Rbf-Based Compensation Of Velocity Effects On Mfl Signals

In the magnetic flux leakage (MFL) non-destructive testing (NDT) technique, the velocity of the inspecting instruments could produce eddy currents during the sensing that might cause the distortion of the MFL signals. In this paper, the distortions of the sensing signals are analysed using a 2-D finite element model with velocity-induced eddy currents, and Radial Basis Function (RBF) neural networks are presented to compensate the velocity effects in order to obtain pure defect signals. This method has been verified by simulation.