I’m trying to recreate a multidispersed bubbling fluidized bed, and have created a pseudo 2D DEM simulation to limit the number of particles and get reasonable solution times. (The reason I say pseudo is that the geometry is still 3D, but very thin, as I believe I read that this was the proper way to do it.) I have a CDF curve for the particle size distribution. I am also monitoring the mean particle diameter on the top outlet, and these results seem reasonable.
However, every now and then the fluidized bed seems to not work as intended, and I don’t see exactly why. Please see the image below. The top row illustrates the problem, and what I expect is the bottom row. Is there a known reason to why the particles down like this? Could the problem be related to the mesh density or the initial particle grid? These two cases are almost identical, and the problem have occured in the bottom case also.
