Efficient Numerical Solver for Simulation of Pulsed Eddy-Current Testing Signals

IEEE Transactions on Magnetics (Impact Factor: 1.39). 12/2011; 47(11):4582 - 4591. DOI: 10.1109/TMAG.2011.2151872
Source: IEEE Xplore


The pulsed eddy-current testing (PECT) method has the promising capabilities for detecting defects and evaluating material properties. It achieves this through its rich variety of frequency components and large driving electric current. Efficient numerical simulation of PECT signals plays an important role in probe optimization and quantitative signal processing. This study primarily focuses on the development of an efficient numerical solver for PECT signals, and its validation via the consideration of the nondestructive testing problems of wall thinning defects in pipes of nuclear power plants. A frequency domain summation method combined with an interpolation strategy was proposed and implemented. It is based on the finite element method with edge elements. The number of total frequencies used in signal summation and the number of selected frequencies for interpolation were thoroughly discussed. In addition, a code using the time domain integration method was also developed for the signal prediction of a transient PECT problem. It was used for comparison with the frequency domain summation method. A comparison of numerical results of the two proposed simulation methods and experimental results indicates that both of these simulation methods can model PECT signals with high precision. However, the frequency domain summation method combined with an interpolation strategy is much more efficient in its use of simulation time.

1 Follower
15 Reads
  • [Show abstract] [Hide abstract]
    ABSTRACT: In this work Fe-1wt%Cu alloy samples exposed to thermal aging and cold-rolling were studied as an attempt to emulate material degradation in neutron irradiated nuclear reactor pressure vessel. The cold rolling was tested by pulsed eddy currents (PEC) using a directional probe which was rotated in the sample plane. Normalised PEC signals represented by typical "figures of eight" on the polar plot, showed very strong anisotropy which has a monotonic dependence on the cold work amount. Since the normalised PEC signals represent response due to electrical conductivity and do discard the ferromagnetic effect, it is concluded that the change is due to anisotropic magnetoresistivy of the rolled steel. The normalised difference PEC responses have been found to be described by quadratic dependence of cosine of in-plane angle shifted by an angle which is believed to correspond to slip planes of dislocations induced by rolling. The technique advantageously combines effects of rolling on electrical conductivity and magnetic permeability and does not require referencing to an untreated standard´┐Ż┬╗ which is important for future material and device health management. Keywords - pulsed eddy current, material ageing monitoring, health management.
    IEEE Transactions on Magnetics 01/2011; 49(1). DOI:10.1109/ICEMI.2011.6037936 · 1.39 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Reconstruction of Stress Corrosion Cracks (SCCs) using conventional Eddy Current Testing method (ECT) shows its limitation especially when dealing with deep SCCs. Recently, a new approach utilizing Pulsed Eddy Current Testing (PECT) signals has been proposed to reconstruct wall thinning defect based on a deterministic optimization method. It is because PECT is found advantageous over the conventional ECT due to its features of abundant frequency components and large exciting currents. In this study, stochastic optimization methods of neural network, tabu search, simulated annealing and genetic algorithm are introduced to reconstruct the SCC profile from the PECT signals. The efficiency and accuracy of these stochastic methods are evaluated and discussed.
    Electromagnetic Field Problems and Applications (ICEF), 2012 Sixth International Conference on; 01/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: The detection of concealed weapons is one of the biggest challenges facing homeland security. It has been shown that each weapon can have a unique fingerprint, which is an electromagnetic signal determined by its size, shape, and physical composition. Extracting the signature of each weapon is one of the major tasks of any detection system. In this paper, feature extraction of a new metal detector signal is conducted using a Wavelet and Fourier Transform. These features are used to classify two different groups of threat objects. Artificial Neural Network (ANN) and Support Vector Machines (SVM) classification techniques are used to classify the metal objects towards automatic threat object detection and classification. Promising classification accuracy rates are obtained from using individual and combined features.
    Imaging Systems and Techniques (IST), 2012 IEEE International Conference on; 01/2012
Show more