Identification of Multiple Cracks from Eddy Current Testing Signal with Noise Source by Image Processing and Inverse Analysis

摘要

This paper describes a novel method identifying the number and positions of cracks and reconstructing the crack shape. Using the ECT signal obtained by two-dimensional scanning as a picture image, a template matching method with help of genetic algorithms is applied to predict the number and positions of cracks. The present method employs a superposition of crack signals and a nonlinear scaling technique of a signal profile on crack length which are verified by numerical simulation. The number and positions of cracks are predicted sufficiently. Crack shape reconstructions from the predicted positions with help of inverse analysis are achieved with satisfactory accuracy.