[CeesH]

Hello all,

I've seen quite some papers using particle tracking, and quite often people simply state `we used 100.000 particles and this was sufficient' or something likewise, without much backup.

My question is, are there any guidelines on how many particles to use (without going to the brute force approach of simply trying different numbers until your solution is invariant) when particles interact with the eulerian field (by mass, momentum and/or heat exchange). Does anyone have ideas or good references for this?

Best,
Cees

[edoan]

One reason for using stochastic tracking is to facilitate convergence. I would set your number of tries to a value that gives a good balance between convergence and computational time. I think there is no general rule for number of particles because it depends on the case.

[sarp]

there is no general rule but you should select sufficient number of particles in order to get averaged result stochastic tracking.
Also you should consider total particle mass to check fraction of dispersed phase<%10 for dilute phase.

[fishball]

Originally Posted by CeesH

Hello all,

I've seen quite some papers using particle tracking, and quite often people simply state `we used 100.000 particles and this was sufficient' or something likewise, without much backup.

My question is, are there any guidelines on how many particles to use (without going to the brute force approach of simply trying different numbers until your solution is invariant) when particles interact with the eulerian field (by mass, momentum and/or heat exchange). Does anyone have ideas or good references for this?

Best,
Cees

number of particles should depend on the pollutant you want to simulate