Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
We consider linear and nonlinear convergence acceleration techniques in the framework of Newton or inexact Newton methods. The proposed procedure is based on a new dynamic preconditioner to be used in combination with the GMRES method for reducing the cost of solving a sequence of linear systems. A nonlinear convergence acceleration technique based on a previous work by Washio et al. is also added. The benefits of this combination of acceleration techniques is shown on fluid-structure interaction problems in haemodynamics for two- and three-dimensional applications.
Alfio Quarteroni, Simone Deparis, Andrea Manzoni, Niccolò Dal Santo