This thread is locked.Only browsing is available.
Top Page > Browsing
Is the Orbital Optimization variational with rispect to the Total Energy?
Date: 2010/06/02 00:46
Name: Mauro   <mantegm@tcd.ie>

Dear Prof. Ozaki,

I performed calculations with OpenMx on several different systems but I still don't fully understand if the orbital optimization procedure is variational with respect to the total energy, in the case of a "restarted" calculation.
This is just an example (I can show a lot of them) where, after I restart the calculation, there is a strong variation of the total energy:

***********************************************************
***********************************************************
History of geometry optimization
***********************************************************
***********************************************************

MD_iter SD_scaling |Maximum force| Maximum step Utot
(Hartree/Bohr) (Ang) (Hartree)

1 0.09448630 0.01336337 0.00066817 -2118.84888991
2 0.09448630 0.00759660 0.00037983 -2118.84927638
3 0.09448630 0.01309462 0.00065473 -2118.84905295
4 0.09448630 0.00690073 0.00034504 -2118.84942645
5 0.09448630 0.01224749 0.00061237 -2118.84920579
6 0.09448630 0.01080701 0.00054035 -2118.85014586
7 0.09448630 0.00761054 0.00038053 -2118.85281106
8 0.09448630 0.00616471 0.00030824 -2118.85317501
9 0.09448630 0.00404163 0.00020208 -2118.85406198
10 0.09448630 0.00461742 0.00023087 -2118.85477119
11 0.09448630 0.00506093 0.00025305 -2118.85508478
12 0.09448630 0.00285542 0.00014277 -2118.85484215
13 0.09448630 0.00231183 0.00011559 -2118.85510565
14 0.09448630 0.00302907 0.00015145 -2118.85535628
15 0.09448630 0.00471748 0.00023587 -2118.85592002
16 0.09448630 0.00493121 0.00024656 -2118.85616241
17 0.09448630 0.00571419 0.00028571 -2118.85688621
18 0.09448630 0.00926938 0.00046347 -2118.85367628
19 0.09448630 0.01046145 0.00052307 -2118.85573153
20 0.09448630 0.00427271 0.00021364 -2118.85621658
21 0.09448630 0.00721642 0.00036082 -2118.85700705

***********************************************************
***********************************************************
History of geometry optimization
***********************************************************
***********************************************************

MD_iter SD_scaling |Maximum force| Maximum step Utot
(Hartree/Bohr) (Ang) (Hartree)

1 0.09448630 0.00702033 0.00035102 -2118.85311350
2 0.09448630 0.00693983 0.00034699 -2118.85313224
3 0.09448630 0.00683646 0.00034182 -2118.85314942
4 0.09448630 0.00675874 0.00033794 -2118.85316558
5 0.09448630 0.00651362 0.00032568 -2118.85318098
6 0.09448630 0.00563209 0.00028160 -2118.85331740
7 0.09448630 0.00232997 0.00011650 -2118.85354347
8 0.09448630 0.00436749 0.00021837 -2118.85351803
9 0.09448630 0.00368519 0.00018426 -2118.85331589
10 0.09448630 0.00257001 0.00012850 -2118.85346714
11 0.09448630 0.00220104 0.00011005 -2118.85341500
12 0.09448630 0.00357679 0.00017884 -2118.85335177
13 0.09448630 0.00168147 0.00008407 -2118.85337236
14 0.09448630 0.00226040 0.00011302 -2118.85341395
15 0.09448630 0.00182393 0.00009120 -2118.85338776
16 0.09448630 0.00085448 0.00004272 -2118.85341094
17 0.09448630 0.00081785 0.00004089 -2118.85341250
18 0.09448630 0.00078089 0.00003904 -2118.85340161
19 0.09448630 0.00069217 0.00003461 -2118.85340966
20 0.09448630 0.00030674 0.00001534 -2118.85342731
21 0.09448630 0.00030895 0.00001545 -2118.85343013

***********************************************************
***********************************************************
History of geometry optimization
***********************************************************
***********************************************************

MD_iter SD_scaling |Maximum force| Maximum step Utot
(Hartree/Bohr) (Ang) (Hartree)

1 0.09448629 0.00081843 0.00004092 -2118.86721295
2 0.09448629 0.00080319 0.00004016 -2118.86721442
3 0.09448629 0.00077649 0.00003882 -2118.86721663
4 0.09448629 0.00075450 0.00003772 -2118.86721774
5 0.09448629 0.00071510 0.00003576 -2118.86721947
6 0.09448629 0.00050104 0.00002505 -2118.86723459
7 0.09448629 0.00030122 0.00001506 -2118.86727119
8 0.09448629 0.00027228 0.00001361 -2118.86727683
9 0.09448629 0.00120669 0.00006033 -2118.86740609
10 0.09448629 0.00051929 0.00002596 -2118.86729332
11 0.09448629 0.00046655 0.00002333 -2118.86729858
12 0.09448629 0.00024582 0.00001229 -2118.86731088
13 0.09448629 0.00046206 0.00002310 -2118.86724706
14 0.09448629 0.00030428 0.00001521 -2118.86729322
15 0.09448629 0.00028415 0.00001421 -2118.86730507
16 0.09448629 0.00029036 0.00001452 -2118.86731663
17 0.09448629 0.00079748 0.00003987 -2118.86735756
18 0.09448629 0.00030321 0.00001516 -2118.86731383
19 0.09448629 0.00017211 0.00000861 -2118.86731706
20 0.09448629 0.00014406 0.00000720 -2118.86731951
21 0.09448629 0.00008315 0.00000416 -2118.86731586



(to restart the calculation I just copied the *.dat# file into the *.dat one.)

Regards

Mauro
e
Page: [1]