Reduced Model for Properties of Multiscale Porous Media with Changing Geometry

In this paper, we consider an important problem for modeling complex coupled phenomena in porous media at multiple scales. particular, flow and transport the void space between pores when pore is altered by new solid obstructions formed microbial growth or reactive transport, are mostly interested pore-coating pore-filling type obstructions, observed applications to biofilm hydrate crystal formation, respectively. We impact of these on macroscopic properties medium, such as porosity, permeability tortuosity, which build experimental probability distribution with reduced models, involves three steps: (1) generation independent realizations followed by, (2) simulations pore-scale, (3) upscaling. For first step, approaches: (1A) direct numerical (DNS) PDE model actual physical process called BN forms two non-DNS methods, call (1B) CLPS (1C) LP. LP a lattice Ising-type model, constrained version Allen–Cahn phase separation localization term. Both approximations BN, they seek local minima some nonconvex energy functional, provide plausible obstructed geometry tuned heuristically deliver either obstructions. Our methods work rock-void geometries obtained imaging, but bypass need imaging real-time, fairly inexpensive, can be tailored other applications. The models less computationally expensive than DNS, desired fidelity distributions upscaled quantities.