Numerical Approximation and Analysis of the Inverse Robin Problem Using the Kohn-Vogelius Method

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Burman, Erik, Cen, Siyu, Jin, Bangti, Zhou, Zhi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918050571223040
author Burman, Erik
Cen, Siyu
Jin, Bangti
Zhou, Zhi
author_facet Burman, Erik
Cen, Siyu
Jin, Bangti
Zhou, Zhi
contents In this work, we numerically investigate the inverse Robin problem of recovering a piecewise constant Robin coefficient in an elliptic or parabolic problem from the Cauchy data on a part of the boundary, a problem that commonly arises in applications such as non-destructive corrosion detection. We employ a Kohn-Vogelius type variational functional for the regularized reconstruction, and discretize the resulting optimization problem using the Galerkin finite element method on a graded mesh. We establish rigorous error estimates on the recovered Robin coefficient in terms of the mesh size, temporal step size and noise level. This is achieved by combining the approximation error of the direct problem, a priori estimates on the functional, and suitable conditional stability estimates of the continuous inverse problem. We present several numerical experiments to illustrate the approach and to complement the theoretical findings.
format Preprint
id arxiv_https___arxiv_org_abs_2506_07370
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Numerical Approximation and Analysis of the Inverse Robin Problem Using the Kohn-Vogelius Method
Burman, Erik
Cen, Siyu
Jin, Bangti
Zhou, Zhi
Numerical Analysis
In this work, we numerically investigate the inverse Robin problem of recovering a piecewise constant Robin coefficient in an elliptic or parabolic problem from the Cauchy data on a part of the boundary, a problem that commonly arises in applications such as non-destructive corrosion detection. We employ a Kohn-Vogelius type variational functional for the regularized reconstruction, and discretize the resulting optimization problem using the Galerkin finite element method on a graded mesh. We establish rigorous error estimates on the recovered Robin coefficient in terms of the mesh size, temporal step size and noise level. This is achieved by combining the approximation error of the direct problem, a priori estimates on the functional, and suitable conditional stability estimates of the continuous inverse problem. We present several numerical experiments to illustrate the approach and to complement the theoretical findings.
title Numerical Approximation and Analysis of the Inverse Robin Problem Using the Kohn-Vogelius Method
topic Numerical Analysis
url https://arxiv.org/abs/2506.07370