Page Menu
Home
c4science
Search
Configure Global Search
Log In
Files
F77668935
ElementQuad4.hpp
No One
Temporary
Actions
Download File
Edit File
Delete File
View Transforms
Subscribe
Mute Notifications
Award Token
Subscribers
None
File Metadata
Details
File Info
Storage
Attached
Created
Thu, Aug 15, 19:12
Size
19 KB
Mime Type
text/x-c
Expires
Sat, Aug 17, 19:12 (2 d)
Engine
blob
Format
Raw Data
Handle
19906039
Attached To
rGOOSEFEM GooseFEM
ElementQuad4.hpp
View Options
/* =================================================================================================
(c - GPLv3) T.W.J. de Geus (Tom) | tom@geus.me | www.geus.me | github.com/tdegeus/GooseFEM
================================================================================================= */
#ifndef GOOSEFEM_ELEMENTQUAD4_CPP
#define GOOSEFEM_ELEMENTQUAD4_CPP
// -------------------------------------------------------------------------------------------------
#include "ElementQuad4.h"
// =================================== GooseFEM::Element::Quad4 ====================================
namespace GooseFEM {
namespace Element {
namespace Quad4 {
// ================================ GooseFEM::Element::Quad4::Gauss ================================
namespace Gauss {
// --------------------------------- number of integration points ----------------------------------
inline size_t nip()
{
return 4;
}
// ----------------------- integration point coordinates (local coordinates) -----------------------
inline ArrD xi()
{
size_t nip = 4;
size_t ndim = 2;
ArrD xi({nip,ndim});
xi(0,0) = -1./std::sqrt(3.); xi(0,1) = -1./std::sqrt(3.);
xi(1,0) = +1./std::sqrt(3.); xi(1,1) = -1./std::sqrt(3.);
xi(2,0) = +1./std::sqrt(3.); xi(2,1) = +1./std::sqrt(3.);
xi(3,0) = -1./std::sqrt(3.); xi(3,1) = +1./std::sqrt(3.);
return xi;
}
// ----------------------------------- integration point weights -----------------------------------
inline ArrD w()
{
size_t nip = 4;
ArrD w({nip});
w(0) = 1.;
w(1) = 1.;
w(2) = 1.;
w(3) = 1.;
return w;
}
// -------------------------------------------------------------------------------------------------
}
// ================================ GooseFEM::Element::Quad4::Nodal ================================
namespace Nodal {
// --------------------------------- number of integration points ----------------------------------
inline size_t nip()
{
return 4;
}
// ----------------------- integration point coordinates (local coordinates) -----------------------
inline ArrD xi()
{
size_t nip = 4;
size_t ndim = 2;
ArrD xi({nip,ndim});
xi(0,0) = -1.; xi(0,1) = -1.;
xi(1,0) = +1.; xi(1,1) = -1.;
xi(2,0) = +1.; xi(2,1) = +1.;
xi(3,0) = -1.; xi(3,1) = +1.;
return xi;
}
// ----------------------------------- integration point weights -----------------------------------
inline ArrD w()
{
size_t nip = 4;
ArrD w({nip});
w(0) = 1.;
w(1) = 1.;
w(2) = 1.;
w(3) = 1.;
return w;
}
// -------------------------------------------------------------------------------------------------
}
// =================================================================================================
// ------------------------------------------ constructor ------------------------------------------
inline Quadrature::Quadrature(const ArrD &x, const ArrD &xi, const ArrD &w)
: m_x(x), m_w(w), m_xi(xi)
{
// check input
assert( m_x.rank() == 3 ); // shape: [nelem, nne, ndim]
assert( m_x.shape(1) == m_nne ); // number of nodes per element
assert( m_x.shape(2) == m_ndim ); // number of dimensions
// extract number of elements
m_nelem = m_x.shape(0);
// integration scheme
// - default
if ( m_w.size() == 0 and m_xi.size() == 0 )
{
m_nip = Gauss::nip();
m_xi = Gauss::xi();
m_w = Gauss::w();
}
// - input
else if ( m_w.size() > 0 and m_xi.size() > 0 )
{
m_nip = m_w.size();
}
// - unknown
else
{
throw std::runtime_error("Input integration point coordinates and weights");
}
// check input
assert( m_xi.rank() == 2 ); // shape: [nip, ndim]
assert( m_xi.shape(0) == m_nip ); // number of integration points
assert( m_xi.shape(1) == m_ndim ); // number of dimensions
assert( m_w .rank() == 1 ); // shape: [nip]
assert( m_w .size() == m_nip ); // number of integration points
// allocate arrays
// - shape functions
m_N.resize({m_nip,m_nne});
// - shape function gradients in local coordinates
m_dNxi.resize({m_nip,m_nne,m_ndim});
// - shape function gradients in global coordinates
m_dNx.resize({m_nelem,m_nip,m_nne,m_ndim});
// - integration point volume
m_vol.resize({m_nelem,m_nip});
// shape functions
for ( size_t k = 0 ; k < m_nip ; ++k )
{
m_N(k,0) = .25 * (1.-m_xi(k,0)) * (1.-m_xi(k,1));
m_N(k,1) = .25 * (1.+m_xi(k,0)) * (1.-m_xi(k,1));
m_N(k,2) = .25 * (1.+m_xi(k,0)) * (1.+m_xi(k,1));
m_N(k,3) = .25 * (1.-m_xi(k,0)) * (1.+m_xi(k,1));
}
// shape function gradients in local coordinates
for ( size_t k = 0 ; k < m_nip ; ++k )
{
// - dN / dxi_0
m_dNxi(k,0,0) = -.25*(1.-m_xi(k,1));
m_dNxi(k,1,0) = +.25*(1.-m_xi(k,1));
m_dNxi(k,2,0) = +.25*(1.+m_xi(k,1));
m_dNxi(k,3,0) = -.25*(1.+m_xi(k,1));
// - dN / dxi_1
m_dNxi(k,0,1) = -.25*(1.-m_xi(k,0));
m_dNxi(k,1,1) = -.25*(1.+m_xi(k,0));
m_dNxi(k,2,1) = +.25*(1.+m_xi(k,0));
m_dNxi(k,3,1) = +.25*(1.-m_xi(k,0));
}
// compute the shape function gradients, based on "x"
compute_dN();
}
// --------------------------- integration volume (per tensor-component) ---------------------------
inline ArrD Quadrature::dV(size_t ncomp) const
{
if ( ncomp == 0 ) return m_vol;
ArrD out = ArrD::Zero({m_nelem, m_nip, ncomp});
#pragma omp parallel for
for ( size_t e = 0 ; e < m_nelem ; ++e )
for ( size_t k = 0 ; k < m_nip ; ++k )
for ( size_t i = 0 ; i < ncomp ; ++i )
out(e,k,i) = m_vol(e,k);
return out;
}
// -------------------------------------- number of elements ---------------------------------------
inline size_t Quadrature::nelem() const
{
return m_nelem;
}
// ---------------------------------- number of nodes per element ----------------------------------
inline size_t Quadrature::nne() const
{
return m_nne;
}
// ------------------------------------- number of dimensions --------------------------------------
inline size_t Quadrature::ndim() const
{
return m_ndim;
}
// --------------------------------- number of integration points ----------------------------------
inline size_t Quadrature::nip() const
{
return m_nip;
}
// --------------------------------------- update positions ----------------------------------------
inline void Quadrature::update_x(const ArrD &x)
{
// check input
assert( x.rank() == 3 ); // shape: [nelem, nne, ndim]
assert( x.shape(0) == m_nelem ); // number of elements
assert( x.shape(1) == m_nne ); // number of nodes per element
assert( x.shape(2) == m_ndim ); // number of dimensions
assert( x.size() == m_x.size() ); // total number of components (redundant)
// update positions
m_x.setCopy(x.begin(), x.end());
// update the shape function gradients for the new "x"
compute_dN();
}
// ------------------------ shape function gradients in global coordinates -------------------------
inline void Quadrature::compute_dN()
{
#pragma omp parallel
{
// intermediate quantities and local views
double Jdet;
cppmat::tiny::matrix<double,m_nne,m_ndim> dNx;
cppmat::view::matrix<double,m_nne,m_ndim> dNxi, x;
cppmat::tiny::cartesian::tensor2<double,2> J, Jinv;
// loop over all elements (in parallel)
#pragma omp for
for ( size_t e = 0 ; e < m_nelem ; ++e )
{
// alias nodal positions
x.setMap(&m_x(e));
// loop over integration points
for ( size_t k = 0 ; k < m_nip ; ++k )
{
// - alias shape function gradients (local coordinates)
dNxi.setMap(&m_dNxi(k));
// - Jacobian (loops unrolled for efficiency)
// J(i,j) += dNxi(m,i) * x(m,j);
J(0,0) = dNxi(0,0)*x(0,0) + dNxi(1,0)*x(1,0) + dNxi(2,0)*x(2,0) + dNxi(3,0)*x(3,0);
J(0,1) = dNxi(0,0)*x(0,1) + dNxi(1,0)*x(1,1) + dNxi(2,0)*x(2,1) + dNxi(3,0)*x(3,1);
J(1,0) = dNxi(0,1)*x(0,0) + dNxi(1,1)*x(1,0) + dNxi(2,1)*x(2,0) + dNxi(3,1)*x(3,0);
J(1,1) = dNxi(0,1)*x(0,1) + dNxi(1,1)*x(1,1) + dNxi(2,1)*x(2,1) + dNxi(3,1)*x(3,1);
// - determinant and inverse of the Jacobian
Jdet = J.det();
Jinv = J.inv();
// - shape function gradients wrt global coordinates (loops partly unrolled for efficiency)
// dNx(m,i) += Jinv(i,j) * dNxi(m,j);
for ( size_t m = 0 ; m < m_nne ; ++m )
{
dNx(m,0) = Jinv(0,0) * dNxi(m,0) + Jinv(0,1) * dNxi(m,1);
dNx(m,1) = Jinv(1,0) * dNxi(m,0) + Jinv(1,1) * dNxi(m,1);
}
// - copy to matrix: shape function gradients (global coordinates)
dNx.copyTo(m_dNx.item(e,k));
// - copy to matrix: integration point volume
m_vol(e,k) = m_w(k) * Jdet;
}
}
} // #pragma omp parallel
}
// ------------------- dyadic product "qtensor(i,j) = dNdx(m,i) * elemvec(m,j)" --------------------
template<class T>
inline ArrD Quadrature::gradN_vector(const ArrD &elemvec) const
{
// check input
assert( elemvec.rank() == 3 ); // shape: [nelem, nne, ndim]
assert( elemvec.shape(0) == m_nelem ); // number of elements
assert( elemvec.shape(1) == m_nne ); // number of nodes per element
assert( elemvec.shape(2) == m_ndim ); // number of dimensions
// zero-initialize output: matrix of tensors
ArrD qtensor = ArrD::Zero({m_nelem, m_nip, T::Size()});
#pragma omp parallel
{
// intermediate quantities and local views
T gradu;
cppmat::view::matrix<double,m_nne,m_ndim> dNx, u;
// loop over all elements (in parallel)
#pragma omp for
for ( size_t e = 0 ; e < m_nelem ; ++e )
{
// alias element vector (e.g. nodal displacements)
u.setMap(&elemvec(e));
// loop over all integration points in element "e"
for ( size_t k = 0 ; k < m_nip ; ++k )
{
// - alias shape function gradients (local coordinates)
dNx.setMap(&m_dNx(e,k));
// - evaluate dyadic product (loops unrolled for efficiency)
// gradu(i,j) += dNx(m,i) * u(m,j)
gradu(0,0) = dNx(0,0)*u(0,0) + dNx(1,0)*u(1,0) + dNx(2,0)*u(2,0) + dNx(3,0)*u(3,0);
gradu(0,1) = dNx(0,0)*u(0,1) + dNx(1,0)*u(1,1) + dNx(2,0)*u(2,1) + dNx(3,0)*u(3,1);
gradu(1,0) = dNx(0,1)*u(0,0) + dNx(1,1)*u(1,0) + dNx(2,1)*u(2,0) + dNx(3,1)*u(3,0);
gradu(1,1) = dNx(0,1)*u(0,1) + dNx(1,1)*u(1,1) + dNx(2,1)*u(2,1) + dNx(3,1)*u(3,1);
// - copy resulting integration point tensor
std::copy(gradu.begin(), gradu.end(), qtensor.item(e,k));
}
}
} // #pragma omp parallel
return qtensor;
}
// ---------------------------------- transpose of "GradN_vector" ----------------------------------
template<class T>
inline ArrD Quadrature::gradN_vector_T(const ArrD &elemvec) const
{
// check input
assert( elemvec.rank() == 3 ); // shape: [nelem, nne, ndim]
assert( elemvec.shape(0) == m_nelem ); // number of elements
assert( elemvec.shape(1) == m_nne ); // number of nodes per element
assert( elemvec.shape(2) == m_ndim ); // number of dimensions
// zero-initialize output: matrix of tensors
ArrD qtensor = ArrD::Zero({m_nelem, m_nip, T::Size()});
#pragma omp parallel
{
// intermediate quantities and local views
T gradu;
cppmat::view::matrix<double,m_nne,m_ndim> dNx, u;
// loop over all elements (in parallel)
#pragma omp for
for ( size_t e = 0 ; e < m_nelem ; ++e )
{
// alias element vector (e.g. nodal displacements)
u.setMap(&elemvec(e));
// loop over all integration points in element "e"
for ( size_t k = 0 ; k < m_nip ; ++k )
{
// - alias shape function gradients (global coordinates)
dNx.setMap(&m_dNx(e,k));
// - evaluate dyadic product (loops unrolled for efficiency)
// gradu(j,i) += dNx(m,i) * u(m,j)
gradu(0,0) = dNx(0,0)*u(0,0) + dNx(1,0)*u(1,0) + dNx(2,0)*u(2,0) + dNx(3,0)*u(3,0);
gradu(1,0) = dNx(0,0)*u(0,1) + dNx(1,0)*u(1,1) + dNx(2,0)*u(2,1) + dNx(3,0)*u(3,1);
gradu(0,1) = dNx(0,1)*u(0,0) + dNx(1,1)*u(1,0) + dNx(2,1)*u(2,0) + dNx(3,1)*u(3,0);
gradu(1,1) = dNx(0,1)*u(0,1) + dNx(1,1)*u(1,1) + dNx(2,1)*u(2,1) + dNx(3,1)*u(3,1);
// - copy resulting integration point tensor
std::copy(gradu.begin(), gradu.end(), qtensor.item(e,k));
}
}
} // #pragma omp parallel
return qtensor;
}
// ------------------------------- symmetric part of "GradN_vector" --------------------------------
template<class T>
inline ArrD Quadrature::symGradN_vector(const ArrD &elemvec) const
{
// check input
assert( elemvec.rank() == 3 ); // shape: [nelem, nne, ndim]
assert( elemvec.shape(0) == m_nelem ); // number of elements
assert( elemvec.shape(1) == m_nne ); // number of nodes per element
assert( elemvec.shape(2) == m_ndim ); // number of dimensions
// zero-initialize output: matrix of tensors
ArrD qtensor = ArrD::Zero({m_nelem, m_nip, T::Size()});
#pragma omp parallel
{
// intermediate quantities and local views
T eps;
cppmat::tiny::cartesian::tensor2<double,m_ndim> gradu;
cppmat::view::matrix<double,m_nne,m_ndim> dNx, u;
// loop over all elements (in parallel)
#pragma omp for
for ( size_t e = 0 ; e < m_nelem ; ++e )
{
// alias element vector (e.g. nodal displacements)
u.setMap(&elemvec(e));
// loop over all integration points in element "e"
for ( size_t k = 0 ; k < m_nip ; ++k )
{
// - alias shape function gradients (global coordinates)
dNx.setMap(&m_dNx(e,k));
// - evaluate dyadic product (loops unrolled for efficiency)
// gradu(i,j) += dNx(m,i) * u(m,j)
gradu(0,0) = dNx(0,0)*u(0,0) + dNx(1,0)*u(1,0) + dNx(2,0)*u(2,0) + dNx(3,0)*u(3,0);
gradu(0,1) = dNx(0,0)*u(0,1) + dNx(1,0)*u(1,1) + dNx(2,0)*u(2,1) + dNx(3,0)*u(3,1);
gradu(1,0) = dNx(0,1)*u(0,0) + dNx(1,1)*u(1,0) + dNx(2,1)*u(2,0) + dNx(3,1)*u(3,0);
gradu(1,1) = dNx(0,1)*u(0,1) + dNx(1,1)*u(1,1) + dNx(2,1)*u(2,1) + dNx(3,1)*u(3,1);
// - symmetrize (loops unrolled for efficiency)
// eps(i,j) = .5 * ( gradu(i,j) + gradu(j,i) )
eps(0,0) = gradu(0,0);
eps(0,1) = .5 * ( gradu(0,1) + gradu(1,0) );
eps(1,0) = eps (0,1);
eps(1,1) = gradu(1,1);
// - copy resulting integration point tensor
std::copy(eps.begin(), eps.end(), qtensor.item(e,k));
}
}
} // #pragma omp parallel
return qtensor;
}
// ------- scalar product "elemmat(m*ndim+i,n*ndim+i) = N(m) * qscalar * N(n)"; for all "i" --------
inline ArrD Quadrature::int_N_scalar_NT_dV(const ArrD &qscalar) const
{
// check input
assert( qscalar.rank() == 2 ); // shape: [nelem, nip]
assert( qscalar.shape(0) == m_nelem ); // number of elements
assert( qscalar.shape(1) == m_nip ); // number of integration points
// zero-initialize: matrix of matrices
ArrD elemmat = ArrD::Zero({m_nelem, m_nne*m_ndim, m_nne*m_ndim});
#pragma omp parallel
{
// intermediate quantities and local views
cppmat::tiny::matrix<double,m_nne*m_ndim,m_nne*m_ndim> M;
cppmat::view::vector<double,m_nne> N;
double rho, vol;
// loop over all elements (in parallel)
#pragma omp for
for ( size_t e = 0 ; e < m_nelem ; ++e )
{
// zero-initialize (e.g. mass matrix)
M.setZero();
// loop over all integration points in element "e"
for ( size_t k = 0 ; k < m_nip ; ++k )
{
// - alias shape functions
N.setMap(&m_N(k));
// - alias
vol = m_vol (e,k); // integration point volume
rho = qscalar(e,k); // integration point scalar (e.g. density)
// - evaluate scalar product, for all dimensions, and assemble
// M(m*ndim+i,n*ndim+i) += N(m) * scalar * N(n) * dV
for ( size_t m = 0 ; m < m_nne ; ++m ) {
for ( size_t n = 0 ; n < m_nne ; ++n ) {
M(m*m_ndim+0, n*m_ndim+0) += N(m) * rho * N(n) * vol;
M(m*m_ndim+1, n*m_ndim+1) += N(m) * rho * N(n) * vol;
}
}
}
// copy result to element matrix
std::copy(M.begin(), M.end(), elemmat.item(e));
}
} // #pragma omp parallel
return elemmat;
}
// ------------ integral of dot product "elemvec(m,j) += dNdx(m,i) * qtensor(i,j) * dV" ------------
template<class T>
inline ArrD Quadrature::int_gradN_dot_tensor2_dV(const ArrD &qtensor) const
{
// check input
assert( qtensor.rank() == 3 ); // shape: [nelem, nip, #tensor-components]
assert( qtensor.shape(0) == m_nelem ); // number of elements
assert( qtensor.shape(1) == m_nip ); // number of integration points
assert( qtensor.shape(2) == T::Size() ); // tensor dimensions
// zero-initialize output: matrix of vectors
ArrD elemvec = ArrD::Zero({m_nelem, m_nne, m_ndim});
#pragma omp parallel
{
// intermediate quantities and local views
cppmat::view::matrix<double,m_nne,m_ndim> dNx;
cppmat::tiny::matrix<double,m_nne,m_ndim> f;
double vol;
T sig;
// loop over all elements (in parallel)
#pragma omp for
for ( size_t e = 0 ; e < m_nelem ; ++e )
{
// zero-initialize (e.g. nodal force)
f.setZero();
// loop over all integration points in element "e"
for ( size_t k = 0 ; k < m_nip ; ++k )
{
// - alias
dNx.setMap (&m_dNx (e,k)); // shape function gradients (global coordinates)
sig.setCopy(&qtensor(e,k)); // integration point tensor (e.g. stress)
vol = m_vol (e,k); // integration point volume
// - evaluate dot product, and assemble (loops partly unrolled for efficiency)
// f(m,j) += dNx(m,i) * sig(i,j) * vol;
for ( size_t m = 0 ; m < m_nne ; ++m )
{
f(m,0) += ( dNx(m,0) * sig(0,0) + dNx(m,1) * sig(1,0) ) * vol;
f(m,1) += ( dNx(m,0) * sig(0,1) + dNx(m,1) * sig(1,1) ) * vol;
}
}
// copy result to element vector
std::copy(f.begin(), f.end(), elemvec.item(e));
}
} // #pragma omp parallel
return elemvec;
}
// ---------------------- wrappers with default storage (no template needed) -----------------------
inline ArrD Quadrature::gradN_vector(const ArrD &elemvec) const
{
return gradN_vector<cppmat::tiny::cartesian::tensor2<double,2>>(elemvec);
}
// -------------------------------------------------------------------------------------------------
inline ArrD Quadrature::gradN_vector_T(const ArrD &elemvec) const
{
return gradN_vector_T<cppmat::tiny::cartesian::tensor2<double,2>>(elemvec);
}
// -------------------------------------------------------------------------------------------------
inline ArrD Quadrature::symGradN_vector(const ArrD &elemvec) const
{
return symGradN_vector<cppmat::tiny::cartesian::tensor2s<double,2>>(elemvec);
}
// -------------------------------------------------------------------------------------------------
inline ArrD Quadrature::int_gradN_dot_tensor2_dV(const ArrD &qtensor) const
{
assert( qtensor.rank() == 3 ); // shape: [nelem, nip, #tensor-components]
if ( qtensor.shape(2) == m_ndim*m_ndim )
return int_gradN_dot_tensor2_dV<cppmat::tiny::cartesian::tensor2<double,2>>(qtensor);
else if ( qtensor.shape(2) == (m_ndim+1)*m_ndim/2 )
return int_gradN_dot_tensor2_dV<cppmat::tiny::cartesian::tensor2s<double,2>>(qtensor);
else
throw std::runtime_error("assert: qtensor.shape(2) == 4 or qtensor.shape(2) == 3");
}
// -------------------------------------------------------------------------------------------------
inline ArrD Quadrature::int_gradN_dot_tensor2s_dV(const ArrD &qtensor) const
{
return int_gradN_dot_tensor2_dV<cppmat::tiny::cartesian::tensor2s<double,2>>(qtensor);
}
// -------------------------------------------------------------------------------------------------
}}} // namespace ...
// =================================================================================================
#endif
Event Timeline
Log In to Comment