Traditional Nondestructive Evaluation Research Articles

Interpretable & Explainable Machine Learning for Ultrasonic Defect Sizing.

Despite its popularity in literature, there are few examples of machine learning (ML) being used for industrial nondestructive evaluation (NDE) applications. A significant barrier is the 'black box' nature of most ML algorithms. This paper aims to improve the interpretability and explainability of ML for ultrasonic NDE by presenting a novel dimensionality reduction method: Gaussian feature approximation (GFA). GFA involves fitting a 2D elliptical Gaussian function an ultrasonic image and storing the seven parameters that describe each Gaussian. These seven parameters can then be used as inputs to data analysis methods such as the defect sizing neural network presented in this paper. GFA is applied to ultrasonic defect sizing for inline pipe inspection as an example application. This approach is compared to sizing with the same neural network, and two other dimensionality reduction methods (the parameters of 6 dB drop boxes and principal component analysis), as well as a convolutional neural network applied to raw ultrasonic images. Of the dimensionality reduction methods tested, GFA features produce the closest sizing accuracy to sizing from the raw images, with only a 23% increase in RMSE, despite a 96.5% reduction in the dimensionality of the input data. Implementing ML with GFA is implicitly more interpretable than doing so with principal component analysis or raw images as inputs, and gives significantly more sizing accuracy than 6 dB drop boxes. Shapley additive explanations (SHAP) are used to calculate how each feature contributes to the prediction of an individual defect's length. Analysis of SHAP values demonstrates that the GFA-based neural network proposed displays many of the same relationships between defect indications and their predicted size as occur in traditional NDE sizing methods.

Assessment of Composite Aluminum Adhesive Joints Using Digital Image Correlation

Traditional nondestructive evaluation (NDE) methods present significant challenges to detecting and characterizing kissing or weak bonds in adhesively bonded structures. These kissing or weak bonds also cannot transmit shear stresses or handle complex loading modes and, if not detected, can present a significant threat to the structural integrity of the components or systems. This paper demonstrates the digital image correlation (DIC) technique for evaluating adhesively bonded dissimilar materials joints subjected to kissing or weak bonds. The study employed four adhesively bonded carbon fiber reinforced plastics and aluminum (CFRP-Al) lap-shear test coupons with varied bond quality (i.e., with no contamination and three simulated kissing bond defects). The novelty of the approach presented in this paper was that this technique could detect and demonstrate changes in the normal strain (εyy) contour map of the contaminated specimens at relatively lower load levels. This load level corresponds to 15% of the failure load for the silicone and hydraulic oil contaminated sample and around 30% for the polyvinyl alcohol (PVA) contaminated sample. In addition, higher compressive strains along the overlap edges were observed in the strain map for the single lap joints due to the higher peeling stresses of the adherend and the stress concentration at the edges of an adhesively bonded joint.

Imaging method of interfacial anomalies in layered structures

Layered structures are progressively critical to several fields like aerospace, pipeline transportation, and petrochemical engineering. However, the anomalies located on the interfaces between the layers are troublesome yet problematic to image through traditional non-destructive evaluation methods. This study presented an imaging method utilizing interface waves and visualized the interfacial anomalies in layered structures. The paper first derived and solved the dispersion equations of a layered plate. The wave structure of the interface waves and their interaction mechanisms with interfacial anomalies make them inherently sensitive to interfacial anomalies. Moreover, the interface wave modes are non-dispersive at high frequencies, causing macroscopic differences between the signals with or without abnormal features. On that basis, correlation analysis and probabilistic reconstruction formed the imaging algorithm in this work. Lastly, the experimental system was tested on two different layered metal plates. The results show that the system can precisely depict interfacial anomalies of different shapes. Especially for the circular anomalies, the produced images coincide with the pre-manufactured defects with relatively minor errors of location and area.

Assessing the challenges of condition assessment of steel-concrete (SC) composite elements using NDE

Steel-concrete (SC) is a structural typology and construction method in which continuous steel plates are used on the surfaces of concrete walls or slabs, having dual role of formwork and longitudinal reinforcement. The continuous steel-plates on both sides of the concrete element improve economy by prefabrication, but hinder the use of traditional non-destructive evaluation (NDE) to assess the condition of the concrete. In this research, conventional NDE techniques were investigated with regards to their suitability for SC elements, with focus on detection of mechanical damage. Tests were performed on large-scale, mechanically damaged, SC elements. The deployed NDE methods were based on ultrasonic measurements using: the pulse echo technique; radiographic inspection; ultrasound 3D tomographer; impact-echo (IE); impulse-response (IR) and ground penetrating radar (GPR). It was assessed whether cracks of different sizes in the SC beams could be detected through the steel-plated surface. Perpendicular and oblique cracks of 0.1–2.5 mm width were studied. Resultsshow that ultrasonic testing and GPR were not suitable for testing through the steel surface, due to loss of the transmitted energy in the steel plates. Mechanical impact-based testing systems revealed promising results to detect cracks and de-bonding. Radiographic testing was successful in detecting the presence of cracks in cases when these had sufficient opening and length.

Nondestructive Evaluation of Special Defects Based on Ultrasound Metasurface

We demonstrate the nondestructive evaluation by means of directional ultrasound emitted from a planar metasurface. The ultrasound metasurface is designed to generate the collimated and directional ultrasound efficiently in a planar configuration, which is endowed with the full-2π-range phase manipulation ability and high transmittance up to 80%. We employ the directional emission based on the ultrasound metasurface to innovate the traditional nondestructive evaluation methods, benefited from the freely controlled directivity and the superior fitness to sample surface of the planar metasurface. Merits of this innovative application are evidenced by the remarkable accuracy (higher than 98%) in the thickness evaluation, and precise detection (accuracy higher than 96%) of the special defect inside the V-shaped workpiece which is intractable to be inspected conventionally. The implementation of the metasurface-based directional ultrasound emission in the nondestructive evaluation bears the advantages of high coupling efficiency, superior fitness, high accuracy, and applicability to special defect, providing new solutions to the challenges in conventional defect detection and promotes the development in the nondestructive evaluation applications.

An Overlapping Array Coil for Eddy Current Non-destructive Evaluation

An overlapping array coil for eddy current nondestructive evaluation (NDE) is proposed in this paper. Compared with the traditional NDE probe, our proposed structure works at 6.78 MHz, and generates a more stable and uniform magnetic field by interlacing and stacking coil units. Based on the propagation characteristics of a time-varying magnetic field, when a metal plate is placed in such a magnetic field, eddy currents are induced on the surface. If a scratched metal plate is placed in the same magnetic field, different eddy currents will be generated and the port impedance of the structure will be changed. Therefore, defects in the metal can be non-destructively detected.

Piezoresistive Sensing Approaches for Structural Health Monitoring of Polymer Composites—A Review

Structural health monitoring (SHM) is an emerging paradigm of real-time in situ structural evaluation for the detection of damage and structural degradation. This is achieved while the structure is kept in-service as against traditional non-destructive evaluation (NDE) techniques which require scheduled interventions while the structure is kept offline. SHM offers great advantages over traditional regimens of condition monitoring (CM) by improving structural reliability and safety through timely detection of structural defects also known as “diagnosis”. Polymeric composite materials offer the unique opportunity of integrating different phases for designing self-sensing smart systems capable of self-diagnosis. Polymers are unique in the sense that they can be designed in various configurations as they generally have facile manufacturing procedures. Among other properties, piezoresistance is the one that can be detected in composites in real-time as a function of strain. Conductive polymers including intrinsic and extrinsic conductive polymers can be used to induce piezoresistivity in composites. Careful design procedures can be adopted to maximize the sensitivity of these piezoresistive composites in order to fully exploit the potential of this property for SHM. Various manufacturing/integration strategies can be employed to effectively use piezoresistance in composites for structural health monitoring. These include self-sensing in carbon fiber-reinforced composites, use of surface deposited/mounted sensing films and patterns, integration of filaments and yarns during reinforcement manufacturing or lay-up and impregnation of reinforcements with piezoresistive matrices. A comprehensive review of these techniques is presented with the view of their utility in the SHM of composites. A selection criterion for these techniques is also presented based on sensitivity, manufacturing method and detection capability.

Bayesian‐optimized unsupervised learning approach for structural damage detection

AbstractStructural health monitoring (SHM) is developing rapidly to fulfill the world's need for resilient and sustainable communities. Due to the current advancements in machine learning and data science, data‐driven SHM is an attractive solution for real‐time damage detection compared to the traditional nondestructive evaluation techniques. However, most widely available data‐driven SHM methods rely on fully or partially simulated data to train the statistical model, and thus require a number of predefined assumptions and parameters, or are not adapted for post‐extreme events damage diagnosis. In this study, we propose a density‐based unsupervised learning approach for structural damage detection and localization. This approach leverages cumulative intensity measures for damage‐sensitive feature extraction for the first time in an unsupervised learning approach. Furthermore, a statistical model construction process is proposed based on kernel density maximum entropy (KDME) and Bayesian optimization. The framework is evaluated in three case studies. The first two involve a numerical three‐story building and a numerical nine‐story asymmetrical building that are both subjected to 100 ground motion excitations while considering environmental variations. The proposed framework is able to detect and localize damage in those case studies with an average accuracy of 92%. The third case study, which contains 44 shake‐table tests of a three‐story frame structure with masonry infill, is used to experimentally validate the proposed framework in damage detection. The three case studies demonstrate the potential and robustness of the proposed Bayesian‐optimized, multivariate KDME novelty detection framework for detecting and localizing structural damage, especially after extreme events.

Repeatable part authentication using impedance based analysis for side-channel monitoring

The rise of the interconnected manufacturing enterprise has extended its vulnerability space to malicious cyber-physical attacks. To improve the resilience of a manufacturing enterprise to cyber-to-physical attacks, a robust set of quality tools for part and process authentication and verification are needed. However, current Non-Destructive Evaluation (NDE) methodologies, which focus on verifying a small subset of geometric features of a part such as its length, width or depth among others, are easier to circumvent by an attacker. This work presents a novel approach to improve the resilience of a manufacturing enterprise via side channels, specifically using electromechanical impedance signatures measured by piezoelectric transducers (PZTs) to verify if a part conforms to its design intent. Since electromechanical impedance signatures take into account all geometric features and material characteristics in a holistic manner, they can provide a new class of robust quality metrics that can be used in tandem with traditional NDE methodologies to verify a part or a process. Two major contributions of this work include (1) demonstrate the feasibility of using a PZT mounted to a fixture to facilitate side channel analysis for part authentication and compare its performance to other PZT mounting methods, and (2) develop a new damage metric that properly distinguishes between nominal parts and various altered groups. The results of our study show that using PZTs as a side channel has the potential for part and process authentication in a manufacturing setting. Furthermore, we show that a PZT mounted to a secondary device, such as a fixture, can be used for NDE in a manufacturing setting.

Integration of scheduled structural health monitoring with airline maintenance program based on risk analysis

The aerospace industry is striving to reduce the aircraft operating costs while maintaining required safety level. Emerging technologies such as the structural health monitoring to reduce long-term maintenance cost and increase aircraft availability are promoted by the manufacturers. To successfully integrate the structural health monitoring technology into the current maintenance process of modern commercial aviation, a clear definition of the structural-health-monitoring-based maintenance operational concept and the system level requirements is required. This article proposed a structural health monitoring operational concept and the associated maintenance cost modeling and risk assessment methods for the implementation of the structural health monitoring in commercial aviation industry. The developed methodology provides a tool to determine the optimal scheduled structural health monitoring inspection interval and repair decision thresholds for approved scheduled structural health monitoring task. A simulated case study is carried out to demonstrate the structural health monitoring operational concept and how an optimal maintenance strategy can be determined using the proposed methodology. Preliminary results show that the integration of the structural health monitoring into the existing maintenance process can reduce the maintenance cost compared to that of the current practice using the traditional Non-Destructive Evaluation (NDE) techniques while maintaining the risk below an acceptable level.

Phased Array Imaging of Complex-Geometry Composite Components.

Progress in computational fluid dynamics and the availability of new composite materials are driving major advances in the design of aerospace engine components which now have highly complex geometries optimized to maximize system performance. However, shape complexity poses significant challenges to traditional nondestructive evaluation methods whose sensitivity and selectivity rapidly decrease as surface curvature increases. In addition, new aerospace materials typically exhibit an intricate microstructure that further complicates the inspection. In this context, an attractive solution is offered by combining ultrasonic phased array (PA) technology with immersion testing. Here, the water column formed between the complex surface of the component and the flat face of a linear or matrix array probe ensures ideal acoustic coupling between the array and the component as the probe is continuously scanned to form a volumetric rendering of the part. While the immersion configuration is desirable for practical testing, the interpretation of the measured ultrasonic signals for image formation is complicated by reflection and refraction effects that occur at the water-component interface. To account for refraction, the geometry of the interface must first be reconstructed from the reflected signals and subsequently used to compute suitable delay laws to focus inside the component. These calculations are based on ray theory and can be computationally intensive. Moreover, strong reflections from the interface can lead to a thick dead zone beneath the surface of the component which limits sensitivity to shallow subsurface defects. This paper presents a general approach that combines advanced computing for rapid ray tracing in anisotropic media with a 256-channel parallel array architecture. The full-volume inspection of complex-shape components is enabled through the combination of both reflected and transmitted signals through the part using a pair of arrays held in a yoke configuration. Experimental results are provided for specimens of increasing complexity relevant to aerospace applications such as fan blades. It is shown that PA technology can provide a robust solution to detect a variety of defects including porosity and waviness in composite parts.

Nonlinear ultrasonic stimulated thermography for damage assessment in isotropic fatigued structures

Traditional non-destructive evaluation (NDE) and structural health monitoring (SHM) systems are used to analyse that a structure is free of any harmful damage. However, these techniques still lack sensitivity to detect the presence of material micro-flaws in the form of fatigue damage and often require time-consuming procedures and expensive equipment. This research work presents a novel "nonlinear ultrasonic stimulated thermography" (NUST) method able to overcome some of the limitations of traditional linear ultrasonic/thermography NDE-SHM systems and to provide a reliable, rapid and cost effective estimation of fatigue damage in isotropic materials. Such a hybrid imaging approach combines the high sensitivity of nonlinear acoustic/ultrasonic techniques to detect micro-damage, with local defect frequency selection and infrared imaging. When exciting structures with an optimised frequency, nonlinear elastic waves are observed and higher frictional work at the fatigue damaged area is generated due to clapping and rubbing of the crack faces. This results in heat at cracked location that can be measured using an infrared camera. A Laser Vibrometer (LV) was used to evaluate the extent that individual frequency components contribute to the heating of the damage region by quantifying the out-of-plane velocity associated with the fundamental and second order harmonic responses. It was experimentally demonstrated the relationship between a nonlinear ultrasound parameter (βratio) of the material nonlinear response to the actual temperature rises near the crack. These results demonstrated that heat generation at damaged regions could be amplified by exciting at frequencies that provide nonlinear responses, thus improving the imaging of material damage and the reliability of NUST in a quick and reproducible manner.

Accurate Microwave Measurement of Coating Thickness on Carbon Composite Substrates

This paper describes the utility of a microwave-based nondestructive evaluation (NDE) method for accurate thickness measurement of thin dielectric coatings (e.g., paint and primer) on a carbon composite (CC) substrate and the results are compared with those using a conducting (aluminum) substrate. The small thickness and the diverse dielectric properties associated with these coatings, along with the anisotropy associated with CC substrates, significantly limit the use of traditional NDE methods. This method not only overcomes these limitations but also provides for simultaneously evaluating the thicknesses of multiple coating layers (and also their dielectric constants). The electromagnetic simulation results for simultaneously estimating the thicknesses of a top and a bottom coating (i.e., primer and paint) as well as experimentally evaluating thicknesses of several coatings on such substrates are presented. In addition, thickness evaluation sensitivity to several critical parameters such as CC fiber orientation and variation in the measured coating dielectric constant used in the thickness recalculation is also investigated and presented.

Damage detection sensitivity characterization of acousto-ultrasound-based structural health monitoring techniques

Reliability quantification is a critical and necessary process for the evaluation and assessment of any inspection technology that may be classified either as a nondestructive evaluation or structural health monitoring technique. Based on the sensitivity characterization of nondestructive evaluation techniques, appropriate processes have been developed and established for the reliability quantification of their performance with respect to damage/flaw detection in materials or structures. However, in the case of structural health monitoring methods, no such well-defined and general applicable approaches have been established for neither active nor passive sensing techniques that allow for their accurate reliability quantification. The objective of this study is to characterize the sensitivity of active-sensing acousto-ultrasound-based structural health monitoring techniques with respect to damage detection, as well as to identify the parameters that influence their sensitivity. With such an understanding, it is believed that adequate quantitative methods could then be established to enable the practical use of acousto-ultrasound structural health monitoring methods in the aerospace and mechanical engineering communities. In order to evaluate the sensitivity of a pre-selected active-sensing acousto-ultrasound structural health monitoring system, both numerical simulations and experiments were performed on 30 aluminum coupons each outfitted with a pair of lead zirconate titanate piezoelectric sensors/actuators. A damage index versus damage size relationship was investigated numerically and experimentally to assess the applicability of the traditional nondestructive evaluation linear regression framework for probability of detection for an active-sensing structural health monitoring system. The results of the study show that the position of each sensor–actuator pair with respect to a known damage location and the damage growth pattern are the two most critical parameters influencing the reliability of the same structural health monitoring system applied to identical structural components under the same environmental conditions.

Multi-frequency local wavenumber analysis and ply correlation of delamination damage

Wavenumber domain analysis through use of scanning laser Doppler vibrometry has been shown to be effective for non-contact inspection of damage in composites. Qualitative and semi-quantitative local wavenumber analysis of realistic delamination damage and quantitative analysis of idealized damage scenarios (Teflon inserts) have been performed previously in the literature. This paper presents a new methodology based on multi-frequency local wavenumber analysis for quantitative assessment of multi-ply delamination damage in carbon fiber reinforced polymer (CFRP) composite specimens. The methodology is presented and applied to a real world damage scenario (impact damage in an aerospace CFRP composite). The methodology yields delamination size and also correlates local wavenumber results from multiple excitation frequencies to theoretical dispersion curves in order to robustly determine the delamination ply depth. Results from the wavenumber based technique are validated against a traditional nondestructive evaluation method.

Delamination and concrete quality assessment of concrete bridge decks using a fully autonomous RABIT platform

One of the main causes of a limited use of nondestructive evaluation (NDE) technologies in bridge deck assessment is the speed of data collection and analysis. The paper describes development and implementation of the RABIT (Robotics Assisted Bridge Inspection Tool) for data collection using multiple NDE technologies. The system is designed to characterize three most common deterioration types in concrete bridge decks: rebar corrosion, delamination, and concrete degradation. It implements four NDE technologies: electrical resistivity (ER), impact echo (IE), ground-penetrating radar (GPR), and ultrasonic surface waves (USW) method. The technologies are used in a complementary way to enhance the interpretation. In addition, the system utilizes advanced vision to complement traditional visual inspection. Finally, the RABIT collects data at a significantly higher speed than it is done using traditional NDE equipment. The robotic system is complemented by an advanced data interpretation. The associated platform for the enhanced interpretation of condition assessment in concrete bridge decks utilizes data integration, fusion, and deterioration and defect visualization. This paper concentrates on the validation and field implementation of two NDE technologies. The first one is IE used in the delamination detection and characterization, while the second one is the USW method used in the assessment of concrete quality. The validation of performance of the two methods was conducted on a 9 m long and 3.6 m wide fabricated bridge structure with numerous artificial defects embedded in the deck

Terahertz frequency optical properties of acoustic materials

An investigation of the terahertz frequency electromagnetic absorption coefficient spectra and refractive index spectra is presented for many materials employed in acoustic applications. A brief discussion on the use of these material properties for terahertz imaging applications is also included. Terahertz-frequency electromagnetic radiation has the potential to image through and inside many acoustically damping materials, unlike traditional nondestructive evaluation techniques such as ultrasound. The optical properties discussed herein are necessary to quantify potential terahertz imaging performance for these and other acoustic materials.

Nondestructive assessment of pore size in foam-based hybrid composite material

In situ nondestructive evaluation (NDE) during processing of high-temperature polymer-based hybrids offers great potential to gain close control of and achieve the desired level of pore size, with low overall development cost. During the polymer-curing cycle, close control over the evolution of volatiles would be beneficial to avoid the presence of pores or control of their sizes. Traditional NDE methods cannot realistically be expected to evaluate individual pores in such components as each pore evolves and grows during curing. However, NDE techniques offer the potential to detect and quantify the macroscopic response of many pores that are undesirable or intentionally introduced into these advanced materials. In this article, preliminary results will be presented for nondestructive assessment of pore size in foam-based hybrid composite materials using ultrasonic techniques. The ultrasonic method was chosen due to its high sensitivity to the change of composition in material, ability to provide the required depth of penetration and ability to provide the needed modulus information for the cured material through appropriate velocity measurements. Pore size was evaluated through the frequency content of ultrasonic signal. The effects of pore size on the attenuation of ultrasound were studied. Feasibility of this method was demonstrated on two types of foams with various pore sizes.

A multi-mode structural health monitoring system for wind turbine blades and components

The Adverse Event Detection (AED) system described in this paper supports nondestructive evaluation (NDE) systems and evaluates advanced composite structures such as wind turbine blades. AED is a joint effort by Extreme Diagnostics and Virginia Polytechnic Institute and State University (Virginia Tech). AED uses the impedance method to monitor bulk structural integrity, wave propagation methods to assess surfaces, and acoustic emission (AE) structural health monitoring (SHM) to detect adverse events such as impacts. The incorporation of AE methods significantly increases the sensor coverage area, which is crucial in health monitoring of large-scale structures like wind turbine blades. Our AED system provides on-line assessment of structural integrity during normal operations, as opposed to traditional nondestructive evaluation (NDE) methods that are commonly applied off-line—that is, systems must be shut down for inspection by conventional NDE. Our AED system not only provides timely information during operation, it can also reveal defects that only become apparent under operational stress—such defects can be overlooked by traditional off-line inspection methods. AED actively examines structures across several length and time scales in an autonomous fashion, thus greatly reducing the number of sensors required and lowering system complexity and cost. Our early AED prototype demonstrated impedance-based SHM in wind turbine blades. This project integrates three previously independent SHM approaches, and demonstrates damage detection on a composite structure. Our current AED prototype is a low-power, wireless, and embeddable sensor that detects incipient damage in near real-time and automatically provides early warning of structural damage. Our AED system is also suited to a variety of aerospace applications that include composite overwrapped pressure vessels. AED also supports Homeland Security, and furthers national preparedness by monitoring infrastructure integrity and disaster response by providing damage assessment.

Spatial Recognition of a Superconducting Quantum Interference Devices Nondestructive Evaluation System Using a Small Room-Temperature Probe

A superconducting-qantum-interference-device (SQUID) nondestructive evaluation (NDE) system using a small room-temperature probe is developed for active scanning rather than for a massive movement occurring in a traditional SQUID NDE system. The small room-temperature probe is composed of a quadruple excitation coil and a double D-shaped pickup coil. It is connected to the input coil surrounding a high-Tc rf SQUID, immersed in liquid nitrogen, and shielded by a shielding can. Beyond the NDE function, the SQUID NDE scheme has spatial recognition functions, including the detection of the orientation and depth of a narrow line crack using different parameters, and the scanning of images of large objects with arbitrary shapes. Furthermore, the spatial sensitivity, limited by the size of the probe, reaches up to only 7 µm in the aspect of crack widths and 1 mm in the aspect of spatial intervals for precision NDE on a printed circuit board.