This is an early access version, the complete PDF, HTML, and XML versions will be available soon.
Open AccessArticle
Multi-Agent Deep Reinforcement Learning-Based HVAC and Electrochromic Window Control Framework
by
Hongjian Chen
Hongjian Chen 1
,
Duoyu Sun
Duoyu Sun 2
,
Yuyu Sun
Yuyu Sun 2,
Yong Zhang
Yong Zhang 1
and
Huan Yang
Huan Yang 2,*
1
School of Electronic Science and Engineering (School of Microelectronics), South China Normal University, Foshan 528225, China
2
School of Artificial Intelligence, South China Normal University, Foshan 528225, China
*
Author to whom correspondence should be addressed.
Buildings 2025, 15(17), 3114; https://doi.org/10.3390/buildings15173114 (registering DOI)
Submission received: 1 August 2025
/
Revised: 28 August 2025
/
Accepted: 29 August 2025
/
Published: 31 August 2025
Abstract
Deep reinforcement learning (DRL)-based HVAC control has shown clear advantages over rule-based and model predictive methods. However, most prior studies remain limited to HVAC-only optimization or simple coordination with operable windows. Such approaches do not adequately address buildings with fixed glazing systems—a common feature in high-rise offices—where the lack of operable windows restricts adaptive envelope interaction. To address this gap, this study proposes a multi-zone control framework that integrates HVAC systems with electrochromic windows (ECWs). The framework leverages the Q-value Mixing (QMIX) algorithm to dynamically coordinate ECW transmittance with HVAC setpoints, aiming to enhance energy efficiency and thermal comfort, particularly in high-consumption buildings such as offices. Its performance is evaluated using EnergyPlus simulations. The results show that the proposed approach reduces HVAC energy use by 19.8% compared to the DQN-based HVAC-only control and by 40.28% relative to conventional rule-based control (RBC). In comparison with leading multi-agent deep reinforcement learning (MADRL) algorithms, including MADQN, VDN, and MAPPO, the framework reduces HVAC energy consumption by 1–5% and maintains a thermal comfort violation rate (TCVR) of less than 1% with an average temperature variation of 0.35 Moreover, the model demonstrates strong generalizability, achieving 16.58–58.12% energy savings across six distinct climatic regions—ranging from tropical (Singapore) to temperate (Beijing)—with up to 48.2% savings observed in Chengdu. Our framework indicates the potential of coordinating HVAC systems with ECWs in simulation, while also identifying limitations that need to be addressed for real-world deployment.
Share and Cite
MDPI and ACS Style
Chen, H.; Sun, D.; Sun, Y.; Zhang, Y.; Yang, H.
Multi-Agent Deep Reinforcement Learning-Based HVAC and Electrochromic Window Control Framework. Buildings 2025, 15, 3114.
https://doi.org/10.3390/buildings15173114
AMA Style
Chen H, Sun D, Sun Y, Zhang Y, Yang H.
Multi-Agent Deep Reinforcement Learning-Based HVAC and Electrochromic Window Control Framework. Buildings. 2025; 15(17):3114.
https://doi.org/10.3390/buildings15173114
Chicago/Turabian Style
Chen, Hongjian, Duoyu Sun, Yuyu Sun, Yong Zhang, and Huan Yang.
2025. "Multi-Agent Deep Reinforcement Learning-Based HVAC and Electrochromic Window Control Framework" Buildings 15, no. 17: 3114.
https://doi.org/10.3390/buildings15173114
APA Style
Chen, H., Sun, D., Sun, Y., Zhang, Y., & Yang, H.
(2025). Multi-Agent Deep Reinforcement Learning-Based HVAC and Electrochromic Window Control Framework. Buildings, 15(17), 3114.
https://doi.org/10.3390/buildings15173114
Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details
here.
Article Metrics
Article Access Statistics
For more information on the journal statistics, click
here.
Multiple requests from the same IP address are counted as one view.