Homogenization of nonlinear scalar conservation laws

DOI10.1007/S00205-008-0123-7MaRDI QIDQ834738zbMATH OpenOpenAlexFDO

Publication date 27 August 2009

Published in Archive for Rational Mechanics and Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/0706.2104

BGK relaxation cell problem convolution arguments kinetic law

Homogenization in context of PDEs; PDEs in media with periodic structure (35B27) Asymptotic behavior of solutions to PDEs (35B40) Hyperbolic conservation laws (35L65)

Abstract: We study the limit as

e o 0

of the entropy solutions of the equation

p_{t} u e + d v_{x} [A (f r a c x e, u e)] = 0

. We prove that the sequence

u e

two-scale converges towards a function

u (t, x, y)

, and

u

is the unique solution of a limit evolution problem. The remarkable point is that the limit problem is not a scalar conservation law, but rather a kinetic equation in which the macroscopic and microscopic variables are mixed. We also prove a strong convergence result in

L_{e x t l o c}^{1}

.

Review text

Let \(u^{\varepsilon }\) be an entropy solution of the scalar conservation law \[ \partial u^{\varepsilon }/\partial t(t,x)+\sum_{i=1}^{N}\partial A_{i}(x/\varepsilon ,u^{\varepsilon }(t,x))/\partial x=0\text{ in } (0,\infty )\times \mathbb{R}^{N}, \] with the initial condition \( u^{\varepsilon }(0,x)=u_{0}(x,x/\varepsilon )\). The purpose of the paper is to describe the asymptotic behaviour of \(u^{\varepsilon }\) when \(\varepsilon \) goes to 0. The function \(A_{i}\) is supposed to be \(Y \)-periodic with \(Y=\prod_{i=1}^{N}(0,T_{i})\) and \(u_{0}\) is supposed to be also periodic with respect to its second variable. In order to ensure the existence of a unique entropy solution of this problem, the author assumes that \(A_{i}\in W_{per,loc}^{2,\infty }(\mathbb{R}^{N+1})\) and \(u_{0}\in L^{\infty }\). The author first writes the kinetic formulation of this problem introducing the function \(f^{\varepsilon }(t,x,\xi )=\mathbf{1}_{\xi <u^{\varepsilon }(t,x)}\). The main result of the paper builds the limit kinetic problem and proves the existence of a unique kinetic solution of this limit problem, provided \(u_{0}\) is such that \(f_{0}(t,x,\xi )=\mathbf{1} _{\xi <u_{0}(t,x)}\) is the kinetic solution of a steady problem and that it is bounded from below and from above by \(L^{\infty }\) functions which can be associated to kinetic solutions of the same steady problem. The first convergence result proves that under some further hypotheses \(\mathbf{1} _{\xi <u^{\varepsilon }(t,x)}\) two-scale converges to \(\mathbf{1}_{\xi <u_{0}(t,x)}\). The second convergence result proves a quite similar result when assuming different hypotheses on the \(A_{i}\) and on the initial data \( u_{0}\). The author also proves a strong convergence result in \(L_{loc}^{1}\). For the proofs of the convergence results, the author mainly uses convolution arguments in order to regularize the solution of the original problem. She introduces some cell problem. In the last part of the paper, the author studies a relaxation model of BGK type for some kinetic problem. She concludes the paper with further studies on this limit kinetic problem.

Recommendations

Cites work

Cited in

(23)

This page was built for publication: Homogenization of nonlinear scalar conservation laws

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q834738)