A micromechanics-based artificial neural networks model for elastic properties of short fiber composites

There are a wide variety of microstructural parameters which affect the macro-mechanical response of short fiber reinforced composites. Effects of these parameters could be captured using different micromechanics-based models. However, in some cases, it is very challenging and computationally expensive. In this study, a micromechanics-based Artificial Neural Networks (ANN) model is developed to predict the elastic properties of these materials, accurately and quickly. The required data for training and validating the model is created using a two-step approach, combining Finite Element Analysis and Orientation Averaging. The capability of the model for fair predictions is proven, not only by using the validation data, but also by comparisons to experimental results taken from literature.