Comparative Study between Physics-Informed CNN and PCA in Induction Motor Broken Bars MCSA Detection
In this article, two methods for broken bar detection in induction motors are considered and tested using data collected from the LIAS laboratory at the University of Poitiers. The first approach is Motor Current Signature Analysis (MCSA) with Convolutional Neural Networks (CNN), in which measuremen...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-12-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/22/23/9494 |
_version_ | 1797462142398496768 |
---|---|
author | Abderrahim Boushaba Sebastien Cauet Afzal Chamroo Erik Etien Laurent Rambault |
author_facet | Abderrahim Boushaba Sebastien Cauet Afzal Chamroo Erik Etien Laurent Rambault |
author_sort | Abderrahim Boushaba |
collection | DOAJ |
description | In this article, two methods for broken bar detection in induction motors are considered and tested using data collected from the LIAS laboratory at the University of Poitiers. The first approach is Motor Current Signature Analysis (MCSA) with Convolutional Neural Networks (CNN), in which measurements have to be processed in the frequency domain before training the CNN to ensure that the resulting model is physically informed. A double input CNN has been introduced to perform a 100% detection regardless of the speed and load torque value. A second approach is the Principal Components Analysis (PCA), in which the processing is undertaken in the time domain. The PCA is applied on the induction motor currents to eventually calculate the <i>Q</i> statistic that serves as a threshold for detecting anomalies/faults. Even if obtained results show that both approaches work very well, there are major differences that need to be pointed out, and this is the aim of the current paper. |
first_indexed | 2024-03-09T17:32:17Z |
format | Article |
id | doaj.art-d4f9e5e3d7de472eb942be1e42308dd6 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-09T17:32:17Z |
publishDate | 2022-12-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-d4f9e5e3d7de472eb942be1e42308dd62023-11-24T12:15:19ZengMDPI AGSensors1424-82202022-12-012223949410.3390/s22239494Comparative Study between Physics-Informed CNN and PCA in Induction Motor Broken Bars MCSA DetectionAbderrahim Boushaba0Sebastien Cauet1Afzal Chamroo2Erik Etien3Laurent Rambault4University of Poitiers, ISAE-ENSMA Poitiers, 2 rue Pierre Brousse, TSA41105, CEDEX 9, 86073 Poitiers, FranceUniversity of Poitiers, ISAE-ENSMA Poitiers, 2 rue Pierre Brousse, TSA41105, CEDEX 9, 86073 Poitiers, FranceUniversity of Poitiers, ISAE-ENSMA Poitiers, 2 rue Pierre Brousse, TSA41105, CEDEX 9, 86073 Poitiers, FranceUniversity of Poitiers, ISAE-ENSMA Poitiers, 2 rue Pierre Brousse, TSA41105, CEDEX 9, 86073 Poitiers, FranceUniversity of Poitiers, ISAE-ENSMA Poitiers, 2 rue Pierre Brousse, TSA41105, CEDEX 9, 86073 Poitiers, FranceIn this article, two methods for broken bar detection in induction motors are considered and tested using data collected from the LIAS laboratory at the University of Poitiers. The first approach is Motor Current Signature Analysis (MCSA) with Convolutional Neural Networks (CNN), in which measurements have to be processed in the frequency domain before training the CNN to ensure that the resulting model is physically informed. A double input CNN has been introduced to perform a 100% detection regardless of the speed and load torque value. A second approach is the Principal Components Analysis (PCA), in which the processing is undertaken in the time domain. The PCA is applied on the induction motor currents to eventually calculate the <i>Q</i> statistic that serves as a threshold for detecting anomalies/faults. Even if obtained results show that both approaches work very well, there are major differences that need to be pointed out, and this is the aim of the current paper.https://www.mdpi.com/1424-8220/22/23/9494MCSAPCAdeep learningphysically informedPINNSbroken bars |
spellingShingle | Abderrahim Boushaba Sebastien Cauet Afzal Chamroo Erik Etien Laurent Rambault Comparative Study between Physics-Informed CNN and PCA in Induction Motor Broken Bars MCSA Detection Sensors MCSA PCA deep learning physically informed PINNS broken bars |
title | Comparative Study between Physics-Informed CNN and PCA in Induction Motor Broken Bars MCSA Detection |
title_full | Comparative Study between Physics-Informed CNN and PCA in Induction Motor Broken Bars MCSA Detection |
title_fullStr | Comparative Study between Physics-Informed CNN and PCA in Induction Motor Broken Bars MCSA Detection |
title_full_unstemmed | Comparative Study between Physics-Informed CNN and PCA in Induction Motor Broken Bars MCSA Detection |
title_short | Comparative Study between Physics-Informed CNN and PCA in Induction Motor Broken Bars MCSA Detection |
title_sort | comparative study between physics informed cnn and pca in induction motor broken bars mcsa detection |
topic | MCSA PCA deep learning physically informed PINNS broken bars |
url | https://www.mdpi.com/1424-8220/22/23/9494 |
work_keys_str_mv | AT abderrahimboushaba comparativestudybetweenphysicsinformedcnnandpcaininductionmotorbrokenbarsmcsadetection AT sebastiencauet comparativestudybetweenphysicsinformedcnnandpcaininductionmotorbrokenbarsmcsadetection AT afzalchamroo comparativestudybetweenphysicsinformedcnnandpcaininductionmotorbrokenbarsmcsadetection AT eriketien comparativestudybetweenphysicsinformedcnnandpcaininductionmotorbrokenbarsmcsadetection AT laurentrambault comparativestudybetweenphysicsinformedcnnandpcaininductionmotorbrokenbarsmcsadetection |