|
[Sponsors] |
[ImmersedBoundary] Advantage of using Immersed Boundary method in Polyhedral Mesh structure |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
June 29, 2016, 10:29 |
Advantage of using Immersed Boundary method in Polyhedral Mesh structure
|
#1 |
Member
Pavan
Join Date: Jan 2016
Posts: 53
Rep Power: 10 |
Please correct me if I am wrong, but I am in a sort of dilemma over using IBM solvers in Foam 3.2 extend:
I understand that OpenFOAM uses a polyhedral mesh structure which comes under the category "Unstructured mesh". So I am not being able to gauge the advantage of using an immersed boundary method on an unstructured mesh. Traditionally IBM have a lot of advantages in case of Cartesian meshes, but I am not able to understand the significance of IBM in case of polyhedral mesh structure. Can someone please throw some light on this? |
|
June 29, 2016, 21:29 |
|
#2 | |
Senior Member
Arjun
Join Date: Mar 2009
Location: Nurenberg, Germany
Posts: 1,272
Rep Power: 34 |
Quote:
The main advantage of IBM is the ability to move objects or deform without need to remesh. Now IBM in polyhderal meshes is nothing special, its same stuff. If you dont need polyhderal meshes than dont use IBM in polyhedral set up. It is this simple. Many people need polyhedral or unstructed meshes because they can not mesh their geom by cartesian mesh. |
||
June 29, 2016, 22:08 |
|
#3 | |
Member
Pavan
Join Date: Jan 2016
Posts: 53
Rep Power: 10 |
Quote:
I get the sense that you are referring to Cartesian in terms of cell shape. By Cartesian and Polyhedral I was not referring to the shape of the cell. I believe polyhedral and Cartesian meshes differ in the manner in which the which mesh information is saved the computer memory. In my understanding a Cartesian mesh is something where we do not need to store the connectivity information between the nodes, if there is node a (i,j) then it is implicitly understood that a(i+1,j) will be the next node in x-direction and a(i,j+1) will be the next node in the y-direction. However, in case of unstructured or polyhedral mesh structure we need to save the connectivity information along with with node location. So, IBM on a Cartesian mesh has the advantage of realizing complex boundaries without any change in mesh data structure since we add tha forcing term in the solver for realizing IB. Achieving direct forcing in a polyhedral mesh is not that staight forward. |
||
June 30, 2016, 17:47 |
|
#4 | |
Senior Member
Arjun
Join Date: Mar 2009
Location: Nurenberg, Germany
Posts: 1,272
Rep Power: 34 |
Quote:
Yes correct. I did write one IBM solver for cartesian mesh where we simply had Ni x Nj x Nk system. In this IBM had all the advantages you talk about only problem in our case (note I said our calculations) was that it required mesh size was too much to finally resolved what we needed. If it were a system where we had local refinement (oct tree type) I believe we could have achieved good resolution near boundaries with increasing mesh over all. So total mesh size may be a factor too in many cases. |
||
June 30, 2016, 18:01 |
|
#5 |
Member
Pavan
Join Date: Jan 2016
Posts: 53
Rep Power: 10 |
I am trying to write a IB code within an OpenFOAM solver and thus I need to know if there is any merit in trying to use IB on an unstructured grid
|
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
On the immersed boundary method | sbaffini | Main CFD Forum | 31 | December 5, 2019 04:39 |
Error - Solar absorber - Solar Thermal Radiation | MichaelK | CFX | 12 | September 1, 2016 05:15 |
Wrong flow in ratating domain problem | Sanyo | CFX | 17 | August 15, 2015 06:20 |
Error finding variable "THERMX" | sunilpatil | CFX | 8 | April 26, 2013 07:00 |
[Commercial meshers] Trimmed cell and embedded refinement mesh conversion issues | michele | OpenFOAM Meshing & Mesh Conversion | 2 | July 15, 2005 04:15 |