Home

Context Navigation

← Previous Revision
Latest Revision
Next Revision →
Blame
Revision Log

ros4_ddm.f @ 2696

Last change on this file since 2696 was 2696, checked in by kanani, 6 years ago
Merge of branch palm4u into trunk
File size: 21.0 KB

Line
1	SUBROUTINE INTEGRATE( NSENSIT, Y, TIN, TOUT )
2
3	INCLUDE 'KPP_ROOT_params.h'
4	INCLUDE 'KPP_ROOT_global.h'
5
6	C TIN - Start Time
7	KPP_REAL TIN
8	C TOUT - End Time
9	KPP_REAL TOUT
10	C Y - Concentrations and Sensitivities
11	KPP_REAL Y(NVAR*(NSENSIT+1))
12	C --- Note: Y contains: (1:NVAR) concentrations, followed by
13	C --- (1:NVAR) sensitivities w.r.t. first parameter, followed by
14	C --- etc., followed by
15	C --- (1:NVAR) sensitivities w.r.t. NSENSIT's parameter
16
17	INTEGER INFO(5)
18
19	EXTERNAL FUNC_CHEM, JAC_CHEM
20
21	INFO(1) = Autonomous
22
23	CALL ROS4_DDM(NVAR,NSENSIT,TIN,TOUT,STEPMIN,STEPMAX,
24	+ STEPMIN,Y,ATOL,RTOL,
25	+ Info,FUNC_CHEM,JAC_CHEM)
26
27
28	RETURN
29	END
30
31
32
33
34	SUBROUTINE ROS4_DDM(N,NSENSIT,T,Tnext,Hmin,Hmax,Hstart,
35	+ y,AbsTol,RelTol,
36	+ Info,FUNC_CHEM,JAC_CHEM)
37	IMPLICIT NONE
38	INCLUDE 'KPP_ROOT_params.h'
39	INCLUDE 'KPP_ROOT_global.h'
40	INCLUDE 'KPP_ROOT_sparse.h'
41	C
42	C Four Stages, Fourth Order L-stable Rosenbrock Method,
43	C with embedded L-stable, third order method for error control
44	C Simplified version of E. Hairer's atmros4; the coefficients are slightly different
45	C
46	C Direct decoupled computation of sensitivities.
47	C The global variable DDMTYPE distinguishes between:
48	C DDMTYPE = 0 : sensitivities w.r.t. initial values
49	C DDMTYPE = 1 : sensitivities w.r.t. parameters
50	C
51	C INPUT ARGUMENTS:
52	C y = Vector of: (1:NVAR) concentrations, followed by
53	C (1:NVAR) sensitivities w.r.t. first parameter, followed by
54	C etc., followed by
55	C (1:NVAR) sensitivities w.r.t. NSENSIT's parameter
56	C (y contains initial values at input, final values at output)
57	C [T, Tnext] = the integration interval
58	C Hmin, Hmax = lower and upper bounds for the selected step-size.
59	C Note that for Step = Hmin the current computed
60	C solution is unconditionally accepted by the error
61	C control mechanism.
62	C AbsTol, RelTol = (NVAR) dimensional vectors of
63	C componentwise absolute and relative tolerances.
64	C FUNC_CHEM = name of routine of derivatives. KPP syntax.
65	C See the header below.
66	C JAC_CHEM = name of routine that computes the Jacobian, in
67	C sparse format. KPP syntax. See the header below.
68	C Info(1) = 1 for Autonomous system
69	C = 0 for nonAutonomous system
70	C
71	C OUTPUT ARGUMENTS:
72	C y = the values of concentrations and sensitivities at Tend.
73	C T = equals TENDon output.
74	C Info(2) = # of FUNC_CHEM CALLs.
75	C Info(3) = # of JAC_CHEM CALLs.
76	C Info(4) = # of accepted steps.
77	C Info(5) = # of rejected steps.
78	C
79	C Adrian Sandu, December 2001
80	C
81
82
83	INTEGER NSENSIT
84	KPP_REAL y(NVAR(NSENSIT+1)), ynew(NVAR(NSENSIT+1))
85	KPP_REAL K1(NVAR*(NSENSIT+1))
86	KPP_REAL K2(NVAR*(NSENSIT+1))
87	KPP_REAL K3(NVAR*(NSENSIT+1))
88	KPP_REAL K4(NVAR*(NSENSIT+1))
89	KPP_REAL Fv(NVAR), Hv(NVAR)
90	KPP_REAL DFDT(NVAR*(NSENSIT+1))
91	KPP_REAL DFDP(NVARNSENSIT), DFDPDT(NVARNSENSIT)
92	KPP_REAL DJDP(NVAR*NSENSIT)
93	KPP_REAL JAC(LU_NONZERO), AJAC(LU_NONZERO)
94	KPP_REAL DJDT(LU_NONZERO)
95	KPP_REAL HESS(NHESS)
96	KPP_REAL Hmin,Hmax,Hstart,ghinv,uround
97	KPP_REAL AbsTol(NVAR), RelTol(NVAR)
98	KPP_REAL T, Tnext, Tplus, H, Hnew, elo
99	KPP_REAL ERR, factor, facmax, dround, tau
100	KPP_REAL w, e, beta1, beta2, beta3, beta4
101
102	INTEGER n,nfcn,njac,Naccept,Nreject,i,j,ier
103	INTEGER Info(5)
104	LOGICAL IsReject, Autonomous
105	EXTERNAL FUNC_CHEM, JAC_CHEM, HESS_CHEM
106
107
108	C The method coefficients
109	C DOUBLE PRECISION gamma, gamma2, gamma3, gamma4
110	C PARAMETER ( gamma = 0.57281606D0 )
111	C PARAMETER ( gamma2 = -1.769177067112013949170520D0 )
112	C PARAMETER ( gamma3 = 0.759293964293209853670967D0 )
113	C PARAMETER ( gamma4 = -0.104894621490955803206743D0 )
114	C DOUBLE PRECISION a21, a31, a32, a41, a42, a43
115	C PARAMETER ( a21 = 2.00000000000000000000000D0 )
116	C PARAMETER ( a31 = 1.86794814949823713234476D0 )
117	C PARAMETER ( a32 = 0.23444556851723885002322D0 )
118	C DOUBLE PRECISION alpha2, alpha3, alpha4
119	C PARAMETER ( alpha2 = 1.145632120D0 )
120	C PARAMETER ( alpha3 = 0.655214975973133829477748D0 )
121	C DOUBLE PRECISION c21, c31, c32, c41, c42, c43
122	C PARAMETER ( c21 = -7.137649943349979830369260D0 )
123	C PARAMETER ( c31 = 2.580923666509657714488050D0 )
124	C PARAMETER ( c32 = 0.651629887302032023387417D0 )
125	C PARAMETER ( c41 = -2.137115266506619116806370D0 )
126	C PARAMETER ( c42 = -0.321469531339951070769241D0 )
127	C PARAMETER ( c43 = -0.694966049282445225157329D0 )
128	C DOUBLE PRECISION m1, m2, m3, m4, mhat1, mhat2, mhat3, mhat4
129	C PARAMETER ( m1 = 2.255566228604565243728840D0 )
130	C PARAMETER ( m2 = 0.287055063194157607662630D0 )
131	C PARAMETER ( m3 = 0.435311963379983213402707D0 )
132	C PARAMETER ( m4 = 1.093507656403247803214820D0 )
133	C PARAMETER ( mhat1 = 2.068399160527583734258670D0 )
134	C PARAMETER ( mhat2 = 0.238681352067532797956493D0 )
135	C PARAMETER ( mhat3 = 0.363373345435391708261747D0 )
136	C PARAMETER ( mhat4 = 0.366557127936155144309163D0 )
137	C DOUBLE PRECISION e1, e2, e3, e4
138	c PARAMETER ( e1 = 1.8716706807698191283861888D-01 )
139	c PARAMETER ( e2 = 4.8373711126624835410225955D-02 )
140	c PARAMETER ( e3 = 7.1938617944591554120847832D-02 )
141	c PARAMETER ( e4 = 7.2695052846709262706070831D-01 )
142	C PARAMETER ( e1 = -0.2815431932141155D+00 )
143	C PARAMETER ( e2 = -0.7276199124938920D-01 )
144	C PARAMETER ( e3 = -0.1082196201495311D+00 )
145	C PARAMETER ( e4 = -0.1093502252409163D+01 )
146	C The method coefficients
147	DOUBLE PRECISION gamma, gamma2, gamma3, gamma4
148	PARAMETER ( gamma = 0.5728200000000000D+00 )
149	PARAMETER ( gamma2 = -0.1769193891319233D+01 )
150	PARAMETER ( gamma3 = 0.7592633437920482D+00 )
151	PARAMETER ( gamma4 = -0.1049021087100450D+00 )
152	DOUBLE PRECISION a21, a31, a32, a41, a42, a43
153	PARAMETER ( a21 = 0.2000000000000000D+01 )
154	PARAMETER ( a31 = 0.1867943637803922D+01 )
155	PARAMETER ( a32 = 0.2344449711399156D+00 )
156	DOUBLE PRECISION alpha2, alpha3
157	PARAMETER ( alpha2 = 0.1145640000000000D+01 )
158	PARAMETER ( alpha3 = 0.6552168638155900D+00 )
159	DOUBLE PRECISION c21, c31, c32, c41, c42, c43
160	PARAMETER ( c21 = -0.7137615036412310D+01 )
161	PARAMETER ( c31 = 0.2580708087951457D+01 )
162	PARAMETER ( c32 = 0.6515950076447975D+00 )
163	PARAMETER ( c41 = -0.2137148994382534D+01 )
164	PARAMETER ( c42 = -0.3214669691237626D+00 )
165	PARAMETER ( c43 = -0.6949742501781779D+00 )
166	DOUBLE PRECISION b1, b2, b3, b4
167	PARAMETER ( b1 = 0.2255570073418735D+01 )
168	PARAMETER ( b2 = 0.2870493262186792D+00 )
169	PARAMETER ( b3 = 0.4353179431840180D+00 )
170	PARAMETER ( b4 = 0.1093502252409163D+01 )
171	DOUBLE PRECISION d1, d2, d3, d4
172	PARAMETER ( d1 = -0.2815431932141155D+00 )
173	PARAMETER ( d2 = -0.7276199124938920D-01 )
174	PARAMETER ( d3 = -0.1082196201495311D+00 )
175	PARAMETER ( d4 = -0.1093502252409163D+01 )
176
177
178	c Initialization of counters, etc.
179	Autonomous = Info(1) .EQ. 1
180	uround = 1.d-15
181	dround = DSQRT(uround)
182	IF (Hmax.le.0.D0) THEN
183	Hmax = DABS(Tnext-T)
184	END IF
185	H = DMAX1(1.d-8, Hstart)
186	Tplus = T
187	IsReject = .false.
188	Naccept = 0
189	Nreject = 0
190	Nfcn = 0
191	Njac = 0
192
193	C === Starting the time loop ===
194	10 CONTINUE
195
196	Tplus = T + H
197	IF ( Tplus .gt. Tnext ) THEN
198	H = Tnext - T
199	Tplus = Tnext
200	END IF
201
202	C Initial Function, Jacobian, and Hessian Values
203	CALL FUNC_CHEM(NVAR, T, y, Fv)
204	CALL JAC_CHEM(NVAR, T, y, JAC)
205	CALL HESS_CHEM( NVAR, T, y, HESS )
206	IF (DDMTYPE .EQ. 1) THEN
207	CALL DFUNDPAR(NVAR, NSENSIT, T, y, DFDP)
208	END IF
209
210	C The time derivatives for non-Autonomous case
211	IF (.not. Autonomous) THEN
212	tau = DSIGN(dround*DMAX1( 1.0d0, DABS(T) ), T)
213	CALL FUNC_CHEM(NVAR, T+tau, y, K2)
214	CALL JAC_CHEM(NVAR, T+tau, y, AJAC)
215	nfcn=nfcn+1
216	DO 20 j = 1,NVAR
217	DFDT(j) = ( K2(j)-Fv(j) )/tau
218	20 CONTINUE
219	DO 30 j = 1,LU_NONZERO
220	DJDT(j) = ( AJAC(j)-JAC(j) )/tau
221	30 CONTINUE
222	DO 35 i=1,NSENSIT
223	CALL Jac_SP_Vec (DJDT,y(iNVAR+1),DFDT(iNVAR+1))
224	35 CONTINUE
225	END IF
226
227	11 CONTINUE ! From here we restart after a rejected step
228
229	C Form the Prediction matrix and compute its LU factorization
230	Njac = Njac+1
231	ghinv = 1.0d0/(gamma*H)
232	DO 40 j=1,LU_NONZERO
233	AJAC(j) = -JAC(j)
234	40 CONTINUE
235	DO 50 j=1,NVAR
236	AJAC(LU_DIAG(j)) = AJAC(LU_DIAG(j)) + ghinv
237	50 CONTINUE
238	CALL KppDecomp (AJAC, ier)
239	C
240	IF (ier.ne.0) THEN
241	IF ( H.gt.Hmin) THEN
242	H = 5.0d-1*H
243	GO TO 10
244	ELSE
245	PRINT *,'ROS4: Singular factorization at T=',T,'; H=',H
246	STOP
247	END IF
248	END IF
249
250	C ------------ STAGE 1-------------------------
251	DO 60 j = 1,NVAR
252	K1(j) = Fv(j)
253	60 CONTINUE
254	IF (.NOT. Autonomous) THEN
255	beta1 = H*gamma
256	DO 70 j=1,NVAR
257	K1(j) = K1(j) + beta1*DFDT(j)
258	70 CONTINUE
259	END IF
260	CALL KppSolve (AJAC, K1)
261	C --- If derivative w.r.t. parameters
262	IF (DDMTYPE .EQ. 1) THEN
263	CALL DJACDPAR(NVAR, NSENSIT, T, y, K1(1), DJDP)
264	END IF
265	C --- End of derivative w.r.t. parameters
266
267	DO 100 i=1,NSENSIT
268	CALL Jac_SP_Vec (JAC,y(iNVAR+1),K1(iNVAR+1))
269	CALL Hess_Vec ( HESS, y(i*NVAR+1), K1(1), Hv )
270	DO 80 j=1,NVAR
271	K1(iNVAR+j) = K1(iNVAR+j) + Hv(j)
272	80 CONTINUE
273	IF (.NOT. Autonomous) THEN
274	DO 90 j=1,NVAR
275	K1(iNVAR+j) = K1(iNVAR+j) + beta1DFDT(iNVAR+j)
276	90 CONTINUE
277	END IF
278	C --- If derivative w.r.t. parameters
279	IF (DDMTYPE .EQ. 1) THEN
280	DO 95 j = 1,NVAR
281	K1(iNVAR+j) = K1(iNVAR+j) + DFDP((i-1)*NVAR+j)
282	& + DJDP((i-1)*NVAR+j)
283	95 CONTINUE
284	END IF
285	C --- End of derivative w.r.t. parameters
286	CALL KppSolve (AJAC, K1(i*NVAR+1))
287	100 CONTINUE
288
289	C ----------- STAGE 2 -------------------------
290	DO 110 j = 1,NVAR*(NSENSIT+1)
291	ynew(j) = y(j) + a21*K1(j)
292	110 CONTINUE
293	CALL FUNC_CHEM(NVAR, T+alpha2*H, ynew, Fv)
294	IF (DDMTYPE .EQ. 1) THEN
295	CALL DFUNDPAR(NVAR, NSENSIT, T+alpha2*H, ynew, DFDP)
296	END IF
297	nfcn=nfcn+1
298	beta1 = c21/H
299	DO 120 j = 1,NVAR
300	K2(j) = Fv(j) + beta1*K1(j)
301	120 CONTINUE
302	IF (.NOT. Autonomous) THEN
303	beta2 = H*gamma2
304	DO 130 j=1,NVAR
305	K2(j) = K2(j) + beta2*DFDT(j)
306	130 CONTINUE
307	END IF
308	CALL KppSolve (AJAC, K2)
309	C --- If derivative w.r.t. parameters
310	IF (DDMTYPE .EQ. 1) THEN
311	CALL DJACDPAR(NVAR, NSENSIT, T, y, K2(1), DJDP)
312	END IF
313	C --- End of derivative w.r.t. parameters
314
315	CALL JAC_CHEM(NVAR, T+alpha2*H, ynew, JAC)
316	njac=njac+1
317	DO 160 i=1,NSENSIT
318	CALL Jac_SP_Vec (JAC,ynew(iNVAR+1),K2(iNVAR+1))
319	CALL Hess_Vec ( HESS, y(i*NVAR+1), K2(1), Hv )
320	DO 140 j = 1,NVAR
321	K2(iNVAR+j) = K2(iNVAR+j) + beta1K1(iNVAR+j)
322	& + Hv(j)
323	140 CONTINUE
324	IF (.NOT. Autonomous) THEN
325	DO 150 j=1,NVAR
326	K2(iNVAR+j) = K2(iNVAR+j) + beta2DFDT(iNVAR+j)
327	150 CONTINUE
328	END IF
329	C --- If derivative w.r.t. parameters
330	IF (DDMTYPE .EQ. 1) THEN
331	DO 155 j = 1,NVAR
332	K2(iNVAR+j) = K2(iNVAR+j) + DFDP((i-1)*NVAR+j)
333	& + DJDP((i-1)*NVAR+j)
334	155 CONTINUE
335	END IF
336	C --- End of derivative w.r.t. parameters
337	CALL KppSolve (AJAC, K2(i*NVAR+1))
338	160 CONTINUE
339
340
341	C ------------ STAGE 3 -------------------------
342	DO 170 j = 1,NVAR*(NSENSIT+1)
343	ynew(j) = y(j) + a31K1(j) + a32K2(j)
344	170 CONTINUE
345	CALL FUNC_CHEM(NVAR, T+alpha3*H, ynew, Fv)
346	IF (DDMTYPE .EQ. 1) THEN
347	CALL DFUNDPAR(NVAR, NSENSIT, T+alpha3*H, ynew, DFDP)
348	END IF
349	nfcn=nfcn+1
350	beta1 = c31/H
351	beta2 = c32/H
352	DO 180 j = 1,NVAR
353	K3(j) = Fv(j) + beta1K1(j) + beta2K2(j)
354	180 CONTINUE
355	IF (.NOT. Autonomous) THEN
356	beta3 = H*gamma3
357	DO 190 j=1,NVAR
358	K3(j) = K3(j) + beta3*DFDT(j)
359	190 CONTINUE
360	END IF
361	CALL KppSolve (AJAC, K3)
362	C --- If derivative w.r.t. parameters
363	IF (DDMTYPE .EQ. 1) THEN
364	CALL DJACDPAR(NVAR, NSENSIT, T, y, K3(1), DJDP)
365	END IF
366	C --- End of derivative w.r.t. parameters
367
368	CALL JAC_CHEM(NVAR, T+alpha3*H, ynew, JAC)
369	njac=njac+1
370	DO 220 i=1,NSENSIT
371	CALL Jac_SP_Vec (JAC,ynew(iNVAR+1),K3(iNVAR+1))
372	CALL Hess_Vec ( HESS, y(i*NVAR+1), K3(1), Hv )
373	DO 200 j = 1,NVAR
374	K3(iNVAR+j) = K3(iNVAR+j) + beta1K1(iNVAR+j)
375	& + beta2K2(iNVAR+j) + Hv(j)
376	200 CONTINUE
377	IF (.NOT. Autonomous) THEN
378	DO 210 j=1,NVAR
379	K3(iNVAR+j) = K3(iNVAR+j) + beta3DFDT(iNVAR+j)
380	210 CONTINUE
381	END IF
382	C --- If derivative w.r.t. parameters
383	IF (DDMTYPE .EQ. 1) THEN
384	DO 215 j = 1,NVAR
385	K3(iNVAR+j) = K3(iNVAR+j) + DFDP((i-1)*NVAR+j)
386	& + DJDP((i-1)*NVAR+j)
387	215 CONTINUE
388	END IF
389	C --- End of derivative w.r.t. parameters
390	CALL KppSolve (AJAC, K3(i*NVAR+1))
391	220 CONTINUE
392
393	C ------------ STAGE 4 -------------------------
394	C Note: uses the same function values as stage 3
395	beta1 = c41/H
396	beta2 = c42/H
397	beta3 = c43/H
398	DO 230 j = 1,NVAR
399	K4(j) = Fv(j) + beta1K1(j) + beta2K2(j) + beta3*K3(j)
400	230 CONTINUE
401	IF (.NOT. Autonomous) THEN
402	beta4 = H*gamma4
403	DO 240 j=1,NVAR
404	K4(j) = K4(j) + beta4*DFDT(j)
405	240 CONTINUE
406	END IF
407	CALL KppSolve (AJAC, K4)
408	C --- If derivative w.r.t. parameters
409	IF (DDMTYPE .EQ. 1) THEN
410	CALL DJACDPAR(NVAR, NSENSIT, T, y, K4(1), DJDP)
411	END IF
412	C --- End of derivative w.r.t. parameters
413
414	njac=njac+1
415	DO 270 i=1,NSENSIT
416	CALL Jac_SP_Vec (JAC,ynew(iNVAR+1),K4(iNVAR+1))
417	CALL Hess_Vec ( HESS, y(i*NVAR+1), K4(1), Hv )
418	DO 250 j = 1,NVAR
419	K4(iNVAR+j) = K4(iNVAR+j) + beta1K1(iNVAR+j)
420	& + beta2K2(iNVAR+j) + beta3K3(iNVAR+j)
421	& + Hv(j)
422	250 CONTINUE
423	IF (.NOT. Autonomous) THEN
424	DO 260 j=1,NVAR
425	K4(iNVAR+j) = K4(iNVAR+j) + beta4DFDT(iNVAR+j)
426	260 CONTINUE
427	END IF
428	C --- If derivative w.r.t. parameters
429	IF (DDMTYPE .EQ. 1) THEN
430	DO 265 j = 1,NVAR
431	K4(iNVAR+j) = K4(iNVAR+j) + DFDP((i-1)*NVAR+j)
432	& + DJDP((i-1)*NVAR+j)
433	265 CONTINUE
434	END IF
435	CALL KppSolve (AJAC, K4(i*NVAR+1))
436	270 CONTINUE
437
438
439	C ---- The Solution ---
440	DO 280 j = 1,NVAR*(NSENSIT+1)
441	ynew(j) = y(j) + b1K1(j) + b2K2(j) + b3K3(j) + b4K4(j)
442	280 CONTINUE
443
444
445	C ====== Error estimation -- can be extended to control sensitivities too ========
446
447	ERR = 0.d0
448	DO 290 i=1,NVAR
449	w = AbsTol(i) + RelTol(i)*DMAX1(DABS(ynew(i)),DABS(y(i)))
450	e = d1K1(i) + d2K2(i) + d3K3(i) + d4K4(i)
451	ERR = ERR + ( e/w )**2
452	290 CONTINUE
453	ERR = DMAX1( uround, DSQRT( ERR/NVAR ) )
454
455	C ======= Choose the stepsize ===============================
456
457	elo = 4.0D0 ! estimator local order
458	factor = DMAX1(2.0D-1,DMIN1(6.0D0,ERR**(1.0D0/elo)/.9D0))
459	Hnew = DMIN1(Hmax,DMAX1(Hmin, H/factor))
460
461	C ======= Rejected/Accepted Step ============================
462
463	IF ( (ERR.gt.1).and.(H.gt.Hmin) ) THEN
464	IsReject = .true.
465	H = DMIN1(H/10,Hnew)
466	Nreject = Nreject+1
467	ELSE
468	DO 300 i=1,NVAR*(NSENSIT+1)
469	y(i) = ynew(i)
470	300 CONTINUE
471	T = Tplus
472	IF (.NOT.IsReject) THEN
473	H = Hnew ! Do not increase stepsize if previos step was rejected
474	END IF
475	IsReject = .false.
476	Naccept = Naccept+1
477	END IF
478
479	C ======= End of the time loop ===============================
480	IF ( T .lt. Tnext ) GO TO 10
481
482
483
484	C ======= Output Information =================================
485	Info(2) = Nfcn
486	Info(3) = Njac
487	Info(4) = Naccept
488	Info(5) = Nreject
489	Hstart = H
490
491	RETURN
492	END
493
494
495
496	SUBROUTINE FUNC_CHEM(N, T, Y, P)
497	INCLUDE 'KPP_ROOT_params.h'
498	INCLUDE 'KPP_ROOT_global.h'
499	KPP_REAL T, Told
500	KPP_REAL Y(NVAR), P(NVAR)
501	Told = TIME
502	TIME = T
503	CALL Update_SUN()
504	CALL Update_RCONST()
505	CALL Fun( Y, FIX, RCONST, P )
506	TIME = Told
507	RETURN
508	END
509
510
511	SUBROUTINE DFUNDPAR(N, NSENSIT, T, Y, P)
512	C --- Computes the partial derivatives of FUNC_CHEM w.r.t. parameters
513	INCLUDE 'KPP_ROOT_params.h'
514	INCLUDE 'KPP_ROOT_global.h'
515	C --- NCOEFF, JCOEFF useful for derivatives w.r.t. rate coefficients
516	INTEGER NCOEFF, JCOEFF(NREACT)
517	COMMON /DDMRCOEFF/ NCOEFF, JCOEFF
518
519	KPP_REAL T, Told
520	KPP_REAL Y(NVAR), P(NVAR*NSENSIT)
521	Told = TIME
522	TIME = T
523	CALL Update_SUN()
524	CALL Update_RCONST()
525	C
526	IF (DDMTYPE .EQ. 0) THEN
527	C --- Note: the values below are for sensitivities w.r.t. initial values;
528	C --- they may have to be changed for other applications
529	DO j=1,NSENSIT
530	DO i=1,NVAR
531	P(i+NVAR*(j-1)) = 0.0D0
532	END DO
533	END DO
534	ELSE
535	C --- Example: the call below is for sensitivities w.r.t. rate coefficients;
536	C --- JCOEFF(1:NSENSIT) are the indices of the NSENSIT rate coefficients
537	C --- w.r.t. which one differentiates
538	CALL dFun_dRcoeff( Y, FIX, NCOEFF, JCOEFF, P )
539	END IF
540	TIME = Told
541	RETURN
542	END
543
544	SUBROUTINE JAC_CHEM(N, T, Y, J)
545	INCLUDE 'KPP_ROOT_params.h'
546	INCLUDE 'KPP_ROOT_global.h'
547	INTEGER N
548	KPP_REAL Told, T
549	KPP_REAL Y(NVAR), J(LU_NONZERO)
550	Told = TIME
551	TIME = T
552	CALL Update_SUN()
553	CALL Update_RCONST()
554	CALL Jac_SP( Y, FIX, RCONST, J )
555	TIME = Told
556	RETURN
557	END
558
559
560	SUBROUTINE DJACDPAR(N, NSENSIT, T, Y, U, P)
561	C --- Computes the partial derivatives of JAC w.r.t. parameters times user vector U
562	INCLUDE 'KPP_ROOT_params.h'
563	INCLUDE 'KPP_ROOT_global.h'
564	C --- NCOEFF, JCOEFF useful for derivatives w.r.t. rate coefficients
565	INTEGER N
566	INTEGER NCOEFF, JCOEFF(NREACT)
567	COMMON /DDMRCOEFF/ NCOEFF, JCOEFF
568
569	KPP_REAL T, Told
570	KPP_REAL Y(NVAR), U(NVAR)
571	KPP_REAL P(NVAR*NSENSIT)
572	Told = TIME
573	TIME = T
574	CALL Update_SUN()
575	CALL Update_RCONST()
576	C
577	IF (DDMTYPE .EQ. 0) THEN
578	C --- Note: the values below are for sensitivities w.r.t. initial values;
579	C --- they may have to be changed for other applications
580	DO j=1,NSENSIT
581	DO i=1,NVAR
582	P(i+NVAR*(j-1)) = 0.0D0
583	END DO
584	END DO
585	ELSE
586	C --- Example: the call below is for sensitivities w.r.t. rate coefficients;
587	C --- JCOEFF(1:NSENSIT) are the indices of the NSENSIT rate coefficients
588	C --- w.r.t. which one differentiates
589	CALL dJac_dRcoeff( Y, FIX, U, NCOEFF, JCOEFF, P )
590	END IF
591	TIME = Told
592	RETURN
593	END
594
595
596	SUBROUTINE HESS_CHEM(N, T, Y, HESS)
597	INCLUDE 'KPP_ROOT_params.h'
598	INCLUDE 'KPP_ROOT_global.h'
599	INTEGER N
600	KPP_REAL Told, T
601	KPP_REAL Y(NVAR), HESS(NHESS)
602	Told = TIME
603	TIME = T
604	CALL Update_SUN()
605	CALL Update_RCONST()
606	CALL Hessian( Y, FIX, RCONST, HESS )
607	TIME = Told
608	RETURN
609	END
610
611
612
613
614
615

Note: See TracBrowser for help on using the repository browser.

Context Navigation

source: palm/trunk/UTIL/chemistry/gasphase_preproc/kpp/int/oldies/ros4_ddm.f @ 2696

Download in other formats: