Infinite-energy solutions for the Navier-Stokes equations in a strip revisited (Q2438960)

The infinite energy solutions to the Navier-Stokes equations in a 2D strip is studied in the paper \[ \frac{\partial v}{\partial t}+(v\cdot\nabla)v-\Delta v+\nabla p=f,\quad \text{div}\,v=0,\quad t>0,\;x\in \Omega, \] \[ \left. v\right|_{\partial \Omega}=0,\quad \left. v\right|_{t=0}=v_0. \] Here \(\Omega=\mathbb{R}\times(-1,1)\). The problem was investigated in [the second author, Glasg. Math. J. 49, No. 3, 525--588 (2007; Zbl 1142.35067)]. This investigation is found erroneous. It is proved in the present paper that for \(f\in L^2_b(\Omega)\), \(v_0\in L^2_b(\Omega)\) the problem has a unique solution \(v(t)\in L^2_b(\Omega)\). This solution satisfies to the dissipative estimate \[ \| v(t)\|_{L^2_b(\Omega)}\leq Q\left(\| v_0\|_{L^2_b(\Omega)} \right) e^{-\alpha t}+C_1+C_2\| f\|_{L^2_b(\Omega)}, \] where \(C_1\), \(C_2\), \(\alpha\) are positive constants, \(Q\) is the monotone function, \(L^2_b(\Omega)\) is the space with the norm \[ \| u \|_{L^2_b(\Omega)}=\sup\limits_{s\in\mathbb{R}}\| u \|_{L^2(\Omega_s)}, \quad \Omega_s=(s,s+1)\times(0,1). \] The authors are sure that the presented proof does not contain errors.