Publication

Abstract : The present study is based on the development of an inverse approach in the domain of Recurrent Neural Networks (RNNs) to identify and quantify multiple cracks on a cantilever beam structure subjected to transit mass. First, the responses of the multicrack structure subjected to transit load were determined using fourth-order Runge-Kutta numerical method and Finite Element Analysis (FEA) executed using ANSYS software to authenticate the employed numerical method. The existence and positions of cracks were identified from the measured dynamic excitation of the structure. The crack severities were found as a forward problem through FEA. The modied Elman's Recurrent Neural Networks (ERNNs) approach was implemented to predict the locations and severity of cracks in the structure as an inverse problem by applying Levenberg-Marquardt (L-M) back propagation algorithm. The analogy was carried out in a supervised manner to check the convergence of the proposed algorithm. The results of the proposed ERNNs method were in good agreement with the theoretical and FEA results.