Advancing Numerical Methods and Error Analysis in Complex Systems

The recent developments in the research area primarily focus on advancing numerical methods and error analysis for complex physical systems. There is a notable trend towards enhancing the accuracy and stability of existing numerical techniques, particularly in the context of nonlinear and non-convex problems. For instance, second-order accurate methods are being introduced to improve the efficiency of simulations in micromagnetics, addressing the limitations of first-order methods. Additionally, the use of harmonic functions and discrete harmonics is being explored to achieve higher convergence rates in fluid dynamics problems. Another significant direction is the application of discontinuous Galerkin isogeometric methods to electronic structure calculations, providing a unified analysis framework for elliptic eigenvalue problems. Furthermore, there is a strong emphasis on deriving both a priori and a posteriori error estimates for discontinuous Galerkin approximations of time-harmonic Maxwell's equations, ensuring minimal regularity assumptions and optimal accuracy. These advancements collectively push the boundaries of computational methods, making them more robust and applicable to a wider range of scientific and engineering challenges.

Noteworthy papers include one introducing second-order Gauss-Seidel projection methods for micromagnetics simulations, demonstrating unconditional stability. Another highlights the use of discrete harmonics to achieve order one error for the L2 norm of vorticity in Stokes problems.

Sources

Enhanced second-order Gauss-Seidel projection methods for the Landau-Lifshitz equation

Discrete harmonics for stream function-vorticity Stokes problem

Numerical Analysis of Multi-patch Discontinuous Galerkin Isogeometric Method for Full-potential Electronic Structure Calculations

A priori and a posteriori analysis of the discontinuous Galerkin approximation of the time-harmonic Maxwell's equations under minimal regularity assumptions

Built with on top of