Pub Date : 2024-01-28DOI: 10.1007/s10921-023-01041-w
Abstract
Artificial intelligence is providing new possibilities for analysis in the field of industrial radiography. As capabilities evolve, there is the need for knowledge concerning how to deploy these technologies in practice and benefit from the new automatically generated information. In this study, automatic defect recognition based on machine learning was deployed as an aid in industrial radiography of laser welds in an aerospace component, and utilized to produce statistics for improved quality control. A multi-model approach with an added weld segmentation step improved the inference speed and decreased false calls to improve field use. A user interface with visualization options was developed to display the evaluation results. A dataset of 451 radiographs was automatically analysed, yielding 10037 indications with size and location information, providing capability for statistical analysis beyond what is practical to carry out with manual annotation. The distribution of indications was modeled as a product of the probability of detection and an exponentially decreasing underlying flaw distribution, opening the possibility for model reliability assessment and predictive capabilities on weld defects. An analysis of the indications demonstrated the capability to automatically detect both large-scale trends and individual components and welds that were more at risk of failing the inspection. This serves as a step towards smarter utilization of non-destructive evaluation data in manufacturing.
{"title":"Deploying Machine Learning for Radiography of Aerospace Welds","authors":"","doi":"10.1007/s10921-023-01041-w","DOIUrl":"https://doi.org/10.1007/s10921-023-01041-w","url":null,"abstract":"<h3>Abstract</h3> <p>Artificial intelligence is providing new possibilities for analysis in the field of industrial radiography. As capabilities evolve, there is the need for knowledge concerning how to deploy these technologies in practice and benefit from the new automatically generated information. In this study, automatic defect recognition based on machine learning was deployed as an aid in industrial radiography of laser welds in an aerospace component, and utilized to produce statistics for improved quality control. A multi-model approach with an added weld segmentation step improved the inference speed and decreased false calls to improve field use. A user interface with visualization options was developed to display the evaluation results. A dataset of 451 radiographs was automatically analysed, yielding 10037 indications with size and location information, providing capability for statistical analysis beyond what is practical to carry out with manual annotation. The distribution of indications was modeled as a product of the probability of detection and an exponentially decreasing underlying flaw distribution, opening the possibility for model reliability assessment and predictive capabilities on weld defects. An analysis of the indications demonstrated the capability to automatically detect both large-scale trends and individual components and welds that were more at risk of failing the inspection. This serves as a step towards smarter utilization of non-destructive evaluation data in manufacturing.</p>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2024-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139583598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2024-01-24DOI: 10.1007/s10921-023-01038-5
Lawrence Yule, Nicholas Harris, Martyn Hill, Bahareh Zaghari
Ultrasonic guided waves offer a promising method of monitoring the online temperature of plate-like structures in extreme environments, such as aero-engine nozzle guide vanes (NGVs), and can provide the resolution, response rate, and robust operation that is required in aerospace. Previous investigations have shown the potential of such a system but the effect of the complex physical environment on wave propagation is yet to be considered. This article uses a numerical approach to investigate how thermal barrier coatings (TBCs) applied to the surface of many components designed for extreme thermal conditions will affect ultrasonic guided wave propagation, and how a system can be employed to monitor through-thickness temperature changes. The top coat/bond coat boundary in NGVs has been shown to be a temperature critical point that is difficult to monitor with traditional temperature sensors, which highlights the potential of ultrasonic guided waves. Differences in application method and layer thickness are considered, and analysis of through-thickness displacement profiles and dispersion curves are used to predict signal response and determine the most suitable mode of operation. Heat transfer simulations (COMSOL) have been used to predict temperature gradients within a TBC, and dispersion curves have been produced from the temperature dependant material properties. Time dependant simulations of wave propagation are in good agreement with dispersion curve predictions of wave velocity for the two lowest order modes in three thicknesses of TBC top coat (100, 250, and 500 (upmu hbox {m})). When wave velocity measurements from the simulations are compared to dispersion curves generated at isotropic temperatures, the corresponding temperature represents the average temperature of a gradient system well. Such a measurement system could, in principle, be used in conjunction with surface temperature measurement systems to monitor through-thickness temperature changes.
{"title":"Temperature Monitoring of Through-Thickness Temperature Gradients in Thermal Barrier Coatings Using Ultrasonic Guided Waves","authors":"Lawrence Yule, Nicholas Harris, Martyn Hill, Bahareh Zaghari","doi":"10.1007/s10921-023-01038-5","DOIUrl":"https://doi.org/10.1007/s10921-023-01038-5","url":null,"abstract":"<p>Ultrasonic guided waves offer a promising method of monitoring the online temperature of plate-like structures in extreme environments, such as aero-engine nozzle guide vanes (NGVs), and can provide the resolution, response rate, and robust operation that is required in aerospace. Previous investigations have shown the potential of such a system but the effect of the complex physical environment on wave propagation is yet to be considered. This article uses a numerical approach to investigate how thermal barrier coatings (TBCs) applied to the surface of many components designed for extreme thermal conditions will affect ultrasonic guided wave propagation, and how a system can be employed to monitor through-thickness temperature changes. The top coat/bond coat boundary in NGVs has been shown to be a temperature critical point that is difficult to monitor with traditional temperature sensors, which highlights the potential of ultrasonic guided waves. Differences in application method and layer thickness are considered, and analysis of through-thickness displacement profiles and dispersion curves are used to predict signal response and determine the most suitable mode of operation. Heat transfer simulations (COMSOL) have been used to predict temperature gradients within a TBC, and dispersion curves have been produced from the temperature dependant material properties. Time dependant simulations of wave propagation are in good agreement with dispersion curve predictions of wave velocity for the two lowest order modes in three thicknesses of TBC top coat (100, 250, and 500 <span>(upmu hbox {m})</span>). When wave velocity measurements from the simulations are compared to dispersion curves generated at isotropic temperatures, the corresponding temperature represents the average temperature of a gradient system well. Such a measurement system could, in principle, be used in conjunction with surface temperature measurement systems to monitor through-thickness temperature changes.</p>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2024-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139561736","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this paper, Monte Carlo simulations are performed based on the two-dimensional Ising model with the objective of matching the simulated magnetic Barkhausen noise (MBN) signals with the measured MBN signals obtained from empirical research on bearing steel of different hardness levels. Firstly, the methods for obtaining simulated MBN signals based on the Ising model are studied. This paper suggests that simulated MBN signals obtained by applying a digital filter to the simulated magnetization curve, both in the time domain and frequency spectrum, are closer to the actual measured signals. Secondly, the influencing factors of the two-dimensional Ising model are studied, including lattice size (N), temperature (T), neighbor interaction (J), external magnetic field (H(t)), number of simulation points per period ((P_{sim})) and Monte Carlo step (MCS). Furthermore, the simulated MBN signals and their feature diagrams under different temperatures and neighbor interactions are plotted. Finally, a method is proposed to match the simulated MBN signals with the actual measured MBN signals using scaling and shifting, reducing the relative error between the simulated and measured MBN signal features to within 7%. This method makes it possible to generate simulated MBN signals at different hardness levels.
{"title":"Ising Model Simulation and Empirical Research of Barkhausen Noise","authors":"Cheng Hang, Wenbo Liu, Gerd Dobmann, Yin Wu, Wangcai Chen, Ping Wang","doi":"10.1007/s10921-023-01037-6","DOIUrl":"https://doi.org/10.1007/s10921-023-01037-6","url":null,"abstract":"<p>In this paper, Monte Carlo simulations are performed based on the two-dimensional Ising model with the objective of matching the simulated magnetic Barkhausen noise (MBN) signals with the measured MBN signals obtained from empirical research on bearing steel of different hardness levels. Firstly, the methods for obtaining simulated MBN signals based on the Ising model are studied. This paper suggests that simulated MBN signals obtained by applying a digital filter to the simulated magnetization curve, both in the time domain and frequency spectrum, are closer to the actual measured signals. Secondly, the influencing factors of the two-dimensional Ising model are studied, including lattice size (<i>N</i>), temperature (<i>T</i>), neighbor interaction (<i>J</i>), external magnetic field (<i>H</i>(<i>t</i>)), number of simulation points per period (<span>(P_{sim})</span>) and Monte Carlo step (<i>MCS</i>). Furthermore, the simulated MBN signals and their feature diagrams under different temperatures and neighbor interactions are plotted. Finally, a method is proposed to match the simulated MBN signals with the actual measured MBN signals using scaling and shifting, reducing the relative error between the simulated and measured MBN signal features to within 7%. This method makes it possible to generate simulated MBN signals at different hardness levels.</p>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2024-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139561691","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Additive manufacturing brings inspection issues for quality assurance of final parts because non-destructive testing methods are faced with shape complexity, size, and high surface roughness. Thus, to drive additive manufacturing forward, advanced non-destructive testing methods are required. Methods based on resonant ultrasound spectroscopy (RUS) can take on all the challenges that come with additive manufacturing. Indeed, these full body inspection methods are adapted to shape complexity, to nearly any size, and to high degrees of surface roughness. Furthermore, they are easy to implement, fast and low cost. In this paper, we present the benefit of a resonant ultrasound spectroscopy method, combined with a statistical analysis through Z score implementation, to classify supposedly identical parts, from a batch comprised of several individual builds. We also demonstrate that the inspection can be further accelerated and automated, to make the analysis operator independent, whether the analysis of the resonant ultrasound spectroscopy data is performed supervised or unsupervised with machine learning algorithms.
增材制造给最终零件的质量保证带来了检测问题,因为非破坏性检测方法要面对形状复杂、尺寸大和表面粗糙度高的问题。因此,要推动增材制造向前发展,就需要先进的无损检测方法。基于共振超声波谱(RUS)的方法可以应对增材制造带来的所有挑战。事实上,这些全身检测方法可适应形状复杂性、几乎任何尺寸和高度表面粗糙度。此外,它们易于实施、速度快、成本低。在本文中,我们介绍了共振超声波光谱方法的优点,该方法结合了通过 Z 分数实施的统计分析,可对由多个单个构建组成的批次中假定相同的部件进行分类。我们还证明,无论是使用机器学习算法对共振超声波谱数据进行监督式分析还是非监督式分析,都可以进一步加快检测速度并实现自动化,从而使分析不受操作人员的影响。
{"title":"Statistical Analysis and Automation Through Machine Learning of Resonant Ultrasound Spectroscopy Data from Tests Performed on Complex Additively Manufactured Parts","authors":"Anne-Françoise Obaton, Nasim Fallahi, Anis Tanich, Louis-Ferdinand Lafon, Gregory Weaver","doi":"10.1007/s10921-023-01035-8","DOIUrl":"https://doi.org/10.1007/s10921-023-01035-8","url":null,"abstract":"<p>Additive manufacturing brings inspection issues for quality assurance of final parts because non-destructive testing methods are faced with shape complexity, size, and high surface roughness. Thus, to drive additive manufacturing forward, advanced non-destructive testing methods are required. Methods based on resonant ultrasound spectroscopy (RUS) can take on all the challenges that come with additive manufacturing. Indeed, these full body inspection methods are adapted to shape complexity, to nearly any size, and to high degrees of surface roughness. Furthermore, they are easy to implement, fast and low cost. In this paper, we present the benefit of a resonant ultrasound spectroscopy method, combined with a statistical analysis through Z score implementation, to classify supposedly identical parts, from a batch comprised of several individual builds. We also demonstrate that the inspection can be further accelerated and automated, to make the analysis operator independent, whether the analysis of the resonant ultrasound spectroscopy data is performed supervised or unsupervised with machine learning algorithms.</p>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2024-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139561708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2024-01-05DOI: 10.1007/s10921-023-01034-9
Tino Band, Benedikt Karrasch, Markus Patzold, Chia-Mei Lin, Ralph Gottschalg, Kai Kaufmann
Magnetic field measurements play a vital role in various industries, particularly in the detection of cracks using magnetic field images, also known as magnetic field leakage testing. This paper presents an approach to automate the extraction of crack signals in magnetic field imaging by using neural networks. The proposed method relies on simulation-based training using the lightweight Python library Magpylib to calculate the three-dimensional static magnetic field of permanent magnets with surface defects. This approach has numerous advantages. It allows control of training data set variance by tuning simulation input parameters such as sample magnetization, measurement parameters, and defect properties to cover a wide range of cracks in size and position. Starting data acquisition before system operation allows investigating potential changes in sample shape or measurement parameters. Importantly, simulation-based data generation eliminates the need for physical measurements, leading to significant time savings. The study presents and discusses results obtained on two different ferromagnetic samples with surface cracks, a hollow cylinder and a steel sheet.
{"title":"Simulation-Trained Neural Networks for Automatable Crack Detection in Magnetic Field Images","authors":"Tino Band, Benedikt Karrasch, Markus Patzold, Chia-Mei Lin, Ralph Gottschalg, Kai Kaufmann","doi":"10.1007/s10921-023-01034-9","DOIUrl":"10.1007/s10921-023-01034-9","url":null,"abstract":"<div><p>Magnetic field measurements play a vital role in various industries, particularly in the detection of cracks using magnetic field images, also known as magnetic field leakage testing. This paper presents an approach to automate the extraction of crack signals in magnetic field imaging by using neural networks. The proposed method relies on simulation-based training using the lightweight Python library Magpylib to calculate the three-dimensional static magnetic field of permanent magnets with surface defects. This approach has numerous advantages. It allows control of training data set variance by tuning simulation input parameters such as sample magnetization, measurement parameters, and defect properties to cover a wide range of cracks in size and position. Starting data acquisition before system operation allows investigating potential changes in sample shape or measurement parameters. Importantly, simulation-based data generation eliminates the need for physical measurements, leading to significant time savings. The study presents and discusses results obtained on two different ferromagnetic samples with surface cracks, a hollow cylinder and a steel sheet.</p></div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2024-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139104311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this work, we present a system architecture that is especially designed for remote ultrasound testing inspections (UT). The system itself is realized using a real-time session service and a web service based on a REST API (REST: Representational State Transfer; API: application programming interface). This web service is used to store data persistently, e.g., sample geometries, raw nondestructive testing (NDT) data and derived inspection results in a shared dataspace. In the current development state, the results consist of textures mapped onto the sample geometry. This approach allows us to display the UT results directly on the real sample using mixed reality technologies. We also implemented a feature to assist the inspector remotely by making use of the availability of this digital representation. Hence, it is necessary to share additional data like the current UT-signal, temporary position marks, user and device positions, etc. between the different participants. A real-time distribution of this highly dynamic data is required to create an effective assistance environment. Therefore, a separate session service is used. The inspection data generated in this temporary session can also be transferred to the afore mentioned dataspace to be saved persistently. The system features mixed reality visualization for the inspector and optionally a virtual reality or a 3D Desktop environment for one or more remote assistants.
在这项工作中,我们提出了一种专为远程超声检测(UT)设计的系统架构。系统本身是通过实时会话服务和基于 REST API 的网络服务实现的(REST:Representational State Transfer;API:Application Programming Interface):REST:Representational State Transfer;API:Application Programming Interface)。该网络服务用于在共享数据空间中持久存储数据,如样品几何形状、原始无损检测(NDT)数据和衍生检测结果。在当前的开发阶段,检测结果由映射到样品几何图形上的纹理组成。通过这种方法,我们可以使用混合现实技术在真实样品上直接显示 UT 结果。我们还实施了一项功能,利用这一数字表示的可用性远程协助检测人员。因此,有必要在不同参与者之间共享其他数据,如当前的 UT 信号、临时位置标记、用户和设备位置等。要创建有效的辅助环境,就必须实时分发这些高度动态的数据。因此,需要使用单独的会话服务。在该临时会话中生成的检测数据也可以传输到上述数据空间中进行持久保存。该系统的特点是为检查员提供混合现实可视化,并为一个或多个远程助手提供虚拟现实或 3D 桌面环境。
{"title":"3D Remote Assistance for NDT Inspections","authors":"Jörg Rehbein, Sebastian-Johannes Lorenz, Jens Holtmannspötter, Bernd Valeske","doi":"10.1007/s10921-023-01020-1","DOIUrl":"10.1007/s10921-023-01020-1","url":null,"abstract":"<div><p>In this work, we present a system architecture that is especially designed for remote ultrasound testing inspections (UT). The system itself is realized using a real-time session service and a web service based on a REST API (REST: Representational State Transfer; API: application programming interface). This web service is used to store data persistently, e.g., sample geometries, raw nondestructive testing (NDT) data and derived inspection results in a shared dataspace. In the current development state, the results consist of textures mapped onto the sample geometry. This approach allows us to display the UT results directly on the real sample using mixed reality technologies. We also implemented a feature to assist the inspector remotely by making use of the availability of this digital representation. Hence, it is necessary to share additional data like the current UT-signal, temporary position marks, user and device positions, etc. between the different participants. A real-time distribution of this highly dynamic data is required to create an effective assistance environment. Therefore, a separate session service is used. The inspection data generated in this temporary session can also be transferred to the afore mentioned dataspace to be saved persistently. The system features mixed reality visualization for the inspector and optionally a virtual reality or a 3D Desktop environment for one or more remote assistants.</p></div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2023-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10921-023-01020-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138946545","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-12-21DOI: 10.1007/s10921-023-01029-6
Yongkun Li, Yan Lyu, Bin Wu, Jie Gao, Zeqi Bian, Cunfu He
The experimental study of ultrasonic transmission coefficient measurement method was carried out for the problem of bonding quality inspection of bonded structures of composites with thick adhesive layer. For the uniaxial CFRP bonded specimens, the ultrasonic waves’ phase velocity distribution and the ultrasonic transmission coefficient spectrums were measured by the water-immersion ultrasonic transmission method. Then their complex elastic constants were inversed by the particle swarm optimization algorithm based on simulated annealing. Based on this, CFRP bonded specimens in perfect bonding status and having single weak bonding interface, dual weak bonding interfaces were manufactured. Then the ultrasonic transmission coefficient spectrums were measured, and the distinction of bonding quality between different bonding specimens was predicted. At the same time, according to the shift characteristics of the measured ultrasonic transmission coefficient spectra, the influence of the symmetry of the CFRP bonded structures with thick adhesive layers on the ultrasonic transmission coefficient spectrums was investigated experimentally. In addition, by comparing the time-domain and coefficient spectra characteristics of transmitted waves, the advantage of using the coefficient spectrum measurement method in bonding quality detection was expounded.
{"title":"Experimental Investigation of Bonding Quality for CFRP Bonded Structures with Thick Adhesive Layers Based on Ultrasonic Transmission Coefficient Spectrums","authors":"Yongkun Li, Yan Lyu, Bin Wu, Jie Gao, Zeqi Bian, Cunfu He","doi":"10.1007/s10921-023-01029-6","DOIUrl":"10.1007/s10921-023-01029-6","url":null,"abstract":"<div><p>The experimental study of ultrasonic transmission coefficient measurement method was carried out for the problem of bonding quality inspection of bonded structures of composites with thick adhesive layer. For the uniaxial CFRP bonded specimens, the ultrasonic waves’ phase velocity distribution and the ultrasonic transmission coefficient spectrums were measured by the water-immersion ultrasonic transmission method. Then their complex elastic constants were inversed by the particle swarm optimization algorithm based on simulated annealing. Based on this, CFRP bonded specimens in perfect bonding status and having single weak bonding interface, dual weak bonding interfaces were manufactured. Then the ultrasonic transmission coefficient spectrums were measured, and the distinction of bonding quality between different bonding specimens was predicted. At the same time, according to the shift characteristics of the measured ultrasonic transmission coefficient spectra, the influence of the symmetry of the CFRP bonded structures with thick adhesive layers on the ultrasonic transmission coefficient spectrums was investigated experimentally. In addition, by comparing the time-domain and coefficient spectra characteristics of transmitted waves, the advantage of using the coefficient spectrum measurement method in bonding quality detection was expounded.</p></div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2023-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138949112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2023-12-19DOI: 10.1007/s10921-023-01028-7
Shaohua Wang, Lihua Tang, Yinling Dou, Zhaoyu Li, Kean C. Aw
In this paper, the vibration-based image representation and data fusion demonstrates distinctive benefit in feature extraction, yielding superior performance for damage identification in railway engineering. Specifically, based on vehicle-track coupled dynamics, the rail vibration datasets under diverse fastener damage conditions are generated. By converting 1-D vibration signals into 2-D grayscale images with recurrence plots (RPs) and the aid of conditional variational autoencoder (CVAE), the acceleration RPs and displacement RPs are fused for enhancing feature extraction. It is demonstrated that detecting the variation in texture patterns and color distribution of the vibration-based images facilitates effective damage identification, mitigating the sensitivity of damage recognition to the deterioration of track irregularity. The results show that the displacement RPs characterised by quasi-static features are more suitable for fastener damage identification. Further, by employing the data fusion that combines both the random dynamic features of the acceleration RPs and quasi-static features of the displacement RPs, the tolerance of measurement range for accurate fastener damage identification can be extended. The robustness of the proposed method is validated after testing different sampling frequencies and additional noise.
{"title":"Enhancement of Track Damage Identification by Data Fusion of Vibration-Based Image Representation","authors":"Shaohua Wang, Lihua Tang, Yinling Dou, Zhaoyu Li, Kean C. Aw","doi":"10.1007/s10921-023-01028-7","DOIUrl":"10.1007/s10921-023-01028-7","url":null,"abstract":"<div><p>In this paper, the vibration-based image representation and data fusion demonstrates distinctive benefit in feature extraction, yielding superior performance for damage identification in railway engineering. Specifically, based on vehicle-track coupled dynamics, the rail vibration datasets under diverse fastener damage conditions are generated. By converting 1-D vibration signals into 2-D grayscale images with recurrence plots (RPs) and the aid of conditional variational autoencoder (CVAE), the acceleration RPs and displacement RPs are fused for enhancing feature extraction. It is demonstrated that detecting the variation in texture patterns and color distribution of the vibration-based images facilitates effective damage identification, mitigating the sensitivity of damage recognition to the deterioration of track irregularity. The results show that the displacement RPs characterised by quasi-static features are more suitable for fastener damage identification. Further, by employing the data fusion that combines both the random dynamic features of the acceleration RPs and quasi-static features of the displacement RPs, the tolerance of measurement range for accurate fastener damage identification can be extended. The robustness of the proposed method is validated after testing different sampling frequencies and additional noise.</p></div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2023-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138818444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2023-12-19DOI: 10.1007/s10921-023-01027-8
Satish Sonwane, Shital Chiddarwar
This study presents a Decision Support System (DSS) designed for Non-Destructive Online Evaluation in welding. Based on the Multi-Scale Dense Cross Block Network (MDCBNet), it is able to detect, classify, and recommend remedial actions to prevent surface defects in welding. The performance of the network architecture is enhanced with synthetic defect samples generated through image augmentation techniques. By employing gradient attribution and t-SNE plot methods, we gained insights into the network’s predictions and comprehensively analyzed decision-making process. Comparative evaluations against pre-trained deep learning techniques revealed that our proposed model exhibits significant improvements, ranging from 2 to 10% across various performance metrics. Extensive comparisons with state-of-the-art methods underscored the effectiveness of our approach in detecting and classifying weld defects. Notably, our network, initially trained on Gas Tungsten Arc Welding images, demonstrated remarkable adaptability and versatility by effectively classifying images from Gas Metal Arc Welding processes. These findings emphasize the potential of the MDCBNet-based DSS to enhance welding practices, thereby contributing to producing high-quality weldments. The successful implementation of our DSS recommendations further supports its capacity to optimize the welding process and facilitate improved weld quality.
{"title":"Developing a DSS for Enhancing Weldment Defect Detection, Classification, and Remediation Using HDR Images and Adaptive MDCBNet Neural Network","authors":"Satish Sonwane, Shital Chiddarwar","doi":"10.1007/s10921-023-01027-8","DOIUrl":"10.1007/s10921-023-01027-8","url":null,"abstract":"<div><p>This study presents a Decision Support System (DSS) designed for Non-Destructive Online Evaluation in welding. Based on the Multi-Scale Dense Cross Block Network (MDCBNet), it is able to detect, classify, and recommend remedial actions to prevent surface defects in welding. The performance of the network architecture is enhanced with synthetic defect samples generated through image augmentation techniques. By employing gradient attribution and t-SNE plot methods, we gained insights into the network’s predictions and comprehensively analyzed decision-making process. Comparative evaluations against pre-trained deep learning techniques revealed that our proposed model exhibits significant improvements, ranging from 2 to 10% across various performance metrics. Extensive comparisons with state-of-the-art methods underscored the effectiveness of our approach in detecting and classifying weld defects. Notably, our network, initially trained on Gas Tungsten Arc Welding images, demonstrated remarkable adaptability and versatility by effectively classifying images from Gas Metal Arc Welding processes. These findings emphasize the potential of the MDCBNet-based DSS to enhance welding practices, thereby contributing to producing high-quality weldments. The successful implementation of our DSS recommendations further supports its capacity to optimize the welding process and facilitate improved weld quality.</p></div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2023-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138818486","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2023-12-16DOI: 10.1007/s10921-023-01030-z
Saurabh Gupta, Siddesh Sutrave
The ice formation over the aerofoil structure of the aircraft wing has been an obstruction as they abrupt the airflow, acting as drag. The investigation will intend to determine ice accumulation on carbon fiber-reinforced polymer (CFRP), approximated as ice build-up on aircraft wings. The observation is carried out over quasi-isotropic composite laminates using ultrasonic-guided waves with a central working frequency regime of 100 kHz. The three-dimensional (3D) finite element (FE) simulations are performed to observe the scattering effect to explore the reflection site in the far field. This effect was quite prominent for different thicknesses of Glaze ice (G-Ice) and was found to be strongly linked with the wave propagation and dispersion effect. The scattering results for the reflection of Lamb mode, when it interacted with the G-Ice interface, were quite noteworthy along the angular region rather than on the center line, indicating that the scattering was more prominent due to the presence of a 45° or (− 45)-degree fiber orientation in that laminate. A similar but complex scattering phenomenon was observed for different stacking sequences where the wave propagation angle and its amplitude at the receiver nodes are found to be closely bound with the exponential decay in group/phase velocity for the ice thicknesses studied. The FE approach is verified, and the results are validated analytically. Analytically, we have investigated a much-closed approximation with the detectability obtained from three-dimensional studies. Where the dispersion study performed has also contributed to verifying the present investigation in the long wavelength limits. This study can reveal the various optimized locations for placing the sensor for ice detection and quantification, which can be further helpful for practical guided wave inspection in ice detection and its removal.
{"title":"Scattering Analysis of Glaze Ice Accretion on CFRP Laminated Composite Plate Structures Using Ultrasonic Lamb Waves: Towards Aviation Safety","authors":"Saurabh Gupta, Siddesh Sutrave","doi":"10.1007/s10921-023-01030-z","DOIUrl":"10.1007/s10921-023-01030-z","url":null,"abstract":"<div><p>The ice formation over the aerofoil structure of the aircraft wing has been an obstruction as they abrupt the airflow, acting as drag. The investigation will intend to determine ice accumulation on carbon fiber-reinforced polymer (CFRP), approximated as ice build-up on aircraft wings. The observation is carried out over quasi-isotropic composite laminates using ultrasonic-guided waves with a central working frequency regime of 100 kHz. The three-dimensional (3D) finite element (FE) simulations are performed to observe the scattering effect to explore the reflection site in the far field. This effect was quite prominent for different thicknesses of Glaze ice (G-Ice) and was found to be strongly linked with the wave propagation and dispersion effect. The scattering results for the reflection of Lamb mode, when it interacted with the G-Ice interface, were quite noteworthy along the angular region rather than on the center line, indicating that the scattering was more prominent due to the presence of a 45° or (− 45)-degree fiber orientation in that laminate. A similar but complex scattering phenomenon was observed for different stacking sequences where the wave propagation angle and its amplitude at the receiver nodes are found to be closely bound with the exponential decay in group/phase velocity for the ice thicknesses studied. The FE approach is verified, and the results are validated analytically. Analytically, we have investigated a much-closed approximation with the detectability obtained from three-dimensional studies. Where the dispersion study performed has also contributed to verifying the present investigation in the long wavelength limits. This study can reveal the various optimized locations for placing the sensor for ice detection and quantification, which can be further helpful for practical guided wave inspection in ice detection and its removal.</p></div>","PeriodicalId":655,"journal":{"name":"Journal of Nondestructive Evaluation","volume":null,"pages":null},"PeriodicalIF":2.8,"publicationDate":"2023-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138678747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}