English

Domain Variations for Moving Boundary Problems

Analysis of PDEs 2018-08-14 v3

Abstract

In the past few decades maximal regularity theory has successfully been applied to moving boundary problems. The basic idea is to reduce the system with varying domains to one in a fixed domain. This is done by a transformation (the so-called Hanzawa transformation) and yields a typically nonlocal and nonlinear coupled system of (evolution) equations. Well-posedness results can then often be established as soon as it is proved that the relevant linearization is the generator of an analytic semigroup or admits maximal regularity. To implement this program, it is necessary to somehow parametrize to space of boundaries/domains (typically the space of compact hypersurfaces Γ\Gamma in Rn\mathbb{R}^n, in the Euclidean setting). This has traditionally been achieved by means of the already mentioned Hanzawa transformation. The approach, while successful, requires the introduction of a smooth manifold Γ\Gamma_\infty close to the manifold Γ0\Gamma_0 in which one cares to linearize. This prevents one to use coordinates in which Γ0\Gamma_0 lies at their "center". As a result formul{\ae} tend to contain terms that would otherwise not be present were one able to linearize in a neighborhood emanating from Γ0\Gamma_0 instead of from Γ\Gamma_\infty. In this paper it is made use of flows (curves of diffeomorphisms) to obtain a general form of the relevant linearization in combination with an alternative coordinatization of the manifold of hypersurfaces, which circumvents the need for the introduction of a "phantom" reference manifold Γ\Gamma_\infty by, in its place, making use of a "phantom geometry" on Γ0\Gamma_0. The upshot is a clear insight into the structure of the linearization, simplified calculations, and simpler formul{\ae} for the resulting linear operators, which are useful in applications.

Keywords

Cite

@article{arxiv.1607.05774,
  title  = {Domain Variations for Moving Boundary Problems},
  author = {Patrick Guidotti},
  journal= {arXiv preprint arXiv:1607.05774},
  year   = {2018}
}
R2 v1 2026-06-22T14:58:59.629Z