Gevrey regularity of random attractors for stochastic reaction-diffusion equations (Q4516138)

Let us consider the stochastic semilinear parabolic equation \[ \partial_{t}u(t,x)-\nu\Delta u(t,x)+\nu u(t,x)+f(u(t,x))=\partial_{t} W(t,x), \quad x\in R^{n},\;t>t_0, \] with periodic boundary conditions \(u(t,x+Le_{j})=u(t,x),j=1,\ldots,n,\) and the initial condition \(u(t_0,x)= u_0(x)\), where \(\{e_{j}\}\) is the standard basis in \(R^{n}\), \(L>0\), \(\Delta\) is the Laplace operator, \(\nu>0\), \(\partial_{t}W(t,x)\) is a spatially smooth white noise process, \(u_0(x)\) is an \(L\)-periodic function, \(f(u)=\sum_{k=0}^{2p-1}b_{k}u^{k}\), \(b_{2p-1}>0\). The author proves existence of the Gevrey regular absorbing set for the random dynamical system generated by the given boundary value problem. In particular, this result implies that elements of the random attractor are real analytic functions in the spatial variables. In one spatial dimension case the considered problem possesses two determining nodes.