Alberto F. Martin

icon indicating a website link https://amartinhuertas.github.io/

icon company Australian National University icon location Canberra, ACT, Australia icon location Computational scientist working in the interface of Scientific Computing (in particular numerical solution of PDEs), HPC, and ML. See my web below.

Results 74 comments of Alberto F. Martin

HDG for advection-diffusion on the sphere

> They are doing material form advection \beta.\nabla u, whereas I am doing flux form \nabla\cdot(\beta u) For a divergence free vector field, \nabla\cdot\beta = 0, the two are the...

HDG for advection-diffusion on the sphere

> We could do un continuous or discrete. I would do it continuous (as far as possible). Is there any reason you need to interpolate it into a FE space?...

HDG for advection-diffusion on the sphere

> I don't understand the differences in the mesh skeleton term w.r.t. Kang et al 2020. (ie the use of |\bea\cdot n|) - do you have any thoughts on this??...

HDG for advection-diffusion on the sphere

The general form of the upwind HDG discretization for problem (2.1) reads as: ![Screenshot from 2022-07-06 21-51-22](https://user-images.githubusercontent.com/38347633/177543943-9ee29a4d-5cc2-470a-8ef4-e1a1f6164782.png) In the case of problem (3.1), A is diagonal matrix, with \beta_k n_k...