Reinforcement learning for energy conservation and comfort in buildings

Reinforcement Learning for Building Environmental Control

“…It is also evident that the potential benefits both in terms of energy conservation as well as in terms of comfort are significant. Nevertheless the results presented above and earlier in (Dalamagkidis et al, 2007) signify that there is still room for improvements. Besides better fine-tuning of the parameters and further training, the authors would like to propose some other ideas for future work.…”

“…On the other hand finding ways for the occupants to interact with the controller is far more complicated. In (Dalamagkidis et al, 2007) an additional module was proposed that is trained by occupant input whenever the latter is available. This module is then used as another component in determining the reward function.…”

Reinforcement learning for optimal control of low exergy buildings

“…They use a preliminary learning phase, which is conducted off-line on a simulation model to guide the controller before it is implemented into the actual environment. A linear reinforcement learning controller (LRLC) has been introduced in [24], with the aim of achieving energy savings, high comfort and indoor air quality. After a simulated period of four years, LRLC only manages to perform close the ON/OFF and fuzzy-logic controllers.…”

Predictive control techniques for energy and indoor environmental quality management in buildings

“…In the ''black box'' or non-physical model approaches, self-learning algorithms, reinforced learning [9] or neural networks [10] are some of the methodologies found in the literature. The benefits of the mentioned approaches are low computational time and the fact that they do not require any specific building modeling expertise, while their limitations are (i) the fact that neural networks require reliable training data that may not be available and (ii) self-learning algorithms cannot move beyond the limits of their experience.…”