DEFECT SIGNAL ENHANCEMENT IN INSPECTION LINES BYMAGNETIC FLUX LEAKAGE

J. Etcheverry, A. Pignotti, G. Sanchez, and P. Stickar

Centre de Investigation IndustrialL. Alem 1067, 1001 Buenos Aires, Argentina

ABSTRACT. The detection of flaws that involve 5% or more of the pipe wall thickness is not easyto achieve for internal defects inspected from the outside. In this work we focus on a relativelystraightforward technique, based on obtaining the characteristic signature of relevant defects, andprojecting the actual signals on these "standard" defect configurations, thus increasing the signal-to-noise ratio and providing an alternative way to determine the nature of the defect. Severaloptions are discussed, including some that are computationally less demanding, and are susceptibleof being implemented on-line.

INTRODUCTION

Magnetic flux leakage (MFL) is one of the preferred techniques for nondestructiveflaw detection in ferromagnetic materials [1]. It is used as one of the basic tools in steelpipe inspection at the manufacturing plants, where it has to be carried out at a pacecompatible with the plant productivity, and is usually implemented by comparing thepeak-to-peak signal amplitude with a predetermined threshold. In this application, it isusually necessary to detect very small flaws in the pipe wall, such as those that affect 5%or more of the wall thickness [2]. For thick-walled tubes and internal defects inspectedfrom the outside, this 5% detection is difficult to achieve because of the presence of noiseof comparable magnitude. As a consequence, a sizable fraction of the number of tubes hasto be sent for manual inspection by magnetic particles, with the consequent productivityloss.

Several alternatives involving filtering and other approaches such as neural-netbased recognition have been proposed to improve the performance of this inspection [3-4].In this work we focus on a relatively straightforward technique, based on obtaining thecharacteristic signature of the relevant defects, and projecting the actual signal on thesereference defect signals.

REFERENCE SIGNALS

The detection thresholds are set at calibration time, before a given batch of tubes isinspected, and are based on the signals recorded from notches machined on a calibrationpipe. Figs. 1-2 show such signals, one from an "external 12.5%" notch, (i.e., from a notch

CP657, Review of Quantitative Nondestructive Evaluation Vol. 22, ed. by D. O. Thompson and D. E. Chimenti 2003 American Institute of Physics 0-7354-0117-9/03/S20.00

1721

1000

500

Oi

-500

-1000

i V

1000

500

0 20 40 60 80 100Samples (4 kHz sampling rate)

FIGURE 1. Reference signal for an internal12.5% notch.

-500

-1000

0 20 40 60 80 100Samples (4 kHz sampling rate)

FIGURE 2. Reference signal for an external 12.5%notch.

machined on the external surface, whose depth is equal to 12.5% of the pipe wallthickness), and the other one from an "internal 12.5%" notch. It is clear from the figuresthat signals from external notches are larger than those from internal ones of comparablesize. This is of course not surprising, because the inspection probes are located at a smalldistance from the external pipe surface, and are therefore much closer to an external flawthan to an internal one. Besides, it is clearly seen from Figs. 1-2 that signals from externalflaws are also narrower than those from internal ones. These features imply that: 1)different thresholds have to be used depending on the nature (external/internal) of theflaw; 2) the determination of whether the defect is external or internal has to precede thedecision of whether it is likely to surpass the maximum admissible defect size. Thealgorithm discussed below is an attempt at facilitating the detection and sizing of defectsby simultaneously increasing the signal-to-noise ratio and providing an alternative way fordetermining the nature of the flaw.

SIGNAL PROJECTION

In an environment in which the analog signals picked up by the inspection probesare digitized, a defect signal is an N-component segment of a data string, and may bethought of as a vector u in an N-dimensional vector space that we represent as

. ,UN] (1)The components of this vector are the evenly sampled values obtained by digitizing theanalog defect signal.

1722

Following a standard mathematical notation, the scalar product of two vectors uand v is

= ui.v i + ... + UN.VN (2)

and the vector norm is defined to be

H = 1/2. (3)

The "cosine of the angle" between two signals u and v is therefore given by

cos(a) = /||u||||v||. (4)If D is the vector associated to a given reference signal, the normalized reference signal is

D* = D/||D||. (5)The projection of a signal u in the direction of defect D is just the scalar product.

STRING PROJECTION AND NOISE REDUCTION

In this context, the data string generated by a given probe may be denoted by theindefinite sequence of the evenly sampled data values:

S = (..., Sb Si+i, Si+2v.) (6)

and we may construct a "projected" string SD, by replacing each value si by the projectionof the preceding N-component substring in the direction of D.

Thus,SD = (..., SDI> SDi+1,...) (7)

withSDi = (8)

In this way, the elements in the projected string are always smaller or equal to those in theoriginal string, the equality holding only when the substring considered is proportional toD, i.e., it points in the same direction. The net result is that the signal-to-noise ratio for"D-type" defects is improved when replacing string S by SD, as shown in Figs. 3-4.

Significant noise is still occasionally present in SD, i.e., there are noisy datasegments that have a sizable projection in the direction of D, even though they do not pointin the same direction, and, therefore, do not qualify as defects. To further dampen thiskind of noise, we multiply the ith element of SD by the factor

1723

1650 1700 1750 1800 1850

2000 4000 6000 8000 10000 12000 14000 16000Samples

FIGURE 3. Original data string, containing a large external defect and weaker signals hardlydistinguishable from noise. The inset depicts one hundred samples, containing the defect signal.

2000 4000 6000 8000 10000 12000 14000 16000Samples

FIGURE 4. Same data string as in Fig. 3, after projection on the external reference defect.

1724

10.8

7B '

I0-4

0.2" >

0

Q-0.4

-0.6-0.8

0 2000 4000 6000 8000 10000 12000 14000 16000Samples

FIGURE 5. Same data string as in Figs. 3 and 4, after further noise reduction with the cos2((X|) factor.

cos2(oci) = < [Si_N+i,..., sj , D*>2/ ||[Si_N+1,...,Si]||2

The final proposal is therefore to replace the string element Sj by

SDI = SDi cos2(oci) = < [Si_N+1,..., sj , D*>3/ ||[Si_N+1,...

(9)

(10)

Fig. 5 shows the result of introducing this further noise reducing factor. In order tovisualize better the effect of these operations on the signal-to-noise ratio, the three stringsshown in Figs. 3-5 are multiplied by normalization coefficients such that the peak value ofthe string is in all cases equal to 1. The fact that weak signals, impossible to be singled outin the original string, are visible after being subject to this process, is precisely the pointthat we want to make.

DEFECT IDENTIFICATION

We have shown in the previous section a procedure that can be used to improve thesignal-to-noise ratio in a data string. But the same procedure can also be used todiscriminate whether a signal that has been extracted from one such string by any othermethod, is more likely to originate from a particular type of flaw. This "defect

1725

identification" is done by comparing the angles that are obtained by projection of thesignal in the direction of the reference signals under investigation, and choosing thesmaller one.

Defect characterization is an open problem, because there is a wide variety ofpossible defect morphologies. As an initial task, we have set the goal of discriminatingbetween external and internal defects, even realizing that other geometrical characteristicsof the flaws may mask their internal or external nature. The reason behind setting this goalis the fact - already mentioned above - that different thresholds should be used todetermine the criticality of one defect, depending on which wall surface it is located. Thetraditional way of discriminating between these defects is based on the fact that externalsignals are usually sharper than internal ones, therefore having a higher average frequency,which can be determined by carrying out a FFT in real time. This is currently done atsome industrial inspection lines, but implies a heavy computational burden, especiallybecause not knowing a priori the nature of the defect, a low threshold has to be used, andtherefore string segments that are candidates to qualify as defects do appear quitefrequently. Alternatively, identification of signals by neural network techniques has alsobeen attempted with satisfactory results in a study of artificial defects [4], but the problemof retraining the net every time there is a change in the inspected batch of products has yetto be solved. The advantage of the scheme proposed here is that retraining of the algorithmwhen there is a change in the product under inspection is fairly straightforward, because itamounts to updating the reference signal strings using the signals recorded at calibrationtime, which is not significantly different from the usual resetting of the rejectionthresholds during the calibration procedure that precedes the inspection of each newproduction batch.

A final point to be considered is the choice of N, the number of elements kept inthe data segments used to represent a defect signal. A fine sampling of the referencesignals with N=100, such as shown in Figs. 1-2, is quite effective for noise suppression,but is likely to put too much of a burden on the CPU for on-line inspection. We werefavorably surprised to realize, however, that there is no need to use all the N componentsof the selected segment of the input data string, and that as few as 4 or 5 convenientlychosen samples do provide a significant improvement for noise reduction and defectidentification in the examples examined.

CONCLUSIONS

In spite of the wide variety of natural defects that may be present in the inspectionof seamless steel pipes, we have found that the proposed scheme, based on the projectionof digitized data streams on signals obtained from calibration notches, is a promising wayof enhancing weak signals in a noisy environment, thus improving the detection andidentification of the flaws that originate them.

ACKNOWLEDGMENTS

This work was done as part of a program sponsored by the tube manufacturersTamsa and Siderca. The authors want to acknowledge the continuous support and adviceprovided by Alberto Nicolini.

1726

REFERENCES

1. Lord, W., and Hwang, J. H., Brit. J NDT 19, 14 (1977).2. American Petroleum Institute Specification 5CT for casing and tubing, Section 9,

1992.3. Afzal, M., Udpa, S., Udpa, L., and Lord, W., "Rejection of seamless pipe noise in

Magnetic Flux Leakage data obtained from gas pipeline inspection", in Review ofProgress in QNDE, Vol. 19, Eds. D. O. Thompson and D. E. Chimenti, AIPConference Proceedings 509, Melville, New York, 2000, p. 1589.

4. Perazzo, R., Pignotti, A., Reich, S., and Stickar, P., "Feature extraction in MFL signalsof machined defects in steel tubes", in Review of Progress in QNDE, Vol. 20, Eds. D.O. Thompson and D. E. Chimenti, AIP Conference Proceedings 557, Melville, NewYork, 2001, p. 619.

1727

WelcomeContentsMINISYMPOSIUM: IMPORTANT TECHNOLOGY ADVANCES FOR NDEAdvances in Information TechnologyNDE of Material Degradation by Embrittlement and FatiguePractical Long Range Guided Wave Inspection Managing Complexity

ELASTIC WAVES AND ULTRASONIC TECHNIQUESSection A. Ultrasonic ModelingA Coupling Method of Boundary Element Method and Generalized Ray Theory for Elastic Wave Scattering in a Thick PlateAn Extension of the Spring Model to Nonlinear InterfacesDiffraction of an Elastic Wave by a Rough Crack EdgeScattering of Elastic Waves by a Rectangular Crack in a Thick Walled Anisotropic SolidHybrid Modeling of Elastic Wave Scattering in a Welded CylinderA Formal Approach to Include Multiple Scattering in the Estimation of Ultrasonic Backscattered SignalsA Multilevel Multipole Method for Modeling Elastic-Wave Multiple Scattering in Fiber-Reinforced CompositesA Multilevel Multipole Method for Modeling Elastic-Wave Multiple Scattering in Fiber-Reinforced CompositesModelling of the Ultrasonic Response of Inclusions in SteelsUltrasonic Attenuation as Influenced by Elongated GrainsComputational Study of Grain Scattering Effects in Ultrasonic MeasurementsAdvances in Modeling Ultrasonic Noise Induced by Machining Advances in Modeling Ultrasonic Noise Induced by Machining Roughness: Development and ValidUltrasonic Detection of Damage in Heterogeneous MediaWaves in Anisotropic Elastic Media

Section B. Guided WavesScattering of the SHO Mode from Geometrical Discontinuities in PlatesRemote Monitoring of Plate-Like Structures Using Guided Wave ArraysSimulations of Laser-Generated Guided Waves in Two-Layer Bonded Plate with a Weak InterfaceCrack Detection for Aircraft Holes with Limited Accessibility Containing Fasteners and SealantModels for Crack Detection in a Cylindrical Hole Containing an Elastic Layer and a Fluid-Filled Cylindrical HoleOn the Detectability of Fatigue Crack Growth at Fastener Holes Using Guided WavesReflection of the S0 Lamb Mode from a Part-Depth Circular Defect in a Plate, When the Incident Wave is Created by a Small SourceHigh Frequency Guided Wave Virtual Array S A F TFast Techniques for Calculating Dispersion Relations of Circumferential Waves in Annular StructuresThe Scattering of Ultrasonic Guided Waves in Partly Embedded Cylindrical StructuresDetection of Scale Inside of Water Supply Pipes Using Guided WavesLong Range Inspection of Rail Using Guided WavesNon-Axisymmetric Wave Focusing in Pipe InspectionSemianalytical Finite Element Analysis for Ultrasonic Focusing in a PipeThe Application of Finite Element Modelling to Guided Wave Testing Systems

Section C. Laser UltrasonicsLaser Ultrasonic System for On-Line Steel Tube GaugingDoppler Frequency-Shift Compensated Photorefractive Interferometer for Ultrasound Detection on Objects in MotionA Theoretical Model for the Ultrasonic Detection of Surface-Breaking Cracks with the Scanning Laser Source TechniqueCalculation of Ultrasound Excited by a Pulsed Thermal Source Distributed Along the Depth DirectionThe Optimization of Lamb and Rayleigh Wave Generation Using Wideband-Low-Frequency EMATsLaser Generation of Focused Ultrasonic WaveAn Effective Generation of Lamb Waves in a Thin Plate Using Laser Line Array Illumination and Their Propagation CharacteristicsHigh Frequency Ultrasound Generation Using a Femtosecond LaserA New Thermoelastic Source Model for Non-MetalsLamb Wave Dispersion Characterization Using Multiplexed Two-Wave Mixing InterferometryUsing Computer Vision to Map Laser Ultrasound onto CAD Geometries

ELECTROMAGNETIC, THERMAL, AND X-RAY TECHNIQUESSection A. Eddy Current Modeling, Probes, and ArraysCoupled Transient Thermal and Pulsed EC Modeling for NDT of Materials Subjected to Laser Based Heat TreatmentSimulation of Eddy-Current Corrosion Detection Using a Sensor ArraySimulation of Ultrasonic and Electromagnetic Nondestructive Evaluation of CorrosionElectric Potential in Eddy Current TestingAdvances in Modeling Eddy-Current NDE of Ferromagnetic BodiesPulsed Eddy-Current NDE at Iowa State University Recent Progress and ResultsDevelopment of Eddy Current Probe for Thick-Walled Plates and Quantitative Evaluation of CracksAn RFECT Probe with a Superconducting ShieldA New Eddy Current Surface Probe for Short Flaws with Minimal Lift-Off NoiseGMR Magnetic Sensor Arrays for NDE Eddy-Current TestingEddy Current Testing of Thick Aluminum Plates with Hidden CracksExperimental Verification of Model-Based ECT Signal Interpretation for Quantitative Flaw Characterization in Steam Generator Tubes

Section B. Microwave and SQUID NDENear-Field Microwave and Embedded Modulated Scattering Technique (MST) for Dielectric Characterization of MaterialsNano Spatial Resolution with 60 GHz Near-Field Scanning Millimeter-Wave MicroscopeNear-Field Scanning Microwave Microscope Using a Dielectric ResonatorNear-Field Microwave Detection of Corrosion Precursor Pitting under Thin Dielectric Coatings in Metallic SubstrateInfluence of Cyclical Soaking in Chloride Bath and Drying of Mortar on its Microwave Dielectric Properties: The Forward ModelMicrowave Radar Detection of Gas Pipeline LeaksDetecting Incipient Fatigue Damage with Scanning SQUID Microscopy

Section C. Thermosonics and Thermal Wave ApplicationsLaser Vibrometry and Strain Gage Measurements of Thermosonic ActivationRecent Developments in Sonic IR ImagingOptimizing the Thermosonics SignalSonic IR Imaging and Vibration Pattern Studies of Cracks in an Engine DiskAcousto-Thermal Microstructure CharacterizationModeling of Effects of Excitation Velocities on the Thermal Image Obtained for Thermosonic N D EUltrasound Burst Phase Thermography (UBP) for Applications in the Automotive Industry

Section D. X-Ray Techniques and ApplicationsSpecial Features in Radiography Accessed by 3D Monte Carlo ModelA Monte Carlo Analysis of the Lead Screen Impact on Film Radiography Image FormationNew Developments in Virtual X-Ray Imaging: Fast Simulation Using a Deterministic ApproachIssues with the High Energy Radiography SimulationsAutomated Identification of Intergranular Corrosion in X-Ray CT Images

SIGNAL AND IMAGE PROCESSING, SIZING, INVERSION, AND RECONSTRUCTIONSection A. Signal and Image ProcessingData Fusion Method for the Optimal Mixing of Multi-Frequency Eddy Current SignalsDesign of an NDE Integrated Data Acquisition System (NIDAS)Time-of-Flight Measurements from Eddy Current TestsWavelet for Ultrasonic Flaw Enhancement and Image CompressionNeural Network Analysis for Evaluating Welding ProcessNew Modular Ultrasonic Signal Processing Building Blocks for Real-Time Data Acquisition and Post ProcessingSwept Frequency Multiplication (SFM) Techniques for Improved Air-Coupled Ultrasonic NDEModel-Based Enhancement of the TIFD for Flaw Signal Identification in Ultrasonic Testing of Welded JointsMechanical and Electrical Noise in the PVLAS ExperimentStatistical Characterization of Multi-Phase Flow by Dynamic TomographyApplication of Taguchi Methodology for Optimizing Test Parameters in Magneto-Optic ImagingUltrasonic Array Approach for the Evaluation of Electrofusion Joints of Polyethylene Gas PipingUltrasonic Tomographic Imaging of Air Flow in Pipes Using an Electrostatic Transducer ArrayIdentification of Thermal and Optical Effects for the Detection of Open-Crack in Photothermal Nondestructive TestingA New Approach to Thermal TomographyAutomatic Evaluation of Welded Joints Using Image Processing on Radiographs3D Modeling of a Magneto-Optic Imager by a Dyadic Green's Functions Approach

Section B. Sizing, Inversion, and ReconstructionQuantitative Approaches to Flaw Sizing Based-On Ultrasonic Testing ModelsGroove Sizing Using a Robust Neural Network ApproachEstimation of Defect's Geometric Parameters with a Thermal MethodApplication of Linearized Inverse Scattering Methods to the Material with Flat Measurement SurfaceThree Dimensional Born and Kirchhoff Inversions for Shape Reconstruction of DefectsApplication of Optimization Methods to Crack Profile Inversion Using Eddy CurrentsReconstruction of 3D Flaws in Eddy Current Quantitative Nondestructive Evaluation

ULTRASONIC ARRAYS, TRANSDUCERS, SENSORS, TECHNIQUES, AND DEVICESSection A. Ultrasonic ArraysGuided Wave Beam Steering from Omni-Directional Transducer ArraysComparative Evaluation between Ultrasonic Phased Array and Synthetic Aperture Focusing TechniquesUltrasonic NDT Simulation Tools for Phased Array TechniquesDevelopment and Optimization of a Rotating Phased Array Inspection SystemPhased Arrays Techniques and Split Spectrum Processing for Inspection of Thick Titanium Casting ComponentsGroundwork for Rail Flaw Detection Using Ultrasonic Phased Array InspectionCorrosion Monitoring of Airframe Structures Using Ultrasonic Arrays and Guided WavesAn EMAT Array for the Rapid Inspection of Large Structures Using Guided Waves

Section B. Transducers, Sensors, and ProbesNew Piezocomposite Transducers for Improvement of Ultrasonic InspectionsA Modular Multi-Gaussian Beam Model for Isotropic and Anisotropic MediaSimilarity of Gaussian and Piston Transducer VoltagesApplication of a Giant Magnetoresistive (GMR) Sensor for Characterization of Corrosion in a Laboratory SpecimenLow-Coherence Optical Probe for Non-Contact Detection of Photothermal and Photoacoustic Phenomena in BiomaterialsDesign of a Fiber Bragg Based Measurement System for Strain and Temperature Monitoring

Section C. Techniques and DevicesModulation Enhanced Detectability of Cracks Using Surface Acoustic WavesAcoustic Interferometer for Localized Rayleigh Wave Velocity MeasurementsDevelopment of a Manual Air-Coupled Ultrasonic Inspection Instrument for Use on Aeronautical Structures Under In-Service ConditionsDevelopment of a High Performance Acousto-Ultrasonic Scan SystemAcquisition of Time of Flight Diffraction Data by Mobile RobotsSequential Evaluation of QNDE Devices for Underground Storage TanksDevelopment of Modeling and Simulation for Magnetic Particle Inspection Using Finite ElementsThree-Channel Non-Force Magnetic MicroscopeDevelopment of a Magnetic NDE Imaging System Using Magnetoresistive DevicesQuantitative Evaluation of a Crack Inside of Pressure Pipeline by Shearography and ESPIDevelopment and Testing of Prototype Giant Magnetoresistive (GMR) Rotating Probe System

ENGINEERED MATERIALS AND COMPONENTSSection A. Composites and ComponentsPrediction of Ultrasonic Fields into Composite Multi-Layered Structures: Homogenization Approach for the Direct Field and Statistical Approach for theOptical Absorption of Epoxy Resin and its Role in the Laser Ultrasonic Generation Mechanism in Composite MaterialsNon-Contact Inspection of Composites Using Air-Coupled UltrasoundThermographic Depth Profiling of Delaminations in CompositesImaging of Fatigue Damage in CFRP Composite Laminates Using Nonlinear Harmonic GenerationAn Experimental Study on the Impact Collapse Characteristics of CF/Epoxy Circular TubesInfrared Scanning of FRP Composite MembersNondestructive Evaluation of Ceramic Matrix Composite Combustor ComponentsDevelopment of Nondestructive Inspection Methods for Composite Repair

Section B. NDE for Bonds, Interfaces, and CoatingsModeling of Transmitted Ultrasonic Signals Through Bonded Aluminum SkinA Comparison of the Detectability of Dry Contact Kissing Bonds in Adhesive Joints Using Longitudinal, Shear and High Power Ultrasonic TechniquesNonlinear Angle Beam Ultrasonic Evaluation of Adhesive BondsThe Development of Shear and Compression Elastic Moduli in Curing Epoxy Adhesives Measured Using Non-Contact Ultrasonic TransducersRoughness-Caused Dispersion of High Frequency Surface Acoustic Waves on Crystal MaterialsThe Interaction of Lamb Waves with Solid-Solid InterfacesThe Use of Ultrasound to Measure Contact Stiffness and Pressure in Large Contacting InterfacesDetection of Weak Interface Signals for Same Material Bond/Weld InspectionTransmission and Reflection of AO Mode Lamb Wave in a Plate OverlapA Comparison of Ultrasonic Wave Reflection/Transmission Models from Isotropic Multi-Layered Structures by Transfer-Matrix and Stiffness-Matrix RecursiApplication of Fast Multipole Boundary Element Method to Scattering Analysis of SH Waves by a Lap JointCharacterization of Weathering Degradation in Aircraft Polymeric Coatings Using NDE Imaging Techniques

Section C. Infrastructure Materials, Components, and SystemsUltrasonic Methodology to Characterize the State of CureCrack Depth Measurement in Concrete Using Surface Wave TransmissionMicrowave Reflection and Dielectric Properties of Mortar Exposed Periodically to 2% Chloride Salt Solution and Compression ForceInspection of Steel Tendons in Concrete Using Guided WavesRadar Detection of Rebars Including Use of Neural Networks and Horn AntennasCalculation and Measurement of Ultrasonic Attenuation for Distributed Cracks in a SolidNondestructive Evaluation of Double Bevel T-Joint by Tandem Array Ultrasonic TransducerNDE of FRP Wrapped Timber Bridge Components Using Infrared ThermographyNDE of FRP Bridge Beams and DecksImproving the Accuracy of Impact-Echo in Testing Post-Tensioning DuctsThree-Dimensional Steady-State Green's Function for a Layered Isotropic PlateAxisymmetric Modes that Propagate in Buried Iron Water PipesDesigning and Building to "Impossible" Tolerances for Vibration Sensitive Equipment

MATERIALS CHARACTERIZATIONSection A. Materials Properties, Microstructure, and DegradationPrincipal Surface Wave Velocities in the Point Focus Acoustic Materials Signature V(z) of an Anisotropic SolidErrors in the Measurement of Ultrasonic Phase Velocity in the Context of Materials EvaluationA Guided Wave Technique for the Characterization of Highly Attenuative Viscoelastic MaterialsParabolic Mirror and Air-Coupled Transducer for Multimodal Plate Wave DetectionCharacterization of Solution Annealing Behaviour in Titanium Alloys by Ultrasonic Velocity MeasurementsIdentification of Sintered Irons with Ultrasonic NonlinearityUltrasonic Backscattering Profiles on Periodically Rough InterfaceMaterials Characterization Using Reconstructed Thermographic DataInvestigation of Velocity Surface of Bulk Acoustic Waves for Proton-Exchanged LiNbO3 CrystalCalculation of the Saw Velocity Change of Proton-Exchanged LiNb03CrystalInvestigation of Thermal Diffusivity of Nano-Structured TiO2FilmsA Study on Point Defects of ZnSe/Ga Sa Epilayer Obtained from Photoluminescience MeasurementA Study on Properties of Infrared Detector for a HGCDTE Epilayers Using Photocurrent MeasurementElectrostatic Detection of Density Variations in Green-State Powder Metallurgy CompactsA Study of Drying and Cleaning Methods Used in Preparation for Fluorescent Penetrant Inspection-Part IA Study of Drying and Cleaning Methods Used in Preparation for Fluorescent Penetrant Inspection-Part IIModeling Ultrasonic Grain Noise within Ti-6Al-4V ForgingsSimultaneous Measurement of Grain Size and Shape from Ultrasonic Backscattering Measurements Made from a Single SurfaceCorrelation between Local Ultrasonic Properties and Grain Size within Jet-Engine Nickel Alloy BilletsEffect of Surface Curvature on Backscattered Ultrasonic Grain Noise in Titanium ForgingsLocalization of Dwell Fatigue Cracks in Ti-6242 Alloy SamplesA Lamb Wave Study on Thermal Damage in a Degraded PlateDependence of the Perpendicular Residual Leakage Magnetic Flux Density on Fatigue Damage in an Austenitic Stainless SteelMonitoring Fatigue Damage Accumulation with Rayleigh Wave Harmonic Generation MeasurementsEstimation of Degradation of Strength Properties of a Material of Structures Intended for a Long Service Life Using a Combination of Mathematical ModeComparison of Ultrasound with Tensile Testing of Thermally Damaged Polyimide Insulated Wiring (MIL-W-81381)Non-Baseline Damage Detection from Changes in Strain Energy Mode Shapes. Experiments on Armored Vehicle Launched Bridge

Section B. Applications of Acoustic Emission for Materials CharacterizationMultifrequency Acoustic Emissions (AE) for Monitoring the Time Evolution of Microprocesses within SolidsRelation between Amplitude and Duration of Acoustic Emission SignalsAcoustic Emission Technique for Characterizing Deformation and Fatigue Crack Growth in Austenitic Stainless SteelsAcoustic Emission Performance for Damage Monitoring of Impacted FRP Composite LaminatesDetection and Identification of Concrete Cracking in Reinforced Concrete by Acoustic Emission

Section C. Applications of Laser Ultrasonics for Materials CharacterizationLaser Ultrasound: An Inspection Tool of Soft Porous Low-Dielectric Constant Films for Microelectronic InterconnectLaser Ultrasonic Inspection of Compositionally Graded Mullite CoatingsNonlinear Laser Ultrasonic Measurements of Localized Fatigue DamageLaser Ultrasonic Study of Crack Tip DiffractionVisualization of Surface-Breaking Tight Cracks by Laser-Ultrasonic F-SAFTLaser Ultrasonic Characterization of Residual Stresses in Thin FilmsRepeatability of an Ablation Source Using a Time-Frequency Representation

Section D. Stress and TextureCharacterization of Plastically Deformed Steel Utilizing EMAT Ultrasonic Velocity MeasurementsMonitoring of the Level of Residual Stress in Surface-Treated Specimens by a Noncontacting Thermoelectric TechniqueResidual Stress Relaxation Due to Fretting Fatigue in Shot Peened Surfaces of Ti-6Al-4VA New Model Equation for Interpreting the Magnetomechanical Effect Using a Generalization of the Rayleigh LawStress Evaluation by Chaotic Characteristic of Barkhausen NoiseThe Effect of Plastic Deformation and Residual Stress on Magnetic Barkhausen Noise Signals in Mild SteelExperimental Observation of Linear and Non-linear Guided Wave Propagation in Rolled Aluminum SheetsTheoretical Aspects of Linear and Non-linear Guided Wave Propagation in Rolled Aluminum Sheets

NDE APPLICATIONSSection A. "Smart" Systems for Structural Health MonitoringA Hybrid Approach to Structural Health Monitoring Using Nondestructive Evaluation (NDE) and Active Damage InterrogationSensor Technology for Integrated Vehicle Health Management of Aerospace VehiclesLasernet Fines Wear Debris Analysis Technology: Application to Mechanical Fault DetectionFatigue and Stress Monitoring Using Scanning and Permanently Mounted MWM-ArraysConcepts for an Integrated Vehicle Health Monitoring SystemPlate-Wave Diffraction Tomography for Structural Health MonitoringEmbedded Ultrasonic Transducer Design and Wireless Communications for Intelligent Monitoring of StructuresDevelopment of an Intelligent Electromagnetic Sensor to Detect Ferrous Corrosion Products Under Structural Coatings

Section B. NDE for Process Characterization and ControlUsing Ultrasonic Diffraction Grating Spectroscopy to Characterize Fluids and SlurriesCharacterization of Solid Liquid Suspensions Utilizing Ultrasonic MeasurementsArtificial Neural Network Based Algorithm for Acoustic Impact Based Nondestructive Process Monitoring of Composite ProductsThe Consistency of Phenomenological Models of Ultrasonic Wave Propagation in a Curing ThermosetA Laboratory Laser-Ultrasonic Instrument for Measuring the Mechanical Properties of Paper WebsAn Ultrasonic Meter to Characterize Degree of Fouling and Cleaning in Reverse Osmosis FiltersLaser-Ultrasonic Characterization of the Annealing Process of Low-Carbon SteelWide Bandwidth Air-Coupled Ultrasonic Testing of Food Containers in AirDepth Profiling of Machined Surfaces Using Cross Correlation of Barkhausen Noise Butterfly CurvesLocating LCD Glass for Robotic HandlingOptical Low-Coherence Reflectometry for Nondestructive Process MeasurementsDefect Signal Enhancement in Inspection Lines by Magnetic Flux LeakageOn-Line NDE for Advanced Reactor Designs

Section C. NDE for Bio MaterialsNeural and Decision Theoretic Approaches for the Automated Segmentation of Radiodense Tissue in Digitized MammogramsCT Based Radiography Simulations for Both Industrial and Medical RadiographyNoninvasive Imaging for Tissue Characterization and Hyperthermia ThermometryAttenuation Coefficient Estimation Using Equivalent Diffraction Points with Multiple Interface Reflections

BENCHMARK COMPARISONS, RELIABILITY, AND EDUCATIONSection A. Benchmark Problems2002 Ultrasonic Benchmark Problem: Overview and Discussion of ResultsModeling Ultrasonic Problems for the 2002 Benchmark SessionPrediction of Insonifying Velocity Fields and Flaw Signals of the 2002 Ultrasonic Benchmark ProblemsPrediction of Transient Flaw Signals of the Ultrasonic Benchmark ProblemUltrasonic Benchmark Problem: Application of a Paraxial Model to Side-Drilled Holes and Oblique IncidenceEvaluation of Standard Configurations for Nondestructive Eddy-Current TestingSimulation of the World Federation's Second Eddy Current Benchmark ProblemMFL Benchmark Problem 2: Laboratory MeasurementsNumerical Predication of Signal for Magnetic Flux Leakage Benchmark Task

Section B. Reliability POD Jack Lincoln Memorial SessionStatistical Analysis of Probability of Detection Hit/Miss Data for Small Data SetsRecent Studies on the POD Analysis of "a vs. a" NDI DataReliability Assessment of Ultrasonic Nondestructive Inspection Data Using Monte Carlo SimulationThe Application of Case-Based Reasoning to Ultrasonic Data for Use in Inspection QualificationEvaluation of Multizone Inspection Variability at the Supply Base for 8 Inch-Diameter Ti-6Al-4V: A Round-Robin StudyAutomated Error Interpretation in Flaw Response Models for Ultrasonic NDTInteractive Software Tools for Inspection QualificationFluorescent Penetrant Inspection Probability of Detection Demonstrations Performed for Space Propulsion

Section C. NDE EducationCollaboration for Nondestructive Testing EducationExtending the Reach

HelpSearchExit