scientific article; zbMATH DE number 1002438

From MaRDI portal
Publication:3130097





The physical problem considered is the two-dimensional flow of an incompressible viscous fluid in the strip \(0< x_1<L\), \(-\infty<x_2<+\infty\); the region \(x_2<0\) being occupied by a porous medium. The structure of the porous medium is obtained in the usual way extending by periodicity a basic cell. The basic cell is constructed rescaling by a factor \(\varepsilon\) a unit cell, in which the subset available for the flow is connected and has a smooth boundary \(S\). The flow is supposed to be slow enough, so that on both regions \(x_2<0\), \(x_2>0\) it can be described by Stokes equation with no slip conditions on \(x_1=0\), \(x_1=L\) and on \(S\). Denoting by \(\Omega_1\) the free fluid domain, by \(\Omega^\varepsilon_2\) the porous medium, and by \(\Omega^\varepsilon\) the entire flow region, the governing system for a fixed \(\varepsilon>0\) is \(-\Delta u^\varepsilon+ Dp^\varepsilon= F^\varepsilon\), \(\text{div }u^\varepsilon=0\) in \(\Omega^\varepsilon\), where \(F^\varepsilon=f\) in \(\Omega_2\), \(F^2+\varepsilon^\gamma f\) in \(\Omega_1\). Here \(u^\varepsilon\) is the velocity vector, and \(p^\varepsilon\) is the pressure field. The authors perform a detailed analysis of the limit case \(\varepsilon\to 0\) with the principal aim at describing the behaviour of the effective flow at the interface \((x=0)\). Needless to say, the paper is very technical. The mere statement of the results requires thirty pages, despite the extreme conciseness of exposition. The procedure for constructing suitable approximations of \((u^\varepsilon,p^\varepsilon)\) is very complicated, and it requires the solution of a number of auxiliary problems both in the porous medium and in the free flow region. As expected, the choice of the exponent \(\gamma\) is crucial. For instance, a typical result about the limit flow is the following: take \(\gamma=2\); then in the porous region one finds Darcy's law \(v_D= K(f-\nabla p)\), where \(K\) is the permeability matrix, obtained through the solution of an auxiliary problem in the unit cell in a way which is rather typical for homogenization in porous media. In the free flow region, the velocity field is not just a Stokes flow, but there are correcting terms due to the influences of a boundary layer. The conditions at the interface are the continuity of the normal velocity and a jump condition for the tangential velocity. It is remarkable that the jump is calculated explicitly. These results are discussed and compared to the ones proposed in the literature. This paper is a very valuable contribution in the research area of flows with interfaces, and it contains quite original and interesting results which unfortunately cannot be summarized in a short review.



Cites work


Cited in
(89)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3130097)