dynamic boundary condition

Issues and examples using this model (available in the Models Repository)

dynamic boundary condition

Postby kingpo on Tue May 09, 2017 6:46 pm

when I use the SPHyscis to simulate dambreak,fluid particles are too far away from the boundary particles (use the dynamic boundary condition + SPS viscosity). Meanwhile fluid particles are rebound and climb at the same time instead of climb along the wall and then bounce back. The .txt file which contains the parameters I used and the result will be attached. Why it just like this? how can i modified it? please help me!

0 Choose Starting options: 0=new, 1=restart, 2=new with CheckPointg, 3=restart with CheckPointing
3 Kernel: 1=gaussian, 2=quadratic; 3=cubic; 5=Wendland
1 Time-stepping algorithm: 1=predictor-corrector, 2=verlet, 3=symplectic, 4=Beeman
1 Density Filter: 0=none, 1=Shepard filter, 2=MLS
30 ndt_FilterPerform (if density filter is used) ?
0 Kernel correction 0=None, 1=Kernel correction, 2=Gradient kernel Correction
3 Viscosity treatment 1=artificial; 2=laminar; 3=laminar + SPS
1.e-6 Viscosity value( if visc.treatment=1 it's alpha, if not kinem. visc approx 1.e-6)
0 Vorticity printing ? (1=yes)
1 Equation of State: 1=Tait's equation, 2=Ideal Gas, 3= Morris
2 Maximum Depth (h_SWL) to calculate B
20 coefficient ( 10 , 40 ) ??
2 Boundary Conditions: 1=Repulsive Force; 2=Dalrymple
10 ndt_DBCPerform ? (1 means no correction)
1 Geometry of the zone: 1=BOX, 2=BEACH, 3=COMPLEX GEOMETRY
2 Initial Fluid Particle Structure: 1= SC, 2= BCC
4.0,1.0,3.0 Box dimension LX,LY,LZ?
0.048,0.048,0.048 Spacing dx,dy,dz?
0 inclination of floor in X ( beta ) ??
0 inclination of floor in Y ( tita ) ??
0,0,0 Periodic Lateral boundaries in X, Y, & Z-Directions ? (1=yes)
0 If wavemaker will be added, left wall is not needed
0 Add wall (1=y)
0 Add wall with slot (1=y)
0 Add wall with round hole (1=y)
0 Add obstacle (1=yes)
0 Add wavemaker
0 Add gate (1=yes)
0 Add Floating Body (1=yes,0=no)
2 Initial conditions: 1) Set of particles without grid. 2) particles on a staggered grid without filling the box
0 Correct pressure at boundaries ?? (1=y)
0.04,1.0 XMin, Xmax ??
0.04,1.0 YMin, Ymax ??
0.04,2.0 ZMin, Zmax ??
0 Fill a new region
1.0,0.01 Input the tmax and out
0. initial time of outputting general data
0.0,1.0,-1.0 For detailed recording during RUN: out_detail, start, end
0.0001,1 input dt ??, variable dt ??
0.1 CFL number (0.1-0.5)
0.75 h=coefficient*sqrt(dx*dx+dy*dy+dz*dz): coefficient ???
0 Use of Riemann Solver: 0=None, 1=Conservative (Vila), 2=NonConservative (Parshikov)
4 Which compiler is desired: 1=gfortran, 2=ifort, 3=win_ifort, 4=Silverfrost FTN95
1 Precision of XYZ Variables: 1=Single, 2=Double
txt.jpg (124.26 KiB) Viewed 7606 times
0.8sXZ.jpg (239.37 KiB) Viewed 7606 times
0.8sXY.jpg (324.3 KiB) Viewed 7606 times
0.5sXZ.jpg (251.32 KiB) Viewed 7606 times
0.5sXY.jpg (287.69 KiB) Viewed 7606 times
Posts: 1
Joined: Mon May 08, 2017 8:30 am

Re: dynamic boundary condition

Postby sebastian.koch on Mon Aug 21, 2017 6:53 am

I have also observed this phenomenon. In my opinion it is caused by the density change which is calculated from the speed (1.16). When the fluid particles streams above the boundary particles, a high density and a high pressure results. I think that can be avoided using repulsive boundaries. In the test examples, the effect is avoided by placing water on the ground.
Posts: 5
Joined: Wed Jan 27, 2016 8:46 am

Return to SPHysics User Discussions

Who is online

Users browsing this forum: No registered users and 1 guest