Hello everyone,
the Common Information Model (CIM) is a data model for power systems promoted by entso-e. LinkML is a powerful python tool to create data schemas and export them to various different formats.
I know that some initiatives already exist to use linkml in the energy domain, especially for CIM, like here or here.
Do any of you work in the same direction? Are there people or projects that use LinkML in the energy domain?
Hi @FloKo
This probably isn’t the answer you had in mind, but there are legal considerations too.
The Common Information Model (CIM) (IEC 2020, Wikipedia EN, ENTSO‑E) is proprietary and closed access — with a cover price of nearly €500. So not surprisingly, I could not find a copy to download free‑of‑charge.
I would be careful about using CIM in open science myself.
It is possible that any copyright in CIM could transfer to your codebase and databank (here I use databank as a blanket term to describe your research data management system).
Creative Commons lists four aspects that are potentially under copyright in this general context:
Specifically:
In this 2023 presentation, I traverse some of the issues as I see them:
Indeed, I think just the extensive use of CIM field names could render your model framework and/or supporting databank derivative works of CIM under copyright law. Not an issue for in‑house use, but one which could become problematic when you later try to publish your source code and/or datasets under open licenses.
As mentioned earlier, the cover price for CIM is substantial.
The ECJ judgment in case C-588/21 P, also known as the “Malamud” case, established that standards that are an integral part of EU law must be freely accessible to citizens. This means at zero cost but not necessarily open licensed (Wikipedia EN, Mendis and Kanevskaia 2024).
It is unclear to me whether the use of CIM by ENTSO‑E would fall under these provisions. ENTSO‑E was established by legal mandate in 2009 yet is not covered by the PSI directive. So is ENTSO‑E’s legal status sufficient to bring it under the Malamud ruling? While noting that copyrights must be balanced against public interests to determine when they are enforceable (Mendis and Kanevskaia 2024). I’ll try to find out from ENTSO‑E but I suspect the answer will be a fuzzy no.
Maybe? The Open Energy Ontology (Booshehri et al 2021) might fit the bill. Otherwise this community could develop its own dedicated semantic standards with suitable open licensing.
I’ve heard that the data model in OpenStreetMap and overlaying projects is becoming a de facto standard for some of the more routine aspects of electricity hardware. See the recently launched Oh my Grid! project based on OSM.
There may also be something available and legally unencumbered from public agencies, such as the US Energy Information Administration. However the UK ofgem regulator said in a recent webinar on smart meter data (ES Catapult 2025) that they would not take a lead on data semantics.
In the Europe Union, there is a large discrepancy between official statements on transparency and the reality of usable open data for public interest analysis. Much of this information is generated using public funds or mandated by statute, so the best solutions lie with our science funders and lawmakers — and not with user communities developing citizen science projects to somehow compensate.
I have not seen anything from the European Commission that suggests that they want their information economy to coexist with a thriving information commons. That hybrid model would be possible but it would require substantial interest and commitment from the Commission to bring it to life.
I am pretty sure non‑open semantic standards will become increasingly problematic in our domain (as evidenced by the Malamud case, which left key legal issues unresolved).
If anyone has comments or questions, please reply!
Blind, Knut, Mirko Böhm, and Nikolaus Thumm (October 2022). Chapter 11: Open source software in standard setting: the role of intellectual property right regimes. In Amanda Brock (editor). Open source law, policy and practice (2nd edition). Oxford, United Kingdom: Oxford University Press. ISBN 978-0-19-886234-5. doi:10.1093/oso/9780198862345.003.0011. Pages 256–272. Open access.
Booshehri, Meisam, Lukas Emele, Simon Flügel, Hannah Förster, Johannes Frey, Ulrich Frey, Martin Glauer, Janna Hastings, Christian Hofmann, Carsten Hoyer‑Klick, Ludwig Hülk, Anna Kleinau, Kevin Knosala, Leander Kotzur, Patrick Kuckertz, Till Mossakowski, Christoph Muschner, Fabian Neuhaus, Michaja Pehl, Martin Robinius, Vera Sehn, and Mirjam Stappel (1 September 2021). “Introducing the Open Energy Ontology: enhancing data interpretation and interfacing in energy systems analysis”. Energy and AI. 5: 100074. ISSN 2666-5468. doi:10.1016/j.egyai.2021.100074. Open access.
ES Catapult (1 May 2025). Weave: smart meter data unlocked webinar. London, United Kingdom: Energy Systems Catapult. YouTube video of duration 01:26:26. Streamed 28 April 2025.
IEC (26 June 2020). IEC 61970:301 — Common Information Model (CIM) core package — Edition 7.0. Geneva, Switzerland: International Electrotechnical Commission (IEC). 554 pages. Cover price €480. 
Mendis, Sunimal and Olia Kanevskaia (22 July 2024). Harmonized technical standards under EU copyright: the public.resource.org judgment. The IPKat. London, United Kingdom. Guest post.
â–˘
Hi Flo,
I’m not working in exactly the same direction, but I have been diving into the Common Grid Model Exchange Specification (CGMES) to recreate the BZR for Germany in PyPSA. From my experience, finding up-to-date public documentation is hard, especially since a GitHub page disappeared this January, and providing it in a good format.
While analysing the BZR Grid Model Data, I’ve come across several inconsistencies. E.g. I’ve found HVDC lines represented under the ACLine class, and the generator data is “creative”. So far, my focus has been on Germany, so I can’t speak for other Bidding Zones included in the BZR. My approach for the thesis was to sift through the data, which worked out in the end, but took a lot of time.
Do you think that LinkML would improve the process of importing CGMES models into PyPSA? And out of curiosity, what are you trying to accomplish?
Regards
Stefan
Right now I’m trying to convert this dataset to CGMES. LinkML was one approach that I tried, but other approaches might also work.