Vectorial Field: Online
Contents
Vectorial Field: Online#
To be used only after VectorialPOD_Offline: this solver implements a POD projection to vector fields.
Preparation#
The structure of the case study folder is the following (in this example Folder3 and Folder4 are the test case folders)
>> ./Study_case
>> /Folder_1
>> /Folder_2
>> /Folder_3
>> /Folder_4
>> /POD_(fieldName)
>> /0
>> /system
controlDict
blockMeshDict
...
PODsolverDict <--- Dictionary needed for the input parameters
>> /constant
>> /(fieldName)_POD_Offline_files
The PODsolverDict must be put inside ./Study_case/POD_(fieldName)/system/
An example of PODsolverDict can be found in application/POD/VectorialPOD_Online, which requires the following entries:
Online_parameters
{
field U; <---- ScalarField on which POD is performed
BasisNumber 20; <---- number of POD modes to use
foldersList (
"Folder_3"
"Folder_4" ) ; <---- List of folder names containig the snapshots to be reconstructed
}
Usage#
Inside ./Study_case/POD_(fieldName) launch
VectorialPOD_Online
To include folder “0” use
VectorialPOD_Online -withZero
To perform on a specified region (for multi-region cases) use
VectorialPOD_Online -region <regionName>
Results#
The residual field is defined as the absolute difference between the test snapshot and the reconstruction and it is stored in the snapshot folder, as well.
>> ./Study_case
>> /Folder_1
>> /Folder_2
>> /Folder_3
>> /0
UPODreconstruct <--- U POD reconstruction obtained with BasisNumber basis
UPODresidual <--- U POD reconstruction obtained with BasisNumber basis
>> /1
UPODreconstruct
UPODresidual
>> ...
>> /Folder_4
>> /0
UPODreconstruct <---(fieldName) POD reconstruction obtained with BasisNumber basis
UPODresidual <---(fieldName) POD reconstruction obtained with BasisNumber basis
>> /1
UPODreconstruct
UPODresidual
>> ...
>> /POD_U
>> /0
>> /system
>> /constant
>> /U_POD_Offline_files
>> /U_POD_Online_files
maximum_L2_relative_error.txt <---- max L2 absolute error as a function of basis number
average_L2_relative_error.txt <---- max L2 relative error as a function of basis number
The absolute and relative error are computed as
\[E_N^{L^2} = || \mathbf{u}-\mathbf{u}_{N}^{POD}||_{L^2}\qquad
\epsilon_N^{L^2} = \frac{|| \mathbf{u}-\mathbf{u}_{N}^{POD}||_{L^2}}{|| \mathbf{u} ||_{L^2}}\]