Recently, physics informed neural networks have successfully been applied to a broad variety of problems in mathematics and engineering. The principle idea is the usage network as global ansatz function for partial differential equations. Due approximation, difficulties displaying localized effects strong nonlinear solution fields by optimization. In this work we consider stress displacement in...

A microstructural model of steady-state creep that couples grain-boundary transport, micromechanics, and grain sliding is employed to investigate the grain-boundary diffusional creep response of an idealized microstructure with variable boundary diffusivities. Both numerical and analytical methods were used to determine the stress state and, in some cases, the strain rate associated with an app...