This PDF file contains the front matter associated with SPIE Proceedings Volume10971, including the Title Page, Copyright information, Table of Contents, Introduction, and Conference Committee listing.

Research in ultrasonic NDE over the past several decades has been supported by a growing use of specialist modeling tools, to calculate wave propagation behavior, the influences of materials, guided waves, and the scattering of waves from features and defects. Model capabilities are now so good that simulations are being used in a similar manner to experiments, and some important research objectives are not possible at all without them. The NDE research group at Imperial College has worked over many years on the long-term development of some general purpose modeling tools, which have provided essential underpinning to the creation of new capabilities in NDE. This presentation will use some examples of research achievements in NDE to illustrate the vital role of advanced modeling tools in their success.

Highways consist of many large structures requiring vigilant inspection and maintenance. While significant research attention has been focused on the health management of bridges, comparatively less attention has been paid to other highway structures including retaining walls. In the United States, there has been a recent emphasis on extending the use of risk management methods to the extremely large national inventory of retaining wall structures. In this paper, a long-term wireless monitoring system is developed as a cost-efficient approach to collecting data and information required for risk assessment of retaining wall structures. The study focuses on two reinforced concrete (RC) retaining walls to highlight the monitoring system design and to illustrate how measurement data offers insight to wall performance. The first wall is a caisson supported retaining wall along the M-10 freeway in Detroit, MI; the second is a classical reinforced concrete cantilever wall along I-696 in Southfield, MI. The wireless monitoring system installed on each wall system uses a cellular-based wireless sensor node termed Urbano that is solar powered and customized to measure wall tilt using inclinometers, wall strain using strain gages, and wall temperature using thermistors. The monitoring systems have been valuable in assessing the behavior of the M-10 and I-696 wall systems for a broader risk management framework. The monitoring results reveal both wall systems are operating as designed with limited tilt and strain responses to normal environmental factors including moisture and temperature.

The layout of structural bolts used in infrastructures are determined based on the mechanical and design requirements. Minimum spacing is set to facilitate construction and control stress concentration, and maximum spacing is set to prevent the intrusion of water between plates and provide sufficiently equal force distribution to each bolt. Ultrasonic testing is one of the most widely used and rapid in-situ inspection methods to evaluate the status of bolted connections; however, scattering from hole boundaries may mask reflections from the cracked and corroded surfaces. In current practice, the ultrasonic-based inspectability of bolted connections for the presence of crack and corrosion as well as pretension loss is not a design criterion. In this paper, the inspectability is added as a design variable within the boundaries of design limits that control spacing as well as distribution of bolts at the connection. The ideal spatial distribution of different bolt groups is proposed to detect the critical crack length and area loss hidden between plates. The proposed bolt distribution is numerically tested to show the minimized influence of hole scatters to the ultrasonic inspectability of defects. The regression model is built to predict crack position and size.

Microwave/radar sensors and techniques are widely used for detecting underground or subsurface targets in archeology, geophysics, and civil engineering. Among existing microwave/radar sensors and techniques, synthetic aperture radar (SAR) imaging enables researchers and engineers to conduct surface and subsurface detection of metallic objects with improved detectability. The noncontact, remote sensing feature of SAR imaging provides a safer approach in a dangerous mission, such as demining. The objective of this paper is to investigate the depth (d) effect of a metallic object buried in dry sand. A steel disk specimen of 15-cm diameter was buried inside a box (sandbox) filled up with dry sand at various depths (d = 10 cm, 18 cm, and 26 cm) and scanned by a 10-GHz SAR system. Three ranges (R = 15 cm, 30 cm, and 60 cm) between the SAR antenna and the sandbox were also considered in this research. It was found that the SAR amplitude and its distribution decrease with the increase of buried depth and the increase of range. Distribution of SAR amplitudes representing the buried metal disk specimen also changed with the increase of buried depth. Empirical models were also proposed for range and depth effects of subsurface metallic objects in SAR images.

In recent years, textiles are used as a structural material in externally strengthening/retrofitting deteriorated and damaged concrete structures. Formation of externally strengthened/retrofitted concrete structures creates a new type of multi-layer dielectric system for their condition assessment using non-destructive evaluation (NDE) techniques. The objective of this paper is to investigate the use of microwave/radar NDE on a one-layer textileconcrete system for condition assessment. In this paper, we use a synthetic aperture radar (SAR) imaging system at 10 GHz to study the effect of an externally attached textile layer on the SAR images of two concrete panels. One type of textile was used on a 30.48 cm by 30.48 cm by 2.54 cm concrete panels to form a one-layer textileconcrete system. Various ranges (20 cm, 30 cm, 40 cm, 50 cm and 60 cm) were considered. Our experiment results demonstrated that the SAR imaging can successfully distinguish the type of textiles. Furthermore, it was found that electromagnetic pattern of the textile layer varies with range in SAR images. Empirical models were developed to characterize the range effect on the SAR images by using textile applied on concrete panels.

Corrosion of steel rebar in reinforced concrete (RC) structures introduces internal stress at the interface between rebar and concrete, ultimately leading to the failure of structures. Detection of early-stage corrosion of steel rebar can significantly reduce maintenance cost and risks. An active photoacoustic fiber optic sensor system had been proposed for early-stage corrosion detection of steel rebars by generating and receiving surface ultrasonic waves. However, the implementation of a corrosion detection method requires knowledge of surface ultrasonic waves propagating at rebar-concrete interface. The objective of this study is to investigate the interaction of surface ultrasonic waves with local geometries (of a number four rebar) and concrete covers using the finite element method (FEM). In this study, seven three-dimensional finite element models were created to simulate surface ultrasonic waves propagating in three different cross-sections of a steel rebar with different concrete cover. Three lug locations and three types of concrete (differed by Youngs modulus) were considered. The pitch-catch mode was adopted, in which one source and one receiver were deployed at each rib of the rebar. 1 MHz sinusoidal pulse was introduced at the source while time domain radial displacements were collected at the sensor. Short-time Fourier transform was used to analyze collected time domain radial displacements. From our simulation results, it was found that high frequencies of ultrasonic waves were affected by lugs more than lower frequencies. Presence of concrete cover suppresses the amplitude of surface ultrasonic waves.

This paper presents the improved thickness estimation technique using Steady-state excitation Continuous-scanning Laser Doppler Vibrometery(SCLDV). In this study, the wavenumber sensitivity with respect to the thickness variations of a structure, along with the information of wave modes, is utilized to find an optimal interrogation frequency band for SCLDV. In addition, the use of multi-frequency steady-state response is used to improve the accuracy of the thickness variation. By utilizing the wavenumber sensitivity along with the multi-frequency excitations, the proposed SCLDV shows the improved depth estimation, compared to the previous approaches which empirically select the interrogation frequency. For validation of this technique, several experiments were performed on steel plates, which contain corrosion damage with various depth variations. The results showed that the proposed technique is very efficient in detecting and visualizing very small thickness variations of a structure at high speed.

Low reliability and high maintenance cost of using power and data cables are two main reasons motivating the application of the self-powered wireless sensors for structural health monitoring (SHM) systems in bridge structures. On the other hand, energy harvesting systems have been introduced as a solution for the current limitations of the batterypowered wireless sensors associated with the finite life-span of batteries and their replacements. The objective in this paper is to propose a new optimized nonlinear energy harvesting concept, namely Bistable Energy Harvesting (BEH) system, for smart SHM of bridge structures. In this study, a dynamic analysis of the energy harvesting system for cablesupported bridges subject to wind-induced vibration is carried out and the feasibility of the energy harvesting device is investigated. This paper presents efficient linear and nonlinear energy harvesting systems for wireless monitoring of long-span cable-supported bridges. It is shown that level of the extracted energy from such energy harvesting system is quite sufficient to supply energy for self-powered sensors of a bridge health monitoring system. This study is to promote the recent line of research on self-powered sensor networks for smart bridge monitoring being performed at the Florida International University.

Coda waves experiencing long propagation time and travel path are sensitive to weak changes in a medium. Coda wave interferometry (CWI) is an efficient method to analyze coda waveform variations. In this paper, the CWI technique is used to detect minute cracks and stress changes in a 6-meter reinforced concrete beam. Specifically, four-bending tests with varied loads are conducted on the beam, and a couple of sensors are installed sparsely to collect ultrasonic wave signals. Then for each source-receiver pair, the coda waveform variations between load steps are quantified using the CWI technique. The results show that the stress changes and minute cracks in the beam can be detected through the velocity changes and decorrelations of the coda waveforms. The presented study may provide a useful tool for concrete structural nondestructive evaluation and testing (NDT) applications.

Deep learning-based defect feature recognition from 2D image datasets, has recently been a very active research area and deep Convolutional Neural Networks have brought breakthroughs toward object detection and recognition. Due to CNN’s outstanding performance, several recent studies applied it for defect detection in either routine or post-earthquake infrastructure inspections and have reported competitive performance and potential toward automating infrastructure safety assessment. Despite their benefits, the majority of 2D approaches do not leverage or provide 3D depth information directly from the content of the images and as a result do not enable the 3D measurement of severity of these defects. With the increased popularity of 3D scanning and reconstruction technologies, there is pressing need for defect recognition models that operate on 3D data. In this paper, a novel framework using Deep 3D Convolutional Neural networks (3DCNNs) termed 3D InspectionNet is introduced to learn 3D defects features from an artificially generated 3D dataset, intended to mimic defects on the surface of concrete columns such as either cracks or spalls. InspectionNet has the capability of learning the distribution of complex defect features from a large 3D dataset, and distinguishing defects features. For training 3D InspectionNet, a large simulated 3D defect dataset of 3D CAD models was automatically constructed with labeled defect features. The proposed framework can distinguish defect features from the geometric data such as voxels with a high accuracy. The results of this preliminary work demonstrate and emphasize the feasibility and potentials of this approach for 3D defect detection in automated inspection applications.

The research objective was to evaluate the temperature dissipation during the Hot Mix Asphalt construction from the surface, base, and subbase layers in Southern Taiwan. The scope was to install a multilayered temperature monitoring station along with the weather station that allows temperature data to be recorded and stored continuously. The multilayered temperature monitoring station was installed and comprised of K-type thermocouples. Thermocouples were installed from the surface in-depth 0, 5, 10, 15, 17, 20, 30, 42.5, 72.5, and 90-cm along with the construction sequences starting from base and subbase compaction on soils and graded aggregates in May 2018 and thereafter the paving task of HMA surface layer in August, 2018. It has to be noted that the temperature data was recorded immediately and continuously when thermocouples were installed which enables the monitoring of pavement temperature dissipation during the HMA construction. In this paper, we aim to present a temperature dissipation pattern that is able to improve and facilitate the field HMA paving task. It is nine hours or more, instead of six hour of which roadways agencies often regulate, that is required for HMA pavement materials to cool down and open for traffic.

This article aims to develop a pressure sensing method by utilizing both a contacting active sensor and a non-contacting laser ultrasound transmitter. An overloaded stress in an industrial pressure tank such as a nuclear reactor may cause a catastrophic explosion; thus, it is essential to monitor the mechanical stress in a reliable manner for the structural safety. Among many different types of stress sensing methods, ultrasound sensing has been attractive due to its non-invasive measurement feature. For the recent decades, subsurface longitudinal (SSL) ultrasonic wave has been widely used since it is not only less dependent on the internal medium and the surface condition, but also has the fastest wave speed without wave distortion. In our work, laser source and Aluminum nitride (AlN) wafer are used to generate and to receive SSL ultrasonic waves, respectively. In order to increase the photoacoustic efficacy, a composite of carbon-soot nanoparticles (CSNP) and polydimethylsiloxane (PDMS) was attached onto the intermediate wedge at the transmitter side. The photoacoustic experiment results demonstrate a reasonable linear relationship between the stress level and the time-of-flight variation of the propagated wave signal.

This study proposes an acoustic emission (AE) monitoring approach for old steel bridges strengthened with postinstalled shear connectors. In this strengthening method, cyclic traffic loads gradually fatigue the connectors and increase the bridge deflection at which the connectors engage in shear transfer. The key parameter that this study aims to estimate is how much the connectors need to slide before they engage in shear transfer. To estimate this parameter, this paper leverages the difference in AEs from shear connectors before and after they engage in shear transfer. Specifically, the b-value of AEs a k-mean clustering approach are used. To validate this novel approach, a full-scale, two-span steel girder with a concrete deck was used. The girder was strengthened with post-installed shear connectors and subjected to 20 cycles of sequential loads representing moving trucks. The results confirmed the effectiveness of the approach based on AE during the unloading of the girder.

The compressive strength of concrete structure is always influenced by the composition of varied materials, casting process, and curing period, etc. Among these variables, an optimal mix of different materials will achieve better structural compressive strength. Thus, understanding the non-linearity of concrete and its variables is paramount for improving and predicting the performance of concrete structures. Due to the expensive and time-consuming laboratory analysis, the use of post-processing and data analysis provides an excellent opportunity to explore and predict optimal models for concrete compressive strength performance. However, given the inadequacy of traditional regression models and other analytic techniques in modeling non-linear regression problems, there is still a need to achieve a better predictive model with minimal errors as well as the capability to estimate partial effects of characteristics on response variables. In this study, a predictive analysis was carried out to investigate the performance of concrete compressive strength at 28 days with a new machine learning model called boosting smooth transition regression trees (BooST). It is observed from the experimental results that the BooST model provides a better prediction accuracy in comparison with the state-of-the-art techniques used for concrete compressive strength prediction. Thus, there is a great potential to apply the BooST model for predicting the compressive strength of concrete in practice.

A probabilistic risk assessment method to assess the failure possibilities of aircraft fatigue critical components due to fatigue damage initiation and propagation, as well as the effect of complex maintenance scenarios throughout the aircraft’s service life (including multiple repair types and various nondestructive inspection (NDI) techniques), needs to be developed for aircraft fatigue life management. The traditional Monte Carlo simulation (MCS) offers the most robust and reliable solution; however, MCS is time consuming and unable to support prompt risk decisions. To relieve the computational burden, a novel probabilistic method—AMETA (Aircraft Maintenance Event Tree Analysis)—was developed, which combines the generality of random simulations with the efficiency of analytical probabilistic methods. AMETA consists of a fatigue maintenance event tree and a probabilistic algorithm comprising a set of probabilistic equations. AMETA systematically transforms a complex random maintenance pattern requiring a large number (in the order of billions) of MCSs to more logical and manageable fatigue paths represented by a finite set of probabilistic events to achieve the required computational accuracy and efficiency. Furthermore, the Importance Sampling Method (ISM) can be used for efficiency improvement. In this paper, the accuracy, efficiency and robustness of AMETA are verified and demonstrated. A procedure was provided to select the most suitable sampling functions for ISM. It is found that AMETA is several orders of magnitude more efficient than MCS for the same level of accuracy.

Development of new approaches in the nondestructive evaluation of welded steel plates for large naval structures continues to be an area of interest for the DoD and the military complex. In this article we will evaluate an elastic-microwave based detection and imaging method using numerical and experimental methods. The evaluation was performed on two test articles that were designed to represent critical structural components with flaws of interest. The flaws, cracks and weld defects have been included in the test articles to determine detection sensitivity and accuracy of the proposed method. Our approach uses a microwave interferometer (MI) to record the scattered response of flaws in the steel plated as it is driven by an incident elastic field. The MI can “see” (penetrate) through the viscoelastic coating to the upper surface of the plate. Out-of-plane displacement amplitudes of 10nm in the frequency range (25kHz to 42kHz) are readily observable, with the key feature being, that the surface bond between the coating and the steel plate are undisturbed. The non-contact aspect of the interferometer allows for large surface regions to be accurately and efficiently scanned in space and time. These spatiotemporal data coupled with specialized wavefield processing algorithms provide powerful detection and imaging capabilities. From these wavefield data sets, a plate thickness mapping capability has been demonstrated that can detect thickness changes on the order of 0.79 mm (1/32”) with spatial resolution on the order of the spatial sampling rate, 7.5 mm (~1/4”). We have also shown that a topological energy analysis of the wavefield data can detect and locate small flaws, on the order of 5-10 mm (0.025-0.40”) in the welded joint of a 1.5” thick T-plate. Note, all of these results are obtained through a 2” thick viscoelastic coating without disturbing the coating or the coating bond. The current results indicate that we are detecting and locating damage (flaws) in the plate that are smaller than the wavelength of the propagating guided modes. Classical scattering theory places a (λ/2) resolution limit on the detectability of flaws in terms of the incident field, however, since the spatial resolution of the scanned region is much smaller, (▵x=▵y=15mm) and the inherent natural focusing of the time reversal operation, we are able to detect smaller flaws on the order of (λ/10). It is important to realize that we are not imaging the flaw but detecting and localizing a difference between a reference (pristine) sample and the measured (damaged) sample, relative to a spatial grid on the surface. The current scan resolution is 15mm × 15mm. At present we cannot expect to resolve individual flaws within a grid space only their cumulative effect. Even with the current limitations, this imaging approach appears to be a promising alternative to current methods where the coating layer is removed.

The paper provides an engineering analysis approach for assessing reliability of NDE flaw detection using smaller number of demonstration data points. It explores interdependence of probability of detection (POD), probability of false positive (POF), contrast-to-noise ratio, and net decision threshold-to-noise ratio in a simulated data; and draws some generically applicable inferences to devise the approach. ASTM nondestructive evaluation standards provide requirements on signal-to-noise ratio and/or contrast-to-noise ratio in order to provide reliable flaw detection and limit false positive calls. POD analysis of inspection test data results in a flaw size, denoted by a_(90/95). The flaw size has 90% POD and minimum 95% confidence. POF is also estimated in the analysis. POD demonstration requires specimen with flaws of known size. In many situations, it is very expensive to produce the large number of flaws required for the POD analysis. In some situations, only real flaws can truly represent the flaws for demonstration. Real flaws of correct size and location within part may be difficult to produce, if not impossible. Here, an engineering analysis approach is devised to assess reliability of NDE technique when a limited number of flaws are available for demonstration. A technique is considered reliable, if it provides flaw detectability equal to or better than a_(90/95) and also provides a POF less than or equal to a chosen value. The paper uses simulated signal response versus flaw size data to devise the approach. Linear correlation is used between the signal response data and flaw size. POD software mh1823 uses generalized linear model (GLM) in POD analysis after transforming the flaw size and signal response, if needed, using logarithm. Therefore, this approach is in agreement with the linear signal correlation used in mh1823. Using the POD analysis of data, generic conditions on contrast-to-noise ratio and net decision threshold-to-noise ratio are derived for reliable flaw detection. In order to assess technique reliability using the engineering approach, signal response-to-flaw size correlation about the flaw size of concern is needed. In addition, measurement of noise is also needed. If the technique meets the above requirements, assumption of linear signal –to-flaw size correlation and conditions on noise, then the technique can be assessed using this analysis as it fits the underlying POD model used here. The approach is conservative and is designed to provide a larger flaw size compared to POD approach. Such NDE technique assessment approach, although, not as rigorous as POD, can be cost effective if the larger flaw size can be tolerated. Typically, this is a situation for all quality control NDE inspections. Here, an NDE technique needs to be reliable and the true a₉₀ is not known, but the assessed flaw size is assumed to be larger than the true a₉₀ due to conservative factors or margins. Applicability of the approach for assessing reliability of flaw detection in x-ray radiography and 2D imaging in general is also explored.

The paper provides a procedure for validating flaw detection size using smaller number of signal response data points or number of flaws. MIL-HDBK-1823 probability of detection (POD) analysis of inspection test data results in a flaw size estimate, denoted as a_90/95. The a_90/95. flaw size has 90% POD and minimum 95% confidence. Here, a case is considered where the inspection test data points are not sufficient to run the POD analysis but are sufficient for limited validation approach provided here. The procedure is based on an approach developed by the author. The approach is conservative and it validates a flaw size that is greater than a theoretical a₉₀ used in the simulation data. By using simulated signal response data, the paper shows that conditions based on limits on ratios, called merit ratios, such as contrast-to-noise ratio (CNR), and net decision threshold-to-noise ratio (TNR), and contrast to threshold ratio (CTR) can be used in validating a reliably detectable flaw size. Monte Carlo random data samples from a population of data of a signal response characteristics are used. Decision threshold for each sample and corresponding merit ratios are computed. Using the decision threshold computed from each sample and the signal to flaw size characteristics, the theoretical a_th90/95 is computed for each sample. The merit ratios and theoretical a_th90/95 for each sample are plotted as distributions. These distributions are analyzed to calculate confidence in the meeting the merit ratio conditions. The target flaw size is compared with the distribution of the theoretical a_th90/95 to calculate confidence in validating the target size. The procedure requires measurement of noise. Results of the simulated data prove validity of the approach. Minimum six data points are recommended. Although, small number of flaws are sufficient in this validation procedure, these flaw sizes are carefully selected based on experimental work to find the smallest flaw size that can meet the merit ratio conditions. Therefore, some trial-and-error effort is needed to choose correct target flaw size so that the merit conditions can be met.

We present an approach for quality control using dimensional metrology based on structured light applied to laser beam melting. This application serves as a model for in-line monitoring under adverse conditions such as elevated temperatures, stray light and thermal gradients in the process chamber making closed-loop control very ambitious. Nevertheless, we demonstrate the implementation of precise dimensional measurement of printed parts and powder deposition allowing in- or off-line process assessment. As data acquisition and processing must be fast, we investigate the so-called structure function to extract global or local parameters such as roughness or geometrical periodicities like waviness of the powder deposition.

In the field of structural health monitoring (SHM), innovative methods of non-destructive evaluation (NDE) are currently being investigated with the purpose of safer, longer lasting structures. While current SHM is dominated by acoustic emission and vibration-based methods, it is desirable to integrate NDE techniques with existing structural reinforcement techniques which increase structural service life. Multifunctional materials that can detect internal structural damage while also increasing structural service life offer an inherent advantage as construction materials to be integrated into new and existing structures. Embedding shape memory alloy (SMA) wires in concrete components offers the potential to monitor their structural health via external magnetic field sensing. SMAs have been used to close internal cracks, reinforce concrete structures, and reduce fatigue under cyclic loading, so the addition of such a multi-functional material for the purpose of structural evaluation is very desirable. Thus an evaluation of SMAs for magnetic sensing is required for both the structural and magnetic domains. A concrete beam containing iron-based magnetic SMA (MSMA) wire is subjected to a three point bend test where structural damage is induced, resulting in a localized phase change of the MSMA wire. Magnetic field lines passing through the embedded MSMA domain are altered by this phase change and can thus be used to detect damage within the structure. A good correlation is observed between the computational and experimental results, and the magnetic sensing sensitivity is explored via a robust computational model to evaluate the effectiveness of external magnetic sensing.

Mitigating the structural damage caused by thermal expansion cycles is a primary objective in the design of concrete structures, such as bridges or buildings. One method to achieve this goal is the introduction of shape memory alloys (SMAs) as a replacement of traditional steel reinforcements in concrete structures. SMAs exhibit a characteristic known as “shape memory effect,” which allows the recovery of large deformations through the alloy’s martensite and austenite phase transformations. This effect gives SMAs an inherent advantage over steel. The purpose of this paper is to characterize the effect of an embedded SMA rod on a concrete system undergoing a thermal cycle, and to optimize the configuration of these materials. To achieve these ends, a system is modeled in Abaqus, a software suite for finite element analysis, consisting of a concrete block with an embedded, prestrained SMA rod, in which the concrete and SMA material properties have been determined from experimentation and secondary research. A set of the SMA’s properties (max transformation strain, coefficient of thermal expansion, stress influence coefficients, and volume fraction of SMA to concrete) are iteratively altered to produce characterization of the rod’s effect on the system, and then the same set are again altered using a multi-objective optimization tool to minimize deflection and maximize the temperature where concrete damage occurs. This approach is a cost-effective method to characterize the effects of these material properties and produce results that can be utilized in future projects where SMAs are deployed in large-scale concrete structures.

In this paper, development of a nonlinear vibro-acoustic modulation technique based on non-contact piezoelectric sensors was investigated to detect the crack progression of concrete cracking caused by thermal treatments. Experimental results show that defined ultrasonic nonlinear parameter is in agreement with the accumulation of thermal crack. The phase velocity of Rayleigh wave and resonance frequency of vibrations were measured and compared with ultrasonic nonlinear parameter to validate the sensitivity of developed method. X-ray Computed Tomography (CT) technique is applied to visualize microstructure of thermal damage. The CT images show that proposed nonlinear parameter is reliable and well correlated with the microstructural defects of concrete specimen. Due to the advantage of removable characteristic of non-contact ultrasonic measurements, the developed non-contact nonlinear wave modulation method could be promising for quick and convenient damage assessment of concrete structures in engineering practice.

Masonry structures are widely used for their low cost, durability, fire-resistance, sound isolation and other properties in civil engineering and architecture. Inspection of masonry structures is vital for maintaining their structural performance and long-term safety. Many traditional inspection technologies (e.g., acoustic/ultrasonic, thermographic, electromagnetic) have been applied for the structure health monitoring of masonry structures. The objective of this paper is to apply a synthetic aperture radar (SAR) system for characterizing the dielectric constant of masonry specimens. A 10-GHz imaging radar system was used. A masonry wall was selected, and five ranges (150 cm, 250 cm, 500 cm, 1000 cm and 1500 cm) were considered in collecting SAR images of the masonry wall. From our result, it was found that attenuation of integrated SAR amplitude exhibits a nonlinear pattern as a function of range. An algorithm was developed to estimate the dielectric constant of the masonry wall. Field collected SAR images were compared with a portable 1.6-GHz ground penetrating radar (GPR) system. It was demonstrated that SAR images can be used to estimate dielectric constant of masonry structures in the field.

The aging bridge infrastructure network is in critical need of maintenance, rehabilitation or replacement (MR and R) as nearly half of this inventory is approaching the end their design service lives. Agencies responsible for managing this network have limited resources that are insufficient for the scale of the problem, highlighting the need for smart, system-level decision-making strategies that can be integrated with current practice. A large amount of rich information on element level condition descriptions are buried in bridge inspection reports, but this local information is seldom used holistically to infer system performance. Current decision-making strategies are constrained by limitations in bridge deterioration prediction models, which lack comprehensive and well-structured databases needed for automation of processes associated with high resolution forecasting. How to draw meaningful information from the details of these localized reports to assist system-level bridge condition comparison and maintenance prioritization still remains unclear and warrants further study. To bridge this gap, this paper proposes a Natural Language Processing framework to extract information from the raw textual data in bridge inspection reports. This raw data provides a source for capturing the experience-driven metric inherent to the bridge inspection process. The proposed framework constructs an innovative bi-directional Long-short Term Memory neural network that automatically reads inspection reports into different condition categories and achieves 96.2% accuracy when examined on inspection reports collected by Virginia Department of Transportation. The extracted information forms a well-structured bridge condition inventory that contains rich historical and local condition information, and hence enables smart, system-level bridge MR and R decision-making.

Knowing the grain angle of structural wood members relative to the direction of loading is important in estimating their mechanical performance. The strength of a structural wood member can vary greatly depending on the orientation of grain, making the understanding of grain angle an essential requirement for machine grading and for structural analysis. Condition assessment of structural wood members using non-destructive evaluation (NDE) techniques (e.g., microwave/radar, ultrasonic, stress wave, and X-ray) is a major approach for existing wood structures. Among existing NDE techniques, synthetic aperture radar (SAR) imaging offers technical advantages include remote sensing and subsurface sensing. The objective of this paper is to use SAR imaging to quantify the change in grain direction of various size wood dowels relative to the imaging radar. Two wood specimens were produced each 14 in. long with diameters of 1.25 in. and 1.5 in. Each wood specimen was placed vertically inside an anechoic chamber and imaged at different orientations using a 10 GHz SAR system. It was found that the integrated SAR amplitude and amplitude distribution were affected by the grain orientation of the wood specimens. Further analysis was conducted by estimating the area of contour slices of the SAR images taken in the amplitude direction and were used to confirm the findings.

Photonic crystal-based strain visualization film is promising for detecting the age-related deterioration of large man-made structures and public infrastructures. However, as the number of target structures increases, monitoring them all will become a major problem. We propose two solutions: (1) a portable solar-battery-powered automated monitoring station to monitor the color of photonic coatings, and (2) the application of real-time image analysis using mobile phones to record color changes. Both solutions make use of the power of small computers, while the former assists us with efficient data collection, and the latter helps non-experts to inspect structures without using expensive spectroscopes. The portable monitoring station consists of a micro-computer connected to a 3G mobile network, a USB camera, and a solar battery system installed in a waterproof box. Photographs of the strain visualization film are taken once every hour and, at all other times, the computer disconnects the camera to save electricity. We placed four monitoring stations in the shade of a bridge or a tree and ran them continuously for more than a year. The application displays a real-time image in which only the strain-free area of the film is extracted. As a result, the region under strain and the background appear in white. This software runs on many mobile computers with built-in cameras and the OSs including Android, iOS, Windows, and Linux. This is possible due to the versatility of the computer vision library we used, namely OpenCV, which is widely used in robotics and automatic car-driving.

Based on the modern image recognition technology, this study used the fractal theory to examine the surface cracks of concrete and the fractal dimension was used as an index to identify the bearing capability of concrete components. Through processing and simulating the experimental data obtained by other researchers, the relationship of crack fractal dimension D and the loading-ratio ξ of the concrete component was calculated. The D-ξ relationship of concrete beams is D = 0.1684 ξ + 1.0307. Based on this relationship, a fast evaluation method based on the fractal dimension of the crack surface to judge the bearing capacity of concrete components was proposed in this study, and its accuracy was verified by the analysis of experimental results.

Wood structures have been recently considered as one of the important building styles due to their green use of natural resource, low energy consumption, and less waste emission. But wood structures are vulnerable to near surface material deterioration over long duration, as the result of mechanical impact, chemical corrosion and fungi attack. In order to maintain such natural structure in designed mechanical performance, it is necessary to detect the near surface wood defect with non-destructive evaluation (NDE) methods on a regular basis. Among the existing NDE methods, acoustic-laser technique has been demonstrated for detecting near surface detect in a variety of structures. When using this technique in wood structure detection, the measurement accuracy may be affected by the wood grain. Taking into account this issue, the present research work aims to understand the wood grain on the dynamic behavior of near surface layer in wood structure under acoustic excitation and assess the grain effect on the defect detection via acoustic-laser technique. In this study, artificial defects are fabricated in two wood beams, with the transverse and parallel directions to the grain respectively. Vibration characteristics measured by acoustic-laser technique for the two grain cases are compared and discussed. This paper also provides the suggestions of practical application of the technique in detecting wood structure in considering the grain effect.

Piezoelectric cement, instead of PZT (lead zirconate titanate) sensors and smart aggregate, has been developed as a new piezoelectric sensor that particularly applies to monitor concrete structures. Piezoelectric cement is a 0-3 type cementbased piezoelectric composite with 50% PZT for improving the incompatibility of acoustic impedance and volume deformation between conventional piezoelectric sensors and concrete structure. Piezoelectric cement was installed in concrete to monitor the strength development with the age and to detect the damage of concrete by electromechanical impedance technique. The PZT sensor was the counterpart in the experiments. Results indicate that, similar to PZT sensors, piezoelectric cement exhibits the capability of monitoring concrete structures, and the sensitivity of monitoring for piezoelectric cement even better than for the PZT if the piezoelectric cement with suitable piezoelectric strain factor d₃₃. Piezoelectric cement embedded in concrete structures show no resonant frequency in the conductance-frequency spectra that causes to assess the conductance change easily. For the piezoelectric cement with d₃₃ = 101 pC/N, the intervals of frequency are 300–660 kHz and 1000–2000 kHz for the strength monitoring and the damage detection, respectively. Broad effective frequency range provides larger RMSD value of conductance.

Composite materials play important roles in multifunctional applications, and thus, the diagnosis of damage patterns in composite materials becomes crucial to avoid critical events" such as structural or functional failures. The impact of an individual damage in composite materials has been extensively studied, however, the interaction of defects/cracks, which leads to critical fracture paths, has not been understood well. In this paper, we develop a Bayesian estimation based statistical analysis technique that estimates the damage pattern of a composite material, in particular, the relative positions of defects in the material, by measuring its through-thickness dielectric properties. We first explain the fundamental dielectric principle that leads to the detection of defect patterns. A capacitance model is then built to measure the material permittivity, and the relationship between the dielectric permittivity and relative positions are found using COMSOL Multiphysics. The interaction effects between defects observed in the simulation are interpreted using the fundamental dielectric principle. A Bayesian estimation based statistical analysis model is then developed to estimate the relative positions of defects in composite materials from the measured global dielectric properties.

Traditional physical model-based nondestructive evaluation (NDE) and damage detection methods are often unreliable due to the complex dependence of model parameters on minor differences in material properties (e.g., thickness, temperature, or loading effects). While classic data-driven approaches appear to eliminate model complexity, their performance highly depends on feature extraction, for which domain-expertise-based data preprocessing is required. Wavefield analysis is a promising alternative for non-contact NDE but suffers from the problem of slow data acquisition. As a result, effective structural health monitoring (SHM) based on wavefield analysis of guided waves in large-scale systems, such as mechanical, civil, or aerospace structures, has remained challenging. To address these challenges, we present a deep convolutional neural network (DCNN)-based transfer learning approach to interpret ultrasonic guided waves with small training data sets, thereby achieving rapid, effective, and automated SHM. Specifically, the proposed learning framework includes a pre-trained DCNN for automated feature extraction from the raw inputs (i.e., wavelet-transformed time-frequency images) and a fully connected classification stage that is trained with partial wavefield scans. Experiments on full wavefield scans of various thin metal plates demonstrate the effectiveness and efficiency of the proposed approach: >95% classification accuracy is obtained with only 5% training data, thus enabling fast scanning and fully automated damage detection of large-scale structures.

3D printing of multi-material objects enables the design of complex 3D architectures such as printed electronics and devices. The ability to detect the composition of multi-material printed inks in real time is an enabling feature in a wide range of manufacturing sectors. In this study, dielectric properties of microscale embedded metal particles in a dielectric matrix have been characterized using impedance measurements as a function of particle size, shape, volume percentage and frequency. Measurements were found to agree well with calculations based on an anisotropic Maxwell-Garnett dielectric function model. Despite the metal loading exceeding the theoretical percolation threshold, a percolation transition was not observed in the experimental results. With this data, a calibration curve can be established to correlate metal loading with impedance or capacitance, which can be used with an in situ sensor for ink composition measurements during extrusion-based 3D printing. We demonstrate how an in situ sensor can locally measure the composition of the ink, allowing greater control over the resulting properties and functionality of printed materials.

Soft-matter technologies are essential for emerging applications in wearable computing, human-machine interaction, and soft robotics. However, as these technologies gain adoption in society and interact with unstructured environments, material and structure damage becomes inevitable. Here, we present a robotic material that mimics soft tissues found in biological systems to identify, compute, and respond to damage. This system is composed of liquid metal droplets dispersed in soft elastomers that rupture when damaged, creating electrically conductive pathways that are identified with a soft active-matrix grid. This presents new opportunities to autonomously identify damage, calculate severity, and respond to prevent failure within robotic systems.

Infrared thermography (IRT) is a matured tool, and it can be employed to monitor the health conditions of structures by measuring surface temperature information in real time and in a non-contact way. The surface temperature information provides an important clue to identifying the defects on the building exterior surfaces. According to the surface temperature measurements, for those parts covered by shadows, the surface temperature information is smaller than it is supposed to be. Similarly, glare effects in IRT can be defined as the excessive and uncontrolled brightness illustrated in IRT such that the surface temperature information is larger than it is supposed to be. In general, the shadow and glare effects are often introduced in the thermal images obtained using the passive IRT when the solar energy is the main heat source. The current study proposes an image model in a multiplicative way to evaluate the shadow and glare effects presented in IRT. The experimental results demonstrate that the proposed image model does efficiently remove the shadow or glare effects. A calibrated thermograph can be generated by introducing proper level set functions in the numerical model.

Acoustic tomography method facilitates mapping internal defects in real-time and in-situ without destructive testing. The method requires certain number of transmitter and receiver paths to reconstruct the slowness map of scanned area depending upon the target resolution. Once the hardware component is determined, the major software output to feed into the algorithm is the time of flight. There are sophisticated signal processing methods reported in literature to determine the time of flight (TOF) with better accuracy as compared to conventional threshold-based method. The most common approaches are wavelet-based or energy-based methods, which require transforming time history signal into different domains. Domain transformation is typically applied in laboratory-scale experiments. In this paper, a new arrival time pick-up approach based on defining outliers in the derivative of transient signal in time domain is evaluated in terms of accuracy, computational effort and power as compared to threshold-based and wavelet/energy-based methods reported in literature. The waveforms from experiments is used to study the influence of materials and signal-to-noise ratio on the accuracy of detecting the fastest wave mode. In addition, waveforms are also artificially generated with fixed wave velocity using numerical models to further evaluate the performance of the different methods (outlier-based, threshold-based and energy-based). The influence of tomography quality by using these to this method performs better in accuracy and efficiency.

A novel method is proposed in this paper to extract acoustic emission (AE) source using Helmholtz potentials approach. In order to characterize the source in terms of potentials the underlying physics is to detect the AE signals and then develop an inverse algorithm to characterize the source. According to the new concept, the source can be characterized it provides the excitation potential information from the crack and that can be used to diagnose the crack (crack type and crack growth etc.). An AE experiment was designed to measure the acoustic emissions from the fatigue crack growth. A test specimen was made of 1 mm thick 304-steel material. A small hole (1 mm diameter) was drilled at the center of the specimen to initiate the fatigue crack. The specimen was subjected to the cyclic loadings by using the hydraulic MTS machine. AE waveforms are generated by convolutions of AE source functions, plate transfer functions. An inverse algorithm was developed to characterize the source of AE signal. AE signal analysis are done to determine AE source function using deconvolution process.

This work proposes a nonlinear dual frequency (modulated) phased array technique for fatigue crack evaluation in aluminium and metallic structures, and near field enhancement in composite materials. A standard ultrasound system with multiple transmitting and receiving elements was used to excite an aluminium fatigue sample and impacted composite plate with two frequencies. The method relied on the evaluation of harmonic sidebands which can be correlated to defects/damage in materials, with nonlinear methods having shown increased sensitivity versus standard techniques. An initial pump frequency (f₂) is used to initialize a ‘breathing/ringing’ crack after which a second frequency (f₁) is used to further excite the cracked region and generate sidebands from the modulated frequencies. This method relies on both an amplitude and frequency subtraction techniques to filter out linear ultrasound effects. One of the benefits of modulation techniques in general are reduction of equipment based nonlinearities, commonly produced when conducting single frequency tests. An aluminium coupon with a closed fatigue crack was evaluated using single and dual frequency methods. The results showed that the dual frequency excitation method was able to more accurately define the extent of the crack and improved accuracy versus standard phased array techniques.

Damage such as micro cracks, layer delaminations, corrosion or barely visible impact damage (BVID) could irreparably affect the integrity of the structure. These defects are not ever detectable by the common inspection techniques based on the ultrasonic wave propagation. However, a number of techniques based on nonlinear wave behaviour have been recently developed to improve the sensitivity of ultrasonic methods. The nonlinear acoustic approach proposed in this work relied on generation of new frequency generation due to defects. The spectral changes are caused by nonlinear local dynamics of defects of various scale and nature due to contact between crack surfaces. A standard Air Coupled ultrasound (ACU) system arranged with 88 transmitting elements and 1 receiving element focused on the same point (N=88 mm) with a central frequency (f₀) of 41 KHz was used to excite corroded samples. Results showed that the intact parts of the material outside the defect vibrate linearly, i.e. with no greatly frequency variation in the output spectrum, whilst a small cracked defect behaves as an active radiation source of a new frequency component (2f₀). For the nonlinear ultrasonic testing, the second order nonlinear parameter (β) was chosen as the nonlinear feature to damage identification. In conclusion this research work demonstrated that nonlinear techniques are suitable for numerous classes of defects, such as fatigue cracks and corrosion (micro-cracks).

The design flexibility of aerospace components has been revolutionized with the recent advancements in the manufacturing of composite structures and 3D printed components. Since the manufacturing process is dependent on the control environment, variations from the nominal conditions can result in the degraded surface quality of the manufactured part. It is critical to image these surface features at high-resolutions to enable process feedback, therefore, improving the overall efficiency and process control. In this context, we develop and demonstrate a structured illumination optical fiber probe with embedded speckles for high-resolution surface feature imaging of 3D printed and composite samples. The improved sectioning capabilities by modifying the structured illumination patterns is validated using a plane mirror sample. This method is envisaged for high-resolution surface feature imaging to improve the overall manufacturing process efficiency of critical components.

Carbon fiber composites offer outstanding structural performance with high specific strength and are experiencing significant commercial adoption as the fiber price continues to decrease. Composite research efforts now need to focus on creating multifunctional composites, which can offer sensing capabilities in addition to structural attributes. This work focuses on creating multifunctional carbon fiber composites with structural health monitoring capabilities through the integration of piezoresistive nanoparticles on the surface of carbon fiber. Prior research introduced the development of coating silicon carbide nanoparticles on the surface of carbon fiber in a continuous feed-through process to achieve increased SHM sensitivity with enhanced interlaminar strength and tunable mechanical damping properties. One benefit of that coating process is the compatibility with various nanomaterials. This research capitalizes on that benefit by coating different nanoparticles, such as titanium dioxide, on carbon fiber to further enhance the sensing capabilities. A modification to the prior coating process is made in this research to enable significantly higher nanoparticle loading to be achieved. The resulting composites more accurately measure an applied force by responding with a more profound electrical resistance change. This research lays the foundation for efficiently integrating nanoparticles onto fibers leading to homogenously dispersed nanoparticles throughout a fiber reinforced composite for multifunctional performance.

An innovative wireless passive system for impact detection on large-scale composite airframe structures is presented. The wireless system is designed to operate with a sensor network for onboard of aircraft for structural health monitoring, of composite airframe. The wireless systems efficient design allows for low power consumption, wireless communication capability, system robustness and large sensing area. The system is evaluated on a large-scale stiffened composite fuselage under different operational conditions. It is demonstrated that it is possible to detect impact events with different impact energy levels and impact locations over a large monitoring area. This work provides a potential solution for aircraft on-board structural health monitoring with no human intervention. This sensing system can be also adapted to other Internet of Things and structural health monitoring applications.

Composite materials are susceptible to barely visible impact damages (BVID) due to low-velocity impact. Therefore, an automated damage detection and quantification technique is highly desirable for quick inspection of large number of composite structures. Among various different non-destructive techniques (NDTs), active thermography NDT can be used for detecting damage on aircraft structures, using an infrared camera to capture the temperature distribution on the structure after it is exposed to heat using a flash lamp. In this paper, an image analysis algorithm that analyzes the infrared image, acquired using NDTherm NT, by determining the changes in the colormap values to automate the detection and quantification of the damage size was proposed. An area of the second derivative pre-processed grayscale image acquired using NDTherm NT is scanned in the x-direction and y-direction, and for each scan region the histogram of colormap values is extracted and stored. Irregularities in the structure result in non-uniform temperature distributions, which cause the infrared image to have a wide-range of grayscale colormap values in the damaged area. Therefore, the damage region is identified by monitoring the changes in the number of detected grayscale colormap values. The proposed image analysis technique was implemented for automated damage detection on Boeing 787 skin’s curved CFRP panel, with dimensions of 1.3 × 1.3 m. The proposed method detected the damage and determined the maximum damage length in the x-direction and y-direction to be 70.1 mm and 57.8 mm, respectively. Moreover, the proposed technique is suitable for feature identification applications.

This study presents a low-cost and small-scale Structural Health Monitoring System (SHM) for thin walled carbon fiber reinforced plastics (CFRP) structures based on acoustic emission (AE) analysis. It covers the inherent geometric complexity and anisotropic properties of such structures through the implementation of an artificial neural network (ANN). The system utilizes piezoelectric sensors, a data acquisition unit and a microprocessor with a trained ANN in order to localize events that result from artificial sound sources. Besides high precision in localization the system is scalable and adaptive through adequate design and training of the ANN and system hardware. Especially for CFRP, nowadays well established for lightweight applications in the aerospace and automotive industry, such a system helps to overcome their major downside, their sensitivity towards impact loading. Impact sources like bird strikes, tool drops or stone debris can be the cause for delamination that can result in a severe drop of stiffness and early catastrophic failure. In order to guarantee structural integrity, CFRP structures therefore need to be inspected via nondestructive testing methods on a regular scheme. Due to its passive nature and in-situ capabilities AE-based SHM can reduce cost and down-time that come with regular inspections as an alternative approach that allows for a conditionbased inspection scheme.

Ultrasonic Lamb waves have been proved useful for nondestructive evaluation (NDE) due to their abilities to propagate a long distance with less energy loss as well as their high sensitivity to small defects on the surface or inside the structure. However, there are still many challenging tasks for Lamb wave based NDE due to the complexity involved with Lamb waves propagation and the complexity caused by coupling layer used in traditional contact-type transducers. This paper established a fully non-contact Lamb wave NDE system by using a non-contact air-coupled transducer (ACT) which can eliminate the need for couplant/adhesive and actuate pure fundamental A0 mode Lamb wave; and a non-contact scanning laser Doppler vibrometer (SLDV) for sensing which can provide high spatial resolution wavefield data. Through ACT, pure A0 mode was actuated at selected ACT incident angle based on Snell’s law. By SLDV sensing, multi-dimensional wavefields for one-dimensional or two-dimensional wave propagation were obtained and further used for Lamb waves’ characterization analysis. A specimen with a through-thickness crack was manufactured and adopted to evaluate the inspection capability of the ACT-SLDV system. Waves with normal incident and aligned incident w.r.t the longitudinal dimension of the crack were investigated, and our results showed that cracks were detected successfully in both cases. Moreover, crack lengths can be quantitatively evaluated for both situations with 10% error.

The heterogeneities contained in concrete will cause strongly multiple scattering behaviors during the propagation of ultrasonic waves, forming the so-called coda waves. External loads will slightly change concrete structural size and further introduce stretching effects on coda waveforms. In this paper, coda waves are collected from several concrete samples under different loads, and the waveform variations are quantified through a stretching technique. The results show that their stretching ratios are varied according to external load strengths, which implies that stress changes in concrete can be detected by coda wave measurements. The presented study could be very value for nondestructive testing of concrete structures.

This paper presents phase-space analysis of nonlinear ultrasound in concrete materials subjected to different compression loads. Nonlinearities due to defects and material properties can alter frequency content of transmitted and received ultrasound waveforms. As a result, nonlinear ultrasound waveforms are traditionally analyzed in frequency domain. However, using frequency domain to analyze ultrasound behavior has several shortcomings. Different sources of nonlinearities can make the same change in the frequency domain. This can make the identification of the source of ultrasound nonlinearities impossible. In addition, it is hard to observe and explain complex nonlinearities such as chaotic behavior of ultrasound waveforms in the frequency domain. Analyzing nonlinear ultrasound behavior in phase-space domain allows for a better understanding of ultrasound behavior. Fractal analysis of phase space portrait is used to quantitatively evaluate topography of phase space portrait of ultrasound waveforms. Fractal analysis is a quantitative feature to describe geometry evolution and it can enhance quantitative analysis of phase space portrait. The phase-space along with fractal analysis is proven to be a powerful tool in analyzing ultrasound nonlinearities and determining chaotic behavior.

Guided waves have been extensively explored in the field of Structural Health Monitoring (SHM) and Non-Destructive Testing (NDT) over many years. Guided waves are usually excited and registered by using an array of piezoelectric transducers attached to the surface of the inspected structure. However, in some cases, piezoelectric transducers cannot be used directly on the structure and noncontact wave excitation methods are preferable. This paper is a continuation of authors’ previous research on complete non-contact NDT methods in which low-cost resonant-based ultrasonic transmitters together with scanning laser Doppler vibrometer are used. The aim of the paper is the comparison of guided wave actuation ability in a composite plate by using the single ultrasonic transmitter, flat ultrasonic transmitter array and spherical ultrasonic transmitter array. Full wavefield of propagating guided waves was registered on a fine grid of points by using laser vibrometer and processed by using the wavenumber filtering method. Finally, the accuracy of delamination localization and size was estimated for each actuation setup.

In structural health monitoring (SHM) of composite members, the choice of sensor positions, frequency band and signal interpretations are directly affected by the attenuation levels. Hence, it is important to consider the influence of attenuation for quantitative interpretation of signals in SHM applications. In this paper, attenuation of the two fundamentals Lamb wave modes, namely the symmetric mode S₀ and antisymmetric A₀ were experimentally measured in Carbon Fiber Reinforced Polymer (CFRP) laminates. The stress waves were launched using piezoelectric wafers bonded to the center of the laminate. Symmetric and antisymmetric modes were excitation through appropriate combination of two piezoelectric wafers bonded on opposite surfaces of the plate at the same location. The out of plane displacements corresponding to both modes were measured using a scanning laser vibrometer (SLV), along different orientations starting from 0⁰ to 90⁰ at 15⁰ increments. Attenuation coefficients were obtained from 100 kHz to 350 kHz for both cross-ply as well as quasi-isotropic laminates. The attenuation coefficients ranged from about 5 nepers/m to 40 nepers/m. The experimental results as obtained by the scanning laser vibrometer compared favorably with results available in the literature. The antisymmetric mode was found to undergo significantly higher attenuation compared to the symmetric mode. The attenuation of the two modes further depends on the direction of propagation and their frequency components. Because of the dependence of the attenuation level on the Lamb wave mode, frequency, and direction of propagation, a wide frequency content signal such as acoustic emission will undergo substantial changes before being detected by sensors. Successful acoustic emission monitoring will have to take such attenuation into account to get reliable results.

Acoustic emission (AE) based structural health monitoring relies on detection and analysis of stress waves released by damage growth. The features of AE waveforms such as amplitude, frequency, energy, and rise time could be used characterize the relationship between acoustic emission signals received at a particular instant and the nature of damage growth, and several studies have examined such relationships. Acoustic emission generated by fatigue cracks propagate as a combination of a number of Lamb wave modes in plates, and hence by identifying the modal components it is potentially possible to learn about the conditions under which the crack is propagating. This paper examines the modal features of the acoustic emission signals generated at different stages of fatigue crack growth. A key requirement for this analysis is high fidelity required from the sensor recording AE signals, so that the different modes can be identified. Fatigue crack in a 6061 aluminum plate was monitored using a wideband sensor. The fatigue crack grew over a length of about 3 inches during 74,000 cycles and resulted in over 100,000 waveforms. The waveforms were examined in detail and the features were evaluated. As expected, the waveform’s peak amplitude and energy content were indicative of the rate of fatigue crack growth. More importantly, the modal features of the waveform were found to be indicative of the nature of crack growth. Mode 1 crack growth that resulted in signals that contained almost exclusively the fundamental symmetric mode S₀. When the crack is propagating in shear mode, the waveforms contained dominant A₀ mode. Hence, acoustic emission waveform can potentially identify the transition of a Mode 1 crack to shear crack.

The manufacturing process of carbon fiber reinforced polymer (CFRP) composite structures can introduce many characteristic defects and flaws such as fiber misorientation, fiber waviness and wrinkling. Therefore, it becomes increasingly important to detect the presence of these defects at the earliest stages of development. Eddy current testing (ECT) is a nondestructive inspection (NDI) technique which has been proven quite effective in metallic structures. However, NDI of composite structures has mainly relied on other methods such as ultrasonics and X-ray to name a few, and not much on ECT. In this paper, we explore the possibility of using ECT in NDI of CFRP composites. We base our research on the fact that the CFRP displays some low-level electric conductivity due to the inherent conductivity of the carbon fibers. This low-level conductivity may permit eddy-current pathways that can be exploited for NDI detection. An invention disclosure describing our high-frequency ECT method is being progressed. We use multiphysics FEM simulation to simulate the detection of various types of manufacturing flaws and operational damage in CFRP composites such as fiber misorientation; waviness; wrinkling, etc. ECT experiments were conducted on CFRP specimens with manufacturing flaws using the Eddyfi Reddy eddy current array (ECA) system.

This work proposes a novel technique for the localization of low-velocity impacts in composites without a-priori knowledge of the mechanical properties nor the speed of propagating waves, thus overcoming current limitations of existing impact localization methods. The proposed algorithm is based on the estimation of the power of acoustic emissions generated by impacts on a composite plate instrumented with embedded piezo-transducers. The signal power values calculated at sparse sensor locations are interpolated over the sample by using radial basis function networks. The impact coordinates on the specimen surface are estimated by a center-of-gravity method based on the interpolated power values. Experimental tests were performed by using both an instrumented impact hammer and a drop tower. The results obtained showed the validity of the presented approach, which was able to identify the impact locations with high level of accuracy.

Smart composite structures, which are able to modify their mechanical properties with respect to their environment (e.g. active vibration control), to interact with other structures (e.g. mechatronic) or with human beings (e.g. Human-Machine Interaction), are widely used in the modern industrial fields (e.g. aerospace), due to the intensification of the operational dynamic environment and an increase of durability requirements from the customers. Conventionally, the piezoelectric transducers are glued on the surface of the structure and the power and control electronics are away. To protect the transducer elements and their connections and develop some industrially products in "plug and play" mode, a smart composite structure is designed and manufactured in our lab, a wide distributed network of piezo ceramics elements has been integrated into the heart of the composite during the manufacturing process of composite structures. To meet the technical specifications of smart composite structures, in particular for complex geometries, it is necessary to master the manufacturing process and consequently the material parameters of the manufactured composite. Indeed, during the preliminary design phase, these parameters have to be absolutely known. A design approach based on engineering system theory and uncertainty calculation is applied to characterize the smart composite structures manufactured. In this paper, a Time-of-Flight method is developed in order to extract the elastic properties of smart composite structures. This technique is based on the duration measurements of wave propagation with a simple and low-cost experimental setup. Integrated piezoelectric transducers are used as both transducer and actuator. This method operates the intrinsic abilities of smart composite structures, it is much easier and faster than the model techniques, which are widely used nowadays. Especially for Poisson’s Ratio, this method can extract this parameter rapidly without any complex numerical model by analyzing the phase changes of output signals. In fact, the received waveform contains two types of waves, symmetric and antisymmetric, the elastic properties can be directly calculated based on plate wave propagation theory. In this research, a set of plates with piezo ceramic on each corner are manufactured, the elastic properties of the chosen material are accurately known. Then the Time-of-Flight method will be used for extracting the elastic properties of the plates. A series of sinusoidal burst signals (with different cycles) were chosen as input signals, the idea is to identify the parts where they are in phase from the superposition of the output signals. From the first part of the output signals, there were two areas in phase, they were assumed to be the responses of S0 wave and A0 wave, then the Passion’s Ratio is calculated. At the end, the test results and the known parameters of each plate were compared and discussed.

The localization of structural defects is of great interest in structure health monitoring (SHM). While acoustic emission signals are collected in the practice of SHM, the acquired waveforms inevitably include direct wave as well as reflection and reverberation waveforms. The direct wave actually contains more straightforward information in localizing the sources, so in this work, a deep recurrent denoising autoencoder (DRDA) network is developed. In general, waveform signals are highly correlated at different timescales, so temporally recurrent connections are added to the network structure, which have the memory of recent inputs. Consequently, the proposed DRDA model captures the dependencies across data points, while carrying out denoisng process, and combines the advantages of denoising autoencoders and recurrent neural networks. As the output of the proposed DRDA, direct waveforms are extracted and validated through finite element simulations. A contrived structure with nontrivial shape is excited by simulated pencil break excitations under the ABAQUS environment, then the simulated responses provide training data for the DRDA. The proposed algorithm is effective in filtering the reflected wave and outperforms the conventional denoising autoencoders.

The paper provides an engineering analysis approach for estimating non-destructive evaluation (NDE) flaw size using smaller number of hit-miss data-points. Probability of detection (POD) analysis of inspection test data provides a flaw size, denoted as a_90/95. The flaw size has 90% POD and minimum 95% confidence. In this approach, we do not estimate a_90/95 flaw size. Here, goal is to estimate the flaw detectability size that is larger than the unknown a₉₀ for the NDE technique. The approach is devised using a large set of simulated signal response data and estimating population a_90/95p using an â versus a POD model. The signal response data is converted to hit-miss data using a decision threshold. A hit-miss or â versus a POD model is assumed here. Then a sampling scheme is set-up for generating limited number of data-points and flaw size estimation equations are established by substituting POD model quantities by their approximate estimates based on demonstrated flaw sizes. The data sampling scheme used here is like a Monte Carlo simulation allowing evaluation of engineering estimation approach. The distribution of multiple engineering flaw size estimates provides a confidence on the engineering estimate being larger than a_90/95p used in the simulation. NDE practitioner normally uses some rule of thumb to estimate flaw size based on limited flaw detection data. A typical rule of thumb may be to multiply the smallest detected flaw size by a factor such as two. The rule of thumb is subjective. Compared to the rule of thumb, paper provides more complex equations and demonstrates these equations on simulated randomized data. NDE flaw detection size is used in fracture mechanics analysis of aerospace structures. Fracture mechanics uses flaw growth analysis using NDE flaw size to determine service life of the part. The part design is optimized for performance, weight, service life and cost. In order to tolerate larger NDE flaw size, parts are designed to be stronger and heavier to reduce stress. Therefore, a_90/95 flaw size is desirable. But many times, POD demonstration, requiring a large number of real known size flaws, is cost prohibitive. Thus, if we can estimate risk or confidence associated with the engineering estimate of flaw detectability size, program managers may accept the engineering approach. Here, we need to assess confidence that engineering estimate of flaw size is greater than or equal to estimated a^90/95p of the simulated data. Also, we need to assess smallest possible value of the flaw size estimate and compare it with the estimated a_90/95p. Lastly, the probability of false positive (POF) also needs to be assessed. A technique is considered reliable, if it provides flaw detectability size with 90/95 POD/confidence and also provides a POF less than or equal to a chosen value i.e. 1%. Linear correlation is used between the signal response data and flaw size. POD software mh1823 uses generalized linear model (GLM) in POD analysis after transforming the flaw size and signal response, if needed, using logarithm. Therefore, this simulation approach is in agreement with the linear signal correlation used in mh1823. The approach is conservative and is designed to provide a larger flaw detectability size compared to POD approach with high confidence and takes into account number of data-points used. Such NDE flaw detectability size estimation approach, although, not as rigorous as POD, can be cost effective if the larger estimated flaw size can be tolerated.

As standard method for structural health monitoring (SHM) the electro-mechanical impedance method evaluates the frequency response of a piezoelectric transducer, which is attached to a mechanical structure of interest. The piezoelectric element is excited by a harmonic voltage signal, which causes typically harmonic oscillations on both element and structure. The measured impedance of the piezoelectric element reflects thus the structural response. Consequently, changes of the impedance indicate structural changes, i.e., damage. This contribution investigates linear and possibly non-linear vibrations provoked by contact acoustic non-linearity of a sub-surface crack, a damage typical for composite delamination, in a harmonically excited structure. The considered structure is an aluminum beam with a sub-surface crack, which is introduced artificially according to a specific manufacturing process developed at the authors research group and already presented at SPIE 2018. Numerical studies presented at IWSHM 2017 and SPIE 2018 showed that the considered damage causes non-linear response to harmonic excitation. The proposed work continuous this research by experimental measurements of the vibration response of the considered beam with sub-surface crack to harmonic excitation by a piezoelectric transducer. Laserscanning Vibrometer measurements along the entire beam and in particular at the crack location identify linear and non-linear vibrations, allow its mode shape visualization and to assign the structure as source for nonlinearity. Furthermore, a piezoelectric transducer that simultaneously records the transverse frequency response functions passing through the sub-surface crack shows high potential for vibration-based SHM methods like the electro-mechanical impedance method to assess the non-linear response of this damage type for identification.

Nondestructive testing (NDT), which is applied to the manufacturing/installation of power generation facilities and preventive maintenance, is one of the key element technologies that are essential for "safety and quality assurance" as diagnosis and maintenance technology. Radiography testing (RT) is the main testing method; however, ultrasonic testing (UT) is applied together with RT in most cases. This significantly impacts process delays and economic losses. Therefore, in the industrial field, the introduction and utilization of phased array UT (PAUT) have been aggressively demanded as a new nondestructive volumetric examination technology that can reduce the dependency on RT and possibly replace it. PAUT not only makes it easier to test specimens with complex shapes that are difficult to test with conventional UT, but it also enables high-speed testing because it can electronically control the scanning of the ultrasonic beam. However, in the case of applying PAUT to the testing of a weld zone with a curved shape, such as a piping elbow weld zone, it is not easy to secure the objective and achieve reliable test data because contact with the probe is difficult. Therefore, in this study, flexible PAUT probe was developed that was capable of testing piping elbow weld zones, and its effectiveness was verified by comparing it with the use of a wedge, which is normally used for the testing of pipes with curved structures. Welding specimens were fabricated (including natural defects) in accordance with ASME standards, and artificial notches were machined. The possibility of overcoming the application limit of the current PAUT technology and further improving the reliability of the inspection results were confirmed through signal acquisition and an analysis according to the probe contact position (elbow or pipe surface).

Acoustic emission characteristics according to variety methods for joining PVC pipe during tensile test were investigated. AE amplitudes, the number of AE hits and AE energy were compared through the nonparametric statistics of MannWhitney method. 25A PVC-C pipe and four types of joining were used in this study: only insert, welding, bonding and bonding with welding. The average of the number of AE hits for bonding specimen was 416.2 and for bonding with welding specimen was 522.2. However the result of Mann-Whitney nonparametric static indicated there is no difference (p=0.2963). The average of cumulative AE energy for bonding specimen was 47.8 × 10³ and for bonding with welding specimen was 662.5 × 10³. Therefore the result of Mann-Whitney nonparametric static indicated there is significant difference (p=0.0122). In this study, the analysis of AE signal characteristics confirmed that the tensile damage behavior varies depending on the bonding method of PVC pipe and socket.

Temperature effect is one of the most significant and negative effects on bridges, even worse for long-span bridges. In this study, numerical method for temperature-induced structural strains analysis based on a long-span suspension bridge is investigated. The finite element (FE) models for transient thermal analysis and structural analysis of the long-span suspension bridge are developed, respectively. The variations and distributions of structural temperatures are calculated by applying the thermal boundary conditions on the thermal FE models. Then, structural temperatures are loaded on the structural FE models for structural analysis to obtain the structural strains. The temperature-induced strains of box girder, main cables and towers of the suspension bridge are calculated and analyzed. The results indicated that the temperature effects on the main components of suspension bridge are significant. The structural temperature variations exactly explicate the changes of environmental conditions. The strains of temperature effects not only caused by temperatures of itself, but also the impact of other components. This numerical method can conveniently and effectively calculate the structural temperatures and temperature-induced strains of suspension bridge.

This paper analyzes the working principle, advantages and disadvantages of traditional Boost DC-DC converter firstly, and then proposes a new type of high set-up ratio DC-DC converter topology with the coupled-inductor and voltage-doubler structure. In the following parts, it introduces the composition and the equivalent circuit of this topology in detail, and deduces its set-up ratio expression and the voltage stress expression of the switching tubes and diodes. Finally, this paper designs the main parameters and the relevant models of the topology, and makes the experimental platform. The experiment results indicate that this topology could realize high set-up ratio even under usual duty-ratio conditions, and could effectively reduce the voltage stress and switching-off loss of switching tubes, providing favorable conditions for realizing multilevel conversion of the following inverter in the photovoltaic power generation system.

The wood elastic constants are of great importance in the design of structural elements. In particular, the elastic modulus in the longitudinal direction is often used for strength estimation and grading of wood. Ultrasonic techniques are commonly applied to determine elastic parameters in laboratory experiments and for condition assessment of existing wooden structures. In the present study, the elastic parameters (elastic modulus and shear modulus) for aged and un-aged Mexican pine (Pinus strobus) wood were measured by the ultrasonic emission-transmission technique. The experimental measurements were carried out using longitudinal and shear ultrasonic transducers with a central frequency of 1 MHz and 0.5 MHz respectively. Ultrasonic data and scanning electron microscopy (SEM) were performed, establishing a direct correlation with the measurements of the longitudinal and shear ultrasonic velocities and the elastic properties developed on the naturally aged and un-aged wood. Finally, the elastic and shear modulus of the wood before and after aging conditions were determinate. Therefore, the obtained experimental results not only contribute to the existing knowledge about aged wood but provide useful information about the possibilities of re-using wood as well.

The density of wood is one of the most important physical properties when it comes to understanding its mechanical behavior. The strength of a wood specimen is directly related to the amount of wood material in a given volume, making the accurate determination of density essential for the analysis of wood structures. The use of non-destructive evaluation (NDE) techniques (e.g., microwave/radar, ultrasonic, stress wave, and X-ray) is a powerful approach for condition assessment of existing wood structures. Synthetic aperture radar (SAR) imaging, with its remote and subsurface sensing abilities, provides information about the mechanical properties of wood structures without obstructing their functionality. The objective of this paper is to use SAR imaging to determine the differences in density in a variety of different wood species. Five 14 in.-by-2 in.-by-0.75 in. wood specimens were manufactured. Each wood specimen was imaged vertically inside an anechoic chamber using a 10 GHz SAR system. It was found that SAR amplitude distribution was affected by the density of wood specimens. It was also found that the increase of wood density leads to the increase of contour area of SAR images.

Accurate determination of anisotropic thermal properties is important for the assessment of the structural integrity of composite members. While the through-thickness thermal conductivity is easily determined using thermography, the inplane thermal conductivity is difficult to measure. This study deals with a new approach of determining the lateral thermal conductivity of carbon/epoxy composites laminates by analyzing the heat flow around a circular shadow in flash thermography. Experiments were performed using quasi-isotropic and unidirectional laminate to record transient heat flow pattern during the cooling period using flash thermography. The experiments were also simulated using finite element analysis for a range of lateral diffusivity values. Data thus obtained from experimental and numerical simulation were normalized using pre-established parameters and contour plots of temporal and radial variation for various thermal parameters around the shadow were obtained. The contour plots thus obtained from experimental analysis were then correlated with those from numerical analysis to determine actual in-plane diffusivity. Furthermore, correlation using logarithmic temperature gradient was observed to give relatable and efficient correlation factor than other parameters.

Wind and solar power generation have nearly become synonymous with green energy in recent years. Vibration analysis is crucial to the structural integrity of wind turbines, especially for those exposed to tropical storms and salt corrosion in the coastal environment. In this study, a one-dimensional linear continuous model is applied to analyze the spectral characteristics of the supporting tower of wind turbines. Simulated frequency-domain displacements were obtained for the vibration of the supporting tower subjected to the reduction in local stiffness, representing various cases of tower defects. To identify the defected section of the supporting tower, the unit angle of each section is used as the indicator of the defect that can be obtained using a signal processing technique developed in this study. The results of the numerical analysis show that the method can effectively expose the defect location Furthermore, the progression of the existing defect and the difference between the existing defect and the new one can also be observed. The results of this study demonstrate a great potential for future applications. In practice, the possible variation of the tower structure can be obtained by comparing the measured signal results at different periods. This would allow for maintenance to target precisely at the defected areas so as to achieve efficient allocation of maintenance funds.