[QE-users] Relaxation: bfgs failed after 4 scf cycles and 3 bfgs steps, convergence not achieved
Paolo Giannozzi
paolo.giannozzi at uniud.it
Thu Sep 19 12:07:40 CEST 2024
QE was not devised to reach the level of convergence that you would like
to achieve. I don't think you can do that, without some insight of the
origin of numerical errors in QE, and maybe not even with: numerical
errors often arise in system and mathematical libraries
Paolo
On 18/09/2024 14:38, aleksandr.doma--- via users wrote:
> Dear QE users & devs!
>
> We get the same error using different pseudopotentials (SSSP, LDA,
> PBEsol) at the final steps of the structure relaxation. I mean the message:
> / bfgs failed after 4 scf cycles and 3 bfgs steps, convergence not
> achieved/
> / (criteria: energy < 1.0E-08 Ry, force < 1.0E-08 Ry/Bohr, cell <
> 1.0E-01 kbar)/
> /
> /
> / End of BFGS Geometry Optimization/
>
>
> Using the last geometry (lattice parameters and atomic coordinates) in
> the next input file, QE takes 4 steps again and stops with the same problem.
>
> We would like to achieve "ideal relaxation" for use in elastic and
> phonon calculations. Highlights of the input file:
>
> /&control/
> / ..../
> / calculation = 'vc-relax',
> /
> / forc_conv_thr = 1.0D-8,/
> / etot_conv_thr = 1.0D-8,/
> / //
> /&SYSTEM/
> / celldm(1) = 15.89938814/
> / celldm(2) = 1.30410009/
> / celldm(3) = 1.47077836/
> / ibrav = 8,/
> / nat = 104,/
> / ntyp = 4,/
> / ecutrho = 800/
> / ecutwfc = 100/
> ///
> /&ELECTRONS/
> / conv_thr = 1.D-10, (or 1.D-11)/
> / mixing_beta=0.5/
> ///
> /
> /
> /&IONS/
> / ion_dynamics = "bfgs"/
> ///
> /&CELL/
> / cell_dofree = "ibrav"/
> / cell_dynamics = "bfgs"/
> / press = 0.0/
> / press_conv_thr = 0.1/
> /
> /
> /K_POINTS AUTOMATIC/
> /6 5 4 0 0 0/
> /(8 6 6 0 0 0)/
>
> Restarting the 'vc-relax' calculation several times, we got the
> following "total force" results:
> Total force = 0.000080 Total SCF correction = 0.000018
> SCF correction compared to forces is large: reduce conv_thr to get
> better values
> ----------------------------------------------------------------------------------------------
> Total force = 0.000084 Total SCF correction = 0.000037
> SCF correction compared to forces is large: reduce conv_thr to get
> better values
> ----------------------------------------------------------------------------------------------
> Total force = 0.000072 Total SCF correction = 0.000002
> ----------------------------------------------------------------------------------------------
> Total force = 0.000072 Total SCF correction = 0.000003
> Computing stress (Cartesian axis) and pressure
>
> total stress (Ry/bohr**3) (kbar) P=
> 0.00
> 0.00000000 0.00000000 0.00000000 0.00 0.00
> 0.00
> 0.00000000 0.00000001 0.00000000 0.00 0.00
> 0.00
> 0.00000000 0.00000000 0.00000002 0.00 0.00
> 0.00
>
> Message from routine bfgs:
> history already reset at previous step: exiting
> Energy error = 2.4E-09 Ry
> Gradient error = 1.3E-05 Ry/Bohr
> Cell gradient error = 3.1E-03 kbar
> ibrav = 8
> celldm(1) = 15.89938798
> celldm(2) = 1.30409950
> celldm(3) = 1.47077720
> Input lattice vectors:
> 0.99999903 0.00000000 0.00000000
> 0.00000000 1.30409824 0.00000000
> 0.00000000 0.00000000 1.47077577
> New lattice vectors in INITIAL alat:
> 0.99999903 0.00000000 0.00000000
> 0.00000000 1.30409824 0.00000000
> 0.00000000 0.00000000 1.47077577
> New lattice vectors in NEW alat (for information only):
> 1.00000000 0.00000000 0.00000000
> 0.00000000 1.30409950 0.00000000
> 0.00000000 0.00000000 1.47077720
> Discrepancy in bohr = 0.000000 0.000000 0.000000
>
> bfgs failed after 4 scf cycles and 3 bfgs steps, convergence
> not achieved
> (criteria: energy < 1.0E-08 Ry, force < 1.0E-08 Ry/Bohr, cell <
> 1.0E-01 kbar)
>
> End of BFGS Geometry Optimization
> *=======================================================*
>
> *next file:*
> *
> *
> Total force = 0.000071 Total SCF correction = 0.000008
> SCF correction compared to forces is large: reduce conv_thr to get
> better values
> -------------------------------------------------------------------------------------------------
> Total force = 0.000077 Total SCF correction = 0.000009
> SCF correction compared to forces is large: reduce conv_thr to get
> better values
> -------------------------------------------------------------------------------------------------
> Total force = 0.000078 Total SCF correction = 0.000000
> --------------------------------------------------------------------------------------------------
> Total force = 0.000082 Total SCF correction = 0.000001
>
>
> Computing stress (Cartesian axis) and pressure
>
> total stress (Ry/bohr**3) (kbar) P=
> -0.00
> -0.00000001 0.00000000 0.00000000 -0.00 0.00
> 0.00
> 0.00000000 -0.00000002 0.00000000 0.00 -0.00
> 0.00
> 0.00000000 0.00000000 -0.00000001 0.00 0.00
> -0.00
>
> Message from routine bfgs:
> history already reset at previous step: exiting
> Energy error = 1.5E-09 Ry
> Gradient error = 2.2E-05 Ry/Bohr
> Cell gradient error = 2.7E-03 kbar
> ibrav = 8
> celldm(1) = 15.89938814
> celldm(2) = 1.30410009
> celldm(3) = 1.47077836
> Input lattice vectors:
> 1.00000001 0.00000000 0.00000000
> 0.00000000 1.30410010 0.00000000
> 0.00000000 0.00000000 1.47077837
> New lattice vectors in INITIAL alat:
> 1.00000001 0.00000000 0.00000000
> 0.00000000 1.30410010 0.00000000
> 0.00000000 0.00000000 1.47077837
> New lattice vectors in NEW alat (for information only):
> 1.00000000 0.00000000 0.00000000
> 0.00000000 1.30410009 0.00000000
> 0.00000000 0.00000000 1.47077836
> Discrepancy in bohr = 0.000000 0.000000 0.000000
>
> bfgs failed after 4 scf cycles and 3 bfgs steps, convergence
> not achieved
> (criteria: energy < 1.0E-08 Ry, force < 1.0E-08 Ry/Bohr, cell <
> 1.0E-01 kbar)
>
> End of BFGS Geometry Optimization
> *==========================================================*
>
>
> We seem to have reached a threshold as the last calculation shows an
> increase in "total force" and stops with the same problem after 4 steps:
>
> Total force = 0.000082 Total SCF correction = 0.000008
> ------------------------------------------------------------------------------------
> Total force = 0.000088 Total SCF correction = 0.000009
> SCF correction compared to forces is large: reduce conv_thr to get
> better values
> ---------------------------------------------------------------------------------------
> Total force = 0.000093 Total SCF correction = 0.000000
> ---------------------------------------------------------------------------------------
> Total force = 0.000105 Total SCF correction = 0.000001
>
> *==================================================*
>
> Questions:
> 1) What is the acceptable threshold for further calculations of elastic
> properties? (for example, with the relaxed "Total force = 0.000081" we
> had error of 200% in elastic constants using thermo_pw for tetraborate
> crystals)
> 2) Can the relaxation error be avoided to obtain a more accurate geometry?
>
> Thanks in advance!
> Oleksandr
>
>
> _______________________________________________
> The Quantum ESPRESSO community stands by the Ukrainian
> people and expresses its concerns about the devastating
> effects that the Russian military offensive has on their
> country and on the free and peaceful scientific, cultural,
> and economic cooperation amongst peoples
> _______________________________________________
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users at lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users
--
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 206, 33100 Udine Italy, +39-0432-558216
X SCUOLA ENERGIE RINNOVABILI 16-21/9/2024
More information about the users
mailing list