[P]arallel [Hi]gh-order [Li]brary for [P]DEs  Latest
Parallel High-Order Library for PDEs through hp-adaptive Discontinuous Galerkin methods
parameters_reduced_order.h
1 #ifndef __PARAMETERS_REDUCED_ORDER_H__
2 #define __PARAMETERS_REDUCED_ORDER_H__
3 
4 #include <deal.II/base/parameter_handler.h>
5 
6 namespace PHiLiP {
7 namespace Parameters {
10 {
11 public:
12 
15  direct,
17  };
18 
21 
23  std::string path_to_search;
24 
27 
30 
33 
36 
38  std::vector<std::string> parameter_names;
39 
41  std::vector<double> parameter_min_values;
42 
44  std::vector<double> parameter_max_values;
45 
48 
51 
54 
57 
60 
62  static void declare_parameters (dealii::ParameterHandler &prm);
64  void parse_parameters (dealii::ParameterHandler &prm);
65 
66 };
67 
68 } // Parameters namespace
69 } // PHiLiP namespace
70 #endif
bool residual_error_bool
Use residual/reduced residual for error indicator instead of DWR. False by default.
std::vector< std::string > parameter_names
Names of parameters.
static void declare_parameters(dealii::ParameterHandler &prm)
Declares the possible variables and sets the defaults.
int number_modes
Number of modes to include in the POD Basis.
int num_halton
Number of Halton sequence points to add to initial snapshot set.
void parse_parameters(dealii::ParameterHandler &prm)
Parses input file and sets the variables.
std::vector< double > parameter_min_values
Minimum value of parameters.
Files for the baseline physics.
Definition: ADTypes.hpp:10
std::string path_to_search
Path to search for snapshots or saved POD basis.
int recomputation_coefficient
Recomputation parameter for adaptive sampling algorithm.
Parameters related to reduced-order model.
LinearSolverEnum
Types of linear solvers available.
std::string file_path_for_snapshot_locations
Path to search for file with pre-determined snapshot locations used to build POD (actual FOM snapshot...
std::vector< double > parameter_max_values
Maximum value of parameters.
double singular_value_threshold
Singular Value Threshold in the POD Basis.
LinearSolverEnum FOM_error_linear_solver_type
Type of linear solver used for first adjoint problem (DWR between FOM and ROM) (direct or gmres) ...
int output_snapshot_every_x_timesteps
Number of timesteps before putting solution in snapshot matrix.
double reduced_residual_tolerance
Tolerance of the reduced-order nonlinear residual.
double adaptation_tolerance
Tolerance for POD adaptation.