smooth particle hydrodynamics (SPH) vs. Finite Volume (FV) approach pros and cons

asalahud · April 19, 2024, 14:48

Can experts comment on the pros and cons on these topics:

1. single phase flow simulation
2. multiphase flow simulation
3. Thermal simulation with energy equation
4. Free surface modeling
5. Computational accuracy
6. Computational time

Examples of fields of applications may be automotive hvac, coastal wave, fluid-structure coupling. Or any other field which you have experience on.

Much appreciated.

inigo999 · December 4, 2024, 08:04

Hello I am not an expert by any means.

But the SPH method can capture free surface violent flows like the ones in coastal engineering at a fraction of the cost.

The problem with SPH is the boundary no slip wall velocity can be challenging and mesh refinement is not really as easy to implement . Until now DualSPHysics did not enable the user to selictively refine through particle size.

But it is clear that in general it is a less mature technology. The development behind FVM CFD is easier to understand that the WCSPH method which dont follow as naturally as the FVM method.

Keep in mind that most CFD workstation are CPU based and GPU acceleration is the only way to make SPH practical.

April 19, 2024, 14:48	smooth particle hydrodynamics (SPH) vs. Finite Volume (FV) approach pros and cons	#1
asalahud New Member anonymous Join Date: Feb 2016 Posts: 3 Rep Power: 10	Can experts comment on the pros and cons on these topics: 1. single phase flow simulation 2. multiphase flow simulation 3. Thermal simulation with energy equation 4. Free surface modeling 5. Computational accuracy 6. Computational time Examples of fields of applications may be automotive hvac, coastal wave, fluid-structure coupling. Or any other field which you have experience on. Much appreciated.

December 4, 2024, 08:04		#2
inigo999 New Member Inigo Zarate Garcia Join Date: Oct 2024 Posts: 1 Rep Power: 0	Hello I am not an expert by any means. But the SPH method can capture free surface violent flows like the ones in coastal engineering at a fraction of the cost. The problem with SPH is the boundary no slip wall velocity can be challenging and mesh refinement is not really as easy to implement . Until now DualSPHysics did not enable the user to selictively refine through particle size. But it is clear that in general it is a less mature technology. The development behind FVM CFD is easier to understand that the WCSPH method which dont follow as naturally as the FVM method. Keep in mind that most CFD workstation are CPU based and GPU acceleration is the only way to make SPH practical.