In DPM/DDPM particle motion is tracked in a Lagrangian manner and does not require a dynamic meshing to move the particle physically.

DDPM is an improved version of earlier DPM model which is capable of handling higher volume fraction of solid phase and has an in-built particle-particle collision sub-model through DEM collision component. The particles are not true particle since flow around them are not resolved.

If a particle motion is expected to track by deforming mesh, that can be done in the Eulerian framework itself using "dynamic meshing" component. It works on a six degree of freedom solver and requires an external UDF to define the forces or moments acting on the particle. The particle in this case would be a true particle since it would have a "no-slip" BC on it and flow around it, is resolved.

It is, however, difficult to introduce more than one particle unless the size of the particle is too small since mesh deformation is too much during simulation which might compromise the model output. The smaller size particles on the other hand would lead to significant increase in the mesh density hence computational cost. A special UDF here would be required to prevent particle-particle and particle-wall contact which otherwise would lead to meshing failure and termination of simulation.

Have a look at these papers where dynamic meshing has been utilized to model particle motion in an Eulerian framework.

http://www.sciencedirect.com/science/article/pii/S0921883114002702

http://onlinelibrary.wiley.com/doi/10.1002/aic.15027/full

Hope it helps,

SM

Thanks for such a nice reply. Yes,  Dear friend, you are right,  i think because up to now, i tried so many ways. Presently, i am able to give rotational as well translation motion to geometry but failed to deform the grids. I am still searching solution. 

You might have better luck using the DEM model. |

https://www.youtube.com/watch?v=XG_HFKiv9t8