Discrete ABP Estimate and Convergence Rates for Linear Elliptic Equations in Non-divergence Form
This work provides rigorous error analysis and convergence rates for finite element solutions of non-divergence form elliptic equations, addressing a known bottleneck in numerical analysis.
The paper develops a two-scale finite element method for linear elliptic PDEs in non-divergence form, proving a discrete Alexandroff-Bakelman-Pucci estimate that yields pointwise error convergence rates up to O(h|ln h|), which is quasi-optimal.
We design a two-scale finite element method (FEM) for linear elliptic PDEs in non-divergence form $A(x) : D^2 u(x) = f(x)$ in a bounded but not necessarily convex domain $Ω$ and study it in the max norm. The fine scale is given by the meshsize $h$ whereas the coarse scale $ε$ is dictated by an integro-differential approximation of the PDE. We show that the FEM satisfies the discrete maximum principle (DMP) for any uniformly positive definite matrix $A$ provided that the mesh is face weakly acute. We establish a discrete Alexandroff-Bakelman-Pucci (ABP) estimate which is suitable for finite element analysis. Its proof relies on a discrete Alexandroff estimate which expresses the min of a convex piecewise linear function in terms of the measure of its sub-differential, and thus of jumps of its gradient. The discrete ABP estimate leads, under suitable regularity assumptions on $A$ and $u$, to pointwise error estimates of the form \begin{equation*} \| u - u^ε_h \|_{L_\infty(Ω)} \leq \, C(A,u) \, h^{2α/(2 + α)} \big| \ln h \big| \qquad 0< α\leq 2, \end{equation*} provided $ε\approx h^{2/(2+α)}$. Such a convergence rate is at best of order $ h \big| \ln h \big|$, which turns out to be quasi-optimal.