Optimal error analysis of Galerkin FEMs for nonlinear Joule heating equations (Q461242): Difference between revisions

The Joule heating equations consist of a nonlinear parabolic equation and a coupled homogeneous elliptic one. Following \textit{B. Li} and \textit{W. Sun} [Int. J. Numer. Anal. Model. 10, No. 3, 622--633 (2013; Zbl 1281.65122)], the author considers a semi-decoupled scheme using Euler backward differences in time and Galerkin finite elements in space and mentions also a fully decoupled approach for the numerical solution of the system along with Dirichlet boundary conditions in \(\mathbb R^2\) and \(\mathbb R^3\). For the first method, and employing a boundedness result of \textit{Y.-Z. Chen} and \textit{L.-C. Wu} [Second order elliptic equations and elliptic systems. Transl. from the Chinese by Bei Hu. Providence, RI: American Mathematical Society (1998; Zbl 0902.35003)], for the solution of the Dirichlet problem of the Poisson problem in \(\mathbb R^3\), he shows unconditional stability (by comparing with the solution of a corresponding time-discrete system -- the Rothe method) and gets optimal error estimates in \(L_2\) and \(H^1\). Numerically, he illustrates his theoretical results presenting numerical experiments for the unit square and unit circle, and for the unit ball. His numerical results also exhibit the unconditional stability of the fully decoupled scheme.