1 /* 2 Code for timestepping with implicit Theta method 3 */ 4 #include <petsc-private/tsimpl.h> /*I "petscts.h" I*/ 5 #include <petscsnes.h> 6 #include <petscdm.h> 7 #include <petscmat.h> 8 9 typedef struct { 10 Vec X,Xdot; /* Storage for one stage */ 11 Vec X0; /* work vector to store X0 */ 12 Vec affine; /* Affine vector needed for residual at beginning of step */ 13 Vec *VecDeltaLam; /* Increment of the adjoint sensitivity w.r.t IC at stage*/ 14 Vec *VecDeltaMu; /* Increment of the adjoint sensitivity w.r.t P at stage*/ 15 Vec *VecSensiTemp; /* Vector to be timed with Jacobian transpose*/ 16 PetscBool extrapolate; 17 PetscBool endpoint; 18 PetscReal Theta; 19 PetscReal stage_time; 20 TSStepStatus status; 21 char *name; 22 PetscInt order; 23 PetscReal ccfl; /* Placeholder for CFL coefficient relative to forward Euler */ 24 PetscBool adapt; /* use time-step adaptivity ? */ 25 } TS_Theta; 26 27 #undef __FUNCT__ 28 #define __FUNCT__ "TSThetaGetX0AndXdot" 29 static PetscErrorCode TSThetaGetX0AndXdot(TS ts,DM dm,Vec *X0,Vec *Xdot) 30 { 31 TS_Theta *th = (TS_Theta*)ts->data; 32 PetscErrorCode ierr; 33 34 PetscFunctionBegin; 35 if (X0) { 36 if (dm && dm != ts->dm) { 37 ierr = DMGetNamedGlobalVector(dm,"TSTheta_X0",X0);CHKERRQ(ierr); 38 } else *X0 = ts->vec_sol; 39 } 40 if (Xdot) { 41 if (dm && dm != ts->dm) { 42 ierr = DMGetNamedGlobalVector(dm,"TSTheta_Xdot",Xdot);CHKERRQ(ierr); 43 } else *Xdot = th->Xdot; 44 } 45 PetscFunctionReturn(0); 46 } 47 48 49 #undef __FUNCT__ 50 #define __FUNCT__ "TSThetaRestoreX0AndXdot" 51 static PetscErrorCode TSThetaRestoreX0AndXdot(TS ts,DM dm,Vec *X0,Vec *Xdot) 52 { 53 PetscErrorCode ierr; 54 55 PetscFunctionBegin; 56 if (X0) { 57 if (dm && dm != ts->dm) { 58 ierr = DMRestoreNamedGlobalVector(dm,"TSTheta_X0",X0);CHKERRQ(ierr); 59 } 60 } 61 if (Xdot) { 62 if (dm && dm != ts->dm) { 63 ierr = DMRestoreNamedGlobalVector(dm,"TSTheta_Xdot",Xdot);CHKERRQ(ierr); 64 } 65 } 66 PetscFunctionReturn(0); 67 } 68 69 #undef __FUNCT__ 70 #define __FUNCT__ "DMCoarsenHook_TSTheta" 71 static PetscErrorCode DMCoarsenHook_TSTheta(DM fine,DM coarse,void *ctx) 72 { 73 74 PetscFunctionBegin; 75 PetscFunctionReturn(0); 76 } 77 78 #undef __FUNCT__ 79 #define __FUNCT__ "DMRestrictHook_TSTheta" 80 static PetscErrorCode DMRestrictHook_TSTheta(DM fine,Mat restrct,Vec rscale,Mat inject,DM coarse,void *ctx) 81 { 82 TS ts = (TS)ctx; 83 PetscErrorCode ierr; 84 Vec X0,Xdot,X0_c,Xdot_c; 85 86 PetscFunctionBegin; 87 ierr = TSThetaGetX0AndXdot(ts,fine,&X0,&Xdot);CHKERRQ(ierr); 88 ierr = TSThetaGetX0AndXdot(ts,coarse,&X0_c,&Xdot_c);CHKERRQ(ierr); 89 ierr = MatRestrict(restrct,X0,X0_c);CHKERRQ(ierr); 90 ierr = MatRestrict(restrct,Xdot,Xdot_c);CHKERRQ(ierr); 91 ierr = VecPointwiseMult(X0_c,rscale,X0_c);CHKERRQ(ierr); 92 ierr = VecPointwiseMult(Xdot_c,rscale,Xdot_c);CHKERRQ(ierr); 93 ierr = TSThetaRestoreX0AndXdot(ts,fine,&X0,&Xdot);CHKERRQ(ierr); 94 ierr = TSThetaRestoreX0AndXdot(ts,coarse,&X0_c,&Xdot_c);CHKERRQ(ierr); 95 PetscFunctionReturn(0); 96 } 97 98 #undef __FUNCT__ 99 #define __FUNCT__ "DMSubDomainHook_TSTheta" 100 static PetscErrorCode DMSubDomainHook_TSTheta(DM dm,DM subdm,void *ctx) 101 { 102 103 PetscFunctionBegin; 104 PetscFunctionReturn(0); 105 } 106 107 #undef __FUNCT__ 108 #define __FUNCT__ "DMSubDomainRestrictHook_TSTheta" 109 static PetscErrorCode DMSubDomainRestrictHook_TSTheta(DM dm,VecScatter gscat,VecScatter lscat,DM subdm,void *ctx) 110 { 111 TS ts = (TS)ctx; 112 PetscErrorCode ierr; 113 Vec X0,Xdot,X0_sub,Xdot_sub; 114 115 PetscFunctionBegin; 116 ierr = TSThetaGetX0AndXdot(ts,dm,&X0,&Xdot);CHKERRQ(ierr); 117 ierr = TSThetaGetX0AndXdot(ts,subdm,&X0_sub,&Xdot_sub);CHKERRQ(ierr); 118 119 ierr = VecScatterBegin(gscat,X0,X0_sub,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 120 ierr = VecScatterEnd(gscat,X0,X0_sub,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 121 122 ierr = VecScatterBegin(gscat,Xdot,Xdot_sub,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 123 ierr = VecScatterEnd(gscat,Xdot,Xdot_sub,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 124 125 ierr = TSThetaRestoreX0AndXdot(ts,dm,&X0,&Xdot);CHKERRQ(ierr); 126 ierr = TSThetaRestoreX0AndXdot(ts,subdm,&X0_sub,&Xdot_sub);CHKERRQ(ierr); 127 PetscFunctionReturn(0); 128 } 129 130 #undef __FUNCT__ 131 #define __FUNCT__ "TSEvaluateStep_Theta" 132 static PetscErrorCode TSEvaluateStep_Theta(TS ts,PetscInt order,Vec U,PetscBool *done) 133 { 134 PetscErrorCode ierr; 135 TS_Theta *th = (TS_Theta*)ts->data; 136 137 PetscFunctionBegin; 138 if (order == 0) SETERRQ(PetscObjectComm((PetscObject)ts),PETSC_ERR_USER,"No time-step adaptivity implemented for 1st order theta method; Run with -ts_adapt_type none"); 139 if (order == th->order) { 140 if (th->endpoint) { 141 ierr = VecCopy(th->X,U);CHKERRQ(ierr); 142 } else { 143 PetscReal shift = 1./(th->Theta*ts->time_step); 144 ierr = VecAXPBYPCZ(th->Xdot,-shift,shift,0,U,th->X);CHKERRQ(ierr); 145 ierr = VecAXPY(U,ts->time_step,th->Xdot);CHKERRQ(ierr); 146 } 147 } else if (order == th->order-1 && order) { 148 ierr = VecWAXPY(U,ts->time_step,th->Xdot,th->X0);CHKERRQ(ierr); 149 } 150 PetscFunctionReturn(0); 151 } 152 153 #undef __FUNCT__ 154 #define __FUNCT__ "TSRollBack_Theta" 155 static PetscErrorCode TSRollBack_Theta(TS ts) 156 { 157 TS_Theta *th = (TS_Theta*)ts->data; 158 PetscErrorCode ierr; 159 160 PetscFunctionBegin; 161 ierr = VecCopy(th->X0,ts->vec_sol);CHKERRQ(ierr); 162 th->status = TS_STEP_INCOMPLETE; 163 PetscFunctionReturn(0); 164 } 165 166 #undef __FUNCT__ 167 #define __FUNCT__ "TSStep_Theta" 168 static PetscErrorCode TSStep_Theta(TS ts) 169 { 170 TS_Theta *th = (TS_Theta*)ts->data; 171 PetscInt its,lits,reject,next_scheme; 172 PetscReal next_time_step; 173 TSAdapt adapt; 174 PetscBool stageok,accept = PETSC_TRUE; 175 PetscErrorCode ierr; 176 177 PetscFunctionBegin; 178 th->status = TS_STEP_INCOMPLETE; 179 ierr = VecCopy(ts->vec_sol,th->X0);CHKERRQ(ierr); 180 for (reject=0; !ts->reason && th->status != TS_STEP_COMPLETE; ts->reject++) { 181 PetscReal shift = 1./(th->Theta*ts->time_step); 182 th->stage_time = ts->ptime + (th->endpoint ? 1. : th->Theta)*ts->time_step; 183 ierr = TSPreStep(ts);CHKERRQ(ierr); 184 ierr = TSPreStage(ts,th->stage_time);CHKERRQ(ierr); 185 186 if (th->endpoint) { /* This formulation assumes linear time-independent mass matrix */ 187 ierr = VecZeroEntries(th->Xdot);CHKERRQ(ierr); 188 if (!th->affine) {ierr = VecDuplicate(ts->vec_sol,&th->affine);CHKERRQ(ierr);} 189 ierr = TSComputeIFunction(ts,ts->ptime,ts->vec_sol,th->Xdot,th->affine,PETSC_FALSE);CHKERRQ(ierr); 190 ierr = VecScale(th->affine,(th->Theta-1.)/th->Theta);CHKERRQ(ierr); 191 } 192 if (th->extrapolate) { 193 ierr = VecWAXPY(th->X,1./shift,th->Xdot,ts->vec_sol);CHKERRQ(ierr); 194 } else { 195 ierr = VecCopy(ts->vec_sol,th->X);CHKERRQ(ierr); 196 } 197 ierr = SNESSolve(ts->snes,th->affine,th->X);CHKERRQ(ierr); 198 ierr = SNESGetIterationNumber(ts->snes,&its);CHKERRQ(ierr); 199 ierr = SNESGetLinearSolveIterations(ts->snes,&lits);CHKERRQ(ierr); 200 ts->snes_its += its; ts->ksp_its += lits; 201 ierr = TSPostStage(ts,th->stage_time,0,&(th->X));CHKERRQ(ierr); 202 ierr = TSGetAdapt(ts,&adapt);CHKERRQ(ierr); 203 ierr = TSAdaptCheckStage(adapt,ts,&stageok);CHKERRQ(ierr); 204 if (!stageok) {accept = PETSC_FALSE; goto reject_step;} 205 206 ierr = TSEvaluateStep(ts,th->order,ts->vec_sol,NULL);CHKERRQ(ierr); 207 th->status = TS_STEP_PENDING; 208 /* Register only the current method as a candidate because we're not supporting multiple candidates yet. */ 209 ierr = TSGetAdapt(ts,&adapt);CHKERRQ(ierr); 210 ierr = TSAdaptCandidatesClear(adapt);CHKERRQ(ierr); 211 ierr = TSAdaptCandidateAdd(adapt,NULL,th->order,1,th->ccfl,1.0,PETSC_TRUE);CHKERRQ(ierr); 212 ierr = TSAdaptChoose(adapt,ts,ts->time_step,&next_scheme,&next_time_step,&accept);CHKERRQ(ierr); 213 if (!accept) { /* Roll back the current step */ 214 ts->ptime += next_time_step; /* This will be undone in rollback */ 215 th->status = TS_STEP_INCOMPLETE; 216 ierr = TSRollBack(ts);CHKERRQ(ierr); 217 goto reject_step; 218 } 219 220 /* ignore next_scheme for now */ 221 ts->ptime += ts->time_step; 222 ts->time_step = next_time_step; 223 ts->steps++; 224 th->status = TS_STEP_COMPLETE; 225 break; 226 227 reject_step: 228 if (!ts->reason && ++reject > ts->max_reject && ts->max_reject >= 0) { 229 ts->reason = TS_DIVERGED_STEP_REJECTED; 230 ierr = PetscInfo2(ts,"Step=%D, step rejections %D greater than current TS allowed, stopping solve\n",ts->steps,reject);CHKERRQ(ierr); 231 } 232 continue; 233 } 234 PetscFunctionReturn(0); 235 } 236 237 #undef __FUNCT__ 238 #define __FUNCT__ "TSStepAdj_Theta" 239 static PetscErrorCode TSStepAdj_Theta(TS ts) 240 { 241 TS_Theta *th = (TS_Theta*)ts->data; 242 Vec *VecDeltaLam = th->VecDeltaLam,*VecDeltaMu = th->VecDeltaMu,*VecSensiTemp = th->VecSensiTemp; 243 PetscInt nadj; 244 PetscErrorCode ierr; 245 Mat J,Jp; 246 KSP ksp; 247 PetscReal shift; 248 249 PetscFunctionBegin; 250 251 th->status = TS_STEP_INCOMPLETE; 252 ierr = SNESGetKSP(ts->snes,&ksp); 253 ierr = TSGetIJacobian(ts,&J,&Jp,NULL,NULL);CHKERRQ(ierr); 254 th->stage_time = ts->ptime + (th->endpoint ? ts->time_step : (1.-th->Theta)*ts->time_step); /* time_step is negative*/ 255 256 ierr = TSPreStep(ts);CHKERRQ(ierr); 257 258 /* Build RHS */ 259 if (ts->vec_costquad) { /* Cost function has an integral (quadrature) term */ 260 if (th->endpoint) { 261 ierr = TSComputeDRDYFunction(ts,ts->ptime,ts->vec_sol,ts->vecs_drdy);CHKERRQ(ierr); 262 }else { 263 ierr = TSComputeDRDYFunction(ts,th->stage_time,th->X,ts->vecs_drdy);CHKERRQ(ierr); 264 } 265 } 266 for (nadj=0; nadj<ts->numberadjs; nadj++) { 267 ierr = VecCopy(ts->vecs_sensi[nadj],VecSensiTemp[nadj]);CHKERRQ(ierr); 268 ierr = VecScale(VecSensiTemp[nadj],-1./(th->Theta*ts->time_step));CHKERRQ(ierr); 269 } 270 271 /* Build LHS */ 272 shift = -1./(th->Theta*ts->time_step); 273 if (th->endpoint) { 274 ierr = TSComputeIJacobian(ts,ts->ptime,ts->vec_sol,th->Xdot,shift,J,Jp,PETSC_FALSE);CHKERRQ(ierr); 275 }else { 276 ierr = TSComputeIJacobian(ts,th->stage_time,th->X,th->Xdot,shift,J,Jp,PETSC_FALSE);CHKERRQ(ierr); 277 } 278 ierr = KSPSetOperators(ksp,J,Jp);CHKERRQ(ierr); 279 280 /* Solve LHS X = RHS */ 281 for (nadj=0; nadj<ts->numberadjs; nadj++) { 282 ierr = KSPSolveTranspose(ksp,VecSensiTemp[nadj],VecDeltaLam[nadj]);CHKERRQ(ierr); 283 } 284 285 /* Update sensitivities */ 286 if(th->endpoint && th->Theta!=1.) { /* two-stage case */ 287 shift = -1./((th->Theta-1.)*ts->time_step); 288 ierr = TSComputeIJacobian(ts,th->stage_time,th->X,th->Xdot,shift,J,Jp,PETSC_FALSE);CHKERRQ(ierr); 289 for (nadj=0; nadj<ts->numberadjs; nadj++) { 290 ierr = MatMultTranspose(J,VecDeltaLam[nadj],ts->vecs_sensi[nadj]);CHKERRQ(ierr); 291 ierr = VecScale(ts->vecs_sensi[nadj],1./shift);CHKERRQ(ierr); 292 } 293 294 if (ts->vecs_sensip) { /* sensitivities wrt parameters */ 295 ierr = TSRHSJacobianP(ts,ts->ptime,ts->vec_sol,ts->Jacp);CHKERRQ(ierr); 296 for (nadj=0; nadj<ts->numberadjs; nadj++) { 297 ierr = MatMultTranspose(ts->Jacp,VecDeltaLam[nadj],VecDeltaMu[nadj]);CHKERRQ(ierr); 298 ierr = VecAXPY(ts->vecs_sensip[nadj],-ts->time_step*th->Theta,VecDeltaMu[nadj]);CHKERRQ(ierr); 299 } 300 ierr = TSRHSJacobianP(ts,th->stage_time,th->X,ts->Jacp);CHKERRQ(ierr); 301 for (nadj=0; nadj<ts->numberadjs; nadj++) { 302 ierr = MatMultTranspose(ts->Jacp,VecDeltaLam[nadj],VecDeltaMu[nadj]);CHKERRQ(ierr); 303 ierr = VecAXPY(ts->vecs_sensip[nadj],-ts->time_step*(1.-th->Theta),VecDeltaMu[nadj]);CHKERRQ(ierr); 304 } 305 } 306 }else { /* one-stage case */ 307 shift = 0.0; 308 ierr = TSComputeIJacobian(ts,th->stage_time,th->X,th->Xdot,shift,J,Jp,PETSC_FALSE);CHKERRQ(ierr); 309 /* When th->endpoint is true and th->Theta==1 (beuler method), the Jacobian is supposed to be evaluated at ts->ptime like this: 310 if(th->endpoint) { 311 ierr = TSComputeIJacobian(ts,ts->ptime,ts->vec_sol,th->Xdot,shift,J,Jp,PETSC_FALSE);CHKERRQ(ierr); 312 } 313 but ts->ptime and ts->vec_sol have the same values as th->stage_time and th->X in this case. So the code is simplified here. 314 */ 315 for (nadj=0; nadj<ts->numberadjs; nadj++) { 316 ierr = MatMultTranspose(J,VecDeltaLam[nadj],VecSensiTemp[nadj]);CHKERRQ(ierr); 317 ierr = VecAXPY(ts->vecs_sensi[nadj],ts->time_step,VecSensiTemp[nadj]);CHKERRQ(ierr); 318 } 319 if (ts->vecs_sensip) { 320 ierr = TSRHSJacobianP(ts,th->stage_time,th->X,ts->Jacp);CHKERRQ(ierr); 321 for (nadj=0; nadj<ts->numberadjs; nadj++) { 322 ierr = MatMultTranspose(ts->Jacp,VecDeltaLam[nadj],VecDeltaMu[nadj]);CHKERRQ(ierr); 323 ierr = VecAXPY(ts->vecs_sensip[nadj],-ts->time_step,VecDeltaMu[nadj]);CHKERRQ(ierr); 324 } 325 } 326 } 327 328 ts->ptime += ts->time_step; 329 ts->steps++; 330 th->status = TS_STEP_COMPLETE; 331 PetscFunctionReturn(0); 332 } 333 334 #undef __FUNCT__ 335 #define __FUNCT__ "TSInterpolate_Theta" 336 static PetscErrorCode TSInterpolate_Theta(TS ts,PetscReal t,Vec X) 337 { 338 TS_Theta *th = (TS_Theta*)ts->data; 339 PetscReal alpha = t - ts->ptime; 340 PetscErrorCode ierr; 341 342 PetscFunctionBegin; 343 ierr = VecCopy(ts->vec_sol,th->X);CHKERRQ(ierr); 344 if (th->endpoint) alpha *= th->Theta; 345 ierr = VecWAXPY(X,alpha,th->Xdot,th->X);CHKERRQ(ierr); 346 PetscFunctionReturn(0); 347 } 348 349 /*------------------------------------------------------------*/ 350 #undef __FUNCT__ 351 #define __FUNCT__ "TSReset_Theta" 352 static PetscErrorCode TSReset_Theta(TS ts) 353 { 354 TS_Theta *th = (TS_Theta*)ts->data; 355 PetscErrorCode ierr; 356 357 PetscFunctionBegin; 358 ierr = VecDestroy(&th->X);CHKERRQ(ierr); 359 ierr = VecDestroy(&th->Xdot);CHKERRQ(ierr); 360 ierr = VecDestroy(&th->X0);CHKERRQ(ierr); 361 ierr = VecDestroy(&th->affine);CHKERRQ(ierr); 362 if(ts->reverse_mode) { 363 ierr = VecDestroyVecs(ts->numberadjs,&th->VecDeltaLam);CHKERRQ(ierr); 364 if(th->VecDeltaMu) { 365 ierr = VecDestroyVecs(ts->numberadjs,&th->VecDeltaMu);CHKERRQ(ierr); 366 } 367 ierr = VecDestroyVecs(ts->numberadjs,&th->VecSensiTemp);CHKERRQ(ierr); 368 } 369 PetscFunctionReturn(0); 370 } 371 372 #undef __FUNCT__ 373 #define __FUNCT__ "TSDestroy_Theta" 374 static PetscErrorCode TSDestroy_Theta(TS ts) 375 { 376 PetscErrorCode ierr; 377 378 PetscFunctionBegin; 379 ierr = TSReset_Theta(ts);CHKERRQ(ierr); 380 ierr = PetscFree(ts->data);CHKERRQ(ierr); 381 ierr = PetscObjectComposeFunction((PetscObject)ts,"TSThetaGetTheta_C",NULL);CHKERRQ(ierr); 382 ierr = PetscObjectComposeFunction((PetscObject)ts,"TSThetaSetTheta_C",NULL);CHKERRQ(ierr); 383 ierr = PetscObjectComposeFunction((PetscObject)ts,"TSThetaGetEndpoint_C",NULL);CHKERRQ(ierr); 384 ierr = PetscObjectComposeFunction((PetscObject)ts,"TSThetaSetEndpoint_C",NULL);CHKERRQ(ierr); 385 PetscFunctionReturn(0); 386 } 387 388 /* 389 This defines the nonlinear equation that is to be solved with SNES 390 G(U) = F[t0+Theta*dt, U, (U-U0)*shift] = 0 391 */ 392 #undef __FUNCT__ 393 #define __FUNCT__ "SNESTSFormFunction_Theta" 394 static PetscErrorCode SNESTSFormFunction_Theta(SNES snes,Vec x,Vec y,TS ts) 395 { 396 TS_Theta *th = (TS_Theta*)ts->data; 397 PetscErrorCode ierr; 398 Vec X0,Xdot; 399 DM dm,dmsave; 400 PetscReal shift = 1./(th->Theta*ts->time_step); 401 402 PetscFunctionBegin; 403 ierr = SNESGetDM(snes,&dm);CHKERRQ(ierr); 404 /* When using the endpoint variant, this is actually 1/Theta * Xdot */ 405 ierr = TSThetaGetX0AndXdot(ts,dm,&X0,&Xdot);CHKERRQ(ierr); 406 ierr = VecAXPBYPCZ(Xdot,-shift,shift,0,X0,x);CHKERRQ(ierr); 407 408 /* DM monkey-business allows user code to call TSGetDM() inside of functions evaluated on levels of FAS */ 409 dmsave = ts->dm; 410 ts->dm = dm; 411 ierr = TSComputeIFunction(ts,th->stage_time,x,Xdot,y,PETSC_FALSE);CHKERRQ(ierr); 412 ts->dm = dmsave; 413 ierr = TSThetaRestoreX0AndXdot(ts,dm,&X0,&Xdot);CHKERRQ(ierr); 414 PetscFunctionReturn(0); 415 } 416 417 #undef __FUNCT__ 418 #define __FUNCT__ "SNESTSFormJacobian_Theta" 419 static PetscErrorCode SNESTSFormJacobian_Theta(SNES snes,Vec x,Mat A,Mat B,TS ts) 420 { 421 TS_Theta *th = (TS_Theta*)ts->data; 422 PetscErrorCode ierr; 423 Vec Xdot; 424 DM dm,dmsave; 425 PetscReal shift = 1./(th->Theta*ts->time_step); 426 427 PetscFunctionBegin; 428 ierr = SNESGetDM(snes,&dm);CHKERRQ(ierr); 429 430 /* th->Xdot has already been computed in SNESTSFormFunction_Theta (SNES guarantees this) */ 431 ierr = TSThetaGetX0AndXdot(ts,dm,NULL,&Xdot);CHKERRQ(ierr); 432 433 dmsave = ts->dm; 434 ts->dm = dm; 435 ierr = TSComputeIJacobian(ts,th->stage_time,x,Xdot,shift,A,B,PETSC_FALSE);CHKERRQ(ierr); 436 ts->dm = dmsave; 437 ierr = TSThetaRestoreX0AndXdot(ts,dm,NULL,&Xdot);CHKERRQ(ierr); 438 PetscFunctionReturn(0); 439 } 440 441 #undef __FUNCT__ 442 #define __FUNCT__ "TSSetUp_Theta" 443 static PetscErrorCode TSSetUp_Theta(TS ts) 444 { 445 TS_Theta *th = (TS_Theta*)ts->data; 446 PetscErrorCode ierr; 447 SNES snes; 448 TSAdapt adapt; 449 DM dm; 450 451 PetscFunctionBegin; 452 ierr = VecDuplicate(ts->vec_sol,&th->X);CHKERRQ(ierr); 453 ierr = VecDuplicate(ts->vec_sol,&th->Xdot);CHKERRQ(ierr); 454 ierr = VecDuplicate(ts->vec_sol,&th->X0);CHKERRQ(ierr); 455 ierr = TSGetSNES(ts,&snes);CHKERRQ(ierr); 456 ierr = TSGetDM(ts,&dm);CHKERRQ(ierr); 457 if (dm) { 458 ierr = DMCoarsenHookAdd(dm,DMCoarsenHook_TSTheta,DMRestrictHook_TSTheta,ts);CHKERRQ(ierr); 459 ierr = DMSubDomainHookAdd(dm,DMSubDomainHook_TSTheta,DMSubDomainRestrictHook_TSTheta,ts);CHKERRQ(ierr); 460 } 461 if (th->Theta == 0.5 && th->endpoint) th->order = 2; 462 else th->order = 1; 463 464 ierr = TSGetAdapt(ts,&adapt);CHKERRQ(ierr); 465 if (!th->adapt) { 466 ierr = TSAdaptSetType(adapt,TSADAPTNONE);CHKERRQ(ierr); 467 } 468 if (ts->reverse_mode) { 469 ierr = VecDuplicateVecs(ts->vecs_sensi[0],ts->numberadjs,&th->VecDeltaLam);CHKERRQ(ierr); 470 if(ts->vecs_sensip) { 471 ierr = VecDuplicateVecs(ts->vecs_sensip[0],ts->numberadjs,&th->VecDeltaMu);CHKERRQ(ierr); 472 } 473 ierr = VecDuplicateVecs(ts->vecs_sensi[0],ts->numberadjs,&th->VecSensiTemp);CHKERRQ(ierr); 474 } 475 PetscFunctionReturn(0); 476 } 477 /*------------------------------------------------------------*/ 478 479 #undef __FUNCT__ 480 #define __FUNCT__ "TSSetFromOptions_Theta" 481 static PetscErrorCode TSSetFromOptions_Theta(PetscOptions *PetscOptionsObject,TS ts) 482 { 483 TS_Theta *th = (TS_Theta*)ts->data; 484 PetscErrorCode ierr; 485 486 PetscFunctionBegin; 487 ierr = PetscOptionsHead(PetscOptionsObject,"Theta ODE solver options");CHKERRQ(ierr); 488 { 489 ierr = PetscOptionsReal("-ts_theta_theta","Location of stage (0<Theta<=1)","TSThetaSetTheta",th->Theta,&th->Theta,NULL);CHKERRQ(ierr); 490 ierr = PetscOptionsBool("-ts_theta_extrapolate","Extrapolate stage solution from previous solution (sometimes unstable)","TSThetaSetExtrapolate",th->extrapolate,&th->extrapolate,NULL);CHKERRQ(ierr); 491 ierr = PetscOptionsBool("-ts_theta_endpoint","Use the endpoint instead of midpoint form of the Theta method","TSThetaSetEndpoint",th->endpoint,&th->endpoint,NULL);CHKERRQ(ierr); 492 ierr = PetscOptionsBool("-ts_theta_adapt","Use time-step adaptivity with the Theta method","",th->adapt,&th->adapt,NULL);CHKERRQ(ierr); 493 ierr = SNESSetFromOptions(ts->snes);CHKERRQ(ierr); 494 } 495 ierr = PetscOptionsTail();CHKERRQ(ierr); 496 PetscFunctionReturn(0); 497 } 498 499 #undef __FUNCT__ 500 #define __FUNCT__ "TSView_Theta" 501 static PetscErrorCode TSView_Theta(TS ts,PetscViewer viewer) 502 { 503 TS_Theta *th = (TS_Theta*)ts->data; 504 PetscBool iascii; 505 PetscErrorCode ierr; 506 507 PetscFunctionBegin; 508 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);CHKERRQ(ierr); 509 if (iascii) { 510 ierr = PetscViewerASCIIPrintf(viewer," Theta=%g\n",(double)th->Theta);CHKERRQ(ierr); 511 ierr = PetscViewerASCIIPrintf(viewer," Extrapolation=%s\n",th->extrapolate ? "yes" : "no");CHKERRQ(ierr); 512 } 513 if (ts->snes) {ierr = SNESView(ts->snes,viewer);CHKERRQ(ierr);} 514 PetscFunctionReturn(0); 515 } 516 517 #undef __FUNCT__ 518 #define __FUNCT__ "TSThetaGetTheta_Theta" 519 PetscErrorCode TSThetaGetTheta_Theta(TS ts,PetscReal *theta) 520 { 521 TS_Theta *th = (TS_Theta*)ts->data; 522 523 PetscFunctionBegin; 524 *theta = th->Theta; 525 PetscFunctionReturn(0); 526 } 527 528 #undef __FUNCT__ 529 #define __FUNCT__ "TSThetaSetTheta_Theta" 530 PetscErrorCode TSThetaSetTheta_Theta(TS ts,PetscReal theta) 531 { 532 TS_Theta *th = (TS_Theta*)ts->data; 533 534 PetscFunctionBegin; 535 if (theta <= 0 || 1 < theta) SETERRQ1(PetscObjectComm((PetscObject)ts),PETSC_ERR_ARG_OUTOFRANGE,"Theta %g not in range (0,1]",(double)theta); 536 th->Theta = theta; 537 PetscFunctionReturn(0); 538 } 539 540 #undef __FUNCT__ 541 #define __FUNCT__ "TSThetaGetEndpoint_Theta" 542 PetscErrorCode TSThetaGetEndpoint_Theta(TS ts,PetscBool *endpoint) 543 { 544 TS_Theta *th = (TS_Theta*)ts->data; 545 546 PetscFunctionBegin; 547 *endpoint = th->endpoint; 548 PetscFunctionReturn(0); 549 } 550 551 #undef __FUNCT__ 552 #define __FUNCT__ "TSThetaSetEndpoint_Theta" 553 PetscErrorCode TSThetaSetEndpoint_Theta(TS ts,PetscBool flg) 554 { 555 TS_Theta *th = (TS_Theta*)ts->data; 556 557 PetscFunctionBegin; 558 th->endpoint = flg; 559 PetscFunctionReturn(0); 560 } 561 562 #if defined(PETSC_HAVE_COMPLEX) 563 #undef __FUNCT__ 564 #define __FUNCT__ "TSComputeLinearStability_Theta" 565 static PetscErrorCode TSComputeLinearStability_Theta(TS ts,PetscReal xr,PetscReal xi,PetscReal *yr,PetscReal *yi) 566 { 567 PetscComplex z = xr + xi*PETSC_i,f; 568 TS_Theta *th = (TS_Theta*)ts->data; 569 const PetscReal one = 1.0; 570 571 PetscFunctionBegin; 572 f = (one + (one - th->Theta)*z)/(one - th->Theta*z); 573 *yr = PetscRealPartComplex(f); 574 *yi = PetscImaginaryPartComplex(f); 575 PetscFunctionReturn(0); 576 } 577 #endif 578 579 #undef __FUNCT__ 580 #define __FUNCT__ "TSGetStages_Theta" 581 static PetscErrorCode TSGetStages_Theta(TS ts,PetscInt *ns,Vec **Y) 582 { 583 TS_Theta *th = (TS_Theta*)ts->data; 584 585 PetscFunctionBegin; 586 *ns = 1; 587 if(Y) { 588 *Y = &(th->X); 589 } 590 PetscFunctionReturn(0); 591 } 592 593 /* ------------------------------------------------------------ */ 594 /*MC 595 TSTHETA - DAE solver using the implicit Theta method 596 597 Level: beginner 598 599 Options Database: 600 -ts_theta_theta <Theta> - Location of stage (0<Theta<=1) 601 -ts_theta_extrapolate <flg> Extrapolate stage solution from previous solution (sometimes unstable) 602 -ts_theta_endpoint <flag> - Use the endpoint (like Crank-Nicholson) instead of midpoint form of the Theta method 603 604 Notes: 605 $ -ts_type theta -ts_theta_theta 1.0 corresponds to backward Euler (TSBEULER) 606 $ -ts_type theta -ts_theta_theta 0.5 corresponds to the implicit midpoint rule 607 $ -ts_type theta -ts_theta_theta 0.5 -ts_theta_endpoint corresponds to Crank-Nicholson (TSCN) 608 609 610 611 This method can be applied to DAE. 612 613 This method is cast as a 1-stage implicit Runge-Kutta method. 614 615 .vb 616 Theta | Theta 617 ------------- 618 | 1 619 .ve 620 621 For the default Theta=0.5, this is also known as the implicit midpoint rule. 622 623 When the endpoint variant is chosen, the method becomes a 2-stage method with first stage explicit: 624 625 .vb 626 0 | 0 0 627 1 | 1-Theta Theta 628 ------------------- 629 | 1-Theta Theta 630 .ve 631 632 For the default Theta=0.5, this is the trapezoid rule (also known as Crank-Nicolson, see TSCN). 633 634 To apply a diagonally implicit RK method to DAE, the stage formula 635 636 $ Y_i = X + h sum_j a_ij Y'_j 637 638 is interpreted as a formula for Y'_i in terms of Y_i and known values (Y'_j, j<i) 639 640 .seealso: TSCreate(), TS, TSSetType(), TSCN, TSBEULER, TSThetaSetTheta(), TSThetaSetEndpoint() 641 642 M*/ 643 #undef __FUNCT__ 644 #define __FUNCT__ "TSCreate_Theta" 645 PETSC_EXTERN PetscErrorCode TSCreate_Theta(TS ts) 646 { 647 TS_Theta *th; 648 PetscErrorCode ierr; 649 650 PetscFunctionBegin; 651 ts->ops->reset = TSReset_Theta; 652 ts->ops->destroy = TSDestroy_Theta; 653 ts->ops->view = TSView_Theta; 654 ts->ops->setup = TSSetUp_Theta; 655 ts->ops->step = TSStep_Theta; 656 ts->ops->interpolate = TSInterpolate_Theta; 657 ts->ops->evaluatestep = TSEvaluateStep_Theta; 658 ts->ops->rollback = TSRollBack_Theta; 659 ts->ops->setfromoptions = TSSetFromOptions_Theta; 660 ts->ops->snesfunction = SNESTSFormFunction_Theta; 661 ts->ops->snesjacobian = SNESTSFormJacobian_Theta; 662 #if defined(PETSC_HAVE_COMPLEX) 663 ts->ops->linearstability = TSComputeLinearStability_Theta; 664 #endif 665 ts->ops->getstages = TSGetStages_Theta; 666 ts->ops->stepadj = TSStepAdj_Theta; 667 668 ierr = PetscNewLog(ts,&th);CHKERRQ(ierr); 669 ts->data = (void*)th; 670 671 th->extrapolate = PETSC_FALSE; 672 th->Theta = 0.5; 673 th->ccfl = 1.0; 674 th->adapt = PETSC_FALSE; 675 ierr = PetscObjectComposeFunction((PetscObject)ts,"TSThetaGetTheta_C",TSThetaGetTheta_Theta);CHKERRQ(ierr); 676 ierr = PetscObjectComposeFunction((PetscObject)ts,"TSThetaSetTheta_C",TSThetaSetTheta_Theta);CHKERRQ(ierr); 677 ierr = PetscObjectComposeFunction((PetscObject)ts,"TSThetaGetEndpoint_C",TSThetaGetEndpoint_Theta);CHKERRQ(ierr); 678 ierr = PetscObjectComposeFunction((PetscObject)ts,"TSThetaSetEndpoint_C",TSThetaSetEndpoint_Theta);CHKERRQ(ierr); 679 PetscFunctionReturn(0); 680 } 681 682 #undef __FUNCT__ 683 #define __FUNCT__ "TSThetaGetTheta" 684 /*@ 685 TSThetaGetTheta - Get the abscissa of the stage in (0,1]. 686 687 Not Collective 688 689 Input Parameter: 690 . ts - timestepping context 691 692 Output Parameter: 693 . theta - stage abscissa 694 695 Note: 696 Use of this function is normally only required to hack TSTHETA to use a modified integration scheme. 697 698 Level: Advanced 699 700 .seealso: TSThetaSetTheta() 701 @*/ 702 PetscErrorCode TSThetaGetTheta(TS ts,PetscReal *theta) 703 { 704 PetscErrorCode ierr; 705 706 PetscFunctionBegin; 707 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 708 PetscValidPointer(theta,2); 709 ierr = PetscUseMethod(ts,"TSThetaGetTheta_C",(TS,PetscReal*),(ts,theta));CHKERRQ(ierr); 710 PetscFunctionReturn(0); 711 } 712 713 #undef __FUNCT__ 714 #define __FUNCT__ "TSThetaSetTheta" 715 /*@ 716 TSThetaSetTheta - Set the abscissa of the stage in (0,1]. 717 718 Not Collective 719 720 Input Parameter: 721 + ts - timestepping context 722 - theta - stage abscissa 723 724 Options Database: 725 . -ts_theta_theta <theta> 726 727 Level: Intermediate 728 729 .seealso: TSThetaGetTheta() 730 @*/ 731 PetscErrorCode TSThetaSetTheta(TS ts,PetscReal theta) 732 { 733 PetscErrorCode ierr; 734 735 PetscFunctionBegin; 736 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 737 ierr = PetscTryMethod(ts,"TSThetaSetTheta_C",(TS,PetscReal),(ts,theta));CHKERRQ(ierr); 738 PetscFunctionReturn(0); 739 } 740 741 #undef __FUNCT__ 742 #define __FUNCT__ "TSThetaGetEndpoint" 743 /*@ 744 TSThetaGetEndpoint - Gets whether to use the endpoint variant of the method (e.g. trapezoid/Crank-Nicolson instead of midpoint rule). 745 746 Not Collective 747 748 Input Parameter: 749 . ts - timestepping context 750 751 Output Parameter: 752 . endpoint - PETSC_TRUE when using the endpoint variant 753 754 Level: Advanced 755 756 .seealso: TSThetaSetEndpoint(), TSTHETA, TSCN 757 @*/ 758 PetscErrorCode TSThetaGetEndpoint(TS ts,PetscBool *endpoint) 759 { 760 PetscErrorCode ierr; 761 762 PetscFunctionBegin; 763 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 764 PetscValidPointer(endpoint,2); 765 ierr = PetscTryMethod(ts,"TSThetaGetEndpoint_C",(TS,PetscBool*),(ts,endpoint));CHKERRQ(ierr); 766 PetscFunctionReturn(0); 767 } 768 769 #undef __FUNCT__ 770 #define __FUNCT__ "TSThetaSetEndpoint" 771 /*@ 772 TSThetaSetEndpoint - Sets whether to use the endpoint variant of the method (e.g. trapezoid/Crank-Nicolson instead of midpoint rule). 773 774 Not Collective 775 776 Input Parameter: 777 + ts - timestepping context 778 - flg - PETSC_TRUE to use the endpoint variant 779 780 Options Database: 781 . -ts_theta_endpoint <flg> 782 783 Level: Intermediate 784 785 .seealso: TSTHETA, TSCN 786 @*/ 787 PetscErrorCode TSThetaSetEndpoint(TS ts,PetscBool flg) 788 { 789 PetscErrorCode ierr; 790 791 PetscFunctionBegin; 792 PetscValidHeaderSpecific(ts,TS_CLASSID,1); 793 ierr = PetscTryMethod(ts,"TSThetaSetEndpoint_C",(TS,PetscBool),(ts,flg));CHKERRQ(ierr); 794 PetscFunctionReturn(0); 795 } 796 797 /* 798 * TSBEULER and TSCN are straightforward specializations of TSTHETA. 799 * The creation functions for these specializations are below. 800 */ 801 802 #undef __FUNCT__ 803 #define __FUNCT__ "TSView_BEuler" 804 static PetscErrorCode TSView_BEuler(TS ts,PetscViewer viewer) 805 { 806 PetscErrorCode ierr; 807 808 PetscFunctionBegin; 809 ierr = SNESView(ts->snes,viewer);CHKERRQ(ierr); 810 PetscFunctionReturn(0); 811 } 812 813 /*MC 814 TSBEULER - ODE solver using the implicit backward Euler method 815 816 Level: beginner 817 818 Notes: 819 TSBEULER is equivalent to TSTHETA with Theta=1.0 820 821 $ -ts_type theta -ts_theta_theta 1. 822 823 .seealso: TSCreate(), TS, TSSetType(), TSEULER, TSCN, TSTHETA 824 825 M*/ 826 #undef __FUNCT__ 827 #define __FUNCT__ "TSCreate_BEuler" 828 PETSC_EXTERN PetscErrorCode TSCreate_BEuler(TS ts) 829 { 830 PetscErrorCode ierr; 831 832 PetscFunctionBegin; 833 ierr = TSCreate_Theta(ts);CHKERRQ(ierr); 834 ierr = TSThetaSetTheta(ts,1.0);CHKERRQ(ierr); 835 ts->ops->view = TSView_BEuler; 836 PetscFunctionReturn(0); 837 } 838 839 #undef __FUNCT__ 840 #define __FUNCT__ "TSView_CN" 841 static PetscErrorCode TSView_CN(TS ts,PetscViewer viewer) 842 { 843 PetscErrorCode ierr; 844 845 PetscFunctionBegin; 846 ierr = SNESView(ts->snes,viewer);CHKERRQ(ierr); 847 PetscFunctionReturn(0); 848 } 849 850 /*MC 851 TSCN - ODE solver using the implicit Crank-Nicolson method. 852 853 Level: beginner 854 855 Notes: 856 TSCN is equivalent to TSTHETA with Theta=0.5 and the "endpoint" option set. I.e. 857 858 $ -ts_type theta -ts_theta_theta 0.5 -ts_theta_endpoint 859 860 .seealso: TSCreate(), TS, TSSetType(), TSBEULER, TSTHETA 861 862 M*/ 863 #undef __FUNCT__ 864 #define __FUNCT__ "TSCreate_CN" 865 PETSC_EXTERN PetscErrorCode TSCreate_CN(TS ts) 866 { 867 PetscErrorCode ierr; 868 869 PetscFunctionBegin; 870 ierr = TSCreate_Theta(ts);CHKERRQ(ierr); 871 ierr = TSThetaSetTheta(ts,0.5);CHKERRQ(ierr); 872 ierr = TSThetaSetEndpoint(ts,PETSC_TRUE);CHKERRQ(ierr); 873 ts->ops->view = TSView_CN; 874 PetscFunctionReturn(0); 875 } 876