Can OpenFoam solve this problem?

salazardetroya · July 9, 2015, 15:18

Hello everybody

I am trying to solve the next problem that I have been struggling with trying to implement it by myself. I was wondering if I could use OpenFoam instead and how long it could take me in order to figure out how to solve it (I'm pretty good with C++ and my knowledge on turbulence models is decent at this point) I am asking this question because I have a time constraint (1 month) and I don't know if I will have time to switch and solve the problem with OpenFoam or try to debug my implementation. So basically I'm asking for two simple questions: Can OpenFoam solve this problem? how long would it take me given my current knowledge on turbulence models and C++?

The problem is this described in this paper (http://onlinelibrary.wiley.com/doi/1...20126/abstract) I will describe it here as well, but you can find the equations there.

I'm solving a turbidity current vertical profile. From the 2D RANS equations, I neglect the convective terms because I assume the flow has developed and does not change in the longitudinal direction. With that assumption and the conservation equation, I am left with just the streamwise velocity component's equation. This mean flow equation has a source term which is the sediment concentration. The concentration follows a conservation equation. As a turbulence model, I'm using k-epsilon.

One thing different from the paper mentioned above is that in my case there is no roof, the turbidity current is in the lower part of a larger domain that it is mostly at rest.

The tricky part of the implementation is the wall function. Special care needs to be taken because of the buoyant forces produced by the concentration. Can OpenFoam take this into account? The common log-law is not accurate since it ignores buoyant forces.

So, can OpenFoam solve this? If so, how long will it take me?

Please let me know if you might need more details. Thanks in advance.

Edit: High Reynolds number is assumed.

Miguel

Navip · July 29, 2015, 23:34

Hi Miguel, did you solve your problem ?

If there's settling within your turbidity current you can use settlingFoam. I use it and reproduce well experiment found in literature except that the advective term in the AlphaEqn.H are missing. It could explain the early the stop of my turbidity current.

I try to implement the advectiveFvPatchField.H within this solver and see how it does affect the motion.

Xavier

July 9, 2015, 15:18	Can OpenFoam solve this problem?	#1
salazardetroya New Member Miguel Salazar Join Date: Jul 2015 Posts: 9 Rep Power: 11	Hello everybody I am trying to solve the next problem that I have been struggling with trying to implement it by myself. I was wondering if I could use OpenFoam instead and how long it could take me in order to figure out how to solve it (I'm pretty good with C++ and my knowledge on turbulence models is decent at this point) I am asking this question because I have a time constraint (1 month) and I don't know if I will have time to switch and solve the problem with OpenFoam or try to debug my implementation. So basically I'm asking for two simple questions: Can OpenFoam solve this problem? how long would it take me given my current knowledge on turbulence models and C++? The problem is this described in this paper (http://onlinelibrary.wiley.com/doi/1...20126/abstract) I will describe it here as well, but you can find the equations there. I'm solving a turbidity current vertical profile. From the 2D RANS equations, I neglect the convective terms because I assume the flow has developed and does not change in the longitudinal direction. With that assumption and the conservation equation, I am left with just the streamwise velocity component's equation. This mean flow equation has a source term which is the sediment concentration. The concentration follows a conservation equation. As a turbulence model, I'm using k-epsilon. One thing different from the paper mentioned above is that in my case there is no roof, the turbidity current is in the lower part of a larger domain that it is mostly at rest. The tricky part of the implementation is the wall function. Special care needs to be taken because of the buoyant forces produced by the concentration. Can OpenFoam take this into account? The common log-law is not accurate since it ignores buoyant forces. So, can OpenFoam solve this? If so, how long will it take me? Please let me know if you might need more details. Thanks in advance. Edit: High Reynolds number is assumed. Miguel Last edited by salazardetroya; July 9, 2015 at 15:40. Reason: Forgot details

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Problem when installing Openfoam 2.0.x with mac	giovaniharyadi	OpenFOAM Installation	0	March 24, 2012 19:45
Problem installing OpenFOAM 1.5 installation on RHEL 4.	vwsj84	OpenFOAM Installation	4	April 23, 2009 05:48
Adventure of fisrst openfoam installation on Ubuntu 710	jussi	OpenFOAM Installation	0	April 24, 2008 15:25
Problem in installation of OpenFOAM	sachin	OpenFOAM Installation	7	January 22, 2008 02:40
Can FLUENT solve this type of problem?	Whitaker	FLUENT	3	October 23, 2000 04:01

July 29, 2015, 23:34		#2
Navip New Member Xavier Pivan Join Date: May 2015 Posts: 10 Rep Power: 11	Hi Miguel, did you solve your problem ? If there's settling within your turbidity current you can use settlingFoam. I use it and reproduce well experiment found in literature except that the advective term in the AlphaEqn.H are missing. It could explain the early the stop of my turbidity current. I try to implement the advectiveFvPatchField.H within this solver and see how it does affect the motion. Xavier