Prediction of the collapse modes of PVC cylindrical shells under compressive axial loads using Artificial Neural Networks

Abstract

In the present paper Artificial Neural Networks (ANN) are applied in order to predict the buckling modes of thin-walled PVC tubes under compressive axial forces. For the development of the models the neural network toolbox of Matlab® was applied. The results show that these models can satisfactorily face these problems and they constitute not only a fast method, compared to time consuming experiments, but also a reliable tool that can be used for the studying of such parts which are usually employed as structural elements for the absorption of the energy of an impact, in automotive and aerospace applications. © 2007 International Federation for Information Processing.