1 #include <petsc/private/tsimpl.h> /*I "petscts.h" I*/ 2 #include <petscdm.h> 3 4 static const PetscInt TSEIMEXDefault = 3; 5 6 typedef struct { 7 PetscInt row_ind; /* Return the term T[row_ind][col_ind] */ 8 PetscInt col_ind; /* Return the term T[row_ind][col_ind] */ 9 PetscInt nstages; /* Numbers of stages in current scheme */ 10 PetscInt max_rows; /* Maximum number of rows */ 11 PetscInt *N; /* Harmonic sequence N[max_rows] */ 12 Vec Y; /* States computed during the step, used to complete the step */ 13 Vec Z; /* For shift*(Y-Z) */ 14 Vec *T; /* Working table, size determined by nstages */ 15 Vec YdotRHS; /* f(x) Work vector holding YdotRHS during residual evaluation */ 16 Vec YdotI; /* xdot-g(x) Work vector holding YdotI = G(t,x,xdot) when xdot =0 */ 17 Vec Ydot; /* f(x)+g(x) Work vector */ 18 Vec VecSolPrev; /* Work vector holding the solution from the previous step (used for interpolation) */ 19 PetscReal shift; 20 PetscReal ctime; 21 PetscBool recompute_jacobian; /* Recompute the Jacobian at each stage, default is to freeze the Jacobian at the start of each step */ 22 PetscBool ord_adapt; /* order adapativity */ 23 TSStepStatus status; 24 } TS_EIMEX; 25 26 /* This function is pure */ 27 static PetscInt Map(PetscInt i, PetscInt j, PetscInt s) 28 { 29 return ((2 * s - j + 1) * j / 2 + i - j); 30 } 31 32 static PetscErrorCode TSEvaluateStep_EIMEX(TS ts, PetscInt order, Vec X, PetscBool *done) 33 { 34 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 35 const PetscInt ns = ext->nstages; 36 PetscFunctionBegin; 37 PetscCall(VecCopy(ext->T[Map(ext->row_ind, ext->col_ind, ns)], X)); 38 PetscFunctionReturn(PETSC_SUCCESS); 39 } 40 41 static PetscErrorCode TSStage_EIMEX(TS ts, PetscInt istage) 42 { 43 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 44 PetscReal h; 45 Vec Y = ext->Y, Z = ext->Z; 46 SNES snes; 47 TSAdapt adapt; 48 PetscInt i, its, lits; 49 PetscBool accept; 50 51 PetscFunctionBegin; 52 PetscCall(TSGetSNES(ts, &snes)); 53 h = ts->time_step / ext->N[istage]; /* step size for the istage-th stage */ 54 ext->shift = 1. / h; 55 PetscCall(SNESSetLagJacobian(snes, -2)); /* Recompute the Jacobian on this solve, but not again */ 56 PetscCall(VecCopy(ext->VecSolPrev, Y)); /* Take the previous solution as initial step */ 57 58 for (i = 0; i < ext->N[istage]; i++) { 59 ext->ctime = ts->ptime + h * i; 60 PetscCall(VecCopy(Y, Z)); /* Save the solution of the previous substep */ 61 PetscCall(SNESSolve(snes, NULL, Y)); 62 PetscCall(SNESGetIterationNumber(snes, &its)); 63 PetscCall(SNESGetLinearSolveIterations(snes, &lits)); 64 ts->snes_its += its; 65 ts->ksp_its += lits; 66 PetscCall(TSGetAdapt(ts, &adapt)); 67 PetscCall(TSAdaptCheckStage(adapt, ts, ext->ctime, Y, &accept)); 68 } 69 PetscFunctionReturn(PETSC_SUCCESS); 70 } 71 72 static PetscErrorCode TSStep_EIMEX(TS ts) 73 { 74 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 75 const PetscInt ns = ext->nstages; 76 Vec *T = ext->T, Y = ext->Y; 77 SNES snes; 78 PetscInt i, j; 79 PetscBool accept = PETSC_FALSE; 80 PetscReal alpha, local_error, local_error_a, local_error_r; 81 82 PetscFunctionBegin; 83 PetscCall(TSGetSNES(ts, &snes)); 84 PetscCall(SNESSetType(snes, "ksponly")); 85 ext->status = TS_STEP_INCOMPLETE; 86 87 PetscCall(VecCopy(ts->vec_sol, ext->VecSolPrev)); 88 89 /* Apply n_j steps of the base method to obtain solutions of T(j,1),1<=j<=s */ 90 for (j = 0; j < ns; j++) { 91 PetscCall(TSStage_EIMEX(ts, j)); 92 PetscCall(VecCopy(Y, T[j])); 93 } 94 95 for (i = 1; i < ns; i++) { 96 for (j = i; j < ns; j++) { 97 alpha = -(PetscReal)ext->N[j] / ext->N[j - i]; 98 PetscCall(VecAXPBYPCZ(T[Map(j, i, ns)], alpha, 1.0, 0, T[Map(j, i - 1, ns)], T[Map(j - 1, i - 1, ns)])); /* T[j][i]=alpha*T[j][i-1]+T[j-1][i-1] */ 99 alpha = 1.0 / (1.0 + alpha); 100 PetscCall(VecScale(T[Map(j, i, ns)], alpha)); 101 } 102 } 103 104 PetscCall(TSEvaluateStep(ts, ns, ts->vec_sol, NULL)); /*update ts solution */ 105 106 if (ext->ord_adapt && ext->nstages < ext->max_rows) { 107 accept = PETSC_FALSE; 108 while (!accept && ext->nstages < ext->max_rows) { 109 PetscCall(TSErrorWeightedNorm(ts, ts->vec_sol, T[Map(ext->nstages - 1, ext->nstages - 2, ext->nstages)], ts->adapt->wnormtype, &local_error, &local_error_a, &local_error_r)); 110 accept = (local_error < 1.0) ? PETSC_TRUE : PETSC_FALSE; 111 112 if (!accept) { /* add one more stage*/ 113 PetscCall(TSStage_EIMEX(ts, ext->nstages)); 114 ext->nstages++; 115 ext->row_ind++; 116 ext->col_ind++; 117 /*T table need to be recycled*/ 118 PetscCall(VecDuplicateVecs(ts->vec_sol, (1 + ext->nstages) * ext->nstages / 2, &ext->T)); 119 for (i = 0; i < ext->nstages - 1; i++) { 120 for (j = 0; j <= i; j++) PetscCall(VecCopy(T[Map(i, j, ext->nstages - 1)], ext->T[Map(i, j, ext->nstages)])); 121 } 122 PetscCall(VecDestroyVecs(ext->nstages * (ext->nstages - 1) / 2, &T)); 123 T = ext->T; /*reset the pointer*/ 124 /*recycling finished, store the new solution*/ 125 PetscCall(VecCopy(Y, T[ext->nstages - 1])); 126 /*extrapolation for the newly added stage*/ 127 for (i = 1; i < ext->nstages; i++) { 128 alpha = -(PetscReal)ext->N[ext->nstages - 1] / ext->N[ext->nstages - 1 - i]; 129 PetscCall(VecAXPBYPCZ(T[Map(ext->nstages - 1, i, ext->nstages)], alpha, 1.0, 0, T[Map(ext->nstages - 1, i - 1, ext->nstages)], T[Map(ext->nstages - 1 - 1, i - 1, ext->nstages)])); /*T[ext->nstages-1][i]=alpha*T[ext->nstages-1][i-1]+T[ext->nstages-1-1][i-1]*/ 130 alpha = 1.0 / (1.0 + alpha); 131 PetscCall(VecScale(T[Map(ext->nstages - 1, i, ext->nstages)], alpha)); 132 } 133 /*update ts solution */ 134 PetscCall(TSEvaluateStep(ts, ext->nstages, ts->vec_sol, NULL)); 135 } /*end if !accept*/ 136 } /*end while*/ 137 138 if (ext->nstages == ext->max_rows) PetscCall(PetscInfo(ts, "Max number of rows has been used\n")); 139 } /*end if ext->ord_adapt*/ 140 ts->ptime += ts->time_step; 141 ext->status = TS_STEP_COMPLETE; 142 143 if (ext->status != TS_STEP_COMPLETE && !ts->reason) ts->reason = TS_DIVERGED_STEP_REJECTED; 144 PetscFunctionReturn(PETSC_SUCCESS); 145 } 146 147 /* cubic Hermit spline */ 148 static PetscErrorCode TSInterpolate_EIMEX(TS ts, PetscReal itime, Vec X) 149 { 150 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 151 PetscReal t, a, b; 152 Vec Y0 = ext->VecSolPrev, Y1 = ext->Y, Ydot = ext->Ydot, YdotI = ext->YdotI; 153 const PetscReal h = ts->ptime - ts->ptime_prev; 154 PetscFunctionBegin; 155 t = (itime - ts->ptime + h) / h; 156 /* YdotI = -f(x)-g(x) */ 157 158 PetscCall(VecZeroEntries(Ydot)); 159 PetscCall(TSComputeIFunction(ts, ts->ptime - h, Y0, Ydot, YdotI, PETSC_FALSE)); 160 161 a = 2.0 * t * t * t - 3.0 * t * t + 1.0; 162 b = -(t * t * t - 2.0 * t * t + t) * h; 163 PetscCall(VecAXPBYPCZ(X, a, b, 0.0, Y0, YdotI)); 164 165 PetscCall(TSComputeIFunction(ts, ts->ptime, Y1, Ydot, YdotI, PETSC_FALSE)); 166 a = -2.0 * t * t * t + 3.0 * t * t; 167 b = -(t * t * t - t * t) * h; 168 PetscCall(VecAXPBYPCZ(X, a, b, 1.0, Y1, YdotI)); 169 170 PetscFunctionReturn(PETSC_SUCCESS); 171 } 172 173 static PetscErrorCode TSReset_EIMEX(TS ts) 174 { 175 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 176 PetscInt ns; 177 178 PetscFunctionBegin; 179 ns = ext->nstages; 180 PetscCall(VecDestroyVecs((1 + ns) * ns / 2, &ext->T)); 181 PetscCall(VecDestroy(&ext->Y)); 182 PetscCall(VecDestroy(&ext->Z)); 183 PetscCall(VecDestroy(&ext->YdotRHS)); 184 PetscCall(VecDestroy(&ext->YdotI)); 185 PetscCall(VecDestroy(&ext->Ydot)); 186 PetscCall(VecDestroy(&ext->VecSolPrev)); 187 PetscCall(PetscFree(ext->N)); 188 PetscFunctionReturn(PETSC_SUCCESS); 189 } 190 191 static PetscErrorCode TSDestroy_EIMEX(TS ts) 192 { 193 PetscFunctionBegin; 194 PetscCall(TSReset_EIMEX(ts)); 195 PetscCall(PetscFree(ts->data)); 196 PetscCall(PetscObjectComposeFunction((PetscObject)ts, "TSEIMEXSetMaxRows_C", NULL)); 197 PetscCall(PetscObjectComposeFunction((PetscObject)ts, "TSEIMEXSetRowCol_C", NULL)); 198 PetscCall(PetscObjectComposeFunction((PetscObject)ts, "TSEIMEXSetOrdAdapt_C", NULL)); 199 PetscFunctionReturn(PETSC_SUCCESS); 200 } 201 202 static PetscErrorCode TSEIMEXGetVecs(TS ts, DM dm, Vec *Z, Vec *Ydot, Vec *YdotI, Vec *YdotRHS) 203 { 204 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 205 206 PetscFunctionBegin; 207 if (Z) { 208 if (dm && dm != ts->dm) { 209 PetscCall(DMGetNamedGlobalVector(dm, "TSEIMEX_Z", Z)); 210 } else *Z = ext->Z; 211 } 212 if (Ydot) { 213 if (dm && dm != ts->dm) { 214 PetscCall(DMGetNamedGlobalVector(dm, "TSEIMEX_Ydot", Ydot)); 215 } else *Ydot = ext->Ydot; 216 } 217 if (YdotI) { 218 if (dm && dm != ts->dm) { 219 PetscCall(DMGetNamedGlobalVector(dm, "TSEIMEX_YdotI", YdotI)); 220 } else *YdotI = ext->YdotI; 221 } 222 if (YdotRHS) { 223 if (dm && dm != ts->dm) { 224 PetscCall(DMGetNamedGlobalVector(dm, "TSEIMEX_YdotRHS", YdotRHS)); 225 } else *YdotRHS = ext->YdotRHS; 226 } 227 PetscFunctionReturn(PETSC_SUCCESS); 228 } 229 230 static PetscErrorCode TSEIMEXRestoreVecs(TS ts, DM dm, Vec *Z, Vec *Ydot, Vec *YdotI, Vec *YdotRHS) 231 { 232 PetscFunctionBegin; 233 if (Z) { 234 if (dm && dm != ts->dm) PetscCall(DMRestoreNamedGlobalVector(dm, "TSEIMEX_Z", Z)); 235 } 236 if (Ydot) { 237 if (dm && dm != ts->dm) PetscCall(DMRestoreNamedGlobalVector(dm, "TSEIMEX_Ydot", Ydot)); 238 } 239 if (YdotI) { 240 if (dm && dm != ts->dm) PetscCall(DMRestoreNamedGlobalVector(dm, "TSEIMEX_YdotI", YdotI)); 241 } 242 if (YdotRHS) { 243 if (dm && dm != ts->dm) PetscCall(DMRestoreNamedGlobalVector(dm, "TSEIMEX_YdotRHS", YdotRHS)); 244 } 245 PetscFunctionReturn(PETSC_SUCCESS); 246 } 247 248 /* 249 This defines the nonlinear equation that is to be solved with SNES 250 Fn[t0+Theta*dt, U, (U-U0)*shift] = 0 251 In the case of Backward Euler, Fn = (U-U0)/h-g(t1,U)) 252 Since FormIFunction calculates G = ydot - g(t,y), ydot will be set to (U-U0)/h 253 */ 254 static PetscErrorCode SNESTSFormFunction_EIMEX(SNES snes, Vec X, Vec G, TS ts) 255 { 256 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 257 Vec Ydot, Z; 258 DM dm, dmsave; 259 260 PetscFunctionBegin; 261 PetscCall(VecZeroEntries(G)); 262 263 PetscCall(SNESGetDM(snes, &dm)); 264 PetscCall(TSEIMEXGetVecs(ts, dm, &Z, &Ydot, NULL, NULL)); 265 PetscCall(VecZeroEntries(Ydot)); 266 dmsave = ts->dm; 267 ts->dm = dm; 268 PetscCall(TSComputeIFunction(ts, ext->ctime, X, Ydot, G, PETSC_FALSE)); 269 /* PETSC_FALSE indicates non-imex, adding explicit RHS to the implicit I function. */ 270 PetscCall(VecCopy(G, Ydot)); 271 ts->dm = dmsave; 272 PetscCall(TSEIMEXRestoreVecs(ts, dm, &Z, &Ydot, NULL, NULL)); 273 274 PetscFunctionReturn(PETSC_SUCCESS); 275 } 276 277 /* 278 This defined the Jacobian matrix for SNES. Jn = (I/h-g'(t,y)) 279 */ 280 static PetscErrorCode SNESTSFormJacobian_EIMEX(SNES snes, Vec X, Mat A, Mat B, TS ts) 281 { 282 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 283 Vec Ydot; 284 DM dm, dmsave; 285 PetscFunctionBegin; 286 PetscCall(SNESGetDM(snes, &dm)); 287 PetscCall(TSEIMEXGetVecs(ts, dm, NULL, &Ydot, NULL, NULL)); 288 /* PetscCall(VecZeroEntries(Ydot)); */ 289 /* ext->Ydot have already been computed in SNESTSFormFunction_EIMEX (SNES guarantees this) */ 290 dmsave = ts->dm; 291 ts->dm = dm; 292 PetscCall(TSComputeIJacobian(ts, ts->ptime, X, Ydot, ext->shift, A, B, PETSC_TRUE)); 293 ts->dm = dmsave; 294 PetscCall(TSEIMEXRestoreVecs(ts, dm, NULL, &Ydot, NULL, NULL)); 295 PetscFunctionReturn(PETSC_SUCCESS); 296 } 297 298 static PetscErrorCode DMCoarsenHook_TSEIMEX(DM fine, DM coarse, void *ctx) 299 { 300 PetscFunctionBegin; 301 PetscFunctionReturn(PETSC_SUCCESS); 302 } 303 304 static PetscErrorCode DMRestrictHook_TSEIMEX(DM fine, Mat restrct, Vec rscale, Mat inject, DM coarse, void *ctx) 305 { 306 TS ts = (TS)ctx; 307 Vec Z, Z_c; 308 309 PetscFunctionBegin; 310 PetscCall(TSEIMEXGetVecs(ts, fine, &Z, NULL, NULL, NULL)); 311 PetscCall(TSEIMEXGetVecs(ts, coarse, &Z_c, NULL, NULL, NULL)); 312 PetscCall(MatRestrict(restrct, Z, Z_c)); 313 PetscCall(VecPointwiseMult(Z_c, rscale, Z_c)); 314 PetscCall(TSEIMEXRestoreVecs(ts, fine, &Z, NULL, NULL, NULL)); 315 PetscCall(TSEIMEXRestoreVecs(ts, coarse, &Z_c, NULL, NULL, NULL)); 316 PetscFunctionReturn(PETSC_SUCCESS); 317 } 318 319 static PetscErrorCode TSSetUp_EIMEX(TS ts) 320 { 321 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 322 DM dm; 323 324 PetscFunctionBegin; 325 if (!ext->N) { /* ext->max_rows not set */ 326 PetscCall(TSEIMEXSetMaxRows(ts, TSEIMEXDefault)); 327 } 328 if (-1 == ext->row_ind && -1 == ext->col_ind) { 329 PetscCall(TSEIMEXSetRowCol(ts, ext->max_rows, ext->max_rows)); 330 } else { /* ext->row_ind and col_ind already set */ 331 if (ext->ord_adapt) PetscCall(PetscInfo(ts, "Order adaptivity is enabled and TSEIMEXSetRowCol or -ts_eimex_row_col option will take no effect\n")); 332 } 333 334 if (ext->ord_adapt) { 335 ext->nstages = 2; /* Start with the 2-stage scheme */ 336 PetscCall(TSEIMEXSetRowCol(ts, ext->nstages, ext->nstages)); 337 } else { 338 ext->nstages = ext->max_rows; /* by default nstages is the same as max_rows, this can be changed by setting order adaptivity */ 339 } 340 341 PetscCall(TSGetAdapt(ts, &ts->adapt)); 342 343 PetscCall(VecDuplicateVecs(ts->vec_sol, (1 + ext->nstages) * ext->nstages / 2, &ext->T)); /* full T table */ 344 PetscCall(VecDuplicate(ts->vec_sol, &ext->YdotI)); 345 PetscCall(VecDuplicate(ts->vec_sol, &ext->YdotRHS)); 346 PetscCall(VecDuplicate(ts->vec_sol, &ext->Ydot)); 347 PetscCall(VecDuplicate(ts->vec_sol, &ext->VecSolPrev)); 348 PetscCall(VecDuplicate(ts->vec_sol, &ext->Y)); 349 PetscCall(VecDuplicate(ts->vec_sol, &ext->Z)); 350 PetscCall(TSGetDM(ts, &dm)); 351 if (dm) PetscCall(DMCoarsenHookAdd(dm, DMCoarsenHook_TSEIMEX, DMRestrictHook_TSEIMEX, ts)); 352 PetscFunctionReturn(PETSC_SUCCESS); 353 } 354 355 static PetscErrorCode TSSetFromOptions_EIMEX(TS ts, PetscOptionItems *PetscOptionsObject) 356 { 357 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 358 PetscInt tindex[2]; 359 PetscInt np = 2, nrows = TSEIMEXDefault; 360 361 PetscFunctionBegin; 362 tindex[0] = TSEIMEXDefault; 363 tindex[1] = TSEIMEXDefault; 364 PetscOptionsHeadBegin(PetscOptionsObject, "EIMEX ODE solver options"); 365 { 366 PetscBool flg; 367 PetscCall(PetscOptionsInt("-ts_eimex_max_rows", "Define the maximum number of rows used", "TSEIMEXSetMaxRows", nrows, &nrows, &flg)); /* default value 3 */ 368 if (flg) PetscCall(TSEIMEXSetMaxRows(ts, nrows)); 369 PetscCall(PetscOptionsIntArray("-ts_eimex_row_col", "Return the specific term in the T table", "TSEIMEXSetRowCol", tindex, &np, &flg)); 370 if (flg) PetscCall(TSEIMEXSetRowCol(ts, tindex[0], tindex[1])); 371 PetscCall(PetscOptionsBool("-ts_eimex_order_adapt", "Solve the problem with adaptive order", "TSEIMEXSetOrdAdapt", ext->ord_adapt, &ext->ord_adapt, NULL)); 372 } 373 PetscOptionsHeadEnd(); 374 PetscFunctionReturn(PETSC_SUCCESS); 375 } 376 377 static PetscErrorCode TSView_EIMEX(TS ts, PetscViewer viewer) 378 { 379 PetscFunctionBegin; 380 PetscFunctionReturn(PETSC_SUCCESS); 381 } 382 383 /*@C 384 TSEIMEXSetMaxRows - Set the maximum number of rows for `TSEIMEX` schemes 385 386 Logically Collective 387 388 Input Parameters: 389 + ts - timestepping context 390 - nrows - maximum number of rows 391 392 Level: intermediate 393 394 .seealso: [](ch_ts), `TSEIMEXSetRowCol()`, `TSEIMEXSetOrdAdapt()`, `TSEIMEX` 395 @*/ 396 PetscErrorCode TSEIMEXSetMaxRows(TS ts, PetscInt nrows) 397 { 398 PetscFunctionBegin; 399 PetscValidHeaderSpecific(ts, TS_CLASSID, 1); 400 PetscTryMethod(ts, "TSEIMEXSetMaxRows_C", (TS, PetscInt), (ts, nrows)); 401 PetscFunctionReturn(PETSC_SUCCESS); 402 } 403 404 /*@C 405 TSEIMEXSetRowCol - Set the number of rows and the number of columns for the tableau that represents the T solution in the `TSEIMEX` scheme 406 407 Logically Collective 408 409 Input Parameters: 410 + ts - timestepping context 411 . row - the row 412 - col - the column 413 414 Level: intermediate 415 416 .seealso: [](ch_ts), `TSEIMEXSetMaxRows()`, `TSEIMEXSetOrdAdapt()`, `TSEIMEX` 417 @*/ 418 PetscErrorCode TSEIMEXSetRowCol(TS ts, PetscInt row, PetscInt col) 419 { 420 PetscFunctionBegin; 421 PetscValidHeaderSpecific(ts, TS_CLASSID, 1); 422 PetscTryMethod(ts, "TSEIMEXSetRowCol_C", (TS, PetscInt, PetscInt), (ts, row, col)); 423 PetscFunctionReturn(PETSC_SUCCESS); 424 } 425 426 /*@C 427 TSEIMEXSetOrdAdapt - Set the order adaptativity for the `TSEIMEX` schemes 428 429 Logically Collective 430 431 Input Parameters: 432 + ts - timestepping context 433 - flg - index in the T table 434 435 Level: intermediate 436 437 .seealso: [](ch_ts), `TSEIMEXSetRowCol()`, `TSEIMEX` 438 @*/ 439 PetscErrorCode TSEIMEXSetOrdAdapt(TS ts, PetscBool flg) 440 { 441 PetscFunctionBegin; 442 PetscValidHeaderSpecific(ts, TS_CLASSID, 1); 443 PetscTryMethod(ts, "TSEIMEXSetOrdAdapt_C", (TS, PetscBool), (ts, flg)); 444 PetscFunctionReturn(PETSC_SUCCESS); 445 } 446 447 static PetscErrorCode TSEIMEXSetMaxRows_EIMEX(TS ts, PetscInt nrows) 448 { 449 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 450 PetscInt i; 451 452 PetscFunctionBegin; 453 PetscCheck(nrows >= 0 && nrows <= 100, ((PetscObject)ts)->comm, PETSC_ERR_ARG_OUTOFRANGE, "Max number of rows (current value %" PetscInt_FMT ") should be an integer number between 1 and 100", nrows); 454 PetscCall(PetscFree(ext->N)); 455 ext->max_rows = nrows; 456 PetscCall(PetscMalloc1(nrows, &ext->N)); 457 for (i = 0; i < nrows; i++) ext->N[i] = i + 1; 458 PetscFunctionReturn(PETSC_SUCCESS); 459 } 460 461 static PetscErrorCode TSEIMEXSetRowCol_EIMEX(TS ts, PetscInt row, PetscInt col) 462 { 463 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 464 465 PetscFunctionBegin; 466 PetscCheck(row >= 1 && col >= 1, ((PetscObject)ts)->comm, PETSC_ERR_ARG_OUTOFRANGE, "The row or column index (current value %" PetscInt_FMT ",%" PetscInt_FMT ") should not be less than 1 ", row, col); 467 PetscCheck(row <= ext->max_rows && col <= ext->max_rows, ((PetscObject)ts)->comm, PETSC_ERR_ARG_OUTOFRANGE, "The row or column index (current value %" PetscInt_FMT ",%" PetscInt_FMT ") exceeds the maximum number of rows %" PetscInt_FMT, row, col, 468 ext->max_rows); 469 PetscCheck(col <= row, ((PetscObject)ts)->comm, PETSC_ERR_ARG_OUTOFRANGE, "The column index (%" PetscInt_FMT ") exceeds the row index (%" PetscInt_FMT ")", col, row); 470 471 ext->row_ind = row - 1; 472 ext->col_ind = col - 1; /* Array index in C starts from 0 */ 473 PetscFunctionReturn(PETSC_SUCCESS); 474 } 475 476 static PetscErrorCode TSEIMEXSetOrdAdapt_EIMEX(TS ts, PetscBool flg) 477 { 478 TS_EIMEX *ext = (TS_EIMEX *)ts->data; 479 PetscFunctionBegin; 480 ext->ord_adapt = flg; 481 PetscFunctionReturn(PETSC_SUCCESS); 482 } 483 484 /*MC 485 TSEIMEX - Time stepping with Extrapolated IMEX methods {cite}`constantinescu_a2010a`. 486 487 These methods are intended for problems with well-separated time scales, especially when a slow scale is strongly nonlinear such that it 488 is expensive to solve with a fully implicit method. The user should provide the stiff part of the equation using `TSSetIFunction()` and the 489 non-stiff part with `TSSetRHSFunction()`. 490 491 Level: beginner 492 493 Notes: 494 The default is a 3-stage scheme, it can be changed with `TSEIMEXSetMaxRows()` or -ts_eimex_max_rows 495 496 This method currently only works with ODE, for which the stiff part $ G(t,X,Xdot) $ has the form $ Xdot + Ghat(t,X)$. 497 498 The general system is written as 499 500 $$ 501 G(t,X,Xdot) = F(t,X) 502 $$ 503 504 where G represents the stiff part and F represents the non-stiff part. The user should provide the stiff part 505 of the equation using TSSetIFunction() and the non-stiff part with `TSSetRHSFunction()`. 506 This method is designed to be linearly implicit on G and can use an approximate and lagged Jacobian. 507 508 Another common form for the system is 509 510 $$ 511 y'=f(x)+g(x) 512 $$ 513 514 The relationship between F,G and f,g is 515 516 $$ 517 G = y'-g(x), F = f(x) 518 $$ 519 520 .seealso: [](ch_ts), `TSCreate()`, `TS`, `TSSetType()`, `TSEIMEXSetMaxRows()`, `TSEIMEXSetRowCol()`, `TSEIMEXSetOrdAdapt()`, `TSType` 521 M*/ 522 PETSC_EXTERN PetscErrorCode TSCreate_EIMEX(TS ts) 523 { 524 TS_EIMEX *ext; 525 526 PetscFunctionBegin; 527 528 ts->ops->reset = TSReset_EIMEX; 529 ts->ops->destroy = TSDestroy_EIMEX; 530 ts->ops->view = TSView_EIMEX; 531 ts->ops->setup = TSSetUp_EIMEX; 532 ts->ops->step = TSStep_EIMEX; 533 ts->ops->interpolate = TSInterpolate_EIMEX; 534 ts->ops->evaluatestep = TSEvaluateStep_EIMEX; 535 ts->ops->setfromoptions = TSSetFromOptions_EIMEX; 536 ts->ops->snesfunction = SNESTSFormFunction_EIMEX; 537 ts->ops->snesjacobian = SNESTSFormJacobian_EIMEX; 538 ts->default_adapt_type = TSADAPTNONE; 539 540 ts->usessnes = PETSC_TRUE; 541 542 PetscCall(PetscNew(&ext)); 543 ts->data = (void *)ext; 544 545 ext->ord_adapt = PETSC_FALSE; /* By default, no order adapativity */ 546 ext->row_ind = -1; 547 ext->col_ind = -1; 548 ext->max_rows = TSEIMEXDefault; 549 ext->nstages = TSEIMEXDefault; 550 551 PetscCall(PetscObjectComposeFunction((PetscObject)ts, "TSEIMEXSetMaxRows_C", TSEIMEXSetMaxRows_EIMEX)); 552 PetscCall(PetscObjectComposeFunction((PetscObject)ts, "TSEIMEXSetRowCol_C", TSEIMEXSetRowCol_EIMEX)); 553 PetscCall(PetscObjectComposeFunction((PetscObject)ts, "TSEIMEXSetOrdAdapt_C", TSEIMEXSetOrdAdapt_EIMEX)); 554 PetscFunctionReturn(PETSC_SUCCESS); 555 } 556