A Hybrid Deep Transfer Learning Framework for Delamination Identification in Composite Laminates

Structural health monitoring (SHM) has proven to be an effective technique to maintain the safety and reliability of laminated composites. Recently, both deep learning and machine learning methodologies have gained popularity in sensor-based SHM. However, machine learning approaches often require te...

Full description

Saved in:
Bibliographic Details
Main Authors: Muhammad Haris Yazdani, Muhammad Muzammil Azad, Salman Khalid, Heung Soo Kim
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/3/826
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849717699658121216
author Muhammad Haris Yazdani
Muhammad Muzammil Azad
Salman Khalid
Heung Soo Kim
author_facet Muhammad Haris Yazdani
Muhammad Muzammil Azad
Salman Khalid
Heung Soo Kim
author_sort Muhammad Haris Yazdani
collection DOAJ
description Structural health monitoring (SHM) has proven to be an effective technique to maintain the safety and reliability of laminated composites. Recently, both deep learning and machine learning methodologies have gained popularity in sensor-based SHM. However, machine learning approaches often require tedious manual feature extraction, while deep learning models require large training datasets, which may not be feasible. To overcome these limitations, this study presents a hybrid deep transfer learning (HTL) framework to identify delamination in composite laminates. The proposed framework enhances SHM performance by utilizing pre-trained EfficientNet and ResNet models to allow for deep feature extraction with limited data. EfficientNet contributes to this by efficiently scaling the model to capture multi-scale spatial features, while ResNet contributes by extracting hierarchical representations through its residual connections. Vibration signals from piezoelectric (PZT) sensors attached to the composite laminates, consisting of three health states, are used to validate the approach. Compared to the existing transfer learning approaches, the suggested method achieved better performance, hence improving both the accuracy and robustness of delamination detection in composite structures.
format Article
id doaj-art-21cdb4fdc48741549ca1add6ea02e643
institution DOAJ
issn 1424-8220
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-21cdb4fdc48741549ca1add6ea02e6432025-08-20T03:12:35ZengMDPI AGSensors1424-82202025-01-0125382610.3390/s25030826A Hybrid Deep Transfer Learning Framework for Delamination Identification in Composite LaminatesMuhammad Haris Yazdani0Muhammad Muzammil Azad1Salman Khalid2Heung Soo Kim3Department of Mechanical Engineering, Dongguk University-Seoul, 30 Pildong-ro 1-gil, Jung-gu, Seoul 04620, Republic of KoreaDepartment of Mechanical Engineering, Dongguk University-Seoul, 30 Pildong-ro 1-gil, Jung-gu, Seoul 04620, Republic of KoreaDepartment of Mechanical, Robotics and Energy Engineering, Dongguk University-Seoul, 30 Pildong-ro 1-gil, Jung-gu, Seoul 04620, Republic of KoreaDepartment of Mechanical, Robotics and Energy Engineering, Dongguk University-Seoul, 30 Pildong-ro 1-gil, Jung-gu, Seoul 04620, Republic of KoreaStructural health monitoring (SHM) has proven to be an effective technique to maintain the safety and reliability of laminated composites. Recently, both deep learning and machine learning methodologies have gained popularity in sensor-based SHM. However, machine learning approaches often require tedious manual feature extraction, while deep learning models require large training datasets, which may not be feasible. To overcome these limitations, this study presents a hybrid deep transfer learning (HTL) framework to identify delamination in composite laminates. The proposed framework enhances SHM performance by utilizing pre-trained EfficientNet and ResNet models to allow for deep feature extraction with limited data. EfficientNet contributes to this by efficiently scaling the model to capture multi-scale spatial features, while ResNet contributes by extracting hierarchical representations through its residual connections. Vibration signals from piezoelectric (PZT) sensors attached to the composite laminates, consisting of three health states, are used to validate the approach. Compared to the existing transfer learning approaches, the suggested method achieved better performance, hence improving both the accuracy and robustness of delamination detection in composite structures.https://www.mdpi.com/1424-8220/25/3/826vibration signalsdelamination detectiondelamination identificationtransfer learningdeep learninghybrid model
spellingShingle Muhammad Haris Yazdani
Muhammad Muzammil Azad
Salman Khalid
Heung Soo Kim
A Hybrid Deep Transfer Learning Framework for Delamination Identification in Composite Laminates
Sensors
vibration signals
delamination detection
delamination identification
transfer learning
deep learning
hybrid model
title A Hybrid Deep Transfer Learning Framework for Delamination Identification in Composite Laminates
title_full A Hybrid Deep Transfer Learning Framework for Delamination Identification in Composite Laminates
title_fullStr A Hybrid Deep Transfer Learning Framework for Delamination Identification in Composite Laminates
title_full_unstemmed A Hybrid Deep Transfer Learning Framework for Delamination Identification in Composite Laminates
title_short A Hybrid Deep Transfer Learning Framework for Delamination Identification in Composite Laminates
title_sort hybrid deep transfer learning framework for delamination identification in composite laminates
topic vibration signals
delamination detection
delamination identification
transfer learning
deep learning
hybrid model
url https://www.mdpi.com/1424-8220/25/3/826
work_keys_str_mv AT muhammadharisyazdani ahybriddeeptransferlearningframeworkfordelaminationidentificationincompositelaminates
AT muhammadmuzammilazad ahybriddeeptransferlearningframeworkfordelaminationidentificationincompositelaminates
AT salmankhalid ahybriddeeptransferlearningframeworkfordelaminationidentificationincompositelaminates
AT heungsookim ahybriddeeptransferlearningframeworkfordelaminationidentificationincompositelaminates
AT muhammadharisyazdani hybriddeeptransferlearningframeworkfordelaminationidentificationincompositelaminates
AT muhammadmuzammilazad hybriddeeptransferlearningframeworkfordelaminationidentificationincompositelaminates
AT salmankhalid hybriddeeptransferlearningframeworkfordelaminationidentificationincompositelaminates
AT heungsookim hybriddeeptransferlearningframeworkfordelaminationidentificationincompositelaminates