The editorial frames open-tyndp as a watershed moment because the TYNDP justifies €584 billion in cross-border transmission investment through 2030, yet until now the underlying model was a proprietary chain of PLEXOS, Antares, and in-house spreadsheets owned by transmission system operators. By shipping scenario logic, network topology, cost assumptions, and the optimisation formulation in Python, stakeholders like regulators, NGOs, and member-state ministries can finally interrogate and challe
The project explicitly scopes itself to reproducing TYNDP outputs rather than producing an alternative plan, with the first milestone being replication of the 2024 TYNDP electricity scenarios. This positioning matters because it lets the tool function as an audit and verification layer that ENTSO-E itself cannot easily dismiss, rather than as a rival modelling effort that would be easy to sideline politically.
By surfacing the repository on HN with the framing 'EU Open Sources Ten-Year Network Development Planning Tools,' the submitter highlights the reproducibility angle as the headline value. The 143-point score suggests the developer audience sees the fork-and-extend approach building on PyPSA-Eur as a credible, pragmatic path rather than a vanity rewrite.
The editorial emphasizes that open-tyndp is a fork-and-extend of PyPSA-Eur, the academic European power-system model that's been kicking around since 2017, retooled to consume the exact input datasets ENTSO-E and ENTSOG publish. This lineage matters because it inherits validated solver formulations and a research community already familiar with the codebase, rather than starting from a blank slate that would take years to reach credibility.
Open Energy Transition, a nonprofit spun out of the PyPSA research group, has published open-tyndp — a from-scratch open-source implementation of the modelling stack behind ENTSO-E's Ten-Year Network Development Plan (TYNDP). The repo hit 143 on Hacker News this week and sits at github.com/open-energy-transition/open-tyndp. It's a fork-and-extend of PyPSA-Eur, the academic European power-system model that's been kicking around since 2017, retooled to consume the exact input datasets ENTSO-E and ENTSOG publish for their biennial planning cycle.
The TYNDP is not a small document. It is the single planning artefact that justifies, prioritises, and sequences essentially every cross-border transmission project the EU funds — a pipeline currently estimated at €584 billion through 2030 for electricity alone, before you add the hydrogen and offshore-grid annexes. Until now, the model that produces it has been a black box: a proprietary chain of PLEXOS, Antares, and in-house spreadsheets owned by the European transmission system operators themselves. Stakeholders — regulators, NGOs, industrial consumers, member-state ministries — got a PDF and a stakeholder workshop. If you wanted to challenge a scenario, you wrote a position paper and hoped someone read it.
open-tyndp ships the scenario logic, the network topology import, the cost assumptions, and the optimisation formulation in Python. The README is explicit that the goal is reproducibility of TYNDP results, not a competing plan. The first milestone listed is replicating the 2024 TYNDP electricity scenarios; the hydrogen and gas modules are flagged as in progress.
Grid planning is the most consequential modelling problem nobody outside the sector argues about, because nobody outside the sector could. The TYNDP determines which interconnectors get built, which countries get stranded with curtailment, and which generation mixes are considered "adequate" for the next decade — and the assumptions driving it have until now been auditable only by the people producing it. That's a governance gap large enough to drive a 525kV HVDC line through.
The parallels to other domains where open-sourcing the model reframed the debate are worth dwelling on. When the IPCC's climate models were opened up, climate-policy arguments moved from "trust us" to "here's the parameter you're disputing." When OpenStreetMap challenged Navteq, routing stopped being a vendor monoculture. The same shift is now plausible for European grid planning. A think tank in Berlin can rerun TYNDP 2024 with a 20% lower demand growth assumption and publish the diff. An offshore-wind consortium can test whether moving a substation 80km changes the case for a North Sea hub. A Polish coal-region MEP can ask, on the floor, why a specific scenario assumes 2034 lignite phase-out and get a falsifiable answer.
The technical bet is sound. PyPSA is the only open energy-system modelling framework that has been independently benchmarked against commercial tools like PLEXOS and shown to produce comparable optimal-dispatch results on European-scale problems — typically within 1-2% on system cost. That credibility is what makes open-tyndp more than an academic exercise. ENTSO-E can't easily wave it away as "toy modelling," because the same solver core has already been cited in peer-reviewed work the Commission funds.
The HN thread surfaced the obvious sceptical take: ENTSO-E will simply ignore it. Possibly. But the EU's own Network Code on Demand Response and the ACER opinions on the 2022 TYNDP both flagged transparency as a regulatory deficiency. Once an open reproduction exists, "we cannot share the model" stops being a defensible answer in a stakeholder consultation — and ACER, which has formal oversight, now has a reference implementation to point at.
If you're building anything that touches European electricity markets — battery dispatch optimisers, virtual power plants, PPA pricing tools, industrial demand-response — open-tyndp gives you something you previously had to either pay six figures for or reconstruct from public PDFs: a defensible long-term price-and-flow forecast you can actually inspect. The locational marginal prices, the cross-border flows, the residual load curves — all derivable from a reproducible run rather than scraped from an ENTSO-E transparency dashboard with a four-month lag.
For practitioners outside utilities, the more interesting unlock is the input pipeline. The repo standardises the messy ETL between ENTSO-E's published datasets (which arrive as a mix of Excel, XML, and PDF) and a clean network model. That alone is months of grunt work most independent analysts never had time to do. If you've been considering a side project around European energy data, the marginal cost of getting to a working model just collapsed.
A caveat: this is not a turnkey tool. Solving a continental electricity model with hourly resolution and unit commitment is a 16-core, 128GB-RAM job at minimum, and the commercial solvers (Gurobi, CPLEX) still outperform the open HiGHS solver on the harder MIP variants by a meaningful margin. Plan on either a Gurobi academic licence or accepting longer solve times. The PyPSA community has been pragmatic about this trade-off for years.
The interesting question isn't whether open-tyndp produces numbers identical to ENTSO-E's — it won't, and the gaps will themselves be informative. The question is whether the first credible third-party rerun, probably from Agora Energiewende or Ember or one of the Brussels think tanks that already lives in this data, produces a scenario that contradicts an official one on something politically expensive — a delayed interconnector, an overbuilt hydrogen corridor, an underestimated demand peak. When that happens, and the methodology is on GitHub rather than behind an NDA, the politics of European grid planning changes in a way that a decade of transparency directives hasn't managed.
Nice. Better energy grid seems like an easy win for Europe. It has enough diversity of sources that just connecting it cleverly can get you crazy efficiency.e.g. Right now the UK grid is running on just over 80%+ renewable. But when the wind isn't cooperating french nuclear helps.Broad reliance
Cringe. China and Russia are building energy infrastructure and here we have Europe open sourcing tools to help cope with its self-inflicted bureaucratic burdens instead.Yay, sustainability!
Educate me, what’s the point of (open-sourcing) this?
What a great present for Russians planning their drone attacks on Europe. Why not open source military bases and personal development too? It's all theoretically could be found in public sources anyway, so why not help Ivan save some time and put it all together. As a EU citizen, I should be ab
Top 10 dev stories every morning at 8am UTC. AI-curated. Retro terminal HTML email.
And where did they put it? On the biggest US platform.Someday we'll get there, someday.