In this application of the statistical pattern recognition paradigm, a prediction model of a chosen feature is developed from the time domain response of a baseline structure. After the model is developed, subsequent feature sets are tested against the model to determine if a change in the feature has occurred. In the proposed statistical inference for damage identification there are two basic hypotheses; (1) the model can predict the feature, in which case the structure is undamaged or (2) the model can not accurately predict the feature, suggesting that the structure is damaged. The Sequential Probability Ratio Test (SPRT) develops a statistical method that quickly arrives at a decision between these two hypotheses and is applicable to continuous monitoring. In the original formulation of the SPRT algorithm, the feature is assumed to be Gaussian and thresholds are set accordingly. It is likely, however, that the feature used for damage identification is sensitive to the tails of the distribution and that the tails may not necessarily be governed by Gaussian characteristics. By modeling the tails using the technique of Extreme Value Statistics, the hypothesis decision thresholds for the SPRT algorithm may be set avoiding the normality assumption. The SPRT algorithm is utilized to decide if the test structure is undamaged or damaged and which joint is exhibiting the change.

For monitoring and evaluating structural health for the three long-span cable-supported bridges, namely, Tsing Ma Bridge, Kap Shui Mun Bridge and Ting Kua Bridge in Hong Kong, Wind And Structural Health Monitoring System (WASHMS) has been established by the Highway Department of the local Government in Hong Kong. To investigate the possibilities of further enhancing the functions of the existing system in storage, retrieve and distribute huge amount of monitoring data for the WASHMS, a study of developing GIS-based bridge structural health monitoring and management system is carried out and presented in this paper. This system is developed by integrating GIS, large database and network techniques. The system is able to run on a Local Area Network or Internet. GIS is applied to manage spatial information on location of bridges on a regional map and position of sensors on a 3D bridge model. It also provides users with interactive interface between user and the system. Large database system, such as SQL Sever or Oracle, is used to manage huge amount of dataset captured from the sensors. In the Sever-Side, by using the data input and database maintain module, the raw data come from WASHMS and the analyzed results can be imported into the central database in real time, and is backup from time to time. For those data that are not changed or changed infrequently, such as regional map for bridges, location information and attributes about sensors, a static database is established to store them. In the Client-Side, user can not only access the raw data from various sensors from the central database at any time, but also visualize or even further process the data. Based on above designed prototype of GIS-based bridge structural health monitoring and management system. It is demonstrated by the prototype that the developed system is feasible. Furthermore, the implementation issues including system design, technique solutions, database structure design and function design are also presented in this paper.

Non-destructive health monitoring of structures may be achieved by system identification to determine structural parameters based on measured dynamic response and excitation. A hybrid approach of combining genetic algorithms (GA) and a local search (LS) method is developed, which has some advantages over the direct GA (i.e. without LS). The proposed LS algorithm is GA-compatible and hence easy to implement. The deviation size of local search can be adjusted adaptively in both the global search and local search phases. Extensive numerical studies accounting for the effects of measurement noise have shown that the hybrid GA-LS method leads to considerable improvement in the identification results. The hybrid method is applied to plate/shell structures in this study. As an illustration, an aircraft wing model is presented in the numerical simulation study. The accuracy of identified parameters is very good generally. A sensitivity study is carried out by perturbing the identified parameter and computing the corresponding change in fitness value. It is found that, to some extent, the accuracy of identification results is well correlated with the sensitivity in terms of the fitness value used.

Ceramic matrix composites (CMC) are engineered materials filled with manufacturing anomalies, such as voids, delamination, or fiber cracking. In this paper a non- destructive evaluation (NDE) of a CMC tensile specimen is coupled with a finite element analysis to locate the failure location prior to the actual testing. The tensile CMC specimen is scanned with computed tomography (CT) along various planes. The majority of the observed anomalies are porosities in the matrix. The CT images are then used to reconstruct a 3-D volume of the specimen's gage section using velocity² (an image processing software). Subsequently, a three dimensional finite element analysis (FEA) is carried out to include the scanned porosities. The stress variations along the scanned CT planes are determined, comparison of the FEA results with those extracted via NDS, and the test data are reported.

The electrical resistance change of CFRP composite under mechanical loading condition can be used as an effective parameter to evaluate the accumulated internal damage. The internal electrically conductive network, which is constructed by the contacts between conductive carbon fibers, is very important in correlating the resistance change to mechanical damage state. The electrical ineffective length (delta) _ec, the mean distance between adjacent contact points, can be used as a useful parameter to evaluate the internal conductive network. In this study, a new methodology for the evaluation of the internal conducting network is proposed. For this purpose, the electrically anisotropic characteristics of the CFRP composite are measured using the DC 4 probe and the DC 6 probe methods and the values of (delta) _ec are estimated using the mechanical tensile test for various fiber volume fractions. Based on the experimental results, the empirical relationship between the electrical anisotropy and (delta) _ec is established. We also conduct the DC network circuit analysis using Kirchhoff's rule and the Monte Carlo simulation of the contact point distribution, and calculate the (delta) _ec for various contact configurations. By comparing the analytical and experimental results, the distribution configuration and the number of contact points, which constitute the internal electrically conductive network of CFRP composites, can be evaluated quantitatively.

In this paper, an experimental study was performed to investigate the possibility of a new NDE system - Tapping Sound Analysis (TSA). The tapping sound and contact force of healthy laminated composite structure and defective laminated composite structure were measured using tapping device. The feature extraction method based on the wavelet packet transform was used to extract features from the tapping sound. Comparing the feature of the tapping sound of healthy structure and defective structure, a feature index could be derived, which indicates the existence of defect inside the laminated composite structure. Using the feature index, the difference between the tapping sound data of the present specimens could be expressed as a single value.

When fiber Bragg (FBG) sensors are embedded in carbon fiber reinforced plastic (CFRP) laminates, the reflection spectrum from the FBG sensors split into two peaks because of the non-axisymmetric thermal residual stress. This deformation of the spectrum will lead to misreading in strain measurements or crack detection in the laminates. In the present research, three types of FBG sensors: uncoated normal, polyimide-coated normal, and polyimide-coated small- diameter FBG sensors were embedded in CFRP laminates, and reflection spectra from the sensors were measured during the fabrication process of the laminates in order to evaluate the effect of thermal residual stress on the reflection spectra. Through the comparison of results obtained for the three FBG sensors, it was found that the effect of thermal residual stress on the reflection spectrum could be decreased when the optical fiber was coated with polyimide in the present laminate configuration and embedment position. Furthermore, these changes in the spectra during the curing process were simulated theoretically. Thermal residual stress components at the core were obtained by FEM analysis, and the spectra were calculated using the couple mode theory and the transfer matrix method. Considering the birefringence effect, the calculated spectra reproduced the measured spectra very well.

AS4/PEEK [0/+/- 45/90]_7s composite laminates were subjected to a bending moment. We have obtained the displacement fields by moire interferometry with the aid of Long Distance Microscope. Based on wavelet analysis, the digital moire images were processed and analyzed. The special algorithms of unconstrained nonlinear minimization and the scale parameter were used to eliminate the influence of the noise in the digital image processing. Tension tests were carried out on AS4/PEEK [ 0/+/- 45/90 ]_2s composites. When AS4/PEEK composite laminates were deformed, the images of natural surface of the AS4/PEEK specimen were recorded to study the damage initiation and growth. The experimental results show that the considerable interlaminal stresses between different plies, and these places were more prone to delaminate. The paper presents the microscopic damages and matrix crack growth observed by an optical microscope and the CCD camera.

Conventional methods for analysis of pulsed thermographic NDE sequence data are highly susceptible to noise, nonlinearity of the IR camera response, and the presence of surface features on the sample. Furthermore, the ability of conventional methods to significantly improve the ability to retrieve deep or weak subsurface features beyond the original unmodified image is limited. We have developed a Thermographic Signal Reconstruction (TSR) technique that enhances defect to background contrast, increases the depth range over which pulsed thermography can be applied, and reduces the amount of blurring due to lateral diffusion that is typical of thermographic imaging. The TSR approach also reduces the amount of data that must be stored by an order of magnitude. The reduction in size of the data structure allows simultaneous manipulation of data from numerous locations on a sample, so that fast parallel processing of large structure data is possible. The results of the parallel processed TSR data consistently offer higher spatial resolution, less blurring and more precise depth and size measurement than the original data. Examples on aircraft and power generation components will be presented.

Acoustography is the ultrasonic analog of radiography and photography. A unique 2D detector, called acousto-optic (AO) sensor, is used that is capable of directly converting ultrasound into visual images; much like a fluorescent screen is able to convert x-rays into a visual image. The AO sensor offers exceptionally high resolution and can be fabricated to have a large area. This allows image formation through simple shadow casting (analogous to x-ray image formation) or with acoustic lense (analogous to a photographic or video camera). This paper will report on several new developments, which could allow acoustography to provide a simpler more cost-effective alternative to conventional ultrasonic testing.

Pulsed thermography is a powerful method for the nondestructive evaluation (NDE) of composite materials since it provides a large field means for detecting subsurface material defects and material variations. The main difficulty in the use of this technique, however, is the processing and interpretation of the acquired thermal image data. This paper examines three methods for processing pulsed thermography results concerning a composite plate with material inserts that simulate delamination type defects. Using the same temporal data set, a specimen is analyzed utilizing peak contrast, peak slope and a newly developed thermal image reconstruction technique. Comparisons are made on a composite panel with differing defect sizes and depths. In addition, results are compared to images gathered using through transmission ultrasonics and microfocus radiography. With subjective manipulation, pulsed thermography was able to show the defects more clearly than either the ultrasonic or radiographic techniques. A discussion of this subjectivity and future directions for automation are provided in an effort to better understand the potential of the method.

Thermoelastic Stress Analysis is used to investigation two damage types in composite materials, namely delamination and fibre breakage. The 'damage' is introduced into the material at the manufacturing stage using PTFE patches to model delamination and by cutting fibers to model breakage. Both Glass Fibre-Reinforced Plastic (GRP) and Carbon Fibre-Reinforced Plastic (CFRP) specimens were tested using an Instron 8800 servohydraulic test machine, and the Deltatherm TSA equipment was used to obtain full-field images of the temperature change in the cyclically loaded specimens. Results are presented to show how the damage introduced produces a change in the measured thermoelastic signal. Issues such as non-adiabatic behavior are discussed in the context of quantitative damage assessment.

As the aerospace industry continues to advance the design and use of composite structure, the NDE community faces the difficulties of trying to keep up. The challenges lie in manufacturing evaluation of the newest aerospace structures and materials and the in-service inspection and monitoring of damaged or aging composites. This paper provides examples of several promising NDI applications in the world of aerospace composites. Airborne (or non-contact) Ultrasonic Testing (UT) has been available for decades, but recently has generated new interest due to significant improvements in transducer design and low noise electronics. Boeing is developing inspection techniques for composite joints and core blankets using this technology. In-service inspection techniques for thick, multi-layer structures are also being advanced. One effective technique integrates the S-9 Sondicator, a traditional bond testing device, with Boeing's Mobile Automated Scanner (MAUS) platform. Composite patches have seen limited use on-aircraft, due, in part, to the difficulty of determining the quality of a bonded joint. A unique approach using Electronic Speckle Pattern Interferometry (ESPI) is showing promise as a bonded patch-inspection method. Other NDI techniques currently being developed for aerospace application are also briefly discussed.

This paper describes the study carried out to determine the possibility of healing an AS4/PEEK composite plate that was impacted at a low velocity to create the delaminations. Images of the AS4/PEEK composite plates were obtained prior to the impact, to ensure it is free of any gross defects that could have been imparted during the manufacturing of the composite plate. A drop-weight was used to impact the composite plate at a low velocity and the incident energy of impact was maintained at 2.26 ft-lb. Interior images of the composite plates were obtained after the impact and prior to healing by a C-scan imaging system. The healing process was conducted at controlled temperature and pressure. The healed specimens were imaged again by the C-Scan imaging system. The results obtained by analyzing the images show a reduction in the size of the delaminations.

Bond inspection is one of the most critical NDE tasks addressing the aging aircraft problem. This paper is aimed at reviewing and furthering the use of ultrasonic guided waves for the inspection of bonded components, particularly lap joints between thin plates such as those found in aircraft skins. The transfer matrix theory of wave propagation in multilayered components is used to predict the sensitivity of the wave dispersion behavior to the bond state. The experimental study includes the use of non- contact air-coupled transducers to excite and detect guided waves in a pitch-catch arrangement. It is shown that by monitoring the ultrasound amplitude leaking through the bondline it is possible to detect disbonds as well as low- strength bonds. A photoelastic study confirms this result. It is reaffirmed that those modes with predominant shear- type deformations at the bondline are the most sensitive to the bond state.

Different approaches are used to sense and to localize a damage of rotating structures. Most of the methods take advantage of the dynamic behavior of the structural model [1-7]. This paper uses the modal and sensor norms, as defined in [8], to determine damage locations. The proposed approach allows localization of damaged elements within a structure, and provides information concerning the impact of the damage on the structure's natural modes of vibration.

The purpose of this effort was to develop a system* to detect, discriminate and track fatigue cracks in rotating disks. Aimed primarily at jet engines in flight applications, the system also has value for detecting cracks in a spin pit during low cycle fatigue testing, and for monitoring the health of steam turbines and land-based gas turbine engines for maintenance purposes. The results of this effort produced: a physics-based model that describes the change in the center of mass of a rotating disk using damping ratio, initial unbalance and crack size as parameters; the development of a data acquisition and analysis system that can detect and discriminate a crack using a single cycle of data; and initial validation of the model through testing in a spin pit. The development of the physics-based model also pointed to the most likely regimes for crack detection; identified specific powers of (omega) search for in specific regimes; dictated a particular type of data acquisition for crack discrimination; and demonstrated a need for a higher signal-to-noise ratio in the measurement of the basic vibration signal.

Low Frequency Eddy Current (EC) probes are capable of measurement from 5 MHz down to DC through the use of Magnetoresistive (MR) sensors. Choosing components with appropriate electrical specifications allows them to be matched to the power and impedance characteristics of standard computer connectors. This permits direct attachment of the probe to inexpensive computers, thereby eliminating external power supplies, amplifiers and modulators that have heretofore precluded very low system purchase prices. Such price reduction is key to increased market penetration in General Aviation maintenance and consequent reduction in recurring costs. This paper examines our computer software CANDETECT, which implements this approach and permits effective probe operation. Results are presented to show the intrinsic sensitivity of the software and demonstrate its practical performance when seeking cracks in the underside of a thick aluminum multilayer structure. The majority of the General Aviation light aircraft fleet uses rivets and screws to attach sheet aluminum skin to the airframe, resulting in similar multilayer lap joints.

Electronic Speckle Pattern Interferometry (ESPI) and Digital Shearography are optical interference techniques, suitable for non-destructive inspection procedures. Due to the stringent vibration isolation conditions required for ESPI, the technique is mainly suited for laboratory based inspection procedures, which cannot be said for Digital Shearography. On the other hand, the interference patterns obtained using ESPI exhibit better fringe definition and contrast than those obtained using Digital Shearography. The image quality of Digital Shearography can be improved by introducing phase stepping and unwrapping techniques, but these methods add a level of complexity to the inspection system and reduce the image refresh rate of the overall process. As part of a project to produce a low cost portable ESPI system suitable for industrial applications, this paper investigates various methods of minimizing the impact of environmental vibration on the ESPI technique. This can be achieved by effectively 'freezing' the object movement during the image acquisition process. The methods employed include using a high-powered infra-red laser, which is continuously pulsed using an electronic signal generator as well as a mechanical chopper. The effect of using a variable shutter speed camera in conjunction with custom written software acquisition routines is also studied. The techniques employed are described and are applied to selected samples. The initial results are presented and analyzed. Conclusions are drawn and their impact on the feasibility of a portable ESPI system discussed.

In a previous study by the authors, the ultrasonic spectroscopy technique identified possible disbonds or delaminations that went unsubstantiated by other NDE (nondestructive evaluation) methods. The specimens were polymer matrix composite (PMC) rings sectioned from flywheel rotors. For this study, polymer matrix composite (PMC) rings were further investigated to determine the sensitivity of the ultrasonic spectroscopy technique in detecting tight disbonds or delaminations. The ultrasonic system utilizes a continuous swept sine waveform as the input. After the swept sine wave traverses the material, the captured waveform is subjected to two fast Fourier transforms (FFT); i.e. an FFT operation is performed on the amplitude versus frequency plot obtained by the first FFT. The second FFT along with equalization of the frequency spectrum allows for the evaluation of the fundamental resonant frequency as a function of material properties and thickness. Here, a study of ultrasonic spectroscopy's sensitivity to delaminations was conducted. Data was collected while opening a controlled delamination. The delamination opening was monitored using optical methods. The full thickness resonance, the resonance corresponding to the location of the intentional disbond, and the frequency spectrum were examined in an effort to characterize the sensitivity of the NDE method concerning various delamination conditions.

Acousto-ultrasonics (AU) is a nondestructive evaluation (NDE) technique that utilizes two ultrasonic transducers to interrogate the condition of a test specimen. The sending transducer introduces an ultrasonic pulse at a point on the surface of the specimen while a receiving transducer detects the signal after it has passed through the material. The aim of the method is to correlate certain parameters of the detected waveform to characteristics of the material between the two transducers. The waveform parameter of interest is the attenuation due to internal damping for which information is being garnered from the frequency domain. The parameters utilized to indirectly quantify the attenuation are the ultrasonic decay rate as well as various moments of the frequency power spectrum. Here, the sensitivity for each of the parameters was assessed in respect to changing boundary conditions during the experiments. Three conditions were controlled during the ultrasonic characterization of the specimens. First, issues concerning the contact force of the transducer were studied. Second, the support structure of the specimen was addressed. Lastly, the damage state of the material itself was considered. After analyzing the various AU parameters, the overall sensitivity of the AU technique to material change or damage were quantified and compared to changes in the AU values resulting from the experimental boundary conditions. This investigation showed that certain AU parameters could be utilized to gauge damage in composites, although, the experimental boundary conditions may slightly influence the results. The results of this study are important due to the fact that at this point AU is an empirical method.

We have developed a non-contact ultrasonic system capable of measuring the thickness of aluminium sheet with sub-micron accuracy. The thickness of sheet has been calculated from ultrasonic data obtained using a send-receive, radially polarized shear wave Electromagnetic Acoustic Transducer (EMAT). Sheets in the thickness range between 0.1mm to 0.5mm have been measured using this non-contact approach at a stand-off of up to 1.5mm. The ultrasonic echo trains have been processed using Fourier analysis to extract transit time measurements for the two possible shear wave polarizations within the sheet. Two broadband EMAT systems have been used to perform the measurements with center frequency of approximately 5 MHz and frequency content up to 10 MHz and 20 MHz respectively. The most accurate measurements of thickness on thin sheets have been made using Fourier analysis rather than direct temporal measurements and have yielded thicknesses accurate to within 0.2%. This accuracy is potentially of significant commercial benefit to sheet manufacturers. There are also other potential applications of this approach, particularly for the measurement of very thin layers on aluminium substrates and these will be discussed in the presentation.

This study was aimed at developing a new method for assessing the integrity of concrete structures. Especially acoustic emission technique was used in carrying out both laboratory experiment and field application. From the previous laboratory study, we confirmed that AE analysis provided a promising approach for estimating the level of damage and distress in concrete structures. The Felicity ratio, one of the key parameter for assessing damage, exhibits a favorable correlation with the overall damage level. The total number of AE events under stepwise cyclic loading also showed a good agreement with the damage level. In this study, a new suggested technique was applied to several concrete bridges in Korea in order to verify the applicability in field. The AE response was analyzed to obtain key parameters such as the total number and rate of AE events, AE parameter analysis for each event, and the characteristic features of the waveform as well as Felicity ratio analysis. Stepwise loading-unloading procedure for AE generation was introduced in field test by using each different weight of vehicle. According to the condition of bridge, for instance new or old bridge, AE event rate and AE generation behavior indicated many different aspects. The results showed that the suggested analyzing method would be a promising approach for assessing the integrity of concrete structures.

Implementation of a sophisticated long-term monitoring system on the Tsing Ma suspension bridge in Hong Kong highlights the necessity of developing practical damage detection methodologies for large-scale civil structures. The novelty detection technique has been demonstrated to be greatly promising for damage occurrence detection of long-span bridge structures in operation with noisy measurement data. Some key issues concerning the implementation of this technique to the instrumented Tsing Ma Bridge are explored in this paper: (i) Selection of modal frequencies in constructing novelty neural network input for structural damage alarming. A novelty detection neural network using natural frequencies of only vertical modes as input vector and a novelty detection neural network using natural frequencies of mixed vertical and lateral/torsional modes as input vector are respectively constructed, and their identification sensitivities are compared; (ii) Definition of distance functions to measure difference between the input and output vectors due to anomaly. Both the Euclidean distance and the Mahalanobis distance are used to define the anomaly metric and their sensitivities to damage are identified; (iii) Construction of a novelty index capable of discerning between the novelty shift due to structural damage and the novelty shift due to inconsistent noise level. The focus of the present study is on developing an improved novelty index which uniquely indicates structural damage rather than anomaly due to change of noise level. The existing novelty index cannot distinguish true damage from anomaly due to inconsistent noise level in training and testing stages. As a result, the sequence shift signaled by such a novelty index is definitely indicative of structural damage only when the noise level is exactly same for the training data and for the testing data in statistical sense. On the contrary, the improved novelty index developed in this study always indicates damage information no matter the training noise level is greater or less than the testing noise level.

A damage detection system was developed with commercially available wireless sensors. Statistical process control methods were used to monitor the correlation of vibration data from two accelerometers mounted across a joint. Changes in correlation were used to detect damage to the joint. All data processing was done remotely on a microprocessor integrated with the wireless sensors to allow for the transmission of a simple damaged or undamaged status for each monitored joint. Additionally, a portable demonstration structure was developed to showcase the capabilities of the damage detection system to monitor joint failure in real time.

Fiber optic pH sensors based on the evanescent field spectroscopic technique is studied. Portions of poly (-methyl methacrylate) (PMMA) cladding of plastic clad silica (PCS) optical fibers are replaced with new cladding composed of PMMA doped with a pH sensitive chromophore. Methyl Red, Thymol Blue, Thymolphtalein are used for sensing pH at the acid, neutral and base levels. Changes in the pH of the analyte are detected by measuring the absorption spectrum of the new cladding in the sensing region of the optical fiber.

Smart materials and intelligent devices are promising to revolutionize data collection in civil infrastructure. However, their application has so far been very limited in scope. So far, there is no comprehensive mechanisms to integrate data generated by smart materials and intelligent devices into the overall IT systems of user organizations. This article outlines a framework for reengineering organizations for optimal utilization of smart materials and intelligent devices in the infrastructure development domain. The aim is to integrate data collection and management into the organizational culture. The framework includes three basic modules: assessing current practice, identifying level of deployment, and defining the needed actions.

This paper describes efforts at monitoring microfloor vibrations in a newly-constructed research building. This building is intended to house a variety of delicate precision scientific instruments with performances that are deleteriously affected by even small floor vibrations. The building is five stories with welded steel construction. Upon completion of the construction, the initial occupants anecdotally complained of excessive floor vibrations. This resulted in an effort to measure the vibrations and to reduce them at their sources, including the mechanical systems for the building. The measurements are compared with industry standards and with measurements taken at nearby reinforced concrete buildings. The success of efforts at reducing the vibrations due to the mechanical systems of the building are also assessed.