navier_stokes_elements.h
Go to the documentation of this file.
1 // LIC// ====================================================================
2 // LIC// This file forms part of oomph-lib, the object-oriented,
3 // LIC// multi-physics finite-element library, available
4 // LIC// at http://www.oomph-lib.org.
5 // LIC//
6 // LIC// Copyright (C) 2006-2023 Matthias Heil and Andrew Hazel
7 // LIC//
8 // LIC// This library is free software; you can redistribute it and/or
9 // LIC// modify it under the terms of the GNU Lesser General Public
10 // LIC// License as published by the Free Software Foundation; either
11 // LIC// version 2.1 of the License, or (at your option) any later version.
12 // LIC//
13 // LIC// This library is distributed in the hope that it will be useful,
14 // LIC// but WITHOUT ANY WARRANTY; without even the implied warranty of
15 // LIC// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16 // LIC// Lesser General Public License for more details.
17 // LIC//
18 // LIC// You should have received a copy of the GNU Lesser General Public
19 // LIC// License along with this library; if not, write to the Free Software
20 // LIC// Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
21 // LIC// 02110-1301 USA.
22 // LIC//
23 // LIC// The authors may be contacted at oomph-lib@maths.man.ac.uk.
24 // LIC//
25 // LIC//====================================================================
26 // Header file for Navier Stokes elements
27 
28 #ifndef OOMPH_NAVIER_STOKES_ELEMENTS_HEADER
29 #define OOMPH_NAVIER_STOKES_ELEMENTS_HEADER
30 
31 // Config header generated by autoconfig
32 #ifdef HAVE_CONFIG_H
33 #include <oomph-lib-config.h>
34 #endif
35 
36 
37 // OOMPH-LIB headers
38 #include "../generic/Qelements.h"
39 #include "../generic/fsi.h"
40 #include "../generic/projection.h"
41 
42 #include <algorithm>
43 #include <iterator>
44 
45 namespace oomph
46 {
47  //======================================================================
48  /// Helper class for elements that impose Robin boundary conditions
49  /// on pressure advection diffusion problem required by Fp preconditioner
50  /// (class used to get around some templating issues)
51  //======================================================================
53  {
54  public:
55  /// Constructor
57 
58  /// Empty virtual destructor
60 
61  /// This function returns the residuals for the
62  /// traction function. flag=1 (or 0): do (or don't) compute the
63  /// Jacobian as well.
65  Vector<double>& residuals,
66  DenseMatrix<double>& jacobian,
67  unsigned flag) = 0;
68  };
69 
70 
71  /// ////////////////////////////////////////////////////////////////////
72  /// ////////////////////////////////////////////////////////////////////
73  /// ////////////////////////////////////////////////////////////////////
74 
75 
76  //======================================================================
77  /// A class for elements that allow the imposition of Robin boundary
78  /// conditions for the pressure advection diffusion problem in the
79  /// Fp preconditioner.
80  /// The geometrical information can be read from the FaceGeometry<ELEMENT>
81  /// class and and thus, we can be generic enough without the need to have
82  /// a separate equations class
83  //======================================================================
84  template<class ELEMENT>
86  : public virtual FaceGeometry<ELEMENT>,
87  public virtual FaceElement,
89  {
90  public:
91  /// Constructor, which takes a "bulk" element and the value of the index
92  /// and its limit. Optional boolean flag indicates if it's called
93  // refineable constructor.
95  FiniteElement* const& element_pt,
96  const int& face_index,
97  const bool& called_from_refineable_constructor = false)
98  : FaceGeometry<ELEMENT>(), FaceElement()
99  {
100  // Attach the geometrical information to the element. N.B. This function
101  // also assigns nbulk_value from the required_nvalue of the bulk element
102  element_pt->build_face_element(face_index, this);
103 
104 #ifdef PARANOID
105  {
106  // Check that the element is not a refineable 3d element
107  if (!called_from_refineable_constructor)
108  {
109  // If it's three-d
110  if (element_pt->dim() == 3)
111  {
112  // Is it refineable
113  RefineableElement* ref_el_pt =
114  dynamic_cast<RefineableElement*>(element_pt);
115  if (ref_el_pt != 0)
116  {
117  if (this->has_hanging_nodes())
118  {
119  throw OomphLibError("This flux element will not work correctly "
120  "if nodes are hanging\n",
121  OOMPH_CURRENT_FUNCTION,
122  OOMPH_EXCEPTION_LOCATION);
123  }
124  }
125  }
126  }
127  }
128 #endif
129  }
130 
131  /// Empty destructor
133 
134  /// This function returns the residuals for the
135  /// traction function. flag=1 (or 0): do (or don't) compute the
136  /// Jacobian as well.
138  Vector<double>& residuals, DenseMatrix<double>& jacobian, unsigned flag);
139 
140 
141  /// This function returns just the residuals
143  {
144  std::ostringstream error_message;
145  error_message
146  << "fill_in_contribution_to_residuals() must not be called directly.\n"
147  << "since it uses the local equation numbering of the bulk element\n"
148  << "which calls the relevant helper function directly.\n";
149  throw OomphLibError(
150  error_message.str(), OOMPH_CURRENT_FUNCTION, OOMPH_EXCEPTION_LOCATION);
151  }
152 
153  /// This function returns the residuals and the jacobian
155  DenseMatrix<double>& jacobian)
156  {
157  std::ostringstream error_message;
158  error_message
159  << "fill_in_contribution_to_jacobian() must not be called directly.\n"
160  << "since it uses the local equation numbering of the bulk element\n"
161  << "which calls the relevant helper function directly.\n";
162  throw OomphLibError(
163  error_message.str(), OOMPH_CURRENT_FUNCTION, OOMPH_EXCEPTION_LOCATION);
164  }
165 
166  /// Overload the output function
167  void output(std::ostream& outfile)
168  {
169  FiniteElement::output(outfile);
170  }
171 
172  /// Output function: x,y,[z],u,v,[w],p in tecplot format
173  void output(std::ostream& outfile, const unsigned& nplot)
174  {
175  FiniteElement::output(outfile, nplot);
176  }
177  };
178 
179  /// ////////////////////////////////////////////////////////////////////
180  /// ////////////////////////////////////////////////////////////////////
181  /// ////////////////////////////////////////////////////////////////////
182 
183 
184  //============================================================================
185  /// Get residuals and Jacobian of Robin boundary conditions in pressure
186  /// advection diffusion problem in Fp preconditoner
187  //============================================================================
188  template<class ELEMENT>
191  Vector<double>& residuals, DenseMatrix<double>& jacobian, unsigned flag)
192  {
193  // Storage for local coordinates in FaceElement and associted bulk element
194  unsigned my_dim = this->dim();
195  Vector<double> s(my_dim);
196  Vector<double> s_bulk(my_dim + 1);
197 
198  // Storage for outer unit normal
199  Vector<double> unit_normal(my_dim + 1);
200 
201  // Storage for veloc in bulk element
202  Vector<double> veloc(my_dim + 1);
203 
204  // Set the value of n_intpt
205  unsigned n_intpt = this->integral_pt()->nweight();
206 
207  // Integers to store local equation numbers
208  int local_eqn = 0;
209  int local_unknown = 0;
210 
211  // Get cast bulk element
212  ELEMENT* bulk_el_pt = dynamic_cast<ELEMENT*>(this->bulk_element_pt());
213 
214  // Find out how many pressure dofs there are in the bulk element
215  unsigned n_pres = bulk_el_pt->npres_nst();
216 
217  // Get the Reynolds number from the bulk element
218  double re = bulk_el_pt->re();
219 
220  // Set up memory for pressure shape and test functions
221  Shape psip(n_pres), testp(n_pres);
222 
223  // Loop over the integration points
224  for (unsigned ipt = 0; ipt < n_intpt; ipt++)
225  {
226  // Get the integral weight
227  double w = this->integral_pt()->weight(ipt);
228 
229  // Assign values of local coordinate in FaceElement
230  for (unsigned i = 0; i < my_dim; i++)
231  s[i] = this->integral_pt()->knot(ipt, i);
232 
233  // Find corresponding coordinate in the the bulk element
234  s_bulk = this->local_coordinate_in_bulk(s);
235 
236  /// Get outer unit normal
237  this->outer_unit_normal(ipt, unit_normal);
238 
239  // Get velocity in bulk element
240  bulk_el_pt->interpolated_u_nst(s_bulk, veloc);
241 
242  // Get normal component of veloc
243  double flux = 0.0;
244  for (unsigned i = 0; i < my_dim + 1; i++)
245  {
246  flux += veloc[i] * unit_normal[i];
247  }
248 
249  // Modify bc: If outflow (flux>0) apply Neumann condition instead
250  if (flux > 0.0) flux = 0.0;
251 
252  // Get pressure
253  double interpolated_press = bulk_el_pt->interpolated_p_nst(s_bulk);
254 
255  // Call the pressure shape and test functions in bulk element
256  bulk_el_pt->pshape_nst(s_bulk, psip, testp);
257 
258  // Find the Jacobian of the mapping within the FaceElement
259  double J = this->J_eulerian(s);
260 
261  // Premultiply the weights and the Jacobian
262  double W = w * J;
263 
264  // Loop over the pressure shape functions in bulk
265  //(wasteful but they'll be zero on the boundary)
266  for (unsigned l = 0; l < n_pres; l++)
267  {
268  local_eqn = bulk_el_pt->p_local_eqn(l);
269 
270  // If not a boundary conditions
271  if (local_eqn >= 0)
272  {
273  residuals[local_eqn] -= re * flux * interpolated_press * testp[l] * W;
274 
275  // Jacobian too?
276  if (flag)
277  {
278  // Loop over the shape functions in bulk
279  for (unsigned l2 = 0; l2 < n_pres; l2++)
280  {
281  local_unknown = bulk_el_pt->p_local_eqn(l2);
282 
283  // If not a boundary conditions
284  if (local_unknown >= 0)
285  {
286  jacobian(local_eqn, local_unknown) -=
287  re * flux * psip[l2] * testp[l] * W;
288  }
289  }
290  } /*End of Jacobian calculation*/
291  } // End of if not boundary condition
292  } // End of loop over l
293  }
294  }
295 
296  /// ////////////////////////////////////////////////////////////////////
297  /// ////////////////////////////////////////////////////////////////////
298  /// ////////////////////////////////////////////////////////////////////
299 
300 
301  //======================================================================
302  /// Template-free base class for Navier-Stokes equations to avoid
303  /// casting problems
304  //======================================================================
307  public virtual FiniteElement
308  {
309  public:
310  /// Constructor (empty)
312 
313  /// Virtual destructor (empty)
315 
316  /// Compute the residuals for the associated pressure advection
317  /// diffusion problem. Used by the Fp preconditioner.
319  Vector<double>& residuals) = 0;
320 
321  /// Compute the residuals and Jacobian for the associated
322  /// pressure advection diffusion problem. Used by the Fp preconditioner.
324  Vector<double>& residuals, DenseMatrix<double>& jacobian) = 0;
325 
326  /// Return the index at which the pressure is stored if it is
327  /// stored at the nodes. If not stored at the nodes this will return
328  /// a negative number.
329  virtual int p_nodal_index_nst() const = 0;
330 
331  /// Access function for the local equation number information for
332  /// the pressure.
333  /// p_local_eqn[n] = local equation number or < 0 if pinned
334  virtual int p_local_eqn(const unsigned& n) const = 0;
335 
336  /// Global eqn number of pressure dof that's pinned in pressure
337  /// adv diff problem
338  virtual int& pinned_fp_pressure_eqn() = 0;
339 
340 
341  /// Pin all non-pressure dofs and backup eqn numbers of all Data
343  std::map<Data*, std::vector<int>>& eqn_number_backup) = 0;
344 
345  /// Build FaceElements that apply the Robin boundary condition
346  /// to the pressure advection diffusion problem required by
347  /// Fp preconditioner
349  const unsigned& face_index) = 0;
350 
351  /// Delete the FaceElements that apply the Robin boundary condition
352  /// to the pressure advection diffusion problem required by
353  /// Fp preconditioner
355 
356 
357  /// Compute the diagonal of the velocity/pressure mass matrices.
358  /// If which one=0, both are computed, otherwise only the pressure
359  /// (which_one=1) or the velocity mass matrix (which_one=2 -- the
360  /// LSC version of the preconditioner only needs that one)
362  Vector<double>& press_mass_diag,
363  Vector<double>& veloc_mass_diag,
364  const unsigned& which_one = 0) = 0;
365  };
366 
367 
368  /// ////////////////////////////////////////////////////////////////////
369  /// ////////////////////////////////////////////////////////////////////
370  /// ////////////////////////////////////////////////////////////////////
371 
372 
373  //======================================================================
374  /// A class for elements that solve the cartesian Navier--Stokes equations,
375  /// templated by the dimension DIM.
376  /// This contains the generic maths -- any concrete implementation must
377  /// be derived from this.
378  ///
379  /// We're solving:
380  ///
381  /// \f$ { Re \left( St \frac{\partial u_i}{\partial t} + (u_j - u_j^{M}) \frac{\partial u_i}{\partial x_j} \right) = - \frac{\partial p}{\partial x_i} - R_\rho B_i(x_j) - \frac{Re}{Fr} G_i + \frac{\partial }{\partial x_j} \left[ R_\mu \left( \frac{\partial u_i}{\partial x_j} + \frac{\partial u_j}{\partial x_i} \right) \right] } \f$
382  ///
383  /// and
384  ///
385  /// \f$ { \frac{\partial u_i}{\partial x_i} = Q } \f$
386  ///
387  /// We also provide all functions required to use this element
388  /// in FSI problems, by deriving it from the FSIFluidElement base
389  /// class.
390  //======================================================================
391  template<unsigned DIM>
393  : public virtual FSIFluidElement,
395  {
396  public:
397  /// Function pointer to body force function fct(t,x,f(x))
398  /// x is a Vector!
399  typedef void (*NavierStokesBodyForceFctPt)(const double& time,
400  const Vector<double>& x,
401  Vector<double>& body_force);
402 
403  /// Function pointer to source function fct(t,x)
404  /// x is a Vector!
405  typedef double (*NavierStokesSourceFctPt)(const double& time,
406  const Vector<double>& x);
407 
408 
409  /// Function pointer to source function fct(x) for the
410  /// pressure advection diffusion equation (only used during
411  /// validation!). x is a Vector!
413  const Vector<double>& x);
414 
415  private:
416  /// Static "magic" number that indicates that the pressure is
417  /// not stored at a node
419 
420  /// Static default value for the physical constants (all initialised to
421  /// zero)
423 
424  /// Static default value for the physical ratios (all are initialised to
425  /// one)
427 
428  /// Static default value for the gravity vector
430 
431  protected:
432  // Physical constants
433 
434  /// Pointer to the viscosity ratio (relative to the
435  /// viscosity used in the definition of the Reynolds number)
437 
438  /// Pointer to the density ratio (relative to the density used in the
439  /// definition of the Reynolds number)
441 
442  // Pointers to global physical constants
443 
444  /// Pointer to global Reynolds number
445  double* Re_pt;
446 
447  /// Pointer to global Reynolds number x Strouhal number (=Womersley)
448  double* ReSt_pt;
449 
450  /// Pointer to global Reynolds number x inverse Froude number
451  /// (= Bond number / Capillary number)
452  double* ReInvFr_pt;
453 
454  /// Pointer to global gravity Vector
456 
457  /// Pointer to body force function
459 
460  /// Pointer to volumetric source function
462 
463  /// Pointer to source function pressure advection diffusion equation
464  /// (only to be used during validation)
466 
467  /// Boolean flag to indicate if ALE formulation is disabled when
468  /// time-derivatives are computed. Only set to true if you're sure
469  /// that the mesh is stationary.
471 
472  /// Storage for FaceElements that apply Robin BC for pressure adv
473  /// diff equation used in Fp preconditioner.
476 
477  /// Global eqn number of pressure dof that's pinned in
478  /// pressure advection diffusion problem (defaults to -1)
480 
481  /// Compute the shape functions and derivatives
482  /// w.r.t. global coords at local coordinate s.
483  /// Return Jacobian of mapping between local and global coordinates.
485  Shape& psi,
486  DShape& dpsidx,
487  Shape& test,
488  DShape& dtestdx) const = 0;
489 
490  /// Compute the shape functions and derivatives
491  /// w.r.t. global coords at ipt-th integration point
492  /// Return Jacobian of mapping between local and global coordinates.
494  const unsigned& ipt,
495  Shape& psi,
496  DShape& dpsidx,
497  Shape& test,
498  DShape& dtestdx) const = 0;
499 
500  /// Shape/test functions and derivs w.r.t. to global coords at
501  /// integration point ipt; return Jacobian of mapping (J). Also compute
502  /// derivatives of dpsidx, dtestdx and J w.r.t. nodal coordinates.
504  const unsigned& ipt,
505  Shape& psi,
506  DShape& dpsidx,
507  RankFourTensor<double>& d_dpsidx_dX,
508  Shape& test,
509  DShape& dtestdx,
510  RankFourTensor<double>& d_dtestdx_dX,
511  DenseMatrix<double>& djacobian_dX) const = 0;
512 
513  /// Calculate the body force at a given time and local and/or
514  /// Eulerian position. This function is virtual so that it can be
515  /// overloaded in multi-physics elements where the body force might
516  /// depend on another variable.
517  virtual void get_body_force_nst(const double& time,
518  const unsigned& ipt,
519  const Vector<double>& s,
520  const Vector<double>& x,
521  Vector<double>& result)
522  {
523  // If the function pointer is zero return zero
524  if (Body_force_fct_pt == 0)
525  {
526  // Loop over dimensions and set body forces to zero
527  for (unsigned i = 0; i < DIM; i++)
528  {
529  result[i] = 0.0;
530  }
531  }
532  // Otherwise call the function
533  else
534  {
535  (*Body_force_fct_pt)(time, x, result);
536  }
537  }
538 
539  /// Get gradient of body force term at (Eulerian) position x. This function
540  /// is virtual to allow overloading in multi-physics problems where the
541  /// strength of the source function might be determined by another system of
542  /// equations. Computed via function pointer (if set) or by finite
543  /// differencing (default)
544  inline virtual void get_body_force_gradient_nst(
545  const double& time,
546  const unsigned& ipt,
547  const Vector<double>& s,
548  const Vector<double>& x,
549  DenseMatrix<double>& d_body_force_dx)
550  {
551  // hierher: Implement function pointer version
552  /* //If no gradient function has been set, FD it */
553  /* if(Body_force_fct_gradient_pt==0) */
554  {
555  // Reference value
556  Vector<double> body_force(DIM, 0.0);
557  get_body_force_nst(time, ipt, s, x, body_force);
558 
559  // FD it
561  Vector<double> body_force_pls(DIM, 0.0);
562  Vector<double> x_pls(x);
563  for (unsigned i = 0; i < DIM; i++)
564  {
565  x_pls[i] += eps_fd;
566  get_body_force_nst(time, ipt, s, x_pls, body_force_pls);
567  for (unsigned j = 0; j < DIM; j++)
568  {
569  d_body_force_dx(j, i) =
570  (body_force_pls[j] - body_force[j]) / eps_fd;
571  }
572  x_pls[i] = x[i];
573  }
574  }
575  /* else */
576  /* { */
577  /* // Get gradient */
578  /* (*Source_fct_gradient_pt)(time,x,gradient); */
579  /* } */
580  }
581 
582 
583  /// Calculate the source fct at given time and
584  /// Eulerian position
585  virtual double get_source_nst(const double& time,
586  const unsigned& ipt,
587  const Vector<double>& x)
588  {
589  // If the function pointer is zero return zero
590  if (Source_fct_pt == 0)
591  {
592  return 0;
593  }
594  // Otherwise call the function
595  else
596  {
597  return (*Source_fct_pt)(time, x);
598  }
599  }
600 
601 
602  /// Get gradient of source term at (Eulerian) position x. This function is
603  /// virtual to allow overloading in multi-physics problems where
604  /// the strength of the source function might be determined by
605  /// another system of equations. Computed via function pointer
606  /// (if set) or by finite differencing (default)
607  inline virtual void get_source_gradient_nst(const double& time,
608  const unsigned& ipt,
609  const Vector<double>& x,
610  Vector<double>& gradient)
611  {
612  // hierher: Implement function pointer version
613  /* //If no gradient function has been set, FD it */
614  /* if(Source_fct_gradient_pt==0) */
615  {
616  // Reference value
617  double source = get_source_nst(time, ipt, x);
618 
619  // FD it
621  double source_pls = 0.0;
622  Vector<double> x_pls(x);
623  for (unsigned i = 0; i < DIM; i++)
624  {
625  x_pls[i] += eps_fd;
626  source_pls = get_source_nst(time, ipt, x_pls);
627  gradient[i] = (source_pls - source) / eps_fd;
628  x_pls[i] = x[i];
629  }
630  }
631  /* else */
632  /* { */
633  /* // Get gradient */
634  /* (*Source_fct_gradient_pt)(time,x,gradient); */
635  /* } */
636  }
637 
638 
639  /// Compute the residuals for the Navier--Stokes equations.
640  /// Flag=1 (or 0): do (or don't) compute the Jacobian as well.
641  /// Flag=2: Fill in mass matrix too.
643  Vector<double>& residuals,
644  DenseMatrix<double>& jacobian,
645  DenseMatrix<double>& mass_matrix,
646  unsigned flag);
647 
648 
649  /// Compute the residuals for the associated pressure advection
650  /// diffusion problem. Used by the Fp preconditioner.
651  /// flag=1(or 0): do (or don't) compute the Jacobian as well.
653  Vector<double>& residuals, DenseMatrix<double>& jacobian, unsigned flag);
654 
655  /// Compute the derivatives of the
656  /// residuals for the Navier--Stokes equations with respect to a parameter
657  /// Flag=1 (or 0): do (or don't) compute the Jacobian as well.
658  /// Flag=2: Fill in mass matrix too.
660  double* const& parameter_pt,
661  Vector<double>& dres_dparam,
662  DenseMatrix<double>& djac_dparam,
663  DenseMatrix<double>& dmass_matrix_dparam,
664  unsigned flag);
665 
666  /// Compute the hessian tensor vector products required to
667  /// perform continuation of bifurcations analytically
669  Vector<double> const& Y,
670  DenseMatrix<double> const& C,
671  DenseMatrix<double>& product);
672 
673 
674  public:
675  /// Constructor: NULL the body force and source function
676  /// and make sure the ALE terms are included by default.
678  : Body_force_fct_pt(0),
679  Source_fct_pt(0),
681  ALE_is_disabled(false),
683  {
684  // Set all the Physical parameter pointers to the default value zero
689  // Set the Physical ratios to the default value of 1
692  }
693 
694  /// Vector to decide whether the stress-divergence form is used or not
695  // N.B. This needs to be public so that the intel compiler gets things
696  // correct somehow the access function messes things up when going to
697  // refineable navier--stokes
699 
700  // Access functions for the physical constants
701 
702  /// Reynolds number
703  const double& re() const
704  {
705  return *Re_pt;
706  }
707 
708  /// Product of Reynolds and Strouhal number (=Womersley number)
709  const double& re_st() const
710  {
711  return *ReSt_pt;
712  }
713 
714  /// Pointer to Reynolds number
715  double*& re_pt()
716  {
717  return Re_pt;
718  }
719 
720  /// Pointer to product of Reynolds and Strouhal number (=Womersley number)
721  double*& re_st_pt()
722  {
723  return ReSt_pt;
724  }
725 
726  /// Viscosity ratio for element: Element's viscosity relative to the
727  /// viscosity used in the definition of the Reynolds number
728  const double& viscosity_ratio() const
729  {
730  return *Viscosity_Ratio_pt;
731  }
732 
733  /// Pointer to Viscosity Ratio
735  {
736  return Viscosity_Ratio_pt;
737  }
738 
739  /// Density ratio for element: Element's density relative to the
740  /// viscosity used in the definition of the Reynolds number
741  const double& density_ratio() const
742  {
743  return *Density_Ratio_pt;
744  }
745 
746  /// Pointer to Density ratio
747  double*& density_ratio_pt()
748  {
749  return Density_Ratio_pt;
750  }
751 
752  /// Global inverse Froude number
753  const double& re_invfr() const
754  {
755  return *ReInvFr_pt;
756  }
757 
758  /// Pointer to global inverse Froude number
759  double*& re_invfr_pt()
760  {
761  return ReInvFr_pt;
762  }
763 
764  /// Vector of gravitational components
765  const Vector<double>& g() const
766  {
767  return *G_pt;
768  }
769 
770  /// Pointer to Vector of gravitational components
772  {
773  return G_pt;
774  }
775 
776  /// Access function for the body-force pointer
778  {
779  return Body_force_fct_pt;
780  }
781 
782  /// Access function for the body-force pointer. Const version
784  {
785  return Body_force_fct_pt;
786  }
787 
788  /// Access function for the source-function pointer
790  {
791  return Source_fct_pt;
792  }
793 
794  /// Access function for the source-function pointer. Const version
796  {
797  return Source_fct_pt;
798  }
799 
800  /// Access function for the source-function pointer for pressure
801  /// advection diffusion (used for validation only).
803  {
805  }
806 
807  /// Access function for the source-function pointer for pressure
808  /// advection diffusion (used for validation only). Const version.
810  const
811  {
813  }
814 
815  /// Global eqn number of pressure dof that's pinned in pressure
816  /// adv diff problem
818  {
819  return Pinned_fp_pressure_eqn;
820  }
821 
822  /// Function to return number of pressure degrees of freedom
823  virtual unsigned npres_nst() const = 0;
824 
825  /// Compute the pressure shape functions at local coordinate s
826  virtual void pshape_nst(const Vector<double>& s, Shape& psi) const = 0;
827 
828  /// Compute the pressure shape and test functions
829  /// at local coordinate s
830  virtual void pshape_nst(const Vector<double>& s,
831  Shape& psi,
832  Shape& test) const = 0;
833 
834  /// Compute the pressure shape and test functions and derivatives
835  /// w.r.t. global coords at local coordinate s.
836  /// Return Jacobian of mapping between local and global coordinates.
838  Shape& ppsi,
839  DShape& dppsidx,
840  Shape& ptest,
841  DShape& dptestdx) const = 0;
842 
843  /// Velocity i at local node n. Uses suitably interpolated value
844  /// for hanging nodes. The use of u_index_nst() permits the use of this
845  /// element as the basis for multi-physics elements. The default
846  /// is to assume that the i-th velocity component is stored at the
847  /// i-th location of the node
848  double u_nst(const unsigned& n, const unsigned& i) const
849  {
850  return nodal_value(n, u_index_nst(i));
851  }
852 
853  /// Velocity i at local node n at timestep t (t=0: present;
854  /// t>0: previous). Uses suitably interpolated value for hanging nodes.
855  double u_nst(const unsigned& t, const unsigned& n, const unsigned& i) const
856  {
857  return nodal_value(t, n, u_index_nst(i));
858  }
859 
860  /// Return the index at which the i-th unknown velocity component
861  /// is stored. The default value, i, is appropriate for single-physics
862  /// problems.
863  /// In derived multi-physics elements, this function should be overloaded
864  /// to reflect the chosen storage scheme. Note that these equations require
865  /// that the unknowns are always stored at the same indices at each node.
866  virtual inline unsigned u_index_nst(const unsigned& i) const
867  {
868  return i;
869  }
870 
871  /// Return the number of velocity components
872  /// Used in FluidInterfaceElements
873  inline unsigned n_u_nst() const
874  {
875  return DIM;
876  }
877 
878  /// i-th component of du/dt at local node n.
879  /// Uses suitably interpolated value for hanging nodes.
880  double du_dt_nst(const unsigned& n, const unsigned& i) const
881  {
882  // Get the data's timestepper
884 
885  // Initialise dudt
886  double dudt = 0.0;
887 
888  // Loop over the timesteps, if there is a non Steady timestepper
889  if (!time_stepper_pt->is_steady())
890  {
891  // Find the index at which the dof is stored
892  const unsigned u_nodal_index = this->u_index_nst(i);
893 
894  // Number of timsteps (past & present)
895  const unsigned n_time = time_stepper_pt->ntstorage();
896  // Loop over the timesteps
897  for (unsigned t = 0; t < n_time; t++)
898  {
899  dudt +=
900  time_stepper_pt->weight(1, t) * nodal_value(t, n, u_nodal_index);
901  }
902  }
903 
904  return dudt;
905  }
906 
907  /// Disable ALE, i.e. assert the mesh is not moving -- you do this
908  /// at your own risk!
909  void disable_ALE()
910  {
911  ALE_is_disabled = true;
912  }
913 
914  /// (Re-)enable ALE, i.e. take possible mesh motion into account
915  /// when evaluating the time-derivative. Note: By default, ALE is
916  /// enabled, at the expense of possibly creating unnecessary work
917  /// in problems where the mesh is, in fact, stationary.
918  void enable_ALE()
919  {
920  ALE_is_disabled = false;
921  }
922 
923  /// Pressure at local pressure "node" n_p
924  /// Uses suitably interpolated value for hanging nodes.
925  virtual double p_nst(const unsigned& n_p) const = 0;
926 
927  /// Pressure at local pressure "node" n_p at time level t
928  virtual double p_nst(const unsigned& t, const unsigned& n_p) const = 0;
929 
930  /// Pin p_dof-th pressure dof and set it to value specified by p_value.
931  virtual void fix_pressure(const unsigned& p_dof, const double& p_value) = 0;
932 
933  /// Return the index at which the pressure is stored if it is
934  /// stored at the nodes. If not stored at the nodes this will return
935  /// a negative number.
936  virtual int p_nodal_index_nst() const
937  {
939  }
940 
941  /// Integral of pressure over element
942  double pressure_integral() const;
943 
944  /// Return integral of dissipation over element
945  double dissipation() const;
946 
947  /// Return dissipation at local coordinate s
948  double dissipation(const Vector<double>& s) const;
949 
950  /// Compute the vorticity vector at local coordinate s
952  Vector<double>& vorticity) const;
953 
954  /// Compute the scalar vorticity at local coordinate s (2D)
955  void get_vorticity(const Vector<double>& s, double& vorticity) const;
956 
957  /// Get integral of kinetic energy over element
958  double kin_energy() const;
959 
960  /// Get integral of time derivative of kinetic energy over element
961  double d_kin_energy_dt() const;
962 
963  /// Strain-rate tensor: 1/2 (du_i/dx_j + du_j/dx_i)
964  void strain_rate(const Vector<double>& s,
966 
967  /// Compute traction (on the viscous scale) exerted onto
968  /// the fluid at local coordinate s. N has to be outer unit normal
969  /// to the fluid.
970  void get_traction(const Vector<double>& s,
971  const Vector<double>& N,
972  Vector<double>& traction);
973 
974  /// Compute traction (on the viscous scale) exerted onto
975  /// the fluid at local coordinate s, decomposed into pressure and
976  /// normal and tangential viscous components.
977  /// N has to be outer unit normal to the fluid.
978  void get_traction(const Vector<double>& s,
979  const Vector<double>& N,
980  Vector<double>& traction_p,
981  Vector<double>& traction_visc_n,
982  Vector<double>& traction_visc_t);
983 
984  /// This implements a pure virtual function defined
985  /// in the FSIFluidElement class. The function computes
986  /// the traction (on the viscous scale), at the element's local
987  /// coordinate s, that the fluid element exerts onto an adjacent
988  /// solid element. The number of arguments is imposed by
989  /// the interface defined in the FSIFluidElement -- only the
990  /// unit normal N (pointing into the fluid!) is actually used
991  /// in the computation.
992  void get_load(const Vector<double>& s,
993  const Vector<double>& N,
994  Vector<double>& load)
995  {
996  // Note: get_traction() computes the traction onto the fluid
997  // if N is the outer unit normal onto the fluid; here we're
998  // exepcting N to point into the fluid so we're getting the
999  // traction onto the adjacent wall instead!
1000  get_traction(s, N, load);
1001  }
1002 
1003  /// Compute the diagonal of the velocity/pressure mass matrices.
1004  /// If which one=0, both are computed, otherwise only the pressure
1005  /// (which_one=1) or the velocity mass matrix (which_one=2 -- the
1006  /// LSC version of the preconditioner only needs that one)
1008  Vector<double>& press_mass_diag,
1009  Vector<double>& veloc_mass_diag,
1010  const unsigned& which_one = 0);
1011 
1012  /// Number of scalars/fields output by this element. Reimplements
1013  /// broken virtual function in base class.
1014  unsigned nscalar_paraview() const
1015  {
1016  return DIM + 1;
1017  }
1018 
1019  /// Write values of the i-th scalar field at the plot points. Needs
1020  /// to be implemented for each new specific element type.
1021  void scalar_value_paraview(std::ofstream& file_out,
1022  const unsigned& i,
1023  const unsigned& nplot) const
1024  {
1025  // Vector of local coordinates
1026  Vector<double> s(DIM);
1027 
1028 
1029  // Loop over plot points
1030  unsigned num_plot_points = nplot_points_paraview(nplot);
1031  for (unsigned iplot = 0; iplot < num_plot_points; iplot++)
1032  {
1033  // Get local coordinates of plot point
1034  get_s_plot(iplot, nplot, s);
1035 
1036  // Velocities
1037  if (i < DIM)
1038  {
1039  file_out << interpolated_u_nst(s, i) << std::endl;
1040  }
1041 
1042  // Pressure
1043  else if (i == DIM)
1044  {
1045  file_out << interpolated_p_nst(s) << std::endl;
1046  }
1047 
1048  // Never get here
1049  else
1050  {
1051 #ifdef PARANOID
1052  std::stringstream error_stream;
1053  error_stream << "These Navier Stokes elements only store " << DIM + 1
1054  << " fields, "
1055  << "but i is currently " << i << std::endl;
1056  throw OomphLibError(error_stream.str(),
1057  OOMPH_CURRENT_FUNCTION,
1058  OOMPH_EXCEPTION_LOCATION);
1059 #endif
1060  }
1061  }
1062  }
1063 
1064 
1065  /// Write values of the i-th scalar field at the plot points. Needs
1066  /// to be implemented for each new specific element type.
1068  std::ofstream& file_out,
1069  const unsigned& i,
1070  const unsigned& nplot,
1071  const double& time,
1072  FiniteElement::UnsteadyExactSolutionFctPt exact_soln_pt) const
1073  {
1074 #ifdef PARANOID
1075  if (i > DIM)
1076  {
1077  // Create an output stream
1078  std::stringstream error_stream;
1079 
1080  // Create the error message
1081  error_stream << "These Navier Stokes elements only store " << DIM + 1
1082  << " fields, but i is currently " << i << std::endl;
1083 
1084  // Throw the error message
1085  throw OomphLibError(
1086  error_stream.str(), OOMPH_CURRENT_FUNCTION, OOMPH_EXCEPTION_LOCATION);
1087  }
1088 #endif
1089 
1090  // Vector of local coordinates
1091  Vector<double> s(DIM + 1, 0.0);
1092 
1093  // Storage for the spatial coordinates
1094  Vector<double> spatial_coordinates(DIM, 0.0);
1095 
1096  // How many plot points do we have in total?
1097  unsigned num_plot_points = nplot_points_paraview(nplot);
1098 
1099  // Loop over plot points
1100  for (unsigned iplot = 0; iplot < num_plot_points; iplot++)
1101  {
1102  // Get the local coordinates of the iplot-th plot point
1103  get_s_plot(iplot, nplot, s);
1104 
1105  // Loop over the spatial coordinates
1106  for (unsigned j = 0; j < DIM; j++)
1107  {
1108  // Assign the i-th spatial coordinate
1109  spatial_coordinates[j] = interpolated_x(s, j);
1110  }
1111 
1112  // Exact solution vector (here it's simply a scalar)
1113  Vector<double> exact_soln(DIM + 1, 0.0);
1114 
1115  // Get the exact solution at this point
1116  (*exact_soln_pt)(time, spatial_coordinates, exact_soln);
1117 
1118  // Output the interpolated solution value
1119  file_out << exact_soln[i] << std::endl;
1120  } // for (unsigned iplot=0;iplot<num_plot_points;iplot++)
1121  } // End of scalar_value_fct_paraview
1122 
1123 
1124  /// Name of the i-th scalar field. Default implementation
1125  /// returns V1 for the first one, V2 for the second etc. Can (should!) be
1126  /// overloaded with more meaningful names in specific elements.
1127  std::string scalar_name_paraview(const unsigned& i) const
1128  {
1129  // Velocities
1130  if (i < DIM)
1131  {
1132  return "Velocity " + StringConversion::to_string(i);
1133  }
1134  // Preussre
1135  else if (i == DIM)
1136  {
1137  return "Pressure";
1138  }
1139  // Never get here
1140  else
1141  {
1142  std::stringstream error_stream;
1143  error_stream << "These Navier Stokes elements only store " << DIM + 1
1144  << " fields,\n"
1145  << "but i is currently " << i << std::endl;
1146  throw OomphLibError(
1147  error_stream.str(), OOMPH_CURRENT_FUNCTION, OOMPH_EXCEPTION_LOCATION);
1148  // Dummy return
1149  return " ";
1150  }
1151  }
1152 
1153  /// Output function: x,y,[z],u,v,[w],p
1154  /// in tecplot format. Default number of plot points
1155  void output(std::ostream& outfile)
1156  {
1157  unsigned nplot = 5;
1158  output(outfile, nplot);
1159  }
1160 
1161  /// Output function: x,y,[z],u,v,[w],p
1162  /// in tecplot format. nplot points in each coordinate direction
1163  void output(std::ostream& outfile, const unsigned& nplot);
1164 
1165  /// C-style output function: x,y,[z],u,v,[w],p
1166  /// in tecplot format. Default number of plot points
1167  void output(FILE* file_pt)
1168  {
1169  unsigned nplot = 5;
1170  output(file_pt, nplot);
1171  }
1172 
1173  /// C-style output function: x,y,[z],u,v,[w],p
1174  /// in tecplot format. nplot points in each coordinate direction
1175  void output(FILE* file_pt, const unsigned& nplot);
1176 
1177  /// Full output function:
1178  /// x,y,[z],u,v,[w],p,du/dt,dv/dt,[dw/dt],dissipation
1179  /// in tecplot format. Default number of plot points
1180  void full_output(std::ostream& outfile)
1181  {
1182  unsigned nplot = 5;
1183  full_output(outfile, nplot);
1184  }
1185 
1186  /// Full output function:
1187  /// x,y,[z],u,v,[w],p,du/dt,dv/dt,[dw/dt],dissipation
1188  /// in tecplot format. nplot points in each coordinate direction
1189  void full_output(std::ostream& outfile, const unsigned& nplot);
1190 
1191 
1192  /// Output function: x,y,[z],u,v,[w] in tecplot format.
1193  /// nplot points in each coordinate direction at timestep t
1194  /// (t=0: present; t>0: previous timestep)
1195  void output_veloc(std::ostream& outfile,
1196  const unsigned& nplot,
1197  const unsigned& t);
1198 
1199 
1200  /// Output function: x,y,[z], [omega_x,omega_y,[and/or omega_z]]
1201  /// in tecplot format. nplot points in each coordinate direction
1202  void output_vorticity(std::ostream& outfile, const unsigned& nplot);
1203 
1204  /// Output exact solution specified via function pointer
1205  /// at a given number of plot points. Function prints as
1206  /// many components as are returned in solution Vector
1207  void output_fct(std::ostream& outfile,
1208  const unsigned& nplot,
1210 
1211  /// Output exact solution specified via function pointer
1212  /// at a given time and at a given number of plot points.
1213  /// Function prints as many components as are returned in solution Vector.
1214  void output_fct(std::ostream& outfile,
1215  const unsigned& nplot,
1216  const double& time,
1218 
1219  /// Compute norm of solution: square of the L2 norm of the velocities
1220  void compute_norm(double& norm);
1221 
1222  /// Compute the vector norm of the FEM solution
1223  void compute_norm(Vector<double>& norm);
1224 
1225  /// Validate against exact solution at given time
1226  /// Solution is provided via function pointer.
1227  /// Plot at a given number of plot points and compute L2 error
1228  /// and L2 norm of velocity solution over element
1229  void compute_error(std::ostream& outfile,
1231  const double& time,
1232  double& error,
1233  double& norm);
1234 
1235  /// Validate against exact solution.
1236  /// Solution is provided via function pointer.
1237  /// Plot at a given number of plot points and compute L2 error
1238  /// and L2 norm of velocity solution over element
1239  void compute_error(std::ostream& outfile,
1241  double& error,
1242  double& norm);
1243 
1244  /// Validate against exact solution. Solution is provided via
1245  /// function pointer. Compute L2 error and L2 norm of velocity solution
1246  /// over element.
1248  const double& time,
1249  double& error,
1250  double& norm);
1251 
1252  /// Validate against exact solution. Solution is provided via
1253  /// function pointer. Compute L2 error and L2 norm of velocity solution
1254  /// over element.
1256  double& error,
1257  double& norm);
1258 
1259  /// Compute the element's residual Vector
1261  {
1262  // Call the generic residuals function with flag set to 0
1263  // and using a dummy matrix argument
1265  residuals,
1268  0);
1269  }
1270 
1271  /// Compute the element's residual Vector and the jacobian matrix
1272  /// Virtual function can be overloaded by hanging-node version
1274  DenseMatrix<double>& jacobian)
1275  {
1276  // Call the generic routine with the flag set to 1
1278  residuals, jacobian, GeneralisedElement::Dummy_matrix, 1);
1279  }
1280 
1281  /// Add the element's contribution to its residuals vector,
1282  /// jacobian matrix and mass matrix
1284  Vector<double>& residuals,
1285  DenseMatrix<double>& jacobian,
1286  DenseMatrix<double>& mass_matrix)
1287  {
1288  // Call the generic routine with the flag set to 2
1290  residuals, jacobian, mass_matrix, 2);
1291  }
1292 
1293  /// Compute the element's residual Vector
1295  double* const& parameter_pt, Vector<double>& dres_dparam)
1296  {
1297  // Call the generic residuals function with flag set to 0
1298  // and using a dummy matrix argument
1300  parameter_pt,
1301  dres_dparam,
1304  0);
1305  }
1306 
1307  /// Compute the element's residual Vector and the jacobian matrix
1308  /// Virtual function can be overloaded by hanging-node version
1310  double* const& parameter_pt,
1311  Vector<double>& dres_dparam,
1312  DenseMatrix<double>& djac_dparam)
1313  {
1314  // Call the generic routine with the flag set to 1
1316  parameter_pt,
1317  dres_dparam,
1318  djac_dparam,
1320  1);
1321  }
1322 
1323  /// Add the element's contribution to its residuals vector,
1324  /// jacobian matrix and mass matrix
1326  double* const& parameter_pt,
1327  Vector<double>& dres_dparam,
1328  DenseMatrix<double>& djac_dparam,
1329  DenseMatrix<double>& dmass_matrix_dparam)
1330  {
1331  // Call the generic routine with the flag set to 2
1333  parameter_pt, dres_dparam, djac_dparam, dmass_matrix_dparam, 2);
1334  }
1335 
1336 
1337  /// Compute the residuals for the associated pressure advection
1338  /// diffusion problem. Used by the Fp preconditioner.
1340  Vector<double>& residuals)
1341  {
1343  residuals, GeneralisedElement::Dummy_matrix, 0);
1344  }
1345 
1346  /// Compute the residuals and Jacobian for the associated
1347  /// pressure advection diffusion problem. Used by the Fp preconditioner.
1349  Vector<double>& residuals, DenseMatrix<double>& jacobian)
1350  {
1352  residuals, jacobian, 1);
1353  }
1354 
1355 
1356  /// Pin all non-pressure dofs and backup eqn numbers
1358  std::map<Data*, std::vector<int>>& eqn_number_backup)
1359  {
1360  // Loop over internal data and pin the values (having established that
1361  // pressure dofs aren't amongst those)
1362  unsigned nint = this->ninternal_data();
1363  for (unsigned j = 0; j < nint; j++)
1364  {
1365  Data* data_pt = this->internal_data_pt(j);
1366  if (eqn_number_backup[data_pt].size() == 0)
1367  {
1368  unsigned nvalue = data_pt->nvalue();
1369  eqn_number_backup[data_pt].resize(nvalue);
1370  for (unsigned i = 0; i < nvalue; i++)
1371  {
1372  // Backup
1373  eqn_number_backup[data_pt][i] = data_pt->eqn_number(i);
1374 
1375  // Pin everything
1376  data_pt->pin(i);
1377  }
1378  }
1379  }
1380 
1381  // Now deal with nodal values
1382  unsigned nnod = this->nnode();
1383  for (unsigned j = 0; j < nnod; j++)
1384  {
1385  Node* nod_pt = this->node_pt(j);
1386  if (eqn_number_backup[nod_pt].size() == 0)
1387  {
1388  unsigned nvalue = nod_pt->nvalue();
1389  eqn_number_backup[nod_pt].resize(nvalue);
1390  for (unsigned i = 0; i < nvalue; i++)
1391  {
1392  // Pin everything apart from the nodal pressure
1393  // value
1394  if (int(i) != this->p_nodal_index_nst())
1395  {
1396  // Backup
1397  eqn_number_backup[nod_pt][i] = nod_pt->eqn_number(i);
1398 
1399  // Pin
1400  nod_pt->pin(i);
1401  }
1402  // Else it's a pressure value
1403  else
1404  {
1405  // Exclude non-nodal pressure based elements
1406  if (this->p_nodal_index_nst() >= 0)
1407  {
1408  // Backup
1409  eqn_number_backup[nod_pt][i] = nod_pt->eqn_number(i);
1410  }
1411  }
1412  }
1413 
1414 
1415  // If it's a solid node deal with its positional data too
1416  SolidNode* solid_nod_pt = dynamic_cast<SolidNode*>(nod_pt);
1417  if (solid_nod_pt != 0)
1418  {
1419  Data* solid_posn_data_pt = solid_nod_pt->variable_position_pt();
1420  if (eqn_number_backup[solid_posn_data_pt].size() == 0)
1421  {
1422  unsigned nvalue = solid_posn_data_pt->nvalue();
1423  eqn_number_backup[solid_posn_data_pt].resize(nvalue);
1424  for (unsigned i = 0; i < nvalue; i++)
1425  {
1426  // Backup
1427  eqn_number_backup[solid_posn_data_pt][i] =
1428  solid_posn_data_pt->eqn_number(i);
1429 
1430  // Pin
1431  solid_posn_data_pt->pin(i);
1432  }
1433  }
1434  }
1435  }
1436  }
1437  }
1438 
1439 
1440  /// Build FaceElements that apply the Robin boundary condition
1441  /// to the pressure advection diffusion problem required by
1442  /// Fp preconditioner
1444  const unsigned& face_index) = 0;
1445 
1446  /// Output the FaceElements that apply the Robin boundary condition
1447  /// to the pressure advection diffusion problem required by
1448  /// Fp preconditioner
1450  std::ostream& outfile)
1451  {
1452  unsigned nel = Pressure_advection_diffusion_robin_element_pt.size();
1453  for (unsigned e = 0; e < nel; e++)
1454  {
1455  FaceElement* face_el_pt =
1457  outfile << "ZONE" << std::endl;
1458  Vector<double> unit_normal(DIM);
1459  Vector<double> x(DIM);
1460  Vector<double> s(DIM - 1);
1461  unsigned n = face_el_pt->integral_pt()->nweight();
1462  for (unsigned ipt = 0; ipt < n; ipt++)
1463  {
1464  for (unsigned i = 0; i < DIM - 1; i++)
1465  {
1466  s[i] = face_el_pt->integral_pt()->knot(ipt, i);
1467  }
1468  face_el_pt->interpolated_x(s, x);
1469  face_el_pt->outer_unit_normal(ipt, unit_normal);
1470  for (unsigned i = 0; i < DIM; i++)
1471  {
1472  outfile << x[i] << " ";
1473  }
1474  for (unsigned i = 0; i < DIM; i++)
1475  {
1476  outfile << unit_normal[i] << " ";
1477  }
1478  outfile << std::endl;
1479  }
1480  }
1481  }
1482 
1483  /// Delete the FaceElements that apply the Robin boundary condition
1484  /// to the pressure advection diffusion problem required by
1485  /// Fp preconditioner
1487  {
1488  unsigned nel = Pressure_advection_diffusion_robin_element_pt.size();
1489  for (unsigned e = 0; e < nel; e++)
1490  {
1492  }
1494  }
1495 
1496  /// Compute derivatives of elemental residual vector with respect
1497  /// to nodal coordinates. Overwrites default implementation in
1498  /// FiniteElement base class.
1499  /// dresidual_dnodal_coordinates(l,i,j) = d res(l) / dX_{ij}
1500  virtual void get_dresidual_dnodal_coordinates(
1501  RankThreeTensor<double>& dresidual_dnodal_coordinates);
1502 
1503 
1504  /// Compute vector of FE interpolated velocity u at local coordinate s
1506  Vector<double>& veloc) const
1507  {
1508  // Find number of nodes
1509  unsigned n_node = nnode();
1510  // Local shape function
1511  Shape psi(n_node);
1512  // Find values of shape function
1513  shape(s, psi);
1514 
1515  for (unsigned i = 0; i < DIM; i++)
1516  {
1517  // Index at which the nodal value is stored
1518  unsigned u_nodal_index = u_index_nst(i);
1519  // Initialise value of u
1520  veloc[i] = 0.0;
1521  // Loop over the local nodes and sum
1522  for (unsigned l = 0; l < n_node; l++)
1523  {
1524  veloc[i] += nodal_value(l, u_nodal_index) * psi[l];
1525  }
1526  }
1527  }
1528 
1529  /// Return FE interpolated velocity u[i] at local coordinate s
1530  double interpolated_u_nst(const Vector<double>& s, const unsigned& i) const
1531  {
1532  // Find number of nodes
1533  unsigned n_node = nnode();
1534  // Local shape function
1535  Shape psi(n_node);
1536  // Find values of shape function
1537  shape(s, psi);
1538 
1539  // Get nodal index at which i-th velocity is stored
1540  unsigned u_nodal_index = u_index_nst(i);
1541 
1542  // Initialise value of u
1543  double interpolated_u = 0.0;
1544  // Loop over the local nodes and sum
1545  for (unsigned l = 0; l < n_node; l++)
1546  {
1547  interpolated_u += nodal_value(l, u_nodal_index) * psi[l];
1548  }
1549 
1550  return (interpolated_u);
1551  }
1552 
1553  /// Return FE interpolated velocity u[i] at local coordinate s
1554  /// at time level t (t=0: present; t>0: history)
1555  double interpolated_u_nst(const unsigned& t,
1556  const Vector<double>& s,
1557  const unsigned& i) const
1558  {
1559  // Find number of nodes
1560  unsigned n_node = nnode();
1561 
1562  // Local shape function
1563  Shape psi(n_node);
1564 
1565  // Find values of shape function
1566  shape(s, psi);
1567 
1568  // Get nodal index at which i-th velocity is stored
1569  unsigned u_nodal_index = u_index_nst(i);
1570 
1571  // Initialise value of u
1572  double interpolated_u = 0.0;
1573  // Loop over the local nodes and sum
1574  for (unsigned l = 0; l < n_node; l++)
1575  {
1576  interpolated_u += nodal_value(t, l, u_nodal_index) * psi[l];
1577  }
1578 
1579  return (interpolated_u);
1580  }
1581 
1582  /// Compute the derivatives of the i-th component of
1583  /// velocity at point s with respect
1584  /// to all data that can affect its value. In addition, return the global
1585  /// equation numbers corresponding to the data. The function is virtual
1586  /// so that it can be overloaded in the refineable version
1588  const unsigned& i,
1589  Vector<double>& du_ddata,
1590  Vector<unsigned>& global_eqn_number)
1591  {
1592  // Find number of nodes
1593  unsigned n_node = nnode();
1594  // Local shape function
1595  Shape psi(n_node);
1596  // Find values of shape function
1597  shape(s, psi);
1598 
1599  // Find the index at which the velocity component is stored
1600  const unsigned u_nodal_index = u_index_nst(i);
1601 
1602  // Find the number of dofs associated with interpolated u
1603  unsigned n_u_dof = 0;
1604  for (unsigned l = 0; l < n_node; l++)
1605  {
1606  int global_eqn = this->node_pt(l)->eqn_number(u_nodal_index);
1607  // If it's positive add to the count
1608  if (global_eqn >= 0)
1609  {
1610  ++n_u_dof;
1611  }
1612  }
1613 
1614  // Now resize the storage schemes
1615  du_ddata.resize(n_u_dof, 0.0);
1616  global_eqn_number.resize(n_u_dof, 0);
1617 
1618  // Loop over th nodes again and set the derivatives
1619  unsigned count = 0;
1620  // Loop over the local nodes and sum
1621  for (unsigned l = 0; l < n_node; l++)
1622  {
1623  // Get the global equation number
1624  int global_eqn = this->node_pt(l)->eqn_number(u_nodal_index);
1625  if (global_eqn >= 0)
1626  {
1627  // Set the global equation number
1628  global_eqn_number[count] = global_eqn;
1629  // Set the derivative with respect to the unknown
1630  du_ddata[count] = psi[l];
1631  // Increase the counter
1632  ++count;
1633  }
1634  }
1635  }
1636 
1637 
1638  /// Return FE interpolated pressure at local coordinate s
1639  virtual double interpolated_p_nst(const Vector<double>& s) const
1640  {
1641  // Find number of nodes
1642  unsigned n_pres = npres_nst();
1643  // Local shape function
1644  Shape psi(n_pres);
1645  // Find values of shape function
1646  pshape_nst(s, psi);
1647 
1648  // Initialise value of p
1649  double interpolated_p = 0.0;
1650  // Loop over the local nodes and sum
1651  for (unsigned l = 0; l < n_pres; l++)
1652  {
1653  interpolated_p += p_nst(l) * psi[l];
1654  }
1655 
1656  return (interpolated_p);
1657  }
1658 
1659 
1660  /// Return FE interpolated pressure at local coordinate s at time level t
1661  double interpolated_p_nst(const unsigned& t, const Vector<double>& s) const
1662  {
1663  // Find number of nodes
1664  unsigned n_pres = npres_nst();
1665  // Local shape function
1666  Shape psi(n_pres);
1667  // Find values of shape function
1668  pshape_nst(s, psi);
1669 
1670  // Initialise value of p
1671  double interpolated_p = 0.0;
1672  // Loop over the local nodes and sum
1673  for (unsigned l = 0; l < n_pres; l++)
1674  {
1675  interpolated_p += p_nst(t, l) * psi[l];
1676  }
1677 
1678  return (interpolated_p);
1679  }
1680 
1681 
1682  /// Return FE interpolated derivatives of velocity component u[i]
1683  /// w.r.t spatial global coordinate direction x[j] at local coordinate s
1685  const unsigned& i,
1686  const unsigned& j) const
1687  {
1688  // Determine number of nodes
1689  const unsigned n_node = nnode();
1690 
1691  // Allocate storage for local shape function and its derivatives
1692  // with respect to space
1693  Shape psif(n_node);
1694  DShape dpsifdx(n_node, DIM);
1695 
1696  // Find values of shape function (ignore jacobian)
1697  (void)this->dshape_eulerian(s, psif, dpsifdx);
1698 
1699  // Get the index at which the velocity is stored
1700  const unsigned u_nodal_index = u_index_nst(i);
1701 
1702  // Initialise value of dudx
1703  double interpolated_dudx = 0.0;
1704 
1705  // Loop over the local nodes and sum
1706  for (unsigned l = 0; l < n_node; l++)
1707  {
1708  interpolated_dudx += nodal_value(l, u_nodal_index) * dpsifdx(l, j);
1709  }
1710 
1711  return (interpolated_dudx);
1712  }
1713 
1714 
1715  /// Output solution in data vector at local cordinates s:
1716  /// x,y [,z], u,v,[w], p
1718  {
1719  // Dimension
1720  unsigned dim = s.size();
1721 
1722  // Resize data for values
1723  data.resize(2 * dim + 1);
1724 
1725  // Write values in the vector
1726  for (unsigned i = 0; i < dim; i++)
1727  {
1728  data[i] = interpolated_x(s, i);
1729  data[i + dim] = this->interpolated_u_nst(s, i);
1730  }
1731  data[2 * dim] = this->interpolated_p_nst(s);
1732  }
1733  };
1734 
1735  /// ///////////////////////////////////////////////////////////////////////////
1736  /// ///////////////////////////////////////////////////////////////////////////
1737  /// ///////////////////////////////////////////////////////////////////////////
1738 
1739 
1740  //==========================================================================
1741  /// Crouzeix_Raviart elements are Navier--Stokes elements with quadratic
1742  /// interpolation for velocities and positions, but a discontinuous linear
1743  /// pressure interpolation. They can be used within oomph-lib's
1744  /// block preconditioning framework.
1745  //==========================================================================
1746  template<unsigned DIM>
1747  class QCrouzeixRaviartElement : public virtual QElement<DIM, 3>,
1748  public virtual NavierStokesEquations<DIM>
1749  {
1750  private:
1751  /// Static array of ints to hold required number of variables at nodes
1752  static const unsigned Initial_Nvalue[];
1753 
1754  protected:
1755  /// Internal index that indicates at which internal data the pressure
1756  /// is stored
1758 
1759 
1760  /// Velocity shape and test functions and their derivs
1761  /// w.r.t. to global coords at local coordinate s (taken from geometry)
1762  /// Return Jacobian of mapping between local and global coordinates.
1763  inline double dshape_and_dtest_eulerian_nst(const Vector<double>& s,
1764  Shape& psi,
1765  DShape& dpsidx,
1766  Shape& test,
1767  DShape& dtestdx) const;
1768 
1769  /// Velocity shape and test functions and their derivs
1770  /// w.r.t. to global coords at ipt-th integation point (taken from geometry)
1771  /// Return Jacobian of mapping between local and global coordinates.
1772  inline double dshape_and_dtest_eulerian_at_knot_nst(const unsigned& ipt,
1773  Shape& psi,
1774  DShape& dpsidx,
1775  Shape& test,
1776  DShape& dtestdx) const;
1777 
1778  /// Shape/test functions and derivs w.r.t. to global coords at
1779  /// integration point ipt; return Jacobian of mapping (J). Also compute
1780  /// derivatives of dpsidx, dtestdx and J w.r.t. nodal coordinates.
1782  const unsigned& ipt,
1783  Shape& psi,
1784  DShape& dpsidx,
1785  RankFourTensor<double>& d_dpsidx_dX,
1786  Shape& test,
1787  DShape& dtestdx,
1788  RankFourTensor<double>& d_dtestdx_dX,
1789  DenseMatrix<double>& djacobian_dX) const;
1790 
1791 
1792  public:
1793  /// Constructor, there are DIM+1 internal values (for the pressure)
1795  {
1796  // Allocate and add one Internal data object that stored DIM+1 pressure
1797  // values;
1798  P_nst_internal_index = this->add_internal_data(new Data(DIM + 1));
1799  }
1800 
1801  /// Number of values (pinned or dofs) required at local node n.
1802  virtual unsigned required_nvalue(const unsigned& n) const;
1803 
1804 
1805  /// Pressure shape functions at local coordinate s
1806  inline void pshape_nst(const Vector<double>& s, Shape& psi) const;
1807 
1808  /// Pressure shape and test functions at local coordinte s
1809  inline void pshape_nst(const Vector<double>& s,
1810  Shape& psi,
1811  Shape& test) const;
1812 
1813  /// Return the i-th pressure value
1814  /// (Discontinous pressure interpolation -- no need to cater for hanging
1815  /// nodes).
1816  double p_nst(const unsigned& i) const
1817  {
1818  return this->internal_data_pt(P_nst_internal_index)->value(i);
1819  }
1820 
1821  /// Return the i-th pressure value
1822  /// (Discontinous pressure interpolation -- no need to cater for hanging
1823  /// nodes).
1824  double p_nst(const unsigned& t, const unsigned& i) const
1825  {
1826  return this->internal_data_pt(P_nst_internal_index)->value(t, i);
1827  }
1828 
1829  /// Return number of pressure values
1830  unsigned npres_nst() const
1831  {
1832  return DIM + 1;
1833  }
1834 
1835  /// Pressure shape and test functions and their derivs
1836  /// w.r.t. to global coords at local coordinate s (taken from geometry)
1837  /// Return Jacobian of mapping between local and global coordinates.
1839  Shape& ppsi,
1840  DShape& dppsidx,
1841  Shape& ptest,
1842  DShape& dptestdx) const;
1843 
1844  /// Return the local equation numbers for the pressure values.
1845  inline int p_local_eqn(const unsigned& n) const
1846  {
1847  return this->internal_local_eqn(P_nst_internal_index, n);
1848  }
1849 
1850  /// Pin p_dof-th pressure dof and set it to value specified by p_value.
1851  void fix_pressure(const unsigned& p_dof, const double& p_value)
1852  {
1853  this->internal_data_pt(P_nst_internal_index)->pin(p_dof);
1854  this->internal_data_pt(P_nst_internal_index)->set_value(p_dof, p_value);
1855  }
1856 
1857 
1858  /// Build FaceElements that apply the Robin boundary condition
1859  /// to the pressure advection diffusion problem required by
1860  /// Fp preconditioner
1861  void build_fp_press_adv_diff_robin_bc_element(const unsigned& face_index)
1862  {
1865  this, face_index));
1866  }
1867 
1868  /// Add to the set \c paired_load_data pairs containing
1869  /// - the pointer to a Data object
1870  /// and
1871  /// - the index of the value in that Data object
1872  /// .
1873  /// for all values (pressures, velocities) that affect the
1874  /// load computed in the \c get_load(...) function.
1875  void identify_load_data(
1876  std::set<std::pair<Data*, unsigned>>& paired_load_data);
1877 
1878  /// Add to the set \c paired_pressure_data pairs
1879  /// containing
1880  /// - the pointer to a Data object
1881  /// and
1882  /// - the index of the value in that Data object
1883  /// .
1884  /// for all pressure values that affect the
1885  /// load computed in the \c get_load(...) function.
1887  std::set<std::pair<Data*, unsigned>>& paired_pressure_data);
1888 
1889 
1890  /// Redirect output to NavierStokesEquations output
1891  void output(std::ostream& outfile)
1892  {
1894  }
1895 
1896  /// Redirect output to NavierStokesEquations output
1897  void output(std::ostream& outfile, const unsigned& nplot)
1898  {
1899  NavierStokesEquations<DIM>::output(outfile, nplot);
1900  }
1901 
1902 
1903  /// Redirect output to NavierStokesEquations output
1904  void output(FILE* file_pt)
1905  {
1907  }
1908 
1909  /// Redirect output to NavierStokesEquations output
1910  void output(FILE* file_pt, const unsigned& nplot)
1911  {
1912  NavierStokesEquations<DIM>::output(file_pt, nplot);
1913  }
1914 
1915 
1916  /// Full output function:
1917  /// x,y,[z],u,v,[w],p,du/dt,dv/dt,[dw/dt],dissipation
1918  /// in tecplot format. Default number of plot points
1919  void full_output(std::ostream& outfile)
1920  {
1922  }
1923 
1924  /// Full output function:
1925  /// x,y,[z],u,v,[w],p,du/dt,dv/dt,[dw/dt],dissipation
1926  /// in tecplot format. nplot points in each coordinate direction
1927  void full_output(std::ostream& outfile, const unsigned& nplot)
1928  {
1930  }
1931 
1932 
1933  /// The number of "DOF types" that degrees of freedom in this element
1934  /// are sub-divided into: Velocity and pressure.
1935  unsigned ndof_types() const
1936  {
1937  return DIM + 1;
1938  }
1939 
1940  /// Create a list of pairs for all unknowns in this element,
1941  /// so that the first entry in each pair contains the global equation
1942  /// number of the unknown, while the second one contains the number
1943  /// of the "DOF type" that this unknown is associated with.
1944  /// (Function can obviously only be called if the equation numbering
1945  /// scheme has been set up.) Velocity=0; Pressure=1
1947  std::list<std::pair<unsigned long, unsigned>>& dof_lookup_list) const;
1948  };
1949 
1950  // Inline functions
1951 
1952  //=======================================================================
1953  /// Derivatives of the shape functions and test functions w.r.t. to global
1954  /// (Eulerian) coordinates. Return Jacobian of mapping between
1955  /// local and global coordinates.
1956  //=======================================================================
1957  template<unsigned DIM>
1959  const Vector<double>& s,
1960  Shape& psi,
1961  DShape& dpsidx,
1962  Shape& test,
1963  DShape& dtestdx) const
1964  {
1965  // Call the geometrical shape functions and derivatives
1966  double J = this->dshape_eulerian(s, psi, dpsidx);
1967  // The test functions are equal to the shape functions
1968  test = psi;
1969  dtestdx = dpsidx;
1970  // Return the jacobian
1971  return J;
1972  }
1973 
1974  //=======================================================================
1975  /// Derivatives of the shape functions and test functions w.r.t. to global
1976  /// (Eulerian) coordinates. Return Jacobian of mapping between
1977  /// local and global coordinates.
1978  //=======================================================================
1979  template<unsigned DIM>
1980  inline double QCrouzeixRaviartElement<
1981  DIM>::dshape_and_dtest_eulerian_at_knot_nst(const unsigned& ipt,
1982  Shape& psi,
1983  DShape& dpsidx,
1984  Shape& test,
1985  DShape& dtestdx) const
1986  {
1987  // Call the geometrical shape functions and derivatives
1988  double J = this->dshape_eulerian_at_knot(ipt, psi, dpsidx);
1989  // The test functions are equal to the shape functions
1990  test = psi;
1991  dtestdx = dpsidx;
1992  // Return the jacobian
1993  return J;
1994  }
1995 
1996 
1997  //=======================================================================
1998  /// 2D
1999  /// Define the shape functions (psi) and test functions (test) and
2000  /// their derivatives w.r.t. global coordinates (dpsidx and dtestdx)
2001  /// and return Jacobian of mapping (J). Additionally compute the
2002  /// derivatives of dpsidx, dtestdx and J w.r.t. nodal coordinates.
2003  ///
2004  /// Galerkin: Test functions = shape functions
2005  //=======================================================================
2006  template<>
2009  const unsigned& ipt,
2010  Shape& psi,
2011  DShape& dpsidx,
2012  RankFourTensor<double>& d_dpsidx_dX,
2013  Shape& test,
2014  DShape& dtestdx,
2015  RankFourTensor<double>& d_dtestdx_dX,
2016  DenseMatrix<double>& djacobian_dX) const
2017  {
2018  // Call the geometrical shape functions and derivatives
2019  const double J = this->dshape_eulerian_at_knot(
2020  ipt, psi, dpsidx, djacobian_dX, d_dpsidx_dX);
2021 
2022  // Loop over the test functions and derivatives and set them equal to the
2023  // shape functions
2024  for (unsigned i = 0; i < 9; i++)
2025  {
2026  test[i] = psi[i];
2027 
2028  for (unsigned k = 0; k < 2; k++)
2029  {
2030  dtestdx(i, k) = dpsidx(i, k);
2031 
2032  for (unsigned p = 0; p < 2; p++)
2033  {
2034  for (unsigned q = 0; q < 9; q++)
2035  {
2036  d_dtestdx_dX(p, q, i, k) = d_dpsidx_dX(p, q, i, k);
2037  }
2038  }
2039  }
2040  }
2041 
2042  // Return the jacobian
2043  return J;
2044  }
2045 
2046 
2047  //=======================================================================
2048  /// 3D
2049  /// Define the shape functions (psi) and test functions (test) and
2050  /// their derivatives w.r.t. global coordinates (dpsidx and dtestdx)
2051  /// and return Jacobian of mapping (J). Additionally compute the
2052  /// derivatives of dpsidx, dtestdx and J w.r.t. nodal coordinates.
2053  ///
2054  /// Galerkin: Test functions = shape functions
2055  //=======================================================================
2056  template<>
2059  const unsigned& ipt,
2060  Shape& psi,
2061  DShape& dpsidx,
2062  RankFourTensor<double>& d_dpsidx_dX,
2063  Shape& test,
2064  DShape& dtestdx,
2065  RankFourTensor<double>& d_dtestdx_dX,
2066  DenseMatrix<double>& djacobian_dX) const
2067  {
2068  // Call the geometrical shape functions and derivatives
2069  const double J = this->dshape_eulerian_at_knot(
2070  ipt, psi, dpsidx, djacobian_dX, d_dpsidx_dX);
2071 
2072  // Loop over the test functions and derivatives and set them equal to the
2073  // shape functions
2074  for (unsigned i = 0; i < 27; i++)
2075  {
2076  test[i] = psi[i];
2077 
2078  for (unsigned k = 0; k < 3; k++)
2079  {
2080  dtestdx(i, k) = dpsidx(i, k);
2081 
2082  for (unsigned p = 0; p < 3; p++)
2083  {
2084  for (unsigned q = 0; q < 27; q++)
2085  {
2086  d_dtestdx_dX(p, q, i, k) = d_dpsidx_dX(p, q, i, k);
2087  }
2088  }
2089  }
2090  }
2091 
2092  // Return the jacobian
2093  return J;
2094  }
2095 
2096 
2097  //=======================================================================
2098  /// 2D :
2099  /// Pressure shape functions
2100  //=======================================================================
2101  template<>
2103  Shape& psi) const
2104  {
2105  psi[0] = 1.0;
2106  psi[1] = s[0];
2107  psi[2] = s[1];
2108  }
2109 
2110 
2111  //==========================================================================
2112  /// 2D :
2113  /// Pressure shape and test functions and derivs w.r.t. to Eulerian coords.
2114  /// Return Jacobian of mapping between local and global coordinates.
2115  //==========================================================================
2116  template<>
2118  const Vector<double>& s,
2119  Shape& ppsi,
2120  DShape& dppsidx,
2121  Shape& ptest,
2122  DShape& dptestdx) const
2123  {
2124  // Initalise with shape fcts and derivs. w.r.t. to local coordinates
2125  ppsi[0] = 1.0;
2126  ppsi[1] = s[0];
2127  ppsi[2] = s[1];
2128 
2129  dppsidx(0, 0) = 0.0;
2130  dppsidx(1, 0) = 1.0;
2131  dppsidx(2, 0) = 0.0;
2132 
2133  dppsidx(0, 1) = 0.0;
2134  dppsidx(1, 1) = 0.0;
2135  dppsidx(2, 1) = 1.0;
2136 
2137 
2138  // Get the values of the shape functions and their local derivatives
2139  Shape psi(9);
2140  DShape dpsi(9, 2);
2141  dshape_local(s, psi, dpsi);
2142 
2143  // Allocate memory for the inverse 2x2 jacobian
2144  DenseMatrix<double> inverse_jacobian(2);
2145 
2146  // Now calculate the inverse jacobian
2147  const double det = local_to_eulerian_mapping(dpsi, inverse_jacobian);
2148 
2149  // Now set the values of the derivatives to be derivs w.r.t. to the
2150  // Eulerian coordinates
2151  transform_derivatives(inverse_jacobian, dppsidx);
2152 
2153  // The test functions are equal to the shape functions
2154  ptest = ppsi;
2155  dptestdx = dppsidx;
2156 
2157  // Return the determinant of the jacobian
2158  return det;
2159  }
2160 
2161 
2162  //=======================================================================
2163  /// Ppressure shape and test functions
2164  //=======================================================================
2165  template<unsigned DIM>
2167  Shape& psi,
2168  Shape& test) const
2169  {
2170  // Call the pressure shape functions
2171  this->pshape_nst(s, psi);
2172  // Test functions are equal to shape functions
2173  test = psi;
2174  }
2175 
2176 
2177  //=======================================================================
2178  /// 3D :
2179  /// Pressure shape functions
2180  //=======================================================================
2181  template<>
2183  Shape& psi) const
2184  {
2185  psi[0] = 1.0;
2186  psi[1] = s[0];
2187  psi[2] = s[1];
2188  psi[3] = s[2];
2189  }
2190 
2191 
2192  //==========================================================================
2193  /// 3D :
2194  /// Pressure shape and test functions and derivs w.r.t. to Eulerian coords.
2195  /// Return Jacobian of mapping between local and global coordinates.
2196  //==========================================================================
2197  template<>
2199  const Vector<double>& s,
2200  Shape& ppsi,
2201  DShape& dppsidx,
2202  Shape& ptest,
2203  DShape& dptestdx) const
2204  {
2205  // Initalise with shape fcts and derivs. w.r.t. to local coordinates
2206  ppsi[0] = 1.0;
2207  ppsi[1] = s[0];
2208  ppsi[2] = s[1];
2209  ppsi[3] = s[2];
2210 
2211  dppsidx(0, 0) = 0.0;
2212  dppsidx(1, 0) = 1.0;
2213  dppsidx(2, 0) = 0.0;
2214  dppsidx(3, 0) = 0.0;
2215 
2216  dppsidx(0, 1) = 0.0;
2217  dppsidx(1, 1) = 0.0;
2218  dppsidx(2, 1) = 1.0;
2219  dppsidx(3, 1) = 0.0;
2220 
2221  dppsidx(0, 2) = 0.0;
2222  dppsidx(1, 2) = 0.0;
2223  dppsidx(2, 2) = 0.0;
2224  dppsidx(3, 2) = 1.0;
2225 
2226 
2227  // Get the values of the shape functions and their local derivatives
2228  Shape psi(27);
2229  DShape dpsi(27, 3);
2230  dshape_local(s, psi, dpsi);
2231 
2232  // Allocate memory for the inverse 3x3 jacobian
2233  DenseMatrix<double> inverse_jacobian(3);
2234 
2235  // Now calculate the inverse jacobian
2236  const double det = local_to_eulerian_mapping(dpsi, inverse_jacobian);
2237 
2238  // Now set the values of the derivatives to be derivs w.r.t. to the
2239  // Eulerian coordinates
2240  transform_derivatives(inverse_jacobian, dppsidx);
2241 
2242  // The test functions are equal to the shape functions
2243  ptest = ppsi;
2244  dptestdx = dppsidx;
2245 
2246  // Return the determinant of the jacobian
2247  return det;
2248  }
2249 
2250 
2251  //=======================================================================
2252  /// Face geometry of the 2D Crouzeix_Raviart elements
2253  //=======================================================================
2254  template<>
2255  class FaceGeometry<QCrouzeixRaviartElement<2>> : public virtual QElement<1, 3>
2256  {
2257  public:
2258  FaceGeometry() : QElement<1, 3>() {}
2259  };
2260 
2261  //=======================================================================
2262  /// Face geometry of the 3D Crouzeix_Raviart elements
2263  //=======================================================================
2264  template<>
2265  class FaceGeometry<QCrouzeixRaviartElement<3>> : public virtual QElement<2, 3>
2266  {
2267  public:
2268  FaceGeometry() : QElement<2, 3>() {}
2269  };
2270 
2271  //=======================================================================
2272  /// Face geometry of the FaceGeometry of the 2D Crouzeix_Raviart elements
2273  //=======================================================================
2274  template<>
2276  : public virtual PointElement
2277  {
2278  public:
2280  };
2281 
2282 
2283  //=======================================================================
2284  /// Face geometry of the FaceGeometry of the 3D Crouzeix_Raviart elements
2285  //=======================================================================
2286  template<>
2288  : public virtual QElement<1, 3>
2289  {
2290  public:
2291  FaceGeometry() : QElement<1, 3>() {}
2292  };
2293 
2294 
2295  /// /////////////////////////////////////////////////////////////////////////
2296  /// /////////////////////////////////////////////////////////////////////////
2297 
2298 
2299  //=======================================================================
2300  /// Taylor--Hood elements are Navier--Stokes elements
2301  /// with quadratic interpolation for velocities and positions and
2302  /// continuous linear pressure interpolation. They can be used
2303  /// within oomph-lib's block-preconditioning framework.
2304  //=======================================================================
2305  template<unsigned DIM>
2306  class QTaylorHoodElement : public virtual QElement<DIM, 3>,
2307  public virtual NavierStokesEquations<DIM>
2308  {
2309  private:
2310  /// Static array of ints to hold number of variables at node
2311  static const unsigned Initial_Nvalue[];
2312 
2313  protected:
2314  /// Static array of ints to hold conversion from pressure
2315  /// node numbers to actual node numbers
2316  static const unsigned Pconv[];
2317 
2318  /// Velocity shape and test functions and their derivs
2319  /// w.r.t. to global coords at local coordinate s (taken from geometry)
2320  /// Return Jacobian of mapping between local and global coordinates.
2321  inline double dshape_and_dtest_eulerian_nst(const Vector<double>& s,
2322  Shape& psi,
2323  DShape& dpsidx,
2324  Shape& test,
2325  DShape& dtestdx) const;
2326 
2327  /// Velocity shape and test functions and their derivs
2328  /// w.r.t. to global coords at local coordinate s (taken from geometry)
2329  /// Return Jacobian of mapping between local and global coordinates.
2330  inline double dshape_and_dtest_eulerian_at_knot_nst(const unsigned& ipt,
2331  Shape& psi,
2332  DShape& dpsidx,
2333  Shape& test,
2334  DShape& dtestdx) const;
2335 
2336  /// Shape/test functions and derivs w.r.t. to global coords at
2337  /// integration point ipt; return Jacobian of mapping (J). Also compute
2338  /// derivatives of dpsidx, dtestdx and J w.r.t. nodal coordinates.
2340  const unsigned& ipt,
2341  Shape& psi,
2342  DShape& dpsidx,
2343  RankFourTensor<double>& d_dpsidx_dX,
2344  Shape& test,
2345  DShape& dtestdx,
2346  RankFourTensor<double>& d_dtestdx_dX,
2347  DenseMatrix<double>& djacobian_dX) const;
2348 
2349  public:
2350  /// Constructor, no internal data points
2352 
2353  /// Number of values (pinned or dofs) required at node n. Can
2354  /// be overwritten for hanging node version
2355  inline virtual unsigned required_nvalue(const unsigned& n) const
2356  {
2357  return Initial_Nvalue[n];
2358  }
2359 
2360 
2361  /// Pressure shape functions at local coordinate s
2362  inline void pshape_nst(const Vector<double>& s, Shape& psi) const;
2363 
2364  /// Pressure shape and test functions at local coordinte s
2365  inline void pshape_nst(const Vector<double>& s,
2366  Shape& psi,
2367  Shape& test) const;
2368 
2369  /// Set the value at which the pressure is stored in the nodes
2370  virtual int p_nodal_index_nst() const
2371  {
2372  return static_cast<int>(DIM);
2373  }
2374 
2375  /// Return the local equation numbers for the pressure values.
2376  inline int p_local_eqn(const unsigned& n) const
2377  {
2378  return this->nodal_local_eqn(Pconv[n], p_nodal_index_nst());
2379  }
2380 
2381  /// Access function for the pressure values at local pressure
2382  /// node n_p (const version)
2383  double p_nst(const unsigned& n_p) const
2384  {
2385  return this->nodal_value(Pconv[n_p], this->p_nodal_index_nst());
2386  }
2387 
2388  /// Access function for the pressure values at local pressure
2389  /// node n_p (const version)
2390  double p_nst(const unsigned& t, const unsigned& n_p) const
2391  {
2392  return this->nodal_value(t, Pconv[n_p], this->p_nodal_index_nst());
2393  }
2394 
2395  /// Pressure shape and test functions and their derivs
2396  /// w.r.t. to global coords at local coordinate s (taken from geometry).
2397  /// Return Jacobian of mapping between local and global coordinates.
2399  Shape& ppsi,
2400  DShape& dppsidx,
2401  Shape& ptest,
2402  DShape& dptestdx) const;
2403 
2404  /// Return number of pressure values
2405  unsigned npres_nst() const
2406  {
2407  return static_cast<unsigned>(pow(2.0, static_cast<int>(DIM)));
2408  }
2409 
2410 
2411  /// Deduce whether or not the provided node is a pressure node
2412  bool is_pressure_node(const unsigned& n) const
2413  {
2414  // The number of pressure nodes
2415  unsigned n_p = npres_nst();
2416 
2417  // See if the value n is in the array Pconv
2418  return std::find(this->Pconv, this->Pconv + n_p, n) != this->Pconv + n_p;
2419  } // End of is_pressure_node
2420 
2421 
2422  /// Pin p_dof-th pressure dof and set it to value specified by p_value.
2423  void fix_pressure(const unsigned& p_dof, const double& p_value)
2424  {
2425  this->node_pt(Pconv[p_dof])->pin(this->p_nodal_index_nst());
2426  this->node_pt(Pconv[p_dof])
2427  ->set_value(this->p_nodal_index_nst(), p_value);
2428  }
2429 
2430 
2431  /// Build FaceElements that apply the Robin boundary condition
2432  /// to the pressure advection diffusion problem required by
2433  /// Fp preconditioner
2434  void build_fp_press_adv_diff_robin_bc_element(const unsigned& face_index)
2435  {
2438  this, face_index));
2439  }
2440 
2441 
2442  /// Add to the set \c paired_load_data pairs containing
2443  /// - the pointer to a Data object
2444  /// and
2445  /// - the index of the value in that Data object
2446  /// .
2447  /// for all values (pressures, velocities) that affect the
2448  /// load computed in the \c get_load(...) function.
2449  void identify_load_data(
2450  std::set<std::pair<Data*, unsigned>>& paired_load_data);
2451 
2452 
2453  /// Add to the set \c paired_pressure_data pairs
2454  /// containing
2455  /// - the pointer to a Data object
2456  /// and
2457  /// - the index of the value in that Data object
2458  /// .
2459  /// for all pressure values that affect the
2460  /// load computed in the \c get_load(...) function.
2462  std::set<std::pair<Data*, unsigned>>& paired_pressure_data);
2463 
2464 
2465  /// Redirect output to NavierStokesEquations output
2466  void output(std::ostream& outfile)
2467  {
2469  }
2470 
2471  /// Redirect output to NavierStokesEquations output
2472  void output(std::ostream& outfile, const unsigned& nplot)
2473  {
2474  NavierStokesEquations<DIM>::output(outfile, nplot);
2475  }
2476 
2477  /// Redirect output to NavierStokesEquations output
2478  void output(FILE* file_pt)
2479  {
2481  }
2482 
2483  /// Redirect output to NavierStokesEquations output
2484  void output(FILE* file_pt, const unsigned& nplot)
2485  {
2486  NavierStokesEquations<DIM>::output(file_pt, nplot);
2487  }
2488 
2489 
2490  /// Returns the number of "DOF types" that degrees of freedom
2491  /// in this element are sub-divided into: Velocity and pressure.
2492  unsigned ndof_types() const
2493  {
2494  return DIM + 1;
2495  }
2496 
2497  /// Create a list of pairs for all unknowns in this element,
2498  /// so that the first entry in each pair contains the global equation
2499  /// number of the unknown, while the second one contains the number
2500  /// of the "DOF type" that this unknown is associated with.
2501  /// (Function can obviously only be called if the equation numbering
2502  /// scheme has been set up.) Velocity=0; Pressure=1
2504  std::list<std::pair<unsigned long, unsigned>>& dof_lookup_list) const;
2505  };
2506 
2507  // Inline functions
2508 
2509  //==========================================================================
2510  /// Derivatives of the shape functions and test functions w.r.t to
2511  /// global (Eulerian) coordinates. Return Jacobian of mapping between
2512  /// local and global coordinates.
2513  //==========================================================================
2514  template<unsigned DIM>
2516  const Vector<double>& s,
2517  Shape& psi,
2518  DShape& dpsidx,
2519  Shape& test,
2520  DShape& dtestdx) const
2521  {
2522  // Call the geometrical shape functions and derivatives
2523  double J = this->dshape_eulerian(s, psi, dpsidx);
2524 
2525  // The test functions are equal to the shape functions
2526  test = psi;
2527  dtestdx = dpsidx;
2528 
2529  // Return the jacobian
2530  return J;
2531  }
2532 
2533 
2534  //==========================================================================
2535  /// Derivatives of the shape functions and test functions w.r.t to
2536  /// global (Eulerian) coordinates. Return Jacobian of mapping between
2537  /// local and global coordinates.
2538  //==========================================================================
2539  template<unsigned DIM>
2541  const unsigned& ipt,
2542  Shape& psi,
2543  DShape& dpsidx,
2544  Shape& test,
2545  DShape& dtestdx) const
2546  {
2547  // Call the geometrical shape functions and derivatives
2548  double J = this->dshape_eulerian_at_knot(ipt, psi, dpsidx);
2549 
2550  // The test functions are equal to the shape functions
2551  test = psi;
2552  dtestdx = dpsidx;
2553 
2554  // Return the jacobian
2555  return J;
2556  }
2557 
2558 
2559  //==========================================================================
2560  /// 2D :
2561  /// Pressure shape and test functions and derivs w.r.t. to Eulerian coords.
2562  /// Return Jacobian of mapping between local and global coordinates.
2563  //==========================================================================
2564  template<>
2566  const Vector<double>& s,
2567  Shape& ppsi,
2568  DShape& dppsidx,
2569  Shape& ptest,
2570  DShape& dptestdx) const
2571  {
2572  // Local storage
2573  double psi1[2], psi2[2];
2574  double dpsi1[2], dpsi2[2];
2575 
2576  // Call the OneDimensional Shape functions
2577  OneDimLagrange::shape<2>(s[0], psi1);
2578  OneDimLagrange::shape<2>(s[1], psi2);
2579  OneDimLagrange::dshape<2>(s[0], dpsi1);
2580  OneDimLagrange::dshape<2>(s[1], dpsi2);
2581 
2582  // Now let's loop over the nodal points in the element
2583  // s1 is the "x" coordinate, s2 the "y"
2584  for (unsigned i = 0; i < 2; i++)
2585  {
2586  for (unsigned j = 0; j < 2; j++)
2587  {
2588  /*Multiply the two 1D functions together to get the 2D function*/
2589  ppsi[2 * i + j] = psi2[i] * psi1[j];
2590  dppsidx(2 * i + j, 0) = psi2[i] * dpsi1[j];
2591  dppsidx(2 * i + j, 1) = dpsi2[i] * psi1[j];
2592  }
2593  }
2594 
2595 
2596  // Get the values of the shape functions and their local derivatives
2597  Shape psi(9);
2598  DShape dpsi(9, 2);
2599  dshape_local(s, psi, dpsi);
2600 
2601  // Allocate memory for the inverse 2x2 jacobian
2602  DenseMatrix<double> inverse_jacobian(2);
2603 
2604  // Now calculate the inverse jacobian
2605  const double det = local_to_eulerian_mapping(dpsi, inverse_jacobian);
2606 
2607  // Now set the values of the derivatives to be derivs w.r.t. to the
2608  // Eulerian coordinates
2609  transform_derivatives(inverse_jacobian, dppsidx);
2610 
2611  // The test functions are equal to the shape functions
2612  ptest = ppsi;
2613  dptestdx = dppsidx;
2614 
2615  // Return the determinant of the jacobian
2616  return det;
2617  }
2618 
2619 
2620  //==========================================================================
2621  /// 2D :
2622  /// Define the shape functions (psi) and test functions (test) and
2623  /// their derivatives w.r.t. global coordinates (dpsidx and dtestdx)
2624  /// and return Jacobian of mapping (J). Additionally compute the
2625  /// derivatives of dpsidx, dtestdx and J w.r.t. nodal coordinates.
2626  ///
2627  /// Galerkin: Test functions = shape functions
2628  //==========================================================================
2629  template<>
2631  const unsigned& ipt,
2632  Shape& psi,
2633  DShape& dpsidx,
2634  RankFourTensor<double>& d_dpsidx_dX,
2635  Shape& test,
2636  DShape& dtestdx,
2637  RankFourTensor<double>& d_dtestdx_dX,
2638  DenseMatrix<double>& djacobian_dX) const
2639  {
2640  // Call the geometrical shape functions and derivatives
2641  const double J = this->dshape_eulerian_at_knot(
2642  ipt, psi, dpsidx, djacobian_dX, d_dpsidx_dX);
2643 
2644  // Loop over the test functions and derivatives and set them equal to the
2645  // shape functions
2646  for (unsigned i = 0; i < 9; i++)
2647  {
2648  test[i] = psi[i];
2649 
2650  for (unsigned k = 0; k < 2; k++)
2651  {
2652  dtestdx(i, k) = dpsidx(i, k);
2653 
2654  for (unsigned p = 0; p < 2; p++)
2655  {
2656  for (unsigned q = 0; q < 9; q++)
2657  {
2658  d_dtestdx_dX(p, q, i, k) = d_dpsidx_dX(p, q, i, k);
2659  }
2660  }
2661  }
2662  }
2663 
2664  // Return the jacobian
2665  return J;
2666  }
2667 
2668 
2669  //==========================================================================
2670  /// 3D :
2671  /// Define the shape functions (psi) and test functions (test) and
2672  /// their derivatives w.r.t. global coordinates (dpsidx and dtestdx)
2673  /// and return Jacobian of mapping (J). Additionally compute the
2674  /// derivatives of dpsidx, dtestdx and J w.r.t. nodal coordinates.
2675  ///
2676  /// Galerkin: Test functions = shape functions
2677  //==========================================================================
2678  template<>
2680  const unsigned& ipt,
2681  Shape& psi,
2682  DShape& dpsidx,
2683  RankFourTensor<double>& d_dpsidx_dX,
2684  Shape& test,
2685  DShape& dtestdx,
2686  RankFourTensor<double>& d_dtestdx_dX,
2687  DenseMatrix<double>& djacobian_dX) const
2688  {
2689  // Call the geometrical shape functions and derivatives
2690  const double J = this->dshape_eulerian_at_knot(
2691  ipt, psi, dpsidx, djacobian_dX, d_dpsidx_dX);
2692 
2693  // Loop over the test functions and derivatives and set them equal to the
2694  // shape functions
2695  for (unsigned i = 0; i < 27; i++)
2696  {
2697  test[i] = psi[i];
2698 
2699  for (unsigned k = 0; k < 3; k++)
2700  {
2701  dtestdx(i, k) = dpsidx(i, k);
2702 
2703  for (unsigned p = 0; p < 3; p++)
2704  {
2705  for (unsigned q = 0; q < 27; q++)
2706  {
2707  d_dtestdx_dX(p, q, i, k) = d_dpsidx_dX(p, q, i, k);
2708  }
2709  }
2710  }
2711  }
2712 
2713  // Return the jacobian
2714  return J;
2715  }
2716 
2717 
2718  //==========================================================================
2719  /// 2D :
2720  /// Pressure shape functions
2721  //==========================================================================
2722  template<>
2724  Shape& psi) const
2725  {
2726  // Local storage
2727  double psi1[2], psi2[2];
2728  // Call the OneDimensional Shape functions
2729  OneDimLagrange::shape<2>(s[0], psi1);
2730  OneDimLagrange::shape<2>(s[1], psi2);
2731 
2732  // Now let's loop over the nodal points in the element
2733  // s1 is the "x" coordinate, s2 the "y"
2734  for (unsigned i = 0; i < 2; i++)
2735  {
2736  for (unsigned j = 0; j < 2; j++)
2737  {
2738  /*Multiply the two 1D functions together to get the 2D function*/
2739  psi[2 * i + j] = psi2[i] * psi1[j];
2740  }
2741  }
2742  }
2743 
2744 
2745  //==========================================================================
2746  /// 3D :
2747  /// Pressure shape and test functions and derivs w.r.t. to Eulerian coords.
2748  /// Return Jacobian of mapping between local and global coordinates.
2749  //==========================================================================
2750  template<>
2752  const Vector<double>& s,
2753  Shape& ppsi,
2754  DShape& dppsidx,
2755  Shape& ptest,
2756  DShape& dptestdx) const
2757  {
2758  // Local storage
2759  double psi1[2], psi2[2], psi3[2];
2760  double dpsi1[2], dpsi2[2], dpsi3[2];
2761 
2762  // Call the OneDimensional Shape functions
2763  OneDimLagrange::shape<2>(s[0], psi1);
2764  OneDimLagrange::shape<2>(s[1], psi2);
2765  OneDimLagrange::shape<2>(s[2], psi3);
2766  OneDimLagrange::dshape<2>(s[0], dpsi1);
2767  OneDimLagrange::dshape<2>(s[1], dpsi2);
2768  OneDimLagrange::dshape<2>(s[2], dpsi3);
2769 
2770  // Now let's loop over the nodal points in the element
2771  // s0 is the "x" coordinate, s1 the "y", s2 is the "z"
2772  for (unsigned i = 0; i < 2; i++)
2773  {
2774  for (unsigned j = 0; j < 2; j++)
2775  {
2776  for (unsigned k = 0; k < 2; k++)
2777  {
2778  /*Multiply the three 1D functions together to get the 3D function*/
2779  ppsi[4 * i + 2 * j + k] = psi3[i] * psi2[j] * psi1[k];
2780  dppsidx(4 * i + 2 * j + k, 0) = psi3[i] * psi2[j] * dpsi1[k];
2781  dppsidx(4 * i + 2 * j + k, 1) = psi3[i] * dpsi2[j] * psi1[k];
2782  dppsidx(4 * i + 2 * j + k, 2) = dpsi3[i] * psi2[j] * psi1[k];
2783  }
2784  }
2785  }
2786 
2787 
2788  // Get the values of the shape functions and their local derivatives
2789  Shape psi(27);
2790  DShape dpsi(27, 3);
2791  dshape_local(s, psi, dpsi);
2792 
2793  // Allocate memory for the inverse 3x3 jacobian
2794  DenseMatrix<double> inverse_jacobian(3);
2795 
2796  // Now calculate the inverse jacobian
2797  const double det = local_to_eulerian_mapping(dpsi, inverse_jacobian);
2798 
2799  // Now set the values of the derivatives to be derivs w.r.t. to the
2800  // Eulerian coordinates
2801  transform_derivatives(inverse_jacobian, dppsidx);
2802 
2803  // The test functions are equal to the shape functions
2804  ptest = ppsi;
2805  dptestdx = dppsidx;
2806 
2807  // Return the determinant of the jacobian
2808  return det;
2809  }
2810 
2811  //==========================================================================
2812  /// 3D :
2813  /// Pressure shape functions
2814  //==========================================================================
2815  template<>
2817  Shape& psi) const
2818  {
2819  // Local storage
2820  double psi1[2], psi2[2], psi3[2];
2821 
2822  // Call the OneDimensional Shape functions
2823  OneDimLagrange::shape<2>(s[0], psi1);
2824  OneDimLagrange::shape<2>(s[1], psi2);
2825  OneDimLagrange::shape<2>(s[2], psi3);
2826 
2827  // Now let's loop over the nodal points in the element
2828  // s0 is the "x" coordinate, s1 the "y", s2 is the "z"
2829  for (unsigned i = 0; i < 2; i++)
2830  {
2831  for (unsigned j = 0; j < 2; j++)
2832  {
2833  for (unsigned k = 0; k < 2; k++)
2834  {
2835  /*Multiply the three 1D functions together to get the 3D function*/
2836  psi[4 * i + 2 * j + k] = psi3[i] * psi2[j] * psi1[k];
2837  }
2838  }
2839  }
2840  }
2841 
2842 
2843  //==========================================================================
2844  /// Pressure shape and test functions
2845  //==========================================================================
2846  template<unsigned DIM>
2848  Shape& psi,
2849  Shape& test) const
2850  {
2851  // Call the pressure shape functions
2852  this->pshape_nst(s, psi);
2853  // Test functions are shape functions
2854  test = psi;
2855  }
2856 
2857 
2858  //=======================================================================
2859  /// Face geometry of the 2D Taylor_Hood elements
2860  //=======================================================================
2861  template<>
2862  class FaceGeometry<QTaylorHoodElement<2>> : public virtual QElement<1, 3>
2863  {
2864  public:
2865  FaceGeometry() : QElement<1, 3>() {}
2866  };
2867 
2868  //=======================================================================
2869  /// Face geometry of the 3D Taylor_Hood elements
2870  //=======================================================================
2871  template<>
2872  class FaceGeometry<QTaylorHoodElement<3>> : public virtual QElement<2, 3>
2873  {
2874  public:
2875  FaceGeometry() : QElement<2, 3>() {}
2876  };
2877 
2878 
2879  //=======================================================================
2880  /// Face geometry of the FaceGeometry of the 2D Taylor Hoodelements
2881  //=======================================================================
2882  template<>
2884  : public virtual PointElement
2885  {
2886  public:
2888  };
2889 
2890 
2891  //=======================================================================
2892  /// Face geometry of the FaceGeometry of the 3D Taylor_Hood elements
2893  //=======================================================================
2894  template<>
2896  : public virtual QElement<1, 3>
2897  {
2898  public:
2899  FaceGeometry() : QElement<1, 3>() {}
2900  };
2901 
2902 
2903  /// /////////////////////////////////////////////////////////////////
2904  /// /////////////////////////////////////////////////////////////////
2905  /// /////////////////////////////////////////////////////////////////
2906 
2907 
2908  //==========================================================
2909  /// Taylor Hood upgraded to become projectable
2910  //==========================================================
2911  template<class TAYLOR_HOOD_ELEMENT>
2913  : public virtual ProjectableElement<TAYLOR_HOOD_ELEMENT>
2914  {
2915  public:
2916  /// Constructor [this was only required explicitly
2917  /// from gcc 4.5.2 onwards...]
2919 
2920 
2921  /// Specify the values associated with field fld.
2922  /// The information is returned in a vector of pairs which comprise
2923  /// the Data object and the value within it, that correspond to field fld.
2924  /// In the underlying Taylor Hood elements the fld-th velocities are stored
2925  /// at the fld-th value of the nodes; the pressures (the dim-th
2926  /// field) are the dim-th values at the vertex nodes etc.
2928  {
2929  // Create the vector
2930  Vector<std::pair<Data*, unsigned>> data_values;
2931 
2932  // Velocities dofs
2933  if (fld < this->dim())
2934  {
2935  // Loop over all nodes
2936  unsigned nnod = this->nnode();
2937  for (unsigned j = 0; j < nnod; j++)
2938  {
2939  // Add the data value associated with the velocity components
2940  data_values.push_back(std::make_pair(this->node_pt(j), fld));
2941  }
2942  }
2943  // Pressure
2944  else
2945  {
2946  // Loop over all vertex nodes
2947  unsigned Pconv_size = this->dim() + 1;
2948  for (unsigned j = 0; j < Pconv_size; j++)
2949  {
2950  // Add the data value associated with the pressure components
2951  unsigned vertex_index = this->Pconv[j];
2952  data_values.push_back(
2953  std::make_pair(this->node_pt(vertex_index), fld));
2954  }
2955  }
2956 
2957  // Return the vector
2958  return data_values;
2959  }
2960 
2961  /// Number of fields to be projected: dim+1, corresponding to
2962  /// velocity components and pressure
2964  {
2965  return this->dim() + 1;
2966  }
2967 
2968  /// Number of history values to be stored for fld-th field. Whatever
2969  /// the timestepper has set up for the velocity components and
2970  /// none for the pressure field.
2971  /// (Note: count includes current value!)
2972  unsigned nhistory_values_for_projection(const unsigned& fld)
2973  {
2974  if (fld == this->dim())
2975  {
2976  // pressure doesn't have history values
2977  return this->node_pt(0)->ntstorage(); // 1;
2978  }
2979  else
2980  {
2981  return this->node_pt(0)->ntstorage();
2982  }
2983  }
2984 
2985  /// Number of positional history values
2986  /// (Note: count includes current value!)
2988  {
2989  return this->node_pt(0)->position_time_stepper_pt()->ntstorage();
2990  }
2991 
2992  /// Return Jacobian of mapping and shape functions of field fld
2993  /// at local coordinate s
2994  double jacobian_and_shape_of_field(const unsigned& fld,
2995  const Vector<double>& s,
2996  Shape& psi)
2997  {
2998  unsigned n_dim = this->dim();
2999  unsigned n_node = this->nnode();
3000 
3001  if (fld == n_dim)
3002  {
3003  // We are dealing with the pressure
3004  this->pshape_nst(s, psi);
3005 
3006  Shape psif(n_node), testf(n_node);
3007  DShape dpsifdx(n_node, n_dim), dtestfdx(n_node, n_dim);
3008 
3009  // Domain Shape
3010  double J = this->dshape_and_dtest_eulerian_nst(
3011  s, psif, dpsifdx, testf, dtestfdx);
3012  return J;
3013  }
3014  else
3015  {
3016  Shape testf(n_node);
3017  DShape dpsifdx(n_node, n_dim), dtestfdx(n_node, n_dim);
3018 
3019  // Domain Shape
3020  double J =
3021  this->dshape_and_dtest_eulerian_nst(s, psi, dpsifdx, testf, dtestfdx);
3022  return J;
3023  }
3024  }
3025 
3026 
3027  /// Return interpolated field fld at local coordinate s, at time
3028  /// level t (t=0: present; t>0: history values)
3029  double get_field(const unsigned& t,
3030  const unsigned& fld,
3031  const Vector<double>& s)
3032  {
3033  unsigned n_dim = this->dim();
3034  unsigned n_node = this->nnode();
3035 
3036  // If fld=n_dim, we deal with the pressure
3037  if (fld == n_dim)
3038  {
3039  return this->interpolated_p_nst(t, s);
3040  }
3041  // Velocity
3042  else
3043  {
3044  // Find the index at which the variable is stored
3045  unsigned u_nodal_index = this->u_index_nst(fld);
3046 
3047  // Local shape function
3048  Shape psi(n_node);
3049 
3050  // Find values of shape function
3051  this->shape(s, psi);
3052 
3053  // Initialise value of u
3054  double interpolated_u = 0.0;
3055 
3056  // Sum over the local nodes at that time
3057  for (unsigned l = 0; l < n_node; l++)
3058  {
3059  interpolated_u += this->nodal_value(t, l, u_nodal_index) * psi[l];
3060  }
3061  return interpolated_u;
3062  }
3063  }
3064 
3065 
3066  /// Return number of values in field fld
3067  unsigned nvalue_of_field(const unsigned& fld)
3068  {
3069  if (fld == this->dim())
3070  {
3071  return this->npres_nst();
3072  }
3073  else
3074  {
3075  return this->nnode();
3076  }
3077  }
3078 
3079 
3080  /// Return local equation number of value j in field fld.
3081  int local_equation(const unsigned& fld, const unsigned& j)
3082  {
3083  if (fld == this->dim())
3084  {
3085  return this->p_local_eqn(j);
3086  }
3087  else
3088  {
3089  const unsigned u_nodal_index = this->u_index_nst(fld);
3090  return this->nodal_local_eqn(j, u_nodal_index);
3091  }
3092  }
3093  };
3094 
3095 
3096  //=======================================================================
3097  /// Face geometry for element is the same as that for the underlying
3098  /// wrapped element
3099  //=======================================================================
3100  template<class ELEMENT>
3102  : public virtual FaceGeometry<ELEMENT>
3103  {
3104  public:
3105  FaceGeometry() : FaceGeometry<ELEMENT>() {}
3106  };
3107 
3108 
3109  //=======================================================================
3110  /// Face geometry of the Face Geometry for element is the same as
3111  /// that for the underlying wrapped element
3112  //=======================================================================
3113  template<class ELEMENT>
3115  : public virtual FaceGeometry<FaceGeometry<ELEMENT>>
3116  {
3117  public:
3119  };
3120 
3121 
3122  //==========================================================
3123  /// Crouzeix Raviart upgraded to become projectable
3124  //==========================================================
3125  template<class CROUZEIX_RAVIART_ELEMENT>
3127  : public virtual ProjectableElement<CROUZEIX_RAVIART_ELEMENT>
3128  {
3129  public:
3130  /// Constructor [this was only required explicitly
3131  /// from gcc 4.5.2 onwards...]
3133 
3134  /// Specify the values associated with field fld.
3135  /// The information is returned in a vector of pairs which comprise
3136  /// the Data object and the value within it, that correspond to field fld.
3137  /// In the underlying Crouzeix Raviart elements the
3138  /// fld-th velocities are stored
3139  /// at the fld-th value of the nodes; the pressures are stored internally
3141  {
3142  // Create the vector
3143  Vector<std::pair<Data*, unsigned>> data_values;
3144 
3145  // Velocities dofs
3146  if (fld < this->dim())
3147  {
3148  // Loop over all nodes
3149  const unsigned n_node = this->nnode();
3150  for (unsigned n = 0; n < n_node; n++)
3151  {
3152  // Add the data value associated with the velocity components
3153  data_values.push_back(std::make_pair(this->node_pt(n), fld));
3154  }
3155  }
3156  // Pressure
3157  else
3158  {
3159  // Need to push back the internal data
3160  const unsigned n_press = this->npres_nst();
3161  // Loop over all pressure values
3162  for (unsigned j = 0; j < n_press; j++)
3163  {
3164  data_values.push_back(std::make_pair(
3165  this->internal_data_pt(this->P_nst_internal_index), j));
3166  }
3167  }
3168 
3169  // Return the vector
3170  return data_values;
3171  }
3172 
3173  /// Number of fields to be projected: dim+1, corresponding to
3174  /// velocity components and pressure
3176  {
3177  return this->dim() + 1;
3178  }
3179 
3180  /// Number of history values to be stored for fld-th field. Whatever
3181  /// the timestepper has set up for the velocity components and
3182  /// none for the pressure field.
3183  /// (Note: count includes current value!)
3184  unsigned nhistory_values_for_projection(const unsigned& fld)
3185  {
3186  if (fld == this->dim())
3187  {
3188  // pressure doesn't have history values
3189  return 1;
3190  }
3191  else
3192  {
3193  return this->node_pt(0)->ntstorage();
3194  }
3195  }
3196 
3197  /// Number of positional history values.
3198  /// (Note: count includes current value!)
3200  {
3201  return this->node_pt(0)->position_time_stepper_pt()->ntstorage();
3202  }
3203 
3204  /// Return Jacobian of mapping and shape functions of field fld
3205  /// at local coordinate s
3206  double jacobian_and_shape_of_field(const unsigned& fld,
3207  const Vector<double>& s,
3208  Shape& psi)
3209  {
3210  unsigned n_dim = this->dim();
3211  unsigned n_node = this->nnode();
3212 
3213  if (fld == n_dim)
3214  {
3215  // We are dealing with the pressure
3216  this->pshape_nst(s, psi);
3217 
3218  Shape psif(n_node), testf(n_node);
3219  DShape dpsifdx(n_node, n_dim), dtestfdx(n_node, n_dim);
3220 
3221  // Domain Shape
3222  double J = this->dshape_and_dtest_eulerian_nst(
3223  s, psif, dpsifdx, testf, dtestfdx);
3224  return J;
3225  }
3226  else
3227  {
3228  Shape testf(n_node);
3229  DShape dpsifdx(n_node, n_dim), dtestfdx(n_node, n_dim);
3230 
3231  // Domain Shape
3232  double J =
3233  this->dshape_and_dtest_eulerian_nst(s, psi, dpsifdx, testf, dtestfdx);
3234  return J;
3235  }
3236  }
3237 
3238 
3239  /// Return interpolated field fld at local coordinate s, at time
3240  /// level t (t=0: present; t>0: history values)
3241  double get_field(const unsigned& t,
3242  const unsigned& fld,
3243  const Vector<double>& s)
3244  {
3245  unsigned n_dim = this->dim();
3246 
3247  // If fld=n_dim, we deal with the pressure
3248  if (fld == n_dim)
3249  {
3250  return this->interpolated_p_nst(s);
3251  }
3252  // Velocity
3253  else
3254  {
3255  return this->interpolated_u_nst(t, s, fld);
3256  }
3257  }
3258 
3259 
3260  /// Return number of values in field fld
3261  unsigned nvalue_of_field(const unsigned& fld)
3262  {
3263  if (fld == this->dim())
3264  {
3265  return this->npres_nst();
3266  }
3267  else
3268  {
3269  return this->nnode();
3270  }
3271  }
3272 
3273 
3274  /// Return local equation number of value j in field fld.
3275  int local_equation(const unsigned& fld, const unsigned& j)
3276  {
3277  if (fld == this->dim())
3278  {
3279  return this->p_local_eqn(j);
3280  }
3281  else
3282  {
3283  const unsigned u_nodal_index = this->u_index_nst(fld);
3284  return this->nodal_local_eqn(j, u_nodal_index);
3285  }
3286  }
3287  };
3288 
3289 
3290  //=======================================================================
3291  /// Face geometry for element is the same as that for the underlying
3292  /// wrapped element
3293  //=======================================================================
3294  template<class ELEMENT>
3296  : public virtual FaceGeometry<ELEMENT>
3297  {
3298  public:
3299  FaceGeometry() : FaceGeometry<ELEMENT>() {}
3300  };
3301 
3302 
3303  //=======================================================================
3304  /// Face geometry of the Face Geometry for element is the same as
3305  /// that for the underlying wrapped element
3306  //=======================================================================
3307  template<class ELEMENT>
3309  : public virtual FaceGeometry<FaceGeometry<ELEMENT>>
3310  {
3311  public:
3313  };
3314 
3315 
3316 } // namespace oomph
3317 
3318 #endif
e
Definition: cfortran.h:571
static char t char * s
Definition: cfortran.h:568
cstr elem_len * i
Definition: cfortran.h:603
char t
Definition: cfortran.h:568
A Class for the derivatives of shape functions The class design is essentially the same as Shape,...
Definition: shape.h:278
A class that represents a collection of data; each Data object may contain many different individual ...
Definition: nodes.h:86
long & eqn_number(const unsigned &i)
Return the equation number of the i-th stored variable.
Definition: nodes.h:367
void pin(const unsigned &i)
Pin the i-th stored variable.
Definition: nodes.h:385
TimeStepper *& time_stepper_pt()
Return the pointer to the timestepper.
Definition: nodes.h:238
void set_value(const unsigned &i, const double &value_)
Set the i-th stored data value to specified value. The only reason that we require an explicit set fu...
Definition: nodes.h:271
unsigned nvalue() const
Return number of values stored in data object (incl pinned ones).
Definition: nodes.h:483
double value(const unsigned &i) const
Return i-th stored value. This function is not virtual so that it can be inlined. This means that if ...
Definition: nodes.h:293
unsigned ntstorage() const
Return total number of doubles stored per value to record time history of each value (one for steady ...
Definition: nodes.cc:879
/////////////////////////////////////////////////////////////////////////
Definition: fsi.h:63
FaceElements are elements that coincide with the faces of higher-dimensional "bulk" elements....
Definition: elements.h:4338
int & face_index()
Index of the face (a number that uniquely identifies the face in the element)
Definition: elements.h:4626
void outer_unit_normal(const Vector< double > &s, Vector< double > &unit_normal) const
Compute outer unit normal at the specified local coordinate.
Definition: elements.cc:6006
double interpolated_x(const Vector< double > &s, const unsigned &i) const
Return FE interpolated coordinate x[i] at local coordinate s. Overloaded to get information from bulk...
Definition: elements.h:4528
//////////////////////////////////////////////////////////////////// ////////////////////////////////...
Definition: elements.h:4998
A general Finite Element class.
Definition: elements.h:1313
virtual unsigned nplot_points_paraview(const unsigned &nplot) const
Return the number of actual plot points for paraview plot with parameter nplot. Broken virtual; can b...
Definition: elements.h:2862
Node *& node_pt(const unsigned &n)
Return a pointer to the local node n.
Definition: elements.h:2175
double nodal_value(const unsigned &n, const unsigned &i) const
Return the i-th value stored at local node n. Produces suitably interpolated values for hanging nodes...
Definition: elements.h:2593
virtual void output(std::ostream &outfile)
Output the element data — typically the values at the nodes in a format suitable for post-processing.
Definition: elements.h:3050
double size() const
Calculate the size of the element (length, area, volume,...) in Eulerian computational coordinates....
Definition: elements.cc:4290
virtual void shape(const Vector< double > &s, Shape &psi) const =0
Calculate the geometric shape functions at local coordinate s. This function must be overloaded for e...
virtual double interpolated_x(const Vector< double > &s, const unsigned &i) const
Return FE interpolated coordinate x[i] at local coordinate s.
Definition: elements.cc:3962
int nodal_local_eqn(const unsigned &n, const unsigned &i) const
Return the local equation number corresponding to the i-th value at the n-th local node.
Definition: elements.h:1432
unsigned dim() const
Return the spatial dimension of the element, i.e. the number of local coordinates required to paramet...
Definition: elements.h:2611
unsigned nnode() const
Return the number of nodes.
Definition: elements.h:2210
void(* SteadyExactSolutionFctPt)(const Vector< double > &, Vector< double > &)
Function pointer for function that computes vector-valued steady "exact solution" as .
Definition: elements.h:1759
Integral *const & integral_pt() const
Return the pointer to the integration scheme (const version)
Definition: elements.h:1963
virtual void get_s_plot(const unsigned &i, const unsigned &nplot, Vector< double > &s, const bool &shifted_to_interior=false) const
Get cector of local coordinates of plot point i (when plotting nplot points in each "coordinate direc...
Definition: elements.h:3148
double dshape_eulerian(const Vector< double > &s, Shape &psi, DShape &dpsidx) const
Compute the geometric shape functions and also first derivatives w.r.t. global coordinates at local c...
Definition: elements.cc:3298
virtual void build_face_element(const int &face_index, FaceElement *face_element_pt)
Function for building a lower dimensional FaceElement on the specified face of the FiniteElement....
Definition: elements.cc:5132
void(* UnsteadyExactSolutionFctPt)(const double &, const Vector< double > &, Vector< double > &)
Function pointer for function that computes Vector-valued time-dependent function as .
Definition: elements.h:1765
bool has_hanging_nodes() const
Return boolean to indicate if any of the element's nodes are geometrically hanging.
Definition: elements.h:2470
Helper class for elements that impose Robin boundary conditions on pressure advection diffusion probl...
virtual ~FpPressureAdvDiffRobinBCElementBase()
Empty virtual destructor.
virtual void fill_in_generic_residual_contribution_fp_press_adv_diff_robin_bc(Vector< double > &residuals, DenseMatrix< double > &jacobian, unsigned flag)=0
This function returns the residuals for the traction function. flag=1 (or 0): do (or don't) compute t...
//////////////////////////////////////////////////////////////////// ////////////////////////////////...
void output(std::ostream &outfile)
Overload the output function.
virtual void fill_in_generic_residual_contribution_fp_press_adv_diff_robin_bc(Vector< double > &residuals, DenseMatrix< double > &jacobian, unsigned flag)
This function returns the residuals for the traction function. flag=1 (or 0): do (or don't) compute t...
FpPressureAdvDiffRobinBCElement(FiniteElement *const &element_pt, const int &face_index, const bool &called_from_refineable_constructor=false)
Constructor, which takes a "bulk" element and the value of the index and its limit....
void fill_in_contribution_to_jacobian(Vector< double > &residuals, DenseMatrix< double > &jacobian)
This function returns the residuals and the jacobian.
void fill_in_contribution_to_residuals(Vector< double > &residuals)
This function returns just the residuals.
void output(std::ostream &outfile, const unsigned &nplot)
Output function: x,y,[z],u,v,[w],p in tecplot format.
static double Default_fd_jacobian_step
Double used for the default finite difference step in elemental jacobian calculations.
Definition: elements.h:1198
unsigned add_internal_data(Data *const &data_pt, const bool &fd=true)
Add a (pointer to an) internal data object to the element and return the index required to obtain it ...
Definition: elements.cc:62
Data *& internal_data_pt(const unsigned &i)
Return a pointer to i-th internal data object.
Definition: elements.h:622
static DenseMatrix< double > Dummy_matrix
Empty dense matrix used as a dummy argument to combined residual and jacobian functions in the case w...
Definition: elements.h:227
unsigned ninternal_data() const
Return the number of internal data objects.
Definition: elements.h:823
int internal_local_eqn(const unsigned &i, const unsigned &j) const
Return the local equation number corresponding to the j-th value stored at the i-th internal data.
Definition: elements.h:267
TimeStepper *& time_stepper_pt()
Access function for pointer to time stepper: Null if object is not time-dependent.
Definition: geom_objects.h:192
virtual double knot(const unsigned &i, const unsigned &j) const =0
Return local coordinate s[j] of i-th integration point.
virtual unsigned nweight() const =0
Return the number of integration points of the scheme.
//////////////////////////////////////////////////////////////////// ////////////////////////////////...
Definition: elements.h:5231
//////////////////////////////////////////////////////////////////// ////////////////////////////////...
unsigned nscalar_paraview() const
Number of scalars/fields output by this element. Reimplements broken virtual function in base class.
void(* NavierStokesBodyForceFctPt)(const double &time, const Vector< double > &x, Vector< double > &body_force)
Function pointer to body force function fct(t,x,f(x)) x is a Vector!
double * Viscosity_Ratio_pt
Pointer to the viscosity ratio (relative to the viscosity used in the definition of the Reynolds numb...
void disable_ALE()
Disable ALE, i.e. assert the mesh is not moving – you do this at your own risk!
virtual double p_nst(const unsigned &t, const unsigned &n_p) const =0
Pressure at local pressure "node" n_p at time level t.
double interpolated_u_nst(const unsigned &t, const Vector< double > &s, const unsigned &i) const
Return FE interpolated velocity u[i] at local coordinate s at time level t (t=0: present; t>0: histor...
void get_vorticity(const Vector< double > &s, Vector< double > &vorticity) const
Compute the vorticity vector at local coordinate s.
NavierStokesSourceFctPt Source_fct_pt
Pointer to volumetric source function.
double kin_energy() const
Get integral of kinetic energy over element.
virtual unsigned u_index_nst(const unsigned &i) const
Return the index at which the i-th unknown velocity component is stored. The default value,...
virtual void build_fp_press_adv_diff_robin_bc_element(const unsigned &face_index)=0
Build FaceElements that apply the Robin boundary condition to the pressure advection diffusion proble...
double * Re_pt
Pointer to global Reynolds number.
double(* NavierStokesSourceFctPt)(const double &time, const Vector< double > &x)
Function pointer to source function fct(t,x) x is a Vector!
NavierStokesSourceFctPt & source_fct_pt()
Access function for the source-function pointer.
static int Pressure_not_stored_at_node
Static "magic" number that indicates that the pressure is not stored at a node.
double *& density_ratio_pt()
Pointer to Density ratio.
NavierStokesBodyForceFctPt body_force_fct_pt() const
Access function for the body-force pointer. Const version.
void fill_in_contribution_to_dresiduals_dparameter(double *const &parameter_pt, Vector< double > &dres_dparam)
Compute the element's residual Vector.
void fill_in_contribution_to_hessian_vector_products(Vector< double > const &Y, DenseMatrix< double > const &C, DenseMatrix< double > &product)
Compute the hessian tensor vector products required to perform continuation of bifurcations analytica...
void get_pressure_and_velocity_mass_matrix_diagonal(Vector< double > &press_mass_diag, Vector< double > &veloc_mass_diag, const unsigned &which_one=0)
Compute the diagonal of the velocity/pressure mass matrices. If which one=0, both are computed,...
void get_vorticity(const Vector< double > &s, double &vorticity) const
Compute the scalar vorticity at local coordinate s (2D)
virtual double dpshape_and_dptest_eulerian_nst(const Vector< double > &s, Shape &ppsi, DShape &dppsidx, Shape &ptest, DShape &dptestdx) const =0
Compute the pressure shape and test functions and derivatives w.r.t. global coords at local coordinat...
void fill_in_contribution_to_residuals(Vector< double > &residuals)
Compute the element's residual Vector.
void interpolated_u_nst(const Vector< double > &s, Vector< double > &veloc) const
Compute vector of FE interpolated velocity u at local coordinate s.
virtual int p_nodal_index_nst() const
Return the index at which the pressure is stored if it is stored at the nodes. If not stored at the n...
void compute_error(std::ostream &outfile, FiniteElement::UnsteadyExactSolutionFctPt exact_soln_pt, const double &time, double &error, double &norm)
Validate against exact solution at given time Solution is provided via function pointer....
double interpolated_dudx_nst(const Vector< double > &s, const unsigned &i, const unsigned &j) const
Return FE interpolated derivatives of velocity component u[i] w.r.t spatial global coordinate directi...
void fill_in_pressure_advection_diffusion_jacobian(Vector< double > &residuals, DenseMatrix< double > &jacobian)
Compute the residuals and Jacobian for the associated pressure advection diffusion problem....
virtual double dshape_and_dtest_eulerian_at_knot_nst(const unsigned &ipt, Shape &psi, DShape &dpsidx, Shape &test, DShape &dtestdx) const =0
Compute the shape functions and derivatives w.r.t. global coords at ipt-th integration point Return J...
virtual void pshape_nst(const Vector< double > &s, Shape &psi) const =0
Compute the pressure shape functions at local coordinate s.
virtual unsigned npres_nst() const =0
Function to return number of pressure degrees of freedom.
NavierStokesPressureAdvDiffSourceFctPt Press_adv_diff_source_fct_pt
Pointer to source function pressure advection diffusion equation (only to be used during validation)
virtual void fill_in_generic_pressure_advection_diffusion_contribution_nst(Vector< double > &residuals, DenseMatrix< double > &jacobian, unsigned flag)
Compute the residuals for the associated pressure advection diffusion problem. Used by the Fp precond...
void fill_in_contribution_to_jacobian_and_mass_matrix(Vector< double > &residuals, DenseMatrix< double > &jacobian, DenseMatrix< double > &mass_matrix)
Add the element's contribution to its residuals vector, jacobian matrix and mass matrix.
virtual void pshape_nst(const Vector< double > &s, Shape &psi, Shape &test) const =0
Compute the pressure shape and test functions at local coordinate s.
virtual void get_dresidual_dnodal_coordinates(RankThreeTensor< double > &dresidual_dnodal_coordinates)
Compute derivatives of elemental residual vector with respect to nodal coordinates....
virtual void get_source_gradient_nst(const double &time, const unsigned &ipt, const Vector< double > &x, Vector< double > &gradient)
Get gradient of source term at (Eulerian) position x. This function is virtual to allow overloading i...
void scalar_value_paraview(std::ofstream &file_out, const unsigned &i, const unsigned &nplot) const
Write values of the i-th scalar field at the plot points. Needs to be implemented for each new specif...
double interpolated_u_nst(const Vector< double > &s, const unsigned &i) const
Return FE interpolated velocity u[i] at local coordinate s.
double *& viscosity_ratio_pt()
Pointer to Viscosity Ratio.
double u_nst(const unsigned &n, const unsigned &i) const
Velocity i at local node n. Uses suitably interpolated value for hanging nodes. The use of u_index_ns...
NavierStokesPressureAdvDiffSourceFctPt & source_fct_for_pressure_adv_diff()
Access function for the source-function pointer for pressure advection diffusion (used for validation...
void scalar_value_fct_paraview(std::ofstream &file_out, const unsigned &i, const unsigned &nplot, const double &time, FiniteElement::UnsteadyExactSolutionFctPt exact_soln_pt) const
Write values of the i-th scalar field at the plot points. Needs to be implemented for each new specif...
void strain_rate(const Vector< double > &s, DenseMatrix< double > &strain_rate) const
Strain-rate tensor: 1/2 (du_i/dx_j + du_j/dx_i)
NavierStokesPressureAdvDiffSourceFctPt source_fct_for_pressure_adv_diff() const
Access function for the source-function pointer for pressure advection diffusion (used for validation...
double d_kin_energy_dt() const
Get integral of time derivative of kinetic energy over element.
const double & re_st() const
Product of Reynolds and Strouhal number (=Womersley number)
virtual void get_body_force_nst(const double &time, const unsigned &ipt, const Vector< double > &s, const Vector< double > &x, Vector< double > &result)
Calculate the body force at a given time and local and/or Eulerian position. This function is virtual...
double interpolated_p_nst(const unsigned &t, const Vector< double > &s) const
Return FE interpolated pressure at local coordinate s at time level t.
void enable_ALE()
(Re-)enable ALE, i.e. take possible mesh motion into account when evaluating the time-derivative....
virtual void get_body_force_gradient_nst(const double &time, const unsigned &ipt, const Vector< double > &s, const Vector< double > &x, DenseMatrix< double > &d_body_force_dx)
Get gradient of body force term at (Eulerian) position x. This function is virtual to allow overloadi...
void compute_norm(double &norm)
Compute norm of solution: square of the L2 norm of the velocities.
double *& re_st_pt()
Pointer to product of Reynolds and Strouhal number (=Womersley number)
void fill_in_contribution_to_djacobian_dparameter(double *const &parameter_pt, Vector< double > &dres_dparam, DenseMatrix< double > &djac_dparam)
Compute the element's residual Vector and the jacobian matrix Virtual function can be overloaded by h...
NavierStokesBodyForceFctPt & body_force_fct_pt()
Access function for the body-force pointer.
const double & re_invfr() const
Global inverse Froude number.
int & pinned_fp_pressure_eqn()
Global eqn number of pressure dof that's pinned in pressure adv diff problem.
int Pinned_fp_pressure_eqn
Global eqn number of pressure dof that's pinned in pressure advection diffusion problem (defaults to ...
const double & viscosity_ratio() const
Viscosity ratio for element: Element's viscosity relative to the viscosity used in the definition of ...
virtual void fix_pressure(const unsigned &p_dof, const double &p_value)=0
Pin p_dof-th pressure dof and set it to value specified by p_value.
NavierStokesSourceFctPt source_fct_pt() const
Access function for the source-function pointer. Const version.
virtual void dinterpolated_u_nst_ddata(const Vector< double > &s, const unsigned &i, Vector< double > &du_ddata, Vector< unsigned > &global_eqn_number)
Compute the derivatives of the i-th component of velocity at point s with respect to all data that ca...
Vector< double > * G_pt
Pointer to global gravity Vector.
static double Default_Physical_Constant_Value
Static default value for the physical constants (all initialised to zero)
void output_pressure_advection_diffusion_robin_elements(std::ostream &outfile)
Output the FaceElements that apply the Robin boundary condition to the pressure advection diffusion p...
void full_output(std::ostream &outfile)
Full output function: x,y,[z],u,v,[w],p,du/dt,dv/dt,[dw/dt],dissipation in tecplot format....
void delete_pressure_advection_diffusion_robin_elements()
Delete the FaceElements that apply the Robin boundary condition to the pressure advection diffusion p...
virtual void fill_in_generic_dresidual_contribution_nst(double *const &parameter_pt, Vector< double > &dres_dparam, DenseMatrix< double > &djac_dparam, DenseMatrix< double > &dmass_matrix_dparam, unsigned flag)
Compute the derivatives of the residuals for the Navier–Stokes equations with respect to a parameter ...
virtual double interpolated_p_nst(const Vector< double > &s) const
Return FE interpolated pressure at local coordinate s.
double * ReInvFr_pt
Pointer to global Reynolds number x inverse Froude number (= Bond number / Capillary number)
Vector< double > *& g_pt()
Pointer to Vector of gravitational components.
void output_veloc(std::ostream &outfile, const unsigned &nplot, const unsigned &t)
Output function: x,y,[z],u,v,[w] in tecplot format. nplot points in each coordinate direction at time...
static double Default_Physical_Ratio_Value
Static default value for the physical ratios (all are initialised to one)
void fill_in_contribution_to_jacobian(Vector< double > &residuals, DenseMatrix< double > &jacobian)
Compute the element's residual Vector and the jacobian matrix Virtual function can be overloaded by h...
void fill_in_pressure_advection_diffusion_residuals(Vector< double > &residuals)
Compute the residuals for the associated pressure advection diffusion problem. Used by the Fp precond...
void get_traction(const Vector< double > &s, const Vector< double > &N, Vector< double > &traction)
Compute traction (on the viscous scale) exerted onto the fluid at local coordinate s....
void pin_all_non_pressure_dofs(std::map< Data *, std::vector< int >> &eqn_number_backup)
Pin all non-pressure dofs and backup eqn numbers.
virtual double get_source_nst(const double &time, const unsigned &ipt, const Vector< double > &x)
Calculate the source fct at given time and Eulerian position.
double *& re_invfr_pt()
Pointer to global inverse Froude number.
double u_nst(const unsigned &t, const unsigned &n, const unsigned &i) const
Velocity i at local node n at timestep t (t=0: present; t>0: previous). Uses suitably interpolated va...
const double & re() const
Reynolds number.
virtual void fill_in_generic_residual_contribution_nst(Vector< double > &residuals, DenseMatrix< double > &jacobian, DenseMatrix< double > &mass_matrix, unsigned flag)
Compute the residuals for the Navier–Stokes equations. Flag=1 (or 0): do (or don't) compute the Jacob...
void output(FILE *file_pt)
C-style output function: x,y,[z],u,v,[w],p in tecplot format. Default number of plot points.
double * Density_Ratio_pt
Pointer to the density ratio (relative to the density used in the definition of the Reynolds number)
double *& re_pt()
Pointer to Reynolds number.
double pressure_integral() const
Integral of pressure over element.
static Vector< double > Default_Gravity_vector
Static default value for the gravity vector.
unsigned n_u_nst() const
Return the number of velocity components Used in FluidInterfaceElements.
double du_dt_nst(const unsigned &n, const unsigned &i) const
i-th component of du/dt at local node n. Uses suitably interpolated value for hanging nodes.
NavierStokesBodyForceFctPt Body_force_fct_pt
Pointer to body force function.
virtual double p_nst(const unsigned &n_p) const =0
Pressure at local pressure "node" n_p Uses suitably interpolated value for hanging nodes.
const double & density_ratio() const
Density ratio for element: Element's density relative to the viscosity used in the definition of the ...
void output(std::ostream &outfile)
Output function: x,y,[z],u,v,[w],p in tecplot format. Default number of plot points.
void output_vorticity(std::ostream &outfile, const unsigned &nplot)
Output function: x,y,[z], [omega_x,omega_y,[and/or omega_z]] in tecplot format. nplot points in each ...
double * ReSt_pt
Pointer to global Reynolds number x Strouhal number (=Womersley)
double(* NavierStokesPressureAdvDiffSourceFctPt)(const Vector< double > &x)
Function pointer to source function fct(x) for the pressure advection diffusion equation (only used d...
bool ALE_is_disabled
Boolean flag to indicate if ALE formulation is disabled when time-derivatives are computed....
NavierStokesEquations()
Constructor: NULL the body force and source function and make sure the ALE terms are included by defa...
virtual double dshape_and_dtest_eulerian_nst(const Vector< double > &s, Shape &psi, DShape &dpsidx, Shape &test, DShape &dtestdx) const =0
Compute the shape functions and derivatives w.r.t. global coords at local coordinate s....
const Vector< double > & g() const
Vector of gravitational components.
void get_load(const Vector< double > &s, const Vector< double > &N, Vector< double > &load)
This implements a pure virtual function defined in the FSIFluidElement class. The function computes t...
double dissipation() const
Return integral of dissipation over element.
Vector< FpPressureAdvDiffRobinBCElementBase * > Pressure_advection_diffusion_robin_element_pt
Storage for FaceElements that apply Robin BC for pressure adv diff equation used in Fp preconditioner...
std::string scalar_name_paraview(const unsigned &i) const
Name of the i-th scalar field. Default implementation returns V1 for the first one,...
virtual double dshape_and_dtest_eulerian_at_knot_nst(const unsigned &ipt, Shape &psi, DShape &dpsidx, RankFourTensor< double > &d_dpsidx_dX, Shape &test, DShape &dtestdx, RankFourTensor< double > &d_dtestdx_dX, DenseMatrix< double > &djacobian_dX) const =0
Shape/test functions and derivs w.r.t. to global coords at integration point ipt; return Jacobian of ...
void fill_in_contribution_to_djacobian_and_dmass_matrix_dparameter(double *const &parameter_pt, Vector< double > &dres_dparam, DenseMatrix< double > &djac_dparam, DenseMatrix< double > &dmass_matrix_dparam)
Add the element's contribution to its residuals vector, jacobian matrix and mass matrix.
void output_fct(std::ostream &outfile, const unsigned &nplot, FiniteElement::SteadyExactSolutionFctPt exact_soln_pt)
Output exact solution specified via function pointer at a given number of plot points....
void point_output_data(const Vector< double > &s, Vector< double > &data)
Output solution in data vector at local cordinates s: x,y [,z], u,v,[w], p.
static Vector< double > Gamma
Vector to decide whether the stress-divergence form is used or not.
Nodes are derived from Data, but, in addition, have a definite (Eulerian) position in a space of a gi...
Definition: nodes.h:906
TimeStepper *& position_time_stepper_pt()
Return a pointer to the position timestepper.
Definition: nodes.h:1022
An OomphLibError object which should be thrown when an run-time error is encountered....
/////////////////////////////////////////////////////////////////////// /////////////////////////////...
Definition: elements.h:3439
Crouzeix Raviart upgraded to become projectable.
Vector< std::pair< Data *, unsigned > > data_values_of_field(const unsigned &fld)
Specify the values associated with field fld. The information is returned in a vector of pairs which ...
int local_equation(const unsigned &fld, const unsigned &j)
Return local equation number of value j in field fld.
ProjectableCrouzeixRaviartElement()
Constructor [this was only required explicitly from gcc 4.5.2 onwards...].
double get_field(const unsigned &t, const unsigned &fld, const Vector< double > &s)
Return interpolated field fld at local coordinate s, at time level t (t=0: present; t>0: history valu...
unsigned nfields_for_projection()
Number of fields to be projected: dim+1, corresponding to velocity components and pressure.
double jacobian_and_shape_of_field(const unsigned &fld, const Vector< double > &s, Shape &psi)
Return Jacobian of mapping and shape functions of field fld at local coordinate s.
unsigned nhistory_values_for_coordinate_projection()
Number of positional history values. (Note: count includes current value!)
unsigned nhistory_values_for_projection(const unsigned &fld)
Number of history values to be stored for fld-th field. Whatever the timestepper has set up for the v...
unsigned nvalue_of_field(const unsigned &fld)
Return number of values in field fld.
Wrapper class for projectable elements. Adds "projectability" to the underlying ELEMENT.
Definition: projection.h:183
///////////////////////////////////////////////////////////////// ///////////////////////////////////...
Vector< std::pair< Data *, unsigned > > data_values_of_field(const unsigned &fld)
Specify the values associated with field fld. The information is returned in a vector of pairs which ...
double jacobian_and_shape_of_field(const unsigned &fld, const Vector< double > &s, Shape &psi)
Return Jacobian of mapping and shape functions of field fld at local coordinate s.
unsigned nhistory_values_for_projection(const unsigned &fld)
Number of history values to be stored for fld-th field. Whatever the timestepper has set up for the v...
unsigned nhistory_values_for_coordinate_projection()
Number of positional history values (Note: count includes current value!)
unsigned nfields_for_projection()
Number of fields to be projected: dim+1, corresponding to velocity components and pressure.
int local_equation(const unsigned &fld, const unsigned &j)
Return local equation number of value j in field fld.
unsigned nvalue_of_field(const unsigned &fld)
Return number of values in field fld.
ProjectableTaylorHoodElement()
Constructor [this was only required explicitly from gcc 4.5.2 onwards...].
double get_field(const unsigned &t, const unsigned &fld, const Vector< double > &s)
Return interpolated field fld at local coordinate s, at time level t (t=0: present; t>0: history valu...
/////////////////////////////////////////////////////////////////////////// /////////////////////////...
virtual unsigned required_nvalue(const unsigned &n) const
Number of values (pinned or dofs) required at local node n.
unsigned npres_nst() const
Return number of pressure values.
void output(std::ostream &outfile)
Redirect output to NavierStokesEquations output.
void pshape_nst(const Vector< double > &s, Shape &psi) const
Pressure shape functions at local coordinate s.
int p_local_eqn(const unsigned &n) const
Return the local equation numbers for the pressure values.
double dpshape_and_dptest_eulerian_nst(const Vector< double > &s, Shape &ppsi, DShape &dppsidx, Shape &ptest, DShape &dptestdx) const
Pressure shape and test functions and their derivs w.r.t. to global coords at local coordinate s (tak...
double dshape_and_dtest_eulerian_at_knot_nst(const unsigned &ipt, Shape &psi, DShape &dpsidx, Shape &test, DShape &dtestdx) const
Velocity shape and test functions and their derivs w.r.t. to global coords at ipt-th integation point...
double p_nst(const unsigned &i) const
Return the i-th pressure value (Discontinous pressure interpolation – no need to cater for hanging no...
unsigned P_nst_internal_index
Internal index that indicates at which internal data the pressure is stored.
void identify_load_data(std::set< std::pair< Data *, unsigned >> &paired_load_data)
Add to the set paired_load_data pairs containing.
void build_fp_press_adv_diff_robin_bc_element(const unsigned &face_index)
Build FaceElements that apply the Robin boundary condition to the pressure advection diffusion proble...
void output(std::ostream &outfile, const unsigned &nplot)
Redirect output to NavierStokesEquations output.
static const unsigned Initial_Nvalue[]
Static array of ints to hold required number of variables at nodes.
void output(FILE *file_pt, const unsigned &nplot)
Redirect output to NavierStokesEquations output.
QCrouzeixRaviartElement()
Constructor, there are DIM+1 internal values (for the pressure)
void full_output(std::ostream &outfile)
Full output function: x,y,[z],u,v,[w],p,du/dt,dv/dt,[dw/dt],dissipation in tecplot format....
double dshape_and_dtest_eulerian_at_knot_nst(const unsigned &ipt, Shape &psi, DShape &dpsidx, RankFourTensor< double > &d_dpsidx_dX, Shape &test, DShape &dtestdx, RankFourTensor< double > &d_dtestdx_dX, DenseMatrix< double > &djacobian_dX) const
Shape/test functions and derivs w.r.t. to global coords at integration point ipt; return Jacobian of ...
void full_output(std::ostream &outfile, const unsigned &nplot)
Full output function: x,y,[z],u,v,[w],p,du/dt,dv/dt,[dw/dt],dissipation in tecplot format....
void identify_pressure_data(std::set< std::pair< Data *, unsigned >> &paired_pressure_data)
Add to the set paired_pressure_data pairs containing.
void get_dof_numbers_for_unknowns(std::list< std::pair< unsigned long, unsigned >> &dof_lookup_list) const
Create a list of pairs for all unknowns in this element, so that the first entry in each pair contain...
double p_nst(const unsigned &t, const unsigned &i) const
Return the i-th pressure value (Discontinous pressure interpolation – no need to cater for hanging no...
void output(FILE *file_pt)
Redirect output to NavierStokesEquations output.
unsigned ndof_types() const
The number of "DOF types" that degrees of freedom in this element are sub-divided into: Velocity and ...
void fix_pressure(const unsigned &p_dof, const double &p_value)
Pin p_dof-th pressure dof and set it to value specified by p_value.
double dshape_and_dtest_eulerian_nst(const Vector< double > &s, Shape &psi, DShape &dpsidx, Shape &test, DShape &dtestdx) const
Velocity shape and test functions and their derivs w.r.t. to global coords at local coordinate s (tak...
/////////////////////////////////////////////////////////////////////// /////////////////////////////...
Definition: Qelements.h:459
/////////////////////////////////////////////////////////////////////////
void output(FILE *file_pt, const unsigned &nplot)
Redirect output to NavierStokesEquations output.
int p_local_eqn(const unsigned &n) const
Return the local equation numbers for the pressure values.
unsigned ndof_types() const
Returns the number of "DOF types" that degrees of freedom in this element are sub-divided into: Veloc...
double dshape_and_dtest_eulerian_at_knot_nst(const unsigned &ipt, Shape &psi, DShape &dpsidx, Shape &test, DShape &dtestdx) const
Velocity shape and test functions and their derivs w.r.t. to global coords at local coordinate s (tak...
double dpshape_and_dptest_eulerian_nst(const Vector< double > &s, Shape &ppsi, DShape &dppsidx, Shape &ptest, DShape &dptestdx) const
Pressure shape and test functions and their derivs w.r.t. to global coords at local coordinate s (tak...
double dshape_and_dtest_eulerian_nst(const Vector< double > &s, Shape &psi, DShape &dpsidx, Shape &test, DShape &dtestdx) const
Velocity shape and test functions and their derivs w.r.t. to global coords at local coordinate s (tak...
double p_nst(const unsigned &n_p) const
Access function for the pressure values at local pressure node n_p (const version)
void identify_pressure_data(std::set< std::pair< Data *, unsigned >> &paired_pressure_data)
Add to the set paired_pressure_data pairs containing.
void output(std::ostream &outfile)
Redirect output to NavierStokesEquations output.
void identify_load_data(std::set< std::pair< Data *, unsigned >> &paired_load_data)
Add to the set paired_load_data pairs containing.
void output(FILE *file_pt)
Redirect output to NavierStokesEquations output.
void pshape_nst(const Vector< double > &s, Shape &psi) const
Pressure shape functions at local coordinate s.
virtual unsigned required_nvalue(const unsigned &n) const
Number of values (pinned or dofs) required at node n. Can be overwritten for hanging node version.
virtual int p_nodal_index_nst() const
Set the value at which the pressure is stored in the nodes.
static const unsigned Pconv[]
Static array of ints to hold conversion from pressure node numbers to actual node numbers.
double p_nst(const unsigned &t, const unsigned &n_p) const
Access function for the pressure values at local pressure node n_p (const version)
void output(std::ostream &outfile, const unsigned &nplot)
Redirect output to NavierStokesEquations output.
void get_dof_numbers_for_unknowns(std::list< std::pair< unsigned long, unsigned >> &dof_lookup_list) const
Create a list of pairs for all unknowns in this element, so that the first entry in each pair contain...
void build_fp_press_adv_diff_robin_bc_element(const unsigned &face_index)
Build FaceElements that apply the Robin boundary condition to the pressure advection diffusion proble...
unsigned npres_nst() const
Return number of pressure values.
QTaylorHoodElement()
Constructor, no internal data points.
void fix_pressure(const unsigned &p_dof, const double &p_value)
Pin p_dof-th pressure dof and set it to value specified by p_value.
static const unsigned Initial_Nvalue[]
Static array of ints to hold number of variables at node.
bool is_pressure_node(const unsigned &n) const
Deduce whether or not the provided node is a pressure node.
double dshape_and_dtest_eulerian_at_knot_nst(const unsigned &ipt, Shape &psi, DShape &dpsidx, RankFourTensor< double > &d_dpsidx_dX, Shape &test, DShape &dtestdx, RankFourTensor< double > &d_dtestdx_dX, DenseMatrix< double > &djacobian_dX) const
Shape/test functions and derivs w.r.t. to global coords at integration point ipt; return Jacobian of ...
////////////////////////////////////////////////////////////////// //////////////////////////////////...
Definition: matrices.h:1701
////////////////////////////////////////////////////////////////// //////////////////////////////////...
Definition: matrices.h:1370
RefineableElements are FiniteElements that may be subdivided into children to provide a better local ...
A Class for shape functions. In simple cases, the shape functions have only one index that can be tho...
Definition: shape.h:76
A Class for nodes that deform elastically (i.e. position is an unknown in the problem)....
Definition: nodes.h:1686
Data *const & variable_position_pt() const
Pointer to variable_position data (const version)
Definition: nodes.h:1765
//////////////////////////////////////////////////////////////////// ////////////////////////////////...
virtual void pin_all_non_pressure_dofs(std::map< Data *, std::vector< int >> &eqn_number_backup)=0
Pin all non-pressure dofs and backup eqn numbers of all Data.
virtual ~TemplateFreeNavierStokesEquationsBase()
Virtual destructor (empty)
virtual int p_local_eqn(const unsigned &n) const =0
Access function for the local equation number information for the pressure. p_local_eqn[n] = local eq...
virtual int p_nodal_index_nst() const =0
Return the index at which the pressure is stored if it is stored at the nodes. If not stored at the n...
virtual void fill_in_pressure_advection_diffusion_jacobian(Vector< double > &residuals, DenseMatrix< double > &jacobian)=0
Compute the residuals and Jacobian for the associated pressure advection diffusion problem....
virtual int & pinned_fp_pressure_eqn()=0
Global eqn number of pressure dof that's pinned in pressure adv diff problem.
virtual void build_fp_press_adv_diff_robin_bc_element(const unsigned &face_index)=0
Build FaceElements that apply the Robin boundary condition to the pressure advection diffusion proble...
virtual void delete_pressure_advection_diffusion_robin_elements()=0
Delete the FaceElements that apply the Robin boundary condition to the pressure advection diffusion p...
virtual void fill_in_pressure_advection_diffusion_residuals(Vector< double > &residuals)=0
Compute the residuals for the associated pressure advection diffusion problem. Used by the Fp precond...
virtual void get_pressure_and_velocity_mass_matrix_diagonal(Vector< double > &press_mass_diag, Vector< double > &veloc_mass_diag, const unsigned &which_one=0)=0
Compute the diagonal of the velocity/pressure mass matrices. If which one=0, both are computed,...
////////////////////////////////////////////////////////////////////// //////////////////////////////...
Definition: timesteppers.h:231
unsigned ntstorage() const
Return the number of doubles required to represent history (one for steady)
Definition: timesteppers.h:601
virtual double weight(const unsigned &i, const unsigned &j) const
Access function for j-th weight for the i-th derivative.
Definition: timesteppers.h:594
bool is_steady() const
Flag to indicate if a timestepper has been made steady (possibly temporarily to switch off time-depen...
Definition: timesteppers.h:389
std::string string(const unsigned &i)
Return the i-th string or "" if the relevant string hasn't been defined.
void dshape< 2 >(const double &s, double *DPsi)
Derivatives of 1D shape functions specialised to linear order (2 Nodes)
Definition: shape.h:616
void shape< 2 >(const double &s, double *Psi)
1D shape functions specialised to linear order (2 Nodes)
Definition: shape.h:608
std::string to_string(T object, unsigned float_precision=8)
Conversion function that should work for anything with operator<< defined (at least all basic types).
//////////////////////////////////////////////////////////////////// ////////////////////////////////...