| layout | title | subtitle | thumbnail-img | tags | |||||
|---|---|---|---|---|---|---|---|---|---|
post |
IAQVEC 2023 at Tokyo |
The 11th International Conference on Indoor Air Quality, Ventilation & Energy Conservation in Buildings |
/assets/img/posts/iaqvec.jpg |
|
I presented our work "Comparing model predictive control and reinforcement learning for the optimal operation of building-PV-battery systems" at IAQVEC 2023. The study compared the performance and transferability of MPC and RL across smart grids with different buildings and under different conditions. The experiments conducted on the CityLearn virtual testbed revealed the practical value of both methods. While both methods achieved promising results, MPC had slightly better performance and could be transferred more smoothly. Given the standardized framework, MPC is more suitable in most cases for the purpose of microgrid operations. However, RL could be preferable for its quickness in making decisions if a large number of energy systems are involved.
Singapore crew at the conference