GNU Octave 10.1.0
A high-level interpreted language, primarily intended for numerical computations, mostly compatible with Matlab
 
Loading...
Searching...
No Matches
DASRT.cc
Go to the documentation of this file.
1////////////////////////////////////////////////////////////////////////
2//
3// Copyright (C) 2002-2025 The Octave Project Developers
4//
5// See the file COPYRIGHT.md in the top-level directory of this
6// distribution or <https://octave.org/copyright/>.
7//
8// This file is part of Octave.
9//
10// Octave is free software: you can redistribute it and/or modify it
11// under the terms of the GNU General Public License as published by
12// the Free Software Foundation, either version 3 of the License, or
13// (at your option) any later version.
14//
15// Octave is distributed in the hope that it will be useful, but
16// WITHOUT ANY WARRANTY; without even the implied warranty of
17// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18// GNU General Public License for more details.
19//
20// You should have received a copy of the GNU General Public License
21// along with Octave; see the file COPYING. If not, see
22// <https://www.gnu.org/licenses/>.
23//
24////////////////////////////////////////////////////////////////////////
25
26#if defined (HAVE_CONFIG_H)
27# include "config.h"
28#endif
29
30#include <cinttypes>
31#include <sstream>
32
33#include "DASRT.h"
34#include "f77-fcn.h"
35#include "lo-error.h"
36#include "quit.h"
37
38typedef F77_INT (*dasrt_fcn_ptr) (const double&, const double *, const double *,
39 double *, F77_INT&, double *, F77_INT *);
40
41typedef F77_INT (*dasrt_jac_ptr) (const double&, const double *, const double *,
42 double *, const double&, double *, F77_INT *);
43
44typedef F77_INT (*dasrt_constr_ptr) (const F77_INT&, const double&,
45 const double *, const F77_INT&,
46 double *, double *, F77_INT *);
47
48extern "C"
49{
52 F77_DBLE *, F77_DBLE *, const F77_DBLE&, F77_INT *,
53 const F77_DBLE *, const F77_DBLE *, F77_INT&,
54 F77_DBLE *, const F77_INT&, F77_INT *,
55 const F77_INT&, F77_DBLE *, F77_INT *,
57 F77_INT *);
58}
59
60static DAEFunc::DAERHSFunc user_fsub;
61static DAEFunc::DAEJacFunc user_jsub;
62static DAERTFunc::DAERTConstrFunc user_csub;
63
64static F77_INT nn;
65
66static F77_INT
67ddasrt_f (const double& t, const double *state, const double *deriv,
68 double *delta, F77_INT& ires, double *, F77_INT *)
69{
70 ColumnVector tmp_state (nn);
71 ColumnVector tmp_deriv (nn);
72
73 for (F77_INT i = 0; i < nn; i++)
74 {
75 tmp_state(i) = state[i];
76 tmp_deriv(i) = deriv[i];
77 }
78
79 octave_idx_type tmp_ires = ires;
80
81 ColumnVector tmp_fval = (*user_fsub) (tmp_state, tmp_deriv, t, tmp_ires);
82
83 ires = octave::to_f77_int (tmp_ires);
84
85 if (tmp_fval.isempty ())
86 ires = -2;
87 else
88 {
89 for (F77_INT i = 0; i < nn; i++)
90 delta[i] = tmp_fval(i);
91 }
92
93 return 0;
94}
95
97ddasrt_j (const double& time, const double *state, const double *deriv,
98 double *pd, const double& cj, double *, F77_INT *)
99{
100 // FIXME: would be nice to avoid copying the data.
101
102 ColumnVector tmp_state (nn);
103 ColumnVector tmp_deriv (nn);
104
105 for (F77_INT i = 0; i < nn; i++)
106 {
107 tmp_deriv.elem (i) = deriv[i];
108 tmp_state.elem (i) = state[i];
109 }
110
111 Matrix tmp_pd = (*user_jsub) (tmp_state, tmp_deriv, time, cj);
112
113 for (F77_INT j = 0; j < nn; j++)
114 for (F77_INT i = 0; i < nn; i++)
115 pd[nn * j + i] = tmp_pd.elem (i, j);
116
117 return 0;
118}
119
120static F77_INT
121ddasrt_g (const F77_INT& neq, const double& t, const double *state,
122 const F77_INT& m_ng, double *gout, double *, F77_INT *)
123{
124 F77_INT n = neq;
125
126 ColumnVector tmp_state (n);
127 for (F77_INT i = 0; i < n; i++)
128 tmp_state(i) = state[i];
129
130 ColumnVector tmp_fval = (*user_csub) (tmp_state, t);
131
132 for (F77_INT i = 0; i < m_ng; i++)
133 gout[i] = tmp_fval(i);
134
135 return 0;
136}
137
138void
139DASRT::integrate (double tout)
140{
141 // I suppose this is the safe thing to do. If this is the first
142 // call, or if anything about the problem has changed, we should
143 // start completely fresh.
144
145 if (! m_initialized || m_restart
146 || DAEFunc::m_reset || DAERTFunc::m_reset || DASRT_options::m_reset)
147 {
148 m_integration_error = false;
149
150 m_initialized = true;
151
152 m_info.resize (dim_vector (15, 1));
153
154 for (F77_INT i = 0; i < 15; i++)
155 m_info(i) = 0;
156
157 F77_INT n = octave::to_f77_int (size ());
158
159 nn = n;
160
161 // DAERTFunc
162
163 user_csub = DAERTFunc::constraint_function ();
164
165 if (user_csub)
166 {
167 ColumnVector tmp = (*user_csub) (m_x, m_t);
168 m_ng = octave::to_f77_int (tmp.numel ());
169 }
170 else
171 m_ng = 0;
172
173 F77_INT maxord = octave::to_f77_int (maximum_order ());
174 if (maxord >= 0)
175 {
176 if (maxord > 0 && maxord < 6)
177 {
178 m_info(8) = 1;
179 m_iwork(2) = maxord;
180 }
181 else
182 {
183 (*current_liboctave_error_handler)
184 ("dassl: invalid value for maximum order");
185 m_integration_error = true;
186 return;
187 }
188 }
189
190 m_liw = 21 + n;
191 m_lrw = 50 + 9*n + n*n + 3*m_ng;
192
193 m_iwork.resize (dim_vector (m_liw, 1));
194 m_rwork.resize (dim_vector (m_lrw, 1));
195
196 m_info(0) = 0;
197
198 if (m_stop_time_set)
199 {
200 m_info(3) = 1;
201 m_rwork(0) = m_stop_time;
202 }
203 else
204 m_info(3) = 0;
205
206 m_restart = false;
207
208 // DAEFunc
209
210 user_fsub = DAEFunc::function ();
211 user_jsub = DAEFunc::jacobian_function ();
212
213 if (user_fsub)
214 {
215 octave_idx_type ires = 0;
216
217 ColumnVector fval = (*user_fsub) (m_x, m_xdot, m_t, ires);
218
219 if (fval.numel () != m_x.numel ())
220 {
221 (*current_liboctave_error_handler)
222 ("dasrt: inconsistent sizes for state and residual vectors");
223
224 m_integration_error = true;
225 return;
226 }
227 }
228 else
229 {
230 (*current_liboctave_error_handler)
231 ("dasrt: no user supplied RHS subroutine!");
232
233 m_integration_error = true;
234 return;
235 }
236
237 m_info(4) = (user_jsub ? 1 : 0);
238
239 DAEFunc::m_reset = false;
240
241 m_jroot.resize (dim_vector (m_ng, 1), 1);
242
243 DAERTFunc::m_reset = false;
244
245 // DASRT_options
246
247 double mss = maximum_step_size ();
248 if (mss >= 0.0)
249 {
250 m_rwork(1) = mss;
251 m_info(6) = 1;
252 }
253 else
254 m_info(6) = 0;
255
256 double iss = initial_step_size ();
257 if (iss >= 0.0)
258 {
259 m_rwork(2) = iss;
260 m_info(7) = 1;
261 }
262 else
263 m_info(7) = 0;
264
265 F77_INT sl = octave::to_f77_int (step_limit ());
266 if (sl >= 0)
267 {
268 m_info(11) = 1;
269 m_iwork(20) = sl;
270 }
271 else
272 m_info(11) = 0;
273
274 m_abs_tol = absolute_tolerance ();
275 m_rel_tol = relative_tolerance ();
276
277 F77_INT abs_tol_len = octave::to_f77_int (m_abs_tol.numel ());
278 F77_INT rel_tol_len = octave::to_f77_int (m_rel_tol.numel ());
279
280 if (abs_tol_len == 1 && rel_tol_len == 1)
281 {
282 m_info.elem (1) = 0;
283 }
284 else if (abs_tol_len == n && rel_tol_len == n)
285 {
286 m_info.elem (1) = 1;
287 }
288 else
289 {
290 (*current_liboctave_error_handler)
291 ("dasrt: inconsistent sizes for tolerance arrays");
292
293 m_integration_error = true;
294 return;
295 }
296
297 DASRT_options::m_reset = false;
298 }
299
300 double *px = m_x.rwdata ();
301 double *pxdot = m_xdot.rwdata ();
302
303 F77_INT *pinfo = m_info.rwdata ();
304
305 double *prel_tol = m_rel_tol.rwdata ();
306 double *pabs_tol = m_abs_tol.rwdata ();
307
308 double *prwork = m_rwork.rwdata ();
309 F77_INT *piwork = m_iwork.rwdata ();
310
311 F77_INT *pjroot = m_jroot.rwdata ();
312
313 double *dummy = nullptr;
314 F77_INT *idummy = nullptr;
315
316 F77_INT tmp_istate = octave::to_f77_int (m_istate);
317
318 F77_XFCN (ddasrt, DDASRT, (ddasrt_f, nn, m_t, px, pxdot, tout, pinfo,
319 prel_tol, pabs_tol, tmp_istate, prwork, m_lrw,
320 piwork, m_liw, dummy, idummy, ddasrt_j,
321 ddasrt_g, m_ng, pjroot));
322
323 m_istate = tmp_istate;
324
325 switch (m_istate)
326 {
327 case 1: // A step was successfully taken in intermediate-output
328 // mode. The code has not yet reached TOUT.
329 case 2: // The integration to TOUT was successfully completed
330 // (T=TOUT) by stepping exactly to TOUT.
331 case 3: // The integration to TOUT was successfully completed
332 // (T=TOUT) by stepping past TOUT. Y(*) is obtained by
333 // interpolation. YPRIME(*) is obtained by interpolation.
334 m_t = tout;
335 break;
336
337 case 4: // The integration was successfully completed
338 // by finding one or more roots of G at T.
339 break;
340
341 case -1: // A large amount of work has been expended.
342 case -2: // The error tolerances are too stringent.
343 case -3: // The local error test cannot be satisfied because you
344 // specified a zero component in ATOL and the
345 // corresponding computed solution component is zero.
346 // Thus, a pure relative error test is impossible for
347 // this component.
348 case -6: // DDASRT had repeated error test failures on the last
349 // attempted step.
350 case -7: // The corrector could not converge.
351 case -8: // The matrix of partial derivatives is singular.
352 case -9: // The corrector could not converge. There were repeated
353 // error test failures in this step.
354 case -10: // The corrector could not converge because IRES was
355 // equal to minus one.
356 case -11: // IRES equal to -2 was encountered and control is being
357 // returned to the calling program.
358 case -12: // DASSL failed to compute the initial YPRIME.
359 case -33: // The code has encountered trouble from which it cannot
360 // recover. A message is printed explaining the trouble
361 // and control is returned to the calling program. For
362 // example, this occurs when invalid input is detected.
363 m_integration_error = true;
364 break;
365
366 default:
367 m_integration_error = true;
368 (*current_liboctave_error_handler)
369 ("unrecognized value of istate (= %" OCTAVE_IDX_TYPE_FORMAT ") "
370 "returned from ddasrt", m_istate);
371 break;
372 }
373}
374
377{
378 DASRT_result retval;
379
380 Matrix x_out;
381 Matrix xdot_out;
382 ColumnVector t_out = tout;
383
384 octave_idx_type n_out = tout.numel ();
385 F77_INT n = octave::to_f77_int (size ());
386
387 if (n_out > 0 && n > 0)
388 {
389 x_out.resize (n_out, n);
390 xdot_out.resize (n_out, n);
391
392 for (F77_INT i = 0; i < n; i++)
393 {
394 x_out(0, i) = m_x(i);
395 xdot_out(0, i) = m_xdot(i);
396 }
397
398 for (octave_idx_type j = 1; j < n_out; j++)
399 {
400 integrate (tout(j));
401
403 {
404 retval = DASRT_result (x_out, xdot_out, t_out);
405 return retval;
406 }
407
408 if (m_istate == 4)
409 t_out(j) = m_t;
410 else
411 t_out(j) = tout(j);
412
413 for (F77_INT i = 0; i < n; i++)
414 {
415 x_out(j, i) = m_x(i);
416 xdot_out(j, i) = m_xdot(i);
417 }
418
419 if (m_istate == 4)
420 {
421 x_out.resize (j+1, n);
422 xdot_out.resize (j+1, n);
423 t_out.resize (j+1);
424 break;
425 }
426 }
427 }
428
429 retval = DASRT_result (x_out, xdot_out, t_out);
430
431 return retval;
432}
433
435DASRT::integrate (const ColumnVector& tout, const ColumnVector& tcrit)
436{
437 DASRT_result retval;
438
439 Matrix x_out;
440 Matrix xdot_out;
441 ColumnVector t_outs = tout;
442
443 octave_idx_type n_out = tout.numel ();
444 F77_INT n = octave::to_f77_int (size ());
445
446 if (n_out > 0 && n > 0)
447 {
448 x_out.resize (n_out, n);
449 xdot_out.resize (n_out, n);
450
451 octave_idx_type n_crit = tcrit.numel ();
452
453 if (n_crit > 0)
454 {
455 octave_idx_type i_crit = 0;
456 octave_idx_type i_out = 1;
457 double next_crit = tcrit(0);
458 double next_out;
459 while (i_out < n_out)
460 {
461 bool do_restart = false;
462
463 next_out = tout(i_out);
464 if (i_crit < n_crit)
465 next_crit = tcrit(i_crit);
466
467 bool save_output = false;
468 double t_out;
469
470 if (next_crit == next_out)
471 {
472 set_stop_time (next_crit);
473 t_out = next_out;
474 save_output = true;
475 i_out++;
476 i_crit++;
477 do_restart = true;
478 }
479 else if (next_crit < next_out)
480 {
481 if (i_crit < n_crit)
482 {
483 set_stop_time (next_crit);
484 t_out = next_crit;
485 save_output = false;
486 i_crit++;
487 do_restart = true;
488 }
489 else
490 {
492 t_out = next_out;
493 save_output = true;
494 i_out++;
495 }
496 }
497 else
498 {
499 set_stop_time (next_crit);
500 t_out = next_out;
501 save_output = true;
502 i_out++;
503 }
504
505 integrate (t_out);
506
508 {
509 retval = DASRT_result (x_out, xdot_out, t_outs);
510 return retval;
511 }
512
513 if (m_istate == 4)
514 t_out = m_t;
515
516 if (save_output)
517 {
518 for (F77_INT i = 0; i < n; i++)
519 {
520 x_out(i_out-1, i) = m_x(i);
521 xdot_out(i_out-1, i) = m_xdot(i);
522 }
523
524 t_outs(i_out-1) = t_out;
525
526 if (m_istate == 4)
527 {
528 x_out.resize (i_out, n);
529 xdot_out.resize (i_out, n);
530 t_outs.resize (i_out);
531 i_out = n_out;
532 }
533 }
534
535 if (do_restart)
536 force_restart ();
537 }
538
539 retval = DASRT_result (x_out, xdot_out, t_outs);
540 }
541 else
542 {
543 retval = integrate (tout);
544
546 return retval;
547 }
548 }
549
550 return retval;
551}
552
553std::string
555{
556 std::string retval;
557
558 std::ostringstream buf;
559 buf << m_t;
560 std::string t_curr = buf.str ();
561
562 switch (m_istate)
563 {
564 case 1:
565 retval = "a step was successfully taken in intermediate-output mode.";
566 break;
567
568 case 2:
569 retval = "integration completed by stepping exactly to TOUT";
570 break;
571
572 case 3:
573 retval = "integration to tout completed by stepping past TOUT";
574 break;
575
576 case 4:
577 retval = "integration completed by finding one or more roots of G at T";
578 break;
579
580 case -1:
581 retval = "a large amount of work has been expended (t =" + t_curr + ')';
582 break;
583
584 case -2:
585 retval = "the error tolerances are too stringent";
586 break;
587
588 case -3:
589 retval = "error weight became zero during problem. (t = " + t_curr +
590 "; solution component i vanished, and atol or atol(i) == 0)";
591 break;
592
593 case -6:
594 retval = "repeated error test failures on the last attempted step (t = "
595 + t_curr + ')';
596 break;
597
598 case -7:
599 retval = "the corrector could not converge (t = " + t_curr + ')';
600 break;
601
602 case -8:
603 retval = "the matrix of partial derivatives is singular (t = " + t_curr +
604 ')';
605 break;
606
607 case -9:
608 retval = "the corrector could not converge (t = " + t_curr +
609 "; repeated test failures)";
610 break;
611
612 case -10:
613 retval = "corrector could not converge because IRES was -1 (t = "
614 + t_curr + ')';
615 break;
616
617 case -11:
618 retval = "return requested in user-supplied function (t = " + t_curr +
619 ')';
620 break;
621
622 case -12:
623 retval = "failed to compute consistent initial conditions";
624 break;
625
626 case -33:
627 retval = "unrecoverable error (see printed message)";
628 break;
629
630 default:
631 retval = "unknown error state";
632 break;
633 }
634
635 return retval;
636}
F77_INT(* dasrt_jac_ptr)(const double &, const double *, const double *, double *, const double &, double *, F77_INT *)
Definition DASRT.cc:41
F77_RET_T F77_FUNC(ddasrt, DDASRT)(dasrt_fcn_ptr
F77_INT(* dasrt_fcn_ptr)(const double &, const double *, const double *, double *, F77_INT &, double *, F77_INT *)
Definition DASRT.cc:38
F77_INT(* dasrt_constr_ptr)(const F77_INT &, const double &, const double *, const F77_INT &, double *, double *, F77_INT *)
Definition DASRT.cc:44
F77_INT ddasrt_j(const double &time, const double *state, const double *deriv, double *pd, const double &cj, double *, F77_INT *)
Definition DASRT.cc:97
T & elem(octave_idx_type n)
Size of the specified dimension.
Definition Array.h:563
void resize(const dim_vector &dv, const T &rfv)
Size of the specified dimension.
bool isempty() const
Size of the specified dimension.
Definition Array.h:652
T * rwdata()
Size of the specified dimension.
octave_idx_type numel() const
Number of elements in the array.
Definition Array.h:418
void resize(octave_idx_type n, const double &rfv=0)
Definition dColVector.h:112
DAEJacFunc jacobian_function() const
Definition DAEFunc.h:83
Matrix(* DAEJacFunc)(const ColumnVector &x, const ColumnVector &xdot, double t, double cj)
Definition DAEFunc.h:45
bool m_reset
Definition DAEFunc.h:102
DAERHSFunc function() const
Definition DAEFunc.h:74
ColumnVector(* DAERHSFunc)(const ColumnVector &x, const ColumnVector &xdot, double t, octave_idx_type &ires)
Definition DAEFunc.h:37
ColumnVector(* DAERTConstrFunc)(const ColumnVector &x, double t)
Definition DAERTFunc.h:37
DAERTConstrFunc constraint_function() const
Definition DAERTFunc.h:70
bool m_reset
Definition DAERTFunc.h:88
DASRT_result integrate(const ColumnVector &tout)
Definition DASRT.cc:376
std::string error_message() const
Definition DASRT.cc:554
void resize(octave_idx_type nr, octave_idx_type nc, double rfv=0)
Definition dMatrix.h:156
ColumnVector m_xdot
Definition base-dae.h:79
double m_stop_time
Definition base-de.h:111
bool m_restart
Definition base-de.h:115
double m_t
Definition base-de.h:109
octave_idx_type m_istate
Definition base-de.h:119
virtual void force_restart()
Definition base-de.h:97
void clear_stop_time()
Definition base-de.h:91
bool m_stop_time_set
Definition base-de.h:113
octave_idx_type size() const
Definition base-de.h:78
bool m_integration_error
Definition base-de.h:117
ColumnVector m_x
Definition base-de.h:107
void set_stop_time(double tt)
Definition base-de.h:84
Vector representing the dimensions (size) of an Array.
Definition dim-vector.h:90
subroutine ddasrt(res, neq, t, y, yprime, tout, info, rtol, atol, idid, rwork, lrw, iwork, liw, rpar, ipar, jac, g, ng, jroot)
Definition ddasrt.f:4
#define F77_XFCN(f, F, args)
Definition f77-fcn.h:45
double F77_DBLE
Definition f77-fcn.h:302
octave_f77_int_type F77_INT
Definition f77-fcn.h:306
F77_RET_T(F77_CONST_CHAR_ARG_DECL, F77_CONST_CHAR_ARG_DECL, F77_CONST_CHAR_ARG_DECL, const F77_INT &, const F77_INT &, const F77_INT &, F77_INT &, F77_INT &, F77_DBLE *, const F77_INT &, F77_DBLE *, const F77_INT &, F77_DBLE *, F77_DBLE *, F77_DBLE *, const F77_INT &, F77_DBLE *, const F77_INT &, F77_DBLE *, const F77_INT &, F77_DBLE *, F77_INT *, F77_INT &F77_CHAR_ARG_LEN_DECL F77_CHAR_ARG_LEN_DECL F77_CHAR_ARG_LEN_DECL)