Journal of Nondestructive Evaluation最新文献

The new brake disc was evaluated for microstructure and hardness by the conventional destructive tests and non-destructive Barkhausen noise method (BNM). Ten non-destructive measurements were carried out in different areas of a brake disc, which were then cut out and made into metallographic test samples. Qualitative and quantitative analysis of graphite precipitates was performed to assess their volume in material matrix, anisotropy and size. Subsequently, graphs showing the relationships between selected stereological parameters of graphite precipitates and parameters determined from the RMS envelope of Barkhausen noise were elucidated. Similar relationships between hardness and parameters coming from non-destructive tests were carried out. Magnetic parameters that specified the size of a graphite precipitate was selected. In addition, repeatability studies using BNM were carried out in the areas of the material with the smallest and largest average size of graphite precipitates. A linear relationship between amplitude of BN and length of graphite flakes was found. The paper presents the possibilities of assessing the volume and size of graphite precipitates, as well as cast iron hardness using BNM.

Neutron imaging technology is a novel non-destructive testing technique that combines nuclear technology with digital imaging technology. Neutron radiation has significant advantages in detecting light elements and isotopes, making it complementary to X-ray imaging. This paper focuses on lithium-ion batteries and addresses the high level of speckle noise and the low brightness and clarity of neutron projection images. To improve the image quality of neutron projection images, this study proposes methods for noise suppression and image enhancement. Firstly, the median filtering algorithm is utilized to remove speckle noise in the image, and then the gradient operator is applied to sharpen the image and reduce the blurring effect caused by the filtering algorithm. In terms of image enhancement, the quality of the image is improved from two aspects: brightness adjustment and edge sharpening, aiming to enhance image details and improve image contrast. This study tests the algorithm using real neutron projection images and compares it with seven typical image processing algorithms, using peak signal-to-noise ratio, image feature similarity index, average gradient, and no-reference structural clarity as evaluation indicators for image quality. The experimental results show that the proposed method can effectively remove speckle noise in neutron projection images of lithium batteries, significantly improve image clarity and contrast. Compared with the comparative methods, the proposed method has the best edge-preserving ability, the highest signal-to-noise ratio, and clearer image details. In addition, testing with neutron projection images of three non-lithium battery samples demonstrates the good universality of the proposed method in enhancing neutron projection images.

Eddy current testing is one of the conventional non-destructive testing (NDT) technologies which is widely used in metal defects detection. Defect imaging by eddy current tomography (ECT) has advantages of visualization of defects, large detection area, fast detection speed and avoiding mechanical scanning imaging error. Sensitivity matrix is crucial in reconstructing defect images of metal materials by ECT. This article presents a sensitivity matrix of high conductivity initial estimate for ECT detecting metal materials. A 4(times )4 eddy current planar coil array and a 2 mm thickness titanium plate with defects were designed by both simulation and experiment. Based on the proposed sensitivity matrix, reliability of ECT forward problem linearization was analyzed and image reconstruction with two typical regularization methods ((L_1) and (L_2)) were investigated. Both simulation and experiment results show that ECT forward problem linearization was more accurate and reliable with the proposed sensitivity matrix especially at higher frequency. And (L_1) regularization method was verified to be more suitable to reconstruct image of small defects in metal materials. This work expands the original assumption of ECT forward problem linearization, which is of great significance to improve the metal defect image accuracy of ECT.

Acoustic emission (AE) technology is an effective method for monitoring the quality of carbon fiber reinforced plastic (CFRP) laser cutting. It is a challenge to decipher the formation mechanism because of the heterogeneity and anisotropy of CFRP. By analyzing the characteristics of AE signals and the images captured by the high-speed camera under different parameters, four types of AE signal sources were found: high-strength carbon fiber fractures, uniform resin ablation, photochemical fractures, and plasma plume impact. Kerf depth can be accurately identified by measuring the amplitude of the AE signal, and the experimental results verify that the error between the predicted and the experiment values is less than 0.1 mm. The laser defocusing amount, the width of the Heat-affected zone (HAZ), and the overall cutting time can be accurately determined by measuring the Root Mean Square (RMS) of the AE signal. The focus position can be adjusted before the laser cutting, decreasing the width of HAZ from 211.3 μm to 131.5 μm and the time of through-hole cutting from 245 s to 207 s.

X-ray inspection of ball grid arrays (BGAs) is typically performed at one or more viewing angles to examine adhesion sites for errors such as voids, joint cracking, or head-in-pillow. During this inspection process, the circuit board assembly is subject to ionizing radiation exposure, which can cause trapped charge within oxide layers of semiconductor devices. Some x-ray machines allow for programmable inspection routines, which could be used to optimize radiation exposure to semiconductor components. Using Monte Carlo methods, x-ray inspection of a BGA was simulated to determine a range of acceptable viewing angles. Dose rates to circuit board components were estimated at each inspection angle to determine the view resulting in optimized radiation exposure. Results showed that for each BGA, the maximum unobstructed viewing times without exceeding a 5 Gy dose limit to a single part ranged from 82 to 94 min. Using a radiation cost function method, optimized viewing across all components was found. It was observed that for a consistent dose limit applied to silicon-based components, performing inspection with BGAs facing the x-ray source was optimal. A third method was applied, assigning individual dose limits based on empirical data from the NASA Goddard Space Flight Center radiation database. This method showed that optimized viewing maximizes the distance between the radiation source and highly sensitive components. It was also observed that cumulative effects from viewing two BGAs will influence viewing angles, causing the optimal view of one BGA to exist nearly 180(^circ ) from the other.

Composite plates are susceptible to various damages in complex conditions and working environments, which may reduce the reliability of the structure and threaten equipment and personal safety. Thus, the implementation of a robust online Structural health monitoring (SHM) system for these composite structures becomes imperative. To enhance reliability and safety, we introduce a robust online SHM system anchored by our newly developed damage detection Bayesian neural network (DD-BNN). The main contribution of this study lies in the DD-BNN to perform precise and reliable damage detection and localization in composite plates using only one actuator-receiver pair without any signal/feature pre-processing and human intervention. The proposed DD-BNN model innovatively combines probabilistic modeling with deep learning to address uncertainty in Lamb wave-based damage detection and model performance for composite plates, featuring a specialized probabilistic layer trained through Bayesian inference to efficiently encapsulate and manage uncertainty in model weights and activation. Notably, our method significantly simplifies the SHM system design and manual operation requirements. In addition, this approach not only reduces overfitting but also enhances robustness to noise, as confirmed by experiments on perturbation analysis of Gaussian and Poisson noise.

A weak circumferential resolution of defects in the corner part of engineering components brings great challenges to quantitative non-destructive testing. Especially for the corner of carbon fiber reinforced plastics (CFRP), the complex wave propagation behaviors caused by the elastic anisotropy, laminate structure, and curved surface make the information of defects hard to be distinguished, which finally results in a poor imaging resolution. The surface adaptive ultrasonic (SAUL) method for CFRP corner is investigated, and an improved strategy, focusing in receiving (FiR) of SAUL signals is proposed here. With an isotropic plexiglass as a comparison, the effectiveness of FiR is verified by finite element simulations and experiments. The elastic properties of CFRP corner are accurately characterized and a finite element model is established. On this basis, the wave propagation behavior in the corner is studied, and the influence of the water distance h on the maximum amplitude (MAD) and signal-to-noise ratio (SNR) at the defect is analyzed. The results show that the structural noise can be eliminated, and the imaging quality and SNR can be improved by optimizing the h. After FiR, the maximum increase of defect amplitude is about 9.5 dB and 13.2 dB for plexiglass and CFRP, respectively. Meanwhile, the maximum relative error in length is reduced by 16.7% in plexiglass, and by 13.4% for the 3-mm delamination in CFRP. The strategy would be promising to improve the detection quality of the corner in curved components.

In the detection of corrosion under insulation (CUI) using pulsed eddy current testing (PECT) method, it is of great significance to reduce the footprint of the probe for improving the spatial resolution to local corrosion. This paper presents a novel method to reduce the probe footprint by modifying the excitation coil into multiple sub-coils and driving them with sequential pulses of different delay time. Finite element simulations are conducted to reveal the underlying mechanism. It is found that by using the sequential excitation scheme, the diffusion and decay of eddy currents in the test piece are regulated, and both the footprint reduction and signal enhancement can be achieved. Afterwards, the effects of the sequence and the delay amount of the applying pulses on the probe footprint are analyzed. Results show that the optimal excitation sequence is to apply pulses with increasing delay time to the sub-coils from outside to inside; the probe footprint decreases with the increase of the delay amount. Experimental work is finally performed to verify the simulation results. A graphical method for measuring the probe footprint is proposed by moving the probe on a step wedge plate and plotting the evaluated thickness against the probe position. Footprint measurement results of a conventional probe and the presented 4-subcoil probe are compared. The effectiveness of the proposed method are validated and the differences between experimental and simulation results are analyzed.

The magnetic flux leakage (MFL) technique is widely employed for nondestructive testing of ferromagnetic specimens and materials, including wire ropes, bridge cables, and pipelines. As regards the MFL testing, extracting features from MFL signals is crucial for defect recognition and estimation of corresponding widths. Deep learning has been extensively used for feature extraction, but it often performs inadequately on a small sample dataset. To address this limitation, this paper develops a network framework that combines the Wavelet Scattering Transform (WST) and Neural Networks (NN) for defect width estimation. The WST is a knowledge-based feature extraction technique with a structure similar to convolutional neural networks. It offers a translation-invariant representation of signal features using a redundant dictionary of wavelets. The NN then maps the WST feature representation to the defect width information. Experiments on real steel plates with defects are carried out to validate the effectiveness of the proposed framework. Quantitative comparisons of the experimental results demonstrate that the proposed framework achieves better estimation performance in handling MFL signals and has superiority in scenarios with limited training samples.