Asymptotic decay under nonlinear and noncoercive dissipative effects for electrical conduction in biological tissues (Q503116)

The authors analyze a model accounting for electrical conduction in biological tissues. They start from a smooth domain \(\Omega \subset \mathbb{ R}^{N}\), \(N\geq 2\) and a periodic open subset \(E\) of \(\mathbb{R}^{N}\). They define \(\Omega _{1}^{\varepsilon }=\Omega \cap \varepsilon E\), \( \Omega_{2}^{\varepsilon }=\Omega \setminus \varepsilon \overline{E}\), \(Y=(0,1)^{N}\), \(E_{1}=E\cap Y\), \(E_{2}=E\setminus \overline{Y}\) and \(\Gamma =\partial E\cap \overline{Y}\). They introduce the piecewise constant conductivity \(\sigma (y)=\sigma _{1}\) in \(E_{1}\) and \(\sigma (y)=\sigma _{2}\) in \(E_{2}\), where \(\sigma _{1}\) and \( \sigma _{2}\) are positive constants. They consider the problem \(-\operatorname{div} (\sigma ^{\varepsilon }\nabla u_{\varepsilon })=0\) in \((\Omega _{1}^{\varepsilon }\cup \Omega _{2}^{\varepsilon })\times (0,T)\) with the boundary conditions \([\sigma ^{\varepsilon }\nabla u_{\varepsilon }\cdot \nu _{\varepsilon }]=0\) on \(\Gamma ^{\varepsilon }\times (0,T)\), \(\frac{\alpha }{ \varepsilon }\frac{\partial }{\partial t}[u_{\varepsilon }]+f(\frac{ [u_{\varepsilon }]}{\varepsilon })=\sigma ^{\varepsilon }\nabla u_{\varepsilon }\cdot \nu _{\varepsilon }\) on \(\Gamma ^{\varepsilon }\times (0,T)\), \(u_{\varepsilon }(x)=\Psi (x,t)\) on \(\partial \Omega \times (0,T)\) and \([u_{\varepsilon }](x,0)=S_{\varepsilon }(x)\) on \(\Gamma ^{\varepsilon }\). Here, \(\alpha \) is a positive constant, \(f\) is a \(C^{1}\) function which is strictly monotone increasing and satisfies \(f(0)=0\) and \(f^{\prime }(s)\geq \delta _{0}>0\) for \(|s|\) large enough and \(\Psi (x,\cdot )\) is 1-periodic for a.e. \(x\in \Omega \). The authors use classical tools in order to prove an existence and uniqueness result for this problem. They first describe the asymptotic behavior of the solution \(u_{\varepsilon }\) when \(t\rightarrow \infty \). They prove that \(u_{\varepsilon }\) converges to the solution \( u_{\varepsilon }^{\#}\in C_{\#}^{1}([0,1];\mathcal{X}^{1}(\Omega _{\varepsilon }))\) of a quite similar problem but now posed in \((\Omega _{1}^{\varepsilon }\cup \Omega _{2}^{\varepsilon })\times \mathbb{R}\) for which they prove an existence and uniqueness result. For the proof, the authors first prove uniform estimates and obtain weak convergences. The main result of the paper describes a two-scale convergence result for the solution \(u_{\varepsilon }\) of the original problem. The authors first propose the limit problem and they establish its limit when \(t\rightarrow \infty \). They prove existence results and a convergence result when \(t\rightarrow \infty \) for this asymptotic problem. For the proof, they use classical tools from the two-scale convergence.