The outputs of the project focus on the advice for decision makers. Thus, the results will be presented in a concise manner and directed towards assisting decision makers on the evolution of the national system.
As secondary outputs, 2 technical notes will be released:
- Potential and impact of heat pump deployment in Belgium in 2030 - 2050
- Comparing electric and thermal batteries for decentralized storage
Finally, FlexMyHeat’s technical innovations in modeling and flex exploitation methods will be shared with the scientific community via scientific papers.
The different reports and scientific papers will also be published on this webpage.
Deliverables
- D1.1 - Characterization of flexible assets, adoption potential and grid constraints and markets: Overview of the gathered data sources used for the analyses, the derivation of representative users and PV generation profiles, the modeling of buildings and relevant flexible assets (heat pump, battery and thermal storage) and the simulations and analyses for the business-as-usual scenarios.
- D2.1 - Overview of models for flexibility, local constraints, energy markets and flex controllers: Overview of evaluation of the smart control of battery systems in function of day-ahead and imbalance markets.
- D3.1 - Simulation results for flexibility potential on Belgian electricity grid: Overview of the results for the individual smart control scenarios for heat pumps and thermal storage systems and for the integrated smart control scenarios (battery storage + heat pump + thermal storage).
- D4.1a - White paper for decision makers: Summary of the main results obtained in the project.
- D4.1b - Technical report on heat pump deployment: Overview of the expected evolution of heat pump deployment in Belgium, and associated electricity consumption, and assessment of the impact on the increase of peak power if heat pumps are installed as is vs their smart control in combination with battery and thermal storage.
- D4.1c - Comparing electric and thermal batteries for decentralized storage: Comparison of the performance and characteristics of electric batteries and thermal storage in decentralized residential settings.
Presentations
- Overview of results from D1.1 as presented on the first review meeting.
Scientific Publications
- Seyed Soroush Karimi Madahi, Toon Van Puyvelde, Gargya Gokhale, Bert Claessens, Chris Develder, Multi-source Transfer Learning in Reinforcement Learning-based Home Battery Controller in Proceedings of the 4th ACM International Workshop on Big Data and Machine Learning for Smart Buildings and Cities (In conjunction with ACM BuildSys 24), Hangzhou, Zhejiang, China, November 6 2024
- Seyed Soroush Karimi Madahi, Gargya Gokhale, Marie-Sophie Verwee, Bert Claessens and Chris Develder, Control policy correction framework for reinforcement learning-based energy arbitrage strategies in Proceedings of the 15th ACM International Conference on Future and Sustainable Energy Systems (e-Energy 2024), Singapore, 4-7 June 2024
- Gargya Gokhale, Seyed Soroush Karimi Madahi, Bert Claessens and Chris Develder, Distill2Explain: Differentiable decision trees for explainable reinforcement learning in energy application controllers, in Proceedings of the 15th ACM International Conference on Future and Sustainable Energy Systems (e-Energy 2024), Singapore, 4-7 June 2024
- Gargya Gokhale, Bert Claessens and Chris Develder, Explainable reinforcement learning-based home energy management systems using differentiable decision trees, in Proceedings of the 1st ACM International Workshop on Trust-worthy ML for Energy Systems (SAFE-ENERGY) at ACM e-Energy 2024, 4 June 2024
- Seyed Soroush Karimi Madahi, Bert Claessens, Chris Develder, Distributional reinforcement learning-based energy arbitrage strategies in imbalance settlement mechanism, Journal of Energy Storage, Vol. 104, Part A, Dec. 2024, pp. 114377.