📄 A new SEEDS publication is now available!
“Price-responsive control using deep reinforcement learning for heating systems: Simulation and living lab experiment”
The study explores how price-responsive Deep Reinforcement Learning (DRL) can be used to autonomously control indoor temperature in buildings, responding to dynamic energy prices while balancing heating cost reduction and thermal comfort.
By developing and testing a DRL-based controller in both simulation and a real-world residential unit, the study highlights the potential of intelligent control strategies to improve energy flexibility, reduce costs, and support decarbonization.
Key insights :
🔹 Experimental results show a 79% reduction in heating costs compared to a rule-based controller.
🔹 Incorporating future price information leads to more informed decisions by the controller and improves the performance of DRL by 11.5%.
The study demonstrates that price-responsive DRL can effectively leverage dynamic electricity prices to lower heating costs and provide flexibility to the energy grid, contributing to building decarbonization.
Comments are closed