Pullback attractors for the non-autonomous FitzHugh-Nagumo system on unbounded domains (Q1015817)

This paper studies the long time behaviour of solutions of a singularly perturbed Fitzhugh-Nagumo system of the form \[ \begin{aligned} \frac{\partial u}{\partial t}&-\nu\Delta u+\lambda u+h(u)+v=f(t), \\ \frac{\partial v}{\partial t}&- \varepsilon (u-\gamma v)= \varepsilon g(t),\end{aligned} \] where \(\nu, \lambda, \varepsilon, \gamma\) are positive constants, \(f, g\) are given functions of \(t\), and \(h\) is a nonlinear function satisfying a dissipativity condition. The system is a non autonomous PDE defined on the entire space \({\mathbb R}^{n}\), with possibly unbounded external forcing terms. In the case where \(\varepsilon>0\), uniform estimates are obtained for the solution of the problem as \(t\rightarrow \infty\) necesary for proving the existence of a bounded pullback absorbing set, and by uniform estimates on the tails of the solutions for large \(|x|\) when \(t\rightarrow \infty\), the pullback asymptotic compactness of the \(\theta\)-cocycle \(\varphi\) associated with the system is established. Then, the existence of bounded pullback global attractor, on which solutions satisfy appropriate exponential bounds as \(t\to \infty\), is proved in \(L^2(\mathbb R^n)\times L^2(\mathbb R^n)\). These results are obtained using conditions on \(f\) and \(g\) that do not necessarily require boundedness of these functions in \(L^2(\mathbb R^n)\) as \(t\to\infty\). Next, the behaviour of the attractor as \(\varepsilon\to 0\) is studied. While in the case of unbounded forcing terms the absence of uniform bounds for the attractor does not allow for conclusive results thus leaving the open the possibility of blow up as \(\varepsilon\to 0\), in the case where \(f\in L^\infty(\mathbb R, L^2(\mathbb R^n))\) and \(g\in L^\infty(\mathbb R, H^1(\mathbb R^n))\) the attractor is uniformly bounded in \(\varepsilon\) thus excluding any possibility for blowup.