Application of Artificial Neural Networks in Beam Damage Diagnosis

Abstract

This paper studies the application of Artificial Neural Networks (ANN) to diagnose cracks in beam structures. In this study, FGM beams are considered. This prediction method is based on input parameters to find the parameters of cracks, which are location and depth in the beam structure. In this study, vibration characteristics are used as input parameters for ANN to give output results. The prediction results from the Neural Network method are compared with previously measured data by experimental methods to evaluate the effectiveness of this method in diagnosing damage in beam structures. This study aims to evaluate the accuracy and effectiveness of the ANN method in solving the current problem of detecting structural damage.