Prediction of the Acoustic Pressure Backscattered by a Steel Tube Using Neural Networks Approach

Abstract

A new approach is used to predict the pressure backscattered by a tube using the artificial neural networks (ANNs) techniques. The studied tube consists of steel. During the development of the network, several configurations are evaluated for various radius ratio b/a (a: outer radius, b: inner radius of tube). The multilayer perceptron (MLP) is used in the current study. The optimal model selected is a network with one hidden layer. This model is able to predict the pressure backscattered with a mean relative error (MRE) of about a 1.6%. The comparison of the obtained and the experimental results indicate that the ANN method is suitable to be used to predict this one.