Appliance Classification using BiLSTM Neural Networks and Feature Extraction

Martha T. Correa-Delval; Hongjian Sun; Peter C. Matthews; Jing Jiang

doi:10.1109/ISGTEurope52324.2021.9640061

Appliance Classification using BiLSTM Neural Networks and Feature Extraction

Martha T. Correa-Delval, Hongjian Sun, Peter C. Matthews, Jing Jiang

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

7 Citations (Scopus)

64 Downloads (Pure)

Abstract

One significant challenge in Non-Intrusive Load Monitoring (NILM) is to identify and classify active appliances used in a building. This research focuses on the classifying process, exploring different approaches for the feature extraction of the appliances’ power load to improve the classification accuracy. In this paper, we present a new method - Spectral Entropy and Instantaneous Frequency-based Bidirectional Long Short Term Memory (SE-IF BiLSTM). It uses feature extraction from the power load to obtain information, such as instant frequency, spectral entropy, spectrogram, Mel spectrogram and signal variation, to feed BiLSTM Neural Network. We also test different options for the BiLSTM to decide the most optimal settings. This method improves the classification performance, achieving up to 98.57% classification accuracy.

Original language	English
Title of host publication	2021 IEEE PES Innovative Smart Grid Technologies Europe (ISGT Europe)
Subtitle of host publication	Smart Grids: Toward a Carbon-free Future
Place of Publication	Piscataway, US
Publisher	IEEE
Pages	180-184
Number of pages	5
ISBN (Electronic)	9781665448758
ISBN (Print)	9781665448765
DOIs	https://doi.org/10.1109/ISGTEurope52324.2021.9640061
Publication status	Published - 18 Oct 2021
Event	ISGT Europe 2021: IEEE PES Innovative Smart Grid Technologies: Smart Grids: Toward a Carbon-free Future - Virtual, Aalto University, Espoo, Finland Duration: 18 Oct 2021 → 21 Oct 2021 https://ieee-isgt-europe.org/

Conference

Conference	ISGT Europe 2021: IEEE PES Innovative Smart Grid Technologies
Abbreviated title	IEEE PES ISGT EUROPE 2021
Country/Territory	Finland
City	Espoo
Period	18/10/21 → 21/10/21
Internet address	https://ieee-isgt-europe.org/

Keywords

BILSTM
Appliance Classification
NILM

Access to Document

10.1109/ISGTEurope52324.2021.9640061

paper_59Accepted author manuscript, 317 KB

Cite this

@inproceedings{b4a6e2c593fd48808c17f82cf531ef1d,

title = "Appliance Classification using BiLSTM Neural Networks and Feature Extraction",

abstract = "One significant challenge in Non-Intrusive Load Monitoring (NILM) is to identify and classify active appliances used in a building. This research focuses on the classifying process, exploring different approaches for the feature extraction of the appliances{\textquoteright} power load to improve the classification accuracy. In this paper, we present a new method - Spectral Entropy and Instantaneous Frequency-based Bidirectional Long Short Term Memory (SE-IF BiLSTM). It uses feature extraction from the power load to obtain information, such as instant frequency, spectral entropy, spectrogram, Mel spectrogram and signal variation, to feed BiLSTM Neural Network. We also test different options for the BiLSTM to decide the most optimal settings. This method improves the classification performance, achieving up to 98.57\% classification accuracy.",

keywords = "BILSTM, Appliance Classification, NILM",

author = "Correa-Delval, \{Martha T.\} and Hongjian Sun and Matthews, \{Peter C.\} and Jing Jiang",

note = "This work is supported by the CONACyT-SENER (Mexican National Council for Science and Technology, along with the Secretary of Energy) energy sustainability programme.; ISGT Europe 2021: IEEE PES Innovative Smart Grid Technologies : Smart Grids: Toward a Carbon-free Future, IEEE PES ISGT EUROPE 2021 ; Conference date: 18-10-2021 Through 21-10-2021",

year = "2021",

month = oct,

day = "18",

doi = "10.1109/ISGTEurope52324.2021.9640061",

language = "English",

isbn = "9781665448765",

pages = "180--184",

booktitle = "2021 IEEE PES Innovative Smart Grid Technologies Europe (ISGT Europe)",

publisher = "IEEE",

address = "United States",

url = "https://ieee-isgt-europe.org/",

}

Correa-Delval, MT, Sun, H, Matthews, PC & Jiang, J 2021, Appliance Classification using BiLSTM Neural Networks and Feature Extraction. in 2021 IEEE PES Innovative Smart Grid Technologies Europe (ISGT Europe): Smart Grids: Toward a Carbon-free Future., 9640061, IEEE, Piscataway, US, pp. 180-184, ISGT Europe 2021: IEEE PES Innovative Smart Grid Technologies, Espoo, Finland, 18/10/21. https://doi.org/10.1109/ISGTEurope52324.2021.9640061

Appliance Classification using BiLSTM Neural Networks and Feature Extraction. / Correa-Delval, Martha T.; Sun, Hongjian; Matthews, Peter C. et al.
2021 IEEE PES Innovative Smart Grid Technologies Europe (ISGT Europe): Smart Grids: Toward a Carbon-free Future. Piscataway, US: IEEE, 2021. p. 180-184 9640061.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Appliance Classification using BiLSTM Neural Networks and Feature Extraction

AU - Correa-Delval, Martha T.

AU - Sun, Hongjian

AU - Matthews, Peter C.

AU - Jiang, Jing

N1 - This work is supported by the CONACyT-SENER (Mexican National Council for Science and Technology, along with the Secretary of Energy) energy sustainability programme.

PY - 2021/10/18

Y1 - 2021/10/18

N2 - One significant challenge in Non-Intrusive Load Monitoring (NILM) is to identify and classify active appliances used in a building. This research focuses on the classifying process, exploring different approaches for the feature extraction of the appliances’ power load to improve the classification accuracy. In this paper, we present a new method - Spectral Entropy and Instantaneous Frequency-based Bidirectional Long Short Term Memory (SE-IF BiLSTM). It uses feature extraction from the power load to obtain information, such as instant frequency, spectral entropy, spectrogram, Mel spectrogram and signal variation, to feed BiLSTM Neural Network. We also test different options for the BiLSTM to decide the most optimal settings. This method improves the classification performance, achieving up to 98.57% classification accuracy.

AB - One significant challenge in Non-Intrusive Load Monitoring (NILM) is to identify and classify active appliances used in a building. This research focuses on the classifying process, exploring different approaches for the feature extraction of the appliances’ power load to improve the classification accuracy. In this paper, we present a new method - Spectral Entropy and Instantaneous Frequency-based Bidirectional Long Short Term Memory (SE-IF BiLSTM). It uses feature extraction from the power load to obtain information, such as instant frequency, spectral entropy, spectrogram, Mel spectrogram and signal variation, to feed BiLSTM Neural Network. We also test different options for the BiLSTM to decide the most optimal settings. This method improves the classification performance, achieving up to 98.57% classification accuracy.

KW - BILSTM

KW - Appliance Classification

KW - NILM

UR - https://www.scopus.com/pages/publications/85123934619

U2 - 10.1109/ISGTEurope52324.2021.9640061

DO - 10.1109/ISGTEurope52324.2021.9640061

M3 - Conference contribution

SN - 9781665448765

SP - 180

EP - 184

BT - 2021 IEEE PES Innovative Smart Grid Technologies Europe (ISGT Europe)

PB - IEEE

CY - Piscataway, US

T2 - ISGT Europe 2021: IEEE PES Innovative Smart Grid Technologies

Y2 - 18 October 2021 through 21 October 2021

ER -

Appliance Classification using BiLSTM Neural Networks and Feature Extraction

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this