TESPy has been part of the oemof organization for about eight years. While it has always been represented at the meetings, it was time for the first ever TESPy community meeting!
Last week around ten of us got together at the University of Applied Sciences in Flensburg to discuss where TESPy is headed and how we can make open source thermodynamics even better. The discussions covered a broad range of topics:
A review of the most recent feature additions
Beginner tutorials
Debugging with tespy version 0.9
Custom equations into your model
Implementation of new components
Showcases of how the software is used
Strategies for the operation of large scale heat pumps
During the hackathon an addition to a component was not only drafted on paper but actually implemented in the framework. And, it has already made it to the latest release of TESPy.
It was really inspiring to meet others working with the software, exchange ideas, and see how the community is growing. Thanks to everyone who took their time to come to Flensburg to join the meeting.
Nigeria’s power system still relies heavily on captive diesel and petrol generators, with sizeable suppressed and unmet demand. The study by Shari et al. takes this reality as the starting point and uses the Open Energy Modelling Framework (oemof) to ask a question: given Nigeria’s 2060 net-zero target, how and where does green hydrogen make sense once you optimise the system transparently? The model is implemented with oemof-solph as a cost-minimising linear optimisation. The energy system is represented in oemof’s graph structure which uses buses for carriers and components connected by flows. Electricity and hydrogen each have dedicated buses. Supply components include gas turbines and diesel sets alongside solar PV, onshore wind, hydropower and biomass. The hydrogen chain is modelled with standard solph components: an electrolyser transformer converts surplus electricity to hydrogen, a hydrogen storage component holds the gas, and a fuel-cell transformer reconverts it to electricity when needed. Batteries provide a parallel short-term storage route. Demand, exports and curtailment are represented as sinks. Figures in the paper show the reference energy system and a simplified green-hydrogen chain built exactly this way.
Before scenario work, the team benchmarks the model against current system. Using a national hourly load profile adapted to Nigeria, the baseline dispatch reproduces the current diesel-heavy mix and visible unmet demand. That validation step is central to the study’s credibility: subsequent policy results are judged against a present-day system the model can already replicate. Three national scenarios are then evaluated as investment and policy constraints on the same system. Business as Usual allows gradual PV growth and treats gas as a transition fuel while diesel persists for longer. The Sustainability scenario raises renewable and efficiency ambition and phases out diesel more quickly. The Carbon-neutral scenario follows the government’s Energy Transition Plan and permits larger roles for hydrogen, bioenergy and nuclear. Long-term demand growth differs across scenarios and is applied in the 2045 and 2060 model periods. Because hydrogen will only be useful where resources and grid conditions warrant it, the study adopts a multi-node variant that aligns the model with five Distribution Companies (Abuja, Jos, Kaduna, Kano, Yola). Each DisCo becomes a regional node with its own resources, hourly demand and transfer links to the others. This lets the optimiser choose locally among four options for surplus variable renewables: charge batteries, export, curtail, or produce hydrogen. The choice depends on costs, efficiencies and future scarcity in each node. The paper motivates this distributed framing and shows hourly dispatch for both 2030 and 2060.
In Business as Usual scenario, diesel remains substantial in 2030 and only fades late, with gas covering much of the residual demand and storage growing mainly to support solar. In the Sustainability case, the crossover comes earlier: solar covers a significant share by 2030, diesel retreats, and by 2060 renewables dominate with hydrogen and batteries providing balancing. In the Carbon-neutral case, demand is higher but the cost-optimal mix by 2060 still prioritises solar, electrolysers, hydrogen storage and fuel cells; nuclear and bioenergy appear but do not dominate the power balance. Crucially, in the DisCo-level 2060 dispatch, diesel no longer appears in the hour-to-hour operation. Excess PV is directed to electrolysers when it outperforms curtailment plus future scarcity, hydrogen is stored, and fuel cells supply evening peaks or low-renewable periods. Inter-DisCo transfers smooth local mismatches without large curtailment spikes.
Figures reproduced from: Shari, B.E., Moumouni, Y., Ohunakin, O.S. et al. (2024). Exploring the role of green hydrogen for distributed energy access planning towards net-zero emissions in Nigeria.Sustainable Energy Research, 11, 16. https://doi.org/10.1186/s40807-024-00107-1. Licensed under Creative Commons Attribution 4.0 International (CC BY 4.0).
For the last three days, we were meeting at HTW in Berlin for our semi-annual developer meeting. Here are some of the key results:
We plan to have an improved lecture/ tutorial session, that might eventually develop into a summer school.
There will be a release of oemof.solph mid November. It will include whatever is done until then.
Costs calculations in oemof.solph will be aligned with VDI 2067. We plan to distinguish between energy-related costs (€/kWh), capacity related operational costs (€/kW/a), fixed operational costs (€/a) for OPEX and capacity related investment costs (€/kW/a) as well as an investment cost offset (€/a) for CAPEX. At the moment, we group by units, and thus have CAPEX and OPEX mixed.
Some of use will create an awareness concept. For upcoming meetings, there will be an awareness team you can contact if you do feel uncomfortable because of other participants or externals. We will also try to organise ourselves so that you can feel save on your trip back to the hotel.
The first ever in-person community meeting of tespy will be held in one month in Flensburg. From October 13. to 15. we will meet at the location of tespy’s origin, Flensburg University of Applied Sciences. There are many interesting topics on the agenda, suitable from very beginners to advanced users and potential future developers.
This week, there was a premiere at the Conference of Northern German Heat Research (in German): When presenting TESPy, our module for thermodynamic simulation, Francesco named the oemof association as an affiliation of Malte, Jonas, and himself.
Our next in-person user meeting will take place in Nordhausen from 11th to 13th of February 2026. It will be hosted by Nordhausen University of Applied Sciences partly parallel to the 9th Regenerative Energy Technology Conference (page in German).
We are posting this save the date already, as the parallel RET.Con also allows to submit contributions to its proceedings. If you wish to do so, please send an abstract (maximum two pages A4) to ret@hs-nordhausen.de. The deadline for abstracts to the RET.Con is the 15th of October. If accepted, full papers will be due at the 15th of January. Registration to the oemof track will be possible later, (as usual) without abstracts and proceedings.
We just got the feedback that the next oemof meeting will be hosted by Reiner Lemoine Institut from September 15th to 17th at HTW Berlin. Due to the limited room capacity it will come as a developer meeting. Save the date and contribute, if you are interested! (Beginners, in particular first-time contributors, will get guidance, of course.)
A major refactoring of the presolving and solving architecture has been carried out. The changes will allow users to debug models much more efficiently, as it is possible to extract the list of variables and the list of equations that have been presolved as well as those that are finally passed to the numerical solver. In this context, the presolver now further reduces the model size by identifying linear relationships between variables and mapping such variables to a single one. PowerConnections and respective components like Motor, Generator and PowerBus replace the Bus architecture, for example enabling the modelling of single shaft gas turbines, where the generator efficiency will only affect the net shaft power of the gas turbine.
Also, a new package in context of thermal engineering has been released under the oemof organization: ExerPy. The exergy analysis features of TESPy have been ported to the new package and it will be developed independently in the future to integrate more methods in the context of exergy analysis. ExerPy provides an easy to use API to read and process the results of your TESPy models.
As developers and users of the oemof software, we’re always interested in efforts that leverage its modular, open-source design for broader, cross-sectoral analyses. A recent paper, OWEFE, open modelling framework for integrated water, energy, food, and environment systems by J. Fleischmann et al. (2024), presents an “integrated-WEFE” layer that extends oemof’s energy system graph to encompass water, energy, food, and environmental systems in one unified model.
Without any public announcement, we shadow-dropped oemof.solph v0.6.0 last Friday (solph package 0.6.0 at PyPI, solph 0.6.0 changelog at GitHub). As yesterday was a public Holiday in Germany, this was an extra-bold move, breaking the dogma “never release on Fridays” in an extreme way. Seems like it wasn’t extra-stupid: Until now, nobody complained.
We decided to do this, as we believe it to be a rather stable release. We have cool new features (time-series aggregation and new result handling), but we marked them as experimental. The most noticeable change probably is the completely revised and reshaped documentation, which we tried to be more beginner-friendly than ever. Other changes include removal of API wrappers that allowed to still use old class names and keywords. Highlights of the release include:
