Search

Fiber Bragg grating (FBG) sensors have been widely applied for structural health monitoring applications. In some applications, remote bonding of the optical fiber is applied, where ultrasonic waves are coupled from the structure to the optical fiber and propagated along the fiber to the FBG sensor. The distance that this signal can propagate along the optical fiber without decaying below a threshold value can be critical to the area of the structure that can be monitored per sensor. In this paper, we develop a method to measure the acoustic mode attenuation of fiber waveguides based on laser Doppler vibrometry (LDV) that is independent of the fiber type. In order to validate the method, we compare attenuation measurements on single-mode optical fibers using both the LDV and FBG sensor methods. Once the method is validated, experimental measurements of different coated and uncoated optical fibers are performed to quantify the role of the fiber diameter on the attenuation coefficient. As the radius of the waveguide decreases, the signal attenuation increases exponentially.

Guided waves (GW) allow fast inspection of a large area and hence have received great interest from the structural health monitoring (SHM) community. Fiber Bragg grating (FBG) sensors offer several advantages but their use has been limited for the GW sensing due to its limited sensitivity. FBG sensors in the edge-filtering configuration have overcome this issue with sensitivity and there is a renewed interest in their use. Unfortunately, the FBG sensors and the equipment needed for interrogation is quite expensive and their number is restricted. In the previous work by the authors the number and location of the actuators was optimized for developing a SHM system with single sensor and multiple actuators. But through the use of the phenomenon of acoustic coupling, multiple locations on the structure may be interrogated with a single FBG sensors. As a result, a sensor network with multiple sensing locations and few actuators is feasible and cost effective. Hence this paper develops the optimization problem for designing an SHM network for use with FBG sensors making use of acoustic coupling. The optimization problem is implemented on a simple aluminum plate. The directionality, bond efficiency and the factors influencing the acoustic coupling are taken into consideration for optimizing the sensor network. A multi-objective optimization problem is defined and solved using non-sorting genetic algorithm (NSGA). The results indicate that indeed a multi-objective optimization is necessary and has potential to improve the SHM system performance.

In the design of hybrid unmanned aerial and underwater vehicles, buoyancy management and weight are two major factors. Large wing volumes used by unmanned air vehicles to fly efficiently drive vehicle buoyancy up, preventing them from submerging. Heavy active buoyancy control systems can overcome this, but cost weight, energy, and time to transition between underwater operation and flight. An alternative design, consisting of a passively flooding and draining wing, is presented in this paper. Relevant dynamic parameters for a full vehicle dynamic model are identified. A dynamic model of a draining structure is developed and verified experimentally on both a simple cylinder and a full wing structure. With proper tuning, the model captures the salient dynamic behavior of passive draining during vehicle egress. A prototype unmanned aerial and underwater vehicle was built, flown, and used to collect flight test data. The model is used to accurately predict the takeoff performance of the vehicle. As given, the model can be incorporated into a full vehicle dynamic model to aid in the design, simulation, and control of hybrid unmanned aerial and underwater vehicles with passively draining components.

Previous studies demonstrated the coupling of Lamb waves to guided traveling waves in an optical fiber through an adhesive bond. However, the mode conversion at the specific coupling location has not been experimentally verified. This work measures the mode conversion of ultrasonic waves through an adhesive bond, implementing a high-resolution 3-dimensional laser Doppler vibrometer. Specifically, the propagation of surface guided S0 and A0 modes in a thin plate and their conversion into L01 and F11 modes in an optical fiber are successfully verified. In addition, the conversion of a shear-horizontal mode in the plate into an orthogonal F11 mode is also identified.

Fiber Bragg grating (FBG) sensors are often applied as Lamb wave detectors for structural health monitoring (SHM) systems. Analyzing the measured signal for the identification of structural damage requires a high signal-to-noise ratio (SNR) because of the low-amplitude Lamb waves. This paper applies a two-dimensional ultrasonic horn between the structure and a remotely bonded FBG sensor to increase the amplitudes of the measured signal. Experimentally we test a variety of ultrasonic geometries and demonstrate a 100% increase in the measured ultrasonic signal amplitude using a metallic ultrasonic horn with step-down geometry. A bonding procedure for the combined ultrasonic horn and optical fiber is also developed that produces repeatable signal measurements. For some horn geometries, an additional vibration signal at the Lamb wave excitation frequency is observed in the measurements. Laser Doppler vibrometry (LDV) measurements and finite element analysis demonstrate that the signal is due to the natural vibration of the horn. The experimental results demonstrate that using an aluminum ultrasonic horn to focus wave is an excellent method to increase the sensitivity of the FBG to the small amplitude Lamb wave, provided the horn vibration characteristics are taken account in the design of the measurement system.

Coherence of acoustic wave coupling through adhesive bond coupler is investigated using fiber ring resonator configuration. Coherent interference is observed between different kinds of fibers, opening up possibilities for sensing in field applications.

We demonstrate sensing based on an acoustic fiber ring resonator using optical fiber adhesive coupling for temperature sensing based on in acoustic path length and water level based on acoustic loss per unit length.

We examine the measurement of acoustic mode attenuation in optical fibers for sensing applications using both an internal fiber Bragg grating and laser Doppler vibrometry.

Guided waves (GW) and acoustic emission (AE) -based structural health monitoring (SHM) have widespread applications in structures, as the monitoring of an entire structure is possible with a limited number of sensors. Optical fiber-based sensors offer several advantages, such as their low weight, small size, ability to be embedded, and immunity to electro-magnetic interference. Therefore, they have long been regarded as an ideal sensing solution for SHM. In this review, the different optical fiber technologies used for ultrasonic sensing are discussed in detail. Special attention has been given to the new developments in the use of FBG sensors for ultrasonic measurements, as they are the most promising and widely used of the sensors. The paper highlights the physics of the wave coupling to the optical fiber and explains the different phenomena such as directional sensitivity and directional coupling of the wave. Applications of the different sensors in real SHM applications have also been discussed. Finally, the review identifies the encouraging trends and future areas where the field is expected to develop.

This paper presents a fiber Bragg grating (FBG) interrogator based on vertical surface cavity emitting lasers (VCSEL). A Fabry–Perot filter technique is developed to directly calibrate the dynamic wavelength behavior of VCSELs at both low and high sweep rates. A broad bandwidth light source is constructed by multiplexing five VCSELs together to increase the number of FBGs that can be tracked. Scanning of the VCSEL-array is accomplished by a high-speed electronic switch circuit. The developed interrogator achieves a sweep bandwidth of 10 nm at a scanning rate of 4 kHz. Low-velocity impact testing of a composite plate with a surface mounted FBG sensor shows that the wavelength detection error was 2.7% when compared with the FBG strain obtained by a high-speed swept laser.

This paper presents a method for obtaining time dependent back face deformation (BFD) data for body armor during ballistic impact using a clear ballistics gelatin backing and high-speed cameras to capture the deformation profile. Using this method, baseline fabric characterization data was obtained for samples comprised of varying layers of 467 g/m 2 Kevlar K29 fabric impacted with 8.24 g steel ball projectile and backed with NATO standard 20% clear ballistics gelatin. For these tests, deformation depths were seen to increase with increasing impact energy and decreasing total areal density. A limited study of the various test parameters was performed by testing one additional fabric, projectile, and ballistics gelatin. From these comparisons, it was observed that 122 g/m 2 Kevlar KM2+ fabric performs better per weight than 467 g/m2 Kevlar K29 fabric in terms of BFD, 9 mm FMJ projectiles produce deeper BFDs than 12.7 mm steel ball projectiles, and backing a sample with FBI standard 10% ballistics gel increases the BFD considerably over NATO standard 20% ballistics gel.

Tendon-to-bone insertion tissues may be considered as functionally-graded connective tissues, providing a gradual transition from soft tendon to hard bone tissue, and functioning to alleviate stress concentrations at the junction of these tissues. The tendon-to-bone insertion tissues demonstrate pronounced viscoelastic behavior, like many other biological tissues, and are designed by the nature to alleviate stress at physiological load rates and strains states. In this paper we present experimental data showing that under biaxial tension tendon-to-bone insertion demonstrates rate-dependent behavior and that stress-strain curves for the in-plane components of stress and strain become less steep when strain rate is increased, contrary to a typical viscoelastic behavior, where the opposite trend is observed. Such behavior may indicate the existence of a protective viscoelastic mechanism reducing stress and strain during a sudden increase in mechanical loading, known to exist in some biological tissues. The main purpose of the paper is to show that such viscoelastic stress reduction indeed possible and is thermodynamically consistent. We, therefore, propose an anisotropic viscoelasticity model for finite strain. We identify the range of parameters for this model which yield negative viscoelastic contribution into in-plane stress under biaxial state of strain and simultaneously satisfy requirements of thermodynamics. We also find optimal parameters maximizing the observed protective viscoelastic effect for this particular state of strain. This model will be useful for testing and describing viscoelastic materials and for developing interfaces for dissimilar materials, considering rate effect and multiaxial loadings.

In structural health monitoring (SHM) applications, fiber Bragg grating (FBG) sensors are typically bonded directly to the surface of a structure to detect ultrasonic waves for damage identification. The sensitivity of the bonded FBGs to guided waves varies significantly with input ultrasound wavelength (λ)-to-FBG grating length (L) ratio, i.e., λ/L. Recently, the authors have demonstrated that the detection sensitivity of an FBG can be potentially increased by remotely bonding the optical fiber at a distance away from the FBG, however its response as a function of λ/L has not been studied in detail. In this work, we investigate the ultrasound detection of directly bonded FBG and remotely bonded FBG with varying λ/L. Specifically, we maintain L constant and change λ by varying the ultrasound excitation frequency. Using a 3D laser Doppler vibrometer (LDV) we first characterize input Lamb waves as a function of frequency, which are excited in a thin plate using a broadband transducer. Next, we measure the output response of directly bonded FBG and remotely bonded FBG to the same input Lamb waves. Finally, we examine the output FBG responses normalized with the LDV measurements of input waves, investigating the FBG sensitivity as a function of λ/L ratio. Understanding this sensitivity is important because many guided wave signals, for example generated by acoustic emission, are broadband. Additionally, multiple frequencies are often used for guided wave imaging of structures.

We amplify the signal coupled from ultrasonic Lamb waves into an optical fiber, for structural damage detection applications, using two devices: an ultrasonic horn and a phononic crystal GRIN lens.

This paper analyzes the conversion of acoustic modes in optical fibers from Lamb waves in structures. Experimental verifications of the physical behavior of these modes using micro-laser Doppler vibrometry is also presented.

We demonstrate a self-referencing FBG sensor for damage detection. The sensor compensates for changing environmental conditions by measuring the reference and distorted signal simultaneously and uses traveling acoustic waves in optical fibers.

Ultrasound measurement for damage detection in practical structural health monitoring (SHM) applications is often affected by varying environmental condition. In this case, a baseline reference signal measured under initial conditions may not be valid for comparison with the distorted signal (due to structural damage) that is measured under different conditions. In this study, we investigate a self-referencing ultrasound detection of fiber Bragg grating (FBG) by bonding an optical fiber at two different locations away from the FBG. We first investigate the extraction of ultrasonic waves from two different adhesive bond locations, which are measured with a single FBG sensor located between the two bonds. Based on understanding the ultrasound coupling mechanism through two adhesive bonds, we test the self-referencing ultrasound with the presence of a damage in a structure, examining the output FBG response that contains a combined signal of distorted and reference signals extracted through each bond.

Guided waves (GW) allow fast inspection of a large area and hence have received great interest from the structural health monitoring (SHM) community. Fiber Bragg grating (FBG) sensors offer several advantage but their use has been limited for the GW sensing due to their limited sensitivity. But the use of the edge-filtering approach in the remote bonded configuration has enhanced their sensitivity and allows use of these sensors for GW measurement. Although these sensors are light in weight and need no additional wiring there is still significant motivation to reduce the number of sensors while maintaining the quality and reliability of the assessment due to their high cost. In addition, for the safety and reliability of the structures it is of utmost importance that the entire structure can be investigated. Hence it is necessary to optimize the locations of the sensors in order to maximize the coverage while limiting the number of sensors used. The problem posed for the optimization of the FBG sensors for GW is different from any other work in this area due to the directional sensitivity shown by these sensors. Hence a novel optimization problem is developed for the application of FBG sensors for the GW based SHM. This paper develops a genetic algorithm based optimization methodology to incorporate the directional sensitivity. The methodology is shown analytically, using inputs from experimental investigations.

This paper presents test results and performance characterization of the first fixed-wing unmanned vehicle capable of full cross-domain operation in both the aerial and underwater environments with repeated transition and low-energy loitering capabilities. This vehicle concept combines the speed and range of an aircraft with the persistence, diving capabilities, and stealth of a submersible. The paper describes the proof-of-concept vehicle including its concept of operations, the approaches employed to achieve the required functions, and the main components and subsystems. Key subsystems include a passively flooding and draining wing, a single motor and propeller combination for propulsion in both domains, and aerodynamic–hydrodynamic control surfaces. Experiments to quantify the vehicle performance, control responses, and energy consumption in underwater, surface, and flight operation are presented and analyzed. Results of several full-cycle tests are presented to characterize and illustrate each stage of operation including surface locomotion, underwater locomotion, water egress, flight, and water ingress. In total, the proof-of-concept vehicle demonstrated 12 full-cycle cross-domain missions including both manually controlled and autonomous operation.

For an energy storage application such as electrical vehicles (EVs), lithium-ion batteries must overcome limited lifetime and performance degradation under specific conditions. Particularly, lithiu...

Recent studies demonstrated the potential of increasing the Lamb wave detection sensitivity of fiber Bragg grating (FBG) sensors by bonding the optical fiber away from the grating location, instead of the conventional method of bonding the FBG directly. The FBG located at a remote location further along the optical fiber collects the guided traveling wave in the optical fiber generated from the Lamb wave signal. This remote bonding method could potentially be extended to a series of multiplexed FBGs. However, previous experiments also detected coupling to guided traveling waves in both directions in the optical fiber, which could have significant effects on multiplexed signals. In this paper, we measure the coupled signal amplitudes in both forward and backward directions, when ultrasonic waves couple from a thin plate to an optical fiber and from an optical fiber to a thin plate. The forward- and backward-induced modes are measured in both the optical fiber and the plate. The same experiment is then performed for the case when ultrasonic signal is coupled from the optical fiber to the plate. In addition, two different types of bonding, cyanoacrylate adhesive and frictional bond, are explored to investigate how the signal conversion depends on the bonding method. The results demonstrate that the coupling of ultrasonic waves from a thin structure to an optical fiber and from an optical fiber to the structure is complex. The coupling does not only occur in the direction of the wave propagation, but can be coupled into both forward and backward modes, depending on the bonding configuration used.

Despite demonstrated success in both the laboratory and in the field, significant questions remain unanswered regarding the durability of Carbon Fiber Reinforced Polymer (CFRP) for strengthened concrete members. Little is known about the impact of pre-existing bond defects when subjected to harsh environmental conditions. The results from a study observing the performance of 18 small-scale CFRP-to-concrete pull test specimens is presented herein. Half of the sample set was stored in ambient laboratory conditions while the remaining sample set was subjected to 50 freeze–thaw cycles. Repeated for each condition were 3 specimens prepared without any intentional bond defects and 3 specimens containing 645 mm2 Teflon inserts, for comparison purposes. It was found that, not only did the freeze–thaw conditioning reduce the debonding capacity of the externally bonded CFRP, but that the presence of defects resulted in a greater reduction in debonding capacity following the environmental conditioning, when compared to specimens exposed to ambient conditions only. Nondestructive evaluation of the bonded interface was also performed by using Pulse Phase Thermography (PPT).

In this study, we investigate coupling of acoustic guided waves from different types of input fibers, through a bonded coupler, to an optical fiber. These acoustic waves can then be detected with conventional fiber Bragg gratings (FBGs). The input waves are measured using a high-resolution 3D laser Doppler vibrometer, and the output waves in the optical fiber are measured using an FBG. We demonstrate that the wave coupling between two waveguides varies with the cross-sectional area and the modulus of elasticity of the fibers.

Ultrasonic inspection based structural health monitoring is a powerful technique for damage detection in a structure. Ultrasonic waves are often excited at different frequencies to detect damages of different sizes, therefore understanding the frequency response of the sensor can be used to optimize the sensor performance. Fiber Bragg gratings (FBGs) are widely used for this ultrasound collection. The sensitivity of FBGs to a particular ultrasonic frequency is a function of the FBG length and the ultrasonic wavelength. Recently the authors demonstrated that its ultrasound sensitivity is improved for some conditions when the FBG is bonded at a distance away from the adhesive bond, referred to as remote bonding. However, the frequency response of this configuration has not been studied. Therefore, in this paper we measure and compare the ultrasonic frequency responses between a conventional directly bonded FBG and remotely bonded FBG. In theory, the FBG sensitivity varies as a function of ultrasound wavelength-to-grating length (λ/L) ratio. Therefore, for this experimental study, we maintain L constant and vary λ by changing the frequency of the input ultrasonic waves. We demonstrate that there is a region, below a cut-off values of λ/L, for which the remotely bonded FBG output has a higher sensitivity to the Lamb wave amplitude than the directly bonded FBG. The exact value of this λ/L cut-off depends on the mechanical properties of the structure, the windowing of the input Lamb wave, and the FBG properties. We also demonstrate that windowing the Lamb wave excitation signal has a similar affect to apodizing the FBG sensor in modifying the sensitivity response curve.

Fiber Bragg grating (FBG) sensors can be multiplexed in large numbers to monitor the performance of large structures. This paper addresses the collection of FBG reflection spectra from wavelength division multiplexed sensors at fast acquisition rates. The spectral and temporal resolution is first derived as a function of the tunable filter and measurement system properties. The method is applied to impact loading investigations of a stiffened composite skin panel. The reflected spectrum of each FBG in an array, embedded in the panel, is collected at 100 kHz during the impact events with a spectral resolution down to 40 pm. Visualization of the FBG responses to these impact events, including the presence of spectral distortion in some FBG spectra, is presented. Future analyses based on the full-spectral data sets can enable the assessment of the localized progression of internal damage in such structures.

This paper will present a new testing method to simulate lightning strikes and capture the progressive damage that occurs in carbon fiber polymer matrix systems. The simulated lightning strike contains two major components that must occur simultaneously: a normal pressure pulse that impacts the carbon specimen, representative of the impulse shock wave associated with a lightning strike, and a rapid temperature rise, which represents the thermal shock from the lightning strike. The pressure pulse is applied using a shock tube, a device that forms a controlled shock that will impact the sample. The thermal shock is simulated with a capacitive discharging circuit, in which capacitors are charged to a maximum of 400 V and then, due to the relatively low resistance associated with carbon fiber materials, quickly discharged into the sample. The distance between the brass contacts will be varied, ultimately changing the resistance of the sample, in order to achieve the correct timescale and temperature spike. Various discharge voltages and fiber orientations relative to the direction of the current flow will be tested in order to study the effect these parameters have on the response of the sample. Through the use of Digital Image Correlation and an infrared thermal camera, independent measurements of deformation and temperature, respectively, will be found as the damage progresses, and is to be compared to a network of Fiber Bragg gratings (FBG) sensors applied to the sample. The results will help produce a better understanding of the temperature profile and localizations that occur during lightning strike events.

Soft body armor works by dispersing impact energy across woven fibers. This dispersed impact energy is translated though the armor to the wearer, which can cause behind armor blunt trauma (BABT). Impacts from high caliber rounds can cause complex pressure states in the body, resulting in internal injury and death [1]. Current testing methods rely on using clay-backing materials to determine peak back face deformation (BFD). However, clay testing can provide only the max deformation depth and volume. There is no information on how we get to that state, how quickly the event occurs. Understanding the live dynamics of BFD formation will aid in improving both armor evaluation and the prevention of blunt force trauma. To improve soft body armor testing methods, non-invasive sensors were developed to track the back face deformation experienced during an impact event. The sensing layer developed features an imbedded Fiber Bragg grating which is used to capture strain measurements during testing. Theses strain measurements can then be used to reconstruct the peak deformation depth as it forms during impact. Testing has shown that the silicone mat and optical fiber are able to survive impacts behind armor consistently and the transient history of the BFD can be collected during impact. The maximum deformation found while testing theses sensors correlates well with independent clay indentation measurements

In structural health monitoring (SHM) applications, one of the advantages of utilizing a surface bonded fiber Bragg grating (FBG) sensor for damage detection is its increased sensitivity in collecting ultrasonic waves. Recent studies have demonstrated that for a certain bonding condition the output FBG response can be increased by bonding the optical fiber at a distance away from the FBG to collect optical fiber guided wave (L01 mode) that is converted from S0 Lamb wave, referred as remote bonding. However in order to apply the remote bonding configuration in practical situations, the S0 mode conversion to L01 mode through an adhesive bond under various conditions must be characterized. This work investigates how the coupled L01 mode changes with varying input S0 mode frequency and angle of incidence through an adhesive bond. The goal of this work is to better understand the S0 mode conversion to the L01 mode in order to implement the remote bonding configuration for an improved SHM of a structure.

A single optical fibre sensor is used to measure the dynamics of an impact. The method consists of sewing the optical fibre onto a woven Kevlar layer and placing it between the shoot pack and backing material. The measurement is accomplished by using the friction between the layer and the optical fibre to relate the optical fibre strain to impact deformation. Tests are done using a backing material of Roma Plastilina No.1 clay, and transparent ballistics gel with independent high-speed imaging. A final calculated BFD average error of 7.75% is presented as well as a timing error of 15.5% between the imaged dynamic BFD and the dynamic BFD determined by the FBG. This method is also tested at the U.S. Army Aberdeen Test Center in Maryland with a final calculated error of 7%.

We applied pulsed phase thermography to image and size damage in adhesively bonded joints. Specifically, the initiation and propagation of fatigue-induced damage in single lap joints with carbon fiber epoxy adherends was investigated. Lap joint specimens with various levels of manufacturing defects were fabricated and loaded in low-cycle fatigue. A calibration specimen with artificial defects was used to design a threshold algorithm for sizing of the damaged regions. The dominant failure mode in specimens without manufacturing defects was fiber-failure, whereas joints failing prematurely demonstrated adhesive failure. Imaging of the lap joints after regular number of fatigue cycles revealed that manufacturing defects could be detected and the resulting, imminent adhesive failure could be identified prior to joint failure. Additionally, the extent of this damage could be accurately estimated through the sizing algorithm. Due to the brittle nature of fiber-failure, it could not be detected prior to failure of the joint, however this was not critical, as the goal was to identify premature failure of the adhesively bonded joint.