Pre-processing methods for eddy current data analysis using Hilbert-Huang Transform

Abstract

An advanced nonlinear, non-stationary signal processing algorithm using Hilbert-Huang Transform is presented for automatic analysis of eddy current data collected during the inspection of steam generator tubes in nuclear power plants. The proposed algorithm reduces noise and enhances defect indication in the signal, thereby yielding increased detection probability as well as reduced false calls. Results of applying the algorithm on field data from SG tube inspection are presented and compared with those of current practice in industry.