A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Developing a Window Control Algorithm Based on Reinforcement Learning for Indoor PM2.5 Mitigation
Smart control of window is a means of effectively reducing concentrations of indoor PM2.5 (particulate matter with aerodynamic diameter less than 2.5 μm) in naturally ventilated residential buildings without air cleaning devices. This study aimed to develop a reinforcement learning (RL) approach to automatically control window behavior in real time for mitigation of indoor PM2.5 pollution. The method trains the window controller using deep Q-network (DQN) in a specific apartment for a month. The trained controller can be employed to control window behavior to mitigate indoor PM2.5 in that apartment. The input data for the controller are the real-time indoor and outdoor PM2.5 concentrations with 1-min resolution. Simulation was conducted in a real apartment in Tianjin. The results show that, the RL algorithm reduced the average indoor PM2.5 concentration by 9.11% when compared with the I/O ratio algorithm and by 7.40% when compared with real window behavior.
Developing a Window Control Algorithm Based on Reinforcement Learning for Indoor PM2.5 Mitigation
Smart control of window is a means of effectively reducing concentrations of indoor PM2.5 (particulate matter with aerodynamic diameter less than 2.5 μm) in naturally ventilated residential buildings without air cleaning devices. This study aimed to develop a reinforcement learning (RL) approach to automatically control window behavior in real time for mitigation of indoor PM2.5 pollution. The method trains the window controller using deep Q-network (DQN) in a specific apartment for a month. The trained controller can be employed to control window behavior to mitigate indoor PM2.5 in that apartment. The input data for the controller are the real-time indoor and outdoor PM2.5 concentrations with 1-min resolution. Simulation was conducted in a real apartment in Tianjin. The results show that, the RL algorithm reduced the average indoor PM2.5 concentration by 9.11% when compared with the I/O ratio algorithm and by 7.40% when compared with real window behavior.
Developing a Window Control Algorithm Based on Reinforcement Learning for Indoor PM2.5 Mitigation
Environ Sci Eng
Wang, Liangzhu Leon (editor) / Ge, Hua (editor) / Zhai, Zhiqiang John (editor) / Qi, Dahai (editor) / Ouf, Mohamed (editor) / Sun, Chanjuan (editor) / Wang, Dengjia (editor) / An, Yuting (author) / Xia, Tongling (author) / You, Ruoyu (author)
International Conference on Building Energy and Environment ; 2022
Proceedings of the 5th International Conference on Building Energy and Environment ; Chapter: 149 ; 1437-1446
2023-09-05
10 pages
Article/Chapter (Book)
Electronic Resource
English
Reinforcement learning , Smart control , PM<sub>2.5</sub> , Natural ventilation , Artificial intelligence and internet of things (AIoT) Engineering , Building Physics, HVAC , Fire Science, Hazard Control, Building Safety , Sustainable Architecture/Green Buildings , Renewable and Green Energy , Environment, general
Impact of the external window crack structure on indoor PM2.5 mass concentration
| Online Contents | 2016
Impact of the external window crack structure on indoor PM2.5 mass concentration
| British Library Online Contents | 2016
Classification prediction model of indoor PM2.5 concentration using CatBoost algorithm
| DOAJ | 2023
Performing indoor PM2.5 prediction with low-cost data and machine learning
| Emerald Group Publishing | 2022