search

ISISNeutronMuon / MDANSE

MDANSE: Molecular Dynamics Analysis for Neutron Scattering Experiments
https://www.isis.stfc.ac.uk/Pages/MDANSEproject.aspx
GNU General Public License v3.0
18 stars 4 forks source link

Add cumulative time-averaged results to thermodynamic jobs #477

Closed ChiCheng45 closed 1 week ago

ChiCheng45 commented 1 week ago

Description of work When we calculate the temperature and density they are the instantaneous temperatures and densities. To get the actual temperatures and densities we should time-average those properties, I have added a cumulative time-average results to the temperature and density jobs.

I think this should help with reading the temperature and density results since they fluctuate quite a lot. We can see in the below trajectory hasn't equilibrated until at least 2.5 ps.

image

image

To test Run density (use NPT MD trajectory) and temperature jobs. Open results, cumulative average thermodynamic results be available.

MBartkowiakSTFC commented 1 week ago

Works great for temperature, but I got something unexpected with density.

avg_atomic_density avg_atomic_density

The trajectory I used was not equilibrated, with a steep temperature ramp. (It was the molybdenum from the tutorial.) Should I expect it not to work correctly in such a case, or is there still a chance to make the method more general?

ChiCheng45 commented 1 week ago

This is expected. Its a cumulative average so that the average at, for example, 70 ps will be the average over all previous time steps.