:
                savename = 'yearlyActivityTOD_byTech' + tt.remove_ext(
                    database) + '.pdf'
            plt.savefig(savename, dpi=resolution)
            # close the figure
            plt.close()

    # Return to original directory
    os.chdir(wrkdir)

    return activity
Esempio n. 5
0
def getEmissions(folders,
                 dbs,
                 conversion=1E-6,
                 save_data='N',
                 create_plots='N',
                 run_name=''):
    # ==============================================================================
    #    inputs:
    #    1) folders         - paths containing dbs (list or single string if all in the same path)
    #    2) dbs             - names of databases (list)
    #    3) conversion      - converts from emission units to Mton
    #           default is conversion from kton to Mton is 1E-6
    #    4) save_data        - 'Y' or 'N', default is 'N'
    #    5) create_plots     - 'Y' or 'N', default is 'N'
    #    6) run_name         - Used for saving results in dedicated folder
    #
    #    outputs:
    #    1) yearlyEmissions     - pandas DataFrame holding yearly emissions
    #    2) avgEmissions        - dictionary holding average emissions
    # ==============================================================================
    print("Analyzing emissions")

    # Save original directory
    wrkdir = os.getcwd()

    # If only a single db and folder provided, change to a list
    if type(dbs) == str and type(folders) == str:
        dbs = [dbs]
        folders = [folders]
    # If a list of folders is provided with one database, only use first folder
    elif type(dbs) == str:
        dbs = [dbs]
        folders = [folders[0]]
    # If only a single folder provided, create a list of the same folder
    elif type(folders) == str:
        fldrs = []
        for db in dbs:
            fldrs.append(folders)
        folders = fldrs

    # Create a dataframe
    yearlyEmissions = pd.DataFrame(dtype='float64')
    avgEmissions = pd.DataFrame(dtype='float64')

    # Iterate through each db
    for folder, db in zip(folders, dbs):
        # Move to folder
        os.chdir(folder)

        # Access costs
        yearlyEmissions_single, avgEmissions_single = SingleDB(
            folder, db, conversion=conversion)

        # Store costs
        yearlyEmissions = pd.concat([yearlyEmissions, yearlyEmissions_single])
        avgEmissions = pd.concat([avgEmissions, avgEmissions_single])

    # Sort data
    # yearlyEmissions = yearlyEmissions.sort_index()

    # Reset index (remove multi-level indexing, easier to use for processing)
    yearlyEmissions = yearlyEmissions.reset_index()
    avgEmissions = avgEmissions.reset_index()

    # Directory to hold results
    if save_data == 'Y' or create_plots == 'Y':
        tt.create_results_dir(wrkdir=wrkdir, run_name=run_name)

    # Save results to CSV
    if save_data == 'Y':
        yearlyEmissions.to_csv('emissions_yearly.csv')
        avgEmissions.to_csv('emissions_average.csv')

    # Plot Results
    if create_plots == 'Y':
        import matplotlib.pyplot as plt
        import seaborn as sns

        # new figure
        plt.figure()
        # set aesthetics
        sns.set_style("white", {
            "font.family": "serif",
            "font.serif": ["Times", "Palatino", "serif"]
        })
        sns.set_context("paper")
        sns.set_style("ticks", {"xtick.major.size": 8, "ytick.major.size": 8})

        # wide to long
        df2 = pd.melt(yearlyEmissions,
                      id_vars=['database', 'scenario'],
                      var_name='var',
                      value_name='value')
        # plot
        ax = sns.lineplot(x='var', y='value', hue='database', data=df2)

        # yearlyEmissions
        ax.set_xlabel("Year [-]")
        ax.set_ylabel("Emissions [kton]")
        fig = ax.get_figure()
        savename = 'emissions_yearly.png'
        fig.savefig(savename, dpi=resolution)

        # close figure
        plt.close()

    # Return to original directory
    os.chdir(wrkdir)

    return yearlyEmissions, avgEmissions