RANFSNormalMechanicalContact

Enforces the zero penetration constraint using the Reduced Active Nonlinear Function Set (RANFS) scheme. The algorithm is as follows: do a min comparison between the penetration and the Lagrange Multiplier (LM) associated with the zero penetration constraint. If the penetration (properly signed) is less than the LM, then we replace the non-linear residual equation for the secondary node with the zero-penetration constraint equation (represented simply by the gap). In this way the gap is required to be zero.

Advantages of the RANFS scheme:

No Lagrange Multipliers, e.g. no saddle-point and consequently we are free to use preconditioners like AMG
No penalty term, so no introduction of ill conditioning into the matrix
Exact satisfaction of the constraints

The RANFSNormalMechanicalContact object includes ping-ponging protection. It detects when a secondary node has alternated back and forth between two different element faces a sufficient number of times (currently face1-face2-face1-face2-face1) and then fixes the issue by applying more RANFS Explicitly, we apply a number of equality constraints equal to the mesh dimension tying together the locations of the secondary node and the nearest primary node (which we assert is a node that both primary faces involved in the ping-pong share). This ping-ponging protection is necessary for solving step 19 of the full-bouncing-block-ranfs test. Without the protection the solve does not converge and the time-step gets cut.

Applies the Reduced Active Nonlinear Function Set scheme in which the secondary node's non-linear residual function is replaced by the zero penetration constraint equation when the constraint is active

Input Parameters

componentThe force component constraint that this object is supplying
C++ Type:MooseEnum
Unit:(no unit assumed)
Options:x, y, z
Controllable:No
Description:The force component constraint that this object is supplying
displacementsThe displacements appropriate for the simulation geometry and coordinate system
C++ Type:std::vector<VariableName>
Unit:(no unit assumed)
Controllable:No
Description:The displacements appropriate for the simulation geometry and coordinate system
variableThe name of the variable that this residual object operates on
C++ Type:NonlinearVariableName
Unit:(no unit assumed)
Controllable:No
Description:The name of the variable that this residual object operates on

Required Parameters

normal_smoothing_distanceDistance from edge in parametric coordinates over which to smooth contact normal
C++ Type:double
Unit:(no unit assumed)
Controllable:No
Description:Distance from edge in parametric coordinates over which to smooth contact normal
normal_smoothing_methodMethod to use to smooth normals (edge_based|nodal_normal_based)
C++ Type:std::string
Unit:(no unit assumed)
Controllable:No
Description:Method to use to smooth normals (edge_based|nodal_normal_based)
orderFIRSTThe finite element order used for projections
Default:FIRST
C++ Type:MooseEnum
Unit:(no unit assumed)
Options:FIRST, SECOND, THIRD, FOURTH
Controllable:No
Description:The finite element order used for projections
ping_pong_protectionFalseWhether to protect against ping-ponging, e.g. the oscillation of the secondary node between two different primary faces, by tying the secondary node to the edge between the involved primary faces
Default:False
C++ Type:bool
Unit:(no unit assumed)
Controllable:No
Description:Whether to protect against ping-ponging, e.g. the oscillation of the secondary node between two different primary faces, by tying the secondary node to the edge between the involved primary faces
primaryThe boundary ID associated with the primary side
C++ Type:BoundaryName
Unit:(no unit assumed)
Controllable:No
Description:The boundary ID associated with the primary side
primary_variableThe variable on the primary side of the domain
C++ Type:std::vector<VariableName>
Unit:(no unit assumed)
Controllable:No
Description:The variable on the primary side of the domain
secondaryThe boundary ID associated with the secondary side
C++ Type:BoundaryName
Unit:(no unit assumed)
Controllable:No
Description:The boundary ID associated with the secondary side
tangential_toleranceTangential distance to extend edges of contact surfaces
C++ Type:double
Unit:(no unit assumed)
Controllable:No
Description:Tangential distance to extend edges of contact surfaces

Optional Parameters

absolute_value_vector_tagsThe tags for the vectors this residual object should fill with the absolute value of the residual contribution
C++ Type:std::vector<TagName>
Unit:(no unit assumed)
Controllable:No
Description:The tags for the vectors this residual object should fill with the absolute value of the residual contribution
extra_matrix_tagsThe extra tags for the matrices this Kernel should fill
C++ Type:std::vector<TagName>
Unit:(no unit assumed)
Controllable:No
Description:The extra tags for the matrices this Kernel should fill
extra_vector_tagsThe extra tags for the vectors this Kernel should fill
C++ Type:std::vector<TagName>
Unit:(no unit assumed)
Controllable:No
Description:The extra tags for the vectors this Kernel should fill
matrix_tagssystemThe tag for the matrices this Kernel should fill
Default:system
C++ Type:MultiMooseEnum
Unit:(no unit assumed)
Options:nontime, system
Controllable:No
Description:The tag for the matrices this Kernel should fill
vector_tagsnontimeThe tag for the vectors this Kernel should fill
Default:nontime
C++ Type:MultiMooseEnum
Unit:(no unit assumed)
Options:nontime, time
Controllable:No
Description:The tag for the vectors this Kernel should fill

Tagging Parameters

control_tagsAdds user-defined labels for accessing object parameters via control logic.
C++ Type:std::vector<std::string>
Unit:(no unit assumed)
Controllable:No
Description:Adds user-defined labels for accessing object parameters via control logic.
enableTrueSet the enabled status of the MooseObject.
Default:True
C++ Type:bool
Unit:(no unit assumed)
Controllable:Yes
Description:Set the enabled status of the MooseObject.
implicitTrueDetermines whether this object is calculated using an implicit or explicit form
Default:True
C++ Type:bool
Unit:(no unit assumed)
Controllable:No
Description:Determines whether this object is calculated using an implicit or explicit form
seed0The seed for the master random number generator
Default:0
C++ Type:unsigned int
Unit:(no unit assumed)
Controllable:No
Description:The seed for the master random number generator
use_displaced_meshTrueWhether or not this object should use the displaced mesh for computation. Note that in the case this is true but no displacements are provided in the Mesh block the undisplaced mesh will still be used.
Default:True
C++ Type:bool
Unit:(no unit assumed)
Controllable:No
Description:Whether or not this object should use the displaced mesh for computation. Note that in the case this is true but no displacements are provided in the Mesh block the undisplaced mesh will still be used.

Advanced Parameters

Input Files

(modules/contact/test/tests/bouncing-block-contact/frictionless-nodal-reduced-active-set.i)
(modules/contact/test/tests/ranfs-and-scaling/bouncing-block-ranfs.i)
(modules/contact/test/tests/bouncing-block-contact/ping-ponging/ranfs-ping-pong.i)

(modules/contact/test/tests/bouncing-block-contact/bouncing-block-ranfs.i)

starting_point = 2e-1
offset = 1e-2

[GlobalParams]
  displacements = 'disp_x disp_y'
  diffusivity = 1e0
  ping_pong_protection = true
[]

[Mesh]
  file = long-bottom-block-no-lower-d.e
[]

[Variables]
  [./disp_x]
  [../]
  [./disp_y]
  [../]
[]

[ICs]
  [./disp_y]
    block = 2
    variable = disp_y
    value = ${fparse starting_point + offset}
    type = ConstantIC
  [../]
[]

[Kernels]
  [./disp_x]
    type = MatDiffusion
    variable = disp_x
  [../]
  [./disp_y]
    type = MatDiffusion
    variable = disp_y
  [../]
[]

[Contact]
  [./top_bottom]
    secondary = 10
    primary = 20

    model = frictionless
    formulation = ranfs
  [../]
[]

[BCs]
  [./botx]
    type = DirichletBC
    variable = disp_x
    boundary = 40
    value = 0.0
  [../]
  [./boty]
    type = DirichletBC
    variable = disp_y
    boundary = 40
    value = 0.0
  [../]
  [./topy]
    type = FunctionDirichletBC
    variable = disp_y
    boundary = 30
    function = '${starting_point} * cos(2 * pi / 40 * t) + ${offset}'
  [../]
  [./leftx]
    type = FunctionDirichletBC
    variable = disp_x
    boundary = 50
    function = '1e-2 * t'
  [../]
[]

[Executioner]
  type = Transient
  end_time = 200
  dt = 5
  dtmin = 2.5
  solve_type = 'PJFNK'
  petsc_options = '-snes_converged_reason -ksp_converged_reason'
  petsc_options_iname = '-pc_type -pc_hypre_type -mat_mffd_err'
  petsc_options_value = 'hypre    boomeramg      1e-5'
  l_max_its = 30
  nl_max_its = 20
  line_search = 'none'
[]

[Debug]
  show_var_residual_norms = true
[]

[Outputs]
  [exo]
    type = Exodus
  []
  checkpoint = true
[]

[Preconditioning]
  [./smp]
    type = SMP
    full = true
  [../]
[]

[Postprocessors]
  [./num_nl]
    type = NumNonlinearIterations
  [../]
  [./cumulative]
    type = CumulativeValuePostprocessor
    postprocessor = num_nl
  [../]
[]

(modules/contact/test/tests/bouncing-block-contact/frictionless-nodal-reduced-active-set.i)

starting_point = .5

[GlobalParams]
  displacements = 'disp_x disp_y'
  diffusivity = 1
[]

[Mesh]
  file = square-blocks-no-offset.e
[]

[Variables]
  [./disp_x]
  [../]
  [./disp_y]
  [../]
[]

[ICs]
  [./disp_y]
    block = 2
    variable = disp_y
    value = ${starting_point}
    type = ConstantIC
  [../]
[]

[Kernels]
  [./disp_x]
    type = MatDiffusion
    variable = disp_x
  [../]
  [./disp_y]
    type = MatDiffusion
    variable = disp_y
  [../]
[]


[Constraints]
  [./disp_x]
    type = RANFSNormalMechanicalContact
    secondary = 10
    primary = 20
    variable = disp_x
    primary_variable = disp_x
    component = x
  [../]
  [./disp_y]
    type = RANFSNormalMechanicalContact
    secondary = 10
    primary = 20
    variable = disp_y
    primary_variable = disp_y
    component = y
  [../]
[]

[BCs]
  [./botx]
    type = DirichletBC
    variable = disp_x
    preset = false
    boundary = 40
    value = 0.0
  [../]
  [./topx]
    type = DirichletBC
    variable = disp_x
    preset = false
    boundary = 30
    value = 0.0
  [../]
  [./boty]
    type = DirichletBC
    variable = disp_y
    preset = false
    boundary = 40
    value = 0.0
  [../]
  [./topy]
    type = FunctionDirichletBC
    variable = disp_y
    preset = false
    boundary = 30
    function = '${starting_point} - t'
  [../]
[]

[Executioner]
  type = Transient
  num_steps = 1
  dt = 1
  dtmin = 1
  solve_type = 'PJFNK'
  petsc_options = '-snes_converged_reason -ksp_converged_reason -ksp_monitor_true_residual -snes_view'
  petsc_options_iname = '-mat_mffd_err -pc_type -pc_hypre_type'
  petsc_options_value = '1e-5          hypre    boomeramg'
  l_max_its = 30
  nl_max_its = 20
  line_search = 'project'
  snesmf_reuse_base = false
[]

[Debug]
  show_var_residual_norms = true
[]

[Outputs]
  [exo]
    type = Exodus
    execute_on = 'nonlinear'
  []
  print_linear_residuals = false
  [dof]
    type = DOFMap
    execute_on = 'initial'
  []
[]

[Postprocessors]
  [./num_nl]
    type = NumNonlinearIterations
  [../]
  [./cumulative]
    type = CumulativeValuePostprocessor
    postprocessor = num_nl
  [../]
[]

(modules/contact/test/tests/ranfs-and-scaling/bouncing-block-ranfs.i)

starting_point = 2e-1
offset = 1e-2

[GlobalParams]
  displacements = 'disp_x disp_y'
[]

[Mesh]
  file = long-bottom-block-no-lower-d-coarse.e
[]

[Variables]
  [./disp_x]
  [../]
  [./disp_y]
  [../]
[]

[ICs]
  [./disp_y]
    block = 2
    variable = disp_y
    value = ${fparse starting_point + offset}
    type = ConstantIC
  [../]
[]

[Physics/SolidMechanics/QuasiStatic]
  [all]
    add_variables = false
    use_automatic_differentiation = true
    strain = SMALL
  []
[]

[Materials]
  [elasticity]
    type = ADComputeIsotropicElasticityTensor
    youngs_modulus = 1e3
    poissons_ratio = 0.3
  []
  [stress]
    type = ADComputeLinearElasticStress
  []
[]

[Constraints]
  [./disp_x]
    type = RANFSNormalMechanicalContact
    secondary = 10
    primary = 20
    variable = disp_x
    primary_variable = disp_x
    component = x
    normal_smoothing_distance = 0.1
  [../]
  [./disp_y]
    type = RANFSNormalMechanicalContact
    secondary = 10
    primary = 20
    variable = disp_y
    primary_variable = disp_y
    component = y
    normal_smoothing_distance = 0.1
  [../]
[]

[BCs]
  [./botx]
    type = DirichletBC
    variable = disp_x
    boundary = 40
    value = 0.0
  [../]
  [./boty]
    type = DirichletBC
    variable = disp_y
    boundary = 40
    value = 0.0
  [../]
  [./topy]
    type = FunctionDirichletBC
    variable = disp_y
    boundary = 30
    function = '${starting_point} * cos(2 * pi / 40 * t) + ${offset}'
  [../]
  [./leftx]
    type = FunctionDirichletBC
    variable = disp_x
    boundary = 50
    function = '1e-2 * t'
  [../]
[]

[Executioner]
  type = Transient
  end_time = 100
  dt = 5
  dtmin = 2.5
  solve_type = 'PJFNK'
  petsc_options = '-snes_converged_reason -ksp_converged_reason'
  petsc_options_iname = '-pc_type -pc_hypre_type -mat_mffd_err'
  petsc_options_value = 'hypre    boomeramg      1e-5'
  l_max_its = 30
  nl_max_its = 20
  line_search = 'none'
  automatic_scaling = true
  verbose = true
  scaling_group_variables = 'disp_x disp_y'
  resid_vs_jac_scaling_param = 1
  nl_rel_tol = 1e-12
  snesmf_reuse_base = false
[]

[Debug]
  show_var_residual_norms = true
[]

[Outputs]
  [exo]
    type = Exodus
  []
[]

[Preconditioning]
  [./smp]
    type = SMP
    full = true
  [../]
[]

[Postprocessors]
  [nl]
    type = NumNonlinearIterations
  []
  [lin]
    type = NumLinearIterations
  []
  [tot_nl]
    type = CumulativeValuePostprocessor
    postprocessor = nl
  []
  [tot_lin]
    type = CumulativeValuePostprocessor
    postprocessor = lin
  []
[]

(modules/contact/test/tests/bouncing-block-contact/ping-ponging/ranfs-ping-pong.i)

starting_point = 2e-1
offset = 1e-2

[GlobalParams]
  displacements = 'disp_x disp_y'
[]

[Mesh]
  file = long-bottom-block-no-lower-d.e
[]

[Variables]
  [./disp_x]
  [../]
  [./disp_y]
  [../]
[]

[ICs]
  [./disp_y]
    block = 2
    variable = disp_y
    value = ${fparse starting_point + offset}
    type = ConstantIC
  [../]
[]

[Physics/SolidMechanics/QuasiStatic]
  [all]
    add_variables = false
    use_automatic_differentiation = true
    strain = SMALL
  []
[]

[Materials]
  [elasticity]
    type = ADComputeIsotropicElasticityTensor
    youngs_modulus = 1e0
    poissons_ratio = 0.3
  []
  [stress]
    type = ADComputeLinearElasticStress
  []
[]

[Constraints]
  [./disp_x]
    type = RANFSNormalMechanicalContact
    secondary = 10
    primary = 20
    variable = disp_x
    primary_variable = disp_x
    component = x
  [../]
  [./disp_y]
    type = RANFSNormalMechanicalContact
    secondary = 10
    primary = 20
    variable = disp_y
    primary_variable = disp_y
    component = y
  [../]
[]

[BCs]
  [./botx]
    type = DirichletBC
    variable = disp_x
    boundary = 40
    value = 0.0
  [../]
  [./boty]
    type = DirichletBC
    variable = disp_y
    boundary = 40
    value = 0.0
  [../]
  [./topy]
    type = FunctionDirichletBC
    variable = disp_y
    boundary = 30
    function = '${starting_point} * cos(2 * pi / 40 * t) + ${offset}'
  [../]
  [./leftx]
    type = FunctionDirichletBC
    variable = disp_x
    boundary = 50
    function = '1e-2 * t'
  [../]
[]

[Executioner]
  type = Transient
  num_steps = 19
  end_time = 200
  dt = 5
  dtmin = 5
  solve_type = 'PJFNK'
  petsc_options = '-snes_converged_reason -ksp_converged_reason'
  petsc_options_iname = '-pc_type -pc_hypre_type -mat_mffd_err'
  petsc_options_value = 'hypre    boomeramg      1e-5'
  l_max_its = 30
  nl_max_its = 20
  line_search = 'none'
  snesmf_reuse_base = false
[]

[Debug]
  show_var_residual_norms = true
[]

[Outputs]
  [exo]
    type = Exodus
  []
[]

[Preconditioning]
  [./smp]
    type = SMP
    full = true
  [../]
[]

Input Parameters
Input Files