CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Population Balance Modeling (PBM) - Ansys Fluent (https://www.cfd-online.com/Forums/fluent/103074-population-balance-modeling-pbm-ansys-fluent.html)

masoud.ravan June 17, 2017 12:30

Quote:

Originally Posted by furqanrk (Post 653639)
Have you implemented PBM model with command
> /define/models/addon-module ?
If Yes... There is manual in fluent, very easy and useful !

Good Luck

Thanks for respond,

Actually no, I didn't use PBM yet. I want to learn the concept of this model
I appreciate if anyone knows a reference.

sepehrkhn July 13, 2017 10:47

invalid data on PBM
 
Hi
I am modeling two phase(water-in-oil) multiphase in a tank with Rushton impeller and I face a problem in aggregation and breakage system of my PBM. the problem is that I use discrete method for population balance and I have 7 bins. the bin fractions are not correct. (bin0=.99 and other bins are totally 0.01) I should mention that bin0 is the bin with largest droplet diameter. it seems that breakage doesn't happen in my tank. even I just use breakage without coalescence I saw no change in droplet diameter.
any help will be appreciated:)

BlnPhoenix July 13, 2017 13:51

Quote:

Originally Posted by sepehrkhn (Post 656948)
Hi
I am modeling two phase(water-in-oil) multiphase in a tank with Rushton impeller and I face a problem in aggregation and breakage system of my PBM. the problem is that I use discrete method for population balance and I have 7 bins. the bin fractions are not correct. (bin0=.99 and other bins are totally 0.01) I should mention that bin0 is the bin with largest droplet diameter. it seems that breakage doesn't happen in my tank. even I just use breakage without coalescence I saw no change in droplet diameter.
any help will be appreciated:)

Which Kernels are you using for water-in-oil breakup and coalescence?

sepehrkhn July 13, 2017 14:46

Quote:

Originally Posted by BlnPhoenix (Post 657000)
Which Kernels are you using for water-in-oil breakup and coalescence?

I use Luo model for both with surface tension of 0.02

furqanrk July 13, 2017 23:45

Quote:

Originally Posted by sepehrkhn (Post 656948)
Hi
I am modeling two phase(water-in-oil) multiphase in a tank with Rushton impeller and I face a problem in aggregation and breakage system of my PBM. the problem is that I use discrete method for population balance and I have 7 bins. the bin fractions are not correct. (bin0=.99 and other bins are totally 0.01) I should mention that bin0 is the bin with largest droplet diameter. it seems that breakage doesn't happen in my tank. even I just use breakage without coalescence I saw no change in droplet diameter.
any help will be appreciated:)

Can you try to use 'LOG NORMAL' values to set your bins value from velocity boundary condition window ?

sepehrkhn July 14, 2017 12:03

Quote:

Originally Posted by furqanrk (Post 657044)
Can you try to use 'LOG NORMAL' values to set your bins value from velocity boundary condition window ?

I dont have inlet. I just patch a volume of the tank as a secondary phase ant start calculation after. but after the bin0 fraction which is the largest bin starts increasing to 0.99 .

furqanrk July 15, 2017 09:09

Quote:

Originally Posted by sepehrkhn (Post 656948)
Hi
I am modeling two phase(water-in-oil) multiphase in a tank with Rushton impeller and I face a problem in aggregation and breakage system of my PBM. the problem is that I use discrete method for population balance and I have 7 bins. the bin fractions are not correct. (bin0=.99 and other bins are totally 0.01) I should mention that bin0 is the bin with largest droplet diameter. it seems that breakage doesn't happen in my tank. even I just use breakage without coalescence I saw no change in droplet diameter.
any help will be appreciated:)

How did you set Bins fraction in fluent ? How do you know that one bin should 0.99 and other total is 0.01 to make it equal to 1. Do you have any experimental data ? Or do you have Mean you particle size ? Let assume mean particle size is 0.25 cm and your desire to solve it around 7 bins. so, my suggestion is :

Bins 7
Ratio 1.3
Let asume Min size 0.1 cm ( better to use experimental value )
then set like this:
Fraction
Bin-0 0
Bin-1 0
Bin-2 0
Bin-3 1 ( The mean particle size 0.25cm will the middle one )
Bin-4 0
Bin-5 0
Bin-6 0

Good Luck

sepehrkhn July 15, 2017 09:45

Quote:

Originally Posted by furqanrk (Post 657198)
How did you set Bins fraction in fluent ? How do you know that one bin should 0.99 and other total is 0.01 to make it equal to 1. Do you have any experimental data ? Or do you have Mean you particle size ? Let assume mean particle size is 0.25 cm and your desire to solve it around 7 bins. so, my suggestion is :

Bins 7
Ratio 1.3
Let asume Min size 0.1 cm ( better to use experimental value )
then set like this:
Fraction
Bin-0 0
Bin-1 0
Bin-2 0
Bin-3 1 ( The mean particle size 0.25cm will the middle one )
Bin-4 0
Bin-5 0
Bin-6 0

Good Luck


thank dude, I model experimental data from another guy... the particle size varies between 50micrometer to 280 micrometer so I use this size in population balance and ratio equal to one.
I patch the volume fraction and I patch bin fractions as follows:
bin0,1,3,4,5,6=0.1
bin2=0.4 because the sauter mean diameter is in this range.

Mojtaba_ August 7, 2017 08:12

please help me
 
hello everyone
do any one explain the formation of crystal step by step?
secondary phase is my crystal phase and i put the volume fraction zero in boundary condition.
i dont have any computation for secondary phase so this mean that crystal phase is not formatting.
why crystal phase dont form??
could you please help me?

masoud.ravan October 26, 2017 08:05

using file for size distribution in PBM model
 
Hey guys,
Do you know how we could specify size distribution for PBM model by using a file in ANSYS FLUENT?
Thanks in Advance,

sina_mech October 26, 2017 11:37

Quote:

Originally Posted by masoud.ravan (Post 669271)
Hey guys,
Do you know how we could specify size distribution for PBM model by using a file in ANSYS FLUENT?
Thanks in Advance,

Population Balance Model > Definition > File > Load File

You need to assign the number of bins before loading the file. The diameters should be ascending, and only one diameter per line! This is the content of a sample file that I used while ago:
Code:

0.0000015
0.000005
0.000029
0.000075
0.000175
0.000375
0.000750

Having the diameters, you can initialize volume fraction of each at inlet/domain.

masoud.ravan December 5, 2017 09:59

Changing discrete method to inhomogeneous method in PBM model
 
Hi everyone
I'm simulating a rectangular duct with velocity inlet and pressure outlet in order to see how water and air (that had been injected from an inlet with each other) separate from each other. I have used the discrete method with 8 bin fractions for solving the population balance equation.
Now I wanna change my method to inhomogeneous and continue the simulation. After adding another phase and defining bin fractions in the third phase, I encounter an error:
================================================== ============================

Node 0: Process 568: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 1: Process 4360: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 2: Process 1952: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 3: Process 3732: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 4: Process 4904: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 5: Process 1448: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 6: Process 3892: Received signal SIGSEGV.

================================================== ============================
MPI Application rank 0 exited before MPI_Finalize() with status 2

The fl process could not be started.

I would be absolutely grateful for your responses...

masoud.ravan December 8, 2017 10:49

Quote:

Originally Posted by masoud.ravan (Post 674003)
Hi everyone
I'm simulating a rectangular duct with velocity inlet and pressure outlet in order to see how water and air (that had been injected from an inlet with each other) separate from each other. I have used the discrete method with 8 bin fractions for solving the population balance equation.
Now I wanna change my method to inhomogeneous and continue the simulation. After adding another phase and defining bin fractions in the third phase, I encounter an error:
================================================== ============================

Node 0: Process 568: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 1: Process 4360: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 2: Process 1952: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 3: Process 3732: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 4: Process 4904: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 5: Process 1448: Received signal SIGSEGV.

================================================== ============================

================================================== ============================

Node 6: Process 3892: Received signal SIGSEGV.

================================================== ============================
MPI Application rank 0 exited before MPI_Finalize() with status 2

The fl process could not be started.

I would be absolutely grateful for your responses...


And I also encountered another problem with the Inhomogeneous method of PBM in AnsysFLUENT.
Do we need special treatment other than enabling aggregation phenomena to make bubbles in a smaller velocity group to go to a bigger velocity group??
I cannot see a transition between velocity groups in my simulations. For instance, when I inject a swarm of bubbles, their diameter would increase till reaching the maximum diameter of their own velocity group and do not go to a bigger velocity group although I have some bigger velocity groups in my problem.
Does anyone have some advice...?
Thanks in advance

masoud.ravan December 16, 2017 13:50

1 Attachment(s)
Hello All!
I am using the Inhomogeneous discrete method in AnsysFluent PBM model. I have two phases and I want to define just one bin for each phase...after setting this condition, phrase "-1.#IND" will be showed in Ratio Exponent and Max Diameter box:confused: (as shown in the picture)
Do you know what it means...and is it an error??

Thanks in Advance...:)

sam_cfdd January 19, 2018 00:07

Quote:

Originally Posted by chittipo (Post 365684)
Dear Members,

If you have any specific questions on PBM implementation please post here. I will try to answer them, as I gained some experience in it.

on
- Existing module (or)
- Implementation through UDF

Hi Dear,

Actually I want to find aggregation kernel using experimental data to link with FLUENT but I am not sure how to do this. Can you please help me in this case. I appreciate your assist. Regards.

sam_cfdd January 19, 2018 00:11

Quote:

Originally Posted by masoud.ravan (Post 653688)
Thanks for respond,

Actually no, I didn't use PBM yet. I want to learn the concept of this model
I appreciate if anyone knows a reference.

Hi,

PBM is actually Bullshit and headache but it used alot in industry. If you want I have pdf books and ANSYS documents I can share with you. email me on e.zhaleh@gmail.com.

masoud.ravan January 19, 2018 00:29

Quote:

Originally Posted by sam_cfdd (Post 678639)
Hi Dear,

Actually I want to find aggregation kernel using experimental data to link with FLUENT but I am not sure how to do this. Can you please help me in this case. I appreciate your assist. Regards.

Hi
Do you mean that you wanna develeop a kernel? Like lou and other researchers?
Please explain more...

sam_cfdd January 19, 2018 01:20

Quote:

Originally Posted by masoud.ravan (Post 678643)
Hi
Do you mean that you wanna develeop a kernel? Like lou and other researchers?
Please explain more...

Yes, exactly. I wanna develop my kernel. I read lots of article such as luo and others but I couldn't undrestand what did they do.

masoud.ravan January 19, 2018 01:52

Quote:

Originally Posted by sam_cfdd (Post 678645)
Yes, exactly. I wanna develop my kernel. I read lots of article such as Luo and others but I couldn't understand what did they do.

They either performed an experimental investigation or used other's experimental data to validate their kernels. If you want to develop a kernel for coalescence or break-up, you have to perform or choose an air-water experiment like bubble injection from recent works and simulate it, and the last step is evaluation the agreement between the numerical and experimental data.

Good Luck,
Masoud Ravan

seinfeld2 January 21, 2018 21:00

Hi all,

I'm trying to simulate CO2 bubble cavitation in water with dissolved CO2 flowing in a venturi tube. Is using this model in the right direction? If anyone can guide me through just a starting point with using this model, that'd be great. What should the From and To Phase be under mass transfer? Is using species appropriate?

Regards.


All times are GMT -4. The time now is 23:46.