Re: [deal.II] Stabilized FEM implementation with bubble function

Hi Wolfgang, Many thanks! The issue is resolved and the code is working. Regards, Lixing On Saturday, January 9, 2021 at 4:07:27 AM UTC+8 Wolfgang Bangerth wrote: > On 1/8/21 12:15 PM, Lixing Zhu wrote: > > > > This way I can freely use bubble functions and only construct local K > and F of >

Re: [deal.II] Stabilized FEM implementation with bubble function

On 1/8/21 12:15 PM, Lixing Zhu wrote: This way I can freely use bubble functions and only construct local K and F of FE_1. However, one issue is that, apparently, triangulation.cell does not support cell->get_dof_indices(local_dof_indices). Is there any way to call the DOFhandler by the cell

Re: [deal.II] Stabilized FEM implementation with bubble function

On 1/7/21 10:22 PM, Lixing Zhu wrote: You are right. I am not bounded by the memory at this moment. The only potential issue I can think of at this stage is the computation of the error norm. Technically, if I feed the slots in global RHS corresponding to bubble support as zero, the solution

Re: [deal.II] Stabilized FEM implementation with bubble function

Hi W and Jean-Paul, Thanks again for your suggestions. I realized that the bubble functions that I prefer are actually the shape functions from other standard elements. Thus I only need to define two sets of FE (i.g. FE_1 and FE_b). Then I use the iterator from triangulation as for (const auto

Re: [deal.II] Stabilized FEM implementation with bubble function

Hi W, You are right. I am not bounded by the memory at this moment. The only potential issue I can think of at this stage is the computation of the error norm. Technically, if I feed the slots in global RHS corresponding to bubble support as zero, the solution at the bubble support should zero.

Re: [deal.II] Stabilized FEM implementation with bubble function

On 1/7/21 6:58 PM, Lixing Zhu wrote: I implemented the local elimination as you suggested and it works. However, this ends up with many extra "redundant" DOFs in the global system. For example, if I am using N*N linear elements for a scalar 2D problem with structured mesh, instead of (N+1)^2

Re: [deal.II] Stabilized FEM implementation with bubble function

Dear Jean-Paul, Thanks for your suggestion. I'll try to test with these two approaches. Regards, Lixing On Friday, January 8, 2021 at 4:36:44 AM UTC+8 Jean-Paul Pelteret wrote: > Hi Lixing, > > Another tutorial that might be of interest to you to look at is step-44 >

Re: [deal.II] Stabilized FEM implementation with bubble function

Dear W. Many thanks for your suggestion. I guess I have to implement a new class for my own bubbles at some point. A follow-up question on (1). I implemented the local elimination as you suggested and it works. However, this ends up with many extra "redundant" DOFs in the global system. For e

Re: [deal.II] Stabilized FEM implementation with bubble function

1. I looked into the step-51 of the tutorial. It does illustrate a paradigm of segregating local DOFs and global DOFs. If I utilize this paradigm, the workflow would solve the local DOFs first (virtual node of the bubble function), which is a block matrix since bubble support from adjacent ce

Re: [deal.II] Stabilized FEM implementation with bubble function

Hi Lixing, Another tutorial that might be of interest to you to look at is step-44 . In that tutorial, two discontinuous fields are condensed out in one of two ways (there's a switch to choose which method is applied). The first appr

Re: [deal.II] Stabilized FEM implementation with bubble function

Dear Wolgang, 1. I looked into the step-51 of the tutorial. It does illustrate a paradigm of segregating local DOFs and global DOFs. If I utilize this paradigm, the workflow would solve the local DOFs first (virtual node of the bubble function), which is a block matrix since bubble support from

Re: [deal.II] Stabilized FEM implementation with bubble function

Many thanks for such an instant response. I'll adopt these ideas and try to eliminate those DOFs related to bubble support following the workflow in the tutorial. Regards, Lixing On Tuesday, January 5, 2021 at 1:21:56 PM UTC+8 Wolfgang Bangerth wrote: > > Lixing, > > > I am trying to implement

Re: [deal.II] Stabilized FEM implementation with bubble function

Lixing, I am trying to implement a stabiliazed weak form (e.g. advection-diffusion) where the stabilization tensor is computed element-wise through a standard bubble: \Pi(1-x_i^2). It seems that FE_Q_Bubbles should provides all I need, but here are two things I am not quite clear about, 1.

[deal.II] Stabilized FEM implementation with bubble function

Dear all, I am trying to implement a stabiliazed weak form (e.g. advection-diffusion) where the stabilization tensor is computed element-wise through a standard bubble: \Pi(1-x_i^2). It seems that FE_Q_Bubbles should provides all I need, but here are two things I am not quite clear about, 1. I