Identification of default from eddy current testing Signals using multi output support vector Machine

This paper demonstrates the identification of crack size using signals obtained from eddy current testing Signals (ECT). The identification method is based on finite elements with the multi-outputs support vector machines (MO-SVM), The MO-SVM is a statistical learning method that has good generalization capability and learning performance. The finite element method (FEM) is used to create the data set required to train the MO-SVM and the particle swarm optimisation (PSO) is used to find the parameters of MO-SVM. Numerical simulations demonstrate that the MO-SVM method can determine the size of defect with an acceptable accuracy.