Showing papers in "Ndt & E International in 2020"

Performance enhancement of convolutional neural network for ultrasonic flaw classification by adopting autoencoder

Abstract: The industrial application of deep neural networks to automate the ultrasonic weldment flaw classification system has some limitations. The major problem that affects the classification performance of deep neural networks is the noise in the ultrasonic signals. So, in this article, a deep neural network, also known as autoencoder is investigated to remove noise from ultrasonic signals before feeding them to deep learning classifiers. A database was generated from specimens that were closely resembled with pipe weldment geometry having counterbore and weldment defects. Those signals were, later on, corrupted with noise to mimic industrial applicability. An autoencoder was then employed to remove noise from counterbore, planer and volumetric weldment defect signals. The classification performance of the convolutional neural network (CNN) was evaluated in three different ways. At first, without employing the autoencoder, secondly, on the denoised outputs of the autoencoder and on third CNN was trained with the noiseless signals but was tested on the denoised outputs of the autoencoder. The results demonstrate that the autoencoder can successfully remove noise from the ultrasonic weldment defect signals, which consequently improve the defect classification accuracy of the artificially intelligent deep learning classifiers.

An integrated investigative approach in health monitoring of masonry arch bridges using GPR and InSAR technologies

Abstract: This paper provides an overview of the existing health monitoring and assessment methods for masonry arch bridges. In addition, a novel “integrated” holistic non-destructive approach for structural monitoring of bridges using ground-based non-destructive testing (NDT) and the satellite remote sensing techniques is presented. The first part of the paper reports a review of masonry arch bridges and the main issues in terms of structural behaviour and functionality as well as the main assessment methods to identify structural integrity-related issues. A new surveying methodology is proposed based on the integration of multi-source, multi-scale and multi-temporal information collected using the Ground Penetrating Radar (GPR – 200, 600 and 2000 MHz central-frequency antennas) and the Interferometric Synthetic Aperture Radar (InSAR – C-band SAR sensors) techniques. A case study (the “Old Bridge” at Aylesford, Kent, UK – a 13th century bridge) is presented demonstrating the effectiveness of the proposed method in the assessment of masonry arch bridges. GPR has proven essential at providing structural detailing in terms of subsurface geometry of the superstructure as well as the exact positioning of the structural ties. InSAR has identified measures of structural displacements caused by the seasonal variation of the water level in the river and the river bed soil expansions. The above process forms the basis for the “integrated” holistic structural health monitoring approach proposed by this paper.

Integration of InSAR and GPR techniques for monitoring transition areas in railway bridges

Abstract: This paper reports the integration of the Ground Penetrating Radar (GPR) and the Interferometric Synthetic Aperture Radar (InSAR) techniques for the monitoring of the rail-abutment transition area in railway bridges. To this purpose, an experimental campaign was conducted on a rail truss bridge located in Puglia, Southern Italy. On one hand, GPR was used to obtain structural details of the subsurface (thickness of the ballasted layer, position of the sleepers, presence of clay/humidity spots) and to identify potential construction-related issues. Parallel to this, InSAR analyses were mainly addressed to monitor subsidence at the rail-abutment transition area. Outcomes of this investigation outlined presence of subsidence at both the areas of transition and have proven the proposed integrated approach as viable to achieve a more comprehensive assessment of the structural integrity of railway bridges.

UAV photogrammetry, infrared thermography and GPR for enhancing structural and material degradation evaluation of the Roman masonry bridge of Ponte Lucano in Italy

Abstract: This paper presents a case study regarding the cooperative use of non-destructive contactless diagnostic investigations as a tool to aid and assist the degradation evaluation of an iconic Roman masonry bridge: Ponte Lucano in Tivoli, Italy. Specifically, unmanned aerial photogrammetric surveys are considered to perform detailed visual inspections and obtain a geometrical 3D model; infrared thermography analyses are carried out to characterize the thermal surface map of the structure detecting anomalies related to material degradation, such as the presence of humidity; ground penetrating radar investigations are performed to improve knowledge of the bridge subsurface structure. The results of the analyses demonstrate that the integration of mentioned diagnostic tools provides information regarding the degradation state of the stones and its causes, as well as regarding the evolution of the structure from its construction up to the present configurations.

Two-stream convolutional neural network for non-destructive subsurface defect detection via similarity comparison of lock-in thermography signals

Abstract: Active infrared thermography is a safe, fast, and low-cost solution for subsurface defects inspection, providing quality control in many industrial production tasks. In this paper, we explore deep learning-based approaches to analyze lock-in thermography image sequences for non-destructive testing and evaluation (NDT&E) of subsurface defects. Different from most existing Convolutional Neural Network (CNN) models that directly classify individual regions/pixels as defective and non-defective ones, we present a novel two-stream CNN architecture to extract/compare features in a pair of 1D thermal signal sequences for accurate classification/differentiation of defective and non-defective regions. In this manner, we can significantly increase the size of the training data by pairing two individually captured 1D thermal signals, thereby greatly easing the requirement for collecting a large number of thermal sequences of specimens with defects to train deep CNN models. Moreover, we experimentally investigate a number of network alternatives, identifying the optimal fusion scheme/stage for differentiating the thermal behaviors of defective and non-defective regions. Experimental results demonstrate that our proposed method, directly learning how to construct feature representations from a large number of real-captured thermal signal pairs, outperforms the well-established lock-in thermography data processing techniques on specimens made of different materials and at various excitation frequencies.

Features extraction and discussion in a novel frequency-band-selecting pulsed eddy current testing method for the detection of a certain depth range of defects

Abstract: Local wall thinning defects are unavoidable defects in actual engineering structures and in some cases it occurs in a certain depth range of the object structures. In this study, a novel frequency-band-selecting pulsed eddy current testing (FSPECT) has been proposed for the detection of local wall thinning defects in a certain depth range. Feature of peak value has been extracted and analyzed. The results demonstrate that the FSPECT possess the comparable performances in terms of detection sensitivity over the traditional square wave pulsed eddy current testing (PECT). In addition, other fruitful detailed features extraction in the numerical calculation and experimental signals of FSPECT are deeply studied. The features obtained by simulation and experiment mainly include lift-off point of intersection (LOI) and zero-crossing time. Furthermore, the influences of the depth of local wall thinning defect on features of LOI and zero-crossing time have been explored, which enhance the accuracy and reliability of FSPECT method for the evaluation of local wall thinning defects.

An experimental and numerical approach to combine Ground Penetrating Radar and computational modeling for the identification of early cracking in cement concrete pavements

Abstract: Road pavements are subject to traffic and temperature variations producing cracks that propagate to the pavement surface, which reduces its life and decreases circulation comfort. Early identification of pavement cracks allows for suitable maintenance and rehabilitation decreasing life cycle cost and increasing pavement life. For the development of an approach to identify early cracking using Ground Penetrating Radar (GPR), a laboratory simulation of different types of cracking was carried out, using a GPR system equipped with a ground-coupled antenna with a 2.3 GHz central frequency. Additionally, numerical simulations were elaborated using a Finite-Difference Time-Domain (FDTD) method-based software package (gprMax2D). Field GPR measurements were also carried out on a cracked rigid pavement of a road section overlaid with an asphalt layer. The results of this work proved the capability of the GPR method to detect cracks in rigid pavements and allowed to develop an approach for application in the identification of early cracking in rigid pavements.

Measurement of coating thickness using lift-off point of intersection features from pulsed eddy current signals

Abstract: In industries, the thickness measurement of non-ferromagnetic coatings on a ferromagnetic substrate is a difficult problem. In this paper, we proposed a simple but liftoff insensitive method for measuring the thickness of non-ferrous coatings using lift-off point of intersection (LOI) extracted from pulsed eddy current (PEC) signals. Firstly, an analytical model was formulated to calculate PEC signals using Fourier transform. Then, the simulations were carried out to investigate the characteristics of LOI points and the applicability of the presented method for measurement of coating thickness. Subsequently, the critical factors on coating thickness measurement were discussed. The results show that the developed new method is feasible for measurement of nonferrous coating thickness. The rising time of pulse excitations could balance the sensitivity and accuracy. The results demonstrate that the relative error for measurement of coating thickness is 4.9% when the rising time increases to 90 μs

Factors Impacting Monitoring Asphalt Pavement Density by Ground Penetrating Radar

Abstract: Real-time asphalt concrete (AC) pavement density monitoring is important for quality control (QC) and quality assurance (QA) purposes, because it allows correction during the compaction process. Ground penetrating radar (GPR) is capable of providing real-time AC mixture density prediction using the Al-Qadi, Lahouar, and Leng (ALL) density prediction model. However, noise sources, such as surface moisture and vibrations, may jeopardize the AC density prediction accuracy. This study proposes a mean reflection coefficient algorithm and digital filter design method to remove the surface moisture and smooth the density profile. In the mean reflection algorithm, the frequency-select bandwidth was selected as 40–70% of the actual peak frequency in the magnitude spectrum through the simulation studies. White Gaussian noise was added in the models for robustness testing. In the digital filter design method, the magnitude spectrum of the GPR predicted density profile was analyzed to decide filter types and corresponding parameters. Thresholding method was used to remove abnormal values, and window-based finite impulse response (FIR) filters were used to smooth the density profile. Lab-controlled and field tests were performed for both algorithms. Estimated aggregate dielectric constant was used to predict pavement density. A sensitivity analysis was performed to evaluate the effect of different aggregate dielectric constant on density (or air void). For surface moisture effect removal, mean reflection coefficient algorithm may be utilized to reconstruct dielectric constant values at an error less than 4%. This algorithm is independent of the antenna central frequency. For the density profile smoothing during continuous GPR survey, results show that various filter types have comparable smoothing performances. For the effect of aggregate dielectric constant on density prediction, sensitivity analysis shows that when aggregate dielectric constant values changes from 6.5 to 7, the predicted air void increases from 2.5% to 6.3%. This indicates the importance of an accurate aggregate dielectric constant estimate when applying ALL model for pavement density predictions; hence, aggregate dielectric constant estimate must be utilized.

Accurate measurements of plate thickness with variable lift-off using a combined inductive and capacitive sensor

Abstract: In eddy current testing, lift-off variations between sensor and sample plate affect sensor inductance and consequently inferred thickness. In this paper, a novel combined inductive and capacitive sensor is designed in which the output of the capacitive sensor can be used to infer the lift-off. An algorithm is proposed to combine the inferred lift-off and the inductance measurement for predicting thickness. The effect of lift-off variations is significantly reduced using this combined sensor approach. Compared to the multi-frequency thickness measurements in our previous work, this only requires a single frequency measurement for each of the capacitive and inductive modalities, leading to a relaxation of the instrument and measurement requirements.

Integration of non-destructive testing methods to assess asphalt pavement thickness

Abstract: The pavement engineering community has consistently drawn its attention to a broadened utilization of advanced Non-Destructive Testing (NDT) systems for pavement evaluation. Amongst these systems, the Falling Weight Deflectometer (FWD) is designed to measure surface deflections and assess the pavement structural condition, while Ground Penetrating Radar (GPR) is utilized to estimate pavement layers’ thicknesses. An integration of the capabilities of these two separate systems has been considered a powerful resource in pavement evaluation. With this in mind, the present study aims to investigate whether this integration could result in the utilization of the derived FWD deflection data to directly assess pavement layers thicknesses. For the research, an integrated regression analysis is conducted considering both GPR and FWD data with the aim to develop a relationship between GPR-estimated AC thicknesses and FWD deflection indexes. The developed relationship is calibrated considering both GPR and core thicknesses and further validated yielding an equivalent and satisfactory performance, with thickness prediction errors around 10%. This finding produces evidence in support of the statement that deflection data is capable of roughly producing pavement layer thicknesses. Such an approach could suggest a practical and cost-effective pavement evaluation tool, when the procurement and transportation logistics of implementing multiple and expensive equipment may be a critical issue.

Efficient Lamb-wave based damage imaging using multiple sparse Bayesian learning in composite laminates

Abstract: Lamb wave techniques have been widely used for structural health monitoring (SHM) and nondestructive testing (NDT). To deal with dispersive and multimodal problems of Lamb wave signals, many signal processing methods have been developed. A spatially distributed array of piezoelectric transducers is generally adopted for both transmission and reception of Lamb waves. When imaging the damage in composite laminates, it is necessary to meet the need of processing array signals with high efficiency. In this paper, the multiple sparse Bayesian learning (M-SBL) strategy is employed for damage imaging. Multiple residual signals including damage-reflection waves are decomposed into a sparse matrix of location-based components simultaneously. An appropriate dictionary is designed to match the damage-reflection waves instead of interference waves. The key to success is to obtain the sparse matrix of weighting coefficients through the M-SBL algorithm. Damage imaging can be achieved efficiently using the delay-and-sum (DAS) method with sparse coefficients in time-domain. Results from the experiment in composite laminates demonstrate the effectiveness of the proposed method.

Feature extraction of wood-hole defects using empirical mode decomposition of ultrasonic signals

Abstract: Holes and knots are common defects that occur in wood that affect its value for both structural and high-end aesthetic applications. When these defects are internal to wood they are rarely evident from visual inspection. It is therefore important to develop techniques to detect and analyse these defects both in standing trees prior to harvesting them and in processed timber and/or completed wooden structures. This paper presents an effective method to detect and analyse hole defects in wood. The method uses the recorded output wave signal from an ultrasonic device tested on rectangular wood samples. The ultrasonic wave signal is decomposed into its constructive modes using Empirical Mode Decomposition (EMD). This process decomposes a non-stationary non-linear wave signal into its semi-orthogonal bases known as intrinsic mode functions (IMFs). A matrix of all IMFs (except the residual IMF) is then assembled and its covariance matrix derived. The research demonstrates through several experimental studies that the maximum eigenvalue of the proposed covariance matrix is more sensitive to hole defects in wood than traditionally used measures such as time-of-flight. The results provide evidence that the proposed damage sensitive feature (DSF) can successfully detect hole defects in hardwood samples but further work is recommended on its application to other materials. It is anticipated that this method will have wide applicability in the forestry and timber industries for aiding in product value determination.

Improved impact damage characterisation in CFRP samples using the fusion of optical lock-in thermography and optical square-pulse shearography images

Abstract: Image fusion methods with optical lock-in thermography (OLT) and optical square-pulse shearography (OSS) images are proposed to characterise impact damages in carbon fibre reinforced plastic (CFRP) plates. The samples were damaged with low-energy impacts and inspected using OLT, OSS and the reference ultrasound (US) time-of-flight C-scans. A total of 1113 combinations of decomposition, preprocessing, segmentation and fusion tools were proposed and compared with OLT, US and OSS results using the equivalent diameter criterion and the Matthews’ correlation coefficient. The results indicated a reduction of 72.21% in the equivalent diameter measurement error and a metric enhancement of 8.05% when using the fusion, showing that one of the developed image fusion methods can successfully perform improved impact damage inspections.

Nondestructive evaluation of thermal barrier coating thickness degradation using pulsed IR thermography and THz-TDS measurements: A comparative study

Abstract: Thermal barrier coatings are extensively used in aircraft engines. During service, the TBC coatings degrade because of erosion and sintering by hot gas flow and also by localized wear due to rubbing of flaps with spacers. It is necessary to assess the condition of the coatings as a function of service life through suitable non-destructive means. Pulse Thermography (PT) and Terahertz-Time Domain Spectroscopy (THz-TDS) techniques are used to evaluate the degree of degradation of the thin Air Plasma Sprayed (APS) TBCs top coat thickness. Infrared thermography has the advantage of fast inspection of a large area. In this work, we used a simplified analytical model aided calibration and development of a regression model to quantitatively analyze the thickness degradation in real-world TBC samples that have endured varying service life. These measurements were later verified using THz-TDS imaging, an emerging technique for accurate thickness measurements. Assuming the refractive index of the topcoat material, Yttria-Stabilized Zirconia (YSZ) as 4.8, the topcoat thickness of the entire specimen has been estimated using THz-TDS reflection mode setup. Results show that, the thickness values are varying between 94.94 μ m - 114.96 μ m for 500 h serviced samples and 32.5 μ m - 91.96 μ m in the case of 1000 h serviced samples demonstrating loss of TBC with increased service life. Comparison of the pulse thermography results with THz-TDS reveals a mean relative error of less than 10.3 % in TBC thickness estimation. Further the results of both the techniques are cross-validated with Eddy current testing and optical microscopy. The proposed non-destructive techniques for TBC estimation will aid in the accurate creation of engine digital twin and will help in scheduling preventive maintenance measures thus increasing the life and safety of key aircraft engine components.

Super-resolution visualization of subwavelength defects via deep learning-enhanced ultrasonic beamforming: A proof-of-principle study

Abstract: Detecting small, subwavelength defect has known to be a challenging task mainly due to the diffraction limit, according to which the minimum resolvable size is in the order of the wavelength of a propagating wave. In this proof-of-concept study, we present a deep learning-enhanced super-resolution ultrasonic beamforming approach that computationally exceeds the diffraction limit and visualizes subwavelength defects. The proposed super-resolution approach is a novel subwavelength beamforming methodology enabled by a hierarchical deep neural network architecture. The first network (the detection network) globally detects defective regions from an ultrasonic beamforming image. Subsequently, the second network (the super-resolution network) locally resolves subwavelength-scale fine details of the detected defects. We validate the proposed approach using two independent datasets: a bulk wave array dataset generated by numerical simulations and guided wave array dataset generated by laboratory experiments. The results demonstrate that our deep learning super-resolution ultrasonic beamforming approach not only enables visualization of fine structural features of subwavelength defects, but also outperforms the existing widely-accepted super-resolution algorithm (time-reversal MUSIC). We also study key factors of the performance of our approach and discuss its applicability and limitations.

Evaluation of disbonds at various interfaces of adhesively bonded aluminum plates using all-optical excitation and detection of zero-group velocity Lamb waves

Abstract: Inspection of adhesively bonded metallic plates, commonly used in aircraft structures, remains challenging for modern non-destructive testing (NDT) techniques. When a probing ultrasound (US) wave interacts with the plate boundaries, it produces multiple propagating guided waves (or Lamb modes). Analysis of these waves can be complicated due to a strong geometrical dispersion. However, recent studies showed that for specific frequencies zero-group velocity (ZGV) modes exist. Any changes in a bounded structure, like delaminations, drastically alter the conditions for zero-group velocity waves, thus making this method highly sensitive for NDT applications. Laser ultrasound (LU) provides a very broad bandwidth of the generated waves, thus it is a feasible tool for a spectroscopy-based investigation of the ZGV modes. In this paper, we demonstrate an application of high-repetition rate LU with a non-contact fiber-optic Sagnac interferometer on receive for high resolution imaging of delaminations in a structure consisting of 3 epoxy-bonded aluminum plates. The investigation is supported by numerical analysis of Lamb waves existing in the structure to determine ZGV modes sensitive to delaminations at particular bonding interfaces. Tracking the selected modes permits imaging and identification of defects. We also show that mean frequency estimation of the ZGV modes can improve the contrast-to-noise ratio compared to an amplitude of the single ZGV frequency.

Automatic delamination segmentation for bridge deck based on encoder-decoder deep learning through UAV-based thermography

Abstract: Concrete deck delamination often demonstrates strong variations in size, shape, and temperature distribution under the influences of outdoor weather conditions. The strong variations create challenges for pure analytical solutions in infrared image segmentation of delaminated areas. The recently developed supervised deep learning approach demonstrated the potentials in achieving automatic segmentation of RGB images. However, its effectiveness in segmenting thermal images remains under-explored. The main challenge lies in the development of specific models and the generation of a large range of labeled infrared images for training. To address this challenge, a customized deep learning model based on encoder-decoder architecture is proposed to segment the delaminated areas in thermal images at the pixel level. Data augmentation strategies were implemented in creating the training data set to improve the performance of the proposed model. The deep learning generated model was deployed in a real-world project to further evaluate the model's applicability and robustness. The results of these experimental studies supported the effectiveness of the deep learning model in segmenting concrete delamination areas from infrared images. It also suggested that data augmentation is a helpful technique to address the small size issue of training samples. The field test with validation further demonstrated the generalizability of the proposed framework. Limitations of the proposed approach were also briefed at the end of the paper.

Phased array ultrasonic imaging using angle beam virtual source full matrix capture-total focusing method

Abstract: Full matrix capture coupled with total focusing method (FMC-TFM) is one of the most advanced and popular phased array ultrasonic techniques used to achieve simultaneous focusing at all depths in a specimen. In case of conventional FMC-TFM technique, a single element is used for transmission that limits the transmitted energy. This leads to reduced sensitivity at larger depths particularly in attenuating materials such as austenitic alloys. An Angle Beam Virtual Source FMC (ABVSFMC) method is proposed in the present paper to enhance the sensitivity by increasing the transmitted energy with specific directivity. The methodology utilizes apertures of a few elements to obtain a diverging beam by focusing very near to the top surface (virtual source). Further, as a group of elements is used in transmission, a divergent beam with specific directivity can be achieved without using a wedge. In reception, all the available channels are used and an algorithm similar to the Total Focusing Method (TFM) is used to reconstruct the image. The methodology is demonstrated by imaging the tip diffracted signals obtained from slot type planar defects at various depths in the range of 25–175 mm in a 200 mm thick nickel base alloy forging.

Fractal analysis of nonlinear ultrasonic waves in phase-space domain as a quantitative method for damage assessment of concrete structures

Abstract: The present paper applies fractal dimension as a mathematical tool to extract a quantitative geometric feature from nonlinear ultrasonic waves in phase-space domain. The feature is then used for damage assessment of concrete material under different service loads after experiencing extreme compressive loads. For this purpose, concrete samples are loaded in two different scenarios: the first loading part is set to initiate micro cracks and generate macro cracks, while the second loading procedure is set to simulate service loads and change the cracks' boundary conditions. Due to nonlinear ultrasonic waves’ sensitivity to cracks in early stages, nonlinear ultrasound test is performed. In contrast to traditional approaches which analyze nonlinear ultrasound waves in frequency domain, in this paper, nonlinear ultrasonic waves in phase-space domain are studied. Phase-space domain is a powerful mathematical tool that allows researchers to analyze data quantitatively and qualitatively using different signal processing techniques like fractal dimension. For the first time, fractal analysis of nonlinear ultrasound waves in phase-space domain is performed to measure the nonlinearity of the waves due to interactions with loaded-induced cracks in concrete materials. In general, fractal analysis makes it possible to assign dimension for sets that do not have integer dimension. To calculate fractal dimension, the box-counting method, as a pragmatic method, is used. A two-dimensional (2D) and three-dimensional (3D) box-counting method for calculating the fractal dimension of nonlinear ultrasonic waves in phase-space domain are utilized. It is shown that fractal dimension is a powerful signal processing tool for analyzing nonlinear ultrasonic signals in phase-space domain and extracting quantitative damage-sensitive features in presence of service load. In contrast, results of frequency domain analysis show a lack of noticeable trend, especially for large service loads. This observation indicates that features extracted in frequency domain may not be reliably used as damage-sensitive features for evaluating the level of cracking in concrete material under service load.

Adaptive depth and receptive field selection network for defect semantic segmentation on castings X-rays

Abstract: In order to ensure the safety of important castings, the ADR (Automatic-Defect-Recognition) system should recognize, locate, and count the area of internal defects that are undetectable to the naked eye. However, small differences between inter-classes, large defect scale change, and uncertainly annotation limit the achievement for ADR system. To solve these challenges, this paper presents an adaptive depth and receptive field selection network for defect semantic segmentation on castings X-rays. Firstly, the Resnet18 with ADSM (adaptive depth selection mechanism) is elaborately designed to extract and adaptively aggregate the different depth features, which is beneficial to distinguish the similar defect. Then the ARFB (adaptive receptive field block) is proposed to select the optimum receptive field in a data-driven manner to adapt to the scale change of defects. To overcome the problem of inaccurate labeling caused by the ambiguity of defect edges and the subjectivity of manual annotation, we propose a data augmentation method called “lazy-label”. Finally, we set up a castings defect segmentation dataset, called SRIF-CDS, to train and evaluate our method. Experiments on this dataset indicate that our method achieves 0.86 mIoU (mean intersection-over-union) and 0.92 mAcc (mean accuracy), which has better performance than the state-of-the-art semantic segmentation baseline.

Enhanced sizing for surface cracks in welded tubular joints using ultrasonic phased array and image processing

Abstract: This study presents an approach, which couples the ultrasonic wave reflection principle and image processing for the ultrasonic phased array, to enhance the sizing surface cracks in welded tubular joints. Ultrasonic phased array, as one of the non-destructive techniques, has wide applications in detecting and sizing of cracks in engineering structures. In detecting surface cracks in a welded tubular joint and potentially other types of welded components, however, the focused emitting ultrasonic wave, which is not normal to the surface crack plane, can create ultrasonic S-scan images with an inaccurate crack size and incorrect orientation. This study aims to improve the accuracy of the ultrasonic measurement through, 1) capturing multiple S-scan ultrasonic images by emitting the ultrasonic waves at different distances from the surface crack location; and 2) developing the algorithm based on the principle of ultrasonic wave reflections to improve the crack sizing and orientation from the ultrasonic S-scan images. Combined with the real-time crack monitoring during the low-cycle fatigue tests of tubular joints, the proposed scheme and algorithm achieves an enhanced agreement with other crack sizing approaches including the Alternating Current Potential Drop and measurements on the reproduced fracture surface using silicone-replica.

Photothermal super resolution imaging: A comparison of different thermographic reconstruction techniques

Abstract: This paper presents different super resolution reconstruction techniques to overcome the spatial resolution limits in thermography. Pseudo-random blind structured illumination from a one-dimensional laser array is used as heat source for super resolution thermography. Pulsed thermography measurements using an infrared camera with a high frame rate sampling lead to a huge amount of data. To handle this large data set, thermographic reconstruction techniques are an essential step of the overall reconstruction process. Four different thermographic reconstruction techniques are analyzed based on the Fourier transform amplitude, principal component analysis, virtual wave reconstruction and the maximum thermogram. The application of those methods results in a sparse basis representation of the measured data and serves as input for a compressed sensing based algorithm called iterative joint sparsity (IJOSP). Since the thermographic reconstruction techniques have a high influence on the result of the IJOSP algorithm, this paper highlights their advantages and disadvantages.

Fully noncontact measurement of inner cracks in thick specimen with fiber-phased-array laser ultrasonic technique

Abstract: To realize a fully noncontact ultrasonic testing method for inner cracks inspection in thick metal specimen, a phased array laser ultrasonic testing system with using a compact optic fiber array bundle and a laser interferometer is developed in this study. The focusing and steering of the shear wave and longitudinal wave generated with seven fiber-phased-array laser sources in thermoelastic regime is investigated by a numerical simulation and validated by the experiment. A non-contact measurement of the inner-surface cracks by both the angle-beam testing method and time-of-flight diffraction method with the fiber-phased-array laser ultrasonic technique have been studied.

A shape factor based ultrasonic measurement method for determination of bolt preload

Abstract: The existing ultrasonic testing models cannot accurately measure bolt preload due to the non-uniform distribution of axial stress in the effective stressed region of bolt and cannot rapidly complete the calibration of detection coefficients due to the difficulty to determine the key parameters such as effective stressed length. In order to eliminate the impact of non-uniform distribution of axial stress on the measurement, realize rapid and accurate calibration of detection coefficients and further improve the measurement accuracy of bolt preload, the concept and determination method of shape factor are proposed. Based on the shape factor, a combination of longitudinal and transverse waves is adopted to establish a mathematical model for measuring the preload applied to a bolt during assembly, avoiding the need to measure ambient temperature. The calculation method of shape factor is obtained by statics simulation of a bolted joint. Through the combination of finite element simulation and experimental calibration, the material factors and temperature factors of 45# steel as well as the shape factors of M20 and M16 bolts are obtained. The experiments show that the relative error of this method is within 5%, which meets the requirements of engineering applications.

Defects identification using the improved ultrasonic measurement model and support vector machines

Abstract: With a combination of the improved ultrasonic measurement model (IUMM) and support vector machines (SVM), a novel method to identify inclusions and cavities in metallic materials using scanning acoustic microscopy is proposed. In the IUMM, a hybrid model of Born approximation and Kirchhoff approximation is developed to calculate the far-field scattering amplitude of cavities, which improves the accuracy in phase and amplitude of the predicted pulse-echo signals of defects. The SVM classifier, with the amplitude and peak frequency of the predicted echo signals as major features, is applied to distinguish inclusions and cavities. The experimental result shows that the echo signals predicted by the proposed IUMM are more accurate than conventional UMM in amplitude and frequency. The SVM classifier, with the predicted signals as the training set, enables the identification of inclusions and cavities in metallic materials successfully. This work improves the performance of SAM in the identification of internal defects in metallic materials and realizes the intelligent analysis of ultrasonic signals.

Assessing overload-induced delaminations in glass fiber reinforced polymers by its geometry and thermal resistance

Abstract: The one-dimensional thermal quadrupole method is used to evaluate a pulsed thermography measurement at delaminations in a glass-fiber reinforced plastic plate quantitatively. The large-scale delaminations have been induced by tension overload and are air-filled and are usually located at the same depth as the notch bottom of a notch on the rear side. While classical evaluation methods like pulsed phase thermography and thermal signal reconstruction are focused on the delamination depth only, the thermal quadrupole method determines spatially resolved two parameters for delaminations, delamination depth and local thermal resistance. Interestingly, lateral heat flows do not disturb this kind of depth evaluation.

Detection of interfacial weakness in a lap-shear joint using shear horizontal guided waves

Abstract: This study aims to develop a shear horizontal guided wave based technique to evaluate the interfacial adhesion of aluminium-epoxy-aluminium tri-layer in a lap shear joint. A 3-D Multi-physics finite element model was developed to investigate the physics of the interaction of SH modes with a tri-layer structure. By employing the boundary stiffness approach, different cases of interfacial adhesion-ranging from perfect bond, intermediate and weak bond, were simulated. Frequency-wavenumber analysis reveals that at the bond overlap region, the incident SH0 wave mode-converts to fundamental (SH0-like) and first-order(SH1-like) modes. The dispersion characteristics of first-order mode (SH1-like) was found to be dependent on the adhesion level, and this influences the time responses collected on a receiver plate in guided wave through-transmission configuration. Experiments were carried out on aluminium-epoxy-aluminium lap shear joints using PPM-EMAT transducers. The analysis shows that this technique can detect and quantify different levels of adhesion, rather than merely classifying as good or bad bonds.

Full factorial analysis of the accuracy of automated quantification of hidden defects in an anisotropic carbon fibre reinforced composite shell using pulse phase thermography

Abstract: Infrared thermography (IRT) allows fast and cost-effective inspection of large structures made of carbon fibre reinforced polymer (CFRP) which must be checked for structural integrity after manufacture or during use. Due to its anisotropic and inhomogeneous microstructure, CFRP poses a major challenge for the quantitative determination of defects. In this paper we present a comprehensive full-factorial analysis of IRT data to verify the achievable accuracy of automated quantification of hidden defects of different size, shape and position in an inhomogeneous and anisotropic composite specimen, using pulse phase thermography.

Pipeline corrosion assessment using piezo-sensors in reusable non-bonded configuration

Abstract: In this paper, a piezoelectric sensor is used in its non-bonded configuration for non-destructive evaluation of a pipeline segment subjected to progressive corrosion damage. A new non-bonded sensor configuration, specially adapted for small diameter pipelines, known as modified non-bonded sensor (MNS), is proposed and evaluated. The new sensor configuration is evaluated for sensitivity in monitoring corrosion induced damage in pipeline segments in controlled laboratory setup through accelerated corrosion tests. The conventional surface-bonded piezo-sensor (SBPS) configuration is used as standard for comparison. Signature analysis using different methods is carried out in the study. Using the principles of the electro-mechanical impedance (EMI) technique, equivalent stiffness, damping and mass parameters are computed from the admittance signatures of both the configurations (MNS & SBPS) and compared. Experimental results show that piezo based a non-dimensional mass parameter from the MNS measurements are very effective in damage detection and quantification. The results are correlated with actual mass loss measurements and consequent corrosion rates are calibrated. The results clearly make case for practical usage of MNS configuration in corrosion detection and health monitoring of pipeline systems. The new MNS configuration proves to be a successful alternative to the conventional bonded piezo configuration.