heatbeat Blog

Newsletter Issue 62
2025/12/03

Supply Temperature Control Optimization via Dynamic Network Simulation

Dear Reader,

As usual, in the 62nd edition of our heatbeat Research Newsletter, we present a recent article by Boghetti et al. on supply temperature optimizations in heating networks. We would also like to use this newsletter to give you a brief overview of the most important developments and news about heatbeat and our Digital Twin.

News from heatbeat and the heatbeat Digital Twin

An absolute highlight for us in November was the launch of the first German TOP TENs of Energy Efficiency list. At the annual conference of the Energy Efficiency and Climate Protection Networks Initiative, climate-friendly technologies and services were honored, and we are delighted that our heatbeat Digital Twin was included in this list. The list recognizes products, projects, and services with exemplary energy-saving potential, cost-effectiveness, reliability, and ease of use. We would like to thank the Federal Ministry for Economic Affairs and Energy, the Federal Environment Agency, and the IEEKN for this award!

In addition, our team was present at numerous other events. We noticed that public participation formats on future heat supply met with particularly great interest and high participant numbers. These included, for example, the public information event on scenario analysis as part of municipal heat planning for Rheinauen, which we are conducting in cooperation with our project partner EMCEL. We would also like to express our gratitude for the numerous exciting discussions and intensive exchanges at our booth at HeatExpo in Dortmund, at the AGFW Trafotage in Kassel, at the Geothermal Congress 2025 in Frankfurt, and at the general meeting of EnergieRegion Nürnberg e.V.

We also made significant progress in the further development of our heatbeat Digital Twin in the weekly updates in November. The most important innovations include advances in the simulation of generation strategies and network dimensioning for multiple feed-in points in the network—you can find more details in our blog in the monthly feature update.

“Optimizing Supply Temperature Control in District Heating Networks via Differentiable Dynamic Simulation and Gradient Descent” Boghetti & Kämpf (Idiap Research Institute, EPFL)

What they did

The study addresses the limitations of conventional heating curve tuning, which often requires months of trial-and-error and leads to prolonged periods of suboptimal operation. Instead, the authors introduce a differentiable thermal simulation of the DHN and apply gradient descent to directly optimize heating curve parameters. The approach was tested on a real meshed network in Verbier, Switzerland, using monitoring data over a 5-day period.

Key results

Gradient-based optimization offers a promising alternative to traditional heating curve tuning, enabling faster and more systematic improvements in DHN efficiency. While tested on a short horizon, the approach is scalable to larger networks and longer periods, provided computational challenges in gradient calculation are addressed. Future work will integrate hydraulic constraints for even greater savings potential.

Why it matters

Gradient-based optimization offers a promising alternative to traditional heating curve tuning, enabling faster and more systematic improvements in DHN efficiency. While tested on a short horizon, the approach is scalable to larger networks and longer periods, provided computational challenges in gradient calculation are addressed. Future work will integrate hydraulic constraints for even greater savings potential.

Further information

As things slowly quiet down in December in terms of upcoming events, we would like to take this opportunity to announce that we will be kicking off the new year on January 15, 2026, with the next edition of our Feature Update Live Webinar. In this webinar series, we present the most important advances in heatbeat Digital Twin to you personally and interactively every three months. We look forward to your participation. You can register for free at this Link.

In addition, our team for municipal heat planning will present the municipal heat plan for Laer in North Rhine-Westphalia at the citizen information event on December 11 and the heat plan for Bad Essen in Lower Saxony on December 16.

The next edition of our newsletter will be published on January 7, 2026. Until then, we wish you happy holidays and a happy New Year 2026.

Sign up to our Newsletter