Resumen de Finite-dimensionality of attractors for dynamicl systems wigh applications: deterministic and random settings
Arthur Cavalcante Cunha
In this work we obtain estimates on the fractal dimension of attractors in three different settings: global attractors associated to autonomous dynamical systems, uniform attractors associated to non-autonomous dynamical systems and random uniform attractors associated to non-autonomous random dynamical systems. Firstly we give a simple proof of a result due to Mañé (Springer LNM 898, 230242, 1981) that the global attractor A (as a subset of a Banach space) for a map S is finite-dimensional if DS(x) =C(x)+L(x), where C is compact and L is a contraction (and both are linear). In particular, we show that if S is compact and differentiable then A is finite-dimensional. Using a smoothing property for the differential DS we also prove that A has finite fractal dimension and we make a comparison of this method with Mañés approach. We give applications to an abstract semilinear parabolic equation and to 2D Navier-Stokes equations. Secondly we prove using a smoothing method that uniform attractors have finite fractal dimension on Banach spaces, with bounds in terms of the dimension of the symbol space and a Kolmogorov entropy number. We also show that the smoothing property is useful to prove the finite-dimensionality of uniform attractors in more regular Banach spaces. In addition, we prove that the finite-dimensionality of the hull of a time-dependent function is fully determined by the tails of the function. We give applications to non-autonomous 2D Navier- Stokes and reaction-diffusion equations. Thirdly we prove using a smoothing and a squeezing method that random uniform attractors have finite fractal dimension. Neither of the two methods implies the other. Estimates on the dimension are given in terms of the dimension of the symbol space plus a term arising from the smoothing/squeezing property; the smoothing is applied also to more regular spaces. In this setting we give applications to a stochastic reaction-diffusion equation with scalar additive noise. In addition, a random absorbing set which absorbs itself after a deterministic period of time is constructed.
