News and recent events
- PhD opportunity:
Adaptive numerical algorithms for PDE problems with random data.
- 9-13 June, 2016:
Norbert Heuer (Pontificia Universidad Católica de Chile, Santiago, Chile) visited us and gave a talk at the
Applied Mathematics Seminar.
- 5-8 April, 2016:
attending the SIAM Conference on Uncertainty Quantification
at EPFL in Lausanne, Switzerland and
speaking at the Minisymposium "Error Estimation and Adaptive Methods for Uncertainty Quantification
in Computational Sciences".
Travel support from The Institute of Mathematics and its Applications
(Small Grant Scheme) to attend the conference is gratefully acknowledged.
- Paper published:
together with Serge Nicaise we studied convergence rates of the h-version of the boundary element method (BEM) on graded meshes for the
electric field integral equation on polyhedral surfaces. We have proved that, even though quasi-optimal convergence of the method holds
under a restriction on the strength of mesh grading,
the h-BEM on sufficiently graded meshes still regains an optimal convergence rate.
Link to the paper
Download the preprint
- 25 February, 2016:
Sergey Dolgov (University of Bath) visited us and gave a talk at the
Optimisation & Numerical Analysis Seminar.
- 22 February, 2016:
Angela Mihai (Cardiff University) visited us and gave a talk at the
Applied Mathematics Seminar.
- 15 February, 2016:
Tony Shardlow (University of Bath) visited us and gave a talk at the
Applied Mathematics Seminar.
- 5-8 January, 2016:
Dirk Praetorius (Vienna University of Technology, Austria) visited us and gave a talk at the
joint Applied Mathematics and Optimisation & Numerical Analysis Seminar.
- 5-6 January, 2016: the workshop
"Adaptive algorithms for computational PDEs"
was held at the University of Birmingham thanks to support by the
London Mathematical Society and
the School of Mathematics.
- New preprint:
together with David Silvester we have designed and implemented an adaptive algorithm
for efficient solution of elliptic PDEs with random data.
The adaptive refinement process is driven by precise estimates of the energy error reductions
that will occur if different refinement strategies are pursued.
Numerical experiments were performed using
our freely available software S-IFISS.
Preprint on MIMS EPrints.