Cargando…

Unsupervised Domain Adaptive 1D-CNN for Fault Diagnosis of Bearing

Fault diagnosis (FD) plays a vital role in building a smart factory regarding system reliability improvement and cost reduction. Recent deep learning-based methods have been applied for FD and have obtained excellent performance. However, most of them require sufficient historical labeled data to tr...

Descripción completa

Detalles Bibliográficos
Autores principales: Shao, Xiaorui, Kim, Chang-Soo
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9185426/
https://www.ncbi.nlm.nih.gov/pubmed/35684777
http://dx.doi.org/10.3390/s22114156
_version_ 1784724721089118208
author Shao, Xiaorui
Kim, Chang-Soo
author_facet Shao, Xiaorui
Kim, Chang-Soo
author_sort Shao, Xiaorui
collection PubMed
description Fault diagnosis (FD) plays a vital role in building a smart factory regarding system reliability improvement and cost reduction. Recent deep learning-based methods have been applied for FD and have obtained excellent performance. However, most of them require sufficient historical labeled data to train the model which is difficult and sometimes not available. Moreover, the big size model increases the difficulties for real-time FD. Therefore, this article proposed a domain adaptive and lightweight framework for FD based on a one-dimension convolutional neural network (1D-CNN). Particularly, 1D-CNN is designed with a structure of autoencoder to extract the rich, robust hidden features with less noise from source and target data. The extracted features are processed by correlation alignment (CORAL) to minimize domain shifts. Thus, the proposed method could learn robust and domain-invariance features from raw signals without any historical labeled target domain data for FD. We designed, trained, and tested the proposed method on CRWU bearing data sets. The sufficient comparative analysis confirmed its effectiveness for FD.
format Online
Article
Text
id pubmed-9185426
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-91854262022-06-11 Unsupervised Domain Adaptive 1D-CNN for Fault Diagnosis of Bearing Shao, Xiaorui Kim, Chang-Soo Sensors (Basel) Article Fault diagnosis (FD) plays a vital role in building a smart factory regarding system reliability improvement and cost reduction. Recent deep learning-based methods have been applied for FD and have obtained excellent performance. However, most of them require sufficient historical labeled data to train the model which is difficult and sometimes not available. Moreover, the big size model increases the difficulties for real-time FD. Therefore, this article proposed a domain adaptive and lightweight framework for FD based on a one-dimension convolutional neural network (1D-CNN). Particularly, 1D-CNN is designed with a structure of autoencoder to extract the rich, robust hidden features with less noise from source and target data. The extracted features are processed by correlation alignment (CORAL) to minimize domain shifts. Thus, the proposed method could learn robust and domain-invariance features from raw signals without any historical labeled target domain data for FD. We designed, trained, and tested the proposed method on CRWU bearing data sets. The sufficient comparative analysis confirmed its effectiveness for FD. MDPI 2022-05-30 /pmc/articles/PMC9185426/ /pubmed/35684777 http://dx.doi.org/10.3390/s22114156 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Shao, Xiaorui
Kim, Chang-Soo
Unsupervised Domain Adaptive 1D-CNN for Fault Diagnosis of Bearing
title Unsupervised Domain Adaptive 1D-CNN for Fault Diagnosis of Bearing
title_full Unsupervised Domain Adaptive 1D-CNN for Fault Diagnosis of Bearing
title_fullStr Unsupervised Domain Adaptive 1D-CNN for Fault Diagnosis of Bearing
title_full_unstemmed Unsupervised Domain Adaptive 1D-CNN for Fault Diagnosis of Bearing
title_short Unsupervised Domain Adaptive 1D-CNN for Fault Diagnosis of Bearing
title_sort unsupervised domain adaptive 1d-cnn for fault diagnosis of bearing
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9185426/
https://www.ncbi.nlm.nih.gov/pubmed/35684777
http://dx.doi.org/10.3390/s22114156
work_keys_str_mv AT shaoxiaorui unsuperviseddomainadaptive1dcnnforfaultdiagnosisofbearing
AT kimchangsoo unsuperviseddomainadaptive1dcnnforfaultdiagnosisofbearing