“Lagrangian discretization methods for flow problems”
Venerdì 16 Gennaio 2025, ore 12:00 - Sala Riunioni 7B1 - Hirofumi Notsu (Kanazawa University)
Abstract
In flow problems, the presence of convection terms often leads to numerical difficulties, particularly in regimes with high Reynolds or Péclet numbers, where stabilization techniques are typically required. In this talk, we review discretization strategies based on the method of characteristics, namely the Lagrangian approach, and discuss its advantages and limitations. A notable advantage of the Lagrangian approach is that, at least for linear convection–diffusion problems, it circumvents the CFL condition. This feature enables the use of adaptive mesh refinement (AMR) techniques without imposing constraints on the time step size. We then present recent advances in Lagrangian discretization methods, including a second-order discretization of the upper-convected time derivative, and highlight their potential for accurately and efficiently simulating complex flow phenomena.
