Skip to content

Improve parameter evaluation in Neumann-type boundary condition

Tom Fischer requested to merge TomFischer/ogs:ImproveNeumannParameterEvaluation into master
  1. Feature description was added to the changelog
  2. Tests covering your feature were added?
    • reference files for existing tests are updated

For every changed benchmark the analytical solution (as described in the prj-file) was generated. Then vtkdiff was used for comparisons of computed solution with analytical solution. Here are some results:

Benchmark 2D_HeatConduction_wedge

Improved Neumann Evaluation

fischeth@modmon107 ~/w/o/build/release % vtkdiff -a temperature -b analytical_solution /tmp/wedge_ang_0.02_ts_10_t_1.000000_with_analytical_solution.vtu 
Comparing data array `temperature' from file `/tmp/wedge_ang_0.02_ts_10_t_1.000000_with_analytical_solution.vtu' to data array `analytical_solution' from file `'.
Computed difference between data arrays:
abs l1 norm      = [1.277802600374363e+00]
abs l2-norm^2    = [1.147061848691392e-02]
abs l2-norm      = [1.071009733238402e-01]
abs maximum norm = [1.649954514900666e-02]

rel l1 norm      = [inf]
rel l2-norm^2    = [inf]
rel l2-norm      = [inf]
rel maximum norm = [inf]
Absolute and relative error (maximum norm) are larger than the corresponding thresholds 2.220446049250313e-16 and 2.220446049250313e-16.

Current Master

fischeth@modmon107 ~/w/o/build/release % vtkdiff -a temperature -b analytical_solution /tmp/wedge_ang_0.02_ts_10_t_1.000000_master_with_analytical_solution.vtu
Comparing data array `temperature' from file `/tmp/wedge_ang_0.02_ts_10_t_1.000000_master_with_analytical_solution.vtu' to data array `analytical_solution' from file `'.
Computed difference between data arrays:
abs l1 norm      = [1.753154520925763e+00]
abs l2-norm^2    = [2.302234454939754e-02]
abs l2-norm      = [1.517311587954087e-01]
abs maximum norm = [2.111845465255369e-02]

rel l1 norm      = [inf]
rel l2-norm^2    = [inf]
rel l2-norm      = [inf]
rel maximum norm = [inf]
Absolute and relative error (maximum norm) are larger than the corresponding thresholds 2.220446049250313e-16 and 2.220446049250313e-16.

Benchmark SteadyStateDiffusion_square_1x1_1e2_axi

Improved Neumann Evaluation

fischeth@modmon107 ~/w/o/build/release % vtkdiff -a temperature -b analytical_solution /tmp/square_1e2_axi_ts_1_t_1.000000_with_analytical_solution.vtu 
Comparing data array `temperature' from file `/tmp/square_1e2_axi_ts_1_t_1.000000_with_analytical_solution.vtu' to data array `analytical_solution' from file `'.
Computed difference between data arrays:
abs l1 norm      = [9.211672948918542e-01]
abs l2-norm^2    = [1.181201970720864e-02]
abs l2-norm      = [1.086831160172022e-01]
abs maximum norm = [2.078181092561411e-02]

rel l1 norm      = [inf]
rel l2-norm^2    = [inf]
rel l2-norm      = [inf]
rel maximum norm = [inf]
Absolute and relative error (maximum norm) are larger than the corresponding thresholds 2.220446049250313e-16 and 2.220446049250313e-16.

Current Master

fischeth@modmon107 ~/w/o/build/release % vtkdiff -a temperature -b analytical_solution /tmp/square_1e2_axi_ts_1_t_1.000000_master_with_analytical_solution.vtu
Comparing data array `temperature' from file `/tmp/square_1e2_axi_ts_1_t_1.000000_master_with_analytical_solution.vtu' to data array `analytical_solution' from file `'.
Computed difference between data arrays:
abs l1 norm      = [1.253254043295376e+00]
abs l2-norm^2    = [2.282820517187984e-02]
abs l2-norm      = [1.510900564957199e-01]
abs maximum norm = [2.677375737428922e-02]

rel l1 norm      = [inf]
rel l2-norm^2    = [inf]
rel l2-norm      = [inf]
rel maximum norm = [inf]
Absolute and relative error (maximum norm) are larger than the corresponding thresholds 2.220446049250313e-16 and 2.220446049250313e-16.

Benchmark SteadyStateDiffusion_square_1x1_1e4_axi_ang_0.02

Improved Neumann Evaluation

fischeth@modmon107 ~/w/o/build/release % vtkdiff -a temperature -b analytical_solution /tmp/square_1e4_axi_ts_1_t_1.000000_with_analytical_solution.vtu
Comparing data array `temperature' from file `/tmp/square_1e4_axi_ts_1_t_1.000000_with_analytical_solution.vtu' to data array `analytical_solution' from file `'.
Computed difference between data arrays:
abs l1 norm      = [7.778561233344679e-01]
abs l2-norm^2    = [9.754565697524973e-05]
abs l2-norm      = [9.876520489284155e-03]
abs maximum norm = [2.450820612146307e-04]

rel l1 norm      = [inf]
rel l2-norm^2    = [inf]
rel l2-norm      = [inf]
rel maximum norm = [inf]
Absolute and relative error (maximum norm) are larger than the corresponding thresholds 2.220446049250313e-16 and 2.220446049250313e-16.

Current Master

fischeth@modmon107 ~/w/o/build/release % vtkdiff -a temperature -b analytical_solution /tmp/square_1e4_axi_ts_1_t_1.000000_master_with_analytical_solution.vtu
Comparing data array `temperature' from file `/tmp/square_1e4_axi_ts_1_t_1.000000_master_with_analytical_solution.vtu' to data array `analytical_solution' from file `'.
Computed difference between data arrays:
abs l1 norm      = [1.050565421970479e+00]
abs l2-norm^2    = [1.861052118723456e-04]
abs l2-norm      = [1.364203840605742e-02]
abs maximum norm = [3.048876172027878e-04]

rel l1 norm      = [inf]
rel l2-norm^2    = [inf]
rel l2-norm      = [inf]
rel maximum norm = [inf]
Absolute and relative error (maximum norm) are larger than the corresponding thresholds 2.220446049250313e-16 and 2.220446049250313e-16.

Benchmark SteadyStateDiffusion_wedge_1e2_ang_0.02

Improved Neumann Evaluation

fischeth@modmon107 ~/w/o/build/release % vtkdiff -a temperature -b analytical_solution /tmp/wedge_ang_0.02_ts_10_t_1.000000_with_analytical_solution.vtu                                  
Comparing data array `temperature' from file `/tmp/wedge_ang_0.02_ts_10_t_1.000000_with_analytical_solution.vtu' to data array `analytical_solution' from file `'.
Computed difference between data arrays:
abs l1 norm      = [1.277802600374363e+00]
abs l2-norm^2    = [1.147061848691392e-02]
abs l2-norm      = [1.071009733238402e-01]
abs maximum norm = [1.649954514900666e-02]

rel l1 norm      = [inf]
rel l2-norm^2    = [inf]
rel l2-norm      = [inf]
rel maximum norm = [inf]
Absolute and relative error (maximum norm) are larger than the corresponding thresholds 2.220446049250313e-16 and 2.220446049250313e-16.

Current Master

fischeth@modmon107 ~/w/o/build/release % vtkdiff -a temperature -b analytical_solution /tmp/wedge_ang_0.02_ts_10_t_1.000000_master_with_analytical_solution.vtu
Comparing data array `temperature' from file `/tmp/wedge_ang_0.02_ts_10_t_1.000000_master_with_analytical_solution.vtu' to data array `analytical_solution' from file `'.
Computed difference between data arrays:
abs l1 norm      = [1.753154520925763e+00]
abs l2-norm^2    = [2.302234454939754e-02]
abs l2-norm      = [1.517311587954087e-01]
abs maximum norm = [2.111845465255369e-02]

rel l1 norm      = [inf]
rel l2-norm^2    = [inf]
rel l2-norm      = [inf]
rel maximum norm = [inf]
Absolute and relative error (maximum norm) are larger than the corresponding thresholds 2.220446049250313e-16 and 2.220446049250313e-16.

Benchmark SteadyStateDiffusion_wedge_1e4_ang_0.02

Improved Neumann Evaluation

fischeth@modmon107 ~/w/o/build/release % vtkdiff -a temperature -b analytical_solution /tmp/wedge_1e4_axi_ang_0.02_ts_1_t_1.000000_with_analytical_solution.vtu 
Comparing data array `temperature' from file `/tmp/wedge_1e4_axi_ang_0.02_ts_1_t_1.000000_with_analytical_solution.vtu' to data array `analytical_solution' from file `'.
Computed difference between data arrays:
abs l1 norm      = [1.532213755200179e+00]
abs l2-norm^2    = [1.701154950494929e-04]
abs l2-norm      = [1.304283309137600e-02]
abs maximum norm = [2.203274695519564e-04]

rel l1 norm      = [inf]
rel l2-norm^2    = [inf]
rel l2-norm      = [inf]
rel maximum norm = [inf]
Absolute and relative error (maximum norm) are larger than the corresponding thresholds 2.220446049250313e-16 and 2.220446049250313e-16.

Current Master

fischeth@modmon107 ~/w/o/build/release % vtkdiff -a temperature -b analytical_solution /tmp/wedge_1e4_axi_ang_0.02_ts_1_t_1.000000_master_with_analytical_solution.vtu
Comparing data array `temperature' from file `/tmp/wedge_1e4_axi_ang_0.02_ts_1_t_1.000000_master_with_analytical_solution.vtu' to data array `analytical_solution' from file `'.
Computed difference between data arrays:
abs l1 norm      = [2.991233405130929e+00]
abs l2-norm^2    = [7.410801202299059e-04]
abs l2-norm      = [2.722278678294906e-02]
abs maximum norm = [3.650811207507661e-04]

rel l1 norm      = [inf]
rel l2-norm^2    = [inf]
rel l2-norm      = [inf]
rel maximum norm = [inf]
Absolute and relative error (maximum norm) are larger than the corresponding thresholds 2.220446049250313e-16 and 2.220446049250313e-16.
Edited by Tom Fischer

Merge request reports