Upper and lower solutions in quasilinear parabolic boundary value problems (Q2576631)

The following problem in \(Q=\{ \Omega \times \mathbb R^N \}\) \[ u_t = f(| x| , t, u, | \nabla u| ^2) \text{div} (g(| \nabla u| ^2) \nabla u ), \tag{1} \] \[ u(x,t) =u_1(x,t), \qquad (x,t) \in \Gamma, \tag{2} \] \[ u(x,0) =u_0(x), \qquad x \in \Omega, \] where \(\Omega\) is a bounded convex domain of \(\mathbb R^N\) with \(C^{2+\alpha}\) and a boundary \(\Gamma = (\partial \Omega \times \mathbb R^{+})\); \(f, g, u_0, u_1\) are positive smooth functions of their arguments. \(u_0\) and \(u_1\) satisfy the capability condition. The right side operator is suggested strongly elliptic. By introducing special auxiliary functions classical upper (sup), lower (sub) solutions of the considered problem are obtained. The supposing upper and lower solutions \(\Phi(x,t)\) are constructed using a solution of one dimensional equation (1) by replacing \(x\) by \(\sigma(x)=\sqrt{\phi_M-\phi(x)},\) where \(\phi\) is the solution of the torsion problem \[ \nabla \phi(x)=-2, \quad x \in \Omega,\qquad \phi(x)=0, \quad x \in \partial \Omega. \] The effectivity of offered approach is demonstrated on example of the well known problem of differential geometry by numerical experiments.