Recognizing energy-related activities using sensors commonly installed in office buildings

M. Milenkovic, O.D. Amft

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

30 Citations (Scopus)

Abstract

Automated control based on user activities and preferences could reduce energy consumption of office buildings. In this paper, we investigated generalisation properties of an office activity recognition approach using sensors that are frequently installed in modern and refurbished office buildings. In particular, per-desk passive infrared (PIR) sensors and power plug meters were considered in an evaluation study including more than 100 hours of data from both, a single-person room and a three-user multi-person office room. Layered hidden Markov models (LHMM) were used for the recognition. Results showed that 30 hours and 50 hours of training data were needed to achieve robust recognition of desk activities and estimate people count, respectively. The recognition can be performed independent of a particular occupant desk. In further simulations considering different energy profiles, we show how energy consumption due to lighting and office appliances is related to occupant behaviour.
Original languageEnglish
Title of host publicationProceedings of the 4th International Conference on Ambient Systems, Networks and Technologies (ANT 2013) and the 3rd International Conference on Sustainable Energy Information Technology (SEIT-2013), 25-28 June 2013, Halifax, Canada
PublisherElsevier/Procedia Computer Science
Pages669-677
DOIs
Publication statusPublished - 2013
Eventconference; The 3rd International Conference on Sustainable Energy Information Technology (SEIT '13); 2013-06-25; 2013-06-28 -
Duration: 25 Jun 201328 Jun 2013

Publication series

NameProcedia Computer Science
ISSN (Print)1877-0509

Conference

Conferenceconference; The 3rd International Conference on Sustainable Energy Information Technology (SEIT '13); 2013-06-25; 2013-06-28
Period25/06/1328/06/13
OtherThe 3rd International Conference on Sustainable Energy Information Technology (SEIT '13)

Fingerprint

Office buildings
Energy utilization
Sensors
Hidden Markov models
Lighting
Infrared radiation

Cite this

Milenkovic, M., & Amft, O. D. (2013). Recognizing energy-related activities using sensors commonly installed in office buildings. In Proceedings of the 4th International Conference on Ambient Systems, Networks and Technologies (ANT 2013) and the 3rd International Conference on Sustainable Energy Information Technology (SEIT-2013), 25-28 June 2013, Halifax, Canada (pp. 669-677). (Procedia Computer Science). Elsevier/Procedia Computer Science. https://doi.org/10.1016/j.procs.2013.06.089
Milenkovic, M. ; Amft, O.D. / Recognizing energy-related activities using sensors commonly installed in office buildings. Proceedings of the 4th International Conference on Ambient Systems, Networks and Technologies (ANT 2013) and the 3rd International Conference on Sustainable Energy Information Technology (SEIT-2013), 25-28 June 2013, Halifax, Canada. Elsevier/Procedia Computer Science, 2013. pp. 669-677 (Procedia Computer Science).
@inproceedings{ebc3fadb8b8348589c668ee83ce7136b,
title = "Recognizing energy-related activities using sensors commonly installed in office buildings",
abstract = "Automated control based on user activities and preferences could reduce energy consumption of office buildings. In this paper, we investigated generalisation properties of an office activity recognition approach using sensors that are frequently installed in modern and refurbished office buildings. In particular, per-desk passive infrared (PIR) sensors and power plug meters were considered in an evaluation study including more than 100 hours of data from both, a single-person room and a three-user multi-person office room. Layered hidden Markov models (LHMM) were used for the recognition. Results showed that 30 hours and 50 hours of training data were needed to achieve robust recognition of desk activities and estimate people count, respectively. The recognition can be performed independent of a particular occupant desk. In further simulations considering different energy profiles, we show how energy consumption due to lighting and office appliances is related to occupant behaviour.",
author = "M. Milenkovic and O.D. Amft",
year = "2013",
doi = "10.1016/j.procs.2013.06.089",
language = "English",
series = "Procedia Computer Science",
publisher = "Elsevier/Procedia Computer Science",
pages = "669--677",
booktitle = "Proceedings of the 4th International Conference on Ambient Systems, Networks and Technologies (ANT 2013) and the 3rd International Conference on Sustainable Energy Information Technology (SEIT-2013), 25-28 June 2013, Halifax, Canada",

}

Milenkovic, M & Amft, OD 2013, Recognizing energy-related activities using sensors commonly installed in office buildings. in Proceedings of the 4th International Conference on Ambient Systems, Networks and Technologies (ANT 2013) and the 3rd International Conference on Sustainable Energy Information Technology (SEIT-2013), 25-28 June 2013, Halifax, Canada. Procedia Computer Science, Elsevier/Procedia Computer Science, pp. 669-677, conference; The 3rd International Conference on Sustainable Energy Information Technology (SEIT '13); 2013-06-25; 2013-06-28, 25/06/13. https://doi.org/10.1016/j.procs.2013.06.089

Recognizing energy-related activities using sensors commonly installed in office buildings. / Milenkovic, M.; Amft, O.D.

Proceedings of the 4th International Conference on Ambient Systems, Networks and Technologies (ANT 2013) and the 3rd International Conference on Sustainable Energy Information Technology (SEIT-2013), 25-28 June 2013, Halifax, Canada. Elsevier/Procedia Computer Science, 2013. p. 669-677 (Procedia Computer Science).

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

TY - GEN

T1 - Recognizing energy-related activities using sensors commonly installed in office buildings

AU - Milenkovic, M.

AU - Amft, O.D.

PY - 2013

Y1 - 2013

N2 - Automated control based on user activities and preferences could reduce energy consumption of office buildings. In this paper, we investigated generalisation properties of an office activity recognition approach using sensors that are frequently installed in modern and refurbished office buildings. In particular, per-desk passive infrared (PIR) sensors and power plug meters were considered in an evaluation study including more than 100 hours of data from both, a single-person room and a three-user multi-person office room. Layered hidden Markov models (LHMM) were used for the recognition. Results showed that 30 hours and 50 hours of training data were needed to achieve robust recognition of desk activities and estimate people count, respectively. The recognition can be performed independent of a particular occupant desk. In further simulations considering different energy profiles, we show how energy consumption due to lighting and office appliances is related to occupant behaviour.

AB - Automated control based on user activities and preferences could reduce energy consumption of office buildings. In this paper, we investigated generalisation properties of an office activity recognition approach using sensors that are frequently installed in modern and refurbished office buildings. In particular, per-desk passive infrared (PIR) sensors and power plug meters were considered in an evaluation study including more than 100 hours of data from both, a single-person room and a three-user multi-person office room. Layered hidden Markov models (LHMM) were used for the recognition. Results showed that 30 hours and 50 hours of training data were needed to achieve robust recognition of desk activities and estimate people count, respectively. The recognition can be performed independent of a particular occupant desk. In further simulations considering different energy profiles, we show how energy consumption due to lighting and office appliances is related to occupant behaviour.

U2 - 10.1016/j.procs.2013.06.089

DO - 10.1016/j.procs.2013.06.089

M3 - Conference contribution

T3 - Procedia Computer Science

SP - 669

EP - 677

BT - Proceedings of the 4th International Conference on Ambient Systems, Networks and Technologies (ANT 2013) and the 3rd International Conference on Sustainable Energy Information Technology (SEIT-2013), 25-28 June 2013, Halifax, Canada

PB - Elsevier/Procedia Computer Science

ER -

Milenkovic M, Amft OD. Recognizing energy-related activities using sensors commonly installed in office buildings. In Proceedings of the 4th International Conference on Ambient Systems, Networks and Technologies (ANT 2013) and the 3rd International Conference on Sustainable Energy Information Technology (SEIT-2013), 25-28 June 2013, Halifax, Canada. Elsevier/Procedia Computer Science. 2013. p. 669-677. (Procedia Computer Science). https://doi.org/10.1016/j.procs.2013.06.089