Approximation for semilinear stochastic evolution equations (Q1868135)

The author considers stochastic evolution equations of the form \[ du(t) = (Au(t) + f(t,u(t)))dt + \sigma (u(t)) dW(t),\tag{1} \] given on a separable Hilbert space and \(W(t)\) is a Wiener process on that space. The operator \(A\) is the infinitesimal generator of an analytic semigroup of negative type. Typical examples of (1) are parabolic stochastic partial differential equations on a smooth domain with Dirichlet or Neumann boundary conditions. Conditions are given such that (1) has a unique mild solution. Explicit solutions of equations such as (1) are rarely found, thus the article is devoted to numerical methods to approximate solutions. The approximation in the space variable is done by the method of moments, which includes Galerkin, collocation and finite element methods. For the resulting system of stochastic ordinary differential equations the author investigates three time discretization schemes, namely the explicit and implicit Euler methods and the Crank-Nicolson scheme. A very careful and detailed convergence analysis of the proposed numerical methods is then carried out.