Multi-task deep reinforcement learning for intelligent multi-zone residential HVAC control

Yan Du; Fangxing Li; Jeffrey Munk; Kuldeep Kurte; Olivera Kotevska; Kadir Amasyali; Helia Zandi

doi:10.1016/j.epsr.2020.106959

Multi-task deep reinforcement learning for intelligent multi-zone residential HVAC control

Yan Du, Fangxing Li, Jeffrey Munk, Kuldeep Kurte, Olivera Kotevska, Kadir Amasyali, Helia Zandi

Research output: Contribution to journal › Article › peer-review

49 Scopus citations

Abstract

In this short communication, a data-driven deep reinforcement learning (deep RL) method is applied to minimize HVAC users’ energy consumption costs while maintaining users’ comfort. The applied deep RL method's efficiency is enhanced by conducting multi-task learning that can achieve an economic control strategy for a multi-zone residential HVAC system in both cooling and heating scenarios. The applied multi-task deep RL method is compared with a rule-based benchmark case and a single-task deep deterministic policy gradient algorithm to verify its effective and generalized application in optimizing HVAC operation.

Original language	English
Article number	106959
Journal	Electric Power Systems Research
Volume	192
DOIs	https://doi.org/10.1016/j.epsr.2020.106959
State	Published - Mar 2021

Funding

This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the US Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan ( http://energy.gov/downloads/doe-public-access-plan ). This work was funded in part by the U.S. Department of Energy, Energy Efficiency and Renewable Energy, Building Technology Office under contract number DE-AC05-00OR22725, in part by CURENT which is an Engineering Research Center (ERC) funded by the U.S. National Science Foundation (NSF) and DOE under the NSF award EEC-1041877, and in part by the U.S. NSF ECCS awards 1809458 and 2033910.

Funders	Funder number
CURENT
Energy Efficiency and Renewable Energy, Building Technology Office	DE-AC05-00OR22725
National Science Foundation
U.S. Department of Energy	1809458, EEC-1041877, 2033910
Engineering Research Centers

Keywords

Data-driven
Deep deterministic policy gradient (DDPG)
Multi-task deep reinforcement learning
Multi-zone HVAC

Access to Document

10.1016/j.epsr.2020.106959

Cite this

@article{95f3f5b74b864aae948cf2bf6ad81921,

title = "Multi-task deep reinforcement learning for intelligent multi-zone residential HVAC control",

abstract = "In this short communication, a data-driven deep reinforcement learning (deep RL) method is applied to minimize HVAC users{\textquoteright} energy consumption costs while maintaining users{\textquoteright} comfort. The applied deep RL method's efficiency is enhanced by conducting multi-task learning that can achieve an economic control strategy for a multi-zone residential HVAC system in both cooling and heating scenarios. The applied multi-task deep RL method is compared with a rule-based benchmark case and a single-task deep deterministic policy gradient algorithm to verify its effective and generalized application in optimizing HVAC operation.",

keywords = "Data-driven, Deep deterministic policy gradient (DDPG), Multi-task deep reinforcement learning, Multi-zone HVAC",

author = "Yan Du and Fangxing Li and Jeffrey Munk and Kuldeep Kurte and Olivera Kotevska and Kadir Amasyali and Helia Zandi",

note = "Publisher Copyright: {\textcopyright} 2020 Elsevier B.V.",

year = "2021",

month = mar,

doi = "10.1016/j.epsr.2020.106959",

language = "English",

volume = "192",

journal = "Electric Power Systems Research",

issn = "0378-7796",

publisher = "Elsevier",

}

TY - JOUR

T1 - Multi-task deep reinforcement learning for intelligent multi-zone residential HVAC control

AU - Du, Yan

AU - Li, Fangxing

AU - Munk, Jeffrey

AU - Kurte, Kuldeep

AU - Kotevska, Olivera

AU - Amasyali, Kadir

AU - Zandi, Helia

PY - 2021/3

Y1 - 2021/3

N2 - In this short communication, a data-driven deep reinforcement learning (deep RL) method is applied to minimize HVAC users’ energy consumption costs while maintaining users’ comfort. The applied deep RL method's efficiency is enhanced by conducting multi-task learning that can achieve an economic control strategy for a multi-zone residential HVAC system in both cooling and heating scenarios. The applied multi-task deep RL method is compared with a rule-based benchmark case and a single-task deep deterministic policy gradient algorithm to verify its effective and generalized application in optimizing HVAC operation.

AB - In this short communication, a data-driven deep reinforcement learning (deep RL) method is applied to minimize HVAC users’ energy consumption costs while maintaining users’ comfort. The applied deep RL method's efficiency is enhanced by conducting multi-task learning that can achieve an economic control strategy for a multi-zone residential HVAC system in both cooling and heating scenarios. The applied multi-task deep RL method is compared with a rule-based benchmark case and a single-task deep deterministic policy gradient algorithm to verify its effective and generalized application in optimizing HVAC operation.

KW - Data-driven

KW - Deep deterministic policy gradient (DDPG)

KW - Multi-task deep reinforcement learning

KW - Multi-zone HVAC

UR - http://www.scopus.com/inward/record.url?scp=85096554866&partnerID=8YFLogxK

U2 - 10.1016/j.epsr.2020.106959

DO - 10.1016/j.epsr.2020.106959

M3 - Article

AN - SCOPUS:85096554866

SN - 0378-7796

VL - 192

JO - Electric Power Systems Research

JF - Electric Power Systems Research

M1 - 106959

ER -

Multi-task deep reinforcement learning for intelligent multi-zone residential HVAC control

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this