CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > OpenFOAM > OpenFOAM Post-Processing

How to access the pressure value of a moving probe when using Overset mesh?

Register Blogs Members List Search Today's Posts Mark Forums Read

Like Tree4Likes
  • 1 Post By wdx_cfd
  • 2 Post By wdx_cfd
  • 1 Post By wdx_cfd

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   September 27, 2019, 10:55
Question How to access the pressure value of a moving probe when using Overset mesh?
  #1
Member
 
Dongxu Wang
Join Date: Sep 2018
Location: China
Posts: 33
Rep Power: 7
wdx_cfd is on a distinguished road
Dear FOAMers,

I want to simulate the interaction between waves and a floating box using the two-phase model interFOAM. I want to sample the pressure of a moving sensor that fixed on the box. This is quite an easy task when using the dynamicMesh class of OpenFOAM v6.0. The probe function could perfectly resolve this problem with something just like this:

"fixedLocation false;"

However, since the dynamicMesh cannot resolve the problem of extrem mesh deformation, the Overset mesh of OpenFOAM v1906 is adopted. Now the biggest problem for me is to access the pressure of a probe on the floating box. I tried the same aforementioned setting, but the result is incorrect. In addition, when using these two mesh classes, the measured surface elevations are closed to each other. Therefore, I believe that the result is correct and there must be some problem on the setting of the probe function in controlDict. I am a new user of ESI-OpenFOAM as well as the Overset mesh, so I hope someone could provide some help to me.

Thanks a lot,
WDX.
wdx_cfd is offline   Reply With Quote

Old   September 30, 2019, 06:48
Default
  #2
Member
 
Dongxu Wang
Join Date: Sep 2018
Location: China
Posts: 33
Rep Power: 7
wdx_cfd is on a distinguished road
Hi, guys

Recently I found that the problem seems to be the location of the probe.

When using dynamicMesh class, the cell stretchs or compresses. Therefore, the location of the probe is changeable according the grid point. However, when using overset mesh, the background mesh is fixed, making the probe fixed even if the fixedlocation has been set to false.

Now, I wonder that, since the background mesh has been merged with the floating body mesh, how to access the probe located on the boundary of the floating body (between hole cells and interpolated cells, movable) rather than the background mesh?

Any helps will be appreciated.
Mohamed Shabara likes this.

Last edited by wdx_cfd; October 3, 2019 at 22:24.
wdx_cfd is offline   Reply With Quote

Old   October 4, 2019, 23:33
Default
  #3
Member
 
Dongxu Wang
Join Date: Sep 2018
Location: China
Posts: 33
Rep Power: 7
wdx_cfd is on a distinguished road
Hi, guys,

I solve this problem by using patchprobe class instead of probe!

Going deep into the code will never let you down!
amgh and Mohamed Shabara like this.
wdx_cfd is offline   Reply With Quote

Old   October 11, 2019, 06:35
Default
  #4
Member
 
Dongxu Wang
Join Date: Sep 2018
Location: China
Posts: 33
Rep Power: 7
wdx_cfd is on a distinguished road
Unfortanately, the calculated p_rgh field seems not correct although I use patchprobe class but the motion state and surface elevation are correct. I will keep trying to find out the reason why this happens.
wdx_cfd is offline   Reply With Quote

Old   October 28, 2019, 18:38
Default
  #5
Senior Member
 
Join Date: Jan 2019
Posts: 125
Blog Entries: 1
Rep Power: 0
Michael@UW is on a distinguished road
Hi wdx_cfd,
Did you solve this problem? I have the same issue to get the pressure on the surface of the moving object when using overset mesh on OpenFOAM v1906.

Best,
Michael
Michael@UW is offline   Reply With Quote

Old   October 28, 2019, 18:39
Default
  #6
Senior Member
 
Join Date: Jan 2019
Posts: 125
Blog Entries: 1
Rep Power: 0
Michael@UW is on a distinguished road
Quote:
Originally Posted by wdx_cfd View Post
Unfortanately, the calculated p_rgh field seems not correct although I use patchprobe class but the motion state and surface elevation are correct. I will keep trying to find out the reason why this happens.
Do it work correctly now? Can you post your setup?
Michael@UW is offline   Reply With Quote

Old   October 28, 2019, 21:27
Default
  #7
Member
 
Dongxu Wang
Join Date: Sep 2018
Location: China
Posts: 33
Rep Power: 7
wdx_cfd is on a distinguished road
Quote:
Originally Posted by Michael@UW View Post
Do it work correctly now? Can you post your setup?
I haven't solve this problem. I use the default setup just like this:
Code:
            patchProbes
	    {
		type            patchProbes;
		libs            ( "libsampling.so" );

		// Name of the directory for probe data
		name            patchProbes;

		// Patches to sample (wildcards allowed)
		patches         ("floatingObject");

		// Write at same frequency as fields
		writeControl    timeStep;
		writeInterval   1;

		interpolationScheme cellPoint;

		// Fields to be probed
		fields          (p_rgh p);

		fixedLocations  false;

		// Locations to probe. These get snapped onto the nearest point
		// on the selected patches
		probeLocations
		(
				(0.9 0.3 0.04)
				(0.9 0.3 0.0)
				(0.8 0.3 -0.03)            
		);
I think presumably there is a problem of the fixed background mesh, which makes the 'fixedLocation' ineffective. Maybe we could access the different meshes through 'zoneID' and 'cellType'. But recently I do not have enough time to work on this. If you have some progress on this problem, please tell me, thank you very much.
Richal Sun likes this.
wdx_cfd is offline   Reply With Quote

Old   October 28, 2019, 21:41
Default
  #8
Senior Member
 
Join Date: Jan 2019
Posts: 125
Blog Entries: 1
Rep Power: 0
Michael@UW is on a distinguished road
Thank you very much for your reply.

I use this setup, and the pressure does no longer keep unchanged with time. But I haven’t inspected if the pressure probed is correct or not. I assume you did this before and conclude it is not.

Cell type and ID is a good idea to tract specific cells, but it needs to write a function object and takes time.

I will let you know if I have some progress on this problem for sure.
Michael@UW is offline   Reply With Quote

Old   November 2, 2019, 04:51
Default
  #9
Member
 
Dongxu Wang
Join Date: Sep 2018
Location: China
Posts: 33
Rep Power: 7
wdx_cfd is on a distinguished road
Quote:
Originally Posted by Michael@UW View Post
Thank you very much for your reply.

I use this setup, and the pressure does no longer keep unchanged with time. But I haven’t inspected if the pressure probed is correct or not. I assume you did this before and conclude it is not.

Cell type and ID is a good idea to tract specific cells, but it needs to write a function object and takes time.

I will let you know if I have some progress on this problem for sure.
As far as I know, the value is not correct. However, I am not sure about it because there maybe some problems on my exprimental data. Recently I am trying my best to post-process my data. I will also let you know if I have progress on this problem.
wdx_cfd is offline   Reply With Quote

Old   November 2, 2019, 10:16
Default
  #10
Senior Member
 
Join Date: Jan 2019
Posts: 125
Blog Entries: 1
Rep Power: 0
Michael@UW is on a distinguished road
The pressure history looks good though it is smaller than the experimental data.
It is probably because the time step is not small enough to capture the peak. So I expect patchprobe works for overset. I am trying to improve the mesh and temporal resolution to confirm the pressure.
Michael@UW is offline   Reply With Quote

Old   November 3, 2019, 23:46
Default
  #11
Member
 
Dongxu Wang
Join Date: Sep 2018
Location: China
Posts: 33
Rep Power: 7
wdx_cfd is on a distinguished road
Quote:
Originally Posted by Michael@UW View Post
The pressure history looks good though it is smaller than the experimental data.
It is probably because the time step is not small enough to capture the peak. So I expect patchprobe works for overset. I am trying to improve the mesh and temporal resolution to confirm the pressure.
I use the dynamicMesh Class simulate the same case (The case crashed because the mesh has been over deformed, but it is enough for comparison). I found that the pressure signals cannot match each other. The pictures in the attachment are the comparison between the different methods. The blue line is the signal calculated by oversetMesh and the other is by dynamicMesh. I also compared them with my experimental data and found that the orange line matches the experimental result much better.

On the other hand, the motion states calculated by both methods are almost identical. Therefore, I believed that there must be some problems in the pressure capture process when using the Probe/Patchprobe function with oversetMesh.
Attached Images
File Type: jpg probe1.jpg (10.7 KB, 54 views)
File Type: jpg probe2.jpg (14.2 KB, 36 views)
File Type: jpg probe3.jpg (12.9 KB, 31 views)
wdx_cfd is offline   Reply With Quote

Old   November 12, 2019, 23:15
Default
  #12
Member
 
Dongxu Wang
Join Date: Sep 2018
Location: China
Posts: 33
Rep Power: 7
wdx_cfd is on a distinguished road
Quote:
Originally Posted by Michael@UW View Post
The pressure history looks good though it is smaller than the experimental data.
It is probably because the time step is not small enough to capture the peak. So I expect patchprobe works for overset. I am trying to improve the mesh and temporal resolution to confirm the pressure.
Hi, michael

Last night, I finally had time to see the code. I think I figured out this problem. Yes, you are right, using patchProbe class could sample the pressure at moving probes. However, one thing should be noted: the fixedLocations should be set to TRUE (i.e., the default value) or the location of the probe will NOT be updated. I used to set it to FALSE therefore the outputed pressure seems incorrect. Moreover, the p_rgh value seems not correct, therefore samping the p field is recommended.

Other errors are likely due to the difference between the experiment and numerical model. But the error always exists, isn't it?

Hope this could be useful for others.

WDX.
wdx_cfd is offline   Reply With Quote

Old   November 13, 2019, 11:58
Default
  #13
Senior Member
 
Join Date: Jan 2019
Posts: 125
Blog Entries: 1
Rep Power: 0
Michael@UW is on a distinguished road
Hi WDX,

Thank you for your digging into the code and confirmation. I didn't realize I used 'fixedLoacations true' until you pointed out that. This setup really contradicts intuition as the probe is moving with mesh. It also explains why 'p-rgh' is incorrect, because the location of the probe is fixed, if p is correct, then p-rgh must be not.

Michael
Michael@UW is offline   Reply With Quote

Reply

Tags
moving probe, oversetmesh, pressure, runtime processing

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
sliding mesh problem in CFX Saima CFX 46 September 11, 2021 08:38
How to set up an overset mesh ? (Moving square inside a rectangular box) jordan.c STAR-CCM+ 5 September 6, 2019 09:53
Moving Body Simulation (Overset Mesh) Kishen STAR-CCM+ 4 December 8, 2015 12:30
[Other] How to set up a dynamic mesh for a piston moving through a tube of variable diameter? karkar OpenFOAM Meshing & Mesh Conversion 0 July 4, 2012 07:54
pressure oscillations in moving mesh cases david OpenFOAM Running, Solving & CFD 5 April 25, 2011 12:16


All times are GMT -4. The time now is 17:17.