Global small analytic solutions of MHD boundary layer equations (Q2656263)

The global well-posedness of a two-dimensional MHD boundary layer problem (1.1) in the upper plane is obtained, for small initial data. When a specific parameter has a particular constant value, (1.1) is reduced to the classical Prandtl equations derived in 1904 -- the boundary layer theory in the absence of the magnetic field. The results of [\textit{F. Xie} and \textit{T. Yang}, Acta Math. Appl. Sin., Engl. Ser. 35, No. 1, 209--229 (2019; Zbl 1414.76044)] are improved, for the case when the background shear velocity is taken to be a Gaussian error function as in [\textit{M. Ignatova} and \textit{V. Vicol}, Arch. Ration. Mech. Anal. 220, No. 2, 809--848 (2016; Zbl 1334.35238)]. An important element is the parameter \(k\) -- the difference between Reynolds number and the magnetic Reynolds number. A cut-off procedure and the function transformations (1.3) are used to get the equivalent non-linear problem (1.4). A priori estimates for smooth enough solutions in the analytical framework are given. Suitable approximate solutions are constructed and uniform estimates are obtained for such approximate solution sequence. A convergence analysis is carefully performed for obtain the limit in the approximate problem. A large number of quite technical and laborious calculations are performed, in order to estimate the diadic operator (1.7), which is related with the partial Fourier transform of some elements of solution. The basic mathematical tools are Besov spaces, time-weighted Chemin-Lerner spaces, an anisotropic Bernstein-type result, the Plancherel equality, the Littlewood-Paley theory, the Bony's decomposition result. Although the text is quite technical and difficult for an uninformed reader, a large list of recent references provides the necessary elements for a good understanding of this very interesting paper.