General
Semi-Implicit and Convex Splitting Schemes for the Surface Cahn--Hilliard Equation on Curved Closed Surfaces
arXiv cs.NA0 views
arXiv:2611.23915v1 Announce Type: new Abstract: The surface Cahn--Hilliard equation describes phase separation and coarsening of a conserved composition field on fixed curved surfaces. We study a mixed $hp$-surface finite element discretization with curvilinear high-order elements, primarily on the sphere, with the stabilized semi-implicit BDF2 integrator and a stabilized convex-splitting variant solved with the same factorization. For the semi-implicit scheme with a quadratic-growth potential and a truncated nonlinear bulk chemical-potential term, we prove exact discrete mass conservation, energy decay for the start-up steps, an exact per-step energy balance for the BDF2 phase, and a conditional monotone law for a modified Lyapunov functional. A manufactured solution with analytic forcing verifies second order in time and the expected $h$- and $p$-convergence. Torus tests include controlled equal-area bands, whose diffuse energies reproduce sharp-interface perimeters within $1\%$, deterministic coarsening, and a viscous R{\"a}tz--Voigt model. Genus-two and genus-three calculations on implicit surfaces conserve mass to roundoff, and their unstable-mode counts agree with area-based Weyl estimates within $1\%$.
Source:https://arxiv.org/abs/2611.23915PreviousOn global error estimates of a local discontinuous Galerkin method of the Boussinesq system of equationsNextThe $S$ and Double-Exponential Transformations with a Moment Expansion for Three-Centre Coulomb and Hybrid Integrals
Related posts
If nobody writes it down, it did not get decided
0dev.to Productivity
Micrograph B-factor estimation via Wilson statistics
0arXiv eess.IV
Greedy Vector Approximation Based on Rotation for Multiplierless Complex Products
0arXiv eess.SP
Toward Massive MIMO HAPS Networks: Antenna Architectures and Design
0arXiv eess.SP
Deep-Learning-Assisted Full Waveform Inversion for Attenuation Reconstruction in Ultrasound Computed Tomography
0arXiv eess.SP
Two is better than one: optimizing binary rating scales for online platforms
0arXiv cs.MA