Quantitative nondestructive evaluation of material defect using GP-based fuzzy inference system

This paper deals with a quantitative nondestructive evaluation in eddy current testing for steam generator tubes of nuclear power plants by using genetic programming (GP) and fuzzy inference system. Defects can be detected as a probe impedance trajectory by scanning a pancake type probe coil. An inference system is proposed for identifying the defect shape inside and/or outside tubes. GP is applied to extract and select effective features from a probe impedance trajectory. Using the extracted features a fuzzy inference system detects presence, position, and size of a defect of a test sample. The effectiveness of the proposed method is demonstrated through computer simulation studies