Example #1
0
def add_storage(it, labels, nodes, busd):
    """Adds oemof.solph.GenericStorage objects to the list of components.

    If attribute `invest` is *True*, the investment version of the Storage is created.

    Parameters
    ----------
    it : pd.DataFrame
        Table of storage attributes of the producers / consumers.
    labels : dict
        Dictonary containing specifications for label-tuple.
    nodes : list
        All oemof.solph components are added to the list.
    busd : dict
        All oemof.solph.Bus objects are given by this dictionary.

    Returns
    -------
    list : Updated list of nodes.
    """

    for _, s in it.iterrows():

        label_storage = oh.Label(labels['l_1'], s['bus'], s['label'], labels['l_4'])
        label_bus = busd[(labels['l_1'], s['bus'], 'bus', labels['l_4'])]

        if s['invest']:

            epc_s = s['capex']

            nodes.append(
                solph.components.GenericStorage(
                    label=label_storage,
                    inputs={label_bus: solph.Flow()},
                    outputs={label_bus: solph.Flow()},
                    loss_rate=s['capacity_loss'],
                    fixed_losses_relative=s['fixed_losses_relative'],
                    invest_relation_input_capacity=s[
                        'invest_relation_input_capacity'],
                    invest_relation_output_capacity=s[
                        'invest_relation_output_capacity'],
                    inflow_conversion_factor=s['inflow_conversion_factor'],
                    outflow_conversion_factor=s[
                        'outflow_conversion_factor'],
                    investment=solph.Investment(ep_costs=epc_s)))

        else:
            nodes.append(
                solph.components.GenericStorage(
                    label=label_storage,
                    inputs={label_bus: solph.Flow()},
                    outputs={label_bus: solph.Flow()},
                    loss_rate=s['capacity_loss'],
                    fixed_losses_relative=s['fixed_losses_relative'],
                    nominal_storage_capacity=s['capacity'],
                    inflow_conversion_factor=s['inflow_conversion_factor'],
                    outflow_conversion_factor=s[
                        'outflow_conversion_factor']))

    return nodes
Example #2
0
def add_heatpipes(it, labels, gd, q, b_in, b_out, nodes):
    """
    Adds *HeatPipeline* objects with *Investment* attribute to the list of oemof.solph components.

    Parameters
    ----------
    it : pd.DataFrame
        Table of *Heatpipeline* attributes of the district heating grid
    labels : dict
        Dictonary containing specifications for label-tuple
    gd : dict
        Settings of the investment optimisation of the ThermalNetwork
    q : pd.Series
        Specific *Pipe* of ThermalNetwork
    b_in : oemof.solph.Bus
        Bus of Inflow
    b_out : oemof.solph.Bus
        Bus of Outflow
    nodes : list
        All oemof.solph components are added to the list

    Returns
    -------
    list : Updated list of nodes.
    """

    for _, t in it.iterrows():

        # definition of tag3 of label -> type of pipe
        labels['l_3'] = t['label_3']

        epc_p = t['capex_pipes'] * q['length']
        epc_fix = t['fix_costs'] * q['length']

        # Heatpipe with binary variable
        nc = bool(t['nonconvex'])

        # bidirectional heatpipelines yes or no
        flow_bi_args = {
            'bidirectional': True, 'min': -1}\
            if gd['bidirectional_pipes'] else {}

        nodes.append(oh.HeatPipeline(
            label=oh.Label(labels['l_1'], labels['l_2'],
                           labels['l_3'], labels['l_4']),
            inputs={b_in: solph.Flow(**flow_bi_args)},
            outputs={b_out: solph.Flow(
                nominal_value=None,
                **flow_bi_args,
                investment=solph.Investment(
                    ep_costs=epc_p, maximum=t['cap_max'],
                    minimum=t['cap_min'], nonconvex=nc, offset=epc_fix,
                ))},
            heat_loss_factor=t['l_factor'] * q['length'],
            heat_loss_factor_fix=t['l_factor_fix'] * q['length'],
        ))

    return nodes
Example #3
0
def add_heatpipes_exist(pipes, labels, gd, q, b_in, b_out, nodes):
    """
    Adds *HeatPipeline* objects with fix capacity for existing pipes
    to the list of oemof.solph components.

    Parameters
    ----------
    pipes
    labels : dict
        Dictonary containing specifications for label-tuple.
    gd : dict
        Settings of the investment optimisation of the ThermalNetwork
    q : pd.Series
        Specific *Pipe* of ThermalNetwork
    b_in : oemof.solph.Bus
        Bus of Inflow
    b_out : oemof.solph.Bus
        Bus of Outflow
    nodes : list
        All oemof.solph components are added to the list

    Returns
    -------
    list : Updated list of nodes.
    """

    # get index of existing pipe label of pipe data
    ind_pipe = pipes[pipes['label_3'] == q['hp_type']].index[0]
    t = pipes.loc[ind_pipe]
    # get label of pipe
    labels['l_3'] = t['label_3']

    hlf = t['l_factor'] * q['length']
    hlff = t['l_factor_fix'] * q['length']

    flow_bi_args = {
        'bidirectional': True, 'min': -1} \
        if gd['bidirectional_pipes'] else {}

    outflow_args = {'nonconvex': solph.NonConvex()} if t['nonconvex'] else {}

    nodes.append(oh.HeatPipeline(
        label=oh.Label(labels['l_1'], labels['l_2'],
                       labels['l_3'], labels['l_4']),
        inputs={b_in: solph.Flow(**flow_bi_args)},
        outputs={b_out: solph.Flow(
            nominal_value=q['capacity'],
            **flow_bi_args,
            **outflow_args,
        )},
        heat_loss_factor=hlf,
        heat_loss_factor_fix=hlff,
    ))

    return nodes
Example #4
0
def add_demand(it, labels, series, nodes, busd):
    """
    Initialisation of oemof.solph.Sinks which represent demands.

    Parameters
    ----------
    it : pd.DataFrame
        Table of attributes for Sources for the producers and consumers.
    labels : dict
        Dictonary containing specifications for label-tuple.
    series : dict
        Contain the heat demand time-series of all consumers.
    nodes : list
        All oemof.solph components are added to the list.
    busd : dict
        All oemof.solph.Bus objects are given by this dictionary.

    Returns
    -------
    list : Updated list of nodes.
    """

    for _, de in it.iterrows():
        labels['l_3'] = 'demand'
        labels['l_2'] = de['label_2']
        # set static inflow values
        inflow_args = {'nominal_value': de['nominal_value'],
                       'fix': series['heat_flow'][
                           labels['l_4'].split('-')[1]].values}

        # create
        nodes.append(
            solph.Sink(label=oh.Label(
                labels['l_1'], labels['l_2'], labels['l_3'], labels['l_4']),
                inputs={busd[(labels['l_1'], labels['l_2'], 'bus',
                              labels['l_4'])]: solph.Flow(**inflow_args)}))

    return nodes
Example #5
0
def add_nodes_dhs(opti_network, gd, nodes, busd):
    """
    Based on the *forks* and *pipes* of the *ThermalNetwork* of the
    *OemofInvestOptimisationModel*, the oemof-solph components for the district heating
    network optimisation are initialised.

    For all *forks*, a `solph.Bus` with the label `infrastructure_heat_bus_forks-<id>`
    (string representation) is generated.

    For all *pipes*, a 'HeatPipeline' with the label
    `infrastructure_heat_<type>_<from_node>-<to_node>` (string representation)
    is generated. Depending on the attributes in *pipes*,
    an investment pipe, or an existing pipe is built.


    Parameters
    ----------
    opti_network : OemofInvestOptimisationModel
    gd : dict
        General optimisation settings.
    nodes : list
        List for collecting all oemof-solph objects.
    busd : dict
        Dictionary collecting all oemof-solph Buses. Keys: labels of the oemof-solph Buses;
        Values: oemof-solph Buses.

    Returns
    -------
    list, dict : List of all oemof-solph objects and dictionary of oemof-solph Buses.
    """

    d_labels = {}

    d_labels['l_2'] = 'heat'
    d_labels['l_3'] = 'bus'

    for n, _ in opti_network.thermal_network.components['forks'].iterrows():
        d_labels['l_4'] = 'forks-' + str(n)
        d_labels['l_1'] = 'infrastructure'
        l_bus = oh.Label(d_labels['l_1'], d_labels['l_2'], d_labels['l_3'],
                         d_labels['l_4'])
        bus = solph.Bus(label=l_bus)
        nodes.append(bus)
        busd[l_bus] = bus

    # add heatpipes for all lines
    for p, q in opti_network.thermal_network.components['pipes'].iterrows():

        pipe_data = opti_network.invest_options['network']['pipes']

        d_labels['l_1'] = 'infrastructure'
        d_labels['l_2'] = 'heat'

        if q['existing']:

            # terminate the first label
            l_1_in = 'infrastructure'
            l_1_out = 'infrastructure'

            typ_from = q['from_node'].split('-')[0]
            typ_to = q['to_node'].split('-')[0]

            if (typ_from == 'forks') and (typ_to == 'consumers'):
                l_1_out = 'consumers'

                start = q['from_node']
                end = q['to_node']
                b_in = busd[(l_1_in, d_labels['l_2'], 'bus', start)]
                b_out = busd[(l_1_out, d_labels['l_2'], 'bus', end)]
                d_labels['l_4'] = start + '-' + end
                nodes = ac.add_heatpipes_exist(pipe_data, d_labels, gd, q,
                                               b_in, b_out, nodes)

            elif (typ_from == 'consumers') and (typ_to == 'forks'):
                raise ValueError(
                    "Pipes must not go from 'consumers' to 'forks'!"
                    " Existing heatpipe id {}".format(p))

            elif (typ_from == 'forks') and (typ_to == 'producers'):
                l_1_in = 'producers'

                start = q['to_node']
                end = q['from_node']
                b_in = busd[(l_1_in, d_labels['l_2'], 'bus', start)]
                b_out = busd[(l_1_out, d_labels['l_2'], 'bus', end)]
                d_labels['l_4'] = start + '-' + end
                nodes = ac.add_heatpipes_exist(pipe_data, d_labels, gd, q,
                                               b_in, b_out, nodes)

            elif (typ_from == 'producers') and (typ_to == 'forks'):
                l_1_in = 'producers'

                start = q['from_node']
                end = q['to_node']
                b_in = busd[(l_1_in, d_labels['l_2'], 'bus', start)]
                b_out = busd[(l_1_out, d_labels['l_2'], 'bus', end)]
                d_labels['l_4'] = start + '-' + end
                nodes = ac.add_heatpipes_exist(pipe_data, d_labels, gd, q,
                                               b_in, b_out, nodes)

            elif (typ_from == 'forks') and (typ_to == 'forks'):

                start = q['from_node']
                end = q['to_node']
                b_in = busd[(l_1_in, d_labels['l_2'], 'bus', start)]
                b_out = busd[(l_1_out, d_labels['l_2'], 'bus', end)]
                d_labels['l_4'] = start + '-' + end
                nodes = ac.add_heatpipes_exist(pipe_data, d_labels, gd, q,
                                               b_in, b_out, nodes)

            else:
                raise ValueError("Something wrong!")

        else:  # calls Investment heatpipeline function
            # connection of houses
            if q['to_node'].split('-')[0] == "consumers":

                start = q['from_node']
                end = q['to_node']
                b_in = busd[(d_labels['l_1'], d_labels['l_2'], 'bus', start)]
                b_out = busd[('consumers', d_labels['l_2'], 'bus', end)]

                d_labels['l_4'] = start + '-' + end

                nodes = ac.add_heatpipes(pipe_data, d_labels, gd, q, b_in,
                                         b_out, nodes)

            elif q['from_node'].split('-')[0] == "consumers":
                raise ValueError("Pipes must not go from 'consumers'!"
                                 " Existing heatpipe id {}".format(p))

            elif q['to_node'].split('-')[0] == "producers":

                start = q['to_node']
                end = q['from_node']
                b_in = busd[('producers', d_labels['l_2'], 'bus', start)]
                b_out = busd[(d_labels['l_1'], d_labels['l_2'], 'bus', end)]

                d_labels['l_4'] = start + '-' + end

                nodes = ac.add_heatpipes(pipe_data, d_labels, gd, q, b_in,
                                         b_out, nodes)

            elif q['from_node'].split('-')[0] == "producers":

                start = q['from_node']
                end = q['to_node']
                b_in = busd[('producers', d_labels['l_2'], 'bus', start)]
                b_out = busd[(d_labels['l_1'], d_labels['l_2'], 'bus', end)]

                d_labels['l_4'] = start + '-' + end

                nodes = ac.add_heatpipes(pipe_data, d_labels, gd, q, b_in,
                                         b_out, nodes)

            elif (q['from_node'].split('-')[0]
                  == 'forks') and (q['to_node'].split('-')[0] == 'forks'):

                b_in = busd[(d_labels['l_1'], d_labels['l_2'], 'bus',
                             q['from_node'])]
                b_out = busd[(d_labels['l_1'], d_labels['l_2'], 'bus',
                              q['to_node'])]
                d_labels['l_4'] = q['from_node'] + '-' + q['to_node']

                nodes = ac.add_heatpipes(pipe_data, d_labels, gd, q, b_in,
                                         b_out, nodes)

                if not gd['bidirectional_pipes']:
                    # the heatpipes from fork to fork need to be created in
                    # both directions in this case bidiretional = False
                    b_in = busd[(d_labels['l_1'], d_labels['l_2'], 'bus',
                                 q['to_node'])]
                    b_out = busd[(d_labels['l_1'], d_labels['l_2'], 'bus',
                                  q['from_node'])]
                    d_labels['l_4'] = q['to_node'] + '-' + q['from_node']

                    nodes = ac.add_heatpipes(pipe_data, d_labels, gd, q, b_in,
                                             b_out, nodes)

            else:
                raise ValueError("Something wrong!")

    return nodes, busd
Example #6
0
def add_sources(on, it, c, labels, nodes, busd):
    """
    The oemof.solph.Source components for the producers and consumers are initialised based on the
    given tabular information of the investment_options of the OemofInvestOptimizationModel.
    Time-series can also be used as attribute values for the outflow of the Source.
    Therefore, a table with the filename 'source_timeseries' must be given.

    Parameters
    ----------
    on : OemofInvestOptimizationModel
    it : DataFrame
        Table of attributes for Sources for the producers and consumers.
    c : Series
        Attributes of specific producer or consumer from the ThermalNetwork.
    labels : dict
        Dictonary containing specifications for label-tuple.
    nodes : list
        All oemof.solph components are added to the list.
    busd : dict
        All oemof.solph.Bus objects are given by this dictionary.

    Returns
    -------
    list : Updated list of nodes.
    """

    # check if timeseries are given
    ts_status = False   # status if timeseries for sources is present
    if 'source_' + 'timeseries' in on.invest_options[labels['l_1']].keys():
        ts = on.invest_options[labels['l_1']]['source_' + 'timeseries']
        ts_status = True

    # generel further flow attributes: check what flow attributes are given
    # by what comes after 'label_2'
    flow_attr = list(it.columns)
    idx = flow_attr.index('label_2')
    flow_attr = flow_attr[idx + 1:]

    for _, cs in it.iterrows():
        labels['l_3'] = 'source'
        labels['l_2'] = cs['label_2']

        outflow_args = {}

        # general additional flow attributes
        for fa in flow_attr:
            outflow_args[fa] = cs[fa]

        # specific flow attributes
        # e.g. check for heat (label 2)
        # e.g. check for source (label 3)
        spec_attr = [x for x in list(on.thermal_network.components[labels['l_1']].columns)
                     if x.split('.')[-1] in on.oemof_flow_attr
                     if x.split('.')[0] == labels['l_2']
                     if x.split('.')[1] == labels['l_3']]

        for sa in spec_attr:
            if sa.split('.')[-1] in outflow_args.keys():
                print(
                    'General attribute <{}> of Label 2 <{}> and Label 3 <{}> (value: {}) will '
                    'be replaced by specific data. New value for <{}>: {}'.format(
                        sa.split('.')[-1], labels['l_2'], labels['l_3'],
                        outflow_args[sa.split('.')[-1]], labels['l_4'], c[sa]))
            outflow_args[sa.split('.')[-1]] = c[sa]

        # add timeseries data if present
        if ts_status:
            ts_key = labels['l_4'].split('-')[1] + '_' + labels['l_2']
            for col in ts.columns.values:
                if col.split('.')[0] == ts_key:
                    outflow_args[col.split('.')[1]] = ts[col].values

        nodes.append(
            solph.Source(
                label=oh.Label(
                    labels['l_1'], labels['l_2'], labels['l_3'], labels['l_4']),
                outputs={
                    busd[(labels['l_1'], cs['label_2'], 'bus',
                          labels['l_4'])]: solph.Flow(**outflow_args)}))

    return nodes
Example #7
0
def add_transformer(it, labels, nodes, busd):
    """Adds oemof.solph.Transformer objects to the list of components.

    If attribute `invest` is *True*, an `Investment` attribute is added to the outflow of the
    Transformer.

    Parameters
    ----------
    it : pd.DataFrame
        Table of transformer attributes of the producers / consumers.
    labels : dict
        Dictonary containing specifications for label-tuple.
    nodes : list
        All oemof.solph components are added to the list.
    busd : dict
        All oemof.solph.Bus objects are given by this dictionary.

    Returns
    -------
    list : Updated list of nodes.
    """

    for _, t in it.iterrows():
        labels['l_2'] = None
        labels['l_3'] = t['label_3']

        # Transformer with 1 Input and 1 Output
        if t['type'] == "1-in_1-out":

            b_in_1 = busd[(labels['l_1'], t['in_1'], 'bus', labels['l_4'])]
            b_out_1 = busd[(labels['l_1'], t['out_1'], 'bus',
                            labels['l_4'])]

            if t['invest']:

                if t['eff_out_1'] == 'series':
                    print('noch nicht angepasst!')

                epc_t = t['capex']

                # create
                nodes.append(
                    solph.Transformer(
                        label=oh.Label(
                            labels['l_1'], labels['l_2'], labels['l_3'], labels['l_4']),
                        inputs={b_in_1: solph.Flow()},
                        outputs={b_out_1: solph.Flow(
                            variable_costs=t['variable_costs'],
                            summed_max=t['in_1_sum_max'],
                            investment=solph.Investment(
                                ep_costs=epc_t + t['service'],
                                maximum=t['max_invest'],
                                minimum=t['min_invest']))},
                        conversion_factors={
                            b_out_1: t['eff_out_1']}))

            else:
                # create
                if t['eff_out_1'] == 'series':
                    print('noch nicht angepasst!')
                    # for col in nd['timeseries'].columns.values:
                    #     if col.split('.')[0] == t['label']:
                    #         t[col.split('.')[1]] = nd['timeseries'][
                    #             col]

                nodes.append(
                    solph.Transformer(
                        label=oh.Label(labels['l_1'], labels['l_2'], labels['l_3'],
                                       labels['l_4']),
                        inputs={b_in_1: solph.Flow()},
                        outputs={b_out_1: solph.Flow(
                            nominal_value=t['installed'],
                            summed_max=t['in_1_sum_max'],
                            variable_costs=t['variable_costs'])},
                        conversion_factors={b_out_1: t['eff_out_1']}))

    return nodes
Example #8
0
def add_buses(it, labels, nodes, busd):
    """
    This function initialises the oemof.solph.Bus classes of the energysystem based on the given
    tabular information. Additionally, a sink or a source can be added to every bus and costs for
    this source (shortage) or sink (excess) can be defined.

    The table must be given as follows:

    .. _bus_table:

    .. table:: Example for table of *Buses*

        +---------+--------+--------+----------+----------------+--------------+
        | label_2 | active | excess | shortage | shortage costs | excess costs |
        +=========+========+========+==========+================+==============+
        | heat    | 1      | 0      | 0        | 99999          | 99999        |
        +---------+--------+--------+----------+----------------+--------------+

    Parameters
    ----------
    it : pd.DateFrame
        Table of attributes for Buses for the producers and consumers.
    labels : dict
        Dictonary containing tag1 and tag4 of label-tuple.
    nodes : list
        All oemof.solph components are added to the list.
    busd : dict
        All buses are added to this dictionary.

    Returns
    -------
    list, dict : Updated list of nodes and dict of Buses.
    """

    for _, b in it.iterrows():

        labels['l_3'] = 'bus'
        labels['l_2'] = b['label_2']
        l_bus = oh.Label(labels['l_1'], labels['l_2'], labels['l_3'], labels['l_4'])

        # check if bus already exists (due to infrastructure)
        if l_bus in busd:
            print('Bus bereits vorhanden:', l_bus)

        else:
            bus = solph.Bus(label=l_bus)
            nodes.append(bus)

            busd[l_bus] = bus

            if b['excess']:
                labels['l_3'] = 'excess'
                nodes.append(
                    solph.Sink(label=oh.Label(
                        labels['l_1'], labels['l_2'], labels['l_3'], labels['l_4']),
                        inputs={busd[l_bus]: solph.Flow(variable_costs=b['excess costs'])}))

            if b['shortage']:
                labels['l_3'] = 'shortage'
                nodes.append(
                    solph.Source(label=oh.Label(
                        labels['l_1'], labels['l_2'], labels['l_3'], labels['l_4']),
                        outputs={busd[l_bus]: solph.Flow(variable_costs=b['shortage costs'])}))

    return nodes, busd