Analysis and Sizing of Fatigue Cracks in Metallic Structures from Eddy Current Probe Signals

Detection and sizing of fatigue cracks in metal structures surfaces are of the most important applications of non-destructive testing. One of non-destructive methods for this purpose, is eddy current. In this technique, an eddy current is created on the surface of the workpiece, using an alternating current (AC) conductor coil. The presence of cracks leads to disruption of distribution of this current to collapse and thus the variation in the impedance of the coil. Detection and sizing the cracks from impedance variations of the coil, which is called a reverse problem, is important to estimate the life of the work piece and its optimal use. In this paper, a phenomenon-based inverse model and an inverse algorithm based on the particle swarm optimization are proposed to detect the shape and depth of the cracks from impedance variations of the coil. The results obtained from the estimation of the shape of the cracks from the sizing data indicate that the proposed model is highly accurate in determining the shape and depth of the cracks