[dani2702]

Hi,

I'm trying to simulate a fludized bed with two inlets for particles (inletParticlesLeft and inletParticlesRight), one inlet for air (inletAir) and one outlet (outletAir). See the attached image #1 for the geometry.

geometry.png


Gas is flowing from inletAir and particles (alpha_Particles = 0.62) are flowing from inletParticlesRight into the geometry.

I'm trying to code the following condition:
If particles are carried out of the geometry (outletAir), then the same amount of particles should be recirculated into the geometry via inletParticlesLeft. So kind of a coupling between outletAir and inletParticlesLeft.

For testing purposes I tried to use groovyBC to implement the scenario:
- If there are no particles at outletAir the gas flow and the volume fraction of particles at inletParticlesLeft shall be 0 and therefore no flow
- If there are particles above a certain value (alpha.particles > 1E-3) at outletAir, the volume fraction of particles at inletParticlesLeft should jump to 0.62 and the velocity should increase to a fixed amount.

Here is my code of the boundary conditions:

alpha.air:

    inletAir
    {
        type            fixedValue;
        value           uniform 1;
    }




    outletAir
    {
		type 		zeroGradient;

    }

    inletParticles
    {
        type            fixedValue;
        value           uniform 0.38;
    }

    inletParticles2
    {
 

	type groovyBC;
	aliases		{
			alphaP	alpha.particles;
				}

	variables "alphaPMean{outletAir}=sum(alphaP*mag(Sf()))/sum(mag(Sf()));";



	valueExpression "(alphaPMean > 1e-3) ? 0.38 : 1";
	value	      	uniform 1;

alpha.particles:

    inletAir
    {
        type            fixedValue;
        value           uniform 0;
    }




    outletAir
    {
		type 		zeroGradient;

    }

    inletParticles
    {
        type            fixedValue;
        value           uniform 0.62;
    }

    inletParticles2
    {

	type groovyBC;
	aliases		{
			alphaP	alpha.particles;
				}

	variables "alphaPMean{outletAir}=sum(alphaP*mag(Sf()))/sum(mag(Sf()));";



	valueExpression "(alphaPMean > 1e-3) ? 0.62 : 0";
	value	      	uniform 0;  

	
    }

U.air

    inletAir
    {
        type               fixedValue;
        value	      	uniform (0 0 5);
    }


    outletAir
    {
		type 			inletOutlet;
		inletValue		uniform (0 0 0);
		value 			uniform (0 0 0);

   }    


    inletParticles
    {
        type                fixedValue;
        value 				uniform (-0.3 0 -0.3);
    }	

    inletParticles2
    {
	type groovyBC;
	aliases		{
			alphaP	alpha.particles;
				}

	variables "alphaPMean{outletAir}=sum(alphaP*mag(Sf()))/sum(mag(Sf()));";



	valueExpression "(alphaPMean > 1e-3) ? vector(0.3,0,-0.3) : vector(0,0,0)";
	value	      	uniform (0 0 0);  
}

U.particles

    inletAir
    {
        type               fixedValue;
        value	      	uniform (0 0 5);
    }


    outletAir
    {
		type 			inletOutlet;
		inletValue		uniform (0 0 0);
		value 			uniform (0 0 0);

   }    


    inletParticles
    {

        type                fixedValue;
        value 				uniform (-0.3 0 -0.3);
		
    }

    inletParticles2
    {
		
	type groovyBC;
	aliases		{
			alphaP	alpha.particles;
				}

	variables "alphaPMean{outletAir}=sum(alphaP*mag(Sf()))/sum(mag(Sf()));";



	valueExpression "(alphaPMean > 1e-3) ? vector(0.3,0,-0.3) : vector(0,0,0)";
	value	      	uniform (0 0 0);  
}

p

    inletAir
    {
		type		zeroGradient;
    } 
    outletAir
    {
		type		fixedValue;
		value		uniform 2.25e5;

	}
    inletParticles
    {
		type		zeroGradient;
    }


    inletParticles2
    {
		type		zeroGradient;
		}

p_rgh

    inletAir
    {
		type		fixedFluxPressure;
		value		$internalField; 
    } 
    outletAir
    {
        type               prghPressure;
	   p				$internalField;
        value              $internalField;
    }    




    inletParticles
    {
		type		fixedFluxPressure;
		value		$internalField; 
    }

    inletParticles2
    {
		type		fixedFluxPressure;
		value		$internalField; 
    }

Now the problem I'm getting, as soon as alpha.Particles art outletAir hits the defined critical value in order to switch alpha.particles and U for the boundary inletParticlesLeft, the time step delta T decreases heavily (see attached image #2).

deltaT.PNG


When looking at the last time step for the boundary inletParticlesLeft (when alpha.particle and U already changed for inletParticlesLeft), the alpha value and velocity was correct, but the pressure p and p_rgh were 'abnormal', see attached images #3 and #4.

p.png
p_rgh.png

Therefore the velocity and pressures right after the boundary condition inletParticlesLeft rise in the domain (example for U.air):

Uair.png

That's why the time step decreases (as the velocity rises in such a magnitude)

I don't know why the pressure differs when the swtiching conditions at the groovyBC applies.

I also tried a simulation without groovyBC and coded the fixedValues for U and alpha for inletParticlesLeft. I manually interrupted the simulation and changed for inletParticlesLeft the alpha and U values at the last time step and continued the simulation with no problem.

Do you guys have any solution for my problem?

Thank you!