ANN-aided incremental multiscale-remodelling-based finite strain poroelasticity

Mechanical modelling of poroelastic media under finite strain is usually carried out via phenomenological models neglecting complex micro-macro scales interdependency. One reason that the mathematical two-scale analysis only straightforward assuming infinitesimal theory. Exploiting potential ANNs for fast and reliable upscaling localisation procedures, we propose an incremental numerical approach considers rearrangement cell properties based on its current deformation, which leads to remodelling macroscopic model after each time increment. This computational framework valid large deformation problems while it ensures increments within steps. The full effects interdependency between response macro micro are considered first providing more accurate predictive fluid-saturated porous studied a consolidation example. Furthermore, (nonlinear) deviation from Darcy's law captured in fluid filtration analyses. Finally, brain tissue mechanical uniaxial cyclic test simulated studied.