scientific article; zbMATH DE number 3897472

From MaRDI portal
Publication:3676557





The author considers the steady Navier-Stokes equations in a bounded domain \(\Omega\) of \({\mathbb{R}}^ 2:\) (1) \(-\nu \Delta u+(u.\nabla)u=f- \nabla p\) in \(\Omega\) ; \(\nabla.u=0\) in \(\Omega\) ; \(u=g\) on \(\partial \Omega\); where g satisfies the compatibility condition \((2)\quad \sum^{m}_{i=1}\int_{\Gamma_ i}g\cdot n_ id\Gamma_ i=0,\) where \(\Gamma_ i\), \(i=1,...,m\) denote the connected components of \(\Gamma\). Under the stronger condition (3) \(\int_{\Gamma_ i}g\cdot n_ id\Gamma_ i=0\), \(i=1,...,m\), the existence of a solution (u,p) for (1) (under suitable smoothness hypothesis on the data) is classical. This work is an attempt to get existence of a solution by assuming only (2). The main result of the paper is that for admissible \(\Omega\) and admissible f and g, one actually gets existence of a solution under hypothesis (2). Here admissible means essentially that the data and \(\Omega\) are symmetric about the line \(\{x_ 2=0\}\), each component \(\Gamma_ i\) intersecting the line \(\{x_ 2=0\}\).




Cited in
(64)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3676557)