[Varada]

Hello REEF3D team,

I am currently working on a heave decay simulation of a cylinder using REEF3D, as detailed in the paper titled "Simulation of floating bodies using a combined immersed boundary with the level set method in REEF3D" (H. Bihs, A. Kamath, J. Lu, and Ø. Arntsen, MARINE VII Conference, 2017).

In my study, I initially achieved good agreement between experimental and numerical results. However, as time progresses, I've noticed a growing deviation between the two sets of results.

I have attached the relevant control and ctrl files used in the simulation, along with a comparative plot for your reference. I would greatly appreciate it if you could spare some time to review these files. If you identify any discrepancies or potential errors in my input files, kindly guide me in rectifying them.

[kamath]

As time proceeds, the motions and waves produced become smaller. Decay tests are generally not really good for replication of experimental results in numerical model.
Check with a higher grid resolution

[Varada]

Help Needed: Choosing the Right X 41 Value for floating body simulations in REEF3D

I'm working with REEF3D and I'm a bit confused about the X 41 parameter. The default value is 0.6, but in the ctrl.txt for heave decay of a sphere given in tutorial (user REEF3D guide), you used 2.1.

I tried both values for a cylinder's heave decay in a 2D wave tank, and the results are different. Any suggestions on which value I should use?

Thanks you

[TrungPham]

Originally Posted by Varada

Hello REEF3D team,

I am currently working on a heave decay simulation of a cylinder using REEF3D, as detailed in the paper titled "Simulation of floating bodies using a combined immersed boundary with the level set method in REEF3D" (H. Bihs, A. Kamath, J. Lu, and Ø. Arntsen, MARINE VII Conference, 2017).

In my study, I initially achieved good agreement between experimental and numerical results. However, as time progresses, I've noticed a growing deviation between the two sets of results.

I have attached the relevant control and ctrl files used in the simulation, along with a comparative plot for your reference. I would greatly appreciate it if you could spare some time to review these files. If you identify any discrepancies or potential errors in my input files, kindly guide me in rectifying them.

Hi Varadar,
I have used your control and ctrl files (just change number of processors) but it did not work (see attachment picture )
Please kindly guide me in rectifying them.
Thank you ever so much

[Varada]

I believe the input files are ok since the simulation is running on my computer. Please try by upgrading to the latest version of REEF3D.

[TrungPham]

Originally Posted by varada

i believe the input files are ok since the simulation is running on my computer. Please try by upgrading to the latest version of reef3d.

control.txt

ctrl.txt
Dear Varada,
Thank you for your replying. I have just updated the latest version of REEF3D and run but there's the same error. May be I will try again.
Best regards

[soydana]

Originally Posted by Varada

Hello REEF3D team,

I am currently working on a heave decay simulation of a cylinder using REEF3D, as detailed in the paper titled "Simulation of floating bodies using a combined immersed boundary with the level set method in REEF3D" (H. Bihs, A. Kamath, J. Lu, and Ø. Arntsen, MARINE VII Conference, 2017).

In my study, I initially achieved good agreement between experimental and numerical results. However, as time progresses, I've noticed a growing deviation between the two sets of results.

I have attached the relevant control and ctrl files used in the simulation, along with a comparative plot for your reference. I would greatly appreciate it if you could spare some time to review these files. If you identify any discrepancies or potential errors in my input files, kindly guide me in rectifying them.

First and foremost, make sure that you are using the latest Release Candidate of REEF3D.

Second, N40 should be 4 in your ctrl file.

Lastly, since the center of the cylinder is placed 0.02454 m above the free surface

center of location of your cylinder should be 1.24454 as below.

X 132 0.0762 0.01 5.0 0.005 1.24454

Then you are supposed to get better results.

[TrungPham]

Originally Posted by soydana

First and foremost, make sure that you are using the latest Release Candidate of REEF3D.

Second, N40 should be 4 in your ctrl file.

Lastly, since the center of the cylinder is placed 0.02454 m above the free surface

center of location of your cylinder should be 1.24454 as below.

X 132 0.0762 0.01 5.0 0.005 1.24454

Then you are supposed to get better results.

Thank you for your help.
Best regards

[Varada]

Thank you for your response. Based on the tutorial, it seems that the value 3 is commonly used for N = 40. I'm curious if using 4 could potentially yield improved results, especially for simulations involving floating bodies. Another parameter I'm unsure about is X 41, which has a default value of 0.6. In the heave decay of a sphere tutorial, the value 2.1 is utilized. Could you kindly provide guidance on selecting an appropriate value for this parameter?

[soydana]

Originally Posted by Varada

Thank you for your response. Based on the tutorial, it seems that the value 3 is commonly used for N = 40. I'm curious if using 4 could potentially yield improved results, especially for simulations involving floating bodies. Another parameter I'm unsure about is X 41, which has a default value of 0.6. In the heave decay of a sphere tutorial, the value 2.1 is utilized. Could you kindly provide guidance on selecting an appropriate value for this parameter?

With the new release candidate of REEF3D, it needs to be used N 40 4 which was N40 3 before.

When it comes to the X 41 parameter, a thinner thickness layer can be used now as some improvements have been applied to the 6DOF algorithm in REEF3D. So I'd recommend you to use 0.6 which is a default value for X 41 now.

[TrungPham]

Originally Posted by soydana

With the new release candidate of REEF3D, it needs to be used N 40 4 which was N40 3 before.

When it comes to the X 41 parameter, a thinner thickness layer can be used now as some improvements have been applied to the 6DOF algorithm in REEF3D. So I'd recommend you to use 0.6 which is a default value for X 41 now.

When using the command N 40 4, I get an error like the image attached (if I use N 40 3, I don't get this error).
Thank you in advance for your help.
N40.JPG

[soydana]

Originally Posted by TrungPham

When using the command N 40 4, I get an error like the image attached (if I use N 40 3, I don't get this error).
Thank you in advance for your help.
Attachment 95869

Can you share your control files?

[valgrinda]

I think you guys are confusing code versions:

Varada is using the master, whereas the change to N 40 4 as low storage RK is only in the latest RC.

In Varada's version, N 40 4 is still IMEX (removed in RC).

[TrungPham]

[QUOTE=soydana;855814]
Can you share your control files?
Here is the control and ctrl files.
Best regards
control.txt
ctrl.txt

Hi All,
By the way, I have a few more questions:
Based on the setup CFD model presented in as detailed in the paper titled "Simulation of floating bodies using a combined immersed boundary with the level set method in REEF3D" (H. Bihs, A. Kamath, J. Lu, and Ø. Arntsen, MARINE VII Conference, 2017), I adapted my example as follows:
1. The spatial domain of the numerical wave tank (NWT) is 30m long 1.2m width and 1.0m high. The water depth is d=0.71m and the water is initially still. The density of the water is ρ=1000kg/m3.
2. A circular cylinder is located at 21.5m far from Wavemaker.
3. Input wave parameters (height, period): 0.17m, 2.27s
3. Water depth 0.71m
4. The body is a horizontal circular cylinder with a radius r=0.115m and half of water density, ρ=500kg/m3. The cylinder is partially submerged with its centroid at h1=0.03m above the free surface.
Is it correct to declare X 131 statement as below?
X 131 0.115 21.615 21.5 0.60 0.74