Can anyone suggest how I can plot Q criterion in fluent to visualize the coherent structures?

Thanks for your help.

Anindya

hi anindya,

you certainly have the definition of that q criterion which is based on velocity gradient. Fluent do not provide it directly,i mean you have to compose it from computed velocity gradient as a custom field function. for that: define>custom field function and type the name of the criterion "q" for example. you will be then able to visualize the q value on a specified surface on the contour panel...

if you want to create an animation of an iso-q surface for example you also have to create an isosurface,

hope it helps. best regards

Said

Thanks a lot for your message. I have already created that custom field function Q.

So when I create iso-surfaces, do I create iso surfaces of Q ( say specific values of Q) and then plot Q on those Q iso-surfaces? Or do I plot Q on iso-surfaces of low pressure, vorticity-maginitude, etc?

Anindya

Fortunately, you don't have to use UDF. Here's a simpler way of visualizing the iso-contours of the second-invariant.

You can use the custom field funtion capbility in FLUENT. First you define the second-invariant of deformation tensor with

Q = 0.5(W*W - S*S) where W is the vorticity magnitude (you can find it under the "Velocity" menu) and S is the mean rate-of-strain (you can find it under "Derivatives" menu).

Once you defined Q, you can generate iso-surfaces of Q for several positive values.

Thanks a lot for your help.

Anindya

[la7low]

Originally Posted by Helper
;122608

Fortunately, you don't have to use UDF. Here's a simpler way of visualizing the iso-contours of the second-invariant.

You can use the custom field funtion capbility in FLUENT. First you define the second-invariant of deformation tensor with

Q = 0.5(W*W - S*S) where W is the vorticity magnitude (you can find it under the "Velocity" menu) and S is the mean rate-of-strain (you can find it under "Derivatives" menu).

Once you defined Q, you can generate iso-surfaces of Q for several positive values.

Hi Guys,

should not it be the following:
Q=0.25*(W*W - S*S)
???

As the mean rate-of-strain (scalar) is defined as:
S=(2Sij*Sij)^0.5
and similarly the vorticity magnitude (scalar):
W=(2Wij*Wij)^0.5
with Sij being the mean rate of strain tensor and with Wij the mean vorticity tensor (besides, Sij=symmetric, while Wij=antisymmetric part of mean velocity gradient tensor).
Finally the second invariant of velocity grad tensor is defined as:
Q=0.5*(Wij*Wij - Sij*Sij). So I think Q should be: Q=0.25*(W*W - S*S). Is that right?

Besides, this formula of Q is only valid for incompressible flows, more precisely for flows with divergence free velocity field so that divUj=0 <=> Sii=0 !

Thanks!

[Dipanjay]

Hi la7low, I completely agree with you.

Cheers..

[ashar_md2001]

Hi la7low,
Could you please give the reference, from where you have taken this Q criterion.

Thanks

[subha_meter]

Originally Posted by ashar_md2001

Hi la7low,
Could you please give the reference, from where you have taken this Q criterion.

Thanks

Q criterion was proposed by Jeong & Hussain (1995), J.Fluid Mech., vol.285, pp.69-94.

[c.m.]

According to "F.R. Menter, Best Practice: Scale-Resolving Simulations in ANSYS CFD, Version 2.00, November 2015" it is: "[...] for historic reasons 0.5 in ANSYS Fluent and 0.25 in ANSYS CFD-Post". However, I don't know how it is in ANSYS CFX.

[subha_meter]

Originally Posted by subha_meter

Q criterion was proposed by Jeong & Hussain (1995), J.Fluid Mech., vol.285, pp.69-94.

My apology for a correction required in the earlier post. Below are the actual references for the three useful criteria often used for detecting vortex:

1. Q criterion: positive second invariant of velocity gradient tensor

HUNT,J . C. R., WRAY, A.A., & MOIN, P. 1988 Eddies, stream, and convergence zones in turbulent flows. Center for Turbulence Research Report CTR-S88, p. 193.

2. Discriminant (DELTA) criterion: complex eigenvalues of velocity gradient tensor

CHONG, M.S., PERRY, A.E. & CANTWELL, B. J. 1990 A general classification of three-dimensional flow field. Phys. Fluids A 2, 765.

3. Lambda 2 criterion: negative second eigenvalue of the S^2 + W^2 tensor where S = strain rate tensor (symmetric part of velocity gradient tensor) and W = vorticity tensor (antisymmetric part of velocity gradient tensor)

Jeong & Hussain (1995), J.Fluid Mech., vol.285, pp.69-94.

[kazemiakk]

How can we plot Q criterion, Discriminant (DELTA) criterion, and Lambda 2 criterion in a 2D plane. what is an important plot to draw?

[Far]

This reference may be helpful for defining different vortex detection criteria in 2d

http://aip.scitation.org/doi/10.1063/1.4927647

[randolph]

Originally Posted by la7low

Hi Guys,

should not it be the following:
Q=0.25*(W*W - S*S)
???

As the mean rate-of-strain (scalar) is defined as:
S=(2Sij*Sij)^0.5
and similarly the vorticity magnitude (scalar):
W=(2Wij*Wij)^0.5
with Sij being the mean rate of strain tensor and with Wij the mean vorticity tensor (besides, Sij=symmetric, while Wij=antisymmetric part of mean velocity gradient tensor).
Finally the second invariant of velocity grad tensor is defined as:
Q=0.5*(Wij*Wij - Sij*Sij). So I think Q should be: Q=0.25*(W*W - S*S). Is that right?

Besides, this formula of Q is only valid for incompressible flows, more precisely for flows with divergence free velocity field so that divUj=0 <=> Sii=0 !

Thanks!

The constant does not really matter for visualization.

Ps: If I am not wrong, I remember is due to some historical reason to use 0.5 and 0.25 in Fluent