Robust Numerical Methods for Partial Differential Equations


Project Description

The general theme of this project is to study numerical methods for systems of partial differential equations which depend on one or more critical parameters. Typically, we are interested in systems which change type as a critical perturbation parameter tend to zero. Our goal is to construct numerical methods with convergence properties which are uniform with respect to the perturbation parameter. We will also study the influence of the parameter on iterative solvers. The project is divided into three sub-projects with one of the PIs of the project managing the progress of the associated  research.


Contact Persons

The program will have three principle investigators (PIs) serving as a program committee:

Professor Magne Espedal, University of Bergen, Department of Mathematics; Magne.Espedal@mi.uib.no;
Professor Xue-Cheng Tail, University of Bergen, Department of Mathematics; Xue-Cheng.Tai@mi.uib.no;
Professor Ragnar Winther, University of Oslo, Department of Mathematics and Department of Informatics, rwinther@ifi.uio.no.

Professor Winther will act as the administrative coordinator of the project and be responsible for the correspondence with the Research Council. The formation of the present group is based on informal contacts between the members over many years. These contacts has lead to some joint papers between Espedal and Tai and between Tai and Winther. The main motivation for making a joint project is to create a research group with a balanced background in mathematical theory, numerical computations and applications. It is also the intention that there will be educational benefits from the cooperation between the groups in Oslo and Bergen.


Sponsor

This project is supported by The Research Council of Norway (Norges forskningsraad) under project BeMatA.


Candidates

The following candidates are taken up for the positions in connection with this project.

Marius will be mainly working on partial differential equations from image processing and inverse problems using different regularization techniques and level-set type of methods.  The equations are have a mixed property of nonloinear convection nd diffusion problems.  He will be mainly working with Xue-Cheng Tai and Ragnar Winther.  Heimsunf has been working with preconditioning techniques and level set method in his master degree thesis. The method he will concentrate in this project will be ELLAM (Euler-Lagrangian localized adjoint methods) for advection-diffusion problems, especially, he will be engaged in implementing a 3D-simulators under the supervision of Magne Espedal.  Kozlov will concentarte on some partial differential equations coming from dynamic systems with constraints and he will mostly work with Ragnar Winther.


Publications

We listed some of the publications that have been done in connections with this project. Some of them are available online and can be downloaded from: http://www.mi.uib.no/~resme,  http://www.mi.uib.no/~tai or  http://www.ifi.uio.no/~rwinther


Maintained by   Xue-Cheng Tai  (Tai@mi.uib.no)