UCD Energy Institute/NexSys Seminar 27 January 2026

Guest Seminar with Dr Stef Jacobs 27 January 2026

We are delighted to announce an upcoming research seminar with (opens in a new window)Dr. Stef Jacobs, a visiting researcher with the NexSys research programme. Stef is a Post-doctoral researcher in collective heating and cooling at University of Antwerp, Belgium. 

Title of the presentation: Optimal collective heating and cooling systems in apartment buildings

  Location: Room 1.12, Science Hub (SCIH), O’Brien Centre for Science, University College Dublin or zoom (link below)

             (opens in a new window)Link

             Meeting ID: 661 9494 1943

             Passcode: 681793

When: Tuesday, 27 January 2026 at 11am

Abstract:
This presentation will first summarise the key findings of Stef Jacobs’ PhD research on water-based change-over systems for collective heating and cooling in apartment buildings, completed at the University of Antwerp in 2025. The work covers the evaluation of multiple system configurations, the analysis of rule-based and data-driven control strategies, including deep reinforcement learning, and the development of a decision-support tool for design and operational decision-making.

The presentation will then introduce the ongoing TETRA projectChangeOver, which translates these academic insights into practical guidelines through collaboration with an industry consortium, while also addressing the associated implementation challenges.

In the second part, Stef will outline his postdoctoral research ambitions. These include the further development of optimal control strategies for collective thermal systems, the characterisation of control logic for integration into holistic optimisation tools, and research on fault detection and diagnosis, with a focus on bridging the simulation-to-reality gap. The seminar will conclude with several proposed project ideas, including system-based energy pricing for demand-side management and the use of transfer learning to accelerate and generalise deep reinforcement learning across different systems.

All welcome!

(also keep an eye on our (opens in a new window)website and (opens in a new window)LinkedIn)