Conservative solutions to a system of variational wave equations

DOI10.1016/J.JDE.2011.11.022MaRDI QIDQ765961zbMATH OpenOpenAlexFDO

Publication date 22 March 2012

Published in Journal of Differential Equations (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.jde.2011.11.022

conservative solutions continuous dependence on data variational wave equations weak solutions

Weak solutions to PDEs (35D30) Initial value problems for second-order hyperbolic systems (35L52) Second-order quasilinear hyperbolic equations (35L72) PDEs with measure (35R06)

Review text

The Cauchy problem is considered \[ u_{ktt}-c(u)\Bigl[c(u)u_{kx}\Bigr]_x=c(u)\sum_{i=1}^n\Bigl(c_{u_i}u_{kx}-c_{u_k}u_{ix}\Bigr)u_{ix},\quad k=1, 2, \ldots, n, \quad t\in \mathbb R, x\in \mathbb R,\tag{1} \] \[ u_k(0, x)=u_{k0}(x)\in H^1(\mathbb R), u_{kt}(0, x)=u_{k1}(x)\in L^2(\mathbb R),\quad k=1, 2, \ldots, n,\quad x\in \mathbb R,\tag{2} \] where \(c:\mathbb R^n\longrightarrow \mathbb R\) is a smooth and positive function satisfying the conditions \(0<c_1\leq c(z)\leq c_2<\infty\), \(\sup_z|\nabla c(z)|<\infty\), for some positive constants \(c_1\) and \(c_2\). The author proves that the problem (1)--(2) has a global weak solution defined for all \((t, x)\in \mathbb R\times \mathbb R\). In the article it is proved that there exists a continuous family \(\{\mu_t:t\in\mathbb R\}\) of positive Radon measures on the real line with \(\mu_t(\mathbb R)={\mathcal E}_0\) for every time \(t\), the absolutely continuous part of \(\mu_t\) has density \({1\over 2}\sum_{i=1}^n \Bigl(u_{it}^2+c^2(u)u_{ix}^2\Bigr)\) with respect to the Lebesgue measure for every \(t\) and for almost every \(t\in\mathbb R\) the singular part of \(\mu_t\) is concentrated on the set where \(|\nabla c(u)|=0\). Here \({\mathcal E}_0={\mathcal E}(0)\), \({\mathcal E}(t)={1\over 2}\int \sum_{i=1}^n \Bigl[u_{it}^2(t, x)+c^2(u(t, x))u_{ix}^2(t, x)\Bigr] dx\). The author proves existence of a family of weak solutions to the problem (1)--(2) for which for every \(t\in\mathbb R\) is valid \({\mathcal E}(t)\leq {\mathcal E}(0)\) and if \(u^{\nu}\) are solutions of \((1)\) which correspond of the initial data \(u_{0}^{\nu}\), \(u_1^{\nu}\) and \(u\) is a solution of (1)--(2) and \[ \sum_{i=1}^n\left(\Bigl|\Bigl|\Bigl(u_{i0}^{\nu}\Bigr)_x-\Bigl(u_{i0}\Bigr)_x\Bigr|\Bigr|_{L^2(\mathbb R)}+\Bigl|\Bigl|u_{i1}^{\nu}-u_{i1}\Bigr|\Bigr|_{L^2(\mathbb R)}\right)\longrightarrow 0 \] and \(u_{k0}^{\nu}\longrightarrow u_{k0}\), \(k=1, 2, \ldots, n,\) uniformly on compact sets, as \(\nu\longrightarrow\infty\), then \(u^{\nu}\longrightarrow u\) uniformly on bounded subsets of the \(t\)-\(x\) plane.

Recommendations

Cites work

Cited in

(23)

This page was built for publication: Conservative solutions to a system of variational wave equations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q765961)