Ultrasonic signal enhancement for coarse grain materials by machine learning analysis

Analysis of the Suitability of Ultrasonic Testing for Verification of Nonuniform Welded Joints of Austenitic-Ferritic Sheets

The purpose of the presented research was to determine the suitability of using ultrasonic testing (UT) to inspect heterogeneous, from a material point of view, welded joints on the example of the joints of a ferritic steel element with elements made of an austenitic steel. The echo technique with transverse (SEK) and longitudinal wave heads (SEL) addressed this issue. Due to the widespread use of 13CrMo4-5 and X2CrNiMo17-12-2 steel grades in the energy industry, they were selected as the test materials for the study. The objects of the presented research were welded joint specimens with thicknesses of 8, 12, and 16 mm and dimensions of 300 × 300 mm, made using the 135 metal active gas (MAG) process with the use of the Lincoln 309LSi wire-a ferritic-austenitic filler material. The stages of the research task were (1) making distance-amplitude curve (DAC) patterns from the test materials; (2) preparation of specimens of welded joints with artificial discontinuities in the form of through-holes; (3) performing UT tests on welded joints with artificial discontinuities using heads with 60° and 70° angles for the transverse wave and angle heads for longitudinal waves with similar beam insertion angles; (4) selection, by radiographic testing (RT), of welded joint specimens with natural discontinuities in the form of a lack of sidewall fusion; (5) performing UT tests on welded joints with natural discontinuities, using heads as welded joints with artificial discontinuities. It was found that (1) the highest sensitivity of discontinuity detection was obtained by performing tests on the ferritic steel side, which is due to the lower attenuation of the ultrasonic wave propagating in ferritic steel compared to austenitic steel; (2) the best detection of discontinuities could be obtained using a longitudinal ultrasonic wave; (3) there is a relationship between the thickness of the welded elements, the angle of the ultrasonic beam introduction, and the effectiveness of discontinuity detection.

Enhancing the Performance of a Large Aperture Ultrasound System (LAUS): A Combined Approach of Simulation and Measurement for Transmitter-Receiver Optimization

The Large Aperture Ultrasound System (LAUS) developed at BAM is known for its ability to penetrate thick objects, especially concrete structures commonly used in nuclear waste storage and other applications in civil engineering. Although the current system effectively penetrates up to ~9 m, further optimization is imperative to enhance the safety and integrity of disposal structures for radioactive or toxic waste. This study focuses on enhancing the system's efficiency by optimizing the transducer spacing, ensuring that resolution is not compromised. An array of twelve horizontal shear wave transducers was used to find a balance between penetration depth and resolution. Systematic adjustments of the spacing between transmitter and receiver units were undertaken based on target depth ranges of known reflectors at depth ranges from 5 m to 10 m. The trade-offs between resolution and artifact generation were meticulously assessed. This comprehensive study employs a dual approach using both simulations and measurements to investigate the performance of transducer units spaced at 10 cm, 20 cm, 30 cm, and 40 cm. We found that for depths up to 5 m, a spacing of 10 cm for LAUS transducer units provided the best resolution as confirmed by both simulations and measurements. This optimal distance is particularly effective in achieving clear reflections and a satisfactory signal-to-noise ratio (SNR) in imaging scenarios with materials such as thick concrete structures. However, when targeting depths greater than 10 m, we recommend increasing the distance between the transducers to 20 cm. This increased spacing improves the SNR in comparison to other spacings, as seen in the simulation of a 10 m deep backwall. Our results emphasize the critical role of transducer spacing in achieving the desired SNR and resolution, especially in the context of depth imaging requirements for LAUS applications. In addition to the transducer spacing, different distances between individual sets of measurement positions were tested. Overall, keeping the minimal possible distance between measurement position offsets provides the best imaging results at greater depths. The proposed optimizations for the LAUS in this study are primarily relevant to applications on massive nuclear structures for nuclear waste management. This research highlights the need for better LAUS efficiency in applications such as sealing structures, laying the foundation for future technological advances in this field.

Modeling ultrasonic wave fields scattered by flaws using a quasi-Monte Carlo method: Theoretical method and experimental verification

The modeling and visualization of wave fields scattered by flaws can be helpful in terms of guiding the testing and evaluation of flaws using an ultrasonic nondestructive method. In this work, the ultrasonic scattering of wave fields from flaws with different shapes is modeled using a quasi-Monte Carlo (QMC) method and measured through experiments for verification. The incident wave fields generated by a transducer can be modeled using the Rayleigh integral expression and calculated using the QMC method. When the size of the flaw is much larger than the wavelength, the incident wave over the lit portion of flaw can be treated as the source for the scattering of wave fields, and these wave fields can also be modeled using the proposed QMC method. In this paper, water is treated as the material and an embedded solid component is considered as the flaw. Numerical examples and results are presented for flaws with different shapes and sizes, and the properties of these scattering wave fields are analyzed and discussed. Experiments are performed to measure the scattering wave fields using a needle transducer, and it is shown that the results agree with the simulations, thus verifying the proposed modeling method. The work presented here can assist in understanding the wave-flaw interaction and can help in optimizing ultrasonic nondestructive testing.

Machine learning for ultrasonic nondestructive examination of welding defects: A systematic review

Recent years have seen a substantial increase in the application of machine learning (ML) for automated analysis of nondestructive examination (NDE) data. One of the applications of interest is the use of ML for the analysis of data from in-service inspection of welds in nuclear power and other industries. These types of inspections are performed in accordance with criteria described in the ASME Boiler and Pressure Vessel Code and require the use of reliable NDE techniques. The rapid growth in ML methods and the diversity of possible approaches indicate a need to assess the current capabilities of ML and automated data analysis for NDE and identify any gaps or shortcomings in current ML technologies as applied to the automated analysis of NDE data. In particular, there is a need to determine the impact of ML on the NDE reliability. This paper discusses the findings from a literature survey on the current state of ML for the automated analysis of data from ultrasonic NDE of weld flaws. It discusses an overview of ultrasonic NDE as used for weld inspections in nuclear power and other industries. Data sets and ML models used in the literature are summarized, along with a generally applicable workflow for ML. Findings on the capabilities, limitations and potential gaps in feature selection, data selection, and ML model optimization are discussed. The paper identified several needs for quantifying and validating the performance of ML methods for ultrasonic NDE, including the need for common data sets.

Ultrasonic testing of thick and thin Inconel 625 alloys manufactured by laser powder bed fusion

Additive manufacturing of alloys enables low-volume production of functional metallic components with complex geometries. Ultrasonic testing can ensure the quality of these components and detect typical defects generated during laser powder bed fusion (LPBF). However, it is difficult to find a single ultrasonic inspection technique that can detect defects in the large variety of geometries generated using LPBF. In this work, phased array ultrasonic testing (PAUT) is suggested to inspect thick LPBF components, while guided waves are explored for thin curved ones. PAUT is used to detect cylindrical lack of fusion defects in thick LPBF rectangular parts. Practical defects are generated by reducing the laser power at prespecified locations in the samples. The defects' shape and density are verified using optical microscopy and X-ray computed tomography. Partially fused defects down to 0.25 mm in diameter are experimentally detected using a 10 MHz PAUT probe with the total focusing method post-processing. The experimental results are compared to defect images predicted by finite element simulations. For thin components with curved geometry, guided waves are used to detect powder-filled cylindrical defects. The waves are generated using piezoelectric transducers, and the spatiotemporal wavefield is measured using a scanning laser Doppler vibrometer. Using root-mean-square imaging of the wavefield, defects down to 1 mm are clearly detected despite the complex internal features in the samples.

A review of ultrasonic sensing and machine learning methods to monitor industrial processes

Supervised machine learning techniques are increasingly being combined with ultrasonic sensor measurements owing to their strong performance. These techniques also offer advantages over calibration procedures of more complex fitting, improved generalisation, reduced development time, ability for continuous retraining, and the correlation of sensor data to important process information. However, their implementation requires expertise to extract and select appropriate features from the sensor measurements as model inputs, select the type of machine learning algorithm to use, and find a suitable set of model hyperparameters. The aim of this article is to facilitate implementation of machine learning techniques in combination with ultrasonic measurements for in-line and on-line monitoring of industrial processes and other similar applications. The article first reviews the use of ultrasonic sensors for monitoring processes, before reviewing the combination of ultrasonic measurements and machine learning. We include literature from other sectors such as structural health monitoring. This review covers feature extraction, feature selection, algorithm choice, hyperparameter selection, data augmentation, domain adaptation, semi-supervised learning and machine learning interpretability. Finally, recommendations for applying machine learning to the reviewed processes are made.