Mohsen Barzegar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Mohsen Barzegar has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Helena Margarida Ramos
Civil Engineering, Architecture and Environment Department, CERIS, Instituto Superior Técnico, University of Lisbon, 1049-001 Lisbon, Portugal
15 shared publications - Artur Lopes Ribeiro
Instituto de Telecomunicaes, Instituto Superior Tcnico, Universidade de Lisboa, Lisbon, Portugal
13 shared publications - Dario Jeronimo Pasadas
Instituto de Telecomunicaes, Instituto Superior Tcnico, Universidade de Lisboa, Lisbon, Portugal
13 shared publications - Muchao Zhang
Instituto de Telecomunicaes, Instituto Superior Tcnico, Universidade de Lisboa, Lisbon, Portugal
6 shared publications Artur L. Ribeiro
2 shared publicationsDario J. Pasadas
2 shared publications
Latest Publications
- Open Access
- 3 November 2026
- Published by Elsevier in Ultrasonics
- Vol. 171
A review on ultrasonic guided wave-based non-destructive testing and structural health monitoring for adhesive joints
- 3 November 2026
- Published by Elsevier in Ultrasonics
- Vol. 171
Open AccessA review on ultrasonic guided wave-based non-destructive testing and structural health monitoring for adhesive joints
- Open Access
- 29 September 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (19) , 9641
Ultrasonic Guided Wave Localization and Classification of Damage-like Perturbations in a CFRP Plate Using a PZT Network
Ultrasonic guided waves acquired using permanently bonded piezoelectric transducers enable large-area monitoring of composite structures. For practical structural health monitoring; however, both the location and the condition of a structural change are critical. This paper presents a data-driven multi-task framework for a carbon-fiber-reinforced polymer plate using a circular PZT array. To generate a large, precise, and repeatable experimental dataset without permanently damaging the specimen, paired magnets were placed on opposite surfaces of the plate and moved using a two-axis positioning system over a uniform grid. Four magnet-pair configurations were used to form four balanced experimental classes. For each case, ultrasonic guided-wave signals from all actuator–sensor paths were concatenated and processed by a shared convolutional neural network and long short-term memory architecture. The model simultaneously estimates the spatial coordinates through two regression branches and identifies the magnet-pair perturbation configuration through a classification branch. An automated hyperparameter search was used to select the principal architectural and training parameters. Performance was evaluated using five-fold training/validation cross-validation. The selected configuration achieved a classification accuracy of and a macro F1-score of . The mean normalized Euclidean localization error was , and of the de-normalized localization errors were at or below . The shared architecture therefore performs simultaneous perturbation-configuration classification and two-dimensional localization from the same multi-path guided-wave measurements under the investigated controlled perturbation conditions.- 29 September 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (19) , 9641
Open AccessUltrasonic Guided Wave Localization and Classification of Damage-like Perturbations in a CFRP Plate Using a PZT Network
Ultrasonic guided waves acquired using permanently bonded piezoelectric transducers enable large-area monitoring of composite structures. For practical structural health monitoring; however, both the location and the condition of a structural change are critical. This paper presents a data-driven multi-task framework for a carbon-fiber-reinforced polymer plate using a circular PZT array. To generate a large, precise, and repeatable experimental dataset without permanently damaging the specimen, paired magnets were placed on opposite surfaces of the plate and moved using a two-axis positioning system over a uniform grid. Four magnet-pair configurations were used to form four balanced experimental classes. For each case, ultrasonic guided-wave signals from all actuator–sensor paths were concatenated and processed by a shared convolutional neural network and long short-term memory architecture. The model simultaneously estimates the spatial coordinates through two regression branches and identifies the magnet-pair perturbation configuration through a classification branch. An automated hyperparameter search was used to select the principal architectural and training parameters. Performance was evaluated using five-fold training/validation cross-validation. The selected configuration achieved a classification accuracy of and a macro F1-score of . The mean normalized Euclidean localization error was , and of the de-normalized localization errors were at or below . The shared architecture therefore performs simultaneous perturbation-configuration classification and two-dimensional localization from the same multi-path guided-wave measurements under the investigated controlled perturbation conditions. - 18 February 2026
- Published by Elsevier in Mechanical Systems and Signal Processing
- Vol. 248
Disbond imaging of adhesive joints using low-rank representation of Lamb waves
- 18 February 2026
- Published by Elsevier in Mechanical Systems and Signal Processing
- Vol. 248
Disbond imaging of adhesive joints using low-rank representation of Lamb waves
- Top Cited PapersOpen Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-9
Unsupervised Disbond Detection in CFRP Joints via Guided Waves and Autoencoders
Ultrasonic guided waves (UGWs) are well-suited for the nondestructive inspection (NDT) of adhesive joints as they can enable the transmission and reception of signals across the joint without requiring direct access to the damaged area. Conventional damage indices (DIs), such as the peak of the signal envelope, are limited for Carbon Fiber Reinforced Polymer (CFRP) adhesive joints due to the material's anisotropy and the complexity of UGW propagation. This is especially challenging for disbond detection and sizing. This study proposes an unsupervised approach using autoencoders for automatic disbond characterization in CFRP joints. Angle beam transducers (ABTs) were used in a pitch-catch setup to collect UGW signals under pristine conditions at two excitation frequencies (320 kHz and 650 kHz). The proposed DI is based on the reconstruction error between the original and autoencoder-reconstructed signals. Three architectures, including fully connected, Transformer-based, and CNN-LSTM autoencoders, were evaluated using 5-fold cross-validation and tested on independent scans. Performance was assessed using different metrics. Results show that the CNN-LSTM model consistently achieved the highest detection accuracy, while the Transformer model exhibited the most stable performance across folds. In a controlled temperature and humidity environment, the proposed DI provides superior localization and sizing of disbonds compared to conventional manually extracted signal features.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 75 (00189456) , 1-9
Top Cited PapersOpen AccessUnsupervised Disbond Detection in CFRP Joints via Guided Waves and Autoencoders
Ultrasonic guided waves (UGWs) are well-suited for the nondestructive inspection (NDT) of adhesive joints as they can enable the transmission and reception of signals across the joint without requiring direct access to the damaged area. Conventional damage indices (DIs), such as the peak of the signal envelope, are limited for Carbon Fiber Reinforced Polymer (CFRP) adhesive joints due to the material's anisotropy and the complexity of UGW propagation. This is especially challenging for disbond detection and sizing. This study proposes an unsupervised approach using autoencoders for automatic disbond characterization in CFRP joints. Angle beam transducers (ABTs) were used in a pitch-catch setup to collect UGW signals under pristine conditions at two excitation frequencies (320 kHz and 650 kHz). The proposed DI is based on the reconstruction error between the original and autoencoder-reconstructed signals. Three architectures, including fully connected, Transformer-based, and CNN-LSTM autoencoders, were evaluated using 5-fold cross-validation and tested on independent scans. Performance was assessed using different metrics. Results show that the CNN-LSTM model consistently achieved the highest detection accuracy, while the Transformer model exhibited the most stable performance across folds. In a controlled temperature and humidity environment, the proposed DI provides superior localization and sizing of disbonds compared to conventional manually extracted signal features.
- 9 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. PP (1530437X) , 1
Damage Imaging in a CFRP Plate using UGW-based Handheld Probe
This paper presents a handheld guided wave testing probe, measurement procedure, and image processing algorithm to detect and locate defects in carbon fiber-reinforced (CFRP) structures. This study was performed by collecting data from a network of piezoelectric sensors controlled in a pairwise pitch-catch configuration. The probe includes sixteen piezoelectric elements spatially distributed in a square shape configuration to generate and receive ultrasonic guided waves (UGW) signals in the vicinity of a region to be inspected. Instead of collecting all the sensor combinations for defect detection, a measurement procedure is presented to record UGW signals from pairs of sensors that share identical distances and path conditions in the square shape configuration of the proposed probe. Additionally, a comparison between both baseline and baseline-free approaches is performed using the Reconstruction Algorithm for Probabilistic Inspection of Defects (RAPID) algorithm for defect detection, localization, and evaluation in a quasi-isotropic CFRP plate.
- 9 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. PP (1530437X) , 1
Damage Imaging in a CFRP Plate using UGW-based Handheld Probe
This paper presents a handheld guided wave testing probe, measurement procedure, and image processing algorithm to detect and locate defects in carbon fiber-reinforced (CFRP) structures. This study was performed by collecting data from a network of piezoelectric sensors controlled in a pairwise pitch-catch configuration. The probe includes sixteen piezoelectric elements spatially distributed in a square shape configuration to generate and receive ultrasonic guided waves (UGW) signals in the vicinity of a region to be inspected. Instead of collecting all the sensor combinations for defect detection, a measurement procedure is presented to record UGW signals from pairs of sensors that share identical distances and path conditions in the square shape configuration of the proposed probe. Additionally, a comparison between both baseline and baseline-free approaches is performed using the Reconstruction Algorithm for Probabilistic Inspection of Defects (RAPID) algorithm for defect detection, localization, and evaluation in a quasi-isotropic CFRP plate.
- 7 July 2025
- Published by Elsevier in NDT & E International
- Vol. 156
Damage imaging using 2D teager-kaiser operator for early-time ultrasonic guided wavefields
- 7 July 2025
- Published by Elsevier in NDT & E International
- Vol. 156
Damage imaging using 2D teager-kaiser operator for early-time ultrasonic guided wavefields
- Top Cited PapersOpen Access
- 13 March 2025
- Published by Elsevier in Mechanical Systems and Signal Processing
- Vol. 229
Baseline-free damage imaging of CFRP lap joints using K-means clustering of guided wave signals
- 13 March 2025
- Published by Elsevier in Mechanical Systems and Signal Processing
- Vol. 229
Top Cited PapersOpen AccessBaseline-free damage imaging of CFRP lap joints using K-means clustering of guided wave signals
- 19 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 33 (9) , 2370-2381
MSDF-Based MAC for Energy-Efficient Neural Networks
This article presents an energy-efficient serial multiply-accumulate (MAC) unit based on the most significant digit first (MSDF) approach, specifically aimed at neural networks operating in resource-constrained and low-energy environments. The proposed MAC unit has been integrated into two neural network architectures: a multilayer perceptron (MLP) and a denoising autoencoder. For the MLP, we employ a pre-trained model for handwritten digit classification using the Modified National Institute of Standard and Technology (MNIST) dataset. Hardware synthesis using 45 nm CMOS technology shows that the MSDF-based MLP achieves a favorable trade-off between hardware metrics, such as reduced circuit area and energy per neuron, and reaches a good classification accuracy of 97.4%. In the case of the autoencoder, the proposed MAC unit is utilized in a pre-trained denoising autoencoder for the same dataset. The autoencoder employs MSDF-based mixed precision across layers to reduce computational resources and loads while preserving high-quality image reconstruction. A comprehensive evaluation of image correctness, signal fidelity, and visual quality metrics was conducted to assess the impact of varying precision levels across different layers. The results indicate that the proposed approach achieves an effective denoising performance with balanced hardware requirements.
- 19 February 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 33 (9) , 2370-2381
MSDF-Based MAC for Energy-Efficient Neural Networks
This article presents an energy-efficient serial multiply-accumulate (MAC) unit based on the most significant digit first (MSDF) approach, specifically aimed at neural networks operating in resource-constrained and low-energy environments. The proposed MAC unit has been integrated into two neural network architectures: a multilayer perceptron (MLP) and a denoising autoencoder. For the MLP, we employ a pre-trained model for handwritten digit classification using the Modified National Institute of Standard and Technology (MNIST) dataset. Hardware synthesis using 45 nm CMOS technology shows that the MSDF-based MLP achieves a favorable trade-off between hardware metrics, such as reduced circuit area and energy per neuron, and reaches a good classification accuracy of 97.4%. In the case of the autoencoder, the proposed MAC unit is utilized in a pre-trained denoising autoencoder for the same dataset. The autoencoder employs MSDF-based mixed precision across layers to reduce computational resources and loads while preserving high-quality image reconstruction. A comprehensive evaluation of image correctness, signal fidelity, and visual quality metrics was conducted to assess the impact of varying precision levels across different layers. The results indicate that the proposed approach achieves an effective denoising performance with balanced hardware requirements.
- Open Access
- 28 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Instrumentation and Measurement
- Vol. 3, 1-8
Baseline-Free Damage Imaging for Structural Health Monitoring of Composite Lap Joint Using Ultrasonic-Guided Waves
Damage imaging algorithms are crucial for evaluating the condition of critical structures such as adhesively bonded joints. Particularly during service, baseline-free structural health monitoring is essential for robust and real-time evaluation. This paper proposes and investigates the impact of the shape of the damage intensity distribution and damage index on the damage imaging of composite lap joints using a baseline-free structural health monitoring system. This system comprises a parallel array of piezoelectric transducers attached to both sides of the lap joint for generating and receiving ultrasonic guided waves. Various features are extracted from the received signals to serve as damage indices, representing the peak amplitude and energy of the signals as well as the time-of-flight. Different shapes of damage intensity distribution, including elliptical, diamond, rectangular, and quadrilateral, are considered between pairs of sensors to investigate their effects on the total damage intensity distribution. To evaluate the impact of these parameters, a 2D correlation coefficient was employed to compare the results obtained from the baseline-free structural health monitoring system with the image containing actual defects. The results reveal that the time-of-flight was ineffective in providing high correlation and considering the signal’s energy with quadrilateral shape achieved the highest correlation.
- 28 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Instrumentation and Measurement
- Vol. 3, 1-8
Open AccessBaseline-Free Damage Imaging for Structural Health Monitoring of Composite Lap Joint Using Ultrasonic-Guided Waves
Damage imaging algorithms are crucial for evaluating the condition of critical structures such as adhesively bonded joints. Particularly during service, baseline-free structural health monitoring is essential for robust and real-time evaluation. This paper proposes and investigates the impact of the shape of the damage intensity distribution and damage index on the damage imaging of composite lap joints using a baseline-free structural health monitoring system. This system comprises a parallel array of piezoelectric transducers attached to both sides of the lap joint for generating and receiving ultrasonic guided waves. Various features are extracted from the received signals to serve as damage indices, representing the peak amplitude and energy of the signals as well as the time-of-flight. Different shapes of damage intensity distribution, including elliptical, diamond, rectangular, and quadrilateral, are considered between pairs of sensors to investigate their effects on the total damage intensity distribution. To evaluate the impact of these parameters, a 2D correlation coefficient was employed to compare the results obtained from the baseline-free structural health monitoring system with the image containing actual defects. The results reveal that the time-of-flight was ineffective in providing high correlation and considering the signal’s energy with quadrilateral shape achieved the highest correlation.