estimator::Prediction< F, dim > Class Template Reference
Prediction hp - adaptive refinement strategy, proposed by Melenk and Wohlmuth. More...
#include <refinement.hh>
Public Types | |
enum | Mode { H = 0, P = 1 } |
Public Member Functions | |
virtual void | apply (concepts::SpaceOnCells< F > &space) |
virtual void | apply (concepts::SpaceOnCells< Real > &space) |
void | buildRefinement (const Marking &mark, const concepts::LocalEstimator< F > &estimator) |
Prediction (const concepts::SpaceOnCells< F > &space, enum Mode mode=P, const Real hdec=sqrt(4.0), const Real pdec=sqrt(0.4), const Real nRdec=1.0) | |
Constructor for the prediction refinement strategy. More... | |
virtual | ~Prediction () |
Protected Member Functions | |
virtual std::ostream & | info (std::ostream &os) const |
Returns information in an output stream. More... | |
Protected Attributes | |
concepts::HashMap< concepts::AdaptiveAdjustP< dim > > | adjusts_ |
Private Attributes | |
concepts::HashMap< const concepts::Cell * > | cells_ |
Real | hdecay_ |
concepts::Sequence< const concepts::Cell * > | hRefCurCells_ |
Real | noRefdecay_ |
concepts::HashMap< concepts::Array< ushort > > | p_ |
Real | pdecay_ |
concepts::HashMap< Real > | pred_ |
Detailed Description
template<class F = Real, ushort dim = 2>
class estimator::Prediction< F, dim >
Prediction hp - adaptive refinement strategy, proposed by Melenk and Wohlmuth.
Currently just isotropic refinement, i.e.
- p refinement : increase in both direction
- h refinement : bisection in each direction
Definition at line 99 of file refinement.hh.
Member Enumeration Documentation
◆ Mode
enum estimator::Prediction::Mode |
Enumerator | |
---|---|
H | |
P |
Definition at line 107 of file refinement.hh.
Constructor & Destructor Documentation
◆ Prediction()
|
inline |
Constructor for the prediction refinement strategy.
- Parameters
-
space mark estimator hdec value for error decay supposed for h refinement pdec value for error decay supposed for p-refinement pval
nRdec decay value for error change if no refinement is applied
default values may set by the ones given in the paper of Melenk/Wolmuth, i.e.
Definition at line 122 of file refinement.hh.
◆ ~Prediction()
|
inlinevirtual |
Definition at line 183 of file refinement.hh.
Member Function Documentation
◆ apply() [1/2]
|
inlinevirtual |
Definition at line 237 of file refinement.hh.
◆ apply() [2/2]
|
inlinevirtualinherited |
Implements estimator::Refinement< Real >.
Definition at line 64 of file refinement.hh.
◆ buildRefinement()
|
inline |
Definition at line 187 of file refinement.hh.
◆ info()
|
inlineprotectedvirtual |
Returns information in an output stream.
Implements estimator::Refinement< Real >.
Definition at line 282 of file refinement.hh.
Member Data Documentation
◆ adjusts_
|
protectedinherited |
Definition at line 87 of file refinement.hh.
◆ cells_
|
private |
Definition at line 306 of file refinement.hh.
◆ hdecay_
|
private |
Definition at line 293 of file refinement.hh.
◆ hRefCurCells_
|
private |
Definition at line 311 of file refinement.hh.
◆ noRefdecay_
|
private |
Definition at line 297 of file refinement.hh.
◆ p_
|
private |
Definition at line 303 of file refinement.hh.
◆ pdecay_
|
private |
Definition at line 295 of file refinement.hh.
◆ pred_
|
private |
Definition at line 300 of file refinement.hh.
The documentation for this class was generated from the following file:
- estimator/refinement.hh