Distribution of fault currents.

<div><p>This study explores the design of an effective fault classification algorithm for 3 phase induction motor, an integral unit in many industrial systems. It is found that traditional fault detection methods and deep learning approaches are both effective; however, current technique...

Full description

Saved in:
Bibliographic Details
Main Author: Zawar Ahmed Khan (22574556) (author)
Other Authors: Muhammad Amir Raza (22391989) (author), Muhammad I. Masud (22574559) (author), Touqeer Ahmed Jumani (12536455) (author), Muhammad Faheem (1767637) (author), Mohammed Aman (19665349) (author)
Published: 2025
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1852015074428846080
author Zawar Ahmed Khan (22574556)
author2 Muhammad Amir Raza (22391989)
Muhammad I. Masud (22574559)
Touqeer Ahmed Jumani (12536455)
Muhammad Faheem (1767637)
Mohammed Aman (19665349)
author2_role author
author
author
author
author
author_facet Zawar Ahmed Khan (22574556)
Muhammad Amir Raza (22391989)
Muhammad I. Masud (22574559)
Touqeer Ahmed Jumani (12536455)
Muhammad Faheem (1767637)
Mohammed Aman (19665349)
author_role author
dc.creator.none.fl_str_mv Zawar Ahmed Khan (22574556)
Muhammad Amir Raza (22391989)
Muhammad I. Masud (22574559)
Touqeer Ahmed Jumani (12536455)
Muhammad Faheem (1767637)
Mohammed Aman (19665349)
dc.date.none.fl_str_mv 2025-11-06T18:36:01Z
dc.identifier.none.fl_str_mv 10.1371/journal.pone.0335367.g005
dc.relation.none.fl_str_mv https://figshare.com/articles/figure/Distribution_of_fault_currents_/30558141
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Space Science
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
principal component analysis
higher computational cost
frequency domain information
fast fourier transform
extreme gradient boosting
deep learning approaches
many industrial systems
knn performs well
7 &# 8201
dt ), k
knn ),
many real
&# 8217
xlink ">
world settings
work serves
various loads
thus making
synthetic minority
study explores
smote ).
sampling technique
rotor faults
results shows
random forest
nearest neighbors
motor specifications
mm bearing
missing values
intelligent framework
integral unit
generally known
future work
feature set
feature selection
fault types
dimensionality reduction
data preprocessing
current variables
current signals
computationally exhaustive
best possible
additional feature
dc.title.none.fl_str_mv Distribution of fault currents.
dc.type.none.fl_str_mv Image
Figure
info:eu-repo/semantics/publishedVersion
image
description <div><p>This study explores the design of an effective fault classification algorithm for 3 phase induction motor, an integral unit in many industrial systems. It is found that traditional fault detection methods and deep learning approaches are both effective; however, current techniques can either be computationally exhaustive, or suffer from low accuracy, thus making them inapplicable in many real-world settings. To address these limitations, this study evaluates different machine learning algorithms for accurate and efficient fault detection using a dataset of triaxial vibrational data converted into current variables. A dataset of triaxial vibrational current data at 0.7 mm bearing and rotor faults at various loads (100W, 200W, and 300W) were considered. For the data preprocessing, we handled with the missing values by interpolation and handle data imbalance fault types with Synthetic Minority Over-sampling Technique (SMOTE). Through Fast Fourier Transform (FFT) techniques, the frequency domain information were extracted, which is key for current signals, adding to the feature set. In addition, dimensionality reduction with Principal Component Analysis (PCA) and feature selection was done with SelectKBest. Then, the different machine learning models such as Random Forest (RF), Decision Tree (DT), k-nearest neighbors (KNN), and eXtreme Gradient Boosting (XGBoost) was trained to optimize the hyperparameters and make them perform to its best possible. The results shows the accuracy and performance of all models, DT and RF show good performance, with 99.95% accuracy, while KNN performs well, but at a higher computational cost in testing. Generally known for its capability to handle all the complex dataset, XGBoost wasn’t able to perform in this scenario as it got an accuracy of 87.13%, indicating potentially more optimization is required for the model. This work serves as the groundwork for future work with a multiplicity of fault types, motor specifications, and the incorporation of additional feature-engineering techniques to develop a more robust and intelligent framework for fault detection.</p></div>
eu_rights_str_mv openAccess
id Manara_e5351bc45ae3cd0c3e34a642fafcb1bc
identifier_str_mv 10.1371/journal.pone.0335367.g005
network_acronym_str Manara
network_name_str ManaraRepo
oai_identifier_str oai:figshare.com:article/30558141
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Distribution of fault currents.Zawar Ahmed Khan (22574556)Muhammad Amir Raza (22391989)Muhammad I. Masud (22574559)Touqeer Ahmed Jumani (12536455)Muhammad Faheem (1767637)Mohammed Aman (19665349)Space ScienceBiological Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedprincipal component analysishigher computational costfrequency domain informationfast fourier transformextreme gradient boostingdeep learning approachesmany industrial systemsknn performs well7 &# 8201dt ), kknn ),many real&# 8217xlink ">world settingswork servesvarious loadsthus makingsynthetic minoritystudy exploressmote ).sampling techniquerotor faultsresults showsrandom forestnearest neighborsmotor specificationsmm bearingmissing valuesintelligent frameworkintegral unitgenerally knownfuture workfeature setfeature selectionfault typesdimensionality reductiondata preprocessingcurrent variablescurrent signalscomputationally exhaustivebest possibleadditional feature<div><p>This study explores the design of an effective fault classification algorithm for 3 phase induction motor, an integral unit in many industrial systems. It is found that traditional fault detection methods and deep learning approaches are both effective; however, current techniques can either be computationally exhaustive, or suffer from low accuracy, thus making them inapplicable in many real-world settings. To address these limitations, this study evaluates different machine learning algorithms for accurate and efficient fault detection using a dataset of triaxial vibrational data converted into current variables. A dataset of triaxial vibrational current data at 0.7 mm bearing and rotor faults at various loads (100W, 200W, and 300W) were considered. For the data preprocessing, we handled with the missing values by interpolation and handle data imbalance fault types with Synthetic Minority Over-sampling Technique (SMOTE). Through Fast Fourier Transform (FFT) techniques, the frequency domain information were extracted, which is key for current signals, adding to the feature set. In addition, dimensionality reduction with Principal Component Analysis (PCA) and feature selection was done with SelectKBest. Then, the different machine learning models such as Random Forest (RF), Decision Tree (DT), k-nearest neighbors (KNN), and eXtreme Gradient Boosting (XGBoost) was trained to optimize the hyperparameters and make them perform to its best possible. The results shows the accuracy and performance of all models, DT and RF show good performance, with 99.95% accuracy, while KNN performs well, but at a higher computational cost in testing. Generally known for its capability to handle all the complex dataset, XGBoost wasn’t able to perform in this scenario as it got an accuracy of 87.13%, indicating potentially more optimization is required for the model. This work serves as the groundwork for future work with a multiplicity of fault types, motor specifications, and the incorporation of additional feature-engineering techniques to develop a more robust and intelligent framework for fault detection.</p></div>2025-11-06T18:36:01ZImageFigureinfo:eu-repo/semantics/publishedVersionimage10.1371/journal.pone.0335367.g005https://figshare.com/articles/figure/Distribution_of_fault_currents_/30558141CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/305581412025-11-06T18:36:01Z
spellingShingle Distribution of fault currents.
Zawar Ahmed Khan (22574556)
Space Science
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
principal component analysis
higher computational cost
frequency domain information
fast fourier transform
extreme gradient boosting
deep learning approaches
many industrial systems
knn performs well
7 &# 8201
dt ), k
knn ),
many real
&# 8217
xlink ">
world settings
work serves
various loads
thus making
synthetic minority
study explores
smote ).
sampling technique
rotor faults
results shows
random forest
nearest neighbors
motor specifications
mm bearing
missing values
intelligent framework
integral unit
generally known
future work
feature set
feature selection
fault types
dimensionality reduction
data preprocessing
current variables
current signals
computationally exhaustive
best possible
additional feature
status_str publishedVersion
title Distribution of fault currents.
title_full Distribution of fault currents.
title_fullStr Distribution of fault currents.
title_full_unstemmed Distribution of fault currents.
title_short Distribution of fault currents.
title_sort Distribution of fault currents.
topic Space Science
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
principal component analysis
higher computational cost
frequency domain information
fast fourier transform
extreme gradient boosting
deep learning approaches
many industrial systems
knn performs well
7 &# 8201
dt ), k
knn ),
many real
&# 8217
xlink ">
world settings
work serves
various loads
thus making
synthetic minority
study explores
smote ).
sampling technique
rotor faults
results shows
random forest
nearest neighbors
motor specifications
mm bearing
missing values
intelligent framework
integral unit
generally known
future work
feature set
feature selection
fault types
dimensionality reduction
data preprocessing
current variables
current signals
computationally exhaustive
best possible
additional feature