Commutability of homogenization and linearization at identity in finite elasticity and applications (Q654167)

From MaRDI portal





scientific article; zbMATH DE number 5992248
Language Label Description Also known as
default for all languages
No label defined
    English
    Commutability of homogenization and linearization at identity in finite elasticity and applications
    scientific article; zbMATH DE number 5992248

      Statements

      Commutability of homogenization and linearization at identity in finite elasticity and applications (English)
      0 references
      0 references
      0 references
      28 December 2011
      0 references
      The purpose of the paper is to prove the commutability of linearization and homogenization at identity in a context of finite elasticity. Let \(D\) be an open, bounded and Lipschitz domain of \(\mathbb{R}^{d}\) and consider the family \(\mathcal{I}_{\varepsilon }:u\mapsto \int_{D}W_{\varepsilon }(x,\nabla u(x))dx\), where \(W_{\varepsilon }:D\times \mathbb{M} ^{d}\rightarrow [ 0,+\infty ]\) is a Borel function which is frame indifferent, minimal at identity, non-degenerate and admitting a quadratic expansion at identity. The quadratic term \(Q_{\varepsilon }\) of this expansion is thus associated to a functional \(\mathcal{E}_{\varepsilon }:u\mapsto \int_{D}Q_{\varepsilon }(x,\nabla g(x))dx\), where \( g(x)=h^{-1}(u(x)-x)\) for a displacement field \(u\) satisfying \(\left| \nabla u-Id\right| \sim h\ll 1\). The main result of the paper proves that the limits of \(\mathcal{G}_{h,\varepsilon }\) when \(h\rightarrow 0\) and when \(\varepsilon \rightarrow 0\) commute, where the functional \(\mathcal{G} _{h,\varepsilon }\) is defined through \(\mathcal{G}_{h,\varepsilon }(g)= \mathcal{I}_{\varepsilon }(\varphi Id+hg)/h^{2}\), the limits being taken in a \(\Gamma \)-convergence sense. The authors here assume that the energy densities \(W_{\varepsilon }\) belong to the space \(\mathcal{W}_{\alpha ,\rho }^{p}(D\times \mathbb{R}^{d})\) of continuous functions which satisfy a growth condition of order \(p\in [ 2,+\infty )\) and which have a modulus of approximation \(\rho \) at \(Id\). This extends a result obtained by \textit{S. Müller} and the second author in [Arch. Ration. Mech. Anal. 201, No. 2, 465--500 (2011; Zbl 1262.74029)]. The proof of this result is mainly based on the properties of the \(\Gamma \)-convergence and of the \(G\)-convergence. The paper ends with an application to stochastic homogenization within this context.
      0 references
      homogenization
      0 references
      nonlinear elasticity
      0 references
      linearization
      0 references
      \(\Gamma \)-convergence
      0 references
      \(\Gamma \)-closure
      0 references
      \(G\)-convergence
      0 references
      Korn inequality
      0 references
      Meyers estimate
      0 references
      stochastic homogenization
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references
      0 references