The problem of changing the shape of the immersion body structure.

[Dewu Yang]

Dear professor Amneet,

     I modified the shape of the fish in the eel2d example, but the new eel2d.vertex file could not make the program run normally, and an error occurred. And the program could run normally after replacing it with the original eel2d.vertex file. Can't I only replace the vertex file？I would appreciate it if you can give me some advice about it.

Best,

Dewu

[Amneet Bhalla]

Eel2d and eel3d kinematics are specifically tied to the geometries and Lagrangian discretization of the bodies in those examples. 

You might want to follow up on eel3d example to write your own eel2d kinematics for a different shaped/discretized body.

--
You received this message because you are subscribed to the Google Groups "IBAMR Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ibamr-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ibamr-users/293c57f0-ad9c-4f42-bf20-89b4db2d2b7an%40googlegroups.com.

--

--Amneet 

[Boyce Griffith]

On Apr 19, 2021, at 10:01 AM, Amneet Bhalla <mail2...@gmail.com> wrote:

Eel2d and eel3d kinematics are specifically tied to the geometries and Lagrangian discretization of the bodies in those examples. 

Do these specializations show up in only a few places in the example? If so, would it be easy to identify what the assumption is (e.g. via a comment)?

— Boyce

You might want to follow up on eel3d example to write your own eel2d kinematics for a different shaped/discretized body.

On Mon, Apr 19, 2021 at 6:47 AM Dewu Yang <yangd...@gmail.com> wrote:

Dear professor Amneet,

     I modified the shape of the fish in the eel2d example, but the new eel2d.vertex file could not make the program run normally, and an error occurred. And the program could run normally after replacing it with the original eel2d.vertex file. Can't I only replace the vertex file？I would appreciate it if you can give me some advice about it.

Best,

Dewu

<4C760BF3788AF35263C909AE5F975923.jpg>

-- 
You received this message because you are subscribed to the Google Groups "IBAMR Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ibamr-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ibamr-users/293c57f0-ad9c-4f42-bf20-89b4db2d2b7an%40googlegroups.com.

-- 

--Amneet 

-- 
You received this message because you are subscribed to the Google Groups "IBAMR Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ibamr-users...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/ibamr-users/CAMETWJ1LvxB2b4MCGQ5ep%2BtP9T3bOcTP8b-DUGVhBbesMQA5PA%40mail.gmail.com.

[Amneet Bhalla]

Yeah only for fish examples, because of non-unique and specific kinematics for every fish. I feel that there are comments in the kinematics files that suggest that. 

To view this discussion on the web visit https://groups.google.com/d/msgid/ibamr-users/82ECC68C-B105-4F67-A3F3-69F5640B392E%40gmail.com.

--

--Amneet 

[Dewu Yang]

Dear Prof. Bhalla, Griffith,

      Thanks so much for your advice. I will try to analyze the eel3d code. However, it currently displays that my computer does not have GNU library. What puzzles me is that GNU is installed on my computer. Next, I will try to compile and re-install the GNU compiler.

       Besides, I encountered a problem when I changed the grid size and calculation domain. When I increased the calculation domain and changed the grid size. To make sure that there is more or less one Lagrangian point per Eulerian cell on the finest mesh level in my changed domain configuration, I adjusted the size of N (actual number of grid cells on coarsest grid level)to meet the above requirements. For example, I chose the grid with 40*16 grid cells in the coarsest-level patch, and I changed the value of N to 16(In this case, I chose the four-level Cartesian grid). But there was an error in .log file:

      // Lower/Upper side of integration box is not aligned with sides on coarsest level in dimension 0/1. Modifying coordinate to nearest box side.//

No matter what changes I made to the coordinates, this error always appears. What caused it？

Best,

Dewu

[Dewu Yang]

N is only in the y-direction. I tried to divide N into Nx and Ny. By changing the corresponding calculation domain, the program can run normally. Does this exactly satisfy the condition that the change of the grid corresponds to the size of the computational domain?

[Dewu Yang]

      I encountered another problem. After adding the changed Eel.vertex file, it couldn't run in the originally defined grid domain environment. I could only reduce the grid refinement level to 2 (the original refinement level is 4). However, the animation that came out was very strange. The shape of the fish changed at the moment of movement, which puzzled me. The following is the effect animation.

On Monday, April 19, 2021 at 10:01:25 PM UTC+8 mail2...@gmail.com wrote:

[LIMING CHAO]

Dear Prof. Amneet, 

Following with your and Dewu's discussion, I have a question how could I simulate the self-propelled fish with the complicated geometry? In eel3d examples, for example,  you control the body shape of the fish with special functions. If my fish can't be described using  the elliptical cross sections, what should I do?  For example, I attached a real fish geometry, how can I control this fish? 

Best,

Li-Ming