A simulation study on energy optimization in building control with reinforcement learning

Bolt, Peter; Ziebart, Volker; Jaeger, Christian; Schmid, Nicolas; Stadelmann, Thilo; Füchslin, Rudolf Marcel

doi:10.21256/zhaw-31129

Please use this identifier to cite or link to this item: https://doi.org/10.21256/zhaw-31129

Publication type:	Conference paper
Type of review:	Peer review (publication)
Title:	A simulation study on energy optimization in building control with reinforcement learning
Authors:	Bolt, Peter Ziebart, Volker Jaeger, Christian Schmid, Nicolas Stadelmann, Thilo Füchslin, Rudolf Marcel
et. al:	No
DOI:	10.21256/zhaw-31129
Conference details:	11th IAPR TC 3 Workshop on Artificial Neural Networks for Pattern Recognition (ANNPR'24), Montreal, Canada, 10-12 October 2024
Issue Date:	10-Oct-2024
Publisher / Ed. Institution:	ZHAW Zürcher Hochschule für Angewandte Wissenschaften
Language:	English
Subjects:	Smart building; Building control; Reinforcement learning
Subject (DDC):	006: Special computer methods 621.04: Energy engineering
Abstract:	We propose and evaluate a deep reinforcement learning control paradigm for building energy systems. In comparison to other advanced control techniques, namely Model Predictive Control, the reinforcement learning paradigm avoids the costs and uncertainties associated with the requirement for a control-oriented model. We apply a mixed agent for the Proximal Policy Optimization algorithm, similar to the algorithm proposed in as well as a non-discounted finite horizon optimization scheme. We investigate the capabilities of the proposed reinforcement learning controller regarding energy efficiency, comparing it against the most widely used rule-based control paradigm as a baseline controller. We verify our proposed paradigm in a simulation study with building models implemented in Dymola.
URI:	https://digitalcollection.zhaw.ch/handle/11475/31129
Fulltext version:	Accepted version
License (according to publishing contract):	Licence according to publishing contract
Departement:	School of Engineering
Organisational Unit:	Centre for Artificial Intelligence (CAI) Institute of Applied Mathematics and Physics (IAMP)
Published as part of the ZHAW project:	Machbarkeitsstudie Reinforcement Learning Control für Heizsysteme
Appears in collections:	Publikationen School of Engineering

Files in This Item:

File	Description	Size	Format
2024_Bolt-etal_Simulation-study-energy-optimization_ANNPR.pdf	Accepted Version	1.02 MB	Adobe PDF	View/Open

Show full item record

Bolt, P., Ziebart, V., Jaeger, C., Schmid, N., Stadelmann, T., & Füchslin, R. M. (2024, October 10). A simulation study on energy optimization in building control with reinforcement learning. 11th IAPR TC 3 Workshop on Artificial Neural Networks for Pattern Recognition (ANNPR′24), Montreal, Canada, 10-12 October 2024. https://doi.org/10.21256/zhaw-31129

Bolt, P. et al. (2024) ‘A simulation study on energy optimization in building control with reinforcement learning’, in 11th IAPR TC 3 Workshop on Artificial Neural Networks for Pattern Recognition (ANNPR′24), Montreal, Canada, 10-12 October 2024. ZHAW Zürcher Hochschule für Angewandte Wissenschaften. Available at: https://doi.org/10.21256/zhaw-31129.

P. Bolt, V. Ziebart, C. Jaeger, N. Schmid, T. Stadelmann, and R. M. Füchslin, “A simulation study on energy optimization in building control with reinforcement learning,” in 11th IAPR TC 3 Workshop on Artificial Neural Networks for Pattern Recognition (ANNPR′24), Montreal, Canada, 10-12 October 2024, Oct. 2024. doi: 10.21256/zhaw-31129.

BOLT, Peter, Volker ZIEBART, Christian JAEGER, Nicolas SCHMID, Thilo STADELMANN und Rudolf Marcel FÜCHSLIN, 2024. A simulation study on energy optimization in building control with reinforcement learning. In: 11th IAPR TC 3 Workshop on Artificial Neural Networks for Pattern Recognition (ANNPR′24), Montreal, Canada, 10-12 October 2024. Conference paper. ZHAW Zürcher Hochschule für Angewandte Wissenschaften. 10 Oktober 2024

Bolt, Peter, Volker Ziebart, Christian Jaeger, Nicolas Schmid, Thilo Stadelmann, and Rudolf Marcel Füchslin. 2024. “A Simulation Study on Energy Optimization in Building Control with Reinforcement Learning.” Conference paper. In 11th IAPR TC 3 Workshop on Artificial Neural Networks for Pattern Recognition (ANNPR′24), Montreal, Canada, 10-12 October 2024. ZHAW Zürcher Hochschule für Angewandte Wissenschaften. https://doi.org/10.21256/zhaw-31129.

Bolt, Peter, et al. “A Simulation Study on Energy Optimization in Building Control with Reinforcement Learning.” 11th IAPR TC 3 Workshop on Artificial Neural Networks for Pattern Recognition (ANNPR′24), Montreal, Canada, 10-12 October 2024, ZHAW Zürcher Hochschule für Angewandte Wissenschaften, 2024, https://doi.org/10.21256/zhaw-31129.