Total number of particle template is not equal to the number of atoms

Submitted by subhodhk on Tue, 06/24/2014 - 18:04

Hi

I am trying simulate a mixing process and I am using two types of atomic particles. After I ran the simulation I found that there is a mismatch between the total number After I ran this simulation my LIGGGHTS log file had these messages:

INFO: Particle insertion ins1: inserted 361 particle templates (mass 0.001588) at step 9999001
- a total of 3372525 particle templates (mass 14.833140) inserted so far.
WARNING: Particle insertion: Less insertions than requested (../fix_insert.cpp:646)
INFO: Particle insertion ins2: inserted 173 particle templates (mass 0.001698) at step 9999001
- a total of 1796625 particle templates (mass 17.638325) inserted so far.
WARNING: Particle insertion: Less insertions than requested (../fix_insert.cpp:646)
10000000 4.2947699e+18 0 0 0.1600096 53.61218
Loop time of 1.06536e+06 on 1 procs for 9999999 steps with 232933 atoms

So:
The number of particle templates inserted for particle type 1= 3372525
The number of particle templates inserted for particle type 2= 1796625
Total number of particle templates inserted= 5169150

But if you look at the last line of the log message it says that there are only 232933 atoms. Which means that about 5 million particle templates were inserted and only about 2 hundred thousand atoms were inserted into the system.

Now this was a very long simulation and it took about 12 days on one processor. On smaller simulations the total number of particle templates do equal the number of atoms (the numbers are very close even if they don't totally match). I don't think that there should be such a huge mismatch between the values in the long simulation.

AttachmentSize
Plain text icon log_liggghts_longrun.txt4.83 MB
Plain text icon log_liggghts_shortrun.txt53.44 KB
Image icon screen_shot_1.png21.17 KB
Image icon screen_shot_2.png71.71 KB
Image icon screen_shot_3.png74.06 KB

Forums:

aaigner's picture

aaigner | Wed, 06/25/2014 - 13:40

Hi!

From your log-files I can not see if it is a bug or just an error in your simulation.
Did you check your simulation with paraview? How looks your imported geometry? Can the Body_mixer.stl overflow or is it a closed body?

Bests
Andreas

subhodhk | Wed, 06/25/2014 - 18:09

Hi Andreas

The Body_mixer.stl has an outlet which is not closed and the simulation runs fine on Paraview. I have attached three screenshots of the simulation taken from Paraview. Please take a look at them.

Thanks
Subhodh

aaigner's picture

aaigner | Thu, 06/26/2014 - 14:39

Hi Subhodh,

that is not a bug. The total number of particles inserted is 5169150 and that is correct. The number of particles in the last line of the output is the current/final number of particles in the simulation domain. Since you have an output, those two numbers are not equal.

Bests
Andreas

aaigner's picture

aaigner | Fri, 06/27/2014 - 15:10

Sorry, I meant outflow not output in the last sentence.

nasato | Thu, 06/26/2014 - 10:03

Hi Subhodh,

This warning message: WARNING: Particle insertion: Less insertions than requested (../fix_insert.cpp:646) tells that LIGGGHTS is not inserting the expected number of particles. This can happen if you are trying to generate a huge amount of particles in a small region so LIGGGHTS try to create the particles but it doesn't find enough physical space in that region - by default LIGGGHTS search for possible overlaps and make a number of attempt to generate the particles. If not possible than the particle is not generated (have a look in fix_insert_stream).
My suggestion is either you increase the insertion region size or decrease the mass flow rate. You could also increase the initial velocity that particles are created.
Cheers,
Nasato

subhodhk | Thu, 06/26/2014 - 19:33

Hi Andreas

The warning is not what I am worried about. My question was basically related to these 3 lines:
1. - a total of 3372525 particle templates (mass 14.833140) inserted so far.
2. - a total of 1796625 particle templates (mass 17.638325) inserted so far.
3. Loop time of 1.06536e+06 on 1 procs for 9999999 steps with 232933 atoms

So statement 1 tells me that: 3372525 particles of type 1 have been successfully inserted into the system.
Statement 2 tells me that: 1796625 particles of type 2 have been successfully inserted into the system. (Am I correct in assuming that they are succesful insertions?)
So why then doesn't the sum of these two numbers (3372525+1796625) match with the number of atoms in the third statement (232933)?

I would also like to add that the above simulation was run for 10 million timesteps. When I run the same simulation for 1 million timesteps I get this message:
4. Loop time of 81832.6 on 1 procs for 999999 steps with 231590 atoms

Now compare statements 3 & 4. Notice that the difference in the number of atoms is very small (about 1300). It can't be true that only 1300 more particles have been inserted in the remaining 9 million time steps!

Thanks
Subhodh

Daniel Queteschiner | Fri, 06/27/2014 - 09:29

As Andreas already stated, your simulation has an inflow and an outflow of particles. The total number of particles inserted over the whole simulation time is 5169150 (3372525+1796625). 232933 is the number of particles present in the simulation in the current (in this case the last) time step. In other words, there was an outflow (loss) of 4936217 (5169150-232933) particles over the whole simulation.

subhodhk | Fri, 06/27/2014 - 22:49

Thanks Daniel. That clears a lot of confusion!
I thought the statement ' Loop time of 1.06536e+06 on 1 procs for 9999999 steps with 232933 atoms' was like a summary. Hence I thought that the number 232933 should be the total number of particles inserted.
I would also like to commend the promptness with with which you, Andreas and Christoph reply to these posts in the forum! It's really helpful to us users!

Regards
Subhodh