Explicit formulation of SIF using neural networks for opening mode of fracture

This study presents the use of neural networks (NNs) to express the explicit formulation of stress intensity factor (SIF) for the opening mode (KI) of fracture mechanics. Explicit formulations for KI values are obtained using the parameters of the trained NNs. Some numerical applications are performed to show the generalization capability of the trained NNs. A stress intensity factor formulation for three different geometries, which are commonly used in fracture mechanics, is obtained. It is shown that the results of the explicit formulation are in good agreement with the finite element method (FEM), which determines SIF using displacement extrapolation method (DEM) developed in this study, ANSYS and previous work in the literature for those common cases.