Comparison of radial basis function and back-error propagation neural networks for crack detection in variable cross-section beams

Abstract

In this study a method for identification of crack in variable cross-section beam is presented. The process of crack identification is consists of three steps. In first step, three natural frequencies of a variable cross-section beam for different locations and depths of cracks are obtained using Finite Element Method (FEM). In second step, two Back-Error Propagation neural networks (BEP) and two Radial Basis Function neural networks (RBF) are created and trained. The inputs of neural networks are first three natural frequencies and the outputs of first and second BEP and also RBF are corresponding locations and depth of cracks, respectively. In third step, some of natural frequencies of variable cross-section beam with distinct crack conditions are applied as inputs to trained neural networks. Finally obtained results of two types of neural networks are compared with each other. Computed results illustrate that computed cracks characteristics are in good agreements with actual data.