scientific article; zbMATH DE number 1436276

MaRDI QIDQ4948801zbMATH OpenFDO

Authors Fengquan Li

Publication date 8 October 2000

zbMATH Keywords

hyperbolic equations limit behaviour

Mathematics Subject Classification ID

Dependence of solutions to PDEs on initial and/or boundary data and/or on parameters of PDEs (35B30) Asymptotic behavior of solutions to PDEs (35B40) Initial-boundary value problems for second-order hyperbolic equations (35L20)

Review text

Let \(\Omega\subset \mathbb{R}^n, n=2,3\), be a bounded set with smooth boundary \(\Gamma =\Gamma _1\cup \Gamma _2, \Gamma _1\neq \emptyset\) being the outer boundary with \(\Gamma _1\cap \Gamma _2 =\emptyset\). For any fixed \(\varepsilon>0\) assume that \(\Gamma _1\) is partitioned into two subsets \(\gamma _{1\varepsilon}\), and \(\gamma _{2\varepsilon}\), \(\gamma _{1\varepsilon}\) containing the origin. Let \(T>0\) be fixed, \(Q=\Omega \times (0,T), \Sigma _i=\Gamma _i\times (0,T), \Sigma _{i\varepsilon}= \gamma _{i\varepsilon}\times (0,T), i=1,2\). Consider the following initial-boundary problem \((I_{\varepsilon})\): \[ u_{\varepsilon}+Lu_{\varepsilon}=0, (x,t)\in Q, \] \[ \partial u_{\varepsilon}/\partial \nu_L =0, (x,t)\in \gamma _{1\varepsilon},\quad u_{\varepsilon}=C_{\varepsilon}(t), (x,t)\in \gamma _{2\varepsilon},\quad \] \[ \int _{\gamma _{2\varepsilon}}\partial u_{\varepsilon}/\partial \nu_L ds= A_{\varepsilon}(t), \text{a.e. }t\in (0,T),\;u_{\varepsilon}=0, (x,t)\in \Sigma _2, \] \[ u_{\varepsilon}(x,0)=u_{\varepsilon}'(x,0)=0, x\in \Omega, \] where \(C_{\varepsilon}(t)\) is an unknown function of \(t\), \(A_{\varepsilon} \in L^1(0,T)\) is given, \[ Lu_{\varepsilon}=- \sum_{i,j=1}^n \partial /\partial x_i(a_{ij}(x) \partial u_{\varepsilon}/\partial x_j),\quad\partial u_{\varepsilon}/\partial \nu_L=\sum_{i,j=1}^n a_{ij}\nu_i\partial u_{\varepsilon}/\partial x_j, \] and \(\nu=(\nu_1,\dots,\nu_n)\) is the unit outward vector on \(\Gamma_1\). Assume that the following conditions hold: \(\sum_{i_,j=1}^n a_{i_j}(x)\xi_i\xi_j\geq \lambda_0|\xi|^2, \xi\in \mathbb{R}^n, \text{ a.e. }x\in \Omega;\) if \(0<\varepsilon_1< \varepsilon_2\), then \(\gamma _{2\varepsilon_1}\subset \gamma _{2\varepsilon_2},\) and \(\gamma _{2\varepsilon} \to \{0\}\) as \(\varepsilon \to 0\). The author proves that, if \(A_{\varepsilon}\to A_0\) as \(\varepsilon \to 0\) weakly in \(L^1(0,T)\), then \(u_{\varepsilon}\to u\) weakly in \(W^{-1,\infty}(0,T;L^2(\Omega))\), and if \(A_{\varepsilon}\to A_0\) strongly in \(L^1(0,T)\), then \(u_{\varepsilon}\to u\) strongly in \(W^{-1,s}(0,T;L^2(\Omega)), 1<s<\infty\), where \(u\) is the solution to the following problem: \[ u+Lu=0 (x,t)\in Q, \] \[ \partial u/\partial \nu_L =A_0\delta _{\Gamma_1}(x), (x,t)\in \Sigma_1,\quad u=0, (x,t)\in \Sigma _2, \] \[ u(x,0)=u'(x,0)=0, x\in \Omega. \] Also are considered the problem \((II_{\varepsilon})\) which is obtained from the problem \((I_{\varepsilon})\) replacing the homogeneous equation by a nonhomogeneous one, \(A_{\varepsilon}=0\) and the initial conditions by the conditions: \(u_{\varepsilon}'(x,T)=u_{\varepsilon}(x,T)=0\), and the problem \((III_{\varepsilon})\) which is obtained from the problem \((I_{\varepsilon})\) replacing the homogeneous initial conditions by nonhomogeneous one and \(A_{\varepsilon}=0\). For the the solutions of problems \((II_{\varepsilon})\) and \((III_{\varepsilon})\) the author proves results of the same kind.

Recommendations

Cited in

(12)

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 Q4948801)