Permeability through a perforated domain for the incompressible 2D Euler equations (Q2259431): Difference between revisions

The authors describe a homogenization result for an ideal fluid in the domain \(\Omega ^{\varepsilon }=\mathbb{R}^{2}\setminus \cup _{ij}\mathcal{K} _{ij}^{\varepsilon \alpha }\) where the \(\mathcal{K}_{ij}^{\varepsilon \alpha }\) are obstacles of size \(\varepsilon \) disposed along a unit line (\(\mu =0\) ) or a unit square (\(\mu =1\)) and separated by a distance \(\varepsilon ^{\alpha }\): \(\mathcal{K}_{ij}^{\varepsilon \alpha }=\varepsilon \mathcal{K} +z_{ij}^{\varepsilon \alpha }\) where \(z_{ij}^{\varepsilon \alpha }=(\varepsilon +2(i-1)(\varepsilon +\varepsilon ^{\alpha }),\varepsilon +2(j-1)(\varepsilon +\varepsilon ^{\alpha }))\). Here \(\mathcal{K}\) is a smooth (\(\partial \mathcal{K}\) is a \(C^{1,1}\) Jordan curve), simply-connected and compact curve, and \(\alpha >0\). The authors consider the motion of an ideal fluid in \(\Omega ^{\varepsilon }\) according to the system \[ \begin{cases} \partial _{t}\omega ^{\varepsilon }+u^{\varepsilon }\cdot \nabla \omega ^{\varepsilon }=0 & \text{in }(0,\infty )\times \Omega ^{\varepsilon }, \\ \text{div}u^{\varepsilon }=0 & \text{in }[0,\infty )\times \Omega ^{\varepsilon }, \\ u^{\varepsilon }\cdot n=0 & \text{in }[0,\infty )\times \partial \Omega ^{\varepsilon }, \\ \text{curl}u^{\varepsilon }=\omega ^{\varepsilon } & \text{in }[0,\infty )\times \Omega ^{\varepsilon }, \\ \lim_{\left| x\right| \rightarrow \infty }\left| u^{\varepsilon }(t,x)\right| =0 & \text{for }t\in [ 0,\infty ), \\ \oint_{\partial \mathcal{K}_{ij}^{\varepsilon \alpha }}u^{\varepsilon }(0,s)\cdot \tau dx=0 & \text{for all }i,j, \\ \omega ^{\varepsilon }(0,x)=\omega _{0}\in C_{c}^{\infty }(\mathbb{R}^{2}) & \text{in }\Omega ^{\varepsilon }. \end{cases} \] The authors quote from \textit{K. Kikuchi} [J. Fac. Sci., Univ. Tokyo, Sect. I A 30, 63--92 (1983; Zbl 0517.76024)] the existence of a unique, global and strong solution to this problem, with \(\omega ^{\varepsilon }\in L^{\infty }(\mathbb{R}^{+};L^{1}\cap L^{\infty }(\Omega ^{\varepsilon }))\). The main result of the paper says that if \(\alpha \in (0,2-\mu )\) then \( u^{\varepsilon }\) (resp. \(\omega ^{\varepsilon }\)) converges to \(u\) (resp. \( \omega \))\ in \(L_{\mathrm{loc}}^{2}(\mathbb{R}^{+}\times \mathbb{R}^{2})\) strong (resp. \(L^{\infty }(\mathbb{R}^{+}\times \mathbb{R}^{2})\) weak\(\ast \)) where \((u,\omega )\) is the unique and global solution to the Euler equation in \( \mathbb{R}^{2}\) with initial vorticity \(\omega _{0}\). For the proof, the authors use the Biot-Savart representation of \(u\) in terms of \(\omega \) and a biholomorphism \(\mathcal{T}\) mapping the complement \(\mathcal{K}^{c}\) of \( \mathcal{K}\) to \(\overline{B}(0,1)^{c}\) and they prove properties of \( \mathcal{T}\). Then they describe the limit of \(u^{\varepsilon }\) at a fixed time, the convergence of the vorticity and finally the convergence of \(u^{\varepsilon }\) using the properties of the Biot-Savart representation.