Averaging for SDE-BSDE with null recurrent fast component Application to homogenization in a non periodic media

TL;DR

This paper develops an averaging principle for systems of SDE-BSDE with a null recurrent fast component, enabling homogenization of non-periodic, non-ergodic media by combining PDE and probabilistic methods.

Contribution

It introduces a novel averaging approach for SDE-BSDE systems with discontinuous limit coefficients and applies it to homogenize semilinear PDEs in complex media.

Findings

Established averaging principle for null recurrent fast components

Derived homogenization results for non-periodic media

Connected averaged BSDEs to PDE solutions probabilistically

Abstract

We establish an averaging principle for a family of solutions$(X^{\varepsilon}, Y^{\varepsilon})$ $ :=$ $(X^{1,\,\varepsilon},\,X^{2,\,\varepsilon},\, Y^{\varepsilon})$ of a system of SDE-BSDEwith a null recurrent fast component $X^{1,\,\varepsilon}$. Incontrast to the classical periodic case, we can not rely on aninvariant probability and the slow forward component$X^{2,\,\varepsilon}$ cannot be approximated by a diffusion process.On the other hand, we assume that the coefficients admit a limit in a\`{C}esaro sense. In such a case, the limit coefficients may havediscontinuity. We show that we can approximate the triplet$(X^{1,\,\varepsilon},\, X^{2,\,\varepsilon},\, Y^{\varepsilon})$ bya system of SDE-BSDE $(X^1, X^2, Y)$ where $X := (X^1, X^2)$ is aMarkov diffusion which is the unique (in law) weak solution of theaveraged forward component and $Y$ is the unique solution to the averaged backward component. This is done with a backward component whosegenerator depends on the variable $z$. Asapplication, we establish an homogenization result for semilinearPDEs when the coefficients can be neither periodic nor ergodic. Weshow that the averaged BDSE is related to the averaged PDE via aprobabilistic representation of the (unique) Sobolev $W\_{d+1,\text{loc}}^{1,2}(\R\_+\times\R^d)$--solution of the limitPDEs. Our approach combines PDE methods and probabilistic argumentswhich are based on stability property and weak convergence of BSDEsin the S-topology.