Liquid Si - MLFF: Difference between revisions

From VASP Wiki
VisualWikitext
(Created page with "{{Template:Machine learning force field - Tutorial}}")
 
No edit summary
Line 1: Line 1:
{{Template:Machine learning force field - Tutorial}}
{{Template:Machine learning force field - Tutorial}}
== Task ==
Generating a machine learning force field for liquid Si.
== Input ==
=== {{TAG|POSCAR}} ===
*In this example we start from a 64 atom super cell of diamond-fcc Si (the same as in this example: {{TAG|Liquid Si - Standard MD}}:
Si cubic diamond 2x2x2 super cell of conventional cell
      5.43090000000000
    2.00000000  0.00000000  0.00000000
    0.00000000  2.00000000  0.00000000
    0.00000000  0.00000000  2.00000000
    Si
    64
Direct
    0.00000000  0.00000000  0.00000000
    0.50000000  0.00000000  0.00000000
    0.00000000  0.50000000  0.00000000
    0.50000000  0.50000000  0.00000000
    0.00000000  0.00000000  0.50000000
    0.50000000  0.00000000  0.50000000
    0.00000000  0.50000000  0.50000000
    0.50000000  0.50000000  0.50000000
    0.37500000  0.12500000  0.37500000
    0.87500000  0.12500000  0.37500000
    0.37500000  0.62500000  0.37500000
    0.87500000  0.62500000  0.37500000
    0.37500000  0.12500000  0.87500000
    0.87500000  0.12500000  0.87500000
    0.37500000  0.62500000  0.87500000
    0.87500000  0.62500000  0.87500000
    0.00000000  0.25000000  0.25000000
    0.50000000  0.25000000  0.25000000
    0.00000000  0.75000000  0.25000000
    0.50000000  0.75000000  0.25000000
    0.00000000  0.25000000  0.75000000
    0.50000000  0.25000000  0.75000000
    0.00000000  0.75000000  0.75000000
    0.50000000  0.75000000  0.75000000
    0.37500000  0.37500000  0.12500000
    0.87500000  0.37500000  0.12500000
    0.37500000  0.87500000  0.12500000
    0.87500000  0.87500000  0.12500000
    0.37500000  0.37500000  0.62500000
    0.87500000  0.37500000  0.62500000
    0.37500000  0.87500000  0.62500000
    0.87500000  0.87500000  0.62500000
    0.25000000  0.00000000  0.25000000
    0.75000000  0.00000000  0.25000000
    0.25000000  0.50000000  0.25000000
    0.75000000  0.50000000  0.25000000
    0.25000000  0.00000000  0.75000000
    0.75000000  0.00000000  0.75000000
    0.25000000  0.50000000  0.75000000
    0.75000000  0.50000000  0.75000000
    0.12500000  0.12500000  0.12500000
    0.62500000  0.12500000  0.12500000
    0.12500000  0.62500000  0.12500000
    0.62500000  0.62500000  0.12500000
    0.12500000  0.12500000  0.62500000
    0.62500000  0.12500000  0.62500000
    0.12500000  0.62500000  0.62500000
    0.62500000  0.62500000  0.62500000
    0.25000000  0.25000000  0.00000000
    0.75000000  0.25000000  0.00000000
    0.25000000  0.75000000  0.00000000
    0.75000000  0.75000000  0.00000000
    0.25000000  0.25000000  0.50000000
    0.75000000  0.25000000  0.50000000
    0.25000000  0.75000000  0.50000000
    0.75000000  0.75000000  0.50000000
    0.12500000  0.37500000  0.37500000
    0.62500000  0.37500000  0.37500000
    0.12500000  0.87500000  0.37500000
    0.62500000  0.87500000  0.37500000
    0.12500000  0.37500000  0.87500000
    0.62500000  0.37500000  0.87500000
    0.12500000  0.87500000  0.87500000
    0.62500000  0.87500000  0.87500000
=== {{TAG|KPOINTS}} ===
*We will start with a single k point in this example:
K-Points
  0
Gamma
  1  1  1
  0  0  0
=== {{TAG|INCAR}} ===
#Basic parameters
ISMEAR = 0
SIGMA = 0.1
LREAL = Auto
PREC = FAST
ALGO = FAST
ISYM = -1
NELM = 100
EDIFF = 1E-4
LWAVE = .FALSE.
LCHARG = .FALSE.
#Parallelization of ab initio calculations
NCORE = 2
#MD
IBRION = 0
MDALGO = 2
ISIF = 2
SMASS = 1.0
TEBEG = 2000
NSW = 10000
POTIM = 3.0
#Machine learning paramters
ML_FF_LMLFF = .TRUE.
ML_FF_ISTART = 0
ML_FF_NWRITE = 2
ML_FF_EATOM = -.70128086E+00
*The user should be familiar at this step how to run a basic molecular dynamics calculations. If not please go through the example here: {{TAG|Liquid Si - Standard MD}}.
*Machine learning is switched on by setting the following tag: {{TAG|ML_FF_LMLFF}}=''.TRUE.''.
*By setting the tag {{TAG|ML_FF_ISTART}} to zero learning from scratch is selected.
*The flag {{TAG|ML_FF_NWRITE}}=2 selects a more verbose output where the error on energies, forces and stress are output to the {{TAG|ML_LOGFILE}} file. This setting is very handy to check the accuracy of the force field.
*The tag {{TAG|ML_FF_EATOM}}=-.70128086E+00 sets the atomic reference energy for each species. How to obtain that energy is explained below.
== Calculation ==
=== Reference energy ===
Before the force field for liquid Si can be calculated, the atomic energy of a single Si atom in a large enough box has to be calculated.
For that the following steps have to be done:
*Create a new directory ''Si_ATOM'' by typing ''mkdir Si_ATOM'' and go to that directory ''cd Si_ATOM''.
*
== Download ==
{{Template:Machine learning force field - Tutorial}}
[[Category:Examples]]

Revision as of 17:30, 23 July 2019

Task

Generating a machine learning force field for liquid Si.

Input

POSCAR

  • In this example we start from a 64 atom super cell of diamond-fcc Si (the same as in this example: Liquid Si - Standard MD:
Si cubic diamond 2x2x2 super cell of conventional cell
     5.43090000000000
    2.00000000   0.00000000   0.00000000
    0.00000000   2.00000000   0.00000000
    0.00000000   0.00000000   2.00000000
   Si
   64
Direct
   0.00000000   0.00000000   0.00000000
   0.50000000   0.00000000   0.00000000
   0.00000000   0.50000000   0.00000000
   0.50000000   0.50000000   0.00000000
   0.00000000   0.00000000   0.50000000
   0.50000000   0.00000000   0.50000000
   0.00000000   0.50000000   0.50000000
   0.50000000   0.50000000   0.50000000
   0.37500000   0.12500000   0.37500000
   0.87500000   0.12500000   0.37500000
   0.37500000   0.62500000   0.37500000
   0.87500000   0.62500000   0.37500000
   0.37500000   0.12500000   0.87500000
   0.87500000   0.12500000   0.87500000
   0.37500000   0.62500000   0.87500000
   0.87500000   0.62500000   0.87500000
   0.00000000   0.25000000   0.25000000
   0.50000000   0.25000000   0.25000000
   0.00000000   0.75000000   0.25000000
   0.50000000   0.75000000   0.25000000
   0.00000000   0.25000000   0.75000000
   0.50000000   0.25000000   0.75000000
   0.00000000   0.75000000   0.75000000
   0.50000000   0.75000000   0.75000000
   0.37500000   0.37500000   0.12500000
   0.87500000   0.37500000   0.12500000
   0.37500000   0.87500000   0.12500000
   0.87500000   0.87500000   0.12500000
   0.37500000   0.37500000   0.62500000
   0.87500000   0.37500000   0.62500000
   0.37500000   0.87500000   0.62500000
   0.87500000   0.87500000   0.62500000
   0.25000000   0.00000000   0.25000000
   0.75000000   0.00000000   0.25000000
   0.25000000   0.50000000   0.25000000
   0.75000000   0.50000000   0.25000000
   0.25000000   0.00000000   0.75000000
   0.75000000   0.00000000   0.75000000 
   0.25000000   0.50000000   0.75000000
   0.75000000   0.50000000   0.75000000
   0.12500000   0.12500000   0.12500000
   0.62500000   0.12500000   0.12500000
   0.12500000   0.62500000   0.12500000
   0.62500000   0.62500000   0.12500000
   0.12500000   0.12500000   0.62500000
   0.62500000   0.12500000   0.62500000
   0.12500000   0.62500000   0.62500000
   0.62500000   0.62500000   0.62500000
   0.25000000   0.25000000   0.00000000
   0.75000000   0.25000000   0.00000000
   0.25000000   0.75000000   0.00000000
   0.75000000   0.75000000   0.00000000
   0.25000000   0.25000000   0.50000000
   0.75000000   0.25000000   0.50000000
   0.25000000   0.75000000   0.50000000
   0.75000000   0.75000000   0.50000000
   0.12500000   0.37500000   0.37500000
   0.62500000   0.37500000   0.37500000
   0.12500000   0.87500000   0.37500000
   0.62500000   0.87500000   0.37500000
   0.12500000   0.37500000   0.87500000
   0.62500000   0.37500000   0.87500000
   0.12500000   0.87500000   0.87500000
   0.62500000   0.87500000   0.87500000

KPOINTS

  • We will start with a single k point in this example:
K-Points
 0
Gamma
 1  1  1
 0  0  0

INCAR

#Basic parameters
ISMEAR = 0
SIGMA = 0.1
LREAL = Auto
PREC = FAST
ALGO = FAST
ISYM = -1
NELM = 100
EDIFF = 1E-4
LWAVE = .FALSE.
LCHARG = .FALSE.

#Parallelization of ab initio calculations
NCORE = 2

#MD
IBRION = 0
MDALGO = 2
ISIF = 2
SMASS = 1.0
TEBEG = 2000
NSW = 10000
POTIM = 3.0

#Machine learning paramters
ML_FF_LMLFF = .TRUE.
ML_FF_ISTART = 0
ML_FF_NWRITE = 2
ML_FF_EATOM = -.70128086E+00
  • The user should be familiar at this step how to run a basic molecular dynamics calculations. If not please go through the example here: Liquid Si - Standard MD.
  • Machine learning is switched on by setting the following tag: ML_FF_LMLFF=.TRUE..
  • By setting the tag ML_FF_ISTART to zero learning from scratch is selected.
  • The flag ML_FF_NWRITE=2 selects a more verbose output where the error on energies, forces and stress are output to the ML_LOGFILE file. This setting is very handy to check the accuracy of the force field.
  • The tag ML_FF_EATOM=-.70128086E+00 sets the atomic reference energy for each species. How to obtain that energy is explained below.

Calculation

Reference energy

Before the force field for liquid Si can be calculated, the atomic energy of a single Si atom in a large enough box has to be calculated. For that the following steps have to be done:

  • Create a new directory Si_ATOM by typing mkdir Si_ATOM and go to that directory cd Si_ATOM.

Download

Retrieved from "https://www.vasp.at/wiki/index.php?title=Liquid_Si_-_MLFF&oldid=9408"