Event-triggered reinforcement learning; an application to buildings’ micro-climate control
Smart buildings have great potential for shaping an energy-efficient, sustainable, and more economic future for our planet as buildings account for approximately 40% of the global energy consumption. However, most learning methods for micro-climate control in buildings are based on Markov Decision P...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
RWTH Aachen University
2021
|
| Online Access: | https://hdl.handle.net/1721.1/129974 |
| _version_ | 1811077126045040640 |
|---|---|
| author | Haji Hosseinloo, Ashkan Dahleh, Munther A |
| author2 | Massachusetts Institute of Technology. Laboratory for Information and Decision Systems |
| author_facet | Massachusetts Institute of Technology. Laboratory for Information and Decision Systems Haji Hosseinloo, Ashkan Dahleh, Munther A |
| author_sort | Haji Hosseinloo, Ashkan |
| collection | MIT |
| description | Smart buildings have great potential for shaping an energy-efficient, sustainable, and more economic future for our planet as buildings account for approximately 40% of the global energy consumption. However, most learning methods for micro-climate control in buildings are based on Markov Decision Processes with fixed transition times that suffer from high variance in the learning phase. Furthermore, ignoring its continuing-task nature the micro-climate control problem is often modeled and solved as an episodic-task problem with discounted rewards. This can result in a wrong optimization solution. To overcome these issues we propose an event-triggered learning control and formulate it based on Semi-Markov Decision Processes with variable transition times and in an average-reward setting. We show via simulation the efficacy of our approach in controlling the micro-climate of a single-zone building. |
| first_indexed | 2024-09-23T10:38:17Z |
| format | Article |
| id | mit-1721.1/129974 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T10:38:17Z |
| publishDate | 2021 |
| publisher | RWTH Aachen University |
| record_format | dspace |
| spelling | mit-1721.1/1299742022-09-30T21:57:26Z Event-triggered reinforcement learning; an application to buildings’ micro-climate control Haji Hosseinloo, Ashkan Dahleh, Munther A Massachusetts Institute of Technology. Laboratory for Information and Decision Systems Smart buildings have great potential for shaping an energy-efficient, sustainable, and more economic future for our planet as buildings account for approximately 40% of the global energy consumption. However, most learning methods for micro-climate control in buildings are based on Markov Decision Processes with fixed transition times that suffer from high variance in the learning phase. Furthermore, ignoring its continuing-task nature the micro-climate control problem is often modeled and solved as an episodic-task problem with discounted rewards. This can result in a wrong optimization solution. To overcome these issues we propose an event-triggered learning control and formulate it based on Semi-Markov Decision Processes with variable transition times and in an average-reward setting. We show via simulation the efficacy of our approach in controlling the micro-climate of a single-zone building. 2021-02-23T16:15:43Z 2021-02-23T16:15:43Z 2020-03 2020-12-07T15:30:40Z Article http://purl.org/eprint/type/ConferencePaper 1613-0073 https://hdl.handle.net/1721.1/129974 Haji Hosseinloo, Ashkan and Munther Dahleh. et al. “Event-triggered reinforcement learning; an application to buildings’ micro-climate control.” Paper in the CEUR Workshop Proceedings, 2587, AAAI Spring Symposium: MLPS, 2020, virtual meeting, March 23-25 2020, RWTH Aachen University © 2020 The Author(s) en http://ceur-ws.org/Vol-2587/article_6.pdf CEUR Workshop Proceedings Creative Commons Attribution 4.0 International license https://creativecommons.org/licenses/by/4.0/ application/pdf RWTH Aachen University CEUR |
| spellingShingle | Haji Hosseinloo, Ashkan Dahleh, Munther A Event-triggered reinforcement learning; an application to buildings’ micro-climate control |
| title | Event-triggered reinforcement learning; an application to buildings’ micro-climate control |
| title_full | Event-triggered reinforcement learning; an application to buildings’ micro-climate control |
| title_fullStr | Event-triggered reinforcement learning; an application to buildings’ micro-climate control |
| title_full_unstemmed | Event-triggered reinforcement learning; an application to buildings’ micro-climate control |
| title_short | Event-triggered reinforcement learning; an application to buildings’ micro-climate control |
| title_sort | event triggered reinforcement learning an application to buildings micro climate control |
| url | https://hdl.handle.net/1721.1/129974 |
| work_keys_str_mv | AT hajihosseinlooashkan eventtriggeredreinforcementlearninganapplicationtobuildingsmicroclimatecontrol AT dahlehmunthera eventtriggeredreinforcementlearninganapplicationtobuildingsmicroclimatecontrol |