Deep Transfer Learning-Based Broken Rotor Fault Diagnosis for Induction Motors

Please use this identifier to cite or link to this item: https://dspace.univ-ouargla.dz/jspui/handle/123456789/39053

Title:	Deep Transfer Learning-Based Broken Rotor Fault Diagnosis for Induction Motors
Authors:	Souri, Samira Kahel, Elhadj chikh Bouzidi, Abdelkahar
Keywords:	Induction motor Broken rotor bars Deep learning Convolutional Neural Network time- frequency analysis
Issue Date:	2025
Publisher:	UNIVERSITY OF KASDI MERBAH OUARGLA
Citation:	FACULTY OF NEW TECHNOLOGIES OF INFORMATION AND COMMUNICATION
Abstract:	Induction motors, vital in industrial systems, are prone to electrical faults like broken rotor bars. Early detection is crucial to prevent failures. This study uses an IEEE dataset of motor signals to classify faults under varying loads via time-frequency analysis (STFT spectrograms) and deep learning. Three pre-trained CNNs (ResNet50, AlexNet, and VGG16) were tested, with ResNet50 achieving the highest accuracy (99.73%). Evaluated using accuracy/loss curves, confusion matrices, and precision/recall metrics, ResNet50 proved most robust for early fault detection, ensuring reliability across operational conditions.
Description:	Electronics of Embedded Systems
URI:	https://dspace.univ-ouargla.dz/jspui/handle/123456789/39053
Appears in Collections:	Département d'Electronique et des Télécommunications - Master

Files in This Item:

File	Description	Size	Format
KAHEL-BOUZIDI.pdf	Electronics of Embedded Systems	3,29 MB	Adobe PDF	View/Open

DSpace JSPUI