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 proceedingConference contributionpeer-review

1 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 languageEnglish
Title of host publication2021 IEEE PES Innovative Smart Grid Technologies Europe (ISGT Europe)
Subtitle of host publicationSmart Grids: Toward a Carbon-free Future
Place of PublicationPiscataway, US
PublisherIEEE
Pages180-184
Number of pages5
ISBN (Electronic)9781665448758
ISBN (Print)9781665448765
DOIs
Publication statusPublished - 18 Oct 2021
EventISGT Europe 2021: IEEE PES Innovative Smart Grid Technologies: Smart Grids: Toward a Carbon-free Future - Virtual, Aalto University, Espoo, Finland
Duration: 18 Oct 202121 Oct 2021
https://ieee-isgt-europe.org/

Conference

ConferenceISGT Europe 2021: IEEE PES Innovative Smart Grid Technologies
Abbreviated titleIEEE PES ISGT EUROPE 2021
Country/TerritoryFinland
CityEspoo
Period18/10/2121/10/21
Internet address

Fingerprint

Dive into the research topics of 'Appliance Classification using BiLSTM Neural Networks and Feature Extraction'. Together they form a unique fingerprint.

Cite this