Search

Accuracy of noise and signal models utilized in eddy current (EC) signal analysis plays a significant role in the performance. While many studies focus on development of accurate models for different EC probe signals, noise is commonly modeled using simplistic Gaussian models. Accurate noise model is as significant as the signal model in detection and classification tasks. In this paper, statistical models for representation and synthesis of EC measurement are developed. Filtering algorithms for different types of noise are also discussed. Considering EC signals as complex data, anomaly signal and random noise are modeled in the complex plane in rectangular form using Gaussian Mixture Models (GMM). Analysis results show that GMM along with noise filtering algorithms enhance performance of noisy EC signal analysis. [ABSTRACT FROM AUTHOR]

For the control of fuel channels of PHWR CANDU type, a novel type of rotating magnetic field eddy current transducer was developed, as well as the afferent electronic equipment. So, the position and the tilt of garter springs using neuro-fuzzy network could be determined, and the presence of the discontinuities on external and internal surface of the tubes can be emphasized. A numerical code for the solving of the forward problem was developed. © 2005 American Institute of Physics [ABSTRACT FROM AUTHOR]

NDE techniques are each characterized by their individual strengths and limitations that are dictated by the underlying physics. Data Fusion techniques that combine information from multiple sensors can help in exploiting the capabilities of multiple NDE modalities. Hence, there is need for a system that can perform multi-sensor data acquisition and processing. This paper describes an NDE Integrated Data Acquisition System (NIDAS) that can perform an ultrasonic and eddy current testing simultaneously. C-scan images from both inspections maybe obtained from the same scan operation, and these images are spatially correlated. Images obtained in this manner are well suited for data fusion applications and other image and data processing algorithms for enhancing the testing results. A data fusion example is presented using an eddy current and an ultrasonic C-scan image obtained from the NIDAS. The result shows significant enhancement on the defect profile. © 2003 American Institute of Physics [ABSTRACT FROM AUTHOR]

Eddy current (EC) testing methods are widely used in a variety of applications including the inspection of steam generator tubes in nuclear power plants, aircraft parts and airframes. A key factor that affects the EC signal is lift-off. In practice, it is difficult to keep track of the actual value of the liftoff, which is essential for accurate interpretation of the signal. Hence it is necessary to have a scheme to render the EC signal invariant to the effects of lift-off. This paper describes a new method for compensating EC signals for variations in lift-off. The signals are transformed to obtain a zero lift-off equivalent signal that can be subsequently used for defect characterization. [ABSTRACT FROM AUTHOR]

Iterative approaches using numerical forward models are commonly used for solving inverse problems in nondestructive evaluation. The drawbacks of these approaches include their high computational cost and the difficulty in computing gradients for updating defect profiles. This paper proposes a finite element neural network (FENN) that embeds finite element models into a neural network format. This approach enables fast and accurate solution of the forward problem. The FENN can then be used as the forward model in an iterative approach to solve the inverse problem. Gradientbased optimization methods are easily applied since the FENN provides an explicit functional mapping between the defect profile and the measured signal. Results of applying the FENN to several simple electromagnetic forward and inverse problems are presented. [ABSTRACT FROM AUTHOR]

In this paper, we present LEARN++, a new algorithm that allows existing classifiers to learn incrementally from new data without forgetting previously acquired knowledge. LEARN++ is based on generating multiple classifiers, each trained with a different subset of the training data and then combining them to form an ensemble of classifiers using weighted majority voting. The fundamental contribution of the algorithm lies in the manner in which the training data is partitioned. The results demonstrate the feasibility and effectiveness of the approach in learning from new data. © 2001 American Institute of Physics. [ABSTRACT FROM AUTHOR]

Magneto-optic imaging is a relatively new sensor application of bubble memory technology to NDI. The Magneto-Optic Imager (MOI) uses a magneto-optic (MO) sensor to produce analog images of magnetic flux leakage from surface and subsurface defects. The flux leakage is produced by eddy current induction techniques in nonferrous metals and magnetic yokes are used in ferromagnetic materials. The technique has gained acceptance in the aircraft maintenance industry for use to detect surface-breaking cracks and corrosion. Until recently, much of the MOI development has been empirical in nature since the electromagnetic processes that produce images are rather complex. The availability of finite element techniques to numerically solve Maxwell’s equations, in conjunction with MOI observations, allows greater understanding of the capabilities of the instrument. In this paper, we present a systematic set of finite element calculations along with MOI measurements on specific defects to quantify the current capability of the MOI as well as its desired performance. Parametric studies including effects of liftoff and proximity of edges are also studied.—This material is based upon work supported by the Federal Aviation Administration under Contract #DTFA03-98-D-00008, Delivery Order #IA013 and performed at Iowa State University’s Center for NDE as part of the Center for Aviation Systems Reliability program. © 2000 American Institute of Physics. [ABSTRACT FROM AUTHOR]

Diffusion tensor magnetic resonance imaging (DT-MRI) has recently gained popularity because of its capabilities in axonography of the central nervous system. Fast imaging sequences are used to reduce motion induced distortion effects on the diffusion signal, resulting in low signal levels, loud acoustic noise, and occasional peripheral nerve stimulation. Eddy current induced by diffusion gradient pulses is also a challenge to DT-MRI. Magnetic field associated with the eddy current is a major source of artifacts in scanner images. This paper introduces a finite element modeling of electromagnetic and acoustic fields in DT-MRI sequences. The analysis involves three dimensional modeling of the scanner and its interaction with pulses applied to gradient coils. Efficient modeling of induced fields is essential in optimizing parameter settings and improving performance of this imaging modality. © 2006 American Institute of Physics [ABSTRACT FROM AUTHOR]

This paper presents computational results of the second eddy current benchmark problem proposed by the World Federation of Nondestructive Evaluation Centers. The problem involves the prediction of the change in impedance of a pancake coil as it scans the inner surface of a metal tube along both the axial and circumferential directions for detection of flaws on the outer surface. A finite element scheme is carried. The real and imaginary part of the impedance is obtained by observing the total stored energy and the dissipated energy, respectively. © 2003 American Institute of Physics [ABSTRACT FROM AUTHOR]

Multi-frequency eddy current techniques are commonly used in inspecting steam generator tubes in nuclear power plants. The measurement usually contains dominant contributions from external tube support plates that tend to mask signals from defects. This paper proposes the use of a non-linear mixing technique using radial basis function neural networks to suppress the TSP signals. Initial results presented in this paper look very promising and demonstrate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

This paper presents an adaptive Wiener filter based technique for detection of defect locations from (noisy) bobbin coil eddy current data. First, the data is de-noised using an improved adaptive Wiener filter that significantly enhances the defect signals. Next, a statistical signal detection technique is used to detect the probable defect points from the enhanced data. The results of experiments conducted on some real world data sets indicate that the method is quite promising and useful for automated processing and classification of eddy current data. [ABSTRACT FROM AUTHOR]

Integral equation (IE)-based techniques have been widely used for computing the magneto-static field in both the linear and non-linear regime. One disadvantage of IE-based techniques is that they result in dense matrices, the solution of which is computationally prohibitive. Indeed, both the computational time and memory scale as O(N[sup 2]), where N is the number of spatial unknowns that the body is discretized into. The computational complexity has been the primary hurdle towards the popularity of such solvers. Our goal is to develop a fast solution scheme for magneto-static field computations. In this paper, a novel integral equation formulation has been used and a version of the fast multipole algorithm has been incorporated in the overall solution procedure. © 2001 American Institute of Physics. [ABSTRACT FROM AUTHOR]

Magneto-optic imaging (MOI) is a relatively new sensor application of an extension of bubble memory technology to NDT and produce easy-to-interpret, real time analog images. MOI systems use a magneto-optic (MO) sensor to produce analog images of magnetic flux leakage from surface and subsurface defects. The instrument’s capability in detecting the relatively weak magnetic fields associated with subsurface defects depends on the sensitivity of the magneto-optic sensor. The availability of a theoretical model that can simulate the MOI system performance is extremely important for optimization of the MOI sensor and hardware system. A nodal finite element model based on magnetic vector potential formulation has been developed for simulating MOI phenomenon. This model has been used for predicting the magnetic fields in simple test geometry with corrosion dome defects. In the case of test samples with multiple discontinuities, a more robust model using the magnetic vector potential A¯ and electrical scalar potential V is required. In this paper, a finite element model based on A-V formulation is developed to model complex circumferential crack under aluminum rivets in dimpled countersink. © 2001 American Institute of Physics. [ABSTRACT FROM AUTHOR]

Neural network based signal classification systems are being used increasingly in the analysis of large volumes of data obtained in NDE applications. One example is in the interpretation on ultrasonic signals obtained from inspection of welds where signals can be due to porosity, slag, lack of fusion and cracks in the weld region. Standard techniques rely on differences in individual A-scans to classify the signals. This paper proposes an ultrasonic signal classification technique based on the information in a group of signals and examining the statistical characteristics of the signals. The method was 2-dimensional signal processing algorithms to analyze the information in B- and B′-scan images. In this paper, 2-dimensional transform based coefficients of the images are used as features and a multilayer perceptron is used to classify them. These results are then combined to get the final classification for the inspected region. Results of applying the technique to data obtained from the inspection of welds are presented. © 2001 American Institute of Physics. [ABSTRACT FROM AUTHOR]

In eddy current inspection of jet engine disk slots, there are several factors contributing to the degradation of eddy current signals. First, cracks often occur in high stress regions near slot edges where contribution of edge signals overwhelms crack signals. Furthermore, the rounded chamfer edges are hand finished and the curvature of the edges varies from slot to slot. In terms of ECI, this translates into different liftoff signals from slot to slot, for any given disk. Some disk slot surfaces are also shot peened which creates a dimple-like surface finish. Depending on the size and depth of dimples relative to the probes used, significant surface noise can be observed in the eddy current signal. In addition, mechanical jittering of the scanner, probe misalignment and fundamental difference in probe variance result in unwanted lift-off and instrument noise. Therefore, detection of cracks in jet engine disk slots requires a fairly robust algorithm. This paper presents a novel approach for crack detection and noise reduction in eddy current engine disk inspection signals. The eddy current signal can be viewed as a curve in 3-dimensional space: real-imaginary-spatial. Noise reduction is done by adjusting coordinates of the points along the curve in the direction that reduces local curvature. As a result, high frequency noise is drastically removed. After noise removal, the filtered eddy current signal is projected onto the impedance plane where embedded crack signals exhibit themselves as small loop features. A simple loop detection algorithm is employed to detect such crack features. The areas of detected loops are used as criteria for crack classification. Experimental results on real data is presented.—This material is based upon work supported by the Federal Aviation Administration under Contract #DTFA03-98-D-00008, Delivery Order #IA015 and performed at lowa State University’s Center for NDE as part of the Center for Aviation Systems Re... [ABSTRACT FROM AUTHOR]

Filtered backpropagation (FBPP) is a well-known technique used in Diffraction Tomography (DT). For accurate reconstruction using FBPP, full 360o angular coverage is necessary. However, it has been shown that using some inherent redundancies in the projection data in a tomographic setup, accurate reconstruction is still possible with 270° coverage which is called the minimal-scan angle range. This can be done by applying weighing functions (or filters) on projection data of the object to eliminate the redundancies and accurately reconstruct the image from 270° coverage. This paper demonstrates procedures to generate many general classes of these weighing filters. These are all equivalent at 270° coverage but vary in performance at lower angular coverages and in presence of noise. This paper does a comparative analysis of different filters when angular coverage is lower than minimal-scan angle of 270°. Simulation studies have been done to find optimum weight filters for sub-minimal angular coverage (<270°). [ABSTRACT FROM AUTHOR]

Microwave tomography has gained significant attention due to its reliability and unhazardous nature in the fields of NDE and medical industry. A new microwave tomography system is presented in this paper, which significantly reduces the design and operational complexities of traditional microwave imaging systems. The major component of the proposed system is a reconfigurable reflectarray antenna which is used for beam steering in order to generate projections from multiple angles. The design, modeling and fabrication of the building block of the antenna, a tunable unit cell, are discussed in this paper. The unit cell is capable of dynamically altering the phase of the reflected field which results in beam steering ability of the reflectarray antenna. A tomographically reconstructed image of a dielectric sample using this new microwave tomography system is presented in this work. [ABSTRACT FROM AUTHOR]

The detection of a crack under fastener heads (CUF) in a multi-layered aircraft structure remains a challenge in non-destructive evaluation (NDE). An EC-GMR system using a linear eddy current (EC) coil with giant magnetoresistive (GMR) sensors located on the axis of symmetry is proposed for detecting discontinuities in conducting materials. The signal received from sensors is greatly influenced by the interrogating field. This paper describes a detailed parametric study, using a finite element model predicted signals in conjunction with a multi-parameter optimization problem for coil design. The sensor performance is assessed using quantitative measures based on Probability of detection (POD) with respect to different crack geometries. [ABSTRACT FROM AUTHOR]

Microwave tomography is becoming a popular imaging modality in nondestructive evaluation and medicine. A commonly encountered challenge in tomography in general is that in many practical situations a full 360o angular access is not possible and with limited access, the quality of reconstructed image is compromised. This paper presents an approach for reconstruction with limited angular access in diffraction tomography. The algorithm takes advantage of redundancies in image Fourier space data obtained from diffracted field measurements and couples it to an error minimization technique using a constrained total variation (CTV) minimization. Initial results from simulated data have been presented here to validate the approach. [ABSTRACT FROM AUTHOR]

Microwave tomography is a fast-growing technique in the fields of NDE and medical industry. This paper presents a new microwave tomography system which reduces the complexities of conventional microwave imaging systems by utilizing a reconfigurable mirror, a tunable reflectarray antenna. In order to build a tunable reflectarray with beam steering capabilities, the unit cell characteristics should dynamically alter. Modelling and experimental results of a single unit cell are presented in this work. [ABSTRACT FROM AUTHOR]

The detection of embedded cracks under fastener heads (CUF) in multilayer structures is a major challenge facing the aviation industry. Eddy current (EC) based techniques, such as Magneto-Optic Imager, Sliding probe and Self-nulling EC probes, and low frequency EC methods have been investigated. The prior work of authors presented an EC system using a uniform field excitation combined with giant magnetoresistive (GMR) sensors for imaging the normal component of magnetic flux density, has improved the effectiveness in detecting 2nd and 3rd layer CUFs with aluminum fasteners. However, detection of CUFs with ferromagnetic fasteners remains a major challenge. This paper presents the experimental study of automatic crack detection under steel fastener sites using low frequency EC with 3D GMR sensors that measure all 3 components of the induced magnetic flux density. A finite element model based study is used to optimize the system design and the experimental validation is presented. [ABSTRACT FROM AUTHOR]

Inspection of steam generator tubes in nuclear power plants is extremely critical for safe operation of the power plant. In the nuclear industry, steam generator tube inspection using eddy current techniques has evolved over the years from a single bobbin coil, to rotating probe coil (RPC) and array probe, in an attempt to improve the speed and reliability of inspection. The RPC probe offers the accurate spatial resolution but involves complex mechanical rotation. This paper presents a novel design of eddy current probes based on rotating fields produced by three identical coils excited by a balanced three-phase supply. The sensor thereby achieves rotating probe functionality by electronic means and eliminates the need for mechanical rotation. The field generated by the probe is largely radial that result in induced currents that flow circularly around the radial axis and rotating around the tube at a synchronous speed effectively producing induced eddy currents that are multidirectional. The probe will consequently be sensitive to cracks of all orientations in the tube wall. The finite element model (FEM) results of the rotating fields and induced currents are presented. A prototype probe is being built to validate simulation results. [ABSTRACT FROM AUTHOR]

Continuous structural health monitoring (SHM) uses permanently mounted sensor networks on critical locations of a structural component. In-situ wired sensors require a large amount of cabling for power and data transfer, which can drive up costs of installation and maintenance. Hence the need for developing wireless sensors for SHM. The major obstacles preventing the widespread use of wireless sensor networks (WSN) for SHM is the availability of portable, low cost, low powered, low footprint, and high SNR based instrumentation. This paper presents a wireless sensor system that could be interfaced with piezoelectric transducers for the identification of anomalous events using ultrasonic techniques. Power aware algorithms are used to coordinate the actuator-sensor network interaction with a central processing server, where appropriate signal processing techniques are used to quantify the damage in terms of severity. [ABSTRACT FROM AUTHOR]

For continuous monitoring of power-plant components, the use of in-situ sensors (i.e., sensors that are permanently mounted on the structure) is necessary. In-situ wired sensors require an unrealistic amount of cabling for power and data transfer, which can drive up costs of installation and maintenance. In addition, the use of cabling in hostile environments (high temperature/pressure environments) is not a viable option. This paper presents a wireless system for continuous monitoring, identification of anomalous events, NDE data acquisition and data transfer. NDE sensors are integrated with a wireless radio unit such as a MICA mote. Measurements from the sensors are typically acquired at prescribed intervals, encoded and compressed, and transmitted to a central processing server, where appropriate signal processing techniques may be used to filter out noise in the measurements, enhance the desired signal and quantify the damage in terms of severity. [ABSTRACT FROM AUTHOR]

Inverse problem solutions in NDE can be broadly classified as model-based approach and system-based approach. In model-based approach an accurate forward model is used in an iterative framework to provide a defect shape that minimizes the error between the measured signal and a simulated signal. However this approach results in repeated executions of a three dimensional forward model in each iteration, making it computationally demanding. This paper presents a direct approach to inversion using principles of time reversal. The feasibility of the approach is demonstrated via application to microwave NDE data. A two-dimensional finite difference time domain model for simulating the propagation of forward and time reversed wave fields is first developed. The key advantage of the approach is that it provides a model-based inversion method that is not iterative. Simulation and experimental results validating the approach are presented. [ABSTRACT FROM AUTHOR]

This paper presents a feasibility study of imaging defects in dielectric materials using principles of time reversal with microwaves. The principle is demonstrated using a two-dimensional finite difference time domain model for simulating the propagation of forward and time reversed wave fields. The scattered electric field is recorded on a linear array of receivers; time reversed and propagated backwards using the model to highlight the scatterer/defect. Simulations results validating the approach are presented. Initial results demonstrate the ability of the technique to image defects. [ABSTRACT FROM AUTHOR]

This paper introduces a model-based approach to reconstruct the three-dimensional defect profiles using eddy-current heat exchanger tube inspection signals. The method uses a Woodbury’s substructure finite element forward model to simulate the underlying physics, a state space defect representation, and a tree search algorithm to solve the inverse problem. The advantage of the substructure method is that it divides the whole solution domain into two substructures and only the region of interest (ROI) with dramatic material changes will be updated in each iterative step. Since the number of elements inside the ROI is very small compared with the number of elements in the entire mesh, the computational effort needed in both LU factorization and coefficient matrix assembly is reduced. Therefore, the execution time is reduced significantly making the inversion very efficient. The initial inversion results are presented to confirm the validity of the approach. [ABSTRACT FROM AUTHOR]

Eddy current testing (ECT) has been widely used in aircraft inspection. However ECT is limited by skin depth in detecting defects in multi-layered complex structures. Advanced electromagnetic sensing methods, such as giant magnetoresistive (GMR) sensors offer much higher sensitivity that allow detection of defects located deep in the test object. Traditional magnitude based GMR signal interpretation methods using symmetry of images developed for detecting cracks under titanium fastener heads are ineffective when analyzing data from steel fastener heads which results in reduced probability of detection (POD). This paper presents a signal processing algorithm based on mixing the in-phase and quadrature components of the GMR sensor signal via an optimum detection angle (ODA). Features derived from the mixed signal are clustered for automated crack detection using the MR sensor data. Performance of the algorithm is evaluated using signal to noise ratio (SNR) that measures efficacy of different features. [ABSTRACT FROM AUTHOR]

This work presents numerical results obtained by simulating the first eddy current benchmark problem proposed by the World Federation of Nondestructive Evaluation Centers (WFNDEC) as part of an international effort. The problem involves prediction of responses generated by eddy current coil sensors scanning an inconel tube containing a defect. A variety of defect geometries are considered both in the presence as well as absence of a ferromagnetic support plate. The simulations involved the use of A-V formulation based finite element method (FEM). © 2004 American Institute of Physics [ABSTRACT FROM AUTHOR]

Data fusion techniques are based on the premise that two or more carefully designed NDE tests are potentially capable of offering additional information concerning the test object relative to what can be garnered from a single test. However, information from a heterogeneous set of transducers cannot be fused unless the data is mapped onto a common ‘format’. The eddy current method cannot, for example, provide time-of-flight information that can be combined from estimates derived from ultrasonic tests. This paper presents a simple solution to address the problem by employing the so-called Q-transform to relate diffusive fields, such as those generated by eddy current probes, and propagating wave fields generated by ultrasonic NDT sensors. The paper illustrates how the distance between a defect and a source can be extracted from eddy current data generated using carefully selected excitation signals. Numerical results and comparisons with analytical predictions are presented. © 2003 American Institute of Physics [ABSTRACT FROM AUTHOR]

© 2002 American Institute of Physics. [ABSTRACT FROM AUTHOR]

This paper presents the numerical simulation results on an eddy current NDE benchmark problem proposed by the World Federation of NDE Centers. The problem involves the calculation of the impedance of a pair of differentially connected coils scanning an infinitely long tube. The excitation source is provided by feeding the coils with an AC current. An edge-based finite element method is used to simulate the problem numerically. Preliminary results obtained using the simulation model are presented. [ABSTRACT FROM AUTHOR]

Finite element methods (FEM) are widely used for modeling a variety of problems in Nondestructive Evaluation (NDE). For example, in the modeling of multilayer aircraft geometry with third layer cracks under fasteners, the reliance of FEM on a mesh leads to several problems, particularly when tight cracks have to be introduced into the sample. Furthermore remeshing is often required in handling probe motion can be time consuming. This paper presents a meshless elementfree Galerkin method (EFG), where the approximation is entirely constructed in terms of a set of nodes. The model is validated and applied to two dimensional magneto-static and eddy current NDT problems. [ABSTRACT FROM AUTHOR]