-
Notifications
You must be signed in to change notification settings - Fork 1
/
rt_propagation_forces.F
368 lines (311 loc) · 18.2 KB
/
rt_propagation_forces.F
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
!--------------------------------------------------------------------------------------------------!
! CP2K: A general program to perform molecular dynamics simulations !
! Copyright 2000-2024 CP2K developers group <https://cp2k.org> !
! !
! SPDX-License-Identifier: GPL-2.0-or-later !
!--------------------------------------------------------------------------------------------------!
! **************************************************************************************************
!> \brief Routines needed for EMD
!> \author Florian Schiffmann (02.09)
! **************************************************************************************************
MODULE rt_propagation_forces
USE admm_types, ONLY: admm_type,&
get_admm_env
USE atomic_kind_types, ONLY: atomic_kind_type,&
get_atomic_kind_set
USE cp_control_types, ONLY: dft_control_type,&
rtp_control_type
USE cp_dbcsr_api, ONLY: &
dbcsr_copy, dbcsr_create, dbcsr_deallocate_matrix, dbcsr_get_block_p, &
dbcsr_iterator_blocks_left, dbcsr_iterator_next_block, dbcsr_iterator_start, &
dbcsr_iterator_stop, dbcsr_iterator_type, dbcsr_multiply, dbcsr_p_type, dbcsr_type, &
dbcsr_type_no_symmetry
USE cp_dbcsr_cp2k_link, ONLY: cp_dbcsr_alloc_block_from_nbl
USE cp_dbcsr_operations, ONLY: copy_fm_to_dbcsr,&
cp_dbcsr_sm_fm_multiply
USE cp_fm_struct, ONLY: cp_fm_struct_type
USE cp_fm_types, ONLY: cp_fm_create,&
cp_fm_release,&
cp_fm_type
USE kinds, ONLY: dp
USE mathconstants, ONLY: one,&
zero
USE parallel_gemm_api, ONLY: parallel_gemm
USE particle_types, ONLY: particle_type
USE qs_environment_types, ONLY: get_qs_env,&
qs_environment_type
USE qs_force_types, ONLY: add_qs_force,&
qs_force_type
USE qs_ks_types, ONLY: qs_ks_env_type
USE qs_neighbor_list_types, ONLY: neighbor_list_set_p_type
USE qs_overlap, ONLY: build_overlap_force
USE qs_rho_types, ONLY: qs_rho_get,&
qs_rho_type
USE rt_propagation_types, ONLY: get_rtp,&
rt_prop_type
#include "./base/base_uses.f90"
IMPLICIT NONE
PRIVATE
PUBLIC :: calc_c_mat_force, &
rt_admm_force
CHARACTER(len=*), PARAMETER, PRIVATE :: moduleN = 'rt_propagation_forces'
CONTAINS
! **************************************************************************************************
!> \brief calculates the three additional force contributions needed in EMD
!> P_imag*C , P_imag*B*S^-1*S_der , P*S^-1*H*S_der
!> \param qs_env ...
!> \par History
!> 02.2014 switched to dbcsr matrices [Samuel Andermatt]
!> 10.2023 merge MO-based and all-atom into one routine [Guillaume Le Breton]
!> \author Florian Schiffmann (02.09)
! **************************************************************************************************
SUBROUTINE calc_c_mat_force(qs_env)
TYPE(qs_environment_type), POINTER :: qs_env
CHARACTER(LEN=*), PARAMETER :: routineN = 'calc_c_mat_force'
REAL(KIND=dp), PARAMETER :: one = 1.0_dp, zero = 0.0_dp
INTEGER :: handle, i, im, ispin, re
INTEGER, ALLOCATABLE, DIMENSION(:) :: atom_of_kind, kind_of
TYPE(atomic_kind_type), DIMENSION(:), POINTER :: atomic_kind_set
TYPE(dbcsr_p_type), DIMENSION(:), POINTER :: C_mat, rho_ao, rho_ao_im, rho_new, &
S_der, SinvB, SinvH, SinvH_imag
TYPE(dbcsr_type), POINTER :: S_inv, tmp
TYPE(dft_control_type), POINTER :: dft_control
TYPE(particle_type), DIMENSION(:), POINTER :: particle_set
TYPE(qs_force_type), DIMENSION(:), POINTER :: force
TYPE(qs_rho_type), POINTER :: rho
TYPE(rt_prop_type), POINTER :: rtp
TYPE(rtp_control_type), POINTER :: rtp_control
CALL timeset(routineN, handle)
NULLIFY (rtp, particle_set, atomic_kind_set, dft_control)
CALL get_qs_env(qs_env, &
rtp=rtp, &
rho=rho, &
particle_set=particle_set, &
atomic_kind_set=atomic_kind_set, &
force=force, &
dft_control=dft_control)
rtp_control => dft_control%rtp_control
CALL get_rtp(rtp=rtp, C_mat=C_mat, S_der=S_der, S_inv=S_inv, &
SinvH=SinvH, SinvB=SinvB)
CALL get_atomic_kind_set(atomic_kind_set=atomic_kind_set, atom_of_kind=atom_of_kind, kind_of=kind_of)
NULLIFY (tmp)
ALLOCATE (tmp)
CALL dbcsr_create(tmp, template=SinvB(1)%matrix)
IF (rtp%linear_scaling) THEN
CALL get_rtp(rtp=rtp, rho_new=rho_new)
ELSE
CALL qs_rho_get(rho_struct=rho, rho_ao=rho_ao, rho_ao_im=rho_ao_im)
END IF
! If SinvH has an imaginary part (the minus sign is already in SinvH_imag)
IF (rtp%propagate_complex_ks) CALL get_rtp(rtp=rtp, SinvH_imag=SinvH_imag)
DO ispin = 1, SIZE(SinvH)
re = 2*ispin - 1
im = 2*ispin
IF (rtp%linear_scaling) THEN
CALL dbcsr_multiply("N", "N", one, SinvH(ispin)%matrix, rho_new(re)%matrix, zero, tmp, &
filter_eps=rtp%filter_eps)
IF (rtp%propagate_complex_ks) &
CALL dbcsr_multiply("N", "N", one, SinvH_imag(ispin)%matrix, rho_new(im)%matrix, one, tmp, &
filter_eps=rtp%filter_eps)
CALL dbcsr_multiply("N", "N", one, SinvB(ispin)%matrix, rho_new(im)%matrix, one, tmp, &
filter_eps=rtp%filter_eps)
CALL compute_forces(force, tmp, S_der, rho_new(im)%matrix, C_mat, kind_of, atom_of_kind)
ELSE
CALL dbcsr_multiply("N", "N", one, SinvH(ispin)%matrix, rho_ao(ispin)%matrix, zero, tmp)
IF (rtp%propagate_complex_ks) &
CALL dbcsr_multiply("N", "N", one, SinvH_imag(ispin)%matrix, rho_ao_im(ispin)%matrix, one, tmp)
CALL dbcsr_multiply("N", "N", one, SinvB(ispin)%matrix, rho_ao_im(ispin)%matrix, one, tmp)
CALL compute_forces(force, tmp, S_der, rho_ao_im(ispin)%matrix, C_mat, kind_of, atom_of_kind)
END IF
END DO
! recall QS forces, at this point have the other sign.
DO i = 1, SIZE(force)
force(i)%ehrenfest(:, :) = -force(i)%ehrenfest(:, :)
END DO
CALL dbcsr_deallocate_matrix(tmp)
CALL timestop(handle)
END SUBROUTINE
! **************************************************************************************************
!> \brief ...
!> \param force ...
!> \param tmp ...
!> \param S_der ...
!> \param rho_im ...
!> \param C_mat ...
!> \param kind_of ...
!> \param atom_of_kind ...
! **************************************************************************************************
SUBROUTINE compute_forces(force, tmp, S_der, rho_im, C_mat, kind_of, atom_of_kind)
TYPE(qs_force_type), DIMENSION(:), POINTER :: force
TYPE(dbcsr_type), POINTER :: tmp
TYPE(dbcsr_p_type), DIMENSION(:), POINTER :: S_der
TYPE(dbcsr_type), POINTER :: rho_im
TYPE(dbcsr_p_type), DIMENSION(:), POINTER :: C_mat
INTEGER, ALLOCATABLE, DIMENSION(:) :: kind_of, atom_of_kind
INTEGER :: col_atom, i, ikind, kind_atom, row_atom
LOGICAL :: found
REAL(dp), DIMENSION(:), POINTER :: block_values, block_values2
TYPE(dbcsr_iterator_type) :: iter
DO i = 1, 3
!Calculate the sum over the hadmard product
!S_der part
CALL dbcsr_iterator_start(iter, tmp)
DO WHILE (dbcsr_iterator_blocks_left(iter))
CALL dbcsr_iterator_next_block(iter, row_atom, col_atom, block_values)
CALL dbcsr_get_block_p(S_der(i)%matrix, row_atom, col_atom, block_values2, found=found)
IF (found) THEN
ikind = kind_of(col_atom)
kind_atom = atom_of_kind(col_atom)
!The block_values are in a vector format,
! so the dot_product is the sum over all elements of the hamand product, that I need
force(ikind)%ehrenfest(i, kind_atom) = force(ikind)%ehrenfest(i, kind_atom) + &
2.0_dp*DOT_PRODUCT(block_values, block_values2)
END IF
END DO
CALL dbcsr_iterator_stop(iter)
!C_mat part
CALL dbcsr_iterator_start(iter, rho_im)
DO WHILE (dbcsr_iterator_blocks_left(iter))
CALL dbcsr_iterator_next_block(iter, row_atom, col_atom, block_values)
CALL dbcsr_get_block_p(C_mat(i)%matrix, row_atom, col_atom, block_values2, found=found)
IF (found) THEN
ikind = kind_of(col_atom)
kind_atom = atom_of_kind(col_atom)
!The block_values are in a vector format, so the dot_product is
! the sum over all elements of the hamand product, that I need
force(ikind)%ehrenfest(i, kind_atom) = force(ikind)%ehrenfest(i, kind_atom) + &
2.0_dp*DOT_PRODUCT(block_values, block_values2)
END IF
END DO
CALL dbcsr_iterator_stop(iter)
END DO
END SUBROUTINE compute_forces
! **************************************************************************************************
!> \brief ...
!> \param qs_env ...
! **************************************************************************************************
SUBROUTINE rt_admm_force(qs_env)
TYPE(qs_environment_type), POINTER :: qs_env
TYPE(admm_type), POINTER :: admm_env
TYPE(cp_fm_type), DIMENSION(:), POINTER :: mos, mos_admm
TYPE(dbcsr_p_type), DIMENSION(:), POINTER :: KS_aux_im, KS_aux_re, matrix_s_aux_fit, &
matrix_s_aux_fit_vs_orb
TYPE(rt_prop_type), POINTER :: rtp
CALL get_qs_env(qs_env, &
admm_env=admm_env, &
rtp=rtp)
CALL get_admm_env(admm_env, matrix_ks_aux_fit=KS_aux_re, &
matrix_ks_aux_fit_im=KS_aux_im, &
matrix_s_aux_fit=matrix_s_aux_fit, &
matrix_s_aux_fit_vs_orb=matrix_s_aux_fit_vs_orb)
CALL get_rtp(rtp=rtp, mos_new=mos, admm_mos=mos_admm)
! currently only none option
CALL rt_admm_forces_none(qs_env, admm_env, KS_aux_re, KS_aux_im, &
matrix_s_aux_fit, matrix_s_aux_fit_vs_orb, mos_admm, mos)
END SUBROUTINE rt_admm_force
! **************************************************************************************************
!> \brief ...
!> \param qs_env ...
!> \param admm_env ...
!> \param KS_aux_re ...
!> \param KS_aux_im ...
!> \param matrix_s_aux_fit ...
!> \param matrix_s_aux_fit_vs_orb ...
!> \param mos_admm ...
!> \param mos ...
! **************************************************************************************************
SUBROUTINE rt_admm_forces_none(qs_env, admm_env, KS_aux_re, KS_aux_im, matrix_s_aux_fit, matrix_s_aux_fit_vs_orb, mos_admm, mos)
TYPE(qs_environment_type), POINTER :: qs_env
TYPE(admm_type), POINTER :: admm_env
TYPE(dbcsr_p_type), DIMENSION(:), POINTER :: KS_aux_re, KS_aux_im, matrix_s_aux_fit, &
matrix_s_aux_fit_vs_orb
TYPE(cp_fm_type), DIMENSION(:), POINTER :: mos_admm, mos
INTEGER :: im, ispin, jspin, nao, natom, naux, nmo, &
re
REAL(KIND=dp), ALLOCATABLE, DIMENSION(:, :) :: admm_force
TYPE(atomic_kind_type), DIMENSION(:), POINTER :: atomic_kind_set
TYPE(cp_fm_struct_type), POINTER :: mstruct
TYPE(cp_fm_type), DIMENSION(2) :: tmp_aux_aux, tmp_aux_mo, tmp_aux_mo1, &
tmp_aux_nao
TYPE(dbcsr_type), POINTER :: matrix_w_q, matrix_w_s
TYPE(neighbor_list_set_p_type), DIMENSION(:), &
POINTER :: sab_aux_fit_asymm, sab_aux_fit_vs_orb
TYPE(qs_force_type), DIMENSION(:), POINTER :: force
TYPE(qs_ks_env_type), POINTER :: ks_env
NULLIFY (sab_aux_fit_asymm, sab_aux_fit_vs_orb, ks_env)
CALL get_qs_env(qs_env, ks_env=ks_env)
CALL get_admm_env(admm_env, sab_aux_fit_asymm=sab_aux_fit_asymm, &
sab_aux_fit_vs_orb=sab_aux_fit_vs_orb)
ALLOCATE (matrix_w_s)
CALL dbcsr_create(matrix_w_s, template=matrix_s_aux_fit(1)%matrix, &
name='W MATRIX AUX S', matrix_type=dbcsr_type_no_symmetry)
CALL cp_dbcsr_alloc_block_from_nbl(matrix_w_s, sab_aux_fit_asymm)
ALLOCATE (matrix_w_q)
CALL dbcsr_copy(matrix_w_q, matrix_s_aux_fit_vs_orb(1)%matrix, &
"W MATRIX AUX Q")
DO jspin = 1, 2
CALL cp_fm_create(tmp_aux_aux(jspin), admm_env%work_aux_aux%matrix_struct, name="taa")
CALL cp_fm_create(tmp_aux_nao(jspin), admm_env%work_aux_orb%matrix_struct, name="tao")
END DO
DO ispin = 1, SIZE(KS_aux_re)
re = 2*ispin - 1; im = 2*ispin
naux = admm_env%nao_aux_fit; nmo = admm_env%nmo(ispin); nao = admm_env%nao_orb
mstruct => admm_env%work_aux_nmo(ispin)%matrix_struct
DO jspin = 1, 2
CALL cp_fm_create(tmp_aux_mo(jspin), mstruct, name="tam")
CALL cp_fm_create(tmp_aux_mo1(jspin), mstruct, name="tam")
END DO
! First calculate H=KS_aux*C~, real part ends on work_aux_aux2, imaginary part ends at work_aux_aux3
CALL cp_dbcsr_sm_fm_multiply(KS_aux_re(ispin)%matrix, mos_admm(re), tmp_aux_mo(re), nmo, 4.0_dp, 0.0_dp)
CALL cp_dbcsr_sm_fm_multiply(KS_aux_re(ispin)%matrix, mos_admm(im), tmp_aux_mo(im), nmo, 4.0_dp, 0.0_dp)
CALL cp_dbcsr_sm_fm_multiply(KS_aux_im(ispin)%matrix, mos_admm(im), tmp_aux_mo(re), nmo, -4.0_dp, 1.0_dp)
CALL cp_dbcsr_sm_fm_multiply(KS_aux_im(ispin)%matrix, mos_admm(re), tmp_aux_mo(im), nmo, 4.0_dp, 1.0_dp)
! Next step compute S-1*H
CALL parallel_gemm('N', 'N', naux, nmo, naux, 1.0_dp, admm_env%S_inv, tmp_aux_mo(re), 0.0_dp, tmp_aux_mo1(re))
CALL parallel_gemm('N', 'N', naux, nmo, naux, 1.0_dp, admm_env%S_inv, tmp_aux_mo(im), 0.0_dp, tmp_aux_mo1(im))
! Here we go on with Ws=S-1*H * C^H (take care of sign of the imaginary part!!!)
CALL parallel_gemm("N", "T", naux, nao, nmo, -1.0_dp, tmp_aux_mo1(re), mos(re), 0.0_dp, &
tmp_aux_nao(re))
CALL parallel_gemm("N", "T", naux, nao, nmo, -1.0_dp, tmp_aux_mo1(im), mos(im), 1.0_dp, &
tmp_aux_nao(re))
CALL parallel_gemm("N", "T", naux, nao, nmo, 1.0_dp, tmp_aux_mo1(re), mos(im), 0.0_dp, &
tmp_aux_nao(im))
CALL parallel_gemm("N", "T", naux, nao, nmo, -1.0_dp, tmp_aux_mo1(im), mos(re), 1.0_dp, &
tmp_aux_nao(im))
! Let's do the final bit Wq=S-1*H * C^H * A^T
CALL parallel_gemm('N', 'T', naux, naux, nao, -1.0_dp, tmp_aux_nao(re), admm_env%A, 0.0_dp, tmp_aux_aux(re))
CALL parallel_gemm('N', 'T', naux, naux, nao, -1.0_dp, tmp_aux_nao(im), admm_env%A, 0.0_dp, tmp_aux_aux(im))
! *** copy to sparse matrix
CALL copy_fm_to_dbcsr(tmp_aux_nao(re), matrix_w_q, keep_sparsity=.TRUE.)
! *** copy to sparse matrix
CALL copy_fm_to_dbcsr(tmp_aux_aux(re), matrix_w_s, keep_sparsity=.TRUE.)
DO jspin = 1, 2
CALL cp_fm_release(tmp_aux_mo(jspin))
CALL cp_fm_release(tmp_aux_mo1(jspin))
END DO
! *** This can be done in one call w_total = w_alpha + w_beta
! allocate force vector
CALL get_qs_env(qs_env=qs_env, natom=natom)
ALLOCATE (admm_force(3, natom))
admm_force = 0.0_dp
CALL build_overlap_force(ks_env, admm_force, &
basis_type_a="AUX_FIT", basis_type_b="AUX_FIT", &
sab_nl=sab_aux_fit_asymm, matrix_p=matrix_w_s)
CALL build_overlap_force(ks_env, admm_force, &
basis_type_a="AUX_FIT", basis_type_b="ORB", &
sab_nl=sab_aux_fit_vs_orb, matrix_p=matrix_w_q)
! add forces
CALL get_qs_env(qs_env=qs_env, atomic_kind_set=atomic_kind_set, &
force=force)
CALL add_qs_force(admm_force, force, "overlap_admm", atomic_kind_set)
DEALLOCATE (admm_force)
! *** Deallocated weighted density matrices
CALL dbcsr_deallocate_matrix(matrix_w_s)
CALL dbcsr_deallocate_matrix(matrix_w_q)
END DO
DO jspin = 1, 2
CALL cp_fm_release(tmp_aux_aux(jspin))
CALL cp_fm_release(tmp_aux_nao(jspin))
END DO
END SUBROUTINE rt_admm_forces_none
END MODULE rt_propagation_forces