Initial failure strength prediction of woven composites using a new yarn failure criterion constructed by deep learning

A new failure criterion for fiber tows (i.e. yarns) is developed based on a micromechanical model using the mechanics of structure genome (MSG) and a deep learning neural network model. The proposed failure criterion can be applied to yarns in mesoscale textile composites modeling while capturing the failure initiation at the fiber and matrix level. A plain weave fiber reinforced composite material example is used to compute the initial failure strength constants of a woven lamina based on the proposed yarn failure criterion. To study the accuracy and efficiency of the failure criterion, a comparison to a meso-micro coupled model explicitly capturing the failure initiation at fiber and matrix level is performed. Moreover, the differences between the mesoscale modeling results based on the proposed criterion and other yarn failure criteria (i.e. maximum stress, Tsai-Wu, and Hashin) are studied. Lastly, the failure envelope analysis of the mesoscale plain weave example is carried out using the MSG solid model to further demonstrate the accuracy and efficiency of the new yarn failure criterion under combined loading conditions.