A machine learning based prediction of elasto-plastic response of a short fiber reinforced polymer (SFRP) composite

Abstract

Several homogenization techniques are available in the literature to compute the mechanical response of the short fiber-reinforced polymer (SFRP) composites. However, in some cases, the complex modeling of the SFRP makes it computationally expensive. In this study, an artificial neural network (ANN) is developed to predict the elasto-plastic response of an SFRP. The datasets for training the ANN model are obtained from Mori-Tanaka mean-field homogenization using the commercial software Digimat. The elasto-plastic response predicted by the ANN model is compared with the experimental results and with different homogenization schemes reported in the literature. Additionally, the effect of significant parameters on the response of the SFRP is extensively studied using the ANN model.