Optimal \(L^{\infty}\) error estimates for finite element Galerkin methods for nonlinear evolution equations (Q949256)

The author considers three classes of nonlinear nonstationary initial-boundary problems of the form \[ \begin{aligned} u_{t} &= A\frac{\partial^{2}u}{\partial x^{2}} +f(u), \quad (x,t)\in Q\\ u(0,t) &= u(1,t)=0; \\ u(x,0) &= u_{0}(t),\end{aligned} \] where \(u=(u_{1},u_{2},\dots,u_{r})\) is a complex vector-function, \(A\) is a complex diagonal matrix, \(f\) a complex vector function and \(u_{0}\) a given complex valued function. Since \(A\) is diagonal we are led to the following three cases: {\parindent=5mm \begin{itemize}\item[1.]If \(Im(A)=0\) we have the reaction-diffusion equation \item[2.]If \(Re(A)>0\) and \(Im(A)\neq0\) we have the Kuramoto-Tsuzuki equation \item[3.]If \(Re(A)>0\) and \(Im(A)=0\) we have the heat equation. \end{itemize}} The results are presented under certain assumptions in the following sections: Crank-Nicolson Galerkin method; Existence; Stability; Uniqueness; Rate of convergence estimates; The extrapolated Crank-Nicolson scheme.