Stochastic convergence analysis on Stokes problem

Read full text Summarise with VIRAT
Authors: Tong Wu
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This paper investigates the inverse problem of recovering a spatially varying viscosity coefficient in the Stokes equations from noisy, pointwise observational data. To stabilize the inversion, we employ Tikhonov regularization with an $H^1(\Omega)$ penalty and utilize a two-mesh finite element strategy for numerical approximation, assigning Taylor-Hood elements on a fine mesh for state variables and polynomial approximations on a coarse mesh for the parameter. Theoretically, by introducing a non-trapping vector field assumption driven by the exact strain tensor, we overcome the limitations of standard energy estimates. Leveraging the theory of regular Lagrangian flows for Sobolev vector fields, we rigorously establish the existence of global minimizers in both continuous and discrete settings, and derive the convergence rates of the reconstruction error under sub-Gaussian noise. The analysis provides an optimal scaling relationship among the regularization parameter, mesh sizes, and the noise level. Finally, comprehensive numerical experiments will be conducted to validate the theoretical convergence rates.

Fields of Study & Subjects

Numerical methods in inverse problems Advanced Numerical Methods in Computational Mathematics Spectral Theory in Mathematical Physics

Publication details

Journal / source
California Digital Library
Publication date
1 January 2026
Language
EN