[minhbaba]

Hello Everybody,

I am currently dealing with the simulation of particles in closed geometries (Cylindrical Pipes). I used the DEM solver in Star-CCM to simulate the flow of particles in vertical pipes. The flow is in dense phase (packing fraction ~ 60%) and is solely driven by gravity. Since the influence of air on the flow is minimal in dense phase flow, I did not use any FVM solvers and just computed the motion of particles using DEM. Thus, there was no coupling between the interstitial air and the particles.

Now, I need to extend this study for heat transfer applications in which, the particles are heated from the pipe wall. In this case, since the interstitial air plays a vital role, I need to adopt a two-way coupled DEM. But the simulation methodology in StarCCM is giving wrong results. Though the simulation was converging for all the time steps, the energy input into the system is less than the energy rise of the system. After a lot of study, I understood that the problem might lie in the DEM methodology employed in StarCCM. In the two-way coupling, when a particle is absorbing/releasing some energy from/to the air, it is considered as a source term. But this source term is only added to the mesh cell which contains the center of the mesh and not to the cells where a portion of the particle still presents. This might be leading to the energy imbalance. Though I can try to solve this by using a mesh size that is greater than particle diameter, I need to compromise on the flow geometry.

Thus, I am currently exploring other options to simulate this case. So I want suggestions as of which OpenSource code is perfect for my case: LIGGGHTS, CFDEM, OpenFoam etc. I have no experience in LIGGGHTS or CFDEM but many people suggested it.

I need to simulate a Two-way coupled DEM flow enabled with particle-particle conduction, particle-wall conduction, air-wall conduction and particle-air convection.

Please suggest which one is the apt for my simulation case.

Thank you very much!!
Yesaswi