https://mooseframework.inl.gov
SymmetricRankFourTensor.h
Go to the documentation of this file.
1 //* This file is part of the MOOSE framework
2 //* https://mooseframework.inl.gov
3 //*
4 //* All rights reserved, see COPYRIGHT for full restrictions
5 //* https://github.com/idaholab/moose/blob/master/COPYRIGHT
6 //*
7 //* Licensed under LGPL 2.1, please see LICENSE for details
8 //* https://www.gnu.org/licenses/lgpl-2.1.html
9 
10 #pragma once
11 
12 #include "Moose.h"
13 #include "ADRankTwoTensorForward.h"
17 #include "MooseUtils.h"
18 #include "MathUtils.h"
19 
20 #include "libmesh/libmesh.h"
21 #include "libmesh/tuple_of.h"
22 #include "libmesh/int_range.h"
23 
24 #include "metaphysicl/raw_type.h"
25 
26 #include <petscsys.h>
27 
28 // Eigen includes
29 #include <Eigen/Core>
30 #include <Eigen/Dense>
31 
32 #include <array>
33 
34 using libMesh::Real;
35 using libMesh::tuple_of;
36 namespace libMesh
37 {
38 template <typename>
39 class TensorValue;
40 template <typename>
41 class TypeTensor;
42 template <typename>
43 class VectorValue;
44 }
45 
46 // Forward declarations
47 class MooseEnum;
48 
49 namespace MathUtils
50 {
55 template <>
57 template <>
59 }
60 
66 template <typename T>
68 {
69 public:
71  static constexpr unsigned int Ndim = LIBMESH_DIM;
72  static constexpr unsigned int N = Ndim + Ndim * (Ndim - 1) / 2;
73  static constexpr unsigned int N2 = N * N;
75 
76  // Full tensor indices in the Mandel/Voigt representation
77  static constexpr unsigned int full_index[6][6][4] = {
78  {{0, 0, 0, 0}, {0, 0, 1, 1}, {0, 0, 2, 2}, {0, 0, 1, 2}, {0, 0, 0, 2}, {0, 0, 0, 1}},
79  {{1, 1, 0, 0}, {1, 1, 1, 1}, {1, 1, 2, 2}, {1, 1, 1, 2}, {1, 1, 0, 2}, {1, 1, 0, 1}},
80  {{2, 2, 0, 0}, {2, 2, 1, 1}, {2, 2, 2, 2}, {2, 2, 1, 2}, {2, 2, 0, 2}, {2, 2, 0, 1}},
81  {{1, 2, 0, 0}, {1, 2, 1, 1}, {1, 2, 2, 2}, {1, 2, 1, 2}, {1, 2, 0, 2}, {1, 2, 0, 1}},
82  {{0, 2, 0, 0}, {0, 2, 1, 1}, {0, 2, 2, 2}, {0, 2, 1, 2}, {0, 2, 0, 2}, {0, 2, 0, 1}},
83  {{0, 1, 0, 0}, {0, 1, 1, 1}, {0, 1, 2, 2}, {0, 1, 1, 2}, {0, 1, 0, 2}, {0, 1, 0, 1}}};
84 
86  static constexpr Real mandelFactor(unsigned int i, unsigned int j)
87  {
90  }
91 
94  {
98  };
99 
106  {
107  symmetric9, // fillSymmetric9FromInputVector
108  symmetric21, // fillSymmetric21FromInputVector
109  symmetric_isotropic, // fillSymmetricIsotropicFromInputVector
110  symmetric_isotropic_E_nu, // fillSymmetricIsotropicEandNu
111  axisymmetric_rz, // fillAxisymmetricRZFromInputVector
112  principal, // fillPrincipalFromInputVector
113  orthotropic // fillGeneralOrthotropicFromInputVector
114  };
115 
116  template <template <typename> class Tensor, typename Scalar>
118  {
119  static const bool value = false;
120  };
121  template <typename Scalar>
123  {
125  };
126  template <typename Scalar>
127  struct TwoTensorMultTraits<TensorValue, Scalar>
128  {
130  };
131  template <typename Scalar>
132  struct TwoTensorMultTraits<TypeTensor, Scalar>
133  {
135  };
136 
139 
142 
144  SymmetricRankFourTensorTempl(const std::vector<T> &, FillMethod);
145 
148 
152  template <typename T2>
154 
157 
159  explicit operator RankFourTensorTempl<T>();
160 
161  // Named constructors
163  {
165  }
167  {
169  };
170 
172  inline T & operator()(unsigned int i, unsigned int j) { return _vals[i * N + j]; }
173 
178  inline const T & operator()(unsigned int i, unsigned int j) const { return _vals[i * N + j]; }
179 
181  void zero();
182 
184  void print(std::ostream & stm = Moose::out) const;
185 
187  void printReal(std::ostream & stm = Moose::out) const;
188 
189  friend std::ostream & operator<<(std::ostream & os, const SymmetricRankFourTensorTempl<T> & t)
190  {
191  t.print(os);
192  return os;
193  }
194 
197 
203  template <typename Scalar>
204  typename std::enable_if<libMesh::ScalarTraits<Scalar>::value,
206  operator=(const Scalar & libmesh_dbg_var(p))
207  {
208  libmesh_assert_equal_to(p, Scalar(0));
209  this->zero();
210  return *this;
211  }
212 
214  template <typename T2>
215  auto operator*(const SymmetricRankTwoTensorTempl<T2> & b) const
217 
219  template <typename T2>
220  auto operator*(const T2 & a) const ->
221  typename std::enable_if<libMesh::ScalarTraits<T2>::value,
223 
226 
228  template <typename T2>
229  auto operator/(const T2 & a) const ->
230  typename std::enable_if<libMesh::ScalarTraits<T2>::value,
231  SymmetricRankFourTensorTempl<decltype(T() / T2())>>::type;
232 
235 
238 
240  template <typename T2>
241  auto operator+(const SymmetricRankFourTensorTempl<T2> & a) const
243 
246 
248  template <typename T2>
249  auto operator-(const SymmetricRankFourTensorTempl<T2> & a) const
250  -> SymmetricRankFourTensorTempl<decltype(T() - T2())>;
251 
254 
256  template <typename T2>
257  auto operator*(const SymmetricRankFourTensorTempl<T2> & a) const
259 
261  T L2norm() const;
262 
268 
275  static SymmetricRankFourTensorTempl<T> rotationMatrix(const TypeTensor<T> & R);
276 
281  void rotate(const TypeTensor<T> & R);
282 
289 
294 
296  static MooseEnum fillMethodEnum();
297 
304  void fillFromInputVector(const std::vector<T> & input, FillMethod fill_method);
305 
307  void fillSymmetricIsotropic(const T & i0, const T & i1);
308  void fillSymmetricIsotropicEandNu(const T & E, const T & nu);
310 
321  template <typename T2>
322  void fillSymmetric9FromInputVector(const T2 & input);
323 
332  template <typename T2>
333  void fillSymmetric21FromInputVector(const T2 & input);
334 
336  T sum3x3() const;
337 
340 
342  bool isSymmetric() const;
343 
345  bool isIsotropic() const;
346 
347 protected:
349  std::array<T, N2> _vals;
350 
358  void fillSymmetricIsotropicFromInputVector(const std::vector<T> & input);
359 
369  void fillSymmetricIsotropicEandNuFromInputVector(const std::vector<T> & input);
370 
378  void fillAxisymmetricRZFromInputVector(const std::vector<T> & input);
379 
394  void fillPrincipalFromInputVector(const std::vector<T> & input);
395 
402  void fillGeneralOrthotropicFromInputVector(const std::vector<T> & input);
403 
404  template <class T2>
405  friend void dataStore(std::ostream &, SymmetricRankFourTensorTempl<T2> &, void *);
406 
407  template <class T2>
408  friend void dataLoad(std::istream &, SymmetricRankFourTensorTempl<T2> &, void *);
409 
410  template <typename T2>
412  template <typename T2>
414  template <typename T2>
415  friend class RankThreeTensorTempl;
416 };
417 
418 namespace MetaPhysicL
419 {
420 template <typename T>
422 {
424 
426  {
427  value_type ret;
428  for (unsigned int i = 0; i < SymmetricRankFourTensorTempl<T>::N; ++i)
429  for (unsigned int j = 0; j < SymmetricRankFourTensorTempl<T>::N; ++j)
430  ret(i, j) = raw_value(in(i, j));
431 
432  return ret;
433  }
434 };
435 }
436 
437 template <typename T1, typename T2>
438 inline auto
439 operator*(const T1 & a, const SymmetricRankFourTensorTempl<T2> & b) ->
440  typename std::enable_if<libMesh::ScalarTraits<T1>::value,
442 {
443  return b * a;
444 }
445 
446 template <typename T>
447 template <typename T2>
450 {
451  for (const auto i : make_range(N2))
452  _vals[i] = copy._vals[i];
453 }
454 
455 template <typename T>
456 template <typename T2>
457 auto
459  typename std::enable_if<libMesh::ScalarTraits<T2>::value,
461 {
462  typedef decltype(T() * T2()) ValueType;
464 
465  for (const auto i : make_range(N2))
466  result._vals[i] = _vals[i] * b;
467 
468  return result;
469 }
470 
471 template <typename T>
472 template <typename T2>
473 auto
476 {
477  typedef decltype(T() * T2()) ValueType;
479 
480  std::size_t index = 0;
481  for (const auto i : make_range(N))
482  {
483  ValueType tmp = 0.0;
484  for (const auto j : make_range(N))
485  tmp += _vals[index++] * b._vals[j];
486  result._vals[i] = tmp;
487  }
488 
489  return result;
490 }
491 
492 template <typename T>
493 template <typename T2>
494 auto
496  typename std::enable_if<libMesh::ScalarTraits<T2>::value,
497  SymmetricRankFourTensorTempl<decltype(T() / T2())>>::type
498 {
499  SymmetricRankFourTensorTempl<decltype(T() / T2())> result;
500  for (const auto i : make_range(N2))
501  result._vals[i] = _vals[i] / b;
502  return result;
503 }
504 
505 template <typename T>
506 template <typename T2>
507 void
509 {
510  mooseAssert(LIBMESH_DIM == 3, "This method assumes LIBMESH_DIM == 3");
511  mooseAssert(input.size() == 9,
512  "To use fillSymmetric9FromInputVector, your input must have size 9.");
513  zero();
514 
515  _vals[0] = input[0]; // C1111
516  _vals[7] = input[3]; // C2222
517  _vals[14] = input[5]; // C3333
518 
519  _vals[1] = _vals[6] = input[1]; // C1122
520  _vals[2] = _vals[12] = input[2]; // C1133
521  _vals[8] = _vals[13] = input[4]; // C2233
522 
523  static constexpr std::size_t C2323 = 21;
524  static constexpr std::size_t C1313 = 28;
525  static constexpr std::size_t C1212 = 35;
526 
527  _vals[C2323] = 2.0 * input[6];
528  _vals[C1313] = 2.0 * input[7];
529  _vals[C1212] = 2.0 * input[8];
530 }
531 
532 template <typename T>
533 template <typename T2>
534 void
536 {
537  // C1111 C1122 C1133 C1123 C1113 C1112
538  // C2222 C2233 C2223 C2213 C2212
539  // C3333 C3323 C3313 C3312
540  // C2323 C2313 C2312
541  // C1313 C1312
542  // C1212
543 
544  mooseAssert(LIBMESH_DIM == 3, "This method assumes LIBMESH_DIM == 3");
545  mooseAssert(input.size() == 21,
546  "To use fillSymmetric21FromInputVector, your input must have size 21.");
547  std::size_t index = 0;
548  for (const auto i : make_range(N))
549  for (const auto j : make_range(i, N))
550  {
551  _vals[i + N * j] = mandelFactor(i, j) * input[index];
552  _vals[j + N * i] = mandelFactor(j, i) * input[index];
553  index++;
554  }
555 }
556 
557 template <typename T>
560 {
561  SymmetricRankFourTensorTempl<T> result(initNone);
562 
563  if constexpr (SymmetricRankFourTensorTempl<T>::N2 * sizeof(T) > EIGEN_STACK_ALLOCATION_LIMIT)
564  {
565  Eigen::Matrix<T, Eigen::Dynamic, Eigen::Dynamic> mat(N, N);
566  for (const auto i : make_range(N))
567  for (const auto j : make_range(N))
568  mat(i, j) = (*this)(i, j);
569  mat = mat.inverse();
570  for (const auto i : make_range(N))
571  for (const auto j : make_range(N))
572  result(i, j) = mat(i, j);
573  }
574  else
575  {
576  const Eigen::Map<const Eigen::Matrix<T, N, N, Eigen::RowMajor>> mat(&_vals[0]);
577  Eigen::Map<Eigen::Matrix<T, N, N, Eigen::RowMajor>> res(&result._vals[0]);
578  res = mat.inverse();
579  }
580 
581  return result;
582 }
void fillAxisymmetricRZFromInputVector(const std::vector< T > &input)
fillAxisymmetricRZFromInputVector takes 5 inputs to fill the axisymmetric Rank-4 tensor with the appr...
RankFourTensorTempl is designed to handle any N-dimensional fourth order tensor, C.
std::array< T, N2 > _vals
The values of the rank-four tensor.
libMesh::VectorValue< T > sum3x1() const
Calculates the vector a[i] = sum over j Ciijj for i and j varying from 0 to 2.
void fillSymmetricIsotropicEandNu(const T &E, const T &nu)
const T & operator()(unsigned int i, unsigned int j) const
Gets the value for the indices specified.
static MooseEnum fillMethodEnum()
Static method for use in validParams for getting the "fill_method".
T & operator()(unsigned int i, unsigned int j)
Gets the value for the indices specified. Takes indices ranging from 0-5 for i and j...
bool isIsotropic() const
checks if the tensor is isotropic
auto operator+(const SymmetricRankFourTensorTempl< T2 > &a) const -> SymmetricRankFourTensorTempl< decltype(T()+T2())>
C_ijkl + a_ijkl.
FillMethod
To fill up the 36 entries in the 4th-order tensor, fillFromInputVector is called with one of the foll...
void fillSymmetricIsotropic(const T &i0, const T &i1)
Vector-less fill API functions. See docs of the corresponding ...FromInputVector methods.
SymmetricRankFourTensorTempl< T > & operator-=(const SymmetricRankFourTensorTempl< T > &a)
C_ijkl -= a_ijkl.
void mooseSetToZero< SymmetricRankFourTensor >(SymmetricRankFourTensor &v)
Helper function template specialization to set an object to zero.
auto operator*(const SymmetricRankTwoTensorTempl< T2 > &b) const -> SymmetricRankTwoTensorTempl< decltype(T() *T2())>
C_ijkl*a_kl.
auto raw_value(const Eigen::Map< T > &in)
Definition: EigenADReal.h:100
void fillSymmetric21FromInputVector(const T2 &input)
fillSymmetric21FromInputVector takes 21 inputs to fill in the Rank-4 tensor with the appropriate crys...
static constexpr Real mandelFactor(unsigned int i, unsigned int j)
returns the 1, sqrt(2), or 2 prefactor in the Mandel notation for the indices i,j ranging from 0-5...
We need to instantiate the following CompareTypes to tell the compiler that ADReal is a subtype of Ch...
The following methods are specializations for using the libMesh::Parallel::packed_range_* routines fo...
std::basic_ostream< charT, traits > * os
Definition: InfixIterator.h:34
const Number zero
static constexpr unsigned int N
T sum3x3() const
Calculates the sum of Ciijj for i and j varying from 0 to 2.
static SymmetricRankFourTensorTempl< T > rotationMatrix(const TypeTensor< T > &R)
Build a 6x6 rotation matrix MEHRABADI, MORTEZA M.
RankThreeTensor is designed to handle any N-dimensional third order tensor, r.
static value_type value(const SymmetricRankFourTensorTempl< T > &in)
auto operator/(const T2 &a) const -> typename std::enable_if< libMesh::ScalarTraits< T2 >::value, SymmetricRankFourTensorTempl< decltype(T()/T2())>>::type
C_ijkl/a.
typename tuple_n< Index, T >::template type<> tuple_of
SymmetricRankFourTensorTempl< T > & operator+=(const SymmetricRankFourTensorTempl< T > &a)
C_ijkl += a_ijkl for all i, j, k, l.
SymmetricRankFourTensorTempl< typename RawType< T >::value_type > value_type
SymmetricRankTwoTensorTempl is designed to handle the Stress or Strain Tensor for an anisotropic mate...
void fillSymmetricIsotropicEandNuFromInputVector(const std::vector< T > &input)
fillSymmetricIsotropicEandNuFromInputVector is a variation of the fillSymmetricIsotropicFromInputVect...
This is a "smart" enum class intended to replace many of the shortcomings in the C++ enum type It sho...
Definition: MooseEnum.h:54
void rotate(const TypeTensor< T > &R)
Rotate the tensor using C_ijkl = R_im R_jn R_ko R_lp C_mnop.
SymmetricRankFourTensorTempl< T > & operator*=(const T &a)
C_ijkl *= a.
static SymmetricRankFourTensorTempl< T > identitySymmetricFour()
void fillPrincipalFromInputVector(const std::vector< T > &input)
fillPrincipalFromInputVector takes 9 inputs to fill a Rank-4 tensor C1111 = input0 C1122 = input1 C11...
void printReal(std::ostream &stm=Moose::out) const
Print the values of the rank four tensor.
void fillSymmetricIsotropicFromInputVector(const std::vector< T > &input)
fillSymmetricIsotropicFromInputVector takes 2 inputs to fill the the symmetric Rank-4 tensor with the...
friend void dataStore(std::ostream &, SymmetricRankFourTensorTempl< T2 > &, void *)
void mooseSetToZero< ADSymmetricRankFourTensor >(ADSymmetricRankFourTensor &v)
SymmetricRankFourTensorTempl< T > transposeMajor() const
Transpose the tensor by swapping the first pair with the second pair of indices This amounts to a reg...
DIE A HORRIBLE DEATH HERE typedef LIBMESH_DEFAULT_SCALAR_TYPE Real
SymmetricRankFourTensorTempl< T > & operator=(const SymmetricRankFourTensorTempl< T > &a)=default
copies values from a into this tensor
void print(std::ostream &stm=Moose::out) const
Print the rank four tensor.
void fillFromInputVector(const std::vector< T > &input, FillMethod fill_method)
fillFromInputVector takes some number of inputs to fill the Rank-4 tensor.
NumberTensorValue Tensor
SymmetricRankFourTensorTempl< T > operator-() const
-C_ijkl
friend void dataLoad(std::istream &, SymmetricRankFourTensorTempl< T2 > &, void *)
void fillGeneralOrthotropicFromInputVector(const std::vector< T > &input)
fillGeneralOrhotropicFromInputVector takes 10 inputs to fill the Rank-4 tensor It defines a general o...
std::enable_if< libMesh::ScalarTraits< Scalar >::value, SymmetricRankFourTensorTempl & >::type operator=(const Scalar &libmesh_dbg_var(p))
Assignment-from-scalar operator.
static constexpr unsigned int full_index[6][6][4]
SymmetricRankFourTensorTempl is designed to handle an N-dimensional fourth order tensor with minor sy...
IntRange< T > make_range(T beg, T end)
SymmetricRankFourTensorTempl< T > invSymm() const
This returns A_ijkl such that C_ijkl*A_klmn = 0.5*(de_im de_jn + de_in de_jm) This routine assumes th...
static SymmetricRankFourTensorTempl< T > identity()
bool isSymmetric() const
checks if the tensor is symmetric
static constexpr Real mandelFactor(unsigned int i)
returns the 1 or sqrt(2) prefactor in the Mandel notation for the index i ranging from 0-5...
static constexpr unsigned int Ndim
tensor dimension, Mandel matrix dimension, and Mandel matrix size
SymmetricRankFourTensorTempl< T > & operator/=(const T &a)
C_ijkl /= a for all i, j, k, l.
SymmetricRankFourTensorTempl()
Default constructor; fills to zero.
void fillSymmetric9FromInputVector(const T2 &input)
fillSymmetric9FromInputVector takes 9 inputs to fill in the Rank-4 tensor with the appropriate crysta...
auto operator*(const T1 &a, const SymmetricRankFourTensorTempl< T2 > &b) -> typename std::enable_if< libMesh::ScalarTraits< T1 >::value, SymmetricRankFourTensorTempl< decltype(T1() *T2())>>::type
SymmetricRankFourTensorTempl< T > transposeIj() const
Transpose the tensor by swapping the first two indices - a no-op.
static constexpr unsigned int N2