Source code for tespy.tools.analyses

# -*- coding: utf-8

"""Module for thermodynamic analyses.

The analyses module provides thermodynamic analysis tools for your simulation.
Different analysis classes are available:

- :py:class:`tespy.tools.analyses.ExergyAnalysis`


This file is part of project TESPy (github.com/oemof/tespy). It's copyrighted
by the contributors recorded in the version control history of the file,
available from its original location tespy/tools/analyses.py

SPDX-License-Identifier: MIT
"""
import numpy as np
import pandas as pd
from tabulate import tabulate

from tespy.tools import helpers as hlp
from tespy.tools import logger
from tespy.tools.fluid_properties import single_fluid
from tespy.tools.global_vars import ERR
from tespy.tools.global_vars import combustion_gases

idx = pd.IndexSlice


[docs] def categorize_fluids(conn): fluid = single_fluid(conn.fluid_data) if fluid is None: cat = "non-combustion-gas" for f, x in conn.fluid.val.items(): if x > ERR : try: if hlp.fluidalias_in_list(f, combustion_gases): cat = "combustion-gas" break except RuntimeError: # CoolProp cannot call aliases on incompressibles pass else: is_incompressible = False try: is_combustion_gas = hlp.fluidalias_in_list(fluid, combustion_gases) except RuntimeError: # CoolProp cannot call aliases on incompressibles is_incompressible = True is_combustion_gas = False if is_combustion_gas: cat = "combustion-gas" elif is_incompressible: cat = "incompressible" else: cat = "two-phase-fluid" return cat
[docs] class ExergyAnalysis: r"""Class for exergy analysis of TESPy models.""" exergy_cats = ["chemical", "physical", "massless"] def __init__(self, network, E_F, E_P, E_L=[], internal_busses=[]): r""" Parameters ---------- E_F : list List containing busses which represent fuel exergy input of the network, e.g. heat exchangers of the steam generator. E_P : list List containing busses which represent exergy production of the network, e.g. the motors and generators of a power plant. E_L : list Optional: List containing busses which represent exergy loss streams of the network to the ambient, e.g. flue gases of a gas turbine. internal_busses : list Optional: List containing internal busses that represent exergy transfer within your network but neither exergy production or exergy fuel, e.g. a steam turbine driven feed water pump. The conversion factors of the bus are applied to calculate exergy destruction which is allocated to the respective components. Note ---- The nomenclature of the variables used in the exergy analysis is according to :cite:`Tsatsaronis2007`. The analysis is carried out by the :py:meth:`tespy.tools.analyses.ExergyAnalysis.analyse` method. Given the ambient state (pressure and temperature), it will - Calculate the values of physical exergy on all connections - Calculate exergy balance for all components. The individual exergy balance methods are documented in the API-documentation of the respective components. - Components for which no exergy balance has yet been implemented, :code:`nan` (not defined) is assigned for fuel and product exergy as well as exergy destruction and exergetic efficiency. - Dissipative components do not have product exergy (:code:`nan`) per definition. - Calculate exergy balances for busses passed to ExergyAnalysis class instance. - Calculate network fuel exergy, product exergy as well as exergy loss from data provided by the busses passed to the instance. - Calculate network exergetic efficiency. - Calculate exergy destruction ratios for components. - :math:`y_\mathrm{D}` compare the rate of exergy destruction in a component to the exergy rate of the fuel provided to the overall system. - :math:`y^*_\mathrm{D}` compare the component exergy destruction rate to the total exergy destruction rate within the system. .. math:: \begin{split} \dot{E}_{\mathrm{D},comp} = \dot{E}_{\mathrm{F},comp} - \dot{E}_{\mathrm{P},comp} \;& \\ \varepsilon_{\mathrm{comp}} = \frac{\dot{E}_{\mathrm{P},comp}}{\dot{E}_{\mathrm{F},comp}} \;& \\ \dot{E}_{\mathrm{D}} = \sum_{comp} \dot{E}_{\mathrm{D},comp} \;& \forall comp \in \text{ network components}\\ \dot{E}_{\mathrm{P}} = \sum_{comp} \dot{E}_{\mathrm{P},comp} \;& \forall comp \in \text{ components of busses in E\_P if 'base': 'component'}\\ \dot{E}_{\mathrm{P}} = \dot{E}_{\mathrm{P}} - \sum_{comp} \dot{E}_{\mathrm{F},comp} \;& \forall comp \in \text{ components of busses in E\_P if 'base': 'bus'}\\ \dot{E}_{\mathrm{F}} = \sum_{comp} \dot{E}_{\mathrm{F},comp} \;& \forall comp \in \text{ components of busses in E\_F if 'base': 'bus'}\\ \dot{E}_{\mathrm{F}} = \dot{E}_{\mathrm{F}} - \sum_{comp} \dot{E}_{\mathrm{P},comp} \;& \forall comp \in \text{ components of busses in E\_F if 'base': 'component'}\\ \dot{E}_{\mathrm{L}} = \sum_{comp} \dot{E}_{\mathrm{D},comp} \;& \forall comp \in \text{ sinks of network components if parameter exergy='loss'} \end{split} The exergy balance of the network must be closed, meaning fuel exergy minus product exergy, exergy destruction and exergy losses must be zero (:math:`\Delta \dot{E}_\text{max}=0.001`). If the balance is violated a warning message is prompted. .. math:: |\dot{E}_{\text{F}} - \dot{E}_{\text{P}} - \dot{E}_{\text{L}} - \dot{E}_{\text{D}}| \leq \Delta \dot{E}_\text{max}\\ \varepsilon = \frac{\dot{E}_{\text{P}}}{\dot{E}_{\text{F}}} y_{\text{D},comp} = \frac{\dot{E}_{\text{D},comp}}{\dot{E}_{\text{F}}}\\ y^*_{\text{D},comp} = \frac{\dot{E}_{\text{D},comp}}{\dot{E}_{\text{D}}} Example ------- In this example a simple clausius rankine cycle is set up and an exergy analysis is performed after simulation of the power plant. Start by defining ambient state and genereral network setup. >>> from tespy.components import (CycleCloser, SimpleHeatExchanger, ... Merge, Splitter, Valve, Compressor, Pump, Turbine) >>> from tespy.connections import Bus >>> from tespy.connections import Connection >>> from tespy.networks import Network >>> from tespy.tools import ExergyAnalysis >>> Tamb = 20 >>> pamb = 1 >>> nw = Network() >>> nw.set_attr(p_unit='bar', T_unit='C', h_unit='kJ / kg', ... iterinfo=False) In order to show all functionalities available we use a feed water pump that is not driven electrically by a motor but instead internally by an own steam turbine. Therefore we split up the live steam from the steam generator and merge the streams after both steam turbines. For simplicity the steam generator and the condenser are modeled as simple heat exchangers. >>> cycle_close = CycleCloser('cycle closer') >>> splitter1 = Splitter('splitter 1') >>> merge1 = Merge('merge 1') >>> turb = Turbine('turbine') >>> fwp_turb = Turbine('feed water pump turbine') >>> condenser = SimpleHeatExchanger('condenser') >>> fwp = Pump('pump') >>> steam_generator = SimpleHeatExchanger('steam generator') >>> fs_in = Connection(cycle_close, 'out1', splitter1, 'in1') >>> fs_fwpt = Connection(splitter1, 'out1', fwp_turb, 'in1') >>> fs_t = Connection(splitter1, 'out2', turb, 'in1') >>> fwpt_ws = Connection(fwp_turb, 'out1', merge1, 'in1') >>> t_ws = Connection(turb, 'out1', merge1, 'in2') >>> ws = Connection(merge1, 'out1', condenser, 'in1') >>> cond = Connection(condenser, 'out1', fwp, 'in1') >>> fw = Connection(fwp, 'out1', steam_generator, 'in1') >>> fs_out = Connection(steam_generator, 'out1', cycle_close, 'in1') >>> nw.add_conns(fs_in, fs_fwpt, fs_t, fwpt_ws, t_ws, ws, cond, ... fw, fs_out) Next step is to set up the busses to later pass them according to the convetions in the list below: - E_F for fuel exergy - E_P for product exergy - internal_busses for internal energy transport - E_L for exergy loss streams to the ambient (sources and sinks go here, in case you use e.g. flue gases or air input) The first bus is for output power, which is only represented by the main steam turbine. The efficiency is set to 0.97. This bus will represent the product exergy. >>> power = Bus('power_output') >>> power.add_comps({'comp': turb, 'char': 0.97}) The second bus is for driving the feed water pump. The total power of this bus is specified to be 0 in order to make sure, the power genrated by the secondary steam turbine is transferred to the feed water pump. For mechanical efficiency we choose 0.985 for both components, but we need to make sure, the :code:`'base'` of the feed water pump is :code:`'bus'` as the energy from the turbine drives the feed water pump. >>> fwp_power = Bus('feed water pump power', P=0) >>> fwp_power.add_comps( ... {'comp': fwp_turb, 'char': 0.985}, ... {'comp': fwp, 'char': 0.985, 'base': 'bus'}) The fuel exergy is the exergy input into the network which is represented by the heat input bus. Here again, as we have an energy input from outside of the network, the :code:`'base'` keyword must be specified to :code:`'bus'`. >>> heat = Bus('heat_input') >>> heat.add_comps({'comp': steam_generator, 'base': 'bus'}) >>> nw.add_busses(power, fwp_power, heat) After setting up the busses, we specify the parameters for components and connections and start the simulation. >>> turb.set_attr(eta_s=0.9) >>> fwp_turb.set_attr(eta_s=0.87) >>> condenser.set_attr(pr=0.98) >>> fwp.set_attr(eta_s=0.75) >>> steam_generator.set_attr(pr=0.89) >>> fs_in.set_attr(m=10, p=120, T=600, fluid={'water': 1}) >>> cond.set_attr(T=Tamb + 3, x=0) >>> nw.solve('design') To evaluate the exergy balance of the network, we create an instance of class :py:class:`tespy.tools.analyses.ExergyAnalysis` passing the network to analyse as well as the respective busses. To run the analysis, the ambient state is passed subsequently. The results of the analysis can be printed using the :py:meth:`tespy.tools.analyses.ExergyAnalysis.print_results` method. The exergy balance should be closed, if you set up your network analysis correctly. If not, an error is prompted. >>> ean = ExergyAnalysis(network=nw, E_F=[heat], E_P=[power], ... internal_busses=[fwp_power]) >>> ean.analyse(pamb=pamb, Tamb=Tamb) >>> abs(round(ean.network_data['E_F'] - ean.network_data['E_P'] - ... ean.network_data['E_L'] - ean.network_data['E_D'], 3)) np.float64(0.0) >>> ();ean.print_results();() # doctest: +ELLIPSIS (...) The exergy data of the passed busses, the network's components and connections as well as the network itself are stored as dataframes and therefore accessible for further investigation. >>> busses = ean.bus_data >>> components = ean.component_data >>> connections = ean.connection_data >>> network = ean.network_data Additionally, if you defined component groups for your components, the exergy data of these groups are accumulated and saved in an own DataFrame. Please note, that the exergy destruction values of the busses are allocated to the component groups in this case. >>> groups = ean.group_data Lastly, the tool offers an automatically generated data input for the sankey plotting functionalities of the plotly library to create a Grassmann diagram of your network. For more information on the usage of the functionality look up the corresponding method documentation: :py:meth:`tespy.tools.analyses.ExergyAnalysis.generate_plotly_sankey_input`. The method returns a dictionary containting the links for the sankey as well as a list of the nodes. >>> links, nodes = ean.generate_plotly_sankey_input() """ if len(E_F) == 0: msg = ('Missing fuel exergy E_F of network.') logger.error(msg) raise hlp.TESPyNetworkError(msg) if len(E_P) == 0: msg = ('Missing product exergy E_P of network.') logger.error(msg) raise hlp.TESPyNetworkError(msg) self.nw = network self.E_F = E_F self.E_P = E_P self.E_L = E_L self.internal_busses = internal_busses bus_labels = [b.label for b in internal_busses + E_F + E_P + E_L] key_exergy_labels = ['E_P', 'E_F', 'E_D', 'E_L'] self.reserved_fkt_groups = key_exergy_labels + bus_labels if len(set(bus_labels).intersection(key_exergy_labels)) > 0: msg = ( "None of your busses may have the label '" + "', '".join(key_exergy_labels) + "' when performing the " + "exergy analysis." ) raise ValueError(msg)
[docs] def analyse(self, pamb, Tamb, Chem_Ex=None): """Run the exergy analysis. Parameters ---------- pamb : float Ambient pressure value for analysis, provide value in network's pressure unit. Tamb : float Ambient temperature value for analysis, provide value in network's temperature unit. """ pamb_SI = hlp.convert_to_SI('p', pamb, self.nw.p_unit) Tamb_SI = hlp.convert_to_SI('T', Tamb, self.nw.T_unit) # reset data dtypes = { "E_F": float, "E_P": float, "E_D": float, "epsilon": float, "group": str } self.component_data = pd.DataFrame( columns=list(dtypes.keys()) ).astype(dtypes) self.bus_data = self.component_data.copy() self.bus_data["base"] = np.nan self.bus_data["base"] = self.bus_data["base"].astype(str) conn_exergy_data_cols = ['e_PH', 'e_T', 'e_M', 'E_PH', 'E_T', 'E_M'] if Chem_Ex is not None: conn_exergy_data_cols += ['e_CH', 'E_CH'] self.connection_data = pd.DataFrame( columns=conn_exergy_data_cols, dtype='float64' ) self.network_data = pd.Series( index=['E_F', 'E_P', 'E_D', 'E_L'], dtype='float64' ) self.network_data[:] = 0 # physical exergy of connections for conn in self.nw.conns['object']: conn.get_physical_exergy(pamb_SI, Tamb_SI) conn.get_chemical_exergy(pamb_SI, Tamb_SI, Chem_Ex) conn_exergy_data = [ conn.ex_physical, conn.ex_therm, conn.ex_mech, conn.Ex_physical, conn.Ex_therm, conn.Ex_mech ] if Chem_Ex is not None: conn_exergy_data += [conn.ex_chemical, conn.Ex_chemical] self.connection_data.loc[conn.label] = conn_exergy_data # todo: überprüfen der sankey data + massless exergy self.sankey_data = {} sankey_columns_dtypes = { 'chemical': float, 'physical': float, 'massless': float } for label in self.reserved_fkt_groups: self.sankey_data[label] = pd.DataFrame( columns=sankey_columns_dtypes.keys(), index=pd.MultiIndex( levels=[[], []], names=["target_group", "category"], codes=[[], []] ) ).astype(sankey_columns_dtypes) # exergy balance of components for cp in self.nw.comps['object']: # save component information cp.exergy_balance(Tamb_SI) self.component_data.loc[cp.label] = [ cp.E_F, cp.E_P, cp.E_D, cp.epsilon, cp.fkt_group ] if cp.fkt_group in self.reserved_fkt_groups: msg = ( 'The labels ' + ', '.join(self.reserved_fkt_groups) + ' ' 'cannot be used by components (if no group was assigned) ' 'or component groups in the exergy analysis. Found ' 'component/group with name ' + cp.fkt_group + '.' ) raise ValueError(msg) elif cp.fkt_group not in self.sankey_data: self.sankey_data[cp.fkt_group] = pd.DataFrame( columns=sankey_columns_dtypes.keys(), index=pd.MultiIndex( levels=[[], []], names=["target_group", "category"], codes=[[], []] ) ).astype(sankey_columns_dtypes) self.evaluate_busses(cp) # create a table that includes exergy destruction attributed to the # components bus_based = self.bus_data[self.bus_data['base'] == 'bus'].index component_based = self.bus_data[ self.bus_data['base'] == 'component' ].index # add aggregated components with respective buses data self.aggregation_data = self.component_data.copy() # E_D is sum of both E_D self.aggregation_data.loc[self.bus_data.index, 'E_D'] = ( self.component_data.loc[self.bus_data.index, 'E_D'] + self.bus_data['E_D'] ) # E_F for bus based components is higher by E_D of bus self.aggregation_data.loc[bus_based, 'E_F'] += ( self.bus_data.loc[bus_based, 'E_D'] ) # E_P of component based components is lower by E_D of bus self.aggregation_data.loc[component_based, 'E_P'] -= ( self.bus_data.loc[component_based, 'E_D'] ) # calculate network results self.network_data.loc['E_D'] = ( self.component_data['E_D'].sum() + self.bus_data['E_D'].sum()) self.network_data.loc['E_F'] = abs(self.network_data.loc['E_F']) self.network_data.loc['E_P'] = abs(self.network_data.loc['E_P']) self.network_data.loc['epsilon'] = ( self.network_data.loc['E_P'] / self.network_data.loc['E_F'] ) # calculate exergy destruction ratios for components/busses E_F = self.network_data.loc['E_F'] E_D = self.network_data.loc['E_D'] for d in [self.component_data, self.bus_data, self.aggregation_data]: d['y_Dk'] = d['E_D'] / E_F d['y*_Dk'] = d['E_D'] / E_D d['epsilon'] = d['E_P'] / d['E_F'] residual = abs( self.network_data.loc['E_F'] - self.network_data.loc['E_P'] - self.network_data.loc['E_D'] - self.network_data.loc['E_L'] ) if residual >= ERR ** 0.5: msg = ( 'The exergy balance of your network is not closed (residual ' 'value is ' + str(round(residual, 6)) + ', but should be ' 'smaller than ' + str(ERR ** 0.5) + '), you should check the ' 'component and network exergy data and check, if network is ' 'properly setup for the exergy analysis.') logger.error(msg) self.create_group_data()
[docs] def evaluate_busses(self, cp): """Evaluate the exergy balances of busses. Parameters ---------- cp : tespy.components.component.Component Component to analyse the bus exergy balance of. """ cp_on_num_busses = 0 for b in self.E_F + self.E_P + self.internal_busses + self.E_L: if cp in b.comps.index: if cp_on_num_busses > 0: msg = ( 'The component ' + cp.label + ' is on multiple ' 'busses in the exergy analysis. Make sure that no ' 'component is connected to more than one of the ' 'busses passed to the exergy_analysis method.') logger.error(msg) raise hlp.TESPyNetworkError(msg) # todo: E_bus als dict mit den versch. werten if b.comps.loc[cp, 'base'] == 'bus': E_bus = sum(e for e in cp.E_bus.values() if e) self.bus_data.loc[cp.label, 'E_P'] = E_bus bus_efficiency = cp.calc_bus_efficiency(b) E_F = E_bus / bus_efficiency self.bus_data.loc[cp.label, 'E_F'] = E_F if b in self.E_F: self.network_data.loc['E_F'] += E_F elif b in self.E_P: self.network_data.loc['E_P'] -= E_F elif b in self.E_L: self.network_data.loc['E_L'] -= E_F for key, value in cp.E_bus.items(): if value == 0: continue if key != "massless": # this should be a source category = categorize_fluids(cp.outl[0]) else: category = "work" if (cp.fkt_group, category) not in self.sankey_data[b.label].index: self.sankey_data[b.label].loc[(cp.fkt_group, category), :] = 0 self.sankey_data[b.label].loc[(cp.fkt_group, category), key] += value / bus_efficiency else: E_bus = sum(e for e in cp.E_bus.values() if e) bus_efficiency = cp.calc_bus_efficiency(b) E_P = E_bus * bus_efficiency self.bus_data.loc[cp.label, 'E_P'] = E_P self.bus_data.loc[cp.label, 'E_F'] = E_bus if b in self.E_F: self.network_data.loc['E_F'] -= E_P elif b in self.E_P: self.network_data.loc['E_P'] += E_P elif b in self.E_L: self.network_data.loc['E_L'] += E_P for key, value in cp.E_bus.items(): if value == 0: continue if key != "massless": # this should be a sink category = categorize_fluids(cp.inl[0]) else: category = "work" if (b.label, category) not in self.sankey_data[cp.fkt_group].index: self.sankey_data[cp.fkt_group].loc[(b.label, category), :] = 0 self.sankey_data[cp.fkt_group].loc[(b.label, category), key] += value * bus_efficiency self.bus_data.loc[cp.label, 'base'] = b.comps.loc[cp, 'base'] self.bus_data.loc[cp.label, 'group'] = cp.fkt_group cp_on_num_busses += 1 self.bus_data['E_D'] = self.bus_data['E_F'] - self.bus_data['E_P']
[docs] def create_group_data(self): """Collect the component group exergy data.""" for group in self.sankey_data.keys(): E_D = self.aggregation_data[ self.aggregation_data['group'] == group ]['E_D'].sum() self.sankey_data[group].loc[('E_D', "E_D"), :] = [0., 0., E_D] # establish connections for fuel exergy via bus balance for b in self.E_F: input_value = self.calculate_group_input_value(b.label) self.sankey_data['E_F'].loc[(b.label, "E_F"), self.exergy_cats] = ( self.sankey_data[b.label].loc[:, self.exergy_cats].sum() - input_value.sum() ) # establish connections for product exergy via bus balance for b in self.E_P: input_value = self.calculate_group_input_value(b.label) self.sankey_data[b.label].loc[('E_P', "E_P"), self.exergy_cats] = ( input_value.sum() - self.sankey_data[b.label].loc[:, self.exergy_cats].sum() ) # establish connections for exergy loss via bus balance for b in self.E_L: input_value = self.calculate_group_input_value(b.label) self.sankey_data[b.label].loc[('E_L', 'E_L'), self.exergy_cats] = ( input_value.sum() - self.sankey_data[b.label].loc[:, self.exergy_cats].sum() ) for fkt_group, data in self.sankey_data.items(): mask = self.component_data['group'] == fkt_group comps = self.component_data.loc[mask].index for comp in comps: comp_obj = self.nw.get_comp(comp) sources = self.nw.conns[self.nw.conns['source'] == comp_obj] for conn in sources['object']: if conn.target.label not in comps: target_group = self.component_data.loc[conn.target.label, 'group'] target_value_chemical = ( conn.Ex_chemical if hasattr(conn, "Ex_chemical") else 0. ) target_value_physical = conn.Ex_physical cat = categorize_fluids(conn) if (target_group, cat) in data.index: self.sankey_data[fkt_group].loc[ (target_group, cat), 'physical'] += target_value_physical self.sankey_data[fkt_group].loc[ (target_group, cat), 'chemical'] += target_value_chemical else: self.sankey_data[fkt_group].loc[(target_group, cat), :] = [ target_value_chemical, target_value_physical, 0 ] # create overview of component groups self.group_data = pd.DataFrame( columns=['E_in', 'E_out', 'E_D'], dtype='float64' ) for fkt_group in self.component_data['group'].unique(): self.group_data.loc[fkt_group, 'E_in'] = ( self.calculate_group_input_value(fkt_group).sum().sum() ) self.group_data.loc[fkt_group, 'E_D'] = ( self.sankey_data[fkt_group].loc[idx['E_D', :], self.exergy_cats].sum().sum() ) # calculate missing values self.group_data['E_out'] = ( self.group_data['E_in'] - self.group_data['E_D']) self.group_data['y_Dk'] = ( self.group_data['E_D'] / self.network_data.loc['E_F']) self.group_data['y*_Dk'] = ( self.group_data['E_D'] / self.network_data.loc['E_D']) # ToDo: Transform this into a test # assert self.group_data['E_D'].sum() == self.network_data["E_D"] return
[docs] def calculate_group_input_value(self, group_label): """Calculate the total exergy input of a component group.""" value = pd.DataFrame( columns=self.exergy_cats, index=[0], data=[[0., 0., 0.]] ) for fkt_group, data in self.sankey_data.items(): if group_label in data.index: value += data.loc[idx[group_label, :], self.exergy_cats].values return value
[docs] def single_group_input(self, group_label, group_data): """Calculate the total exergy input of a component group.""" inputs = [] for fkt_group, data in group_data.items(): if group_label in data.index.get_level_values("target_group") and fkt_group != group_label: inputs += [fkt_group] return inputs
[docs] def remove_transit_groups(self, group_data): """Remove transit only component groups from sankey display. Method is recursively called if a group was removed from display to catch cases, where multiple groups are attached in line without any streams leaving the line. Parameters ---------- group_data : dict Dictionary containing the modified component group data. """ for fkt_group in group_data.copy().keys(): if fkt_group in self.reserved_fkt_groups: continue source_groups = self.single_group_input(fkt_group, group_data) if len(source_groups) == 1 and len(group_data[fkt_group].index.get_level_values("target_group").unique()) == 1: source_group = source_groups[0] group_data[source_group] = group_data[source_group].add(group_data[fkt_group], fill_value=0) to_drop = group_data[source_group].index.get_level_values("target_group") == fkt_group group_data[source_group] = group_data[source_group].loc[~to_drop] del group_data[fkt_group] # recursive call in case multiple components are attached in # a line without any conversion self.remove_transit_groups(group_data) # exit to stop further iterations inside the groups return # remove groups without any connection elif len(source_groups) == 0 and len(group_data[fkt_group]) == 0: del group_data[fkt_group]
[docs] def generate_plotly_sankey_input( self, node_order=[], colors={}, display_thresold=1e-3, disaggregate_flows=False): """Generate input data for sankey plots. Only exergy flow above the display threshold is included. All component groups with transit only (one input and one output) are cut out of the display. Parameters ---------- node_order : list Order for the nodes in the sankey diagram (optional). In case no order is passed, a generic order will be used. colors : dict Dictionary containing a color for every stream type (optional). Stream type is the key, the color is the corresponding value. Stream types are: - :code:`E_F`, :code:`E_P`, :code:`E_L`, :code:`E_D` - names of the pure fluids of the tespy Network - :code:`mix` (used in case of any gas mixture) - labels of internal busses display_threshold : float Minimum value of flows to be displayed in the diagram, defaults to 1e-3. disaggregate_flows : boolean Separate every flow by chemical, physical and massless exergy, defaults to False. Returns ------- tuple Tuple containing the links and node_order for the plotly sankey diagram. """ group_data = self.sankey_data.copy() cols = self.exergy_cats for fkt_group, data in self.sankey_data.items(): mask = data.loc[:, cols].abs().sum(axis=1) >= display_thresold group_data[fkt_group] = group_data[fkt_group].loc[mask] self.remove_transit_groups(group_data) if len(node_order) == 0: node_order = ( ['E_F'] + [b.label for b in self.E_F] + [fkt_group for fkt_group in self.group_data.index] + [b.label for b in self.internal_busses + self.E_P + self.E_L] + ['E_P', 'E_L', 'E_D'] ) else: missing = [] for node in group_data.keys(): if node not in node_order: missing += [node] if len(missing) > 0: msg = ( 'The list of nodes passed is missing the following ' 'nodes: "' + '", "'.join(missing) + '".') logger.error(msg) raise ValueError(msg) colordict = { "E_F": "rgba(242, 142, 43, 0.90)", "E_P": "rgba(118, 183, 178, 0.90)", "E_D": "rgba(176, 122, 161, 0.90)", "E_L": "rgba(156, 117, 95, 0.90)", "combustion-gas": "rgba(237, 201, 72, 0.90)", "non-combustion-gas": "rgba(186, 176, 172, 0.90)", "two-phase-fluid": "rgba(89, 161, 79, 0.90)", "incompressible": "rgba(255, 157, 167, 0.90)", "work": "rgba(78, 121, 167, 0.90)", "heat": "rgba(225, 87, 89, 0.90)", np.nan: "rgba(100, 100, 100, 1.00)" } colordict.update(colors) links = { 'source': [], 'target': [], 'value': [], 'color': [] } for fkt_group, data in group_data.items(): source_id = node_order.index(fkt_group) for target in data.index: for col in cols: # how to aggregate here? if data.loc[target, col] > 0.: links['source'] += [source_id] links['target'] += [node_order.index(target[0])] links['value'] += [data.loc[target, col]] links['color'].append(colordict[target[1]]) return links, node_order
[docs] def print_results( self, sort_desc=True, busses=True, components=True, connections=True, groups=True, network=True, aggregation=True): r"""Print the results of the exergy analysis to prompt. - The results are sorted beginning with the component having the biggest exergy destruction by default. - Components with an exergy destruction smaller than 1000 W is not printed to prompt by default. Parameters ---------- sort_des : boolean Sort the component results descending by exergy destruction. busses : boolean Print bus results, default value :code:`True`. components : boolean Print component results, default value :code:`True`. connections : boolean Print connection results, default value :code:`True`. network : boolean Print network results, default value :code:`True`. aggregation : boolean Print aggregated component results, default value :code:`True`. """ if connections: print('##### RESULTS: Connection specific physical exergy and ' + 'chemical exergy #####') print(tabulate( self.connection_data, headers='keys', tablefmt='psql', floatfmt='.3e')) if components: df = self.component_data.copy() df = df.loc[:, df.columns != 'group'] if sort_desc: df.sort_values(by=['E_D'], ascending=False, inplace=True) print('##### RESULTS: Component exergy analysis #####') print(tabulate( df, headers='keys', tablefmt='psql', floatfmt='.3e')) if busses: df = self.bus_data.copy() df = df.loc[:, (df.columns != 'group') & (df.columns != 'base')] if sort_desc: df.sort_values(by=['E_D'], ascending=False, inplace=True) print('##### RESULTS: Bus exergy analysis #####') print(tabulate( df, headers='keys', tablefmt='psql', floatfmt='.3e')) if aggregation: df = self.aggregation_data.copy() df = df.loc[:, df.columns != 'group'] if sort_desc: df.sort_values(by=['E_D'], ascending=False, inplace=True) print('##### RESULTS: Aggregation of components and busses #####') print(tabulate( df, headers='keys', tablefmt='psql', floatfmt='.3e')) if network: print('##### RESULTS: Network exergy analysis #####') print(tabulate( self.network_data.to_frame().transpose(), headers='keys', tablefmt='psql', floatfmt='.3e', showindex=False)) if groups: df = self.group_data.copy() if sort_desc: df.sort_values(by=['E_D'], ascending=False, inplace=True) print('##### RESULTS: Functional groups exergy flows #####') print(tabulate( df, headers='keys', tablefmt='psql', floatfmt='.3e'))