Shape identification of conductive anomalies by a new ECT data inversion algorithm

Eddy current testing (ECT) is a technique used to reconstruct the unknown resistivity of conductive materials in a non-destructive way. In this technique a time-varying current flowing in an exciting coil placed near to the specimen induces eddy currents in the specimen under testing. The induced eddy currents, depending on the spatial values of the resistivity and magnetic permeability, affect the signal detected by the surrounding pick-up coils or magnetic sensors. Information concerning the spatial distribution of the resistivity is then retrieved by inversion of the measured data. In the present paper we tackle the non-linear inverse problem of retrieving the shape of interfaces separating different phases for a conductive materials made of two different conducting phases. Hence we assume that the resistivity at a given spatial coordinate assumes either the value /spl eta//sub i/ or /spl eta//sub b/. The measurement system consists of a set of coils and the measurements are the mutual impedances between coils. The main contribution of this paper is an extension to ECT measurements of a new inversion algorithm, initially proposed for electrical resistance tomography (ERT) by Rubinacci et al. The key feature of this algorithm is that its computational cost increases linearly with the number of unknown parameters.