Network(
import_name: str | Path = "",
name: str = "Unnamed Network",
ignore_standard_types: bool = False,
**kwargs: Any,
)
Bases: NetworkComponentsMixin, NetworkDescriptorsMixin, NetworkTransformMixin, NetworkIndexMixin, NetworkConsistencyMixin, NetworkGraphMixin, NetworkPowerFlowMixin, NetworkIOMixin
Network container for all pypsa.Components and helds most of PyPSA's functionality.
Parameters:
-
import_name(string or Path, default:'') –Path to netCDF file, HDF5 .h5 store or folder of CSV files from which to import network data. The string could be a URL. If cloudpathlib is installed, the string could be a object storage URI with an
s3,gsorazURI scheme. -
name(string, default:'Unnamed Network') –Network name.
-
ignore_standard_types(boolean, default:False) –If True, do not read in PyPSA standard types into standard types DataFrames.
-
kwargs(Any, default:{}) –Any remaining attributes to set
Examples:
>>> nw1 = pypsa.Network("network.nc")
>>> nw2 = pypsa.Network("/my/folder")
>>> nw3 = pypsa.Network("https://github.com/PyPSA/PyPSA/raw/master/examples/scigrid-de/scigrid-with-load-gen-trafos.nc")
>>> nw4 = pypsa.Network("s3://my-bucket/my-network.nc")
Methods:
-
__str__–Get a string representation of the network.
-
__repr__–Get representation of the network.
-
__add__–Merge all components of two networks.
-
__eq__–Check for equality of two networks.
-
__getitem__–Return a shallow slice of the Network object.
-
equals–Check for equality of two networks.
-
to_crs–Convert the network's geometries and bus coordinates to a new crs.
-
explore–Alias for pypsa.Network.plot.explore.
-
copy–Return a deep copy of Network object.
-
branches–Get branches.
-
passive_branches–Get passive branches.
-
controllable_branches–Get controllable branches.
-
determine_network_topology–Build sub_networks from topology.
-
cycle_matrix–Get the cycles in the network and represent them as a DataFrame.
-
component–Get a component from the network.
-
iterate_components–Iterate over components.
Attributes:
-
optimize(OptimizationAccessor) –Network optimization functionality accessor.
-
cluster(ClusteringAccessor) –Network clustering functionality accessor.
-
statistics(StatisticsAccessor) –Network statistics functionality accessor.
-
plot(PlotAccessor) –Network plotting functionality accessor.
-
name(str) –Name of the network.
-
pypsa_version(str) –PyPSA version of the network.
-
meta(dict) –Dictionary of the network meta data.
-
model(Model) –Access to linopy model object.
-
objective(float | None) –Objective value of the solved network.
-
objective_constant(float | None) –Objective constant of the network.
-
is_solved(bool) –Check if the network has been solved.
-
crs(Any) –Coordinate reference system of the network's geometries.
-
srid(int) –Spatial reference system identifier of the network's geometries.
instance-attribute
¶
optimize: OptimizationAccessor = OptimizationAccessor(self)
Network optimization functionality accessor.
instance-attribute
¶
cluster: ClusteringAccessor = ClusteringAccessor(self)
Network clustering functionality accessor.
instance-attribute
¶
statistics: StatisticsAccessor = StatisticsAccessor(self)
Network statistics functionality accessor.
instance-attribute
¶
plot: PlotAccessor = PlotAccessor(self)
Network plotting functionality accessor.
property
writable
¶
name: str
Name of the network.
The name is set when the network is created. It can also be changed by setting
the name attribute. It is only descriptive and not used for any
functionality.
Examples:
>>> n.name
'AC-DC-Meshed'
>>> n = pypsa.Network(name='Unnamed Network')
>>> n.name
'Unnamed Network'
>>> n.name = 'net'
>>> n.name
'net'
property
¶
pypsa_version: str
PyPSA version of the network.
The PyPSA version is set when the network is created and cannot be changed manually. When a network of an older version is imported, the version is automatically updated to the current version.
Examples:
>>> n = pypsa.Network()
>>> n.pypsa_version
'1.0.0'
property
writable
¶
meta: dict
Dictionary of the network meta data.
Any additional meta data can be added to the network by setting the meta
attribute. Meta data will be saved on export.
Examples:
>>> n = pypsa.Network()
>>> n.meta['description'] = 'This is a test network'
>>> n.meta['any_key'] = 'Any Key can be added'
>>> n.meta
{'description': 'This is a test network', 'any_key': 'Any Key can be added'}
deletable
property
¶
model: Model
Access to linopy model object.
After optimizing a network, the linopy model object is stored in the network and can be accessed via this property. It cannot be set manually.
Examples:
>>> n.model
Linopy LP model
===============
Variables:
----------
* Generator-p_nom (name)
* Line-s_nom (name)
* Link-p_nom (name)
* Generator-p (snapshot, name)
* Line-s (snapshot, name)
* Link-p (snapshot, name)
* objective_constant
Constraints:
------------
* Generator-ext-p_nom-lower (name)
* Line-ext-s_nom-lower (name)
* Link-ext-p_nom-lower (name)
* Generator-ext-p-lower (snapshot, name)
* Generator-ext-p-upper (snapshot, name)
* Line-ext-s-lower (snapshot, name)
* Line-ext-s-upper (snapshot, name)
* Link-ext-p-lower (snapshot, name)
* Link-ext-p-upper (snapshot, name)
* Bus-nodal_balance (name, snapshot)
* Kirchhoff-Voltage-Law (snapshot, cycle)
* GlobalConstraint-co2_limit
Status:
-------
ok
property
¶
objective: float | None
Objective value of the solved network.
The property yields the objective value of the solved network. It is set after
optimizing the network points to the linopy solution (e.g. is an alias for
n.model.objective.value). When loading a network from file and the model
object is not loaded, the objective value is still available, as it is stored
in the network object.
When optimizing for system costs, the total system costs are the sum of the pypsa.Network.objective and the pypsa.Network.objective_constant.
Examples:
>>> n.objective
-47274166...
>>> n.objective + n.objective_constant
<xarray.DataArray ()> Size: 8B
array(18441021...)
property
¶
objective_constant: float | None
Objective constant of the network.
The property yields the fixed part of the objective function. It is set after optimizing the network.
When optimizing for system costs, the total system costs are the sum of the pypsa.Network.objective and the pypsa.Network.objective_constant. When loading a network from file and the model object is not loaded, the objective constant is still available, as it is stored in the network object.
Examples:
>>> n.objective_constant
<xarray.DataArray ()> Size: 8B
array(65715187...)
>>> n.objective + n.objective_constant
<xarray.DataArray ()> Size: 8B
array(18441021...)
property
¶
is_solved: bool
property
writable
¶
crs: Any
Coordinate reference system of the network's geometries.
Examples:
>>> n.crs
<Geographic 2D CRS: EPSG:4326>
Name: WGS 84
Axis Info [ellipsoidal]:
- Lat[north]: Geodetic latitude (degree)
- Lon[east]: Geodetic longitude (degree)
Area of Use:
- name: World.
- bounds: (-180.0, -90.0, 180.0, 90.0)
Datum: World Geodetic System 1984 ensemble
- Ellipsoid: WGS 84
- Prime Meridian: Greenwich
property
writable
¶
srid: int
Spatial reference system identifier of the network's geometries.
Examples:
>>> n.srid
4326
See Also
__str__() -> str
Get a string representation of the network.
Examples:
>>> str(n)
"PyPSA Network 'AC-DC-Meshed'"
>>> str(n_stochastic)
"Stochastic PyPSA Network 'Stochastic-Network'"
__repr__() -> str
Get representation of the network.
Examples:
>>> n
PyPSA Network 'AC-DC-Meshed'
----------------------------
Components:
- Bus: 9
- Carrier: ...
- Generator: ...
- GlobalConstraint: ...
- Line: ...
- Link: ...
- Load: ...
- SubNetwork: ...
Snapshots: 10
__add__(other: Network) -> None
Merge all components of two networks.
Parameters:
-
other(Network) –Network to merge into this one.
See Also
Examples:
>>> n1 = pypsa.Network()
>>> n2 = pypsa.Network()
>>> n1.add("Bus", "bus1")
>>> n2.add("Bus", "bus2")
>>> new_network = n1 + n2
>>> len(new_network.buses)
2
__eq__(other: object) -> bool
Check for equality of two networks.
Parameters:
-
other(Any) –The other network to compare with.
Returns:
-
bool–True if the networks are equal, False otherwise.
See Also
__getitem__(key: str) -> Network
Return a shallow slice of the Network object.
A network can be sliced in three different ways: 1. If the key is a scenario name and the network has scenarios, the corresponding scenario is returned as a new network object. See pypsa.Network.get_scenario. 2. If the network is a collection and the key is a name of one of the networks in the collection, that network is returned. 3. If the key is a bus name or a boolean indexer for buses, a sliced copy of the network is returned, containing only the selected buses and all connected components. See pypsa.Network.slice_network.
A warning will be issued if the key matches multiple of the above entities. In that case, the first match is returned. But it is recommended to use the explicit methods (e.g. get_scenario(), get_network()) or use unique scenario, collection and bus names to avoid ambiguity.
Parameters:
-
key(str or boolean mask) –The key or boolean mask to select a scenario, a network from a collection or slice the network based on buses.
Returns:
-
n(Network) –
Examples:
Select single scenario from a stochastic network:
>>> n_stochastic
Stochastic PyPSA Network 'Stochastic-Network'
---------------------------------------------
Components:
- Bus: 3
- Generator: 12
- Load: 3
Snapshots: 2920
Scenarios: 3
>>> n_stochastic["high"]
PyPSA Network 'Stochastic-Network - Scenario 'high''
----------------------------------------------------
Components:
- Bus: 1
- Generator: 4
- Load: 1
Snapshots: 2920
Select single collection from a network collection:
>>> nc
NetworkCollection
-----------------
Networks: 2
Index name: 'network'
Entries: ['AC-DC-Meshed', 'AC-DC-Meshed-Shuffled-Load']
>>> nc["AC-DC-Meshed"]
PyPSA Network 'AC-DC-Meshed'
----------------------------
Components:
- Bus: 9
- Carrier: 6
- Generator: 6
- GlobalConstraint: 1
- Line: 7
- Link: 4
- Load: 6
- SubNetwork: 3
Snapshots: 10
Select a network slice based on buses:
>>> n
PyPSA Network 'AC-DC-Meshed'
----------------------------
Components:
- Bus: 9
- Carrier: 6
- Generator: 6
- GlobalConstraint: 1
- Line: 7
- Link: 4
- Load: 6
- SubNetwork: 3
Snapshots: 10
>>> n["London"]
PyPSA Network 'AC-DC-Meshed'
----------------------------
Components:
- Bus: 1
- Carrier: 6
- GlobalConstraint: 1
- Load: 1
Snapshots: 10
or use the pandas .loc method to select multiple buses:
>>> n[n.buses.carrier=='AC']
PyPSA Network 'AC-DC-Meshed'
----------------------------
Components:
- Bus: 6
- Carrier: 6
- Generator: 6
- GlobalConstraint: 1
- Line: 4
- Link: 1
- Load: 6
Snapshots: 10
equals(other: Any, log_mode: str = 'silent') -> bool
Check for equality of two networks.
Parameters:
-
other(Any) –The other network to compare with.
-
log_mode(str, default:'silent') –Controls how differences are reported: - 'silent': No logging, just returns True/False - 'verbose': Prints differences but doesn't raise errors - 'strict': Raises ValueError on first difference
Raises:
-
ValueError–If log_mode is 'strict' and components are not equal.
Returns:
-
bool–True if the networks are equal, False otherwise.
Examples:
>>> n1 = pypsa.Network()
>>> n2 = pypsa.Network()
>>> n1.add("Bus", "bus1")
>>> n2.add("Bus", "bus2")
>>> n1.equals(n2)
False
to_crs(new: int | str | CRS) -> None
Convert the network's geometries and bus coordinates to a new crs.
explore(*args: Any, **kwargs: Any) -> Any
Alias for pypsa.Network.plot.explore.
copy(
snapshots: Sequence | None = None,
investment_periods: Sequence | None = None,
ignore_standard_types: bool = False,
) -> Network
Return a deep copy of Network object.
If only default arguments are passed, the copy will be created via
copy.deepcopy and will contain all components and time-varying data.
For most networks this is the fastest way. However, if the network is very
large, it might be better to filter snapshots and investment periods to reduce
the size of the copy. In this case copy.deepcopy is not used and only
the selected snapshots and investment periods are copied to a new object.
Parameters:
-
snapshots(list or tuple or Index, default:self.snapshots) –A list of snapshots to copy, must be a subset of n.snapshots. Pass an empty list ignore all snapshots.
-
investment_periods(list or tuple or Index, default:self.investment_period_weightings.index) –A list of investment periods to copy, must be a subset of n.investment_periods. Pass
-
ignore_standard_types(boolean, default:False) –Ignore the PyPSA standard types.
Returns:
-
n(Network) –The copied network object.
Examples:
With a simple reference the network is not copied:
>>> n = pypsa.examples.ac_dc_meshed()
>>> network_copy = n
>>> id(network_copy) == id(n)
True
Use the copy method to create a new network object:
>>> network_copy = n.copy()
>>> id(network_copy) == id(n)
False
You can also filter on a subset of snapshots (or investment periods):
>>> n.snapshots
DatetimeIndex(['2015-01-01 00:00:00', '2015-01-01 01:00:00',
'2015-01-01 02:00:00', '2015-01-01 03:00:00',
'2015-01-01 04:00:00', '2015-01-01 05:00:00',
'2015-01-01 06:00:00', '2015-01-01 07:00:00',
'2015-01-01 08:00:00', '2015-01-01 09:00:00'],
dtype='datetime64[ns]', name='snapshot', freq=None)
>>> network_copy = n.copy(snapshots=n.snapshots[0])
>>> network_copy.snapshots
DatetimeIndex(['2015-01-01'], dtype='datetime64[ns]', name='snapshot', freq=None)
branches() -> DataFrame
Get branches.
Branches are Lines, Links and Transformers.
Note
This method will return a merged copy of all branches of the network. Changes to the returned DataFrame will not be reflected in the network.
Examples:
>>> n.branches()
active b b_pu ... x x_pu x_pu_eff
component name ...
Line 0 True 0.0 0.0 ... 0.796878 0.000006 0.000006
1 True 0.0 0.0 ... 0.391560 0.000003 0.000003
2 True 0.0 0.0 ... 0.000000 0.000000 0.000000
3 True 0.0 0.0 ... 0.000000 0.000000 0.000000
4 True 0.0 0.0 ... 0.000000 0.000000 0.000000
5 True 0.0 0.0 ... 0.238800 0.000002 0.000002
6 True 0.0 0.0 ... 0.400000 0.000003 0.000003
Link Norwich Converter True NaN NaN ... NaN NaN NaN
Norway Converter True NaN NaN ... NaN NaN NaN
Bremen Converter True NaN NaN ... NaN NaN NaN
DC link True NaN NaN ... NaN NaN NaN
passive_branches() -> DataFrame
Get passive branches.
Passive branches are Lines and Transformers.
Note
This method will return a merged copy of all passive branches of the network. Changes to the returned DataFrame will not be reflected in the network.
Examples:
>>> n.passive_branches()
active b b_pu ... x x_pu x_pu_eff
component ...
0 True 0.0 0.0 ... 0.796878 0.000006 0.000006
1 True 0.0 0.0 ... 0.391560 0.000003 0.000003
2 True 0.0 0.0 ... 0.000000 0.000000 0.000000
3 True 0.0 0.0 ... 0.000000 0.000000 0.000000
4 True 0.0 0.0 ... 0.000000 0.000000 0.000000
5 True 0.0 0.0 ... 0.238800 0.000002 0.000002
6 True 0.0 0.0 ... 0.400000 0.000003 0.000003
[7 rows x 38 columns]
controllable_branches() -> DataFrame
Get controllable branches.
Controllable branches are Links.
Note
This method will return a merged copy of all controllable branches of the network. Changes to the returned DataFrame will not be reflected in the network.
Examples:
>>> n.controllable_branches()
active b b_pu ... x x_pu x_pu_eff
component name ...
Line 0 True 0.0 0.0 ... 0.796878 0.000006 0.000006
1 True 0.0 0.0 ... 0.391560 0.000003 0.000003
2 True 0.0 0.0 ... 0.000000 0.000000 0.000000
3 True 0.0 0.0 ... 0.000000 0.000000 0.000000
4 True 0.0 0.0 ... 0.000000 0.000000 0.000000
5 True 0.0 0.0 ... 0.238800 0.000002 0.000002
6 True 0.0 0.0 ... 0.400000 0.000003 0.000003
[7 rows x 38 columns]
determine_network_topology(
investment_period: int | str | None = None,
skip_isolated_buses: bool = False,
) -> Network
Build sub_networks from topology.
For the default case investment_period=None, it is not taken into account whether the branch components are active (based on build_year and lifetime). If the investment_period is specified, the network topology is determined on the basis of the active branches.
cycle_matrix(
investment_period: str | int | None = None,
apply_weights: bool = False,
) -> DataFrame
Get the cycles in the network and represent them as a DataFrame.
This function identifies a cycle basis of the network topology and returns a DataFrame representation of the cycle matrix. The cycles matrix is a sparse matrix with branches as rows and independent cycles as columns. An entry of +1 indicates the branch is traversed in the direction from bus0 to bus1 in that cycle, -1 indicates the opposite direction, and 0 indicates the branch is not part of the cycle.
Parameters:
-
investment_period(str or int, default:None) –Investment period to use when determining network topology. If not given, all branches are considered regardless of build_year and lifetime.
-
apply_weights(bool, default:False) –Whether to apply weights (e.g., reactance for AC lines, resistance for DC lines) to the cycles.
Returns:
-
DataFrame–A DataFrame with branches as rows (MultiIndex of (component, name)) and cycles as columns. Each column represents an independent cycle in the network.
component(c_name: str) -> Component
Get a component from the network.
Deprecated in v1.0.0
Use n.components.<component> or n.components[component_name] instead.
iterate_components(
components: Collection[str] | None = None,
skip_empty: bool = True,
) -> Iterator[Component]
Iterate over components.
Deprecated in v1.0.0
Use for component in n.components instead.
Examples:
>>> for component in n.iterate_components():
... print(component)
... break
'Bus' Components