Simulated Annealing

I’m simulating a group of polypeptides and would like to “anneal” the system.

It’s been mentioned before on the list, to anneal one would:

use “velocity” with a high enough initial temperature
use fix nve
use fix viscous to drain energy slowly

This makes me wonder, if I’m annealing from a fairly unorganized state, then the pressure could change. NVE does not correct for pressure so is there a method to adjust the pressure. My first thought would be to use fix press/berendsen.

Thanks for your help.

I'm simulating a group of polypeptides and would like to "anneal" the system.

It's been mentioned before on the list, to anneal one would:

use "velocity" with a high enough initial temperature
use fix nve
use fix viscous to drain energy slowly

This makes me wonder, if I'm annealing from a fairly unorganized state, then the pressure could change. NVE does not correct for pressure so is there a method to adjust the pressure. My first thought would be to use fix press/berendsen.

If you pick the proper density according to the desired final state,
then allowing for a variable cell will only take your calculation
longer to converge.

Axel

There are 2 barostat options in LAMMPS. Fix npt and fix press/berendsen.
A more typical annealing procedure is to run with a thermostat at
high temp, then quench. You can also barostat if you like, but turn
it off eventually, as Axel suggested.

Steve