scnc
/
quantum-espresso
mirror of https://gitlab.com/QEF/q-e.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Fixed a problem with variable cell dynamics. Doc updated and cleaned. 

C.S.


git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@1487 c92efa57-630b-4861-b058-cf58834340f0
This commit is contained in:
sbraccia 2004-11-29 07:47:54 +00:00
parent 804ac1fdcf
commit e2a6fbcaa9
2 changed files with 38 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
Doc/INPUT_PW
View File

@ -450,6 +450,13 @@ ion_temperature
tempw REAL ( default = 300.D0 )
starting temperature (Kelvin) in MD runs
ttol REAL ( default = 1.D-3 )
tolerance for velocity rescaling. Velocities are
not rescaled if the ratio of the run-averaged and
target temperature differs from unit less than ttol.
This keyword is used only in the case of variable cell
calculations.
delta_t REAL ( default = 1.D0 )
delta_T = 1 : nothing is done.
delta_T /= 1 && delta_T > 0 : at each step the temperature is
@ -457,16 +464,13 @@ delta_t REAL ( default = 1.D0 )
done rescaling all the velocities.
delta_T < 0 : every 'nraise' steps the temperature
is reduced by -delta_T
This keyword is NOT used in the case of variable cell
calculations.
nraise INTEGER (default = 100 )
the temperature is reduced of -delta_T every 'nraise' steps.
upscale REAL ( default = 10.D0 )
max reduction factor for conv_thr during structural optimization
conv_thr is automatically reduced when the relaxation
approaches convergence so that forces are still accurate,
but conv_thr will not be reduced to less that
conv_thr / upscale
This keyword is NOT used in the case of variable cell
calculations.
potential_extrapolation
CHARACTER
@ -481,6 +485,18 @@ potential_extrapolation
(default for calcualtion='md' and calcualtion='neb')
'wfc2': as above, with second order formula
!
! ... keywords used only in BFGS calculations
!
upscale REAL ( default = 10.D0 )
max reduction factor for conv_thr during structural optimization
conv_thr is automatically reduced when the relaxation
approaches convergence so that forces are still accurate,
but conv_thr will not be reduced to less that
conv_thr / upscale
lbfgs_ndim INTEGER ( default = 1 )
number of old forces and displacements vectors used in the
linear scaling BFGS algorithm. When lbfgs_ndim = 1 the complete
@ -517,6 +533,10 @@ w_1, w_2
parameters used in line search based on the Wolfe conditions
(bfgs only)
!
! ... keywords used only in NEB and SMD calculations
!
num_of_images INTEGER ( default = 0 )
number of points used to discrtize the path

8
PW/input.f90
View File

@ -47,7 +47,7 @@ SUBROUTINE iosys()
USE char, ONLY : title_ => title, &
crystal
!
USE cellmd, ONLY : cmass, omega_old, at_old, ntcheck, &
USE cellmd, ONLY : cmass, ttol, omega_old, at_old, ntcheck, &
cell_factor_ => cell_factor , &
press_ => press, &
calc, lmovecell
@ -202,7 +202,7 @@ SUBROUTINE iosys()
! IONS namelist
!
USE input_parameters, ONLY : ion_dynamics, ion_positions, ion_temperature, &
tempw, delta_t, nraise, upscale, &
tolp, tempw, delta_t, nraise, upscale, &
potential_extrapolation, &
num_of_images, path_thr, CI_scheme, opt_scheme, &
reset_vel, use_multistep, first_last_opt, damp, &
@ -254,13 +254,16 @@ SUBROUTINE iosys()
DO iiarg = 1, ( nargs - 1 )
!
CALL getarg( iiarg, input_file )
!
IF ( TRIM( input_file ) == '-input' .OR. &
TRIM( input_file ) == '-inp' .OR. &
TRIM( input_file ) == '-in' ) THEN
!
CALL getarg( ( iiarg + 1 ) , input_file )
!
OPEN ( UNIT = unit, FILE = input_file, FORM = 'FORMATTED', &
STATUS = 'OLD', IOSTAT = ierr )
!
CALL errore( 'iosys', 'input file ' // TRIM( input_file ) // &
& ' not found' , ierr )
!
@ -716,6 +719,7 @@ SUBROUTINE iosys()
CONTINUE
CASE ( 'rescaling' )
temperature = tempw
ttol = tolp
CASE DEFAULT
CALL errore( ' iosys ', ' unknown ion_temperature ' // &
& TRIM( ion_temperature ), 1 )