Structured dictionary learning for energy disaggregation

Shalini Pandey, George Karypis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The increased awareness regarding the impact of energy consumption on the environment has led to an increased focus on reducing energy consumption. Feedback on the appliance level energy consumption can help in reducing the energy demands of the consumers. Energy disaggregation techniques are used to obtain the appliance level energy consumption from the aggregated energy consumption of a house. These techniques extract the energy consumption of an individual appliance as features and hence face the challenge of distinguishing two similar energy consuming devices. To address this challenge we develop methods that leverage the fact that some devices tend to operate concurrently at specific operation modes. The aggregated energy consumption patterns of a subgroup of devices allows us to identify the concurrent operating modes of devices in the subgroup. Thus, we design hierarchical methods to replace the task of overall energy disaggregation among the devices with a recursive disaggregation task involving device subgroups. Experiments on two real-world datasets show that our methods lead to improved performance as compared to baseline. One of our approaches, Greedy based Device Decomposition Method (GDDM) achieved up to 23.8%, 10% and 59.3% improvement in terms of micro-averaged f score, macro-averaged f score and Normalized Disaggregation Error (NDE), respectively.

Original languageEnglish (US)
Title of host publicatione-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems
PublisherAssociation for Computing Machinery, Inc
Pages24-34
Number of pages11
ISBN (Electronic)9781450366717
DOIs
StatePublished - Jun 15 2019
Event10th ACM International Conference on Future Energy Systems, e-Energy 2019 - Phoenix, United States
Duration: Jun 25 2019Jun 28 2019

Publication series

Namee-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems

Conference

Conference10th ACM International Conference on Future Energy Systems, e-Energy 2019
CountryUnited States
CityPhoenix
Period6/25/196/28/19

Fingerprint

Glossaries
Energy utilization
Macros
Decomposition
Feedback
Experiments

Cite this

Pandey, S., & Karypis, G. (2019). Structured dictionary learning for energy disaggregation. In e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems (pp. 24-34). (e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems). Association for Computing Machinery, Inc. https://doi.org/10.1145/3307772.3328301

Structured dictionary learning for energy disaggregation. / Pandey, Shalini; Karypis, George.

e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems. Association for Computing Machinery, Inc, 2019. p. 24-34 (e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Pandey, S & Karypis, G 2019, Structured dictionary learning for energy disaggregation. in e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems. e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems, Association for Computing Machinery, Inc, pp. 24-34, 10th ACM International Conference on Future Energy Systems, e-Energy 2019, Phoenix, United States, 6/25/19. https://doi.org/10.1145/3307772.3328301
Pandey S, Karypis G. Structured dictionary learning for energy disaggregation. In e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems. Association for Computing Machinery, Inc. 2019. p. 24-34. (e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems). https://doi.org/10.1145/3307772.3328301
Pandey, Shalini ; Karypis, George. / Structured dictionary learning for energy disaggregation. e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems. Association for Computing Machinery, Inc, 2019. pp. 24-34 (e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems).
@inproceedings{4d47216dee9d42bbbd9dc8bd98e1561c,
title = "Structured dictionary learning for energy disaggregation",
abstract = "The increased awareness regarding the impact of energy consumption on the environment has led to an increased focus on reducing energy consumption. Feedback on the appliance level energy consumption can help in reducing the energy demands of the consumers. Energy disaggregation techniques are used to obtain the appliance level energy consumption from the aggregated energy consumption of a house. These techniques extract the energy consumption of an individual appliance as features and hence face the challenge of distinguishing two similar energy consuming devices. To address this challenge we develop methods that leverage the fact that some devices tend to operate concurrently at specific operation modes. The aggregated energy consumption patterns of a subgroup of devices allows us to identify the concurrent operating modes of devices in the subgroup. Thus, we design hierarchical methods to replace the task of overall energy disaggregation among the devices with a recursive disaggregation task involving device subgroups. Experiments on two real-world datasets show that our methods lead to improved performance as compared to baseline. One of our approaches, Greedy based Device Decomposition Method (GDDM) achieved up to 23.8{\%}, 10{\%} and 59.3{\%} improvement in terms of micro-averaged f score, macro-averaged f score and Normalized Disaggregation Error (NDE), respectively.",
author = "Shalini Pandey and George Karypis",
year = "2019",
month = "6",
day = "15",
doi = "10.1145/3307772.3328301",
language = "English (US)",
series = "e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems",
publisher = "Association for Computing Machinery, Inc",
pages = "24--34",
booktitle = "e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems",

}

TY - GEN

T1 - Structured dictionary learning for energy disaggregation

AU - Pandey, Shalini

AU - Karypis, George

PY - 2019/6/15

Y1 - 2019/6/15

N2 - The increased awareness regarding the impact of energy consumption on the environment has led to an increased focus on reducing energy consumption. Feedback on the appliance level energy consumption can help in reducing the energy demands of the consumers. Energy disaggregation techniques are used to obtain the appliance level energy consumption from the aggregated energy consumption of a house. These techniques extract the energy consumption of an individual appliance as features and hence face the challenge of distinguishing two similar energy consuming devices. To address this challenge we develop methods that leverage the fact that some devices tend to operate concurrently at specific operation modes. The aggregated energy consumption patterns of a subgroup of devices allows us to identify the concurrent operating modes of devices in the subgroup. Thus, we design hierarchical methods to replace the task of overall energy disaggregation among the devices with a recursive disaggregation task involving device subgroups. Experiments on two real-world datasets show that our methods lead to improved performance as compared to baseline. One of our approaches, Greedy based Device Decomposition Method (GDDM) achieved up to 23.8%, 10% and 59.3% improvement in terms of micro-averaged f score, macro-averaged f score and Normalized Disaggregation Error (NDE), respectively.

AB - The increased awareness regarding the impact of energy consumption on the environment has led to an increased focus on reducing energy consumption. Feedback on the appliance level energy consumption can help in reducing the energy demands of the consumers. Energy disaggregation techniques are used to obtain the appliance level energy consumption from the aggregated energy consumption of a house. These techniques extract the energy consumption of an individual appliance as features and hence face the challenge of distinguishing two similar energy consuming devices. To address this challenge we develop methods that leverage the fact that some devices tend to operate concurrently at specific operation modes. The aggregated energy consumption patterns of a subgroup of devices allows us to identify the concurrent operating modes of devices in the subgroup. Thus, we design hierarchical methods to replace the task of overall energy disaggregation among the devices with a recursive disaggregation task involving device subgroups. Experiments on two real-world datasets show that our methods lead to improved performance as compared to baseline. One of our approaches, Greedy based Device Decomposition Method (GDDM) achieved up to 23.8%, 10% and 59.3% improvement in terms of micro-averaged f score, macro-averaged f score and Normalized Disaggregation Error (NDE), respectively.

UR - http://www.scopus.com/inward/record.url?scp=85068655028&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068655028&partnerID=8YFLogxK

U2 - 10.1145/3307772.3328301

DO - 10.1145/3307772.3328301

M3 - Conference contribution

T3 - e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems

SP - 24

EP - 34

BT - e-Energy 2019 - Proceedings of the 10th ACM International Conference on Future Energy Systems

PB - Association for Computing Machinery, Inc

ER -