Título: Performance of Scattering Transform Feature Extraction for Electrical Load Classification
Autores: Everton L. de Aguiar, André E. Lazzaretti and Daniel R. Pipa.
Resumo:
The Scattering transform (ST) presents itself as an alternative approach to the classic methods that involve neural networks and deep learning techniques for the feature extraction and classification of signals and images. Among its main advantages, one can emphasize that the coefficients of the ST are determined analytically and do not need to be learned, as typically performed in Convolutional Neural Networks (CNNs). Additionally, ST has time-shifting and small time-warping invariance, which reduces the need for precise temporal localization (detection) for subsequent classification. This paper originally proposes the application of ST to extract features and classify Non-intrusive Load Monitoring (NILM) high-frequency signals. We validate the extraction strategy performance varying several parameters for ST calculation, such as signal length, number of examples, and sampling frequency. The results outperform other state-of-the-art feature extraction techniques, reaching up to 99.98% of accuracy and 99.51% of FScore for a publicly available dataset, demonstrating the feasibility and promising aspects of the ST for NILM problems.
Palavras-chave:
Scattering Transform, Load Classification, Features Extraction.
Páginas: 8
Código DOI: 10.21528/CBIC2021-43
Artigo em pdf: CBIC_2021_paper_43.pdf
Arquivo BibTeX: CBIC_2021_43.bib