Multiclass Incremental Learning for Fault Diagnosis in Induction Motors Using Fine-Tuning with a Memory of Exemplars and Nearest Centroid Classifier

Early detection of fault events through electromechanical systems operation is one of the most attractive and critical data challenges in modern industry. Although these electromechanical systems tend to experiment with typical faults, a common event is that unexpected and unknown faults can be pres...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Magdiel Jiménez-Guarneros, Jonas Grande-Barreto, Jose de Jesus Rangel-Magdaleno
Formato: article
Lenguaje:EN
Publicado: Hindawi Limited 2021
Materias:
Acceso en línea:https://doaj.org/article/018f20ef02214080bb44d38db79a1f61
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:018f20ef02214080bb44d38db79a1f61
record_format dspace
spelling oai:doaj.org-article:018f20ef02214080bb44d38db79a1f612021-11-08T02:36:37ZMulticlass Incremental Learning for Fault Diagnosis in Induction Motors Using Fine-Tuning with a Memory of Exemplars and Nearest Centroid Classifier1875-920310.1155/2021/6627740https://doaj.org/article/018f20ef02214080bb44d38db79a1f612021-01-01T00:00:00Zhttp://dx.doi.org/10.1155/2021/6627740https://doaj.org/toc/1875-9203Early detection of fault events through electromechanical systems operation is one of the most attractive and critical data challenges in modern industry. Although these electromechanical systems tend to experiment with typical faults, a common event is that unexpected and unknown faults can be presented during operation. However, current models for automatic detection can learn new faults at the cost of forgetting concepts previously learned. This article presents a multiclass incremental learning (MCIL) framework based on 1D convolutional neural network (CNN) for fault detection in induction motors. The presented framework tackles the forgetting problem by storing a representative exemplar set from past data (known faults) in memory. Then, the 1D CNN is fine-tuned over the selected exemplar set and data from new faults. Test samples are classified using nearest centroid classifier (NCC) in the feature space from 1D CNN. The proposed framework was evaluated and validated over two public datasets for fault detection in induction motors (IMs): asynchronous motor common fault (AMCF) and Case Western Reserve University (CWRU). Experimental results reveal the proposed framework as an effective solution to incorporate and detect new induction motor faults to already known, with a high accuracy performance across different incremental phases.Magdiel Jiménez-GuarnerosJonas Grande-BarretoJose de Jesus Rangel-MagdalenoHindawi LimitedarticlePhysicsQC1-999ENShock and Vibration, Vol 2021 (2021)
institution DOAJ
collection DOAJ
language EN
topic Physics
QC1-999
spellingShingle Physics
QC1-999
Magdiel Jiménez-Guarneros
Jonas Grande-Barreto
Jose de Jesus Rangel-Magdaleno
Multiclass Incremental Learning for Fault Diagnosis in Induction Motors Using Fine-Tuning with a Memory of Exemplars and Nearest Centroid Classifier
description Early detection of fault events through electromechanical systems operation is one of the most attractive and critical data challenges in modern industry. Although these electromechanical systems tend to experiment with typical faults, a common event is that unexpected and unknown faults can be presented during operation. However, current models for automatic detection can learn new faults at the cost of forgetting concepts previously learned. This article presents a multiclass incremental learning (MCIL) framework based on 1D convolutional neural network (CNN) for fault detection in induction motors. The presented framework tackles the forgetting problem by storing a representative exemplar set from past data (known faults) in memory. Then, the 1D CNN is fine-tuned over the selected exemplar set and data from new faults. Test samples are classified using nearest centroid classifier (NCC) in the feature space from 1D CNN. The proposed framework was evaluated and validated over two public datasets for fault detection in induction motors (IMs): asynchronous motor common fault (AMCF) and Case Western Reserve University (CWRU). Experimental results reveal the proposed framework as an effective solution to incorporate and detect new induction motor faults to already known, with a high accuracy performance across different incremental phases.
format article
author Magdiel Jiménez-Guarneros
Jonas Grande-Barreto
Jose de Jesus Rangel-Magdaleno
author_facet Magdiel Jiménez-Guarneros
Jonas Grande-Barreto
Jose de Jesus Rangel-Magdaleno
author_sort Magdiel Jiménez-Guarneros
title Multiclass Incremental Learning for Fault Diagnosis in Induction Motors Using Fine-Tuning with a Memory of Exemplars and Nearest Centroid Classifier
title_short Multiclass Incremental Learning for Fault Diagnosis in Induction Motors Using Fine-Tuning with a Memory of Exemplars and Nearest Centroid Classifier
title_full Multiclass Incremental Learning for Fault Diagnosis in Induction Motors Using Fine-Tuning with a Memory of Exemplars and Nearest Centroid Classifier
title_fullStr Multiclass Incremental Learning for Fault Diagnosis in Induction Motors Using Fine-Tuning with a Memory of Exemplars and Nearest Centroid Classifier
title_full_unstemmed Multiclass Incremental Learning for Fault Diagnosis in Induction Motors Using Fine-Tuning with a Memory of Exemplars and Nearest Centroid Classifier
title_sort multiclass incremental learning for fault diagnosis in induction motors using fine-tuning with a memory of exemplars and nearest centroid classifier
publisher Hindawi Limited
publishDate 2021
url https://doaj.org/article/018f20ef02214080bb44d38db79a1f61
work_keys_str_mv AT magdieljimenezguarneros multiclassincrementallearningforfaultdiagnosisininductionmotorsusingfinetuningwithamemoryofexemplarsandnearestcentroidclassifier
AT jonasgrandebarreto multiclassincrementallearningforfaultdiagnosisininductionmotorsusingfinetuningwithamemoryofexemplarsandnearestcentroidclassifier
AT josedejesusrangelmagdaleno multiclassincrementallearningforfaultdiagnosisininductionmotorsusingfinetuningwithamemoryofexemplarsandnearestcentroidclassifier
_version_ 1718443063389454336