[amind]

Hello everyone

I am simulating a 2D vertical axis wind turbine in ANSYS Fluent software. A schematic of this turbine can be seen in Figure1. Sliding Mesh is used to rotate the rotor. As we know, when the rotor rotates, flow separation occurs on the blades. One way to control the flow separation on the blades is to use plasma actuators.

Figure1:




These actuators ionize some of the air on the blade and then the electric field applies a force to the ionized air, which increases the momentum in the boundary layer and controls the flow separation. A schematic of the plasma actuator can be seen in Figure2.

Figure2:




Now I want to write a UDF and define a triangular area on each of the airfoils using the DEFINE_SOURCE Macro according to the Shyy model. Then apply body force inside these triangular areas. But the problem is that when the rotor rotates, the triangular region defined in the UDF remains stationary and does not rotate with the rotor; and I do not know what to do to rotate this triangular area with the blades.

Does anyone have an idea to solve this problem?
Or does anyone know a macro other than DEFINE_SOURCE that is useful for applying body force to wind turbine blades?

Thanks in advance




UDF:
#include "udf.h"
DEFINE_SOURCE(plasma,c,t,dS,eqn)
{
real pos[ND_ND];
real source, x, y;
real m, b, Fx, Fy;
C_CENTROID(pos,c,t);
x=pos[0];
y=pos[1];
F=pow(Fx^2+Fy^2,0.5)

/* apply source term to triangular region */
/* y=mx+b is the lineAB equation */

if ((y>=0)&&(y<=(mx+b))&&(x>=0)&&(x<=((y-b)/m)))
{
source=F
dS=0
}
else
{
source=0
dS=0
}
C_UDMI(c,t,0)=source;
returne source;
}


Figure1.png

Figure2.png

[AlexanderZ]

instead of sliding mesh approach you may use moving wall option in momentum tab of wall (airfoil) boundary condition.

in that case you don't need to warry about mesh motion.

[amind]

Unfortunately, the problem is that the UDF does not rotate with the airfoils.

[amind]

When the blades rotate, the body force defined in the UDF must also rotate with the blades. But the UDF stays in the triangular region and does not rotate.

[AlexanderZ]

are you applying this UDF momentum to fluid region?

should it influence rotation speed somehow during simulation?

[amind]

Thanks for the reply.
Yes, this UDF body force is applied to the separated flow on the airfoils. But, this UDF has no effect on rotor rotational speed and wind speed, because these two speeds are constant:
wind speed = 5m/s
rotational speed = 40rad/s

In the following figures, the wind turbine mesh can be seen:









As can be seen, the blades can rotate without any problems. To control the flow separation on the blades, I wrote a UDF in which I first defined a triangular area on one of the airfoils:

((y>=0)&&(y<=(mx+b))&&(x>=0)&&(x<=((y-b)/m)))

Then I applied a force in this triangular area. But when the blades rotate, the triangular area remains stationary.

[amind]

The blades rotate, but the triangular force (defined in the UDF) remains stationary.

[AlexanderZ]

well, I don't see pictures. but that problem comes from my side I think.

as you are saying that rotational speed is constant and = 40rad/s
you don't need to use sliding mesh (which I've mention before)

on the other hand use other approach:
moving wall option in momentum tab of wall (airfoil) boundary condition.

so walls will move in domain, but you will not really see that, however the flow field will be same as in sliding mesh case.

using this approach mesh is same, so you don't need to put any changes in UDF.

If you insist to use sliding mesh, you should modify your UDF:
rotate if condition:
if ((y>=0)&&(y<=(mx+b))&&(x>=0)&&(x<=((y-b)/m)))

with your mesh. it means you need to recalculate all coordinates.

BTW. m and b are not defined in your UDF. So it will not gonna work even without moving. Fix it