Random attractors for a class of stochastic partial differential equations driven by general additive noise (Q550028)

Let \(V\subseteq H\equiv H^*\subseteq V^*\) be a Gelfand triple, \(A: V\to V^*\) be measurable and \((N_t)_{t\in\mathbb R}\) be a \(V\)-valued adapted stochastic process. For \([s,t]\subseteq\mathbb R\) the authors consider the following stochastic evolution equation \(dX_r=A(X_r)dr+dN_r\), \(r\in [s,t]\), \(X_s=x\in H\). If \(A\) satisfies the standard monotonicity and coercivity conditions the authors prove existence and uniqueness of solutions to the considered equation. The authors provide a general result yielding the existence of a unique random attractor for the random dynamic system associated with the considered stochastic equation. This result is applicable also to quasilinear equations like stochastic porous media equations, the stochastic \(p\)-Laplace equation and stochastic reaction-diffusion equations. Besides classical Brownian motion, the authors also include space-time fractional Brownian motion and space-time Lévy noise as admissible random perturbations. Under a further condition on the drift, the authors prove that the random attractor consists of a single point. Hence the existence of a unique stationary solution is also obtained. The presented results are based on a variational approach to stochastic partial differential equations.