Computation of kinetic energy of the flow

feldy77 · November 9, 2011, 20:15

Hi foamers,
I am really puzzled about how tocompute the overall kinetic energy of the flow, i mean I(1/2*U^2)dV, where I is an integral over the given volume.
I understand that I have to sum over all subvolumes but I do not even know how to start . Please help me

nimasam · November 10, 2011, 07:16

volScalarField ke=0.5*(U&U);
dimensionedScalar ketotal = sum(ke*mesh.V());

feldy77 · November 10, 2011, 15:15

Dear nimasam,
Thank you very much for your answer it was very useful.
Just a small comment,
It is better just to compute the Ek just by

dimensionedScalar ketotal = sum(0.5*(U&U)*mesh.V());

In this way you do not need to construct an additional volScalar field
which can save alot of memory
Yuri

feldy77 · November 10, 2011, 15:43

Thinking about the issue once agin, the last change does not reduce the
memory consumption since U&U opration still consumpts the same amount of memory as if volScalarfield is defined explicitly. I would recommend another way , which is longer but allows you to save the memory:

forAll (mesh.C().internalField(), patchI)
{

Ekin_= Ekin_+0.5*
(sqr(U.internalField()[patchI].component(vector::X))+
sqr(U.internalField()[patchI].component(vector::Y))+
sqr(U.internalField()[patchI].component(vector::Z)))
*mesh.V()[patchI];
}
This is of course right for non-slip boundary conditions.
Yuri

November 9, 2011, 20:15	Computation of kinetic energy of the flow	#1
feldy77 Member Yuri Feldman Join Date: Mar 2011 Posts: 30 Rep Power: 15	Hi foamers, I am really puzzled about how tocompute the overall kinetic energy of the flow, i mean I(1/2*U^2)dV, where I is an integral over the given volume. I understand that I have to sum over all subvolumes but I do not even know how to start . Please help me

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
CFX-POST : gradients for turbulent kinetic energy	syler3321	CFX	1	June 16, 2011 05:47
mass flow rate computation in fluent	Kishore	FLUENT	5	June 27, 2007 03:53
Compressible flow computation with STAR	Veera	Siemens	0	June 19, 2004 04:31
Turbulent Kinetic Energy	Olga	FLUENT	2	October 11, 2002 15:05
computation about flow around a yawed cone	Tylor Xie	Main CFD Forum	0	June 9, 1999 07:33

November 10, 2011, 07:16		#2
nimasam Senior Member Nima Samkhaniani Join Date: Sep 2009 Location: Tehran, Iran Posts: 1,266 Blog Entries: 1 Rep Power: 24	volScalarField ke=0.5(U&U); dimensionedScalar ketotal = sum(kemesh.V());

November 10, 2011, 15:15		#3
feldy77 Member Yuri Feldman Join Date: Mar 2011 Posts: 30 Rep Power: 15	Dear nimasam, Thank you very much for your answer it was very useful. Just a small comment, It is better just to compute the Ek just by dimensionedScalar ketotal = sum(0.5(U&U)mesh.V()); In this way you do not need to construct an additional volScalar field which can save alot of memory Yuri

November 10, 2011, 15:43		#4
feldy77 Member Yuri Feldman Join Date: Mar 2011 Posts: 30 Rep Power: 15	Thinking about the issue once agin, the last change does not reduce the memory consumption since U&U opration still consumpts the same amount of memory as if volScalarfield is defined explicitly. I would recommend another way , which is longer but allows you to save the memory: forAll (mesh.C().internalField(), patchI) { Ekin_= Ekin_+0.5* (sqr(U.internalField()[patchI].component(vector::X))+ sqr(U.internalField()[patchI].component(vector::Y))+ sqr(U.internalField()[patchI].component(vector::Z))) *mesh.V()[patchI]; } This is of course right for non-slip boundary conditions. Yuri