[ghost82]

Hi all my friends,
I'm trying to simulate a multiphase problem: a liquid-solid spouted bed; spouted beds are particular type of fluidized bed, in which a fluid is injected from the bottom, centrally into the solid bed: a central spout is generated, the particles flow centrally from the bottom to the top, they create a fountain and they fall into the peripheral "annulus", so they flow from the top to the bottom and they circulate again to the top.

You can see the model in the attached pictures: my problem (2D, axisymmetric, double precision) is unsteady (time step 0,0001 s), eulerian (with three phase: water, granular solid (d=2,8 mm) and air) with gravity enabled in -x direction, explicit solver with courant number set to 0,2 and volume cut off set to 0,0001; solution methods first order upwind, Phase Coupled SIMPLE scheme, gradient Least Squares Cell Based.
Turbulence model is k-epsilon.
At t=0 I patch the inlet tube with only water, the bed with 0,25 volume fraction of solid and 0,75 with water, the freeboard, above the bed, is patched with volume fraction of air equal to 1; turbulent kinetic energy and dissipation rate are initialize to 0, as my velocity at t=0 is 0 m/s.
Boundary conditions are velocity inlet (with a udf to linearly increase velocity from 0 m/s at t=0 to 55 m/s at t=0,1) and pressure outlet (at the top of the bed, P=101325 Pa).
I set also the inlet hydraulic diameter and the turbulence intensity (5%) and the backflow hydraulic diameter and turbulence intensity (0,1%).
Pressure operating condition is set to 101325 at the top of the bed, centrally, on the outlet boundary.
I set the operating density to that of air (1,22 kg/m3) and run my simulation.
No slip velocity at wall for water and air and specularity coefficient of 0,05 for the solid particles, restitution coefficient of 0,9.
Residuals drop below 10^-4 / 10^-5, Global courant number is below 2 after each iteration.
However, it seems that my mixture pressure profile inside my domain has no sense, as you can see from the picture.

I attach the solid volume fraction and the absolute pressure contours.

Please help me understand why I get this unphysical pressure profile..

Thank you to all for reading,

Daniele

[mohsen0488]

Hi Daniele

I'm trying to simulate gas-solid spouted bed, but my results are different with results of other papers.my model is same with paper but i don't know whats the problem.validation reference is results of He et al .
geometry = 2d axisymmetric
Dp=0.00141m
rho_P=2503 kg/m3
drag model= gidaspow with switch function
granular viscosity= gidaspow
solid pressure = lun et al
radial distribution function= lun et al
initial packing = 0.325 m
initial volume fraction of solid = 0.588
maximum volume fraction of solid = 0.59
time steps size =0.0001 s
inlet velocity of gas = 41.472 m/s
inlet B.C = velocity inlet
outlet B.C = outflow
turbulent model = k-epsilon - dispersed
but in my simulation isn't any spouting my result and result of paper attached.
i don't know whats the problem.please help me .
thanks.

[ghost82]

Hi Mohsen,
as you know spouted beds have a characteristic curve flowrate-bed pressure drop; if you start injecting the gas at time 0 in a packed bed you will have a maximum flowrate at a maximum pressure drop; after this point if you increase a bit the flowrate pressure drop will drop and you will have the spout.
From the spout regime if you decrease the flowrate pressure drop will be constant since there is the spout, then decreasing the flowrate, you will have an increase in pressure drop because the spout ceases to exist.
However, when you decrease the flowrate in a spouting bed, the minimum flowrate to have the spout isn't equal to that if you start from the packed bed.
Can you verify this in your simulation?are you sure that in the article the picture is taken from starting at packed bed condition and not from a spouting regime by decreasing the flowrate?

Daniele

[mohsen0488]

Hi Daniele
thanks for your reply.
in article flowrate inlet is constant and there is no increase or decrease in the flow rate.

[ghost82]

But I think this is a transient simulation, am I right?
Perhaps that flow rate could be that equivalent to the minimum spouting velocity.
Maybe the authors of the article increased that flowrate at the beginning of the simulation and once spouting happened they decreased to that equivalent to the minimum spouting velocity.

Daniele

[mohsen0488]

yes simulation is transient.
and velocity equal 1.2*minimum spouting velocity.
But, during the simulation, the velocity has been constant.
and in article written The computation needs 8 s real time for the spouted bed to reach at its steady-state.

[ghost82]

Then, I don't know how to help...recheck your parameter

Daniele

[mohsen0488]

thanks dear daniele.