Data-physics driven multiscale approach for high-pressure resin transfer molding (HP-RTM).

We present a multiscale computational framework for high-pressure resin transfer molding of fiber-reinforced composites. Due to the relatively rapid speed of resin flow and the significant convective effects, this process is governed by the nonlinear steady-state Navier–Stokes equations, as opposed to the linear Stokes equations commonly adopted for the simulation of classical resin transfer molding processes. To overcome the computational challenge of directly solving the high-pressure resin filling problem through the fiber preform, we developed a data-driven nonlinear homogenization approach where the average velocity and instantaneous permeability computed from the microscale representative volume element problem provide solution-dependent coefficients for the macroscale problem. The proposed data-physics driven computational framework has been validated against the direct numerical solution of the steady-state Navier–Stokes equations. [ABSTRACT FROM AUTHOR]