Sublinear growth of the corrector in stochastic homogenization: optimal stochastic estimates for slowly decaying correlations (Q2014310)

From MaRDI portal





scientific article
Language Label Description Also known as
default for all languages
No label defined
    English
    Sublinear growth of the corrector in stochastic homogenization: optimal stochastic estimates for slowly decaying correlations
    scientific article

      Statements

      Sublinear growth of the corrector in stochastic homogenization: optimal stochastic estimates for slowly decaying correlations (English)
      0 references
      0 references
      0 references
      11 August 2017
      0 references
      This paper deals with the stochastic homogenization of linear elliptic equation of the form \(-\nabla\cdot a \nabla u=f\). Under conditions of weak decorrelation and essentially Gaussian coefficient fields, the authors obtain optimal stochastic moments for the minimal radius above which the corrector is sublinear. The main result of the paper is following. Let \(\tilde a=\tilde a (x)\) be a tensor-valued Gaussian random field on \(\mathbb R^{d}\) that is centered and stationary. Assume that the covariance of \(\tilde a\) satisfies the estimate \(|\langle \tilde a (x)\otimes\tilde a(0)\rangle| \leq| x|^{-\beta}\), \(\forall x\in\mathbb R^{d}\) for some \(\beta\in (0,d)\). Let \(\Phi:\;\mathbb R^{d\times d}\to \mathbb R^{d\times d}\) be a Lipschitz map with Lipschitz constant \(\leq 1\), and \(\Phi\) takes values in the set of symmetric matrices subject to the ellipticity and boundedness assumptions. Define the ensemble \(\langle\cdot\rangle \) as the probability distribution of \(a\), where \(a(x):=\Phi(\tilde a(x))\), and let \(\beta\leq 1/C\), where \(C\) denotes a generic constant only depending on \(d\) and ellipticity ratio \(\lambda\in (0,1)\). (i) Consider a linear functional \(F=Fh\) on vector fields \(h=h(x)\) such that \[ | Fh|\leq \left(-\!\!\!\!\!\!\int_{| x|\leq r}| h|^{2d/(d+\beta)}dx\right)^{(d+\beta)/2d} \] for some radius \(r>0\). Then \(\langle I\left(| F\nabla(\phi,\sigma)|\geq M\right)\rangle\leq C\exp\left(-r^{\beta}M^2/C\right)\) for all \(M\leq1\). Here \(I(E)\) is the characteristic function of an event \(E\), \((\phi,\sigma)\) is the extended corrector. (ii) There exists a random radius \(r_{*}\) for which we have \[ {1\over r^2}-\!\!\!\!\!\!\int_{| x|\leq r}\left|(\phi,\sigma)- -\!\!\!\!\!\!\int_{| x|\leq r}(\phi,\sigma)\right|^2dx\leq \left({r_{*}\over r}\right)^{\beta}\log\left(e+\log\left({r\over r_{*}}\right)\right) \] for \(r\geq r_{*}\) and which satisfies \(\left\langle\exp\left(r_{*}^{\beta}/C\right)\right\rangle\leq C\).
      0 references
      homogenization corrector
      0 references
      weak decorrelation
      0 references
      optimal stochastic moments
      0 references
      0 references
      0 references
      0 references
      0 references

      Identifiers

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