Deepcomfort : energy-efficient thermal comfort control in buildings via reinforcement learning

Heating, Ventilation, and Air Conditioning (HVAC) are extremely energy-consuming, accounting for 40% of total building energy consumption. It is crucial to design some energy-efficient building thermal comfort control strategy which can reduce the energy consumption of the HVAC while maintaining the...

Full description

Saved in:
Bibliographic Details
Main Authors: Gao, Guanyu, Li, Jie, Wen, Yonggang
Other Authors: School of Computer Science and Engineering
Format: Article
Language:English
Published: 2021
Subjects:
Online Access:https://hdl.handle.net/10356/152769
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-152769
record_format dspace
spelling sg-ntu-dr.10356-1527692021-09-30T10:52:53Z Deepcomfort : energy-efficient thermal comfort control in buildings via reinforcement learning Gao, Guanyu Li, Jie Wen, Yonggang School of Computer Science and Engineering Engineering::Computer science and engineering Thermal Comfort Control Thermal Comfort Prediction Deep Reinforcement Learning Heating, Ventilation, and Air Conditioning Smart Building Heating, Ventilation, and Air Conditioning (HVAC) are extremely energy-consuming, accounting for 40% of total building energy consumption. It is crucial to design some energy-efficient building thermal comfort control strategy which can reduce the energy consumption of the HVAC while maintaining the comfort of the occupants. However, implementing such a strategy is challenging, because the changes of the thermal states in a building environment are influenced by various factors. The relationships among these influencing factors are hard to model and are always different in different building environments. To address this challenge, we propose a deep reinforcement learning based framework, DeepComfort, for thermal comfort control in buildings. We formulate the thermal comfort control as a cost-minimization problem by jointly considering the energy consumption of the HVAC and the occupants’ thermal comfort. We first design a deep Feedforward Neural Network (FNN) based approach for predicting the occupants’ thermal comfort, and then propose a Deep Deterministic Policy Gradients (DDPG) based approach for learning the optimal thermal comfort control policy. We implement a building thermal comfort control simu- lation environment and evaluate the performance under various settings. The experimental results show that our approaches can improve the performance of thermal comfort prediction by 14.5% and reduce the energy consumption of HVAC by 4.31% while improving the occupants’ thermal comfort by 13.6%. National Research Foundation (NRF) Accepted version This research is supported in part by a project fund from DSAIR@NTU, and a BSEWWT project fund from Na- tional Research Foundation Singapore, administrated through the BSEWWT program office (Ref. BSEWWT2017_2_06), the Green Buildings Innovation Cluster (Grant NO.: NRF2015ENC-GBICRD001-012), administered by Building and Construction Authority (BCA) Singapore. 2021-09-30T10:50:07Z 2021-09-30T10:50:07Z 2020 Journal Article Gao, G., Li, J. & Wen, Y. (2020). Deepcomfort : energy-efficient thermal comfort control in buildings via reinforcement learning. IEEE Internet of Things Journal, 7(9), 8472-8484. https://dx.doi.org/10.1109/JIOT.2020.2992117 2327-4662 https://hdl.handle.net/10356/152769 10.1109/JIOT.2020.2992117 9 7 8472 8484 en NRF2015ENC-GBICRD001-012 BSEWWT2017_2_06 IEEE Internet of Things Journal © 2020 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.2020.2992117. 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
Thermal Comfort Control
Thermal Comfort Prediction
Deep Reinforcement Learning
Heating, Ventilation, and Air Conditioning
Smart Building
spellingShingle Engineering::Computer science and engineering
Thermal Comfort Control
Thermal Comfort Prediction
Deep Reinforcement Learning
Heating, Ventilation, and Air Conditioning
Smart Building
Gao, Guanyu
Li, Jie
Wen, Yonggang
Deepcomfort : energy-efficient thermal comfort control in buildings via reinforcement learning
description Heating, Ventilation, and Air Conditioning (HVAC) are extremely energy-consuming, accounting for 40% of total building energy consumption. It is crucial to design some energy-efficient building thermal comfort control strategy which can reduce the energy consumption of the HVAC while maintaining the comfort of the occupants. However, implementing such a strategy is challenging, because the changes of the thermal states in a building environment are influenced by various factors. The relationships among these influencing factors are hard to model and are always different in different building environments. To address this challenge, we propose a deep reinforcement learning based framework, DeepComfort, for thermal comfort control in buildings. We formulate the thermal comfort control as a cost-minimization problem by jointly considering the energy consumption of the HVAC and the occupants’ thermal comfort. We first design a deep Feedforward Neural Network (FNN) based approach for predicting the occupants’ thermal comfort, and then propose a Deep Deterministic Policy Gradients (DDPG) based approach for learning the optimal thermal comfort control policy. We implement a building thermal comfort control simu- lation environment and evaluate the performance under various settings. The experimental results show that our approaches can improve the performance of thermal comfort prediction by 14.5% and reduce the energy consumption of HVAC by 4.31% while improving the occupants’ thermal comfort by 13.6%.
author2 School of Computer Science and Engineering
author_facet School of Computer Science and Engineering
Gao, Guanyu
Li, Jie
Wen, Yonggang
format Article
author Gao, Guanyu
Li, Jie
Wen, Yonggang
author_sort Gao, Guanyu
title Deepcomfort : energy-efficient thermal comfort control in buildings via reinforcement learning
title_short Deepcomfort : energy-efficient thermal comfort control in buildings via reinforcement learning
title_full Deepcomfort : energy-efficient thermal comfort control in buildings via reinforcement learning
title_fullStr Deepcomfort : energy-efficient thermal comfort control in buildings via reinforcement learning
title_full_unstemmed Deepcomfort : energy-efficient thermal comfort control in buildings via reinforcement learning
title_sort deepcomfort : energy-efficient thermal comfort control in buildings via reinforcement learning
publishDate 2021
url https://hdl.handle.net/10356/152769
_version_ 1713213287048413184