Prediction of elastic wave propagation in composite bars using deep learning techniques

Abstract

In this work, we developed a deep learning model based on a multilayer perceptron to predict the elastic wave output from composite bars. The model takes a vector representing the microstructure of the composite and the input wave applied at the left edge of the bar as features, with the target being the output elastic waves collected at the right edge of the bar. To train the model, we randomly generated composite bars with corresponding input waves and simulated them using finite element modeling. The results indicate that the proposed method can accurately and efficiently predict the output elastic waves.