NETL III Fluidization Curve

blerli_91 · August 14, 2018, 04:55

Hello People,

rigth now i try to simulate the NETL-Challenge III (small fluidized beds with particle distributions). The task is to obtain the fluidization curve and compare it with experimental data.
The thing is now, I am getting quite reasonable results for the pressure drop above the minimal fluidization velocity. It works quite fine with DDPM-KTGF and Euler Granular Model and also with different drag models and approaches for the particle distribution.

However when I try to simulate the section below the minimum fluidization velocity my results are very, very bad. My pressure dop is far to high, also I am not able to recreate the tendecy of the pressure drop in that section.
I already tried a lot different models (bulit in Models like DDPM, Euler Granular, Euler Granular Packed bed), drag models (Wen Yu, Gidaspow, Syamlal Obrien), Granular Viscosity (Gidaspow and Syamlal Obrien) and particle distrbution approaches (Constant Sauter mean diameter, Full distribution with Rosin Rammler) as well as models and values for frictional viscosity and coefficient of restitution. Also tried a bit varying the initial volume fraction in the bed.
But still I am getting somehow around the same wrong results for the calculation.
Maybe someone has an idea or already experience about that challenge? I am quite out of ideas right now... also I am still quite a beginner with Fluent and Multiphase

PS.: Also the task is to try to simulate the pressure drop with the Fluent built in Models... Found some Papers which are using UDF´s for the Drag Models...

Thank you

Micael · August 17, 2018, 17:39

Forget DDPM-KTGF for fixed bed, it's not valid.

For Euler-Euler (Two-Fluid-Model) it can work in fixed bed regime. Take care though of the frictional viscosity. Have you tried Shaeffer for Frictional viscosity in combinaison with Johnson-et-al Frictional pressure?

blerli_91 · August 20, 2018, 05:22

Hi Micael,

yeah i already gave up with KTGF for it.

I also managed to get some better results for it with Euler Granular Packed Bed options (just in case someone else has troubles with it.: it is necessary to fix all the velocities of the solids inside the cell conditions dialog - for ALL zones not just in the packed bed zone, then it should work)

my results are still not that good but at least i got the linear dependency of pressure drop by gas velocity. so i will also try your suggested combination of schaeffer and johnson - thank you for that

(i just tried schaeffer without johnson). Also the results are getting a bit better with more detailled resolution of the particle distribution. But I did it just by adding additional phases...and the computational effort is increasing rapidly of course for that variant.

Is there maybe an "easy" way to create particle distribution for one solid phase for Euler Granular (I guess it will be possible with UDF´s - just don´t have any experience with UDF´s so I cant estimate how much effort it will be to implement it ...)

August 14, 2018, 04:55	NETL III Fluidization Curve	#1
blerli_91 New Member Mario B. Join Date: Feb 2018 Posts: 11 Rep Power: 8	Hello People, rigth now i try to simulate the NETL-Challenge III (small fluidized beds with particle distributions). The task is to obtain the fluidization curve and compare it with experimental data. The thing is now, I am getting quite reasonable results for the pressure drop above the minimal fluidization velocity. It works quite fine with DDPM-KTGF and Euler Granular Model and also with different drag models and approaches for the particle distribution. However when I try to simulate the section below the minimum fluidization velocity my results are very, very bad. My pressure dop is far to high, also I am not able to recreate the tendecy of the pressure drop in that section. I already tried a lot different models (bulit in Models like DDPM, Euler Granular, Euler Granular Packed bed), drag models (Wen Yu, Gidaspow, Syamlal Obrien), Granular Viscosity (Gidaspow and Syamlal Obrien) and particle distrbution approaches (Constant Sauter mean diameter, Full distribution with Rosin Rammler) as well as models and values for frictional viscosity and coefficient of restitution. Also tried a bit varying the initial volume fraction in the bed. But still I am getting somehow around the same wrong results for the calculation. Maybe someone has an idea or already experience about that challenge? I am quite out of ideas right now... also I am still quite a beginner with Fluent and Multiphase PS.: Also the task is to try to simulate the pressure drop with the Fluent built in Models... Found some Papers which are using UDF´s for the Drag Models... Thank you

August 17, 2018, 17:39		#2
Micael Senior Member Micael Join Date: Mar 2009 Location: Canada Posts: 157 Rep Power: 18	Forget DDPM-KTGF for fixed bed, it's not valid. For Euler-Euler (Two-Fluid-Model) it can work in fixed bed regime. Take care though of the frictional viscosity. Have you tried Shaeffer for Frictional viscosity in combinaison with Johnson-et-al Frictional pressure? blerli_91 likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[ICEM] Problems with coedge curves and surfaces	tommymoose	ANSYS Meshing & Geometry	6	December 1, 2020 12:12
[Gmsh] Vertex numbering is dense	KateEisenhower	OpenFOAM Meshing & Mesh Conversion	7	August 3, 2015 11:49
meshing of a compound volume in GMSH	shawn3531	OpenFOAM	4	March 12, 2015 11:45
block-structured mesh for t-junction	Robert@cfd	ANSYS Meshing & Geometry	20	November 11, 2011 05:59
CFX4.3 -build analysis form	Chie Min	CFX	5	July 13, 2001 00:19

August 20, 2018, 05:22		#3
blerli_91 New Member Mario B. Join Date: Feb 2018 Posts: 11 Rep Power: 8	Hi Micael, yeah i already gave up with KTGF for it. I also managed to get some better results for it with Euler Granular Packed Bed options (just in case someone else has troubles with it.: it is necessary to fix all the velocities of the solids inside the cell conditions dialog - for ALL zones not just in the packed bed zone, then it should work) my results are still not that good but at least i got the linear dependency of pressure drop by gas velocity. so i will also try your suggested combination of schaeffer and johnson - thank you for that (i just tried schaeffer without johnson). Also the results are getting a bit better with more detailled resolution of the particle distribution. But I did it just by adding additional phases...and the computational effort is increasing rapidly of course for that variant. Is there maybe an "easy" way to create particle distribution for one solid phase for Euler Granular (I guess it will be possible with UDF´s - just don´t have any experience with UDF´s so I cant estimate how much effort it will be to implement it ...)