Energy Cost Driven Heating Control with Reinforcement Learning

The current energy crisis raised concern about the lack of electricity during wintertime, especially that consumption should be cut at peak hours. For building owners, this is visible as rising prices. Availability near real-time data on performance opening new opportunities to optimize flexibility capabilities buildings. This paper presents a reinforcement learning (RL)-based method control heating for minimizing cost and shifting usage away from demand Simulations are carried out with electrically heated single-family houses. results indicate RL, in case varying prices, it possible save money keep indoor thermal comfort an appropriate level.