Masaaki Kurokawa

Rapid prediction of eddy current testing signals using A-φ Method and database

Abstract This paper describes a new scheme for the fast evaluation of eddy current testing (ECT) signals by using databases and formulation of the A−φ method. As the calculation of flaw signals is localized around the flaw region, substantial computational work can be reduced comparing with the conventional FEM–BEM code even for a conductor with complex geometry. Unlike the approach proposed by Badics et al. (Rapid flaw reconstruction scheme for 3-d inverse in eddy current NDE. In Studies in Applied Electromagnetics and Mechanics, Vol. 12, eds T. Takagi et al., IOS Press, 1997, pp. 303–309), the new scheme can predict ECT signals due to a large and complex-shaped crack rapidly because of its higher order interpolation and the newly introduced special element, say, an element with different media. The efficiency of the new method is verified by comparing its numerical results with the measured impedance for several bench-mark problems. The fast and accurate features make this new forward approach especially feasible in the reconstruction of flaw shapes by combining with the conventional optimization method.

Optimized eddy current detection of small cracks in steam generator tubing

A complete, computer based design methodology is described, aiming to develop an eddy current sensor with increased sensitivity to flaws, and reduced sensitivity to probe lift-off. The first part of the paper contains an analysis performed in order to establish detailed criteria for an effective design. Numerical investigations have been carried out and their results are discussed, regarding various problems of detectability and lift-off noise level. Based on these results, in the second part two probe arrangements are proposed, and it is shown how their performance parameters could be further improved.

A Distinctive Featured Optimization Approach for ECT Probes

The conventional ways to evaluate the detectability of ECT probes are usually based on some type of Maxwell equations with use of the FEM or BEM method[1]. Though these approachs can give relatively accurate eddy current field and corresponding impedance, the numerical calculation needs a lot of computer memory and CPU time. This causes them a drawback in ECT probe optimization procedures. Moreover, probe optimization is not only related to the best choice of some parameters of a given probe configuration, but the configuration parameters such as the number and arrangement of exciting and pick-up coils also need to be adjusted and modified. Since the numerical method can not give us a clear image to connect the eddy current to the exciting field, the discovery of a new excellent probe is difficult if this approach is not improved.