Pub Date : 2025-10-10DOI: 10.1016/j.ndteint.2025.103570
Kuohai Yu , Rui Guo , Saibo She , Lei Xiong , Xinnan Zheng , Xun Zou , Jialong Shen , Wuliang Yin
Sensor tilt is regarded as one of the major causes of noise in eddy current testing. A tilted Pulsed Eddy Current (PEC) probe can lead to signal distortion, resulting in errors in measurements and evaluations. For the first time, this paper investigates the tilt effect on PEC signals by developing an analytical solution for tilted PEC sensors. The analytical solution combines the theory of mutual impedance variation of tilted coils with PEC testing. It is found that the impact of tilt angles on PEC signals follows a double-exponential function in terms of both amplitude and the decreasing rate of transient voltage change. Corresponding experiments have been conducted, which agree well with the numerical results and validate the analytical solutions. Additionally, a sensor tilt angle estimation method based on the double-exponential relationship curve is developed and an average absolute error of 0.2829° has been achieved.
{"title":"Tilt effects analysis and evaluation in Pulsed Eddy Current measurements","authors":"Kuohai Yu , Rui Guo , Saibo She , Lei Xiong , Xinnan Zheng , Xun Zou , Jialong Shen , Wuliang Yin","doi":"10.1016/j.ndteint.2025.103570","DOIUrl":"10.1016/j.ndteint.2025.103570","url":null,"abstract":"<div><div>Sensor tilt is regarded as one of the major causes of noise in eddy current testing. A tilted Pulsed Eddy Current (PEC) probe can lead to signal distortion, resulting in errors in measurements and evaluations. For the first time, this paper investigates the tilt effect on PEC signals by developing an analytical solution for tilted PEC sensors. The analytical solution combines the theory of mutual impedance variation of tilted coils with PEC testing. It is found that the impact of tilt angles on PEC signals follows a double-exponential function in terms of both amplitude and the decreasing rate of transient voltage change. Corresponding experiments have been conducted, which agree well with the numerical results and validate the analytical solutions. Additionally, a sensor tilt angle estimation method based on the double-exponential relationship curve is developed and an average absolute error of 0.2829° has been achieved.</div></div>","PeriodicalId":18868,"journal":{"name":"Ndt & E International","volume":"158 ","pages":"Article 103570"},"PeriodicalIF":4.5,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-10DOI: 10.1016/j.ndteint.2025.103578
Fanfu Wu, Lei Zhou, Claire Davis
An electromagnetic (EM) sensor array system, consisting of four sensor heads, has been used to non-destructively characterise spatial variations in steel strips, with differences in phase transformation rates due to differences in local cooling rates being reported in this work. An S355-grade steel strip was subjected to different local cooling conditions (air cooling and water cooling, with half the strip being insulated) on a lab-based run-out table (ROT). The sensor array was able to monitor the different phase transformation behaviour across the width of the steel strip due to the non-uniform cooling. Thermocouples were used to determine the local cooling rates, and these were used with continuous cooling transformation (CCT) diagrams to predict the local phase transformation behaviour. It is known that the zero-crossing frequency (ZCF) from EM sensors can be related to the phase transformation; therefore, the ZCF values from the separate EM sensor heads have been compared to the predicted phase transformation behaviour. Microstructural validation for the predicted phase transformation products (fractions of ferrite, pearlite, bainite and/or martensite) was performed using optical microscopy. The spatial resolution performance of the EM sensor array has been compared to that of the commercial EMspec™ system for the case of varying phase transformation across a strip sample. This work demonstrates the potential for EM sensors to be used in arrays without interference between signals, allowing the characterisation of spatially varying behaviour in steel during cooling.
{"title":"EM sensor array for non-destructive evaluation of spatially varying steel phase transformation","authors":"Fanfu Wu, Lei Zhou, Claire Davis","doi":"10.1016/j.ndteint.2025.103578","DOIUrl":"10.1016/j.ndteint.2025.103578","url":null,"abstract":"<div><div>An electromagnetic (EM) sensor array system, consisting of four sensor heads, has been used to non-destructively characterise spatial variations in steel strips, with differences in phase transformation rates due to differences in local cooling rates being reported in this work. An S355-grade steel strip was subjected to different local cooling conditions (air cooling and water cooling, with half the strip being insulated) on a lab-based run-out table (ROT). The sensor array was able to monitor the different phase transformation behaviour across the width of the steel strip due to the non-uniform cooling. Thermocouples were used to determine the local cooling rates, and these were used with continuous cooling transformation (CCT) diagrams to predict the local phase transformation behaviour. It is known that the zero-crossing frequency (ZCF) from EM sensors can be related to the phase transformation; therefore, the ZCF values from the separate EM sensor heads have been compared to the predicted phase transformation behaviour. Microstructural validation for the predicted phase transformation products (fractions of ferrite, pearlite, bainite and/or martensite) was performed using optical microscopy. The spatial resolution performance of the EM sensor array has been compared to that of the commercial EMspec™ system for the case of varying phase transformation across a strip sample. This work demonstrates the potential for EM sensors to be used in arrays without interference between signals, allowing the characterisation of spatially varying behaviour in steel during cooling.</div></div>","PeriodicalId":18868,"journal":{"name":"Ndt & E International","volume":"158 ","pages":"Article 103578"},"PeriodicalIF":4.5,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145324881","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-10DOI: 10.1016/j.ndteint.2025.103579
Zhigang Cheng , Zhizhou He , Peng Pan
Detecting subsurface pipes and cavities is important in urban infrastructure management, but existing methods struggle to accurately reconstruct the 3D shapes of deep subsurface objects. This study pioneers a new paradigm for this task by reformulating the ill-posed permittivity regression problem as a 3D semantic segmentation problem. A novel neural network, 3DReconNet, to predict the material type of each subsurface voxel from ground penetrating radar (GPR) data was proposed. This approach leverages the intrinsic relationship between material composition and reflected signal intensity to simultaneously recover both geometry and material properties. A dataset of 3150 synthetic cases was generated using full-scale simulation models and a Markov model-based algorithm to simulate irregular cavities. The 3DReconNet adopts a U-shaped architecture and incorporates residual connections to reduce information loss. The network is trained using the Dice Loss function regularized with total variation (TV) constraints, which enhances geometric consistency and reconstruction accuracy. The proposed method was validated using both simulated and experimental data, and the qualitative as well as quantitative results confirmed its effectiveness, robustness, and generalizability.
{"title":"3D reconstruction of subsurface pipes and cavities using ground penetrating radar based on deep learning","authors":"Zhigang Cheng , Zhizhou He , Peng Pan","doi":"10.1016/j.ndteint.2025.103579","DOIUrl":"10.1016/j.ndteint.2025.103579","url":null,"abstract":"<div><div>Detecting subsurface pipes and cavities is important in urban infrastructure management, but existing methods struggle to accurately reconstruct the 3D shapes of deep subsurface objects. This study pioneers a new paradigm for this task by reformulating the ill-posed permittivity regression problem as a 3D semantic segmentation problem. A novel neural network, 3DReconNet, to predict the material type of each subsurface voxel from ground penetrating radar (GPR) data was proposed. This approach leverages the intrinsic relationship between material composition and reflected signal intensity to simultaneously recover both geometry and material properties. A dataset of 3150 synthetic cases was generated using full-scale simulation models and a Markov model-based algorithm to simulate irregular cavities. The 3DReconNet adopts a U-shaped architecture and incorporates residual connections to reduce information loss. The network is trained using the Dice Loss function regularized with total variation (TV) constraints, which enhances geometric consistency and reconstruction accuracy. The proposed method was validated using both simulated and experimental data, and the qualitative as well as quantitative results confirmed its effectiveness, robustness, and generalizability.</div></div>","PeriodicalId":18868,"journal":{"name":"Ndt & E International","volume":"158 ","pages":"Article 103579"},"PeriodicalIF":4.5,"publicationDate":"2025-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145324880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-09DOI: 10.1016/j.ndteint.2025.103565
M. Weiss , K. Mrzljak , M. von Schmid , G. Erbach , N. Brierley , T. Meisen
The demand for batteries as portable energy storages increases drastically. Especially for electric mobility, battery safety is crucial which begins at seamless quality control during and after manufacturing. Recent developments in high-speed computed tomography (high-speed CT) enable scan times around 10 s, roughly matching the speed of a typical battery production line. While the majority of defects in batteries can be detected using the CT scan data directly, data post-processing such as material identification can reveal further insights. As the complexity of modern battery production grows, traditional material-resolving CT methods face challenges in delivering the precision and efficiency required. To meet these demands, more advanced, data-driven approaches are becoming essential. This has led to an ongoing paradigm shift in material-resolving CT, introducing deep learning techniques that promise enhanced accuracy and processing speed. In the scope of this paper, we propose an end-to-end deep learning approach, which is designed to resolve materials in CT scans in presence of heavy CT artifacts by exploiting context knowledge with a convolution-based neural network. The model computes atomic numbers and densities directly from the dual-energy CT volume slices for each pixel. Our approach uses simulation-generated training data, thereby avoiding the need for manual annotation. CT scans from two fundamentally different systems – one providing slow, high-quality scans and the other fast, medium-quality scans – are compared in terms of material identification performance. Especially for high-speed CT, increasing the scanning time can influence the data quality drastically. We believe, that the combination of a high-speed scanner for pre-screening together with a slower high-quality scanner provides comprehensive in-line inspection, where only critical candidates, revealing anomalies in the high-speed scan, will be send to the high-quality scanner.
{"title":"Material-resolving computed tomography of lithium-ion batteries using deep learning","authors":"M. Weiss , K. Mrzljak , M. von Schmid , G. Erbach , N. Brierley , T. Meisen","doi":"10.1016/j.ndteint.2025.103565","DOIUrl":"10.1016/j.ndteint.2025.103565","url":null,"abstract":"<div><div>The demand for batteries as portable energy storages increases drastically. Especially for electric mobility, battery safety is crucial which begins at seamless quality control during and after manufacturing. Recent developments in high-speed computed tomography (high-speed CT) enable scan times around 10 s, roughly matching the speed of a typical battery production line. While the majority of defects in batteries can be detected using the CT scan data directly, data post-processing such as material identification can reveal further insights. As the complexity of modern battery production grows, traditional material-resolving CT methods face challenges in delivering the precision and efficiency required. To meet these demands, more advanced, data-driven approaches are becoming essential. This has led to an ongoing paradigm shift in material-resolving CT, introducing deep learning techniques that promise enhanced accuracy and processing speed. In the scope of this paper, we propose an end-to-end deep learning approach, which is designed to resolve materials in CT scans in presence of heavy CT artifacts by exploiting context knowledge with a convolution-based neural network. The model computes atomic numbers and densities directly from the dual-energy CT volume slices for each pixel. Our approach uses simulation-generated training data, thereby avoiding the need for manual annotation. CT scans from two fundamentally different systems – one providing slow, high-quality scans and the other fast, medium-quality scans – are compared in terms of material identification performance. Especially for high-speed CT, increasing the scanning time can influence the data quality drastically. We believe, that the combination of a high-speed scanner for pre-screening together with a slower high-quality scanner provides comprehensive in-line inspection, where only critical candidates, revealing anomalies in the high-speed scan, will be send to the high-quality scanner.</div></div>","PeriodicalId":18868,"journal":{"name":"Ndt & E International","volume":"158 ","pages":"Article 103565"},"PeriodicalIF":4.5,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145324883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-08DOI: 10.1016/j.ndteint.2025.103576
Mingqian Xia, Kohei Nishiuchi, Takahiro Hayashi, Naoki Mori
The authors have previously investigated defect imaging using a phased array probe with a buffer consisting of thin plates. Although the phased array probe with a stacked plate buffer works well in defect imaging, there remains the issue of spurious images due to trailing waves generating at the side walls of a plate, and large stacked plate buffers are required to avoid the trailing waves. To solve these issues, a buffer consisting of circular cylinders is introduced. Considering dispersion characteristics of longitudinal vibration mode of guided waves in a circular cylinder and dimensions of phased array probe, cylinder buffers were designed and fabricated. Using the buffer consisting of circular cylinders, defects were well visualized with two imaging algorithms, plane wave imaging and total focusing method.
{"title":"Ultrasonic imaging using a phased array probe with a buffer consisting of a bundle of circular cylinders","authors":"Mingqian Xia, Kohei Nishiuchi, Takahiro Hayashi, Naoki Mori","doi":"10.1016/j.ndteint.2025.103576","DOIUrl":"10.1016/j.ndteint.2025.103576","url":null,"abstract":"<div><div>The authors have previously investigated defect imaging using a phased array probe with a buffer consisting of thin plates. Although the phased array probe with a stacked plate buffer works well in defect imaging, there remains the issue of spurious images due to trailing waves generating at the side walls of a plate, and large stacked plate buffers are required to avoid the trailing waves. To solve these issues, a buffer consisting of circular cylinders is introduced. Considering dispersion characteristics of longitudinal vibration mode of guided waves in a circular cylinder and dimensions of phased array probe, cylinder buffers were designed and fabricated. Using the buffer consisting of circular cylinders, defects were well visualized with two imaging algorithms, plane wave imaging and total focusing method.</div></div>","PeriodicalId":18868,"journal":{"name":"Ndt & E International","volume":"158 ","pages":"Article 103576"},"PeriodicalIF":4.5,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-08DOI: 10.1016/j.ndteint.2025.103577
Fengling Wang , Shuzeng Zhang , Mingzhu Sun , Tribikram Kundu
This study proposes a nonlinear ultrasonic imaging method for the detection of material or structural damage. The contact-based frequency-mismatched pulse-echo sideband peak intensity (PE-SPI) technique is extended and implemented on an ultrasonic immersion C-scan platform, enabling non-contact scanning and imaging based on nonlinear parameters. Fatigue test specimens were examined using both the proposed method and conventional linear scanning approaches. The results indicate that, when linear parameters such as signal amplitude are used, the outcomes from both methods are consistent. However, the proposed method enables the extraction of nonlinear features by measuring the amplitudes of harmonic peaks in the frequency spectrum, thereby realizing an imaging approach fundamentally different from traditional linear ultrasonics. Experimental results demonstrate that the proposed technique more effectively identifies the locations of fatigue cracks, showing particularly enhanced sensitivity in detecting early-stage cracks and assessing crack extension.
{"title":"Nonlinear ultrasonic C-scan imaging based on sideband peak intensity for fatigue damage evaluation","authors":"Fengling Wang , Shuzeng Zhang , Mingzhu Sun , Tribikram Kundu","doi":"10.1016/j.ndteint.2025.103577","DOIUrl":"10.1016/j.ndteint.2025.103577","url":null,"abstract":"<div><div>This study proposes a nonlinear ultrasonic imaging method for the detection of material or structural damage. The contact-based frequency-mismatched pulse-echo sideband peak intensity (PE-SPI) technique is extended and implemented on an ultrasonic immersion C-scan platform, enabling non-contact scanning and imaging based on nonlinear parameters. Fatigue test specimens were examined using both the proposed method and conventional linear scanning approaches. The results indicate that, when linear parameters such as signal amplitude are used, the outcomes from both methods are consistent. However, the proposed method enables the extraction of nonlinear features by measuring the amplitudes of harmonic peaks in the frequency spectrum, thereby realizing an imaging approach fundamentally different from traditional linear ultrasonics. Experimental results demonstrate that the proposed technique more effectively identifies the locations of fatigue cracks, showing particularly enhanced sensitivity in detecting early-stage cracks and assessing crack extension.</div></div>","PeriodicalId":18868,"journal":{"name":"Ndt & E International","volume":"158 ","pages":"Article 103577"},"PeriodicalIF":4.5,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145324879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-07DOI: 10.1016/j.ndteint.2025.103573
Rongyan Wen, Chongcong Tao, Hongli Ji, Yuqing Qiu, Jinhao Qiu
Carbon Fiber Reinforced Plastic (CFRP) composites are susceptible to damage and defects, which may arise during manufacturing or operational stages, potentially compromising structural integrity. This study introduces a novel eddy current detection probe featuring a nine-grid design, which enhances spatial resolution and sensitivity for detecting impact damage in CFRP. The excitation coils of the probe were optimized to concentrate the majority of the eddy current energy in the localized CFRP region directly beneath the probe, thereby significantly enhancing detection sensitivity and performance. Utilizing excitation coils with phase variations, the probe generates an elliptically polarized electric field with rotational characteristics, facilitating more effective detection of impact-induced defects than conventional probes with linearly polarized fields. Validation experiments were carried out where the nine-grid probe showed a significant enhancement in detecting CFRP impact damage. The damage area can be quantified from the eddy current signal with a thresholding method which shows a positive correlation with the impact energy in CFRP orthotropic plates.
{"title":"Enhanced detection of impact damage in CFRP based on a novel eddy current probe","authors":"Rongyan Wen, Chongcong Tao, Hongli Ji, Yuqing Qiu, Jinhao Qiu","doi":"10.1016/j.ndteint.2025.103573","DOIUrl":"10.1016/j.ndteint.2025.103573","url":null,"abstract":"<div><div>Carbon Fiber Reinforced Plastic (CFRP) composites are susceptible to damage and defects, which may arise during manufacturing or operational stages, potentially compromising structural integrity. This study introduces a novel eddy current detection probe featuring a nine-grid design, which enhances spatial resolution and sensitivity for detecting impact damage in CFRP. The excitation coils of the probe were optimized to concentrate the majority of the eddy current energy in the localized CFRP region directly beneath the probe, thereby significantly enhancing detection sensitivity and performance. Utilizing excitation coils with phase variations, the probe generates an elliptically polarized electric field with rotational characteristics, facilitating more effective detection of impact-induced defects than conventional probes with linearly polarized fields. Validation experiments were carried out where the nine-grid probe showed a significant enhancement in detecting CFRP impact damage. The damage area can be quantified from the eddy current signal with a thresholding method which shows a positive correlation with the impact energy <span><math><mrow><mo><</mo><mn>6</mn><mspace></mspace><mi>J</mi></mrow></math></span> in CFRP orthotropic plates.</div></div>","PeriodicalId":18868,"journal":{"name":"Ndt & E International","volume":"158 ","pages":"Article 103573"},"PeriodicalIF":4.5,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145324878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The aim of this work is to enhance the microwave waveguide-based non-destructive detection of glass fibre-reinforced polymer (GFRP) composites by introducing new strategies for probe design and signal processing. The tapering geometry in the waveguide probe design improves the spatial resolution and sensitivity, and the additive manufacturing technique employed reduces the overall cost. In addition, a new approach is proposed for the optimal selection of the inspection frequency in the analysis of the raw frequency-domain data, which achieves high signal contrast. The spatial Fourier transform is introduced to eliminate the undesirable stand-off distance effect in the conventional waveguide-based inspection. Test samples with subsurface grooves and impact-induced damage were examined. It was found that a 1 mm wide groove at a depth of 9 mm and a 10 J barely visible impact damage were well detected and characterised. The results demonstrate the significant potential of microwave testing for the evaluation of composite structures.
{"title":"Enhanced microwave waveguide probe-based methods for damage detection of GFRP composites","authors":"Zhen Li , Zhaozong Meng , Fei Fei , Constantinos Soutis","doi":"10.1016/j.ndteint.2025.103572","DOIUrl":"10.1016/j.ndteint.2025.103572","url":null,"abstract":"<div><div>The aim of this work is to enhance the microwave waveguide-based non-destructive detection of glass fibre-reinforced polymer (GFRP) composites by introducing new strategies for probe design and signal processing. The tapering geometry in the waveguide probe design improves the spatial resolution and sensitivity, and the additive manufacturing technique employed reduces the overall cost. In addition, a new approach is proposed for the optimal selection of the inspection frequency in the analysis of the raw frequency-domain data, which achieves high signal contrast. The spatial Fourier transform is introduced to eliminate the undesirable stand-off distance effect in the conventional waveguide-based inspection. Test samples with subsurface grooves and impact-induced damage were examined. It was found that a 1 mm wide groove at a depth of 9 mm and a 10 J barely visible impact damage were well detected and characterised. The results demonstrate the significant potential of microwave testing for the evaluation of composite structures.</div></div>","PeriodicalId":18868,"journal":{"name":"Ndt & E International","volume":"158 ","pages":"Article 103572"},"PeriodicalIF":4.5,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268785","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-06DOI: 10.1016/j.ndteint.2025.103574
Taemin Lee , Myung-Hun Lee , Jinyoung Hong , Jiyoung Min , Hajin Choi
Glass fiber-reinforced polymer (GFRP) has recently emerged as a promising alternative to traditional steel reinforcement in concrete due to its superior durability. However, conventional non-destructive testing (NDT) methods often face limitations in detecting GFRP rebars, posing challenges for the maintenance and safety assessment of civil infrastructure. This study evaluates the applicability of ground-penetrating radar (GPR) for detecting GFRP reinforcement in concrete through both numerical simulation and experimental validation. The investigation focuses on the influence of antenna polarization and concrete moisture conditions on electromagnetic (EM) wave-based detection. Numerical simulations confirmed that increased moisture in concrete enhances dielectric contrast, thereby improving the visibility of GFRP bars. For experimental validation, two concrete specimens—a beam and a slab embedded with GFRP reinforcement—were prepared and tested. The results revealed that EM wave reflection energy increased by up to 17.0 % and 15.8 % under wet conditions using cross and normal polarizations, respectively. These findings underscore the significance of selecting appropriate antenna polarization and accounting for moisture conditions to improve the detection accuracy of GFRP rebars using GPR.
{"title":"Influence of antenna polarization and moisture content on detection of GFPR bars in concrete using ground penetrating radar","authors":"Taemin Lee , Myung-Hun Lee , Jinyoung Hong , Jiyoung Min , Hajin Choi","doi":"10.1016/j.ndteint.2025.103574","DOIUrl":"10.1016/j.ndteint.2025.103574","url":null,"abstract":"<div><div>Glass fiber-reinforced polymer (GFRP) has recently emerged as a promising alternative to traditional steel reinforcement in concrete due to its superior durability. However, conventional non-destructive testing (NDT) methods often face limitations in detecting GFRP rebars, posing challenges for the maintenance and safety assessment of civil infrastructure. This study evaluates the applicability of ground-penetrating radar (GPR) for detecting GFRP reinforcement in concrete through both numerical simulation and experimental validation. The investigation focuses on the influence of antenna polarization and concrete moisture conditions on electromagnetic (EM) wave-based detection. Numerical simulations confirmed that increased moisture in concrete enhances dielectric contrast, thereby improving the visibility of GFRP bars. For experimental validation, two concrete specimens—a beam and a slab embedded with GFRP reinforcement—were prepared and tested. The results revealed that EM wave reflection energy increased by up to 17.0 % and 15.8 % under wet conditions using cross and normal polarizations, respectively. These findings underscore the significance of selecting appropriate antenna polarization and accounting for moisture conditions to improve the detection accuracy of GFRP rebars using GPR.</div></div>","PeriodicalId":18868,"journal":{"name":"Ndt & E International","volume":"158 ","pages":"Article 103574"},"PeriodicalIF":4.5,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-06DOI: 10.1016/j.ndteint.2025.103569
Ayush Pratap , Neha Sardana , Tao Wu , P. Karthikeyan , Pao-Ann Hsiung
Detection of anomalies in 3D-printed magnesium alloy while printing is difficult because of the reactive nature of the material. In alignment with the principles of Non-Destructive Testing (NDT) 4.0, which emphasizes the inspection of advanced manufacturing processes and fully automated systems, this work presents a novel approach for anomaly detection in additively manufactured parts. Three Mg-based alloy cubes were printed through Selective Laser Melting (SLM) at different scan rates, and X-ray Computed Tomography (XCT) scan was employed to generate the image slices of all three samples. The novel data from all three samples has been selected to segment the anomaly from the printed part. The work has incorporated an innovative approach of adding a saliency map to the model for segmenting the different 3D printed volumes. Incorporating attention layers into the U-net algorithm enhances the learning characteristics of the model by emphasizing the specific region concerning the saliency map. It was found that by using an attention layer in the model, the accuracy in the segmentation of anomalies has been increased compared to simple U-net and other transfer learning approaches as a backbone. The proposed methodology with salient connection has achieved the Dice similarity coefficient (DSC) and Intersection over union (IOU) of 98.29% and 96.67% respectively, demonstrating its effectiveness in the context of NDT 4.0 for the inspection of additively manufactured components. Further aligning the proposed DAL-AD (Deep Attention Learning for Anomaly Detection) framework with broader industrial segments such as Industry 5.0 and ISO 9000, this work enables AI-assisted, sustainable, and in-situ quality control in additive manufacturing.
由于材料的反应性，在打印时检测3d打印镁合金中的异常是困难的。与无损检测（NDT） 4.0的原则一致，该原则强调对先进制造过程和全自动系统的检查，本工作提出了一种用于增材制造零件异常检测的新方法。采用选择性激光熔化法（SLM）以不同的扫描速率打印出3个mg基合金立方体，并采用x射线计算机断层扫描（XCT）扫描生成3个样品的图像切片。从所有三个样本中选择新的数据来从印刷部分中分割异常。这项工作采用了一种创新的方法，即在模型中添加显著性地图，以分割不同的3D打印体积。将注意层纳入U-net算法中，通过强调与显著性图有关的特定区域，增强了模型的学习特性。研究发现，与简单的U-net和其他迁移学习方法作为主干相比，在模型中使用注意层可以提高异常分割的准确性。所提出的具有显著连接的方法分别实现了98.29%的Dice相似系数（DSC）和96.67%的Intersection over union (IOU)，证明了其在无损检测4.0背景下对增材制造部件检测的有效性。进一步将拟议的DAL-AD（深度注意学习异常检测）框架与工业5.0和ISO 9000等更广泛的工业领域结合起来，这项工作使人工智能辅助的、可持续的、原位的增材制造质量控制成为可能。
{"title":"Revolutionizing NDT 4.0 with Deep Attention Learning for Anomaly Detection (DAL-AD) in Mg-based L-PBF components","authors":"Ayush Pratap , Neha Sardana , Tao Wu , P. Karthikeyan , Pao-Ann Hsiung","doi":"10.1016/j.ndteint.2025.103569","DOIUrl":"10.1016/j.ndteint.2025.103569","url":null,"abstract":"<div><div>Detection of anomalies in 3D-printed magnesium alloy while printing is difficult because of the reactive nature of the material. In alignment with the principles of Non-Destructive Testing (NDT) 4.0, which emphasizes the inspection of advanced manufacturing processes and fully automated systems, this work presents a novel approach for anomaly detection in additively manufactured parts. Three Mg-based alloy cubes were printed through Selective Laser Melting (SLM) at different scan rates, and X-ray Computed Tomography (XCT) scan was employed to generate the image slices of all three samples. The novel data from all three samples has been selected to segment the anomaly from the printed part. The work has incorporated an innovative approach of adding a saliency map to the model for segmenting the different 3D printed volumes. Incorporating attention layers into the U-net algorithm enhances the learning characteristics of the model by emphasizing the specific region concerning the saliency map. It was found that by using an attention layer in the model, the accuracy in the segmentation of anomalies has been increased compared to simple U-net and other transfer learning approaches as a backbone. The proposed methodology with salient connection has achieved the Dice similarity coefficient (DSC) and Intersection over union (IOU) of 98.29% and 96.67% respectively, demonstrating its effectiveness in the context of NDT 4.0 for the inspection of additively manufactured components. Further aligning the proposed DAL-AD (Deep Attention Learning for Anomaly Detection) framework with broader industrial segments such as Industry 5.0 and ISO 9000, this work enables AI-assisted, sustainable, and in-situ quality control in additive manufacturing.</div></div>","PeriodicalId":18868,"journal":{"name":"Ndt & E International","volume":"158 ","pages":"Article 103569"},"PeriodicalIF":4.5,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145268788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号