The focus of this study was to apply a robust inspection technique for monitoring damage nucleation and propagation in 7075 aluminum alloy specimens exposed to cyclic loading. A previously developed specimen, linearly tapered in width along the length, was subjected to a sinusoidal tension-tension load while conductivity and strain were measured in-situ. Ex-situ measurements of modulus, hardness, surface potential, digital image correlation strain field, and neutron diffraction were made as a function of fatigue cycles. It is hypothesized that varying levels of induced stress along the length due to equal-force but varying area along the length will create a record of damage which can be probed to intuit a temporal history for the specimen. Baseline, intermediate, and failure sensor measurements for several specimens were compared and analyzed as a function of applied stress (varied linearly along the length) and fatigue cycles (constant). Mechanisms of damage nucleation and propagation due to fatigue cycling are discussed with an emphasis on which inspection methods are most promising for improving structural durability and state monitoring.

Bridge management and maintenance work is an important part for the assessment the health state of bridge. The conventional management and maintenance work mainly relied on experienced engineering staffs by visual inspection and filling in survey forms. However, the human-based visual inspection is a difficult and time-consuming task and its detection results significantly rely on subjective judgement of human inspectors. To address the drawbacks of human-based visual inspection method, this paper proposes an image-based comprehensive maintenance and inspection method for bridges using deep learning. To classify the types of bridges, a convolutional neural network (CNN) classifier established by fine-turning the AlexNet is trained, validated and tested using 3832 images with three types of bridges (arch, suspension and cable-stayed bridge). For the recognition of bridge components (tower and deck of bridges), a Faster Region-based Convolutional Neural Network (Faster R-CNN) based on modified ZF-net is trained, validated and tested by utilizing 600 bridge images. To implement the strategy of a sliding window technique for the crack detection, another CNN from fine-turning the GoogLeNet is trained, validated and tested by employing a databank with cropping 1455 raw concrete images into 60000 intact and cracked images. The performance of the trained CNNs and Faster R-CNN is tested on some new images which are not used for training and validation processes. The test results substantiate the proposed method can indeed recognize the types and components and detect cracks for a bridges.

Knitted Textiles made from Nickel-Titanium (NiTi) shape memory alloy wires are a new structural element with enhanced properties for a variety of applications. Potential advantages of this structural form include enhanced bending flexibility, tailorable in-plane, and through-thickness mechanical performance, and energy absorption and damping. Inspection of the knit pattern reveals a repeating cell structure of interlocking loops. Because of this repeating structure, knits can be evaluated as cellular structures that leverage their loop-based architecture for mechanical robustness and flexibility. The flexibility and robustness of the structure can be further enhanced by manufacturing with superelastic NiTi. The stiffness of superelastic NiTi, however, makes traditional knit manufacturing techniques inadequate, so knit manufacturing in this research is aided by shape setting the superelastic wire to a predefined pattern mimicking the natural curve of a strand within a knit fabric. This predefined shape-set geometry determines the outcome of the knit’s mechanical performance and tunes the mechanical properties. In this research, the impact of the shape setting process on the material itself is explored through axial loading tests to quantify the effect that heat treatment has on a knit sample. A means of continuously shape setting and feeding the wire into traditional knitting machines is described. These processes lend themselves to mass production and build upon previous textile manufacturing technologies. This research also proposes an empirical exploration of superelastic NiTi knit mechanical performance and several new techniques for manufacturing such knits with adjustable knit parameters. Displacement-controlled axial loading tests in the vertical (wale) direction determined the recoverability of each knit sample in the research and were iteratively increased until failure resulted. Knit samples showed recoverable axial strains of 65–140%, which could be moderately altered based on knit pattern and loop parameters. Furthermore, this research demonstrates that improving the density of the knit increases the stiffness of the knit without any loss in recoverable strains. These results highlight the potential of this unique structural architecture that could be used to design fabrics with adjustable mechanical properties, expanding the design space for aerospace structures, medical devices, and consumer products.

With the rapid development of rail traffic, the importance of railway overhaul is becoming increasingly prominent. Making an inventory on tools is an important step that railway workers must take before and after railway inspection. The tools left on the railway will cause great harm to train safety. To avoid this happening, the commonly used method is manual inventory at present, which is time-consuming, laborious and easily leads to omissions. In order to overcome these shortcomings, this paper proposes a Faster Region-based Convolutional Neural Network (Faster R-CNN)-based method for tools inventory. To realize the method, a Faster R-CNN architecture based on ZF-Net is modified and a database including a large number of images for 10 types of tools is built. Then the Faster R-CNN is trained and validated using the built database. The performance of the trained Faster R-CNN is evaluated using some new images which are not be used for training process. The result shows 95.0325% average precision (AP) ratings for 10 different types of tools and proves the proposed method is effective.

Damage nucleation and growth can be complex in hybrid structures composed of layers of metal and laminated composites. Presently there are limited reliable damage growth analytical and empirical methods to evaluate the bond integrity of such structures and to quantify the state of bonding in such joints. Depending on the geometry and accessibility of hybrid joints, ultrasonic nondestructive testing (NDT) techniques are available for inspection of these structures. However there are some limitations for the usage of typical bulk or guided waves to quantify the integrity of bondline in hybrid structures. This work suggests the use of specific forms of ultrasonic guided waves that propagate along the bondline of these hybrid structures. This study is dedicated to modeling of interface guided waves for the purpose of disbond crack damage assessment. The nature of interface waves is discussed and the numerical simulation based on the material properties and geometries of hybrid interfaces as well as composite stacking sequence is verified. A finite element model of a hybrid structure with isotropic and anisotropic multilayer composites is constructed. The behavior of interface guided waves influenced by disbond cracks at free edges of hybrid bonded joints is numerically studied. The propagation characteristics of interface waves is shown to be sensitive to the size of disbond cracks. The velocity of interface waves is shown to have an inverse relation to the disbond damage size. Results show the speed is also a function of the interfacing ply orientation at the bondline. These results suggest that interface waves can be used to monitor the condition of bonded joints in hybrid structures.

For the traditional inspection methods, the visual inspection data is firstly recorded on the inspection forms and then input manually into computer, which is inefficient and creates errors frequently. This research aims at establishing a smartphone-based bridge inspection and management system that can avoid such inputting errors and facilitate the bridge inspection process. The system enables the inspector to complete the inspection information collection in a portable smart phone. The site photos that related to the investigated structures can be easily added and edited during the inspection work with the help of the smart phone. After the investigation, the inspection report and the technical condition rating of the inspected bridge can be automatically generated. The collected data and the GPS information can be uploaded to the terminal server directly via the mobile network. The interface of the mobile software is user-friendly and easy operation, which provides an opportunity for the public to take part in the bridge inspection work, especially for the bridges in rural and mountainous areas. Then, this paper puts forward the relevant ideas on public participation in bridges’ emergency assessment and disposal after the disaster, which can provide data support for the decision-making and disaster relief work.

This conceptual research focuses on identifying damage precursors in carbon fiber reinforced polymer laminates. By coupling integrated in situ sensing elements (magnetostrictive particles) and multi-scale ultrasonic inspection (Quantitative Ultrasonic Imaging) with minimum hardware the authors are able to capture and gain an understanding of damage precursors. Preliminary damage precursors can be identified and quantified and correlated to a particular failure mode(s) and/or fabrication process accelerating the need for a sensing/repair strategy that could be implemented to capture the specific precursor prior to the onset of cracks. Capturing and mitigating damage precursors ahead of micro-crack formulation will slow the evolution of the damage precursors to micro-cracks, mitigate adverse loading environment, and develop techniques to reduce stress and loading below the endurance limit to keep it from being fatigued.

This paper discusses theoretical and experimental analyses of the standing harmonic waves through the electro-mechanical impedance spectroscopy (EMIS) and guided surface acoustic waves (SAW) through the guided wave propagation (GWP) analyses. Both EMIS and GWP analyses have been carried out by utilizing piezoelectric wafer active sensors (PWAS) for in situ structural inspection. PWAS has recently been extensively employed in many applications such as nuclear-structural as well as aero-structural health monitoring and non-destructive evaluations (NDE). EMIS method is utilized for high frequency local modal sensing to determine the dynamic characteristics of PWAS bonded on nuclear-structural component for in-situ ultrasonics. Rayleigh waves a.k.a., SAW, were generated in relatively thick isotropic elastic plates. Rayleigh waves have the property of propagating close to the plate surface, with rapid attenuation with depth. The polarization of Rayleigh waves lies in a plane perpendicular to the surface so that the effective penetration depth is less than a wavelength. Rayleigh waves are a high frequency approximation of the first symmetric (S 0 ) and anti-symmetric (A 0 ) Lamb wave modes. As the frequency becomes very high the S 0 and the A 0 wave speeds coalesce, and both have the same value. This value is exactly the Rayleigh wave speed and becomes constant along the frequency. In the first part of the study, simplified theoretical constrained PWAS-EMIS model is briefly discussed in relatively high frequency range (in MHz order of magnitude) in terms of thickness mode. Analytical predictive thickness mode impedance simulations of PWAS bonded on plate-like host structures are presented in corresponding with the experiments. For the experimental analyses, PWAS transducers are affixed on isotropic elastic plates such as aluminum plate in relatively high thickness and on a rail I-beam. The extent of the agreement between the experimental and analytical EMIS analyses of PWAS in thickness mode is presented. The study is followed with GWP tests through the pitch-catch method. Rayleigh wave signal packets which are generated in the relatively thick plate and a rail I-beam in high frequency region are assessed along with the experimental thickness mode PWAS-EMIS results. The tuning curve of Rayleigh wave is determined to show the tuning effect of the structure thickness on producing a dominant Rayleigh wave mode. The significant usage of the tuned Rayleigh wave mode is essentially discussed for the applications in the in-situ inspection of relatively thick structures such as nuclear power plant structures. The paper ends with summary, conclusions and suggestions for future work.

In the United States, many civil, aerospace, and military aircraft are nearing the end of their service life. Many of these service life predictions were determined by models that were created at the time of the design of the structure, possibly decades ago. As a precaution, these structures are inspected on a regular basis with techniques that tend to be expensive and laborious, such as tear-down inspections of aircraft. To complicate matters, new complex materials have been incorporated in recent structures to take advantage of their desirable properties, but these materials sustain damage in a manner that is different from that of past monolithic materials. One example is fiber-reinforced polymer (FRP) composites, which are heterogeneous, direction-dependent, and tend to manifest damage internal to their laminate structure, thus making the detection of this damage nearly impossible. For these reasons, numerous groups have focused on developing sensors that can be applied to or embedded within these structures to detect this damage. Some of the most promising of these approaches include using piezoelectric materials as passive or active ultrasonic sensors and actuators, fiber optic-based sensors to measure strain and detect cracking, and carbon nanotube-based sensors that can detect strain and cracking. These are mostly point-based sensors that are accurate at the location of application but require interpolative methods to ascertain the structural health elsewhere on the structure. To conduct direct damage detection across a structure, we have coupled the ability to deposit a carbon nanotube thin film across large substrates with a spatially distributed electrical conductivity measurement methodology called electrical impedance tomography. As indicated by previous research on carbon nanotube thin films, the electrical conductivity of these films changes when subjected to strain or become damaged. Our structural health monitoring strategy involves monitoring for changes in electrical conductivity across an applied CNT thin film, which would indicate damage. In this work, we demonstrate the ability of the Electrical Impedance Tomography (EIT) methodology to detect, locate, size, and determine severity of damage from impact events subjected to glass fiber-reinforced polymer composites. This will demonstrate the value and effectiveness of this next-generation structural health monitoring approach.

Wireless ultrasound inspections using Piezoelectric Wafer Active Sensors (PWAS) are attractive for Structural Health Monitoring (SHM). However, the impedance mismatch between the PWAS and the wireless transponder reduces the wirelessly transmitted signal strength. Electrical Impedance Matching (EIM) circuit can be introduced to maximize the power transmission between the PWAS and the wireless transponder. This paper discusses the wireless excitation of ultrasound as well as the design, simulation, and characterization of the EIM networks for PWAS. To maximize power transmission, a two port EIM network was developed using a computerized smith chart. The equivalent circuit of the PWAS and the EIM network were then combined to establish the equivalent circuit of the matched transducer. Computer simulations were carried out to evaluate the gain, the bandwidth, and the sensitivity of the EIM networks. Two-port EIM networks were implemented for both the actuator and the sensor in an ultrasound pitch-catch inspection system. The performance of the pitch-catch systems with and without the EIM networks was compared. Detailed analysis, simulation, hardware implementation, and measurement results are presented.

Several ensembles of sweeping diagnostic impulsive forces were measured in a unidirectional carbon-epoxy composite beam modified locally with a soft viscoelastic patch. The spatial uniformity of the typical ensemble of diagnostic signals is addressed by a systematic spatio-temporal coherence analysis in terms of proper orthogonal decomposition (POD) modes. All samples of spanning ensembles are strongly dominated by the same POD mode characterized by a nearly uniform spatial modulation and a sharp triangular pulse time modulation. The higher POD modes have small amounts of energy. They possess an important statistics property: their spatial modulation mean value is nearly zero with standard deviation nearly identical to the nearly uniform value of the dominant POD mode. The nearly uniform spatial distribution of the dominant POD mode is a fuzzy picture of the ideal or nominal one where the impact-generated diagnostic forces should have a time waveform independent of the site of impact. Despite this energy content deficiency, the ensemble of acceleration signals acquired at a fixed point while the beam is excited by an ensemble of sweeping diagnostic forces has very robust POD modal structure. The POD modes show in a clear manner the presence of a soft viscoelastic patch simulating mass modifications. The POD-based coherence analysis of ensembles of diagnostic forces generated in this practical problem is potentially useful for a real-time verification-inspection of the integrity of networks of embedded and surface-mounted actuators and sensors.

In this paper, the detection for two kinds of cracks is studied: (1) linear notch crack; (2) nonlinear breathing crack. A pitch-catch method with piezoelectric wafer actives sensors (PWAS) is used to interrogate an aluminum plate with a linear notch crack and a nonlinear breathing crack respectively as two cases. The inspection Lamb waves generated by the transmitter PWAS, propagate into the structure, interact with the crack, acquire crack information and are picked up by the receiver PWAS. The linear notch crack case is investigated through: (1) analytical model developed for Lamb waves interacting with a general linear damage; (2) finite element simulation. The breathing crack, which acts as a nonlinear source, is simulated using two approaches: (1) element activation/deactivation technique; (2) contact model. The theory and solving scheme of the proposed element activation/deactivation approach is discussed in detail. The signal features of different damage severities are analyzed. Crack opening, closing, stress concentration, surface collision phenomena are noticed for the breathing cracks. Mode conversion is noticed for both crack cases. The generation mechanism and mode components of the new wave packets are investigated by studying the particle motion through the plate thickness. A damage index is proposed based on the spectral amplitude ratio between the second harmonic and the excitation frequency for the breathing crack. The damage index is found capable of estimating the presence and severity of the breathing crack. The paper finishes with summary and conclusions.

Fiber reinforced composites (FRP) for industrial applications face constantly increasing demands regarding efficiency, reliability and economy. Furthermore, it was shown that FRP’s with tailored reinforcements are superior to metallic or monolithic materials. However, a trustworthy description of load-specific failure behaviour and damage evolution of composite structures can hardly be given, because these processes are very complex and are still not entirely understood. Amongst other things, several research groups have shown that material damages like fiber fracture, delamination, matrix cracking or flaws can be discovered by analyzing the electrical properties of conducting composites, e.g. carbon fiber reinforced plastics (CFRP). Furthermore, it was shown that this method could be used for structural health monitoring or non-destructive testing (NDT) [8–12].Within this work, Magnetic Induction Tomography (MIT), which is a new imaging approach, is introduced into the topic of NDT of CFRP’s. This non-contacting imaging method gains the inner spatial distribution of conductivity of a specimen and depicts material inhomogeneity, like damages, in 2D or 3D images. Numerical and experimental investigations are presented and give a first impression of the performance of this technique. It is demonstrated that MIT is a promising approach for NDT and could be used for fabrication quality control of conductive FRP’s and could potentially be used as a health monitoring system using an integrated setup.

The cost of corrosion to the US military is huge. In 2006–2007 the US Navy spent $3.2 Billion on corrosion related maintenance on ships alone. Additional funds were expended maintaining US Navy aircraft, helicopters and land vehicles. The cost of corrosion must be controlled. It is apparent that extraordinary cost reductions are needed to bring the problem under control. In order to achieve such dramatic improvements changes from conventional methods of construction, inspection and monitoring are necessary. There are ample opportunities for active material technologies if the researcher understands the fundamental corrosion problem. In this paper the author outlines information on US Navy corrosion concerns including examples of recent request for information data calls. It is hoped that this paper will help researchers in the SMASIS community understand the needs associated with US naval corrosion.

Research at the AFRL Space Vehicles Directorate is being conducted to reduce schedule times for assembly, integration, and test, to make satellite-based capabilities more responsive to user needs. Structural Health Monitoring has been pursued as a means for validating workmanship and has been proven on PnPSat-1. Embedded ultrasonic piezoelectric wafer active sensors (PWAS) have been utilized with local and global inspection techniques, developed both in house and by collaborating universities, to detect structural changes that may occur during assembly, integration, and test. Specific attention has focused on interface qualification. It is now reasonable to believe that evaluation of interfaces through the use of such sensors can also be used to indirectly qualify the structure thermally and that tedious thermal-vacuum testing may be truncated or eliminated altogether. This paper focuses on the computational development of extracting thermal properties from ultrasonic transmission records. Methods are validated on simple bolted lap-joint cantilever beams.

This paper presents a microwave antenna sensor that has the ability to monitor fatigue cracks between lap joints where compressive force is present or in any other areas where a crack is hidden between two parts of a structure. The sensor is designed to be low profile, light weight, and conformal, making it ideal for aerospace applications where lap joints are a common design feature, and thus minimizing manual inspection times. The design of the fatigue sample for simulating a lap joint will be presented. The experiment results validated the microwave antenna sensor’s capability to effectively monitor fatigue cracks under lap joints.

Research interests in structural health monitoring have increased due to in-situ monitoring of structural components to detect damage. This can secure personal safety and reduce maintenance effort for mechanical systems. Conventional damage detection techniques known as nondestructive evaluation (NDE) have been conducted to detect and locate damaged area in structures. Ultrasonic testing, using ultrasonic transducers or electromagnetic acoustic transducers, is one of the most widespread NDE techniques, based on monitoring changes in acoustic impedance. Although the ultrasonic testing has advantages such as high sensitivity to discontinuities and evaluation accuracy, it requires testing surface accessibility, close location to the damaged area, and decent skill and training of technicians. In recent years, modal analysis techniques to capture changes of mode shapes and natural frequency of structures have been investigated. However, the technique is relatively insensitive to small amount of damage such as an initial crack which can rapidly grow in structures under cyclic loadings. In addition, structural health monitoring based on guided waves has become a preferred damage detection approach due to its quick examination of large area and simple inspection mechanisms. There are many techniques used to analyze sensor signals to bring out features related to damage. A phased array coupled with the guided wave approach has been introduced to effectively analyze complicated guided wave signals. Phased array theory as a directional filtering technique is usually used in antenna applications. By using phased array signal processing, virtually steering the array to find the largest response of source, the desired signal component can be enhanced while unwanted information is eliminated.

Ultrasonic guided waves, due to theirs capability of interrogating a large structure from a single sensor position, has been proven as a promising tool for structural health monitoring (SHM). In this paper, we present two imaging approaches of utilizing guided wave leave-in-place sensors for early detection of defects in plate-like structures as well as for monitoring the defect growth. The first approach is based on a guided wave tomographic technique, in which the region surrounded by a sparse sensor array is monitored. The second one is a phased array approach, in which sensors are attached to a structure in a compact format to form an array. The region subjected to inspection and monitoring is the region outside the array. Both techniques have shown excellent capability of determining damage size, location, and severity.