Hello everyone,
I am trying to simulate dust particle motion caused by high-speed fluid flow through DEM in a one-way coupled strategy. However, the number of dust particles in a real situation is simply too high to track in a reasonable amount of time while solving for inter-particle interactions. Now I am thinking of shrinking the computational domain to a reasonable size so that dust particle concentration is realistic, and then scale the obtained small-scale results to real flow. Can someone help me recognize some of the parameters that I need to keep the same as real-scale flow if I am to follow this method? Alternatively, if you have any other suggestion for tackling this problem, that would be greatly appreciated too!
Thank you.
If you have access to an HPC (High performance computing) you can model a huge number of particles, and by huge I mean millions, but if you don't you only can model up to 100K, with 20 CPU probably could take days.
You can optimize your model by scaling both the geometry and the particle size. Here is an example Article Numerical modelling of SS316L powder flowability for laser p...
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis