Preprocessing module

Temporary documentation for the content of IGAparametrization class.

geometricSettings corresponds to the content of .NB file

mechanicalSettings corresponds to the content of .inp file

mechanicalSettings is a list with the following components :

[0] : parameters ???

[1] : boundary conditions

[2] : load

[3] : nodes

[4] : IEN connectivity

[5] : material properties

[6] : properties ???

[7] : tables, used for nodal distributions

[8] : shapeparametrization ????

geometricSettings is a list with the following components :

[0] : dimension

[1] : sizes of knot vectors

[2] : knot vectors

[3] : degrees

[4] : knot spans

[5] : weights

[6] : number of elements per patch

[7] : number of nodes ???

[8] : number of elements (per patch ???)

Members of IGAparametrization class

Global model informations

_ELT_TYPE : an array containing the element types for each patch (size : number of patchs)

_NBPINT : an array containing the number of integration points per element for each patch (size : number of patchs)

_TENSOR : an array containing the tensor type for each patch patch (size : number of patchs)

_mcrd : a scalar the number of coordinates of the model

Control points informations

_COORDS : a 2D array containing the coordinates of the control points (shape : 3, number of control points)

_nb_cp : a scalar containing the total number of control points

Elements informations

_IEN : a list containing the connectivity table for each patch (length : number of patchs). Each element of the list _IEN is a 2D array of shape (number of element of the patch, number of control points per element)

Material informations

_MATERIAL_PROPERTIES : a 2D array containing the material properties for each patch (shape : 3, number of patchs). The first dimension is 3 for elastic materials (Young modulus, Poisson ratio, density) but can be set higher for specific material behaviours (gradient elasticity, plasticity, …)

_N_MATERIAL_PROPERTIES : an array containing the numebr of material properties for each patch (size : number of patchs)

Properties informations

_PROPS : a list containing arrays with the properties with real values for each patch (length : number of patchs). Each element of the _PROPS list can have a different size

_JPROPS : an array containing the number of properties for each patch (size : number of patchs)

Geometry informations

_dim : an array containing the dimension of each patch (size : number of patchs)

_Nkv : a 2D array containing the number of knots for each patch (shape : 3, number of patchs)

_Ukv : a list containing the knot vectors for each patch (length : number of patchs). Each element of the list _Ukv is a list of size 3 containing an array representing with the knot values of the patch in each parametric direction

_Jpqr : a 2D array containing the degrees of each patch in each direction (shape : 3, number of patchs)

_Nijk : a 2D array containing the knot span for each element in each direction (shape : 3, total number of elements)

_weight : a list containing the weights for each element (length : total number of elements). Each element of the list is an array whose size is number of control points of the element

_elementsByPatch : an array containing the number of elements for each patch (size : number of patchs)

_nb_patch : a scalar containg the number of patchs

_nnode : an array containing the number of control points per element for each patch (size : number of patchs)

_nb_elem : a scalar containing the total number of elements

Other informations

_indCPbyPatch : an array containing the indices of control points for each patch (size : number of patchs). Each element of the array is an array containing the indices of the control points for a given patch (size : number of control points of the corresponding patch)

Class members (TEST)

class yeti_iga.preprocessing.igaparametrization.IGAparametrization(mechanicalSettings=None, geometricSettings=None, filename=None)

A class containing the necessary parameters for IGA.

_autoset_indCPbyPatch(): Automatic computation of control point indices by patch.

_compute_vectWeight(): Build ordered NURBS weight vector.

_flatten_data()

Flatten all geometric or mechanical attributes arrays.

This method is used for passing arguments to the FORTRAN part of the code, which is taking only flattened arrays.

_get_bcs_info(): Get attributes related to boundary conditions.

_get_indCPbyPatch(): Get control points indices for each patch.

_get_load_info(): Get attributes related to loads.

_get_vectWeight(): Get NURBS weight vector.

_initRefinementMatHistory(): Initialise _refinementMatHistory to track refinement steps.

_set_indCPbyPatch(indCPbyPatch): Set control points indices for each patch.

_updateNodalField(fieldinit)

Update a nodal field according to the previous refinement steps.

Parameters:: fieldinit (array of floats) – The initial field which discretization level corresponds to the initial geometry.
Returns:: field – The same field but discretized as the current geometry.
Return type:: array of floats

_updateRefinementMatHistory(transformationMatrices)

Update _refinementMatHistory according to a refinement step.

Parameters:: transformationMatrices (list of sparse matrices) – The matrices that transform the control points from a coarser to a finer mesh. Those matrices are considered to be as given by the function refine_and_getTransformationMatrices.

_update_dof_info(): Set degrees of freedom information.

add_multiplePatch(patchMechSet, patchGeoSet)

Parameters:

patchMechSet (TYPE) – DESCRIPTION.
patchGeoSet (TYPE) – DESCRIPTION.

Return type:

None.

add_patch(patchMechSet, patchGeoSet, mode=0)

Add a patch to current IGAparametrization object.

Parameters:

patchMechSet (TYPE) – DESCRIPTION.
patchGeoSet (TYPE) – DESCRIPTION.
mode (TYPE, optional) – DESCRIPTION. The default is 0.

Return type:

None.

property coords: Coordinates of control points

property design_parameters: TODO Document this property

property dim

document this property

Type:: TODO

property elt_type

document this property

Type:: TODO

generate_vtk4controlMeshVisu(filename, num_patch, sol=None, output_path='results'): Generate control mesh visualisation for a given patch number.

get_geometricSettings(): Get the geometrical settings.

get_geometricSettings_somePatch(listpatch): Get the geometrical settings of a given patch.

get_inputs4adjointdisp(xi, numpatch=1)

Parameters:

xi (TYPE) – DESCRIPTION.
numpatch (TYPE, optional) – DESCRIPTION. The default is 1.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4area(COORDS=None, activepatch=None)

Parameters:

COORDS (TYPE, optional) – DESCRIPTION. The default is None.
activepatch (TYPE, optional) – DESCRIPTION. The default is None.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4bandInfos(): SHORT DESCRIPTION.

get_inputs4controlMesh(filename, num_patch, output_path='results', sol=None)

Parameters:

filename (TYPE) – DESCRIPTION.
num_patch (TYPE) – DESCRIPTION.
sol (TYPE, optional) – DESCRIPTION. The default is None.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4cplgEmbddEntites(nb_diag=None, NUMDof2Diag=None)

Parameters:

nb_diag (TYPE, optional) – DESCRIPTION. The default is None.
NUMDof2Diag (TYPE, optional) – DESCRIPTION. The default is None.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4cplginfos(nb_pts=3): SHORT DESCRIPTION.

get_inputs4cplgmatrix(COORDS=None): SHORT DESCRIPTION.

get_inputs4cplgmatrixU5(COORDS=None, integrationOrder=0, output_path='results')

Generate input arguments for coupling.cplgmatrix.cplg_matrixu5.

Parameters:

COORDS (list, optional) – Control points coordinates. The default is None.
integrationOrder (int, optional) – Integration order for coupling equation (used only if greater than auto-evaluated one). The default is 0.
output_path (string) – Path to directory where debug files (for projection) should be written.

Returns:

inputs – List of parameters for coupling.cplgmatrix.cplg_matrixu5.

Return type:

list

get_inputs4dirichletmatrix(): SHORT DESCRIPTION.

get_inputs4dlmmat(): SHORT DESCRIPTION.

get_inputs4entities2param(): SHORT DESCRIPTION.

get_inputs4eval_coupling_gpts(filename, npts_u, npts_v)

Return the necessary data to evaluate coupling Gauss points coordinates

Parameters:

filename (str) – base name for resulting files
npts_u (int) – number of points to compute along u parametric direction
npts_v (int) – number of points to compute along v parametric direction

Returns:

the necessary input parameters for Fortran subroutine eval_coupling_gpts

Return type:

dict

get_inputs4evaldisp(sol, xi, numpatch=1)

Parameters:

sol (TYPE) – DESCRIPTION.
xi (TYPE) – DESCRIPTION.
numpatch (TYPE, optional) – DESCRIPTION. The default is 1.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4evaldispmulti(sol, n_xi, n_eta, n_zeta, numpatch=1)

Parameters:

sol (TYPE) – DESCRIPTION.
n_xi (TYPE) – DESCRIPTION.
n_eta (TYPE) – DESCRIPTION.
n_zeta (TYPE) – DESCRIPTION.
numpatch (TYPE, optional) – DESCRIPTION. The default is 1.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4evalstress(sol, xi, numpatch=1)

Parameters:

sol (TYPE) – DESCRIPTION.
xi (TYPE) – DESCRIPTION.
numpatch (TYPE, optional) – DESCRIPTION. The default is 1.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4gauss3D(KNumFace=0): SHORT DESCRIPTION.

get_inputs4geomat(sol, case=1)

Parameters:

sol (TYPE) – DESCRIPTION.
case (TYPE, optional) – DESCRIPTION. The default is 1.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4gradCompliance(sol, epsilon=0.0001, COORDS=None, activeElem=None, activeDir=array([1, 1, 1]))

Parameters:

sol (TYPE) – DESCRIPTION.
epsilon (TYPE, optional) – DESCRIPTION. The default is 1.e-4.
COORDS (TYPE, optional) – DESCRIPTION. The default is None.
activeElem (TYPE, optional) – DESCRIPTION. The default is None.
activeDir (TYPE, optional) – DESCRIPTION. The default is np.ones(3, np.intp).

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4gradCoupling(sol, COORDS=None, activepatch=None)

Parameters:

sol (TYPE) – DESCRIPTION.
COORDS (TYPE, optional) – DESCRIPTION. The default is None.
activepatch (TYPE, optional) – DESCRIPTION. The default is None.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4gradDisplacement(sol, ADJ, COORDS=None, activeElem=None, activeDir=array([1, 1, 1]))

Parameters:

sol (TYPE) – DESCRIPTION.
ADJ (TYPE) – DESCRIPTION.
COORDS (TYPE, optional) – DESCRIPTION. The default is None.
activeElem (TYPE, optional) – DESCRIPTION. The default is None.
activeDir (TYPE, optional) – DESCRIPTION. The default is np.ones(3, np.intp).

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4gradTotalwork(sol, ADJ, COORDS=None, activeElem=None, activeDir=array([1, 1, 1]))

Parameters:

sol (TYPE) – DESCRIPTION.
ADJ (TYPE) – DESCRIPTION.
COORDS (TYPE, optional) – DESCRIPTION. The default is None.
activeElem (TYPE, optional) – DESCRIPTION. The default is None.
activeDir (TYPE, optional) – DESCRIPTION. The default is np.ones(3, np.intp).

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4gradVibration(eigenvect, eigenval, COORDS=None, activeElem=None, activeDir=array([1, 1, 1]))

Parameters:

eigenvect (TYPE) – DESCRIPTION.
eigenval (TYPE) – DESCRIPTION.
COORDS (TYPE, optional) – DESCRIPTION. The default is None.
activeElem (TYPE, optional) – DESCRIPTION. The default is None.
activeDir (TYPE, optional) – DESCRIPTION. The default is np.ones(3, np.intp).

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4grammat(): Generate input for the routine computing Gram matrix

get_inputs4grevphyscoords(activePatch=None): SHORT DESCRIPTION.

get_inputs4indDOF(): Get the settings concerning fixed and free degrees of freedom.

get_inputs4interpolation(activePatch=None, listParam=None)

Parameters:

activePatch (TYPE, optional) – DESCRIPTION. The default is None.
listParam (TYPE, optional) – DESCRIPTION. The default is None.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4lagrangeCplgMatrx(Mult, IGJELEM, GaussCoordsGlo)

Parameters:

Mult (TYPE) – DESCRIPTION.
IGJELEM (TYPE) – DESCRIPTION.
GaussCoordsGlo (TYPE) – DESCRIPTION.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4massmat(COORDS=None, activeElem=None)

Parameters:

COORDS (TYPE, optional) – DESCRIPTION. The default is None.
activeElem (TYPE, optional) – DESCRIPTION. The default is None.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4post_curve_2D(i_patch, i_face, n_sample, sol)

Get the necessary inputs for function postproc.XXXXXXXXXXXXX in order to compute mechanical quantities (displacement and displacement gradient) along a curve for a 2D problem

Parameters:

i_patch (int) – Index of patch to process (starts at 1)
i_face (int) – Index of side curve to process (1, 2, 3 or 4)
n_sample (int) – Number of sample points to generate
sol (numpy array) – Problem solution

Returns:

inputs – Necessary inputs for function postproc.postproc_curve_2D

Return type:

dict

get_inputs4postprocBasisFcts(filename, activepatch=None, dervorder=0, nb_pts=5)

Parameters:

filename (TYPE) – DESCRIPTION.
activepatch (TYPE, optional) – DESCRIPTION. The default is None.
dervorder (TYPE, optional) – DESCRIPTION. The default is 0.
nb_pts (TYPE, optional) – DESCRIPTION. The default is 5.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4postprocCPLG(filename, sol, nb_ref=5, Flag=array([True, False, False]))

Parameters:

filename (TYPE) – DESCRIPTION.
sol (TYPE) – DESCRIPTION.
nb_ref (TYPE, optional) – DESCRIPTION. The default is 5.
Flag (TYPE, optional) – DESCRIPTION. The default is np.array([True, False, False]).

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4postprocPtsCloud(filename, activepatch=None, nb_pts=5)

Parameters:

filename (TYPE) – DESCRIPTION.
activepatch (TYPE, optional) – DESCRIPTION. The default is None.
nb_pts (TYPE, optional) – DESCRIPTION. The default is 5.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4postprocVTU(filename, sol, nb_ref=array([1, 1, 1]), Flag=array([True, True, True]), output_path='results')

Get the necessary inputs for .vtu file generation.

Parameters:

filename (str) – Name of the output .vtu file.
sol (numpy array) – Displacement field to plot. The function get_inputs4solution can be used for this purpose.
nb_ref (numpy array, optional) – Mesh refinement for visualisation. It takes the form: np.array([xi-dir., eta-dir., zeta-dir.]. The default is np.ones(3).
Flag (numpy array, optional) – Field ouput to save. Three are available: Nodal displacement, stresses, and Von Mises stresses. The default is np.array([True, True, True]).
output_path (string) – Path to the directory where the output file should be written

Returns:

inputs – Necessary inputs for generating the .vtu file.

Return type:

list

get_inputs4postproc_bezier(i_patch, filename, sol, output_path='results')

Get the necessary inputs for VTU postprocessing using Bezier cells with function postprocessing.postproc.generate_vtu_bezier Ouptut is made for a given patch of type U1

Parameters:

i_patch (int) – Index of patch to process (starts at 1)
filename (str) – Name of output .vtu file
sol (numpy array) – Displacement field to plot

Returns:

inputs – Necessary inputs for VTU postprocessing using Bezier cells with function postprocessing.postproc.generate_vtu_bezier

Return type:

dict

get_inputs4postproc_cplg_vtu(filename, lgrge_patch_number, sol, nb_ref=array([1., 1.]), output_path='results')

Returns the inputs for the generation of VTU file for results at coupling interface

Parameters:

filename (string) – Name of the file to write without extension
lgrge_patch_number (int) – Index of Lagrange patch number of the interface to process
sol (numpy.array) – Displacement field solution
nb_ref (numpy.array) – mesh refinement level in each direction. default is [1, 1]

Returns:

inputs – List of input parameters for pp.generate_coupling_vtu

Return type:

list

get_inputs4postproc_faces_vtu(filename, nb_ref=array([1., 1., 1.]), output_path='results')

Get the necessary inputs for function postproc.generate_faces_vtu

Parameters:

filename (str) – Name of the output .vtu file
nb_ref (numpy array, optional) – Mesh refinement for the visualisation. It takes the form: np.array([xi-dir., eta-dir., zeta-dir.]). The default is np.ones(3)
output_path (string) – Path to the directory where the output file should be written. The default is ‘results’

Returns:

inputs – Necessary inputs for function postproc.generate_faces_vtu

Return type:

dict

get_inputs4proj_vtu(filename, sol, svars, nb_ref=array([1., 1., 1.]), output_path='results')

Return the inputs for least squer projection in a VTU file.

Parameters:

filename (string) – Name of the file to write without extension
sol (numpy.array) – Displacement field solution
svars (numpy.array) – variables originally defined at integration points and mean-square projected on CP
nb_ref (numpy.array) – Mesh refinement level in each direction. Default is [1, 1, 1]
outout_path (string) – path to output directory, default is ‘results’

Returns:

inputs – list of input parameters for pp.generate_pro_vtu

Return type:

list

get_inputs4regularizationaggreg(ptseval, pnorm=10)

Parameters:

ptseval (TYPE) – DESCRIPTION.
pnorm (TYPE, optional) – DESCRIPTION. The default is 10.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4solution(U_inv): Return the necessary data for reconstructing nodal displacement.

get_inputs4strainExtrmm(sol): SHORT DESCRIPTION.

get_inputs4stressaggreg(sol, pnorm=10)

Parameters:

sol (TYPE) – DESCRIPTION.
pnorm (TYPE, optional) – DESCRIPTION. The default is 10.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4stressaggregDiscrete(sol, ptseval, pnorm=10)

Parameters:

sol (TYPE) – DESCRIPTION.
ptseval (TYPE) – DESCRIPTION.
pnorm (TYPE, optional) – DESCRIPTION. The default is 10.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4svarsrhs(sol)

Generate input for the routine computing RHS vectors of the least square projection of values at Gauss points

Parameters:: sol (np.array) – displacement solution at control points

get_inputs4system(COORDS=None): SHORT DESCRIPTION.

get_inputs4system_bandStorage(nb_diag=None, NUMDof2Diag=None, COORDS=None, activeElem=None)

Parameters:

nb_diag (TYPE, optional) – DESCRIPTION. The default is None.
NUMDof2Diag (TYPE, optional) – DESCRIPTION. The default is None.
COORDS (TYPE, optional) – DESCRIPTION. The default is None.
activeElem (TYPE, optional) – DESCRIPTION. The default is None.

Returns:

inputs – DESCRIPTION.

Return type:

TYPE

get_inputs4system_elemStorage(COORDS=None, activeElem=None)

Return parameters to compute the sparse stiffness matrix.

Data is structured to comply with the function input parameters. The functions is stiffmtrx_elemstorage.sys_linmat_lindef_static

Parameters:

COORDS (numpy array, optional) – Control points coordinates. The default is None. Default correponds to the coordinates of the current geometry.
activeElem (numpy array (dtype=intp), optional) – Array indicating elements considered to build the stiffness matrix. Value 1 stands for active elements. The default is None (all elements are activated).

Returns:

inputs – All necessary inputs to build stiffness matrix.

Return type:

list

get_mechanicalSettings(): Get the object’s mechanical settings.

get_mechanicalSettings_somePatch(listpatch, with_bc=False, updatePROPS=False)

Get the mechanical settings for a given patch.

Parameters:

listpatch (list of ints) – List of indices corresponding to the patche(s) of interest.
with_bc (boolean, optional) – Adds boundary conditions to the returned list if set to True. The default is False.
updatePROPS (boolean, optional) – Updates the object’s properties if set to True. The default is False.

Returns:

mechSet – List of ordered mechanical settings.

Return type:

list

property ien

document this property

Type:: TODO

property ind_cp_by_patch

document this property

Type:: TODO

property ind_dof_free: Indices of unblocked degrees of freedom

property j_pqr: Degrees for all patchs

property n_kv: knot vectors size for all patchs

property nb_cp

document this property

Type:: TODO

property nb_dof_free: Number of unblocked degrees of freedom

property nb_dof_tot: Total number of degrees of freedom

property nb_patch: ” Total number of patchs in the model

property num_integration_points

document this property

Type:: TODO

refine(nb_refinementByDirection, nb_degreeElevationByDirection=array([0, 0, 0]), additional_knots=None)

Refinement by degree elevation and knot insertion.

For more details on the function’s arguments see the function refine_and_getTransformationMatrices. The present function does not return the transformation matrices for refinement.

refine_and_getTransformationMatrices(nb_refinementByDirection, nb_degreeElevationByDirection=[0, 0, 0], additional_knots=None)

Refinement by degree elevation and knot insertion.

For knot insertion, a level of refinement consists in inserting all of the average knots of non-zero intervals. The number of elements is hence doubled.

Parameters:

nb_refinementByDirection (list of ints) –

Level of refinement by direction.
It takes the form: [xi-dir., eta-dir., zeta-dir.].
nb_degreeElevationByDirection (list of ints, optional) –

Number of degree elevation by direction. The default is [0, 0, 0].
It takes the form:[xi-dir., eta-dir., zeta-dir.].
additional_knots (dict, optional) –

Dictionnary for specific knot insertion. The default is None.
The dictionay takes the following keywords: patches, 1, 2, 3.

Example:
Refine the patch indexed as 2 at parameters u=0.5 and w=0.8. Hence we have: additional_knots = {‘patches’: np.array([2]), ‘1’: np.array([0.5]), ‘2’: np.array([]), ‘3’: np.array([0.8])}.

Returns:

transformationMatrices – Resulting transformation matrices for refinement.

Return type:

numpy array

set_geomectricSettings(newGeoSet): Set the geometrical settings.

set_mechanicalSettings(newMechSet): Set the mechanical settings.

shapeupdate()

Apply the shape update.

The shape update is defined by defined by the attribute _shapeparametrization_def.

The design parameters (names and values) are set in the attribute _design_parameters, and are taken as the inputs.

The function updates the control point coordinates _COORDS.

writeCOORDS(filename): Write the control points coordinates to file, ordered by patch.

writeCOORDS_tot(filename): Write all control points coordinates.