Control of HVAC-Systems Using Reinforcement Learning With Hysteresis and Tolerance Control

Christian Blad; Carsten Kallesøe; Simon Bøgh

doi:10.1109/SII46433.2020.9026189

Control of HVAC-Systems Using Reinforcement Learning With Hysteresis and Tolerance Control

Christian Blad^*, Carsten Kallesøe, Simon Bøgh

^*Corresponding author for this work

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

4 Citations (Scopus)

Abstract

This paper presents the idea of using tolerance control in Deep Reinforcement Learning to improve robustness and reduce training time. This paper is a continuation of [1] where it is shown that Reinforcement Learning (RL) can be used to control an underfloor heating (UFH) system. However, it is seen in the study that the initial training time is too high and that the performance during training is not fulfilling the requirements to a UFH system. In this paper the fundamental challenge regarding control of UFH systems is explained, how RL can be beneficial for control of UFH systems, and how the implementation is done. Furthermore, results are presented with a standard hysteresis control, an RL control, and an RL control with tolerance control. These results show that the effect of tolerance control in these types of systems is significant. Finally, we discuss the challenges there are for a real-world implementation of RL-based control in UFH system.

Original language	English
Title of host publication	IEEE/SICE International Symposium on System Integration
Number of pages	5
Publisher	IEEE (Institute of Electrical and Electronics Engineers)
Publication date	2020
Pages	938-942
Article number	9026189
ISBN (Electronic)	9781728166674
DOIs	https://doi.org/10.1109/SII46433.2020.9026189
Publication status	Published - 2020
Event	IEEE/SICE International Symposium on System Integration - Hawaii Convention Center, Honolulu, United States Duration: 12 Jan 2020 → 15 Jan 2020

Conference

Conference	IEEE/SICE International Symposium on System Integration
Location	Hawaii Convention Center
Country/Territory	United States
City	Honolulu
Period	12/01/2020 → 15/01/2020

Series	Proceedings of the 2020 IEEE/SICE International Symposium on System Integration
ISSN	2474-2325

Keywords

Deep Reinforcement Learning
Artificial Intelligence (AI)
Deep Learning
HVAC System
Hysteresis
tolerance control
HVAC-systems
Tolerance Control
Underfloor Heating

Access to Document

10.1109/SII46433.2020.9026189

AUB Link

Search for the material in Aalborg University Library's search engine

Cite this

@inproceedings{0ae9e2a0c4424074a1056fc6a8bdb341,

title = "Control of HVAC-Systems Using Reinforcement Learning With Hysteresis and Tolerance Control",

abstract = "This paper presents the idea of using tolerance control in Deep Reinforcement Learning to improve robustness and reduce training time. This paper is a continuation of [1] where it is shown that Reinforcement Learning (RL) can be used to control an underfloor heating (UFH) system. However, it is seen in the study that the initial training time is too high and that the performance during training is not fulfilling the requirements to a UFH system. In this paper the fundamental challenge regarding control of UFH systems is explained, how RL can be beneficial for control of UFH systems, and how the implementation is done. Furthermore, results are presented with a standard hysteresis control, an RL control, and an RL control with tolerance control. These results show that the effect of tolerance control in these types of systems is significant. Finally, we discuss the challenges there are for a real-world implementation of RL-based control in UFH system.",

keywords = "Deep Reinforcement Learning, Artificial Intelligence (AI), Deep Learning, HVAC System, Hysteresis, tolerance control, HVAC-systems, Tolerance Control, Underfloor Heating",

author = "Christian Blad and Carsten Kalles{\o}e and Simon B{\o}gh",

year = "2020",

doi = "10.1109/SII46433.2020.9026189",

language = "English",

series = "Proceedings of the 2020 IEEE/SICE International Symposium on System Integration",

pages = "938--942",

booktitle = "IEEE/SICE International Symposium on System Integration",

publisher = "IEEE (Institute of Electrical and Electronics Engineers)",

address = "United States",

note = "IEEE/SICE International Symposium on System Integration ; Conference date: 12-01-2020 Through 15-01-2020",

}

Blad, C, Kallesøe, C & Bøgh, S 2020, Control of HVAC-Systems Using Reinforcement Learning With Hysteresis and Tolerance Control. in IEEE/SICE International Symposium on System Integration., 9026189, IEEE (Institute of Electrical and Electronics Engineers), Proceedings of the 2020 IEEE/SICE International Symposium on System Integration, pp. 938-942, IEEE/SICE International Symposium on System Integration, Honolulu, Hawaii, United States, 12/01/2020. https://doi.org/10.1109/SII46433.2020.9026189

Control of HVAC-Systems Using Reinforcement Learning With Hysteresis and Tolerance Control. / Blad, Christian; Kallesøe, Carsten ; Bøgh, Simon.
IEEE/SICE International Symposium on System Integration. IEEE (Institute of Electrical and Electronics Engineers), 2020. p. 938-942 9026189 (Proceedings of the 2020 IEEE/SICE International Symposium on System Integration).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

TY - GEN

T1 - Control of HVAC-Systems Using Reinforcement Learning With Hysteresis and Tolerance Control

AU - Blad, Christian

AU - Kallesøe, Carsten

AU - Bøgh, Simon

PY - 2020

Y1 - 2020

N2 - This paper presents the idea of using tolerance control in Deep Reinforcement Learning to improve robustness and reduce training time. This paper is a continuation of [1] where it is shown that Reinforcement Learning (RL) can be used to control an underfloor heating (UFH) system. However, it is seen in the study that the initial training time is too high and that the performance during training is not fulfilling the requirements to a UFH system. In this paper the fundamental challenge regarding control of UFH systems is explained, how RL can be beneficial for control of UFH systems, and how the implementation is done. Furthermore, results are presented with a standard hysteresis control, an RL control, and an RL control with tolerance control. These results show that the effect of tolerance control in these types of systems is significant. Finally, we discuss the challenges there are for a real-world implementation of RL-based control in UFH system.

AB - This paper presents the idea of using tolerance control in Deep Reinforcement Learning to improve robustness and reduce training time. This paper is a continuation of [1] where it is shown that Reinforcement Learning (RL) can be used to control an underfloor heating (UFH) system. However, it is seen in the study that the initial training time is too high and that the performance during training is not fulfilling the requirements to a UFH system. In this paper the fundamental challenge regarding control of UFH systems is explained, how RL can be beneficial for control of UFH systems, and how the implementation is done. Furthermore, results are presented with a standard hysteresis control, an RL control, and an RL control with tolerance control. These results show that the effect of tolerance control in these types of systems is significant. Finally, we discuss the challenges there are for a real-world implementation of RL-based control in UFH system.

KW - Deep Reinforcement Learning

KW - Artificial Intelligence (AI)

KW - Deep Learning

KW - HVAC System

KW - Hysteresis

KW - tolerance control

KW - HVAC-systems

KW - Tolerance Control

KW - Underfloor Heating

UR - http://www.scopus.com/inward/record.url?scp=85082596603&partnerID=8YFLogxK

U2 - 10.1109/SII46433.2020.9026189

DO - 10.1109/SII46433.2020.9026189

M3 - Article in proceeding

T3 - Proceedings of the 2020 IEEE/SICE International Symposium on System Integration

SP - 938

EP - 942

BT - IEEE/SICE International Symposium on System Integration

PB - IEEE (Institute of Electrical and Electronics Engineers)

T2 - IEEE/SICE International Symposium on System Integration

Y2 - 12 January 2020 through 15 January 2020

ER -

Blad C, Kallesøe C , Bøgh S. Control of HVAC-Systems Using Reinforcement Learning With Hysteresis and Tolerance Control. In IEEE/SICE International Symposium on System Integration. IEEE (Institute of Electrical and Electronics Engineers). 2020. p. 938-942. 9026189. (Proceedings of the 2020 IEEE/SICE International Symposium on System Integration). doi: 10.1109/SII46433.2020.9026189

Control of HVAC-Systems Using Reinforcement Learning With Hysteresis and Tolerance Control

Abstract

Conference

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Reinforcement Learning Based Control for Underfloor Heating

Cite this

Control of HVAC-Systems Using Reinforcement Learning With Hysteresis and Tolerance Control

Abstract

Conference

Keywords

Access to Document

AUB Link

Other files and links

Fingerprint

Projects

Reinforcement Learning Based Control for Underfloor Heating

Cite this