At SimBuild 2024 (Denver, May 21–23), we presented a paper introducing an innovative two-step method for mixed-integer nonlinear model predictive control (MINLP) of fifth-generation district heating and cooling networks (5GDHC).

🔎 Key highlights of the work :

• Applied to a virtual district with four residential buildings and one office
• Compared with a basic integer-selection control approach
• Based on detailed physics-based models built in Modelica

🏆 Goal: minimize energy consumption while ensuring thermal comfort

📊 An 8-month open-loop simulation showed that the proposed controller:

✔️ Significantly reduces energy use
✔️ Makes better use of system flexibility
✔️ Maintains thermal comfort across all buildings

This work demonstrates the potential for smarter and more efficient control strategies in next-generation thermal networks. 🔁🔥❄️