Toward intelligent multizone thermal control with multiagent deep reinforcement learning

Energy usage and thermal comfort are the pillars of smart buildings. Many research works have been proposed to save energy while maintaining a comfortable thermal condition. However, most of them either make the over-simplified assumption on thermal comfort with unsatisfied comfort performance or de...

Full description

Saved in:
Bibliographic Details
Main Authors: Li, Jie, Zhang, Wei, Gao, Guanyu, Wen, Yonggang, Jin, Guangyu, Christopoulos, Georgios
Other Authors: School of Computer Science and Engineering
Format: Article
Language:English
Published: 2021
Subjects:
Online Access:https://hdl.handle.net/10356/152738
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-152738
record_format dspace
spelling sg-ntu-dr.10356-1527382021-09-22T07:56:03Z Toward intelligent multizone thermal control with multiagent deep reinforcement learning Li, Jie Zhang, Wei Gao, Guanyu Wen, Yonggang Jin, Guangyu Christopoulos, Georgios School of Computer Science and Engineering Engineering::Computer science and engineering Multi-agent Deep Reinforcement Learning Neural Network Energy Efficiency Thermal Comfort Smart Building Energy usage and thermal comfort are the pillars of smart buildings. Many research works have been proposed to save energy while maintaining a comfortable thermal condition. However, most of them either make the over-simplified assumption on thermal comfort with unsatisfied comfort performance or deal with the single-zone thermal control only with limited practical impact. A few preliminary pieces of research on multi-zone control are available, but they fail to keep pace with the latest advancements in the deep learning-based control techniques. In this paper, we investigate the multi-zone thermal control with optimized energy usage and canonical thermal comfort modeling. We adopt the emerging multi-agent deep reinforcement learning techniques and propose to model each zone as an agent. A multi-agent framework is established to support the information exchange among the agents and enable intelligent thermal control in the heterogeneous zones. Accordingly, we mathematically formulate a problem to optimize both energy and comfort. A multi- zone thermal control algorithm (MOCA) is proposed to solve the problem by deriving optimal control policies. We validate the performance of MOCA through simulation in professional TRNSYS, configured based on our real-world laboratory. The results are promising with up to 15.4% energy-saving as well as satisfied thermal comfort in different zones. National Research Foundation (NRF) Accepted version This research is funded by National Research Foundation (NRF) via the Green Buildings Innovation Cluster (Grant NO.: NRF2015ENC_GBICRD001-012), administered by Building and Construction Authority (BCA) Singapore. In addition, this research is sponsored by National Research Foundation (NRF) via the Behavioural Studies in Energy, Water, Waste and Transportation Sectors (Grant NO.: BSEWWT2017 2 06), administered by National University of Singapore (NUS). Moreover, this research is funded by Nanyang Technological University (NTU) via the Data Science & Artificial Intelligence Research Centre @ NTU (Grant NO.: DSAIR@NTU). 2021-09-22T07:56:03Z 2021-09-22T07:56:03Z 2021 Journal Article Li, J., Zhang, W., Gao, G., Wen, Y., Jin, G. & Christopoulos, G. (2021). Toward intelligent multizone thermal control with multiagent deep reinforcement learning. IEEE Internet of Things Journal, 8(14), 11150-11162. https://dx.doi.org/10.1109/JIOT.2021.3051400 2327-4662 https://hdl.handle.net/10356/152738 10.1109/JIOT.2021.3051400 14 8 11150 11162 en NRF2015ENC_GBICRD001-012 BSEWWT2017_2_06 DSAIR@NTU IEEE Internet of Things Journal © 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: https://doi.org/10.1109/JIOT.2021.3051400. application/pdf
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Computer science and engineering
Multi-agent Deep Reinforcement Learning
Neural Network
Energy Efficiency
Thermal Comfort
Smart Building
spellingShingle Engineering::Computer science and engineering
Multi-agent Deep Reinforcement Learning
Neural Network
Energy Efficiency
Thermal Comfort
Smart Building
Li, Jie
Zhang, Wei
Gao, Guanyu
Wen, Yonggang
Jin, Guangyu
Christopoulos, Georgios
Toward intelligent multizone thermal control with multiagent deep reinforcement learning
description Energy usage and thermal comfort are the pillars of smart buildings. Many research works have been proposed to save energy while maintaining a comfortable thermal condition. However, most of them either make the over-simplified assumption on thermal comfort with unsatisfied comfort performance or deal with the single-zone thermal control only with limited practical impact. A few preliminary pieces of research on multi-zone control are available, but they fail to keep pace with the latest advancements in the deep learning-based control techniques. In this paper, we investigate the multi-zone thermal control with optimized energy usage and canonical thermal comfort modeling. We adopt the emerging multi-agent deep reinforcement learning techniques and propose to model each zone as an agent. A multi-agent framework is established to support the information exchange among the agents and enable intelligent thermal control in the heterogeneous zones. Accordingly, we mathematically formulate a problem to optimize both energy and comfort. A multi- zone thermal control algorithm (MOCA) is proposed to solve the problem by deriving optimal control policies. We validate the performance of MOCA through simulation in professional TRNSYS, configured based on our real-world laboratory. The results are promising with up to 15.4% energy-saving as well as satisfied thermal comfort in different zones.
author2 School of Computer Science and Engineering
author_facet School of Computer Science and Engineering
Li, Jie
Zhang, Wei
Gao, Guanyu
Wen, Yonggang
Jin, Guangyu
Christopoulos, Georgios
format Article
author Li, Jie
Zhang, Wei
Gao, Guanyu
Wen, Yonggang
Jin, Guangyu
Christopoulos, Georgios
author_sort Li, Jie
title Toward intelligent multizone thermal control with multiagent deep reinforcement learning
title_short Toward intelligent multizone thermal control with multiagent deep reinforcement learning
title_full Toward intelligent multizone thermal control with multiagent deep reinforcement learning
title_fullStr Toward intelligent multizone thermal control with multiagent deep reinforcement learning
title_full_unstemmed Toward intelligent multizone thermal control with multiagent deep reinforcement learning
title_sort toward intelligent multizone thermal control with multiagent deep reinforcement learning
publishDate 2021
url https://hdl.handle.net/10356/152738
_version_ 1712300649709830144