Hi Friends,
I am a beginner with LBM. I generated a code in D2Q9 my code is working fine for channel flow problems and I am getting good results for drag coefficients. But when I use the same code for square lid driven cavity in 2-D with north wall moving my results diverge.
Re=1000
Uw=0.1lu/ts
grid = (100,100)
My code for moving boundary is as given below:
INTEGER Xmax,Ymax
REAL*8 Uw,density,f(0:8,1001,1001),ftemp(0:8,1001,1001)
c
c local variables
INTEGER x,y,i
REAL*8 Dxy
c Bounceback at Top boundary
y = Ymax
DO x = 2, Xmax-1
Dxy = (ftemp(0,x,y)+ftemp(1,x,y)+ftemp(3,x,y)+
& 2*(ftemp(2,x,y)+ftemp(6,x,y)+ftemp(5,x,y)))
Dxy=Dxy*Uw
ftemp(4,x,y)=ftemp(2,x,y)
ftemp(7,x,y)=ftemp(5,x,y) -Dxy/2. +
& .5*(ftemp(1,x,y)-ftemp(3,x,y))
ftemp(8,x,y)=ftemp(6,x,y) +Dxy/2. +
& .5*(ftemp(3,x,y)-ftemp(1,x,y))
END DO
END
The other bounceback conditions that I am applying on the other walls is given below:
DO x = 1, Xmax
DO y = 1, Ymax
IF (wall(x,y)) THEN
f(1,x,y) = ftemp(3,x,y)
f(2,x,y) = ftemp(4,x,y)
f(3,x,y) = ftemp(1,x,y)
f(4,x,y) = ftemp(2,x,y)
f(5,x,y) = ftemp(7,x,y)
f(6,x,y) = ftemp(8,x,y)
f(7,x,y) = ftemp(5,x,y)
f(8,x,y) = ftemp(6,x,y)
c
END IF
END DO
END DO
I am using full bounceback conditions.
After some time in lid driven cavity the code diverges and gives very large negative values of densities. I am puzzled why it is happening. Kindly help me to sort it out. I am going through various validation processes. I would be grateful to you all.
Regards
Prateek