Implement blueprint approach for initializing cim data structure in c…

…im converter (e2nIEE#2269) * implemented blueprint approach for initializing cim data structure * add support for reading cgmes v3.0 files * add entries to changelog
mrifraunhofer · Apr 17, 2024 · 7fd9930 · 7fd9930
1 parent a4f1d90
commit 7fd9930
Show file tree

Hide file tree

Showing 2 changed files with 55 additions and 30 deletions.
diff --git a/CHANGELOG.rst b/CHANGELOG.rst
@@ -6,11 +6,13 @@ Change Log
 - [ADDED] switch results p and q
 - [ADDED] PowerFactory converter: option to export lines with sections as a single line with averaged-out impedance, or export line sections as separate individual lines
 - [ADDED] extend plotly function: add zoomlevel-parameter and hvdc lines
+- [ADDED] added support for reading cgmes v3.0 files
 - [CHANGED] plotting for voltage profile considers also gens that are slacks and only ext_grids and slack gens that are in service
 - [CHANGED] switched from setup.py to pyproject.toml
 - [CHANGED] updated upload_release.py to not call setup.py anymore (see https://packaging.python.org/en/latest/discussions/setup-py-deprecated/)
 - [CHANGED] updated upload_release.py to install the uploadad package and print the version
 - [CHANGED] updated MANIFEST.in to exclude the ci files from the wheel distribution
+- [CHANGED] cim data structure method in cim converter changed to blueprint approach
 - [FIXED] massive performance drag in large grids due to initializing Ybus for FACTS with np.zeros instead of using sparse matrix initialization
 - [FIXED] further futurewarnings and deprecation warnings
 

diff --git a/pandapower/converter/cim/cim_classes.py b/pandapower/converter/cim/cim_classes.py
@@ -8,6 +8,7 @@
 import re
 import tempfile
 import zipfile
+from types import MappingProxyType
 from typing import Dict, List
 import pandas as pd
 import numpy as np
@@ -19,12 +20,14 @@
 
 class CimParser:
 
- def __init__(self, cim: Dict[str, Dict[str, pd.DataFrame]] = None):
+ def __init__(self, cim: Dict[str, Dict[str, pd.DataFrame]] = None, cgmes_version: str = '2.14.5'):
  """
  This class parses CIM files and loads its content to a dictionary of
  CIM profile (dict) -> CIM element type (str) -> CIM elements (DataFrame)
  """
  self.logger = logging.getLogger(self.__class__.__name__)
+ self.cgmes_version = cgmes_version
+ self.__cim_blueprint = self._initialize_cim_data_structure(cgmes_version)
  self.cim: Dict[str, Dict[str, pd.DataFrame]] = cim if cim is not None else self.get_cim_data_structure()
  self.file_names: Dict[str, str] = dict()
  self.report_container = ReportContainer()
@@ -59,7 +62,7 @@ def parse_files(self, file_list: List[str] or str = None, encoding: str = 'utf-8
  self.set_cim_data_types()
  self.logger.info("Finished parsing CIM files.")
  self.report_container.add_log(Report(level=LogLevel.INFO, code=ReportCode.INFO_PARSING,
- message="CIM parser finished parsing CIM files."))
+  message="CIM parser finished parsing CIM files."))
  return self
 
  def set_cim_data_types(self) -> CimParser:
@@ -96,7 +99,8 @@ def set_cim_data_types(self) -> CimParser:
  self.logger.debug("Setting data type of %s from CIM element %s as type %s" %
  (col, cim_element_type, data_type_col_str))
  if col in default_values.keys(): # todo deprecated due to repair function?
- self.cim[profile][cim_element_type][col] = self.cim[profile][cim_element_type][col].fillna(value=default_values[col])
+ self.cim[profile][cim_element_type][col] = self.cim[profile][cim_element_type][col].fillna(
+ value=default_values[col])
  if data_type_col == bool_type:
  self.cim[profile][cim_element_type][col] = \
  self.cim[profile][cim_element_type][col].map(to_bool)
@@ -164,14 +168,14 @@ def prepare_cim_net(self) -> CimParser:
  message="CIM parser finished preparing the CIM data."))
  return self
 
- def get_cim_data_structure(self) -> Dict[str, Dict[str, pd.DataFrame]]:
+ def _initialize_cim16_data_structure(self) -> Dict[str, Dict[str, pd.DataFrame]]:
  """
- Get the cim data structure used by the converter.
+ Get the cim data structure used by the converter for cgmes version less than 3.
  :return Dict[str, Dict[str, pd.DataFrame]]: The cim data structure used by the converter.
  """
  self.logger.debug("Returning the CIM data structure.")
- return dict({
- 'eq': {
+ return MappingProxyType({
+ 'eq': MappingProxyType({
  'ControlArea': pd.DataFrame(columns=['rdfId', 'name', 'type']),
  'TieFlow': pd.DataFrame(columns=['rdfId', 'Terminal', 'ControlArea', 'positiveFlowIn']),
  'ConnectivityNode': pd.DataFrame(columns=['rdfId', 'name', 'description', 'ConnectivityNodeContainer']),
@@ -286,16 +290,16 @@ def get_cim_data_structure(self) -> Dict[str, Dict[str, pd.DataFrame]]:
  'AnalogValue': pd.DataFrame(columns=[
  'rdfId', 'name', 'sensorAccuracy', 'MeasurementValueSource', 'Analog', 'value']),
  'MeasurementValueSource': pd.DataFrame(columns=['rdfId', 'name'])
- },
- 'eq_bd': {
+ }),
+ 'eq_bd': MappingProxyType({
  'ConnectivityNode': pd.DataFrame(columns=['rdfId', 'name', 'ConnectivityNodeContainer']),
  'BaseVoltage': pd.DataFrame(columns=['rdfId', 'name', 'nominalVoltage']),
  'Terminal': pd.DataFrame(
  columns=['rdfId', 'ConnectivityNode', 'ConductingEquipment', 'sequenceNumber']),
  'EnergySource': pd.DataFrame(columns=['rdfId', 'name', 'nominalVoltage', 'EnergySchedulingType']),
  'EnergySchedulingType': pd.DataFrame(columns=['rdfId', 'name'])
- },
- 'ssh': {
+ }),
+ 'ssh': MappingProxyType({
  'ControlArea': pd.DataFrame(columns=['rdfId', 'netInterchange']),
  'ExternalNetworkInjection': pd.DataFrame(columns=[
  'rdfId', 'p', 'q', 'referencePriority', 'controlEnabled']),
@@ -331,36 +335,36 @@ def get_cim_data_structure(self) -> Dict[str, Dict[str, pd.DataFrame]]:
  'LinearShuntCompensator': pd.DataFrame(columns=['rdfId', 'controlEnabled', 'sections']),
  'NonlinearShuntCompensator': pd.DataFrame(columns=['rdfId', 'controlEnabled', 'sections']),
  'EquivalentInjection': pd.DataFrame(columns=['rdfId', 'regulationTarget', 'regulationStatus', 'p', 'q'])
- },
- 'sv': {
+ }),
+ 'sv': MappingProxyType({
  'SvVoltage': pd.DataFrame(columns=['rdfId', 'TopologicalNode', 'v', 'angle']),
  'SvPowerFlow': pd.DataFrame(columns=['rdfId', 'Terminal', 'p', 'q']),
  'SvShuntCompensatorSections': pd.DataFrame(columns=['rdfId', 'ShuntCompensator', 'sections']),
  'SvTapStep': pd.DataFrame(columns=['rdfId', 'TapChanger', 'position'])
- },
- 'tp': {
+ }),
+ 'tp': MappingProxyType({
  'TopologicalNode': pd.DataFrame(columns=[
  'rdfId', 'name', 'description', 'ConnectivityNodeContainer', 'BaseVoltage']),
  'DCTopologicalNode': pd.DataFrame(columns=['rdfId', 'name', 'DCEquipmentContainer']),
  'ConnectivityNode': pd.DataFrame(columns=['rdfId', 'TopologicalNode']),
  'Terminal': pd.DataFrame(columns=['rdfId', 'TopologicalNode']),
  'DCTerminal': pd.DataFrame(columns=['rdfId', 'DCTopologicalNode']),
  'ACDCConverterDCTerminal': pd.DataFrame(columns=['rdfId', 'DCTopologicalNode'])
- },
- 'tp_bd': {
+ }),
+ 'tp_bd': MappingProxyType({
  'TopologicalNode': pd.DataFrame(columns=['rdfId', 'name', 'ConnectivityNodeContainer', 'BaseVoltage']),
  'ConnectivityNode': pd.DataFrame(columns=['rdfId', 'TopologicalNode'])
- },
- 'dl': {
+ }),
+ 'dl': MappingProxyType({
  'Diagram': pd.DataFrame(columns=['rdfId', 'name']),
  'DiagramObject': pd.DataFrame(columns=['rdfId', 'IdentifiedObject', 'Diagram', 'name']),
  'DiagramObjectPoint': pd.DataFrame(columns=[
- 'rdfId', 'sequenceNumber', 'xPosition', 'yPosition', 'DiagramObject'])},
- 'gl': {
+ 'rdfId', 'sequenceNumber', 'xPosition', 'yPosition', 'DiagramObject'])}),
+ 'gl': MappingProxyType({
  'CoordinateSystem': pd.DataFrame(columns=['rdfId', 'name', 'crsUrn']),
  'Location': pd.DataFrame(columns=['rdfId', 'PowerSystemResources', 'CoordinateSystem']),
  'PositionPoint': pd.DataFrame(columns=['rdfId', 'Location', 'sequenceNumber', 'xPosition', 'yPosition'])
- }})
+ })})
 
  def _parse_element(self, element, parsed=None):
  if parsed is None:
@@ -424,19 +428,20 @@ def _get_cgmes_profile_from_xml(self, root: xml.etree.ElementTree.Element, ignor
  profile_list = [profile_list]
  for one_profile in profile_list:
  if '/EquipmentCore/' in one_profile or '/EquipmentOperation/' in one_profile or \
- '/EquipmentShortCircuit/' in one_profile:
+ '/EquipmentShortCircuit/' in one_profile or '/CoreEquipment-EU/' in one_profile:
  return 'eq'
- elif '/SteadyStateHypothesis/' in one_profile:
+ elif '/SteadyStateHypothesis/' in one_profile or '/SteadyStateHypothesis-EU/' in one_profile:
  return 'ssh'
- elif '/StateVariables/' in one_profile:
+ elif '/StateVariables/' in one_profile or '/StateVariables-EU/' in one_profile:
  return 'sv'
- elif '/Topology/' in one_profile:
+ elif '/Topology/' in one_profile or '/Topology-EU/' in one_profile:
  return 'tp'
- elif '/DiagramLayout/' in one_profile:
+ elif '/DiagramLayout/' in one_profile or '/DiagramLayout-EU/' in one_profile:
  return 'dl'
- elif '/GeographicalLocation/' in one_profile:
+ elif '/GeographicalLocation/' in one_profile or '/GeographicalLocation-EU/' in one_profile:
  return 'gl'
- elif '/EquipmentBoundary/' in one_profile or '/EquipmentBoundaryOperation/' in one_profile:
+ elif ('/EquipmentBoundary/' in one_profile or '/EquipmentBoundaryOperation/' in one_profile
+ or '/EquipmentBoundary-EU/' in one_profile):
  return 'eq_bd'
  elif '/TopologyBoundary/' in one_profile:
  return 'tp_bd'
@@ -512,7 +517,7 @@ def _parse_source_file(self, file: str, output: dict, encoding: str, profile_nam
  prf_content[element_type_c][col].values[prf_content[element_type_c][col].str.len().idxmax()]
  # remove the namespace from the literal
  prf_content[element_type_c][col] = \
- prf_content[element_type_c][col].str[name_space.rfind('.')+1:]
+ prf_content[element_type_c][col].str[name_space.rfind('.') + 1:]
  elif col_new.endswith('-about'):
  col_new = 'rdfId'
  prf_content[element_type_c][col] = prf_content[element_type_c][col].str[1:]
@@ -550,3 +555,21 @@ def get_file_names(self) -> Dict[str, str]:
 
  def get_report_container(self) -> ReportContainer:
  return self.report_container
+
+ def _initialize_cim_data_structure(self, cgmes_version: str) -> Dict[str, Dict[str, pd.DataFrame]]:
+ if cgmes_version == '2.14.5':
+ return self._initialize_cim16_data_structure()
+ if cgmes_version == '3.0':
+ return self._initialize_cim100_data_structure()
+ raise NotImplementedError(f"CGMES version {cgmes_version} is not supported.")
+
+ def _initialize_cim100_data_structure(self) -> Dict[str, Dict[str, pd.DataFrame]]:
+ raise NotImplementedError(f"CGMES version {3.0} is not supported yet.")
+
+ def get_cim_data_structure(self) -> Dict[str, Dict[str, pd.DataFrame]]:
+ cim_data_structure = {}
+ for one_profile, one_profile_dict in self.__cim_blueprint.items():
+ cim_data_structure[one_profile] = {}
+ for one_class, one_class_df in one_profile_dict.items():
+ cim_data_structure[one_profile][one_class] = one_class_df.copy(deep=True)
+ return cim_data_structure