Error estimates of semi-discrete and fully discrete finite element methods for the Cahn-Hilliard-Cook equation

TL;DR

This paper establishes strong convergence rates for semi-discrete and fully discrete finite element methods applied to the Cahn-Hilliard-Cook equation, highlighting the influence of noise regularity and addressing challenges posed by the unbounded elliptic operator.

Contribution

It provides the first derivation of explicit convergence rates for finite element methods solving the stochastic Cahn-Hilliard-Cook equation, considering noise regularity effects.

Findings

Convergence rates depend on the spatial regularity of the noise process.

New techniques are developed to handle the unbounded elliptic operator in the analysis.

Numerical examples confirm the theoretical convergence rates.

Abstract

In two recent publications [Kov{\'a}cs, Larsson, and Mesforush, SIAM J. Numer. Anal. 49(6), 2407-2429, 2011] and [Furihata, et al., SIAM J. Numer. Anal. 56(2), 708-731, 2018], strong convergence of the semi-discrete and fully discrete finite element methods is, respectively, proved for the Cahn-Hilliard-Cook (CHC) equation, but without convergence rates revealed. The present work aims to fill the left gap, by recovering strong convergence rates of (fully discrete) finite element methods for the CHC equation. More accurately, strong convergence rates of a full discretization are obtained, based on Galerkin finite element methods for the spatial discretization and the backward Euler method for the temporal discretization. It turns out that the convergence rates heavily depend on the spatial regularity of the noise process. Different from the stochastic Allen-Cahn equation, the presence of the unbounded elliptic operator in front of the cubic nonlinearity in the underlying model makes the error analysis much more challenging and demanding. To address such difficulties, several new techniques and error estimates are developed. Numerical examples are finally provided to confirm the previous findings.