how to set a proper value for froude_cap?

[JimKirby]

I believe that choosing a value of 0.5 would always be safe - Gangfeng or Cheng please respond if I'm incorrect.

[jue010811]

I find the wave overtopping volume is directly influenced by froude_cap. If froude_cap equals 10, the wave overtopping volume calculated by nhwave is alomost 2 times larger than the experiment. If froude_cap equals 0.5, almost no wave overtopping happen. The experiment is Oumeraci's.

[JimKirby]

On 5/17/17 12:26 AM, jue010811 wrote:

I find the wave overtopping volume is directly influenced by froude_cap. If froude_cap equals 10, the wave overtopping volume calculated by nhwave is alomost 2 times larger than the experiment. If froude_cap equals 0.5, almost no wave overtopping happen. The experiment is Oumeraci's.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/JimKirby/NHWAVE/issues/18#issuecomment-301981946, or mute the thread https://github.com/notifications/unsubscribe-auth/AOODxebj6xIyAW5Hw63mD5lRSuGI3EXpks5r6ncRgaJpZM4Nanmo.

Could you share some example calculations and the reference for the data? Overtopping is probably a challenging calculation. I'd ask what your other parameter choices are, but I would probably be guided better by seeing what is happening.

Jim

[jue010811]

@JimKirby Thank you for your reply! I send the input files to your email(udel.edu).

[jue010811]

Sorry to bother you. But,have you received my input files? Is there any parameter wrong?

[gangfengma]

This parameter was added by Cheng to limit the flow speed in shallow water. Overtopping will certainly be affected by this parameter because the flow speed will be adjusted if you use a small value. You need use a large value (for example 10) to not allow this parameter to affect your results. We need find other reasons why the overtopping volume is overpredicted. This is indeed a challenging problem. Could you send a copy of your input files to me (gma@odu.edu)? I will take a look.

[gangfengma]

There is a major issue in your input file. The grid number in x direction Mglob should be divisible by the number of processors PX. In addition, I think wave runup and overtopping can be greatly affected by bottom friction. So you may add a little friction to reduce the overtopping volume. I don't know if it's better to turn on the turbulence model, which might be capable of better predicting energy dissipation during wave breaking. To include turbulence modeling, more vertical layers are recommended.

[fengyanshi]

Gangfeng - Just let you know that FUNWAVE-TVD can use a random number of processors right now. It was the recent work from my visitor Dr. Choi. I have push the code to my github, funwave-tvd repository. It's an easy fix but needs some work to modify Putfile and Getfile because gather and scatter need a uniform number for each processor.

[gangfengma]

Okay. I will update these subroutines in NHWAVE.

Gangfeng Ma, Ph.D. Assistant Professor and Director of Coastal Engineering Institute Department of Civil and Environmental Engineering Old Dominion University Norfolk, VA 23529 Phone: (757) 683-4732 https://sites.google.com/site/gangfma/home

---

From: fengyanshi notifications@github.com Sent: Tuesday, May 23, 2017 2:53:25 PM To: JimKirby/NHWAVE Cc: Ma, Gangfeng; Comment Subject: Re: [JimKirby/NHWAVE] how to set a proper value for froude_cap? (#18)

Gangfeng - Just let you know that FUNWAVE-TVD can use a random number of processors right now. It was the recent work from my visitor Dr. Choi. I have push the code to my github, funwave-tvd repository. It's an easy fix but needs some work to modify Putfile and Getfile because gather and scatter need a uniform number for each processor.

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2FJimKirby%2FNHWAVE%2Fissues%2F18%23issuecomment-303497148&data=01%7C01%7Cgma%40odu.edu%7C0895e8b3ee8048133d5408d4a20cff43%7C48bf86e811a24b8a8cb368d8be2227f3%7C0&sdata=ceirtIzfrlKWb0OklUXUcVPgnvf9UYgZIU90JmaLHI4%3D&reserved=0, or mute the threadhttps://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAQtICH_bvx3fiOZ-GYLc6iDz3aGNqRUDks5r8yslgaJpZM4Nanmo&data=01%7C01%7Cgma%40odu.edu%7C0895e8b3ee8048133d5408d4a20cff43%7C48bf86e811a24b8a8cb368d8be2227f3%7C0&sdata=1Qni7r2CmTWTnmxEpJ8ZblV%2FGGt42Ol7Gi4whRMIkKE%3D&reserved=0.

[jue010811]

Thanks for your correction, I change the number of processors PX now. I turn on the turbulence model,and set 10 vertical layers now,but it does not help much. Wave runup and overtopping is indeed greatly affected by bottom friction, but the experiments are done under the smooth condition, so i don't know if it's appropriate to add a little friction. Now I represent the breakwater by the bathymetry, can i just add friction on the breakwater ,and make other area remain smooth? Later I want to simulate wave overtopping on the breakwater with armor stones.

[gangfengma]

Right now the model does not have this function to specify friction at certain region. But I think it's not difficult to modify the code for your applications. You need change the code in subroutine vel_bc for different boundary conditions at different regions.