Pulsed thermography is an indispensable tool in the field of non-destructive evaluation. However, the data generated by this technique can be challenging to analyze and require expertise to ...interpret. With the rapid progress in deep learning, image segmentation has become a well-established area of research. This has motivated efforts to apply deep learning methods to non-destructive evaluation data processing, including pulsed thermography. Despite this trend, there has been a lack of public pulsed thermography datasets available for the evaluation of various spatial-temporal deep learning models for segmentation tasks. This paper aims to address this gap by presenting the PVC-Infrared dataset for deep learning. In addition, we evaluated the performance of popular deep-learning-based instance segmentation models on this dataset. Furthermore, we examined the effect of the number of frames and data transformations on the performance of these models. The results of this study suggest that appropriate preprocessing techniques can significantly reduce the size of the data while maintaining the performance of deep learning models, thereby speeding up the data processing process. This highlights the potential for using deep learning methods to make non-destructive evaluation data analysis more efficient and accessible to a wider range of practitioners.
The success of deep neural networks in carrying out a wide variety of cognitive tasks also raised expectations regarding the advent of AI for the ultrasonic testing (UT) data interpretation in the ...Non-destructive evaluation (NDE) field. Though it is a growing area of research, we identify two main barriers that hinder research in the field: the lack of real-world, annotated datasets accessible to the public and the scarceness of benchmarked performance of the state-of-the-art deep learning models. To address these issues, we first introduce a new dataset called "USimgAIST" which contains more than 7000 real ultrasonic inspection images with both normal cases and defective ones from 17 types of flaws 1. Using the dataset, we performed a comprehensive evaluation of representative deep learning models. Through the study, we expect to validate whether existing AI models can achieve human-level ultrasonic image understanding for defect characterization. Besides, all detailed benchmarking comparisons, including defect detection accuracy, model complexity, memory usage, and inference time, are shown. We hope this study exhibits an overview of performances of advanced learning models working for ultrasonic image analysis and lays the groundwork for prospective practitioners to compare their methods and results fairly.
Pulse or flash thermography is a method of nondestructive evaluation that finds subsurface flaws in materials by observing a heat pulse and subsequent cooldown using a thermal camera. A fundamental ...constraint of pulse thermography is lateral heat diffusion that tends to blur the shapes of defects. It can be difficult to interpret the thermal image sequence from a pulse thermography test. This paper presents a model-based inversion for pulse thermography that uses the known physics of heat conduction to as a basis for representing the recorded thermal image sequence. The technique provides a means to solve for the reflectivity distribution of defects across multiple layers, such as delaminations in a composite material. The layer reflectivity distributions provide a compact and concrete representation of the thermal image sequence. The technique gives excellent interpretability and resolution with minimal noise gain. Model-based inversion is demonstrated on several carbon fiber reinforced plastic (CFRP) specimens.
Acoustic Emission (AE) is a non-destructive evaluation method that uses transient elastic waves produced by the sudden release of mechanical energy in a material or structure. This method generates ...multiple AE events during testing; therefore, it is important to develop parameters that capture the characteristics of each event (AE hit). The paper introduces new dimensionless parameters to characterize the waveform of AE signals: Earliness, Transitoriness, and Early Transitoriness. The study shows that these parameters provide an accurate description of AE waveforms, in some respects, better than traditional parameters, which makes them suitable for filtering with simple rules or in combination with machine-learning techniques. Two examples of the application of AE hit filtering from sedimentation and soil compression experiments are provided. In the sedimentation test analysis, the proposed parameters were used with K-means clustering to filter AE hits from outside the zone of interest and to calculate the rate of sedimentation. In the compression test of a sand sample under oedometric conditions, a simple filtering rule was applied to discriminate AE hits from unwanted sources and obtain a clear AE energy cumulative curve. In both cases, the dimensionless parameters have shown the capacity to discriminate between different AE sources and paths and the possibility of filtering hits from unwanted sources.
Active thermography has proven successful for non-destructive evaluation of art objects. The aim of this work was to detect the defects of a genuine ancient marquetry sample with both enhanced ...truncated correlation photothermal coherence tomography (eTC-PCT) and the well-known lock-in thermography (LIT) and compare the two modalities. Quantitative analysis of eTC-PCT images for depth measurements was also performed. Both methods revealed worn-off and glue starved areas, cracks and detachments in the sample, all of which are common defects of these types of art objects. In comparison, the eTC-PCT method was proven to be able to provide 3D images of the interrogated sample and more information about deep sub-surface defects with better contrast than LIT. Moreover, the eTC-PCT results of the marquetry sample containing natural defects were analyzed to determine the location of defects, a task that is not possible with LIT. The experiments showed that both methods are safe for non-destructive evaluation of art samples with temperature rise of less than 0.8°C using eTC-PCT, and less than 1.5°C using the LIT method.
The objective of this study is to evaluate the effects of sheet temper, forming conditions, and the resultant post-braze microstructure on the corrosion behavior of Al alloy heat exchangers. ...Accelerated corrosion tests were performed by flowing a corrosive water solution through brazed heat exchanger samples fabricated from O and H24 temper sheet formed between room temperature and 250 °C. Sheet perforations within the brazed samples were found to be concentrated at the coolant channel inlets, due to crevice corrosion, or around other formed features within the coolant channel, which was attributed to erosion corrosion. A clear relationship between corrosion initiation sites and local microstructure was not observed in the accelerated test. Potentiodynamic polarization measurements were performed using the corrosive water solution and sheet samples pre-strained to 4 and 10% and subjected to a simulated brazing heat treatment. The average corrosion current density was the same for the two conditions, but liquid film migration occurred during brazing in the sample strained to 4%, resulting in localized corrosion and greater variability in the polarization measurements. It is concluded that warm forming does not impair the corrosion resistance of H24 temper sheet, but the corrosion resistance of brazed samples fabricated from O temper sheet is altered by warm forming.
Coal-fired power plants operating under Korea's standard supercritical pressure operate in a high-temperature environment, with steam temperatures reaching 540 °C. A standard coal-fired power plant ...has a 30-year design life, and lifespan diagnosis is performed on facilities that have operated for more than 100,000 h or 20 years. Visual inspection, thickness measurements, and hardness measurements in the field are used to assess the degree of material degradation at the time of diagnosis. In this study, aging degradation was assessed using an electromagnetic acoustic transducer to measure the change in transverse ultrasonic propagation speed, and the results were compared to microstructural analysis and tensile test results. Based on the experimental results, it was found that the boiler tube exposed to a high-temperature environment during long-term boiler operation was degraded and damaged, the ultrasonic wave velocity was reduced, and the microstructural grains were coarsened. It was also confirmed through tensile testing that the tensile and yield strengths increased with degradation. Our findings prove that the degree of change in mechanical properties as a function of the material's degradation state is proportional to the change in ultrasonic wave velocity.
This paper investigates characteristics and reliability of SiC power devices used for a start-up method based on multiple short-circuits. A planar-gate and trench-gate SiC-MOSFETs and SiC-JFET ...employed in a practical power converter are tested under the start-up method using multiple short-circuits more than 10,000 cycles. The experimental results exhibit only the planar-gate SiC-MOSFET can survive the 10,000 cycle short-circuit tests with slight degradation under a testing condition. As a result, this paper reveals the planar-gate SiC-MOSFET is suitable for the start-up method due to its characteristic.
Ultrasound was considered as a means for determining mechanical properties of clear wood in six different Acacia mangium provenances from a trial forest planted in Vietnam. A total of 30 trees (5 ...trees from each provenance) with no major defects were selected, and a 50-cm-long log was obtained at 1.3 m above the ground from each tree for the assessment of mechanical properties. The measured average ultrasound velocities for provenances tested in the longitudinal direction ranged from 4094 m/s to 4271 m/s. The predicted average dynamic modulus of elasticity (Ed) values varied from 7.42 GPa to 8.70 GPa among provenances. The Ed indicated significant positive correlation coefficients with modulus of elasticity (0.64 to 0.96), modulus of rupture (0.44 to 0.87), and compression strength (0.54 to 0.92) for provenances examined in this study. The results indicated that the use of ultrasound was feasible to determine the mechanical properties of A. mangium provenances planted in Vietnam.
PDF
