Journal of Nondestructive Evaluation最新文献

This study presents a quantitative analysis of CeO₂ and TiB₂ non-metallic particles within the microstructure of additively manufactured (AM) Ni-superalloys Inconel 718 (IN718), using microfocus X-ray computed tomography (micro-XCT) and a volumetric analysis tool, CGAL VESPA Alpha Wrapping. Focusing on the characterization of CeO₂ and TiB₂ particles embedded within IN718, this method highlights their size and volume fraction variations as well as distinct spatial distributions, which are quantitatively compared to metallographically prepared SEM samples. Quantitative assessments conducted with Paraview served as the basis for optimizing alpha and offset parameters for surface construction. This optimized data processing routine yields volume and surface morphology estimations that more closely align with those obtained from SEM observations, compared to the traditional Marching Cubes algorithm, assuming identical preprocessing and binarization standards. The flexibility to adjust the wrapping parameters also allows for precise control over volumetric and surface area estimations. The results demonstrated that CGAL VESPA Alpha Wrapping, implemented in Paraview for object identification, enables simultaneous evaluation of particle morphology and authentic volumetric information from the same micro-XCT data, particularly for non-uniformly distributed reinforcement particles. This capability supports a more reliable non-destructive evaluation for AM components.

Acoustic Emission (AE) is a well-established and recognised technique for monitoring the degradation of a variety of structures. It is used in a variety of applications, including fatigue monitoring, corrosion monitoring, or detection of pressure leaks. As sensors evolve and databases grow, analysis allows for a better interpretation and understanding of phenomena. Specifically, the usage of Machine Learning (ML) algorithms has proven to be a major tool for interpreting signals. This paper reviews the current usage of ML algorithms used in major Acoustic Emission applications to interpret damage mechanisms, exploring how ML allows the study of more complex phenomena and structures, discussing the conditions, precautions and limitations to its usage as well as future prospects and potentials.

The probability of detection (POD) is a fundamental metric for evaluating the performance of nondestructive evaluation (NDE) techniques. However, traditional empirical approaches to POD estimation often require extensive measurements, making them costly in terms of time, budget, and resources. In scenarios with limited data, conventional estimation methods frequently fail to capture the underlying relationship between signal responses and flaw sizes, as well as the variability introduced by testing conditions, influencing factors, and inherent uncertainties. Moreover, standard linear regression models, commonly used in POD analysis, rely on assumptions that are often violated when sample sizes are small, resulting in biased or imprecise estimates. To overcome these challenges, this study investigates advanced regression techniques and their integration with physics-based models for limited-sample POD (LS-POD) estimation. LS-POD here is defined as POD estimation when the sample size is below the threshold typically required by conventional methods. We explore a range of information-augmentation approaches, including physics-informed regression and Bayesian methods, which incorporate prior knowledge to improve the characterization of the signal-flaw relationship and the variability of NDE procedures. Additionally, we adapt advanced statistical techniques, such as Box-Cox transformation, robust regression, weighted linear regression, and bootstrapping, to mitigate the impact of assumption violations commonly encountered in small-sample contexts. These methods are further integrated to simultaneously leverage existing knowledge and address statistical assumption violations. We conduct comprehensive simulation studies using both synthetic and empirical datasets to evaluate the performance of these approaches under a variety of LS-POD scenarios. The results are benchmarked against conventional POD estimates derived from large-sample data. Our findings indicate that incorporating prior knowledge and employing assumption-resilient regression techniques can significantly enhance the accuracy and precision of LS-POD estimation. The combined use of information-augmentation and assumption-correction strategies yields further improvements. These results provide practical insights for NDE practitioners, facilitating the selection and application of appropriate LS-POD methods tailored to specific data conditions and application needs.

In the production process of additive manufacturing (AM) components, the occurrence of holes, microcracks, and other defects can seriously affect the physical and mechanical properties of AM components. This paper presents an effective method for quality evaluation of AM components utilizing zero-group-velocity (ZGV) Lamb waves. The displacement distribution and propagation characteristics of the S1-ZGV mode in the AM component are analyzed in detail by the finite element (FE) method, and the changes in the S1-ZGV mode under different quality levels (characterized by different Young’s moduli) are investigated. The results indicate that the S1-ZGV mode in the AM component is distributed in the form of standing waves, whose time-domain waveform persists throughout the entire time-domain. As the level of quality deteriorates, a corresponding reduction is observed in both the frequency and spectral amplitude (SA) of the S1-ZGV mode, and notably, the SA at the initial S1-ZGV frequency (in good material condition) significantly decreases. This observation provides a reliable method for conducting effective quality evaluation of AM components. Subsequently, the S1-ZGV mode is experimentally and successfully excited in the AM component using the pitch-catch technique with air-coupled ultrasonic transducers, and the SA at different detected positions is quantitatively observed to validate the effectiveness of the method. The experimental results reveal that compared to the traditional linear ultrasonic technique based on wave velocity measurement, the SA at the initial S1-ZGV frequency can more effectively evaluate the quality level of the AM component, which are verified by the optical microscope images. These results validate the effectiveness of the SA based on ZGV modes in accurately evaluating the quality level of the AM components.

Ultrasonic testing is a widely employed non-destructive technique for material characterization and defect detection. For pervious concrete (PeC), a porous composite material made of cement paste and coarse aggregate, understanding the interaction between material properties and ultrasonic wave propagation remains a challenge. This study implements a three-dimensional finite element model to simulate acoustic wave behavior in PeC, focusing on the effects of porosity P, aggregate size D, elastic modulus E, and density (rho ). The specific goal is to understand the relationship of ultrasonic wave velocity and porosity in PeC. To control porosity, the model is based on a simplified hypothetical contact between particles which may represent the cement paste surrounding the aggregate particles. Several families of models are built by varying porosity between 8% and 40%, and three different values of D, E and (rho ). An analytical model –an equation– is proposed and successfully fitted to the numerical data, and then tested numerically; the equation consists of a theoretical P-wave velocity multiplied by a factor dependent of D and P. Numerical results are partially validated against experimental measurements obtained from PeC samples with porosity values ranging from 14% to 35%. The findings reveal a clear inverse relationship between porosity and ultrasonic wave velocity, emphasizing the influence of aggregate contact areas. This work establishes a foundation for advancing ultrasonic testing as a reliable tool for assessing PeC porosity and performance in field applications.

This study proposes a magnetic flux leakage inspection capable of identifying internal and external defects in rotating pipe inspections. The proposed identification between internal and external defects employs the effect of motion-induced eddy current that has been an adverse effect on the conventional magnetic flux leakage testing. A three-dimensional finite element analysis was conducted to assess the feasibility of detecting and classifying these defects. Two hall sensors, symmetrically positioned from the pole structure, exhibit asymmetric defect signals with inverse signal variations for the internal and external defects. Simulation studies were performed to investigate the effect of flux density and rotational speed on defect signals. A prototype sensor was fabricated, and the measurement shows peak-to-peak variations as − 43.1% for internal defects and + 25.7% for external defects, indicating a strong correlation with the simulation results. These findings suggest that the proposed inspection can represent an effective alternative to the conventional ultrasonic testing for monitoring pipe integrity at the pipe production stage.

Laser ultrasonic (LU) testing has attracted considerable attention in the fields of material characterization and defect detection due to its non-destructive nature. However, acquiring a complete wavefield using LU typically requires significant time and resources, motivating the development of more efficient sampling strategies. In this study, a novel approach based on Physics-Informed Neural Networks (PINNs) is proposed to reconstruct the full Lamb wavefield from sparsely sampled experimental data. By embedding the governing physical laws of wave propagation into the neural network framework, the PINN model is trained to infer the wavefield characteristics from a limited number of measurements. Notably, the proposed method successfully reconstructs the complete Lamb wavefield with an accuracy of 88% while using only one-sixteenth of the full dataset. The results highlight the potential of PINNs to improve both the efficiency and accuracy of wavefield reconstruction, offering a promising solution to the limitations of conventional LU testing.

The adoption of machine vision to replace manual inspection in X-ray non-destructive testing (NDT) for image defect detection has emerged as a significant trend in the advancement of welding defect detection. In this paper, an enhanced strategy is proposed to address the issue of low detection accuracy of YOLOv8 in X-ray weld defect detection. An extra tiny object detection head is added to the detection head, which enables more accurate capture of extremely small defect features, effectively expanding the lower detection limit and significantly enhancing the detection capability for extremely small weld defects. By employing serpentine deformable convolution, the model dynamically adjusts its receptive field, enabling it to flexibly adapt to variations in crack morphology, thereby improving the detection capability for small objects with special shapes. The integration of an advanced BiFPN structure enables three-level feature fusion, optimizing the detection performance for medium and large objects across multiple scales, and expanding the upper detection range. The results show that the proposed improvement strategy achieves the maximum detection scale while also significantly improving detection accuracy, with the overall mAP@50% reaching 97.2%, an increase of 17.1%. The proposed strategy in this study significantly improves the accuracy of weld defect detection. It also enhances the detection performance for small targets with specific shapes, extremely small defects, and expands the model’s scale adaptability. Validation experiments conducted on the GDXray weld dataset further demonstrate its effectiveness.

Ceramic matrix composites represent a novel type of high-temperature structural material. Defects such as porosity, delamination, and cracking generated during the fabrication process significantly impact the structural uniformity and performance, especially in the case of irregular shapes where this issue becomes more pronounced. Conventional methods that rely on overall grayscale values often fail to quantify structural density differences in irregular components. To address this, we propose a dual-mode uniformity characterization method based on block grayscale difference calculation. Considering the high porosity of ceramic matrix composites and the characteristics of pore defects, the tomography image after pore removal is obtained based on the adaptive threshold algorithm. These images are then partitioned into blocks based on their spatial positions, and the average grayscale values of each block are calculated to achieve a digital representation of the composite material’s uniformity. Furthermore, three-dimensional reconstruction of the average grayscale using volume rendering algorithms provides a visual representation of the structural distribution for intuitive analysis. Test analysis of a U-shaped C_f/SiC specimen yielded maximum grayscale values for blocks of 134.81, minimum grayscale values of 92.24, and grayscale differences between blocks of 42.57, effectively characterizing the digital differences in structural uniformity among various blocks of the specimen. The visualization results of three-dimensional reconstruction and color mapping depict the spatial distribution characteristics of the structural specimen. This method offers a new approach to characterizing the uniformity of irregular-shaped ceramic matrix composites.

To address the precision and adaptability requirements for surface extraction in industrial computed tomography (CT) reverse engineering, we proposes a subvoxel-accuracy surface reconstruction method that integrates surface tracking algorithms with analytical gradient computation. Building upon the Marching Triangles framework, our method introduces an adaptive mesh growth strategy driven by analytical curvature and enhance edge-region extraction through curvature consistency verification. We develop a dual-stage projection mechanism, utilizing gray-value coarse projection in the initial stage followed by second-order gradient refinement. Experimental results demonstrate that compared to traditional Marching Cubes methods, our approach produces higher-quality triangular meshes with reduced vertex counts. When compared with conventional threshold-based algorithms, the proposed method shows superior surface accuracy and significant advantages for industrial metrology CT applications.