The recent development and spread of artificial intelligence-based techniques, particularly deep learning algorithms, have made it possible to model phenomena that were previously impossible to handle. Furthermore, the development of the Big Data paradigm is rapidly leading toward new research frontiers in predicting and classifying one-dimensional signals. Anomaly detection plays a crucial role in the various areas that gain from the introduction of these methodologies. This extremely diverse field detects anomalies in both time series and image data. Anomaly detection applications include the detection of failures of grid-connected machinery in industrial environments. The objective of this study was to propose a fault detection methodology based on deep learning, specifically using convolutional autoencoders, using as few features as possible, specifically the current intensity in one of the three phases of an industrial plant. The results showed a high capability of the methodology to detect faults while generating a minimum number of false positives, paving the way for optimizations of the same and online deployment.

Anomaly Detection on Industrial Electrical Systems using Deep Learning

Carratu' M.;Gallo V.;Pietrosanto A.;Sommella P.;
2023-01-01

Abstract

The recent development and spread of artificial intelligence-based techniques, particularly deep learning algorithms, have made it possible to model phenomena that were previously impossible to handle. Furthermore, the development of the Big Data paradigm is rapidly leading toward new research frontiers in predicting and classifying one-dimensional signals. Anomaly detection plays a crucial role in the various areas that gain from the introduction of these methodologies. This extremely diverse field detects anomalies in both time series and image data. Anomaly detection applications include the detection of failures of grid-connected machinery in industrial environments. The objective of this study was to propose a fault detection methodology based on deep learning, specifically using convolutional autoencoders, using as few features as possible, specifically the current intensity in one of the three phases of an industrial plant. The results showed a high capability of the methodology to detect faults while generating a minimum number of false positives, paving the way for optimizations of the same and online deployment.
2023
978-1-6654-5383-7
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11386/4836771
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact