An opportunistic activity-sensing approach to save energy in office buildings

M. Milenkovic, O.D. Amft

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

47 Citations (Scopus)

Abstract

In this work, we recognised office worker activities that are relevant for energy-related control of appliances and building systems using sensors that are commonly installed in new or refurbished office buildings. We considered desk-related activities and people count in office rooms, structured into desk- and room-cells. Recognition was performed using finite state machines (FSMs) and probabilistic layered hidden Markov models (LHMMs). We evaluated our approach in a real living-lab office, including three private and multi-person office rooms. As example devices, we used different ceiling-mounted PIR sensors based on the EnOcean platform and plug-in power meters. In at least five days of study data per office room, including reference sensor data and occupant annotations, we confirmed that activities can be recognised using these sensors. For computer and desk work, an overall recognition accuracy of 95% was achieved. People count was estimated at 87% and 78% for the best-performing two office rooms. We furthermore present building simulation results that compare different control strategies. Compared to modern BEMS, our results show that 21.9% and 19.5% of electrical energy can be saved for controls based on recognised desk activity and estimated people count, respectively. These results confirm the relevance of building energy management based on activity sensing.
Original languageEnglish
Title of host publicationProceedings of the Fourth International Conference on Future Energy Systems (e-Energy '13), 22-24 May 2013, Berkeley, California
Place of PublicationNew York
PublisherAssociation for Computing Machinery, Inc
Pages247-258
ISBN (Print)978-1-4503-2052-8
DOIs
Publication statusPublished - 2013
Eventconference; The fourth international conference on Future energy systems (e-Energy '13); 2013-05-22; 2013-05-24 -
Duration: 22 May 201324 May 2013

Conference

Conferenceconference; The fourth international conference on Future energy systems (e-Energy '13); 2013-05-22; 2013-05-24
Period22/05/1324/05/13
OtherThe fourth international conference on Future energy systems (e-Energy '13)

Fingerprint

Office buildings
Sensors
Ceilings
Energy management
Finite automata
Hidden Markov models

Cite this

Milenkovic, M., & Amft, O. D. (2013). An opportunistic activity-sensing approach to save energy in office buildings. In Proceedings of the Fourth International Conference on Future Energy Systems (e-Energy '13), 22-24 May 2013, Berkeley, California (pp. 247-258). New York: Association for Computing Machinery, Inc. https://doi.org/10.1145/2487166.2487194
Milenkovic, M. ; Amft, O.D. / An opportunistic activity-sensing approach to save energy in office buildings. Proceedings of the Fourth International Conference on Future Energy Systems (e-Energy '13), 22-24 May 2013, Berkeley, California. New York : Association for Computing Machinery, Inc, 2013. pp. 247-258
@inproceedings{1cb3d056d08547349d7a7612050a1c2e,
title = "An opportunistic activity-sensing approach to save energy in office buildings",
abstract = "In this work, we recognised office worker activities that are relevant for energy-related control of appliances and building systems using sensors that are commonly installed in new or refurbished office buildings. We considered desk-related activities and people count in office rooms, structured into desk- and room-cells. Recognition was performed using finite state machines (FSMs) and probabilistic layered hidden Markov models (LHMMs). We evaluated our approach in a real living-lab office, including three private and multi-person office rooms. As example devices, we used different ceiling-mounted PIR sensors based on the EnOcean platform and plug-in power meters. In at least five days of study data per office room, including reference sensor data and occupant annotations, we confirmed that activities can be recognised using these sensors. For computer and desk work, an overall recognition accuracy of 95{\%} was achieved. People count was estimated at 87{\%} and 78{\%} for the best-performing two office rooms. We furthermore present building simulation results that compare different control strategies. Compared to modern BEMS, our results show that 21.9{\%} and 19.5{\%} of electrical energy can be saved for controls based on recognised desk activity and estimated people count, respectively. These results confirm the relevance of building energy management based on activity sensing.",
author = "M. Milenkovic and O.D. Amft",
year = "2013",
doi = "10.1145/2487166.2487194",
language = "English",
isbn = "978-1-4503-2052-8",
pages = "247--258",
booktitle = "Proceedings of the Fourth International Conference on Future Energy Systems (e-Energy '13), 22-24 May 2013, Berkeley, California",
publisher = "Association for Computing Machinery, Inc",
address = "United States",

}

Milenkovic, M & Amft, OD 2013, An opportunistic activity-sensing approach to save energy in office buildings. in Proceedings of the Fourth International Conference on Future Energy Systems (e-Energy '13), 22-24 May 2013, Berkeley, California. Association for Computing Machinery, Inc, New York, pp. 247-258, conference; The fourth international conference on Future energy systems (e-Energy '13); 2013-05-22; 2013-05-24, 22/05/13. https://doi.org/10.1145/2487166.2487194

An opportunistic activity-sensing approach to save energy in office buildings. / Milenkovic, M.; Amft, O.D.

Proceedings of the Fourth International Conference on Future Energy Systems (e-Energy '13), 22-24 May 2013, Berkeley, California. New York : Association for Computing Machinery, Inc, 2013. p. 247-258.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - An opportunistic activity-sensing approach to save energy in office buildings

AU - Milenkovic, M.

AU - Amft, O.D.

PY - 2013

Y1 - 2013

N2 - In this work, we recognised office worker activities that are relevant for energy-related control of appliances and building systems using sensors that are commonly installed in new or refurbished office buildings. We considered desk-related activities and people count in office rooms, structured into desk- and room-cells. Recognition was performed using finite state machines (FSMs) and probabilistic layered hidden Markov models (LHMMs). We evaluated our approach in a real living-lab office, including three private and multi-person office rooms. As example devices, we used different ceiling-mounted PIR sensors based on the EnOcean platform and plug-in power meters. In at least five days of study data per office room, including reference sensor data and occupant annotations, we confirmed that activities can be recognised using these sensors. For computer and desk work, an overall recognition accuracy of 95% was achieved. People count was estimated at 87% and 78% for the best-performing two office rooms. We furthermore present building simulation results that compare different control strategies. Compared to modern BEMS, our results show that 21.9% and 19.5% of electrical energy can be saved for controls based on recognised desk activity and estimated people count, respectively. These results confirm the relevance of building energy management based on activity sensing.

AB - In this work, we recognised office worker activities that are relevant for energy-related control of appliances and building systems using sensors that are commonly installed in new or refurbished office buildings. We considered desk-related activities and people count in office rooms, structured into desk- and room-cells. Recognition was performed using finite state machines (FSMs) and probabilistic layered hidden Markov models (LHMMs). We evaluated our approach in a real living-lab office, including three private and multi-person office rooms. As example devices, we used different ceiling-mounted PIR sensors based on the EnOcean platform and plug-in power meters. In at least five days of study data per office room, including reference sensor data and occupant annotations, we confirmed that activities can be recognised using these sensors. For computer and desk work, an overall recognition accuracy of 95% was achieved. People count was estimated at 87% and 78% for the best-performing two office rooms. We furthermore present building simulation results that compare different control strategies. Compared to modern BEMS, our results show that 21.9% and 19.5% of electrical energy can be saved for controls based on recognised desk activity and estimated people count, respectively. These results confirm the relevance of building energy management based on activity sensing.

U2 - 10.1145/2487166.2487194

DO - 10.1145/2487166.2487194

M3 - Conference contribution

SN - 978-1-4503-2052-8

SP - 247

EP - 258

BT - Proceedings of the Fourth International Conference on Future Energy Systems (e-Energy '13), 22-24 May 2013, Berkeley, California

PB - Association for Computing Machinery, Inc

CY - New York

ER -

Milenkovic M, Amft OD. An opportunistic activity-sensing approach to save energy in office buildings. In Proceedings of the Fourth International Conference on Future Energy Systems (e-Energy '13), 22-24 May 2013, Berkeley, California. New York: Association for Computing Machinery, Inc. 2013. p. 247-258 https://doi.org/10.1145/2487166.2487194