TY - JOUR
T1 - A hybrid prototype selection-based deep learning approach for anomaly detection in industrial machines
AU - de Paula Monteiro, Rodrigo
AU - Cerrada Lozada, Mariela
AU - Mendieta, Diego Roman Cabrera
AU - Loja, René Vinicio Sánchez
AU - Filho, Carmelo José Albanez Bastos
N1 - Publisher Copyright:
© 2022 Elsevier Ltd
PY - 2022/10/15
Y1 - 2022/10/15
N2 - Anomaly detection in time series is an important task to many applications, e.g, the maintenance policies of rotating machines within industries strongly rely on time series monitoring. Rotating machines are vital elements within industries. Therefore, maintenance policies on these critical elements concern the quality of products and safety issues. Condition-based maintenance is an example of those policies. In this context, we propose a novel method to train a deep learning-based feature extractor for the anomaly detection problem on rotating machinery. It consists of using a prototype selection algorithm to improve the training process of a randomly initialized feature extractor. We perform this process iteratively using data belonging to one probability distribution, i.e., the normal class. We carried the prototype selection out with the Nearest Neighbors algorithm, and the feature extractor was a Convolutional Neural Network. We validate the method on three datasets of spectrograms related to gearbox and compressors faults and achieved promising results. We obtained detection rates in anomalous data close to 100%, and the anomaly detectors classified normal instances with accuracy values superior to 95%. Those results were competitive concerning other deep learning-based anomaly detectors in the literature, with the advantage of being an integrated solution.
AB - Anomaly detection in time series is an important task to many applications, e.g, the maintenance policies of rotating machines within industries strongly rely on time series monitoring. Rotating machines are vital elements within industries. Therefore, maintenance policies on these critical elements concern the quality of products and safety issues. Condition-based maintenance is an example of those policies. In this context, we propose a novel method to train a deep learning-based feature extractor for the anomaly detection problem on rotating machinery. It consists of using a prototype selection algorithm to improve the training process of a randomly initialized feature extractor. We perform this process iteratively using data belonging to one probability distribution, i.e., the normal class. We carried the prototype selection out with the Nearest Neighbors algorithm, and the feature extractor was a Convolutional Neural Network. We validate the method on three datasets of spectrograms related to gearbox and compressors faults and achieved promising results. We obtained detection rates in anomalous data close to 100%, and the anomaly detectors classified normal instances with accuracy values superior to 95%. Those results were competitive concerning other deep learning-based anomaly detectors in the literature, with the advantage of being an integrated solution.
KW - Anomaly detection
KW - Deep learning
KW - Prototype selection
KW - Rotating machinery
UR - http://www.scopus.com/inward/record.url?scp=85131065860&partnerID=8YFLogxK
U2 - 10.1016/j.eswa.2022.117528
DO - 10.1016/j.eswa.2022.117528
M3 - Article
AN - SCOPUS:85131065860
SN - 0957-4174
VL - 204
JO - Expert Systems with Applications
JF - Expert Systems with Applications
M1 - 117528
ER -