Solvers¶

FEA Solve¶

class feastruct.solvers.feasolve.Solver(analysis, analysis_cases, solver_settings)[source]¶

Parent class for a finite element solver.

Provides a number of methods suitable for most finite element analysis solver types.

Variables:

analysis (FiniteElementAnalysis) – Analysis object to solve

analysis_cases (list[AnalysisCase]) – List of analysis cases to solve

solver_settings (SolverSettings) – Settings to use in the solver

ndof (int) – Number of active degrees of freedom in the analysis

__init__(analysis, analysis_cases, solver_settings)[source]¶

Inits the Solver class.

Parameters:

analysis (FiniteElementAnalysis) – Analysis object to solve

analysis_cases (list[AnalysisCase]) – List of analysis cases to solve

solver_settings (SolverSettings) – Settings to use in the solver

apply_bcs(K, f_ext, analysis_case)[source]¶

Applies the boundary conditions to the global stiffness matrix and external force vector for the analysis case.

Parameters:

K (scipy.sparse.coo_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

analysis_case (AnalysisCase) – Analysis case

Returns:
The global stiffness matrix and external force vector, modified by the boundary conditions (K, f_ext)

Return type:
tuple(scipy.sparse.lil_matrix, numpy.ndarray)

assemble_fext(analysis_case)[source]¶

Assembles the external force vector for analysis_case.

Parameters: analysis_case (AnalysisCase) – Analysis case used to assemble the external force vector

Returns: The external force vector f_ext and the equivalent nodal loads vector f_eq

Return type: tuple(numpy.ndarray, numpy.ndarray)

assemble_mass_matrix()[source]¶

Assembles the global stiffness using the sparse COO format.

Returns: The global mass matrix

Return type: scipy.sparse.coo_matrix

assemble_stiff_matrix(geometric=False, analysis_case=None)[source]¶

Assembles the global stiffness using the sparse COO format.

Parameters:

geometric (bool) – If set to True, also returns the global geometric stiffness matrix

analysis_case (AnalysisCase) – If geometric is set to True, the static analysis case from which to use to build the geometric stiffness matrix

Returns:
The global stiffness matrix and, if desired, the global geometric stiffness matrix (K, K_g)

Return type:
tuple(scipy.sparse.coo_matrix, scipy.sparse.coo_matrix)

assign_dofs()[source]¶

Method to assign global degrees of freedom to the nodes in the analysis object.

calculate_reactions(K, u, f_eq, analysis_case)[source]¶

Calculates the reactions using the stiffness matrix K and the displacement vector u and saves the reactions for the analysis case.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

u (numpy.ndarray) – Displacement vector

f_eq (numpy.ndarray) – Global equivalent nodal loads vector of size N

analysis_case (AnalysisCase) – Analysis case

calculate_stresses(analysis_case)[source]¶

Calculates and saves the element stresses for the analysis case.

Parameters: analysis_case (AnalysisCase) – Analysis case

cgs_solver(K, f_ext)[source]¶

Solves Ku=f_ext for u using the cgs method.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

Returns:
Displacement vector u

Return type:
numpy.ndarray

Raises:
Exception – If the cgs solver does not converge or if the preconditioner fails

direct_solver(K, f_ext)[source]¶

Solves Ku=f_ext for u using the direct method.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

Returns:
Displacement vector u

Return type:
numpy.ndarray

function_timer(text, function, *args)[source]¶

Displays the message text and returns the time taken for a function, with arguments args, to execute. The value returned by the timed function is also returned.

Parameters:

text (string) – Message to display

function (function) – Function to time and execute

args – Function arguments

Returns:
Value returned from the function

remove_constrained_dofs(K, analysis_case)[source]¶

Given an initial matrix K, returns a new matrix with the constrained degrees of freedom removed. The constrained degrees of freedom are those with supports defined in the freedom case used in the analysis case.

Parameters:

K (scipy.sparse.coo_matrix) – Stiffness matrix

analysis_case (AnalysisCase) – Analysis case

Returns:
A new stiffness matrix with the constrined dofs removed

Return type:
scipy.sparse.lil_matrix

save_buckling_results(w, v, analysis_case)[source]¶

Saves the buckling results w and v to the degree of freedom objects for the analysis case.

Parameters:

w (numpy.ndarray) – Eigenvalues

v (numpy.ndarray) – Eigenvectors

analysis_case (AnalysisCase) – Analysis case

save_displacements(u, analysis_case)[source]¶

Saves the displacements u to the degree of freedom objects for the analysis case.

Parameters:

u (numpy.ndarray) – Displacement vector

analysis_case (AnalysisCase) – Analysis case

save_frequency_results(w, v, analysis_case)[source]¶

Saves the frequency results w and v to the degree of freedom objects for the analysis case.

Parameters:

w (numpy.ndarray) – Eigenvalues

v (numpy.ndarray) – Eigenvectors

analysis_case (AnalysisCase) – Analysis case

solve_eigenvalue(A, M, eigen_settings)[source]¶

Finds eigenvalues and eigenvectors for A * x[i] = w[i] * M * x[i].

Parameters:

A (scipy.sparse.coo_matrix) – A matrix

M (scipy.sparse.coo_matrix) – M matrix

eigen_settings (feastruct.solvers.feasolve.LinearBucklingSettings or feastruct.solvers.feasolve.NaturalFrequencySettings) – A settings object for use with the eigenvalue solver

Returns:
Eigenvalues and corresponding eigenvectors (w, v)

Return type:
tuple(numpy.ndarray, numpy.ndarray)

Raises:
Exception – If the eigenvalue solver does not converge

Linear Static Solver¶

class feastruct.solvers.linstatic.LinearStatic(analysis, analysis_cases, solver_settings=None)[source]¶

Bases: feastruct.solvers.feasolve.Solver

Class for a linear static solver.

Variables:

analysis (FiniteElementAnalysis) – Analysis object to solve

analysis_cases (list[AnalysisCase]) – List of analysis cases to solve

solver_settings (SolverSettings) – Settings to use in the solver

ndof (int) – Number of degrees of freedom in the analysis

__init__(analysis, analysis_cases, solver_settings=None)[source]¶

Inits the LinearStatic class.

Parameters:

analysis (FiniteElementAnalysis) – Analysis object to solve

analysis_cases (list[AnalysisCase]) – List of analysis cases to solve

solver_settings (SolverSettings) – Settings to use in the solver - if not supplied, the default settings are adopted

apply_bcs(K, f_ext, analysis_case)¶

Applies the boundary conditions to the global stiffness matrix and external force vector for the analysis case.

Parameters:

K (scipy.sparse.coo_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

analysis_case (AnalysisCase) – Analysis case

Returns:
The global stiffness matrix and external force vector, modified by the boundary conditions (K, f_ext)

Return type:
tuple(scipy.sparse.lil_matrix, numpy.ndarray)

assemble_fext(analysis_case)¶

Assembles the external force vector for analysis_case.

Parameters: analysis_case (AnalysisCase) – Analysis case used to assemble the external force vector

Returns: The external force vector f_ext and the equivalent nodal loads vector f_eq

Return type: tuple(numpy.ndarray, numpy.ndarray)

assemble_mass_matrix()¶

Assembles the global stiffness using the sparse COO format.

Returns: The global mass matrix

Return type: scipy.sparse.coo_matrix

assemble_stiff_matrix(geometric=False, analysis_case=None)¶

Assembles the global stiffness using the sparse COO format.

Parameters:

geometric (bool) – If set to True, also returns the global geometric stiffness matrix

analysis_case (AnalysisCase) – If geometric is set to True, the static analysis case from which to use to build the geometric stiffness matrix

Returns:
The global stiffness matrix and, if desired, the global geometric stiffness matrix (K, K_g)

Return type:
tuple(scipy.sparse.coo_matrix, scipy.sparse.coo_matrix)

assign_dofs()¶

Method to assign global degrees of freedom to the nodes in the analysis object.

calculate_reactions(K, u, f_eq, analysis_case)¶

Calculates the reactions using the stiffness matrix K and the displacement vector u and saves the reactions for the analysis case.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

u (numpy.ndarray) – Displacement vector

f_eq (numpy.ndarray) – Global equivalent nodal loads vector of size N

analysis_case (AnalysisCase) – Analysis case

calculate_stresses(analysis_case)¶

Calculates and saves the element stresses for the analysis case.

Parameters: analysis_case (AnalysisCase) – Analysis case

cgs_solver(K, f_ext)¶

Solves Ku=f_ext for u using the cgs method.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

Returns:
Displacement vector u

Return type:
numpy.ndarray

Raises:
Exception – If the cgs solver does not converge or if the preconditioner fails

direct_solver(K, f_ext)¶

Solves Ku=f_ext for u using the direct method.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

Returns:
Displacement vector u

Return type:
numpy.ndarray

function_timer(text, function, *args)¶

Displays the message text and returns the time taken for a function, with arguments args, to execute. The value returned by the timed function is also returned.

Parameters:

text (string) – Message to display

function (function) – Function to time and execute

args – Function arguments

Returns:
Value returned from the function

remove_constrained_dofs(K, analysis_case)¶

Given an initial matrix K, returns a new matrix with the constrained degrees of freedom removed. The constrained degrees of freedom are those with supports defined in the freedom case used in the analysis case.

Parameters:

K (scipy.sparse.coo_matrix) – Stiffness matrix

analysis_case (AnalysisCase) – Analysis case

Returns:
A new stiffness matrix with the constrined dofs removed

Return type:
scipy.sparse.lil_matrix

save_buckling_results(w, v, analysis_case)¶

Saves the buckling results w and v to the degree of freedom objects for the analysis case.

Parameters:

w (numpy.ndarray) – Eigenvalues

v (numpy.ndarray) – Eigenvectors

analysis_case (AnalysisCase) – Analysis case

save_displacements(u, analysis_case)¶

Saves the displacements u to the degree of freedom objects for the analysis case.

Parameters:

u (numpy.ndarray) – Displacement vector

analysis_case (AnalysisCase) – Analysis case

save_frequency_results(w, v, analysis_case)¶

Saves the frequency results w and v to the degree of freedom objects for the analysis case.

Parameters:

w (numpy.ndarray) – Eigenvalues

v (numpy.ndarray) – Eigenvectors

analysis_case (AnalysisCase) – Analysis case

solve()[source]¶

Executes the linear static finite element solver and saves the relevant results.

solve_eigenvalue(A, M, eigen_settings)¶

Finds eigenvalues and eigenvectors for A * x[i] = w[i] * M * x[i].

Parameters:

A (scipy.sparse.coo_matrix) – A matrix

M (scipy.sparse.coo_matrix) – M matrix

eigen_settings (feastruct.solvers.feasolve.LinearBucklingSettings or feastruct.solvers.feasolve.NaturalFrequencySettings) – A settings object for use with the eigenvalue solver

Returns:
Eigenvalues and corresponding eigenvectors (w, v)

Return type:
tuple(numpy.ndarray, numpy.ndarray)

Raises:
Exception – If the eigenvalue solver does not converge

Linear Buckling Solver¶

class feastruct.solvers.linbuckling.LinearBuckling(analysis, analysis_cases, solver_settings=None)[source]¶

Bases: feastruct.solvers.feasolve.Solver

Class for a linear buckling solver.

Variables:

analysis (FiniteElementAnalysis) – Analysis object to solve

analysis_cases (list[AnalysisCase]) – List of analysis cases to solve

solver_settings (SolverSettings) – Settings to use in the solver

ndof (int) – Number of degrees of freedom in the analysis

__init__(analysis, analysis_cases, solver_settings=None)[source]¶

Inits the LinearBuckling class.

Parameters:

analysis (FiniteElementAnalysis) – Analysis object to solve

analysis_cases (list[AnalysisCase]) – List of analysis cases to solve

solver_settings (SolverSettings) – Settings to use in the solver - if not supplied, the default settings are adopted

apply_bcs(K, f_ext, analysis_case)¶

Applies the boundary conditions to the global stiffness matrix and external force vector for the analysis case.

Parameters:

K (scipy.sparse.coo_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

analysis_case (AnalysisCase) – Analysis case

Returns:
The global stiffness matrix and external force vector, modified by the boundary conditions (K, f_ext)

Return type:
tuple(scipy.sparse.lil_matrix, numpy.ndarray)

assemble_fext(analysis_case)¶

Assembles the external force vector for analysis_case.

Parameters: analysis_case (AnalysisCase) – Analysis case used to assemble the external force vector

Returns: The external force vector f_ext and the equivalent nodal loads vector f_eq

Return type: tuple(numpy.ndarray, numpy.ndarray)

assemble_mass_matrix()¶

Assembles the global stiffness using the sparse COO format.

Returns: The global mass matrix

Return type: scipy.sparse.coo_matrix

assemble_stiff_matrix(geometric=False, analysis_case=None)¶

Assembles the global stiffness using the sparse COO format.

Parameters:

geometric (bool) – If set to True, also returns the global geometric stiffness matrix

analysis_case (AnalysisCase) – If geometric is set to True, the static analysis case from which to use to build the geometric stiffness matrix

Returns:
The global stiffness matrix and, if desired, the global geometric stiffness matrix (K, K_g)

Return type:
tuple(scipy.sparse.coo_matrix, scipy.sparse.coo_matrix)

assign_dofs()¶

Method to assign global degrees of freedom to the nodes in the analysis object.

calculate_reactions(K, u, f_eq, analysis_case)¶

Calculates the reactions using the stiffness matrix K and the displacement vector u and saves the reactions for the analysis case.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

u (numpy.ndarray) – Displacement vector

f_eq (numpy.ndarray) – Global equivalent nodal loads vector of size N

analysis_case (AnalysisCase) – Analysis case

calculate_stresses(analysis_case)¶

Calculates and saves the element stresses for the analysis case.

Parameters: analysis_case (AnalysisCase) – Analysis case

cgs_solver(K, f_ext)¶

Solves Ku=f_ext for u using the cgs method.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

Returns:
Displacement vector u

Return type:
numpy.ndarray

Raises:
Exception – If the cgs solver does not converge or if the preconditioner fails

direct_solver(K, f_ext)¶

Solves Ku=f_ext for u using the direct method.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

Returns:
Displacement vector u

Return type:
numpy.ndarray

function_timer(text, function, *args)¶

Displays the message text and returns the time taken for a function, with arguments args, to execute. The value returned by the timed function is also returned.

Parameters:

text (string) – Message to display

function (function) – Function to time and execute

args – Function arguments

Returns:
Value returned from the function

remove_constrained_dofs(K, analysis_case)¶

Given an initial matrix K, returns a new matrix with the constrained degrees of freedom removed. The constrained degrees of freedom are those with supports defined in the freedom case used in the analysis case.

Parameters:

K (scipy.sparse.coo_matrix) – Stiffness matrix

analysis_case (AnalysisCase) – Analysis case

Returns:
A new stiffness matrix with the constrined dofs removed

Return type:
scipy.sparse.lil_matrix

save_buckling_results(w, v, analysis_case)¶

Saves the buckling results w and v to the degree of freedom objects for the analysis case.

Parameters:

w (numpy.ndarray) – Eigenvalues

v (numpy.ndarray) – Eigenvectors

analysis_case (AnalysisCase) – Analysis case

save_displacements(u, analysis_case)¶

Saves the displacements u to the degree of freedom objects for the analysis case.

Parameters:

u (numpy.ndarray) – Displacement vector

analysis_case (AnalysisCase) – Analysis case

save_frequency_results(w, v, analysis_case)¶

Saves the frequency results w and v to the degree of freedom objects for the analysis case.

Parameters:

w (numpy.ndarray) – Eigenvalues

v (numpy.ndarray) – Eigenvectors

analysis_case (AnalysisCase) – Analysis case

solve()[source]¶

Executes the linear buckling finite element solver and saves the relevant results.

solve_eigenvalue(A, M, eigen_settings)¶

Finds eigenvalues and eigenvectors for A * x[i] = w[i] * M * x[i].

Parameters:

A (scipy.sparse.coo_matrix) – A matrix

M (scipy.sparse.coo_matrix) – M matrix

eigen_settings (feastruct.solvers.feasolve.LinearBucklingSettings or feastruct.solvers.feasolve.NaturalFrequencySettings) – A settings object for use with the eigenvalue solver

Returns:
Eigenvalues and corresponding eigenvectors (w, v)

Return type:
tuple(numpy.ndarray, numpy.ndarray)

Raises:
Exception – If the eigenvalue solver does not converge

Natural Frequency Solver¶

class feastruct.solvers.naturalfrequency.NaturalFrequency(analysis, analysis_cases, solver_settings=None)[source]¶

Bases: feastruct.solvers.feasolve.Solver

Class for a natural frequency solver.

Variables:

analysis (FiniteElementAnalysis) – Analysis object to solve

analysis_cases (list[AnalysisCase]) – List of analysis cases to solve

solver_settings (SolverSettings) – Settings to use in the solver

ndof (int) – Number of degrees of freedom in the analysis

__init__(analysis, analysis_cases, solver_settings=None)[source]¶

Inits the NaturalFrequency class.

Parameters:

analysis (FiniteElementAnalysis) – Analysis object to solve

analysis_cases (list[AnalysisCase]) – List of analysis cases to solve

solver_settings (SolverSettings) – Settings to use in the solver - if not supplied, the default settings are adopted

apply_bcs(K, f_ext, analysis_case)¶

Applies the boundary conditions to the global stiffness matrix and external force vector for the analysis case.

Parameters:

K (scipy.sparse.coo_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

analysis_case (AnalysisCase) – Analysis case

Returns:
The global stiffness matrix and external force vector, modified by the boundary conditions (K, f_ext)

Return type:
tuple(scipy.sparse.lil_matrix, numpy.ndarray)

assemble_fext(analysis_case)¶

Assembles the external force vector for analysis_case.

Parameters: analysis_case (AnalysisCase) – Analysis case used to assemble the external force vector

Returns: The external force vector f_ext and the equivalent nodal loads vector f_eq

Return type: tuple(numpy.ndarray, numpy.ndarray)

assemble_mass_matrix()¶

Assembles the global stiffness using the sparse COO format.

Returns: The global mass matrix

Return type: scipy.sparse.coo_matrix

assemble_stiff_matrix(geometric=False, analysis_case=None)¶

Assembles the global stiffness using the sparse COO format.

Parameters:

geometric (bool) – If set to True, also returns the global geometric stiffness matrix

analysis_case (AnalysisCase) – If geometric is set to True, the static analysis case from which to use to build the geometric stiffness matrix

Returns:
The global stiffness matrix and, if desired, the global geometric stiffness matrix (K, K_g)

Return type:
tuple(scipy.sparse.coo_matrix, scipy.sparse.coo_matrix)

assign_dofs()¶

Method to assign global degrees of freedom to the nodes in the analysis object.

calculate_reactions(K, u, f_eq, analysis_case)¶

Calculates the reactions using the stiffness matrix K and the displacement vector u and saves the reactions for the analysis case.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

u (numpy.ndarray) – Displacement vector

f_eq (numpy.ndarray) – Global equivalent nodal loads vector of size N

analysis_case (AnalysisCase) – Analysis case

calculate_stresses(analysis_case)¶

Calculates and saves the element stresses for the analysis case.

Parameters: analysis_case (AnalysisCase) – Analysis case

cgs_solver(K, f_ext)¶

Solves Ku=f_ext for u using the cgs method.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

Returns:
Displacement vector u

Return type:
numpy.ndarray

Raises:
Exception – If the cgs solver does not converge or if the preconditioner fails

direct_solver(K, f_ext)¶

Solves Ku=f_ext for u using the direct method.

Parameters:

K (scipy.sparse.lil_matrix) – Global stiffness matrix

f_ext (numpy.ndarray) – External force vector

Returns:
Displacement vector u

Return type:
numpy.ndarray

function_timer(text, function, *args)¶

Displays the message text and returns the time taken for a function, with arguments args, to execute. The value returned by the timed function is also returned.

Parameters:

text (string) – Message to display

function (function) – Function to time and execute

args – Function arguments

Returns:
Value returned from the function

remove_constrained_dofs(K, analysis_case)¶

Given an initial matrix K, returns a new matrix with the constrained degrees of freedom removed. The constrained degrees of freedom are those with supports defined in the freedom case used in the analysis case.

Parameters:

K (scipy.sparse.coo_matrix) – Stiffness matrix

analysis_case (AnalysisCase) – Analysis case

Returns:
A new stiffness matrix with the constrined dofs removed

Return type:
scipy.sparse.lil_matrix

save_buckling_results(w, v, analysis_case)¶

Saves the buckling results w and v to the degree of freedom objects for the analysis case.

Parameters:

w (numpy.ndarray) – Eigenvalues

v (numpy.ndarray) – Eigenvectors

analysis_case (AnalysisCase) – Analysis case

save_displacements(u, analysis_case)¶

Saves the displacements u to the degree of freedom objects for the analysis case.

Parameters:

u (numpy.ndarray) – Displacement vector

analysis_case (AnalysisCase) – Analysis case

save_frequency_results(w, v, analysis_case)¶

Saves the frequency results w and v to the degree of freedom objects for the analysis case.

Parameters:

w (numpy.ndarray) – Eigenvalues

v (numpy.ndarray) – Eigenvectors

analysis_case (AnalysisCase) – Analysis case

solve()[source]¶

Executes the natural frequency finite element solver and saves the relevant results.

solve_eigenvalue(A, M, eigen_settings)¶

Finds eigenvalues and eigenvectors for A * x[i] = w[i] * M * x[i].

Parameters:

A (scipy.sparse.coo_matrix) – A matrix

M (scipy.sparse.coo_matrix) – M matrix

eigen_settings (feastruct.solvers.feasolve.LinearBucklingSettings or feastruct.solvers.feasolve.NaturalFrequencySettings) – A settings object for use with the eigenvalue solver

Returns:
Eigenvalues and corresponding eigenvectors (w, v)

Return type:
tuple(numpy.ndarray, numpy.ndarray)

Raises:
Exception – If the eigenvalue solver does not converge

Solver Settings¶

class feastruct.solvers.feasolve.SolverSettings[source]¶

Class for settings to be used in a finite element analysis.

Variables:

linear_static (feastruct.solvers.feasolve.LinearStaticSettings) – Linear static settings

linear_buckling (feastruct.solvers.feasolve.LinearBucklingSettings) – Linear buckling settings

natural_frequency (feastruct.solvers.feasolve.NaturalFrequencySettings) – Natural frequency settings

__init__()[source]¶

Inits the SolverSettings class.

Linear Static Settings¶

class feastruct.solvers.feasolve.LinearStaticSettings(time_info=False, solver_type='direct', cgs_tol=1e-05, cgs_maxiter=None, cgs_precond=True)[source]¶

Bases: object

Class for settings to be used in a linear static analysis.

Variables:

time_info (bool) – If set to True, a detailed description of the computation and the time cost is printed to the terminal
solver_type (string) – Solver type used to solve Ku = f - ‘direct’ or ‘cgs’
cgs_tol (float) – Tolerance to achieve for the cgs solver
cgs_maxiter (int) – Maximum number of iterations for the cgs solver
cgs_precond (bool) – If set to True, calculates a preconditioner for K using an ILU decomposition

__init__(time_info=False, solver_type='direct', cgs_tol=1e-05, cgs_maxiter=None, cgs_precond=True)[source]¶

Inits the LinearStaticSettings class.

Parameters:

time_info (bool) – If set to True, a detailed description of the computation and the time cost is printed to the terminal
solver_type (string) – Solver type used to solve Ku = f - ‘direct’ or ‘cgs’
cgs_tol (float) – Tolerance to achieve for the cgs solver
cgs_maxiter (int) – Maximum number of iterations for the cgs solver
cgs_precond (bool) – If set to True, calculates a preconditioner for K using an ILU decomposition

Linear Buckling Settings¶

class feastruct.solvers.feasolve.LinearBucklingSettings(time_info=False, num_modes=4, shift=0, maxiter=None, tol=1e-09)[source]¶

Bases: object

Class for settings to be used in a linear buckling analysis.

__init__(time_info=False, num_modes=4, shift=0, maxiter=None, tol=1e-09)[source]¶

Inits the LinearBucklingSettings class.

Parameters:

time_info (bool) – If set to True, a detailed description of the computation and the time cost is printed to the terminal
num_modes (int) – Number of modes to compute
shift (float) – Finds eigenvalues near shift using the shift-invert mode
maxiter (int) – Maximum number of iterations allowed
tol (float) – Relative accuracy for eigenvalues (stopping criterion)

Natural Frequency Settings¶

class feastruct.solvers.feasolve.NaturalFrequencySettings(time_info=False, num_modes=4, shift=0, maxiter=None, tol=1e-09)[source]¶

Bases: object

Class for settings to be used in a natural frequency analysis.

__init__(time_info=False, num_modes=4, shift=0, maxiter=None, tol=1e-09)[source]¶

Inits the NaturalFrequencySettings class.

Parameters:

time_info (bool) – If set to True, a detailed description of the computation and the time cost is printed to the terminal
num_modes (int) – Number of modes to compute
shift (float) – Finds eigenvalues near shift using the shift-invert mode
maxiter (int) – Maximum number of iterations allowed
tol (float) – Relative accuracy for eigenvalues (stopping criterion)