CGMES2PGM-Suite

cgmes2pgm_suite provides additional tools for cgmes2pgm_converter to integrate PowerGridModel with the Common Grid Model Exchange Standard (CGMES). It focuses on performing the state estimation on CGMES datasets.

Project Status

The SOPTIM PGM Converter is provided by SOPTIM AG.

This repository is intentionally kept small and primarily serves demonstration, evaluation, and reproducibility purposes. While SOPTIM AG actively develops solutions in the area of power system analysis and grid data integration, this repository is not intended to represent the full functionality, performance, or support offering of commercial SOPTIM products.

The code is made available under the Apache-2.0 license. The availability of this repository does not imply any warranty, maintenance obligation, support commitment, or product roadmap by SOPTIM AG.

For additional information please contact SOPTIM AG directly.

Features

• Start an Apache Jena Fuseki as docker container
• Upload Datasets to a SPARQL endpoint
• Human readable exports of PGM Datasets in TXT and Excel
• Create SV-Profile from PGM Results
• Debug state estimation by manipulating datasets (e.g., subnet splitting)
• Configure conversion and state estimation via a configuration file
• Simulate measurements:
  • when real measurements are not provided via an Operation Profile, but a State Variable (SV) Profile is available
  • generates an Operation Profile with distorted measurements based on the SV Profile

Installation

The package can be installed via pip:

pip install cgmes2pgm_suite

To start an Apache Jena Fuseki server via this package, Docker is required. See Docker installation guide.

Usage

This package can be run as a standalone application, performing the conversion and running PGM's state estimation. To do so, you need to install the package and then run the following command:

python -m cgmes2pgm_suite --config <path_to_config_file>

The provided configuration file contains the dataset configuration and the parameters for the conversion and state estimation. An example configuration file can be found in /example.

Quick Start

For a quick start, we recommend cloning this project and using the provided test cases.

If the project is cloned, setup the environment and install the package:

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
pip install -e ".[dev]"
pre-commit install

To run the Conformity Datasets, you need to download Test Configurations v3.0.3 from ENTSO-E CIM Conformity and Interoperability respecting their License. Place the rdf/xml files of each dataset in the respective subdirectory of tests/datasets.

Afterwards, you can run all datasets using:

pytest -m "integration"

Note: Running the tests creates a Fuseki Docker container on port 3030. The container is automatically removed after the tests are finished.

The results of the tests can be found in the tests/out directory.

If you want to add your own datasets, the following steps are required:

• Place the rdf/xml files in the tests/data/ directory
• Create a configuration file in the tests/configs/ directory, by copying an existing one
• Update the name, output directory and location of the rdf/xml files in the new configuration file

Datasets

The conversion, measurement simulation and state estimation has been tested with the CGMES conformity datasets.

The following datasets have been tested:

Dataset	Size (Nodes)	Estimation Result	Comment
PowerFlow	2	🟢
PST	2	🟢	All three Scenarios
MiniGrid	13	🟢
MicroGrid	13	🟢	PST with AsymmetricalPhaseTapChanger (BE-TR2_2) has been split
SmallGrid	167	🟢
Svedala	191	🟢
RealGrid	6051	🟡	Requires smaller sigmas in measurement simulation to converge
FullGrid	26	?	SV-Profile does not contain power flows for all branches, resulting in an insufficient amount of simulated measurements

Dataset Version: CGMES Conformity Assessment Scheme Test Configurations v3.0.2

The used configuration files can be found in the /tests/configs directory.

License

This project is licensed under the Apache License 2.0.

Dependencies

This project includes third-party dependencies, which are licensed under their own respective licenses.

• cgmes2pgm_converter (Apache License 2.0)
• bidict (Mozilla Public License 2.0)
• numpy (BSD License)
• pandas (BSD License)
• power-grid-model (Mozilla Public License 2.0)
• power-grid-model-io (Mozilla Public License 2.0)
• SPARQLWrapper (W3C License)
• XlsxWriter (BSD License)
• PyYAML (MIT License)
• StrEnum (MIT License)
• docker (Apache License 2.0)

This project includes code from jena-fuseki-docker in the src/cgmes2pgm_suite/resources/docker directory, which is redistributed under the original Apache License 2.0. See the root‑level NOTICE file for full attribution.