Computational Contact Mechanics Yastrebov

Introduction to
Computational Contact Mechanics

Part I. Basics
Vladislav A. Yastrebov
Centre des Matériaux, MINES ParisTech, CNRS UMR 7633
WEMESURF short course on contact mechanics and tribology

Paris, France, 21-24 June 2010
Intro Detection Geometry Discretization Solution FEA Examples
Preface
To whom the course is aimed?
Developpers and users.
What is the aim?
Accurate contact modeling, correct interpretation, etc.
FEM - Finite Element Method, FEA - Finite Element Analysis
V.A. Yastrebov | MINES ParisTech Computational Contact Mechanics Paris, 21-24 June 2010 2/77
Outline
Mathematical foundation
contact geometry;
optimization methods.
Inside the Finite Element programme

contact detection;
contact discretization;
account of contact.
Contact problem resolution with FEM: guide for engineer

master-slave approach;
boundary conditions;
spurious cases.
Content
demonstrative;
simple;
general.
Plan
1 Introduction
2 Contact detection
3 Contact geometry
4 Contact discretization methods
5 Solution of contact problem
6 Finite Element Analysis of contact problems
7 Numerical examples
Finite Element Method

FEM a powerfull and multi-purpose tool for linear and
non-linear dynamic and static continuum mechanical and
multi-physic problems
[Rousselier et al.] [Klyavin et al.] [Brinkmeiera et al.]

FEM a powerfull and multi-purpose tool for linear and
non-linear dynamic and static continuum mechanical and
multi-physic problems
[Rousselier et al.] [Klyavin et al.] [Brinkmeiera et al.]
Contact problems are not continuum and they require:

new rigorous mathematical basis: geometry,
optimization, non-smooth analysis;
particular treatment of the nite element algorithms;
smart use of the Finite Element Analysis (FEA).
Finite element analysis of contact problems

assembled components; Disk-blade contact
T.Dick, G.Cailletaud
Centre des Matériaux

assembled components; Rolling bearing
bearings;
F.Massi et al.
LaMCoS, INSA-Lyon et al.

assembled components; Tire rolling noise simulation
bearings;
rolling contact;
M.Brinkmeiera, U.Nackenhorst et al.

University of Hannover et al.

assembled components; Deep drawing
bearings;
rolling contact;
forming processes;
G.Rousselier et al.
Centre des Matériaux et al.

assembled components; Post seismic relaxation
bearings;
rolling contact;
forming processes;
geomechanical contact;
J.D.Garaud, L.Fleitout, G.Cailletaud

Centre des Matériaux, ENS

assembled components; Crash-test
bearings;
rolling contact;
forming processes;
crash tests;
O.Klyavin, A.Michailov, A.Borovkov

St Petersbourg State University

assembled components; Knee simulation
bearings;
rolling contact;
forming processes;
crash tests;
human joints;
E.Peña, B.Calvoa et al.

University of Zaragoza

assembled components; Knee simulation
bearings;
rolling contact;
forming processes;
crash tests;
human joints;
and many others.
E.Peña, B.Calvoa et al.

University of Zaragoza
They trust in the FEA

Leading international industrial companies
Mechanical problem
From a real life problem to an engineering problem
Need to determine:
the problematic:
strength/life-time/fracture;
vibration/buckling;
thermo-electro-mechanical.
relevant geometry;
relevant loads:
static/quasi-static/dynamic;
mechanical/thermic;
volume/surface.
relevant material:
rigid/elastic/plastic/visco-plastic;
brittle/ductile.
relevant scale:
macro/meso/micro.
Mechanical problem
Need to determine:
the problematic:
strength/life-time/fracture; For example: Brinell hardness
vibration/buckling; test
thermo-electro-mechanical.
relevant geometry;
relevant loads:
static/quasi-static/dynamic;
mechanical/thermic;
volume/surface.
relevant material:
rigid/elastic/plastic/visco-plastic; Scheme of the Brinell hardness test and
dierent types of impression
brittle/ductile. [Harry Chandler, Hardness testing]
relevant scale:
macro/meso/micro.
Mechanical problem
Need to determine:
problematic:
strength. For example: Brinell hardness
test
geometry;
sphere + half-space.
loads:
mechanical surface quasi-static.
material:
rigid + elasto-visco-plastic.
scale:
macro. Scheme of the Brinell hardness test and
dierent types of impression
[Harry Chandler, Hardness testing]
Mechanical problem
From an engineering problem to a nite element model
Problem: FE model:
problematic: analysis type:
strength/life-time/fracture; stress-strain state;
vibration/buckling; eigen values;
thermo-electro-mechanical. coupled physics.
Mechanical problem
Problem: FE model:
geometry; nite element mesh;
Mechanical problem
Problem: FE model:
loads: analysis type and BC:
static/quasi-static/dynamic; static/quasi-static/dynamic;
mechanical/thermic; mechanical/thermic;
volume/surface. volume/surface.
Mechanical problem
Problem: FE model:
material: material model:
rigid/elastic/plastic/visco- rigid/elastic/plastic/visco-
plastic; plastic;
brittle/ductile. brittle/ductile.
Mechanical problem
Problem: FE model:
rigid/elastic/plastic/visco- rigid/elastic/plastic/visco-
plastic; plastic;
brittle/ductile. brittle/ductile.
scale: microstructure:
macro/meso/micro. RVE/microstructure/.
Mechanical problem
From engineering problem to a nite element model
Problem: FE model:
strength. stress-strain state.
geometry: nite element mesh:
Mechanical problem
Problem: FE model:
sphere + half-space. sphere + large block.
Mechanical problem
Problem: FE model:
mechanical surface mechanical surface
quasi-static. quasi-static.
Mechanical problem
Problem: FE model:
rigid + elasto-visco-plastic. rigida + elasto-visco-plastic
scale: model.
microstructure:
a rigid in FEA: much more harder than another solid,
special boundary conditions or geometrical
representation.
Mechanical problem
Problem: FE model:
rigid + elasto-visco-plastic. rigida + elasto-visco-plastic
scale: model.
macro. microstructure:
homogeneous > RVE.
a
rigid in FEA: much more harder than another solid,
special boundary conditions or geometrical
representation.
Mechanical problem
Problem: FE model:
problematic analysis type
geometry nite element mesh
loads analysis type and BC
material material model
scale microstructure
Mechanical problem
Problem: FE model:
problematic analysis type
geometry nite element mesh
loads analysis type and BC
material material model
scale microstructure
Finite element mesh : )
Symmetry and plane problems

How to solve problems faster
Main ideas:
symmetry
geometry AND loading;
3D to 2D:
axisymmetry/plane
strain/plane stress;
3D to smaller 3D:
half/quarter/sector;

Axisymmetry
Main ideas:
symmetry
3D to 2D:
axisymmetry/plane
3D to smaller 3D:
Axisymmetry of geometry
axisymmetry of load

Mirror symmetry
Main ideas:
symmetry
3D to 2D:
axisymmetry/plane
3D to smaller 3D:
mirror symmetry of load

No symmetry
Main ideas:
symmetry
3D to 2D:
axisymmetry/plane
3D to smaller 3D:
no symmetry of load

Plain strain
Main ideas:
symmetry
3D to 2D:
axisymmetry/plane
3D to smaller 3D:
Mirror symmetry of geometry
very long structure or xed edges

Plain stress
Main ideas:
symmetry
3D to 2D:
axisymmetry/plane
3D to smaller 3D:
half/quarter/sector; Mirror symmetry of geometry
very thin structure

Full 3D mesh Axisymmetric 2D mesh
Finite element mesh

Basics
In general the nite element mesh should

full the required solution precision;
correctly represent relevant geometry;
not be enormous;
be ne where strain is large;
be rough where strain is small;
avoid too oblong elements.
In contact problems the nite element mesh should
not allow corners at master surface;
be very ne and precise on both contacting surfaces;
use carefully quadratic elements.
Boundary conditions
Two solids Ω1 and Ω2 .

Volumetric forces Fv : e.g. inertia mü.
Neumann (static, force) boundary
conditions: distributed and
concentrated loads.
Dirichlet (kinematic, displacement)
boundary conditions: displacements.
In each solid we full div (σ) + Fv =0
Boundary conditions

concentrated loads.
Boundary conditions

concentrated loads.
Boundary conditions

concentrated loads.
Boundary conditions

concentrated loads.
What are the contact boundary
conditions?
Boundary conditions

concentrated loads.
What are the contact boundary
conditions?
Account of contact
Signorini conditions
Signorini conditions of nonpenetration gn > 0 and non-adhesion

σn ≤ 0
gn σ n = 0, gn ≥ 0, σ ≤ 0, σn = σ · n
Contact boundary conditions ∼ unknown Neumann (force) boundary

conditions depending on the geometry.
Account of contact
Signorini conditions
Signorini conditions of nonpenetration gn > 0 and non-adhesion

σn ≤ 0
gn σ n = 0, gn ≥ 0, σ ≤ 0, σn = σ · n

Account of contact
Coulomb's friction
Coulomb's friction conditions
σt
|ġt | (|σ t | + µσn ) = 0; |σ t | ≤ −µσn ; ġt = |ġt | , σt = σ · t
|σ t |

Account of contact
Coulomb's friction
σt
|σ t |

Account of contact
Coulomb's friction
σt
|σ t |

Contact detection
Two FE meshes penetrate each other

What penetrates?
Nodes, lines, elements?
Into what it penetrates?
Into elements, under surface?
How do we detect such penetration?
Contact detection

What penetrates?
Contact detection

What penetrates?
Contact detection

What penetrates?
Contact detection

What penetrates?
Zoom on penetration
Contact detection

What penetrates?
Volume intersection
Contact detection

What penetrates?
Surface-in-volume 1
Contact detection

What penetrates?
Surface-in-volume 2
Contact detection

What penetrates?
Nodes-in-volume 1
Contact detection

What penetrates?
Nodes-in-volume 2
Contact detection

What penetrates?
Nodes-to-surface 1
Contact detection

What penetrates?
Nodes-to-surface 2
Contact discretization
What is elementary contact contrubutor?
Local contacting unit?

Node-to-node
Node-to-segment/Node-to-edge/Node-
to-surface Node-to-node discretization:
small deformation/small sliding
Gauss point to surface

Node-to-node
Large deformation/large sliding

Node-to-node
Node-to-segment discretization
large deformation/large sliding

Node-to-node
Contact domain method:


Node-to-node
Conception of the contact element
Contact domain method:

Plan
1 Introduction
2 Contact detection
3 Contact geometry
Spatial search and local detection

Spatial search Local contact detection
Detection of contacting solids: Detection of contacting nodes and
multibody systems surfaces of two discretized solids
Discrete Element Method Finite Element Method
Molecular dynamics Smoothed Particle
Hydrodynamics
Example of DEM application

[Williams,O'Connor,1999]
Toruscylinder impact problem

[B. Yang, T.A. Laursen, 2006]
Spatial search and local detection

Spatial search Local contact detection
Detection of contacting solids: Detection of contacting nodes and
multibody systems surfaces of two discretized solids
Discrete Element Method Finite Element Method
Molecular dynamics Smoothed Particle
Hydrodynamics
3rd body layer modeling Buckling with self-contact

[V.-D. Nguyen, J. Fortin et al., 2009] [T. Belytschko, W.K. Liu, B. Moran, 2000]
Local contact detection

Basic ideas
How to detect contact?
What penetrates and where?
What/where approach

Basic ideas
What/where approach, for example,
What? Slave nodes
Where? Under master surface

Basic ideas
What? Slave nodes
Assymetry of contacting surfaces

Basic ideas
What? Slave nodes
New paradigm
BUT! We need to detect contact before any
penetration occures!

Basic ideas
What? Slave nodes
New paradigm
Contact detection idea

In general the detection consists in checking if
slave nodes are close enough
to the master surface

Basic ideas
What? Slave nodes
New paradigm
Contact detection idea

In general the detection consists in checking if
slave nodes are close enougha
to the master surface
a What does it mean close enough?
Maximal detection distance
Maximal detection distance concept dmax

If a slave node is closer than dmax to the master surface, then this slave
node is considered as possibly contacting at current load step and a
contact element is to be created.
Solid with slave nodes and solid with a master surface.

External detection zone dist < dmax and gn ≥ 0

Internal detection zone dist < dmax and gn < 0

Contact detection zone dist < dmax

Verciation if slave nodes are in the detection zone

No slave nodes in the detection zone

No slave nodes in the detection zone

Solid with slave nodes and solid with a master surface.

Contact detection zone dist < dmax

Verciation if slave nodes are in the detection zone

Some slave nodes are in the detection zone. One node is missed!

Create contact elements with detected slave nodes.

How to choose dmax ?

Dangerous solution
Make the user responsible for this choice
Practical solutions: choose accordingly to
master surface mesh;
boundary conditions (maximal variation of displacement of
contacting surface nodes during one iteration/increment)
use both criterions.
Discretized master surface

Master surface contsist of many elementary master surfaces.
The aim is to nd for each slave node the associated master surface.
Triangles - slave nodes and circles - master nodes which are connected by
master surfaces.

Triangles - slave nodes and circles - master nodes which are connected by
master surfaces.

Contact detection zone.

What does it consist of?

Zoom on the master contact surface and

the associated detection zone.

Contact detection zone is nothing but the region, where each point is
closer to the master surface than dmax

Each contact element consists of a slave node and of the master surface
onto which it projects, i.e. the closest master surface.



Closest points denition

How to dene the closest point ρ∗ onto the master surface Γm for a
given slave node rs ?

Denition of the closest point ρ∗
ρ∗ ∈ Γm : ∀ρ ∈ Γm , |rs − ρ∗ | ≤ |rs − ρ|

ρ∗ ∈ Γm : ∀ρ ∈ Γm , |rs − ρ∗ | ≤ |rs − ρ|
Functional for parametrized surface ρ = ρ(ξ):

1
F (rs , ξ) = (rs − ρ(ξ))
2
2

ρ∗ ∈ Γm : ∀ρ ∈ Γm , |rs − ρ∗ | ≤ |rs − ρ|
Functional for parametrized surface ρ = ρ(ξ):

1
F (rs , ξ) = (rs − ρ(ξ))
2
2
In general case, nonlinear minimization problem:

minξ∈[0;1] F (rs , ξ) ⇒ ξ ∗ : ∀ξ ∈ [0; 1], |rs − ρ(ξ ∗ )| ≤ |rs − ρ(ξ)|

∂ρ
minξ∈[0;1] F (rs , ξ) ⇔ (rs − ρ(ξ )) · ∗
=0
∂ξ ξ∗


∂ρ
minξ∈[0;1] F (rs , ξ) ⇔ (rs − ρ(ξ ∗ )) · =0
∂ξ ξ∗
Existance/uniqueness of the closest point

Does the projection always exist?

Does the projection always exist? No!

Only if the master surface is smooth ρ(ξ) ∈ C 1 (Γm )
˛
∂ρ ˛˛
(rs − ρ(ξ∗ )) · =0
∂ξ ˛ξ∗
In the FEM the surface is not obligatory smooth, only continuous

ρ(ξ) ∈ C 0 (Γm )
Even if a projection exists it can be non-unique.

˛
∂ρ ˛˛
(rs − ρ(ξ∗ )) · =0
∂ξ ˛ξ∗

ρ(ξ) ∈ C 0 (Γm )

˛
∂ρ ˛˛
(rs − ρ(ξ∗ )) · =0
∂ξ ˛ξ∗

ρ(ξ) ∈ C 0 (Γm )

˛
∂ρ ˛˛
(rs − ρ(ξ∗ )) · =0
∂ξ ˛ξ∗

ρ(ξ) ∈ C 0 (Γm )

˛
∗ ∂ρ ˛˛
(rs − ρ(ξ )) · =0
∂ξ ˛ξ∗

ρ(ξ) ∈ C 0 (Γm )
For more details see Contact geometry section
Blind spots
Blind spot problem

Blind spots gaps in the detection zone.
Do not miss slave nodes situated in blind spots.
Types of blind spots: external, internal, due to symmetry.
due to symmetry
?
external
s
?
s
?
internal
s
master projection zone blind spots
normals to master s symmetry
Who is master, who is slave?

Social inequality in contact problems
Master-slave denition
The choice of master and slave surfaces is not random.
Incorrect choice leads to meaningless solutions.
Initial FE mesh Incorrect master-slave Correct master-slave

choice / choice ,
Contact detection in global algorithm

At the beginning of each At the beginning of each
increment (loading step) iteration (convergence step)
+ fast; slow;
+ good convergence; innite looping;
+ stable; not stable;
lack of accuracy. + more accurate.
Contact detection in global algorithm

At the beginning of each At the beginning of each
increment (loading step) iteration (convergence step)
+ fast; slow;
+ good convergence; innite looping;
+ stable; not stable;
lack of accuracy. + more accurate.
Contact detection techniques

All-to-all detection Elaborated techniques
Direct all-to-all detection Bounding boxes
Slave nodes to master Account only close
segments. regions.
Indirect all-to-all detection Regular grid methods:
Slave nodes to master bucket [Benson].
nodes. Detect only into a cell
Slave node to attached and in neighbouring cells
master segments.
Sorting methods:
Advantages: heap sort, octree [Williams,
simple implementation. O'Connor].
Data tree construction
Drawbacks: and tree search.
time consumming O (N ); 2
Advantages:
blind spots/passing by nodes.
relatively fast O (N ),
O (N log N )..
Drawbacks:
blind spots/passing by nodes.

Example: two curved contacting surfaces
1 million of slave nodes against 1 million of master segments.


Indirect all-to-all technique

187 hours(!)

Indirect all-to-all technique Bounding box+grid detection

187 hours(!) 1 minute
Summary
Contact detection
Global search/local contact detection;
slave-master or what-where approach;
conception of the maximal detection distance and its choice;
contact geometry and contact detection - closest point denition;
existance and uniqueness of the closest point;
from continuous and smooth to discretized C0 surface;
attention - blind spots;
detect contact at the beginning of increment;
dierent detection techniques;
self-contact detection.
contact detection is strongly connected
with contact geometry and
contact discretization.
Plan
1 Introduction
2 Contact detection
3 Contact geometry
Introduction
Geometry is a foundation for
master-slave approach;
detection;
discretization method;
solution.
Geometrical quantities
penetration or normal gap gn - normal contact;
tangential sliding ∆gt - frictional eects.
Challenges in contact geometry
Challenges
requirement of smooth surface.
non-uniqueness of projection;
Challenges
assymetry of contacting surfacesa ;
a something penetrates into something, something slides over something
Challenges
Challenges
Smoothness of contact surface
Do real surfaces are smooth or not?

It depends on the scale.
The smaller scale, the higher roughness.
Fractal surface.
Surface deforms even without contact Polished metal surface

macro deformation; 400x600 microns specimen
escaping dislocations and twins;
relaxation at nano-scale.
Finite element surface is not smooth

convergence problems;
unphysical oscilations; Fractal surface
remedy - special smoothing techniques.

Fractal surface.



Fractal surface.



Fractal surface.



Fractal surface.



Fractal surface.



Fractal surface.
Surface deforms even without contact

macro deformation;
escaping dislocations and twins; Deformation twinnings in coating
relaxation at nano-scale. [Forest, 2000]

unphysical oscilations;
Escaping dislocations
remedy - special smoothing techniques. [Fivel, 2009]

Fractal surface.

macro deformation;
escaping dislocations and twins; Deformation twinnings in coating
relaxation at nano-scale. [Forest, 2000]

unphysical oscilations; Escaping dislocations
remedy - special smoothing techniques. [Fivel, 2009]

Fractal surface.

macro deformation;
2D surface smoothing with Bezier
Finite element surface is not smooth curves

Fractal surface.

macro deformation;
Finite element surface is not smooth 3D surface smoothing with Bezier

convergence problems; surface
Account of geometrical quantities

First order variations
Contact in the weak form

Geometrical quantities: normal gap gn = gn (x) and tangential sliding
0
ġt = ġt (x , x)
Virtual work at contact surface δ W cont = δ Wn + δ Wt

normal contact Wn (σn , gn ) ⇒ δ Wn (σn , gn , δσn , δ gn )
frictional contact Wt (σ t , ġt ) ⇒ δ Wt (σ t , ġt , δσ t , δ ġt )
Resulting nonlinear equation R δ W int , δ W ext , δ W cont = 0
“ ”
Need of analytical expressions

First order variation of the normal gap and tangential sliding
∂ gn ∂ ġt
δ gn = · δx δ ġt = · δx
∂x ∂x
Account of geometrical quantities

Second order variations
Linearization of the weak form

R δ W int , δ W ext , δ W cont = 0
“ ”
Resulting nonlinear equation
Linearization R (x) = 0 ⇒ R |x0 + ∆R (x )|x0 ≈0
∆R (x) = ∆δ W int + ∆δ W ext + ∆δ W cont
normal contact ∆δ Wn (σn , gn , δσn , δ gn , ∆δσn , ∆δ gn )
frictional contact ∆δ Wt (σ t , ġt , δσ t , δ ġt , ∆δσ t , ∆δ ġt )
Need of analytical expressions

Second order variation of the normal gap and tangential sliding
∂ 2 gn ∂ 2 ġt
∆δ gn = ∆x · · δx ∆δ ġt = ∆x · · δx
∂ x2 ∂ x2
Point and surface

Master-slave approach
Master-slave
Slave node penetrates under and slides over the master surface.
Slave node point rs
Master surface ρ(ξ)
Surface parametrization ξ = {ξ1 , ξ2 }
Projection of the slave node
ρ(ξp )
Normal to the master surface
n(ξ p )
Normal gap gn = (rs − ρ) · n
Tangential sliding ġt dt = δρ(ξ)
Point and surface

Master-slave approach
Master-slave
Slave node penetrates under and slides over the master surface.
Slave node point rs = rs (t )
Master surface ρ(ξ) = ρ(t , ξ)
Surface parametrization ξ = {ξ1 , ξ2 }
Projection of the slave node
ρ(ξp ) = ρ(t , ξ(rs (t )))
Normal to the master surface
n(ξ p ) = n(t , ξ(rs (t )))
Normal gap gn (t ) = (rs (t ) − ρ(t , ξ(rs (t ))) · n(t , ξ(rs (t )))
Tangential sliding ġt (t , ξ(rs (t )))dt = δρ(t , ξ(rs (t )))
Continuum geometrical formulation

Covariant and contravariant bases, fundamental surface tensors
Covariant nsurface basis

o
∂ρ ∂ρ ∂ρ
∂ξ
= ∂ξ 1
; ∂ξ 2
Contravariant
n surfaceo basis
∂ ρ̂ ∂ρ ∂ρ
∂ξ
= ∂ξ1
; ∂ξ 2
Basis change
∂ ρ̂
= A−1 ∂ρ ∂ρ
∂ξ ∂ξi
= aij ∂ξ
∂ρ
∂ξ j
1st fundamental covariant surface tensor
A ∼ aij = ∂ξ · ∂ξ
∂ρ ∂ρ
i j
1st fundamental contravariant surface tensor
A
−1
∼ aij = ∂ξ
∂ρ ∂ρ
i · ∂ξj
2nd fundamental surface tensor
∂2ρ
H ∼ hij = n · ∂ξ ∂ξ
i j

First order variations
First order variation of the normal gap δ gn
δ gn = n · (δ rs − δρ)
First order variation of the surface parameter δξ

∂ρ ∂ρ
δξ = (A − gn H) −1
· · (δ rs − δρ) + gn n · δ
∂ξ ∂ξ

Second order variations
Second order variation of the normal gap δ gn

∂ 2ρ
„ «
∂ρ ∂ρ
∆δ gn = − n · δ · ∆ξ + ∆ · δξ + δξ · 2 · ∆ξ +
∂ξ ∂ξ ∂ξ
+ gn δξ · H · A−1 · H · ∆ξ+ (1)
„ « „ «
∂ρ ∂ρ
+ gn n · δ · A−1 · ∆ ·n
∂ξ ∂ξ
Second order variation of the surface parameter δξ

∂2ρ
„ » «
∂ρ ∂ρ ∂ρ
∆δξ = (gn H − A)−1 · · δ · ∆ξ + ∆ · δξ + δξ · · ∆ξ
∂ξ ∂ξ ∂ξ ∂ξ2
„ 2 2 ∂3ρ
«
∂ ρ ∂ ρ
− gn n · δ 2 · ∆ξ + ∆ 2 · δξ + δξ · · ∆ξ
∂ξ ∂ξ ∂ξ3
∂2ρ
 ff „  ff«
∂ρ ∂ρ ∂ρ
+ ∆ · n + H · ∆ξ · I {n · (δρ − δ rs )} + gn A−1 · · δ + · δξ
∂ξ ∂ξ ∂ξ ∂ξ2
∂2ρ
 ff „  ff«–
∂ρ ∂ρ ∂ρ
+ δ · n + H · δξ · I {n · (∆ρ − ∆rs )} + gn A−1 · · ∆ + · ∆ξ
∂ξ ∂ξ ∂ξ ∂ξ2
(2)
Approximation
Small penetration
Approximation of zero penetration

Normal gap is small gn ≈ 0.
∂ρ
δ gn = n · (δ rs − δρ) δξ = A−1 · · (δ rs − δρ)
∂ξ
∂ 2ρ

∂ρ ∂ρ
∆δ gn = −n · δ · ∆ξ + ∆ · δξ + δξ · 2 · ∆ξ
∂ξ ∂ξ ∂ξ
∂2ρ
» „ «
∂ρ ∂ρ ∂ρ
∆δξ = −A−1 · · δ · ∆ξ + ∆ · δξ + δξ · · ∆ξ +
∂ξ ∂ξ ∂ξ ∂ξ2
 ff
∂ρ
+ ∆ · n + H · ∆ξ · (I {n · (δρ − δ rs )}) + (3)
∂ξ
 ff –
∂ρ
+ δ · n + H · δξ · (I {n · (∆ρ − ∆rs )})
∂ξ
Discretized contact geometry

From continuum formulation to the Finite Element Method
Finite element method formalism

N N
r= φi (ξ)x i = φi (ξ1 , ξ2 )x i
X X
i =1 i =1
[X ] = [X (t )] = [x 0 (t ), x 1 (t ), ..., x N (t )]T ;
[Φ] = [Φ(ξ)] = [0, φ1 (ξ), . . . , φN (ξ)] ;
T
T

∂Φ(ξ)
0
[Φi ] = = [0, φ1,i , . . . , φN ,i ] ;
∂ξi
rs = rs (t ) = x 0 (t ) = [S0 ]T [X ] , where [S0 ] = [1, 0, . . . , 0]T .

T T
ρ = ρ(t , ξ p ) = φi (ξ p )x i = Φ(ξ p ) [X (t )] , = [Φ] [X ]

#T
∂ρ(t , ξ)
"
T

[X (t )] = [Φ0i ] [X ]
∂ρ ∂Φ(ξ)
ρi = = =
∂ξi ∂ξi ∂ξi ξ
ξ p p

From continuum formulation to the Finite Element Method II
First variations of geometrical quantities
T 
n δx 0
 
 −φ1 n   δx 1 
δ gn =  . 
    T
·  .  = [∇gn ] · δ [X ] (4)
 ..   .. 
−φN n δx N
T 
δx 0
 ∂ρ 
∂ξj
n 1, j   δ x 1 
 − ∂ρ φ + g nφ  
= [∇ξi ]T · δ [X ]
 ∂ξj 1
δξi = cij  .
 · . 
  .  (5)
 .
. . 
δx N
 
∂ρ
− ∂ξ φ + gn nφN ,j
j N

Second order variation of the normal gap ∆δ gn
ˆ ˜T
∆δ gn =δ [X ]T · −n Φ0i ⊗ [∇ξi ]T − [∇ξi ] ⊗ Φ0i
n
n
ˆ ˜
− hij − gn hik akm hmj [∇ξi ] ⊗ [∇ξj ]T

“ ”
(6)
ˆ ˜ h iT
+gn aij n Φ0i ⊗ Φ0j
ff
n · ∆ [X ] =
= δ [X ]T · [∇∇gn ] · ∆ [X ]

Second order variation of surface parameter ∆δξ
T T T
8 2
cij 4 Φ0k
i ∂ρ ∂ρ h 0 i
∆δξ = δ [X ]
i
<
k k
h
·
:
−
∂ξ j
⊗ [∇ξ ] + [∇ξ ] ⊗
∂ξ
Φ
j k
∂2 ρ ∂3 ρ
A [∇ξ ] ⊗ [∇ξm ]T
0 1
+ gn n ·
∂ρ
+@
m
·
k
j k∂ξ ∂ξ ∂ξ ∂ξ
k j ∂ξm
∂ξ
T 00 T
+ gn Φ
jk n ⊗ [∇ξk ] + gn [∇ξk ] ⊗ n Φjk
00
h i h i
0 T T “h 0 i T
kj [∇gn ] ⊗ n Φk + hks [∇ξs ] + Φk n + hks [∇ξs ] ⊗ [∇gn ]
„ „ h i « ” «
−δ
(7)
km hΦ0 i ∂ρ ⊗ n hΦ0 iT + h [∇ξs ]T + “n hΦ0 i + h [∇ξs ]” ⊗ ∂ρ hΦ0 iT

!
+ gn a
„ «
j ∂ξm k ks k ks ∂ξm
j
km ∂2 ρ 0 T T
0 1
+ gn a
k + hks [∇ξs ] +
∂ρ „ „ h «
l
[∇ξ ] ⊗ n Φ
i
·
∂ξm ∂ξj ∂ξ
@ A
l
T ”io · ∆ [X ] =
k + hks [∇ξs ] ⊗ [∇ξl ]
0
+ n Φ
“ h i ”
T ˆ
= δ [X ] · ∇∇ξi · ∆ [X ]
˜
Plan
1 Introduction
2 Contact detection
3 Contact geometry
Introduction
Discretization
Discretization of the contact area into elementary units responsible
for the contact stress transmission from one contacting surface to
another.
Units: Discretization method and assymetry:

surface nodes; methods which enforce assymetry of
surfaces:
surface of elements;
node-to-segment.
Gauss points onto
methods which reduce this assymetry:
surfaces;
segment-to-segment, mortar, Nitsche;
Problem types: contact domain method.
small deformation -
no slip;
large deformation -
arbitrary slip.
NTN Node-to-node
Node-to-node discretization
[Francavilla & Zienkiewicz, 1975], [Oden, 1981], [Kikuchi & Oden, 1988]
Advantages:
, very simple;
, passes Taylor's test1 .
Scheme of two conforming meshes.

Pairing nodes form NTN contact elements.
1 Taylor's patch test requires that a
uniform contact stress transmittes correctly
from one contacting surface to another.
See section Finite Element Analysis.
NTN Node-to-node
Node-to-node discretization
[Francavilla & Zienkiewicz, 1975], [Oden, 1981], [Kikuchi & Oden, 1988]
Advantages:
, very simple;
, passes Taylor's test1 .
Drawbacks:
/ small deformation;
/ small slip;
/ requires conforming FE meshes.
Scheme of two conforming meshes.
Pairing nodes form NTN contact elements.
1 Taylor's patch test requires that a
uniform contact stress transmittes correctly
from one contacting surface to another.
See section Finite Element Analysis.
NTN Node-to-node
Denition of the normal
Denition of the normal for the node-to-node

discretization
Two FE mesh with matching nodes in the contact zone
NTN Node-to-node

discretization
NTN contact element detection
NTN Node-to-node

discretization
Master-slave discretization
NTN Node-to-node

discretization
Normals on the master surface
NTN Node-to-node

discretization
Denition of the normals at master nodes as an average of the

normals of adjacent segments
NTN Node-to-node

discretization
Denition of the normals at master nodes
NTS Node-to-Segment/Node-to-Surface
[Hughes, 1977] [Hallquist, 1979] [Bathe & Chaudhary, 1985] [Wriggers et al., 1990]
Advantages:
, simple;
, large deformations and slip;
, mesh independent.
Scheme of two non-matching meshes
Advantages:
, simple;
, mesh independent.
Drawbacks:
/ does not pass1 Taylor's test.
1 [G. Zavarise, L. De Lorenzis, 2009]

A modied NTS algorithm passing the
contact patch test
Advantages:
, simple;
, mesh independent.
Drawbacks:

Detection of contact elements
contact patch test
Advantages:
, simple;
, mesh independent.
Drawbacks:

NTS contact elements
contact patch test
Advantages:
, simple;
, mesh independent.
Drawbacks:
Remark: for a stable large deformation NTS contact elements

implementation NTS should be
supplemented with Node-to-Vertex and
Node-to-Edge.
Segment-to-segment
Segment-to-segment discretization
[Simo et al., 1985], [Zavarise & Wriggers, 1998]
Advantages:
, avoids some spurious modes of
NTS;
, use of higher order shape
functions;
, mesh independent.
Segment-to-segment
Advantages:
NTS;
functions;
, mesh independent.
Drawbacks:
/ complicated segment denition;
/ only 2D version;
/ constant contact pressure
within one segment.
Segment-to-segment
Advantages:
NTS;
functions;
, mesh independent.
Projections of both surfaces
Drawbacks:
/ only 2D version;
within one segment.
Segment-to-segment
Advantages:
NTS;
functions;
, mesh independent.
Contact sub-elements
Drawbacks:
/ only 2D version;
within one segment.
Segment-to-segment
Advantages:
NTS;
functions;
, mesh independent.
Integration line
Drawbacks:
/ only 2D version;
within one segment.
Mortar and Nitsche methods
Mortar and Nitsche discretizations

[Bernadi et al., 1994][Wohlmuth, 2000][Puso&Laursen, 2003][Becker&Hansbo, 1999]
Advantages:
, passes Taylor's test;
, correct contact stress
distribution within contact
element;
, use of any order shape
functions;
, mesh independent.
Mortar and Nitsche methods
Mortar and Nitsche discretizations

[Bernadi et al., 1994][Wohlmuth, 2000][Puso&Laursen, 2003][Becker&Hansbo, 1999]
Advantages: Drawbacks:
, passes Taylor's test; / very complicated
, correct contact stress implementation1 ;
distribution within contact / stability problem for curved
element; surfaces.
, use of any order shape
functions;
1 3D implementation is a nightmare, but
it's feasible.
, mesh independent. T.A. Laursen about mortar method,
ECCM, 2010
CDM Contact Domain Method
Contact domain method for discretization

[Oliver, Hartmann et al. 2009]
Advantages:
, continuous formulation of
contact elements;
, large deformations and slip.

Advantages:
contact elements;
Drawbacks:
/ requirements on the FE mesh in
3D1 ;
/ not elaborated.
1 Triangluationproblems for arbitrary

contacting surfaces in 3D.

Advantages:
contact elements;
Drawbacks:
Projections of both surfaces
3D1 ;
/ not elaborated.


Advantages:
contact elements;
Drawbacks:
Contact domain construction
3D1 ;
/ not elaborated.

Smoothing technique
Smothing of the master surface with
Hermite polynoms;
P-slines;
Bézier curves;
etc.
Consequences
fulls requirements of C 1 -smoothness
all along the master surface;
nonphysical edge eects;
complicated in 3D requires special
FE discretizations.
Smoothing technique
Hermite polynoms;
P-slines;
Bézier curves;
etc.
Consequences
FE discretizations.
Smoothing technique
Hermite polynoms;
P-slines;
Bézier curves;
etc.
Consequences
FE discretizations.
Smoothing technique
Smothing of the master surface with Examples of NTS contact
Hermite polynoms; elements smoothed with
P-slines; Bézier curves
Bézier curves;
etc.
Consequences
Scheme of two non-matching
nonphysical edge eects; meshes
FE discretizations.
Smoothing technique
Bézier curves;
etc.
Consequences
Smoothing of the master
nonphysical edge eects; surface
FE discretizations.
Smoothing technique
Bézier curves;
etc.
Consequences
Contact detection
FE discretizations.
Smoothing technique
Bézier curves;
etc.
Consequences
Contact element construction
nonphysical edge eects; (edge contact element)
FE discretizations.
Smoothing technique
Bézier curves;
etc.
Consequences
Constructed smoothed contact
nonphysical edge eects; elements
FE discretizations.
Smoothing technique
Bézier curves;
etc.
Consequences
FE discretizations.
Ordinary (top) and smothed

(bottom) NTS contact element
in 2D
Smoothing technique
Bézier curves;
etc.
Consequences
FE discretizations.
Ordinary (top) and smothed

(bottom) NTS contact element
in 3D
Plan
1 Introduction
2 Contact detection
3 Contact geometry
Introduction
Boundary value problem with constraints
Continous formulation of boundary value problem

Partial dierential equation
∇ · σ + fv = 0 in Ω1,2
Neumann and Dirichlet boundary conditions
σ · n = f0 at Γ1N,2 , 1, 2
u = u0 at ΓD
Contact constraints: non-penetration and non-adhesion at Γc
Signorini's conditions
gn σ n = 0, gn ≥ 0, σ ≤ 0, σ n = σ · n
Contact constraints: Coulomb's friction at Γc
σt
|σ t |
Introduction

∇ · σ + fv = 0 in Ω1,2
σ · n = f0 at Γ1N,2 , 1, 2
u = u0 at ΓD
gn σ n = 0, gn ≥ 0, σ ≤ 0, σ n = σ · n
σt
|σ t |
Introduction

∇ · σ + fv = 0 in Ω1,2
σ · n = f0 at Γ1N,2 , 1, 2
u = u0 at ΓD
gn σ n = 0, gn ≥ 0, σ ≤ 0, σ n = σ · n
σt
|σ t |
Introduction

∇ · σ + fv = 0 in Ω1,2
σ · n = f0 at Γ1N,2 , 1, 2
u = u0 at ΓD
gn σ n = 0, gn ≥ 0, σ ≤ 0, σ n = σ · n
σt
|σ t |
Introduction

∇ · σ + fv = 0 in Ω1,2
σ · n = f0 at Γ1N,2 , 1, 2
u = u0 at ΓD
gn σ n = 0, gn ≥ 0, σ ≤ 0, σ n = σ · n
σt
|σ t |
Introduction
Weak form with contact terms
Continous formulation of the weak form for contact problems

Weak form
Z Z Z
σ · ·δεd Ω − fv · δ ud Ω − 1, 2
f0 · δ ud Γ = 0
Ω 1, 2 Ω 1, 2 ΓN
Contact term in the weak form, contact pressure σnc = σ · n at Γc
σnc δ gn d Γ = 0
Z Z Z Z
σ · ·δεd Ω − fv · δ ud Ω − 1, 2
f0 · δ ud Γ −
Ω 1, 2 Ω 1, 2 ΓN Γc
Variational inequality (σnc δ gn ≥ 0)

Z Z Z
σ · ·δεd Ω ≥ fv · δ ud Ω + f0 · δ u d Γ
Ω 1, 2 Ω 1, 2 Γ1N,2
Variational equality
Z Z Z
σ · ·δεd Ω − fv · δ ud Ω − f0 · δ u d Γ + C = 0
Ω 1, 2 Ω 1, 2 Γ1N,2
Introduction

Weak form
Z Z Z
σ · ·δεd Ω − fv · δ ud Ω − 1, 2
f0 · δ ud Γ = 0
Ω 1, 2 Ω 1, 2 ΓN
σnc δ gn d Γ = 0
Z Z Z Z
σ · ·δεd Ω − fv · δ ud Ω − 1, 2
f0 · δ ud Γ −
Ω 1, 2 Ω 1, 2 ΓN Γc

Z Z Z
Ω 1, 2 Ω 1, 2 Γ1N,2
Z Z Z
Ω 1, 2 Ω 1, 2 Γ1N,2
Introduction

Weak form
Z Z Z
σ · ·δεd Ω − fv · δ ud Ω − 1, 2
f0 · δ ud Γ = 0
Ω 1, 2 Ω 1, 2 ΓN
σnc δ gn d Γ = 0
Z Z Z Z
σ · ·δεd Ω − fv · δ ud Ω − 1, 2
f0 · δ ud Γ −
Ω 1, 2 Ω 1, 2 ΓN Γc

Z Z Z
Ω 1, 2 Ω 1, 2 Γ1N,2
Z Z Z
Ω 1, 2 Ω 1, 2 Γ1N,2
Introduction

Weak form
Z Z Z
σ · ·δεd Ω − fv · δ ud Ω − 1, 2
f0 · δ ud Γ = 0
Ω 1, 2 Ω 1, 2 ΓN
σnc δ gn d Γ = 0
Z Z Z Z
σ · ·δεd Ω − fv · δ ud Ω − 1, 2
f0 · δ ud Γ −
Ω 1, 2 Ω 1, 2 ΓN Γc

Z Z Z
Ω 1, 2 Ω 1, 2 Γ1N,2
Z Z Z
Ω 1, 2 Ω 1, 2 Γ1N,2
Methods for contact resolution
Variational inequality
[Duvaut & Lions, 1976], [Kikuchi &
Oden, 1988]
Variational equality1
optimization methods
[Kikuchi & Oden, 1988], [Bertsekas,
1984], [Luenberger, 1984], [Curnier
& Alart, 1991], [Wriggers, 2006]
mathematical programming methods
[Conry & Siereg, 1971], [Klarbring,
1986]
1 Oftenused with so-called active set strategy, which determines which contact
elements are active (in contact) and which are not.
Optimization methods
Function to minimize f (x) and constraint gi (x) ≥ 0, i = 1, N
Penalty method
Lagrange multipliers method
Augmented Lagrangian method
Penalty method
fp (x) = f (x) + r hg (x)i2

∇fp (x̄) = ∇f (x̄) + 2r ∇ hg (x̄)i ∇g (x̄) = 0
r −→∞
x̄ −−−−→ x∗
Penalty method
L(x, λ) = f (x) + λi gi (x)
min {f (x)} −→ x∗ = x̄ ←− min{L(x, λ)}

g (x)≥0
∇x f (x) + λi ∇x gi (x)

∇x,λ L = = 0, λ i ≤ 0
gi (x)
Penalty method
La (x, λ) = f (x) + λi gi (x) + r hgi (x)i2
min {f (x)} −→ x∗ = x̄ ←− min{La (x, λ)}

g (x)≥0
∇x f (x) + λi ∇x gi (x) + 2r ∇ hgi (x)i ∇gi (x)

∇x,λ La = =0
gi (x)
Demonstration
Function : ( ) = + 2 + 1 2
Constrain : ( ) = ≥ 0
f x x x
Solution : = 0
g x x
∗
x
Demonstration
Function : ( ) = + 2 + 1 2
Constrain : ( ) = ≥ 0
f x x x
Solution : = 0
g x x
∗
x
Demonstration :: penalty method
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Penalty method
fp (x ) = f (x ) + r h−g (x )i2
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Penalty method
fp (x ) = f (x ) + r h−g (x )i2
r=0
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Penalty method
fp (x ) = f (x ) + r h−g (x )i2
r=1
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Penalty method
fp (x ) = f (x ) + r h−g (x )i2
r = 10
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Penalty method
fp (x ) = f (x ) + r h−g (x )i2
r = 50
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Penalty method
fp (x ) = f (x ) + r h−g (x )i2
Advantages , Drawbacks /
simple physical interpretation; solution is not exact:
no additional degrees of too small penalty →
freedom; large penetration;
too large penalty →
smooth functional. ill-conditioning of the global
matrix;
user has to choose penalty r
properly.
Demonstration :: Lagrange multipliers method
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
L(x , λ) = f (x ) + λg (x ) → Saddle point → minx maxλ L(x , λ)
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Additional unknown λ
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Additional unknown λ Lagrangian
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Additional unknown λ Lagrangian
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Lagrangian Lagrangian (isolines)
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
exact solution. Lagrangian is not smooth;
additional degrees of freedom
increase the problem.
Demonstration :: Augmented Lagrangian method
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
L(x , λ) = f (x ) + r h−g (x )i2 + λg (x ) → minx maxλ L(x , λ)
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Additional unknown λ
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
Standard Lagrangian r =0 Isolines of the standard Lagrangian

r =0
f (x ) = x 2 + 2 + 1,
x g x ( )=x ≥ , 0 x
∗
= 0
Augmented Lagrangian r =1 Isolines of the augmented

Lagrangian r = 1
f (x ) = x 2 + 2 + 1,
x g x ( )=x ≥ , 0 x
∗
= 0
Augmented Lagrangian r = 10 Isolines of the augmented

Lagrangian r = 10
f (x ) = x 2 + 2 + 1,
x g x ( )=x ≥ , 0 x
∗
= 0
Augmented Lagrangian r = 50 Isolines of the augmented

Lagrangian r = 50
f (x ) = x 2 + 2 + 1,
x g x( )=x ≥ , 0 x
∗
= 0
exact solution; additional degrees of freedom
smoothed functional. increase the problem.
Augmented Lagrangian method + Uzawa algorithm
Augmented Lagrangian
L(x , λ) = f (x ) + r h−g (x )i + λg (x )
2
Necessary conditions of the solution
∇x L(x , λ) ∇x f (x ) −2r h−g (x )i ∇x g (x ) λ∇g (x )

=0 + +
∇λ L(x , λ) 0 0 g (x )
Uzawa algorithm
λi +1 = λi − 2r h−g (x )i
L(x , λ) = f (x ) + r h−g (x )i + λg (x )
2
∇x L(x , λ) ∇x f (x ) −2r h−g (x )i ∇x g (x ) + λ∇x g (x )

=0 +
∇λ L(x , λ) 0 g (x )
Uzawa algorithm
λi +1 = λi − 2r h−g (x )i
L(x , λ) = f (x ) + r h−g (x )i + λg (x )
2
∇x L(x , λ) ∇x f (x ) −2r h−g (x )i ∇x g (x ) + λ∇x g (x )

=0 +
∇λ L(x , λ) 0 g (x )
Uzawa algorithm
λi +1 = λi − 2r h−g (x )i
L(x , λ) = f (x ) + r h−g (x )i + λg (x )
2
∇x L(x , λ) ∇x f (x ) (−2r h−g (x )i + λi )∇x g (x )

=0 +
∇λ L(x , λ) 0 g (x )
Uzawa algorithm
λi +1 = λi − 2r h−g (x )i
L(x , λ) = f (x ) + r h−g (x )i + λg (x )
2

i +1
∇x L(x , λ) ∇x f (x ) λ ∇x g (x )

=0 +
∇λ L(x , λ) 0 g (x )
Uzawa algorithm
λi +1 = λi − 2r h−g (x )i
L(x , λ) = f (x ) + r h−g (x )i + λg (x )
2
Uzawa algorithm
λi +1 = λi − 2r h−g (x )i
exact solution;
smoothed functional;
,
no additional degrees of
freedom.
Plan
1 Introduction
2 Contact detection
3 Contact geometry
Introduction
FEA requires
good nite element mesh
represents the real geometry;
ne enough to represent correctly stress-strain eld;
rough enough to solve the problem in reasonable terms.
comprehension how close we are to the real solution;
careful apposition of boundary conditions.
Introduction
FEA requires
Introduction
FEA requires
Introduction
FEA requires
Introduction
FEA requires
Introduction
FEA requires
Introduction
FEA requires
Example of contact problem solved in Example of contact problem solved in ABAQUS (no
ANSYS (no FE mesh presented) FE mesh presented)
Convergence by mesh
Basics
One dimensional example on mesh renement
Convergence by mesh
Basics
Convergence by mesh
Basics
Convergence by mesh
Basics
Convergence by mesh
Basics
Convergence by mesh
Basics
Convergence by mesh
Basics
Convergence by mesh
Basics
Convergence by mesh
Basics
Smart meshing with linear elements
Convergence by mesh
Basics
Smart meshing with quadratic elements
Convergence by mesh
Basics
Case of singularity
Convergence by mesh
Basics
Case of hidden maximum
Convergence by mesh
Basics
Convergence by mesh
Convergence by mesh
Basics
Convergence by mesh
Convergence by mesh
Basics
No convergence by mesh (singularity)
Convergence by mesh
Basics
Case of two maximums
Boundary conditions
How fast can we go?
General thinks
Contact problems are always nonlinear
Nonlinear problems requires slow change of boundary conditions
innite looping;
convergence to the wrong solution.
Boundary conditions
How fast can we go?
General thinks
Resolution of nonlinear problem
Departure point R (x0 , f0 ) = 0
Boundary conditions
How fast can we go?
General thinks
Change of boundary conditions R (x0 , f1 ) 6= 0
Boundary conditions
How fast can we go?
General thinks
Newton-Raphson iterations
R (x0 , f1 ) + ∂∂Rx x δx = 0 → δx = − ∂∂Rx x R (x0 , f1 ) → x 1 = x0 + δx

0 0
Boundary conditions
How fast can we go?
General thinks
R (x 1 , f1 ) + ∂∂Rx x δx = 0 → δx = − ∂∂Rx x R (x 1 , f1 ) → x 2 = x 1 + δx

1 1
Boundary conditions
How fast can we go?
General thinks
Convergence kx i +1 − x i k ≤ ε → x1 = x i +1
Boundary conditions
How fast can we go?
General thinks
Innite looping
Boundary conditions
How fast can we go?
General thinks
Innite looping
Too fast change of boundary conditions R (x0 , f1 ) 6= 0
Boundary conditions
How fast can we go?
General thinks
Innite looping

0 0
Boundary conditions
How fast can we go?
General thinks
Innite looping

1 1
Boundary conditions
How fast can we go?
General thinks
Innite looping
Innite looping
Boundary conditions
How fast can we go?
General thinks
Convergence to the wrong solution
Boundary conditions
How fast can we go?
General thinks
Too fast change of boundary conditions R (x0 , f1 ) 6= 0
Boundary conditions
How fast can we go?
General thinks

0 0
Boundary conditions
How fast can we go?
General thinks

1 1
Boundary conditions
How fast can we go?
General thinks
Solution. Correct?
Patch test
Methods passing Taylor's patch test
mortar;
Nitsche;
node-to-node;
Contact domain method.
Possible schemes for contact patch

test
Patch test
mortar;
Nitsche;
node-to-node;
Method not passing Taylor's patch
test
node-to-segment.
NTS not passing patch test -

oscilation of contact pressure (top)
Nitsche method passing patch test
Patch test
mortar;
Nitsche;
node-to-node;
Method not passing Taylor's patch
test
node-to-segment.
But!
NTS passes the patch test in
two pass;
NTS not passing patch test -
NTS passing patch test [G. oscilation of contact pressure (top)
Zavarise, L. De Lorenzis, 2009]; Nitsche method passing patch test
Revisiting Taylor's patch test
[Criseld].
Master-slave discretization
General rules
Rule 1
Contacting surface with higher mesh density is always slave surface.
Rule 1
If the mesh densities are equal on two surfaces, master surface is the
surface which deforms less.
Initial FE mesh Incorrect master-slave Correct master-slave

choice / choice ,
Plan
1 Introduction
2 Contact detection
3 Contact geometry
Validation
3D contact
Increments
Validation
3D contact
Increments
y y
x x
z z
Validation
3D contact
Increments
y y
y
x x x
z z z
Validation
Shallow ironing
Finite element mesh Description

Plane strain
E i = 68.96 · 108 Pa
Es = 68.96 · 107 Pa
νi = νs = 0.32
µ = 0.3
∆uv = 1mm/10 incr
∆uh = 10mm/500incr
NN = 3840
Validation
Shallow ironing
Results <Stress12 >
Validation
Shallow ironing
Results <Stress12 >
Validation
Shallow ironing
Results <Stress12 >
Validation
Shallow ironing
Results <Stress12 >
Validation
Shallow ironing
Results <Stress12 >
Validation
Shallow ironing
Results <Stress12 >
Validation
Shallow ironing
Comparison
K.A.Fischer, P. Wriggers [2006]
Validation
Shallow ironing
Comparison
J. Oliver, S. Hartmann [2009]
Validation
Shallow ironing
Comparison
Our results [2009]
Validation
Shallow ironing
Comparison
Our results [2009]
Validation
Klang's problem
Description
Finite element mesh
Plane stress
E = 2.1 · 1011 Pa
ν = 0.3
µ = 0.4
r = 5.999 cm
R = 6 cm
F = 18750 N
α = 120◦
NN = 2500
Validation
Klang's problem
Finite element mesh
Validation
Klang's problem
Finite element mesh
Validation
Klang's problem
Results <Stress22 >
Validation
Klang's problem
Results <Stress22 >
Validation
Klang's problem
Results <Stress22 >
Validation
Klang's problem
Results <Stress22 >
Validation
Klang's problem
Results <Stress22 >
Validation
Klang's problem
Results
Semianalytical, K.M.Klang [1979]. Simulation, P.Alart and A.Curnier [1990]
Validation
Klang's problem
Results
ZEBULON, 1 increment
Validation
Klang's problem
Results
Validation
Klang's problem
Results
Validation
Klang's problem
Results
Validation
Klang's problem
Results
Validation
Klang's problem
Results
Performance
Disk-blade contact
Disk-blade frictional contact, elasto-plastic material
Performance
Multi contact
Multi plate frictionless contact
Performance
Multi contact
Performance
Multi contact
Performance
Multi contact
Performance
Multi contact
Performance
Multi contact
Performance
Multi contact
Performance
Multi contact

Computational Contact Mechanics Yastrebov

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Computational Contact Mechanics Yastrebov

Hochgeladen von

Copyright:

Verfügbare Formate

Introduction to

Computational Contact Mechanics

Centre des Matériaux, MINES ParisTech, CNRS UMR 7633

WEMESURF short course on contact mechanics and tribology

Inside the Finite Element programme

Contact problem resolution with FEM: guide for engineer

4 Contact discretization methods

5 Solution of contact problem

6 Finite Element Analysis of contact problems

Finite Element Method

[Rousselier et al.] [Klyavin et al.] [Brinkmeiera et al.]

Finite Element Method

[Rousselier et al.] [Klyavin et al.] [Brinkmeiera et al.]

Contact problems are not continuum and they require:

Finite Element Method

Finite element analysis of contact problems

Finite Element Method

Finite Element Method

Finite element analysis of contact problems

M.Brinkmeiera, U.Nackenhorst et al.

Finite Element Method

Finite element analysis of contact problems

Finite Element Method

Finite element analysis of contact problems

J.D.Garaud, L.Fleitout, G.Cailletaud

Finite Element Method

Finite element analysis of contact problems

O.Klyavin, A.Michailov, A.Borovkov

Finite Element Method

Finite element analysis of contact problems

E.Peña, B.Calvoa et al.

Finite Element Method

Finite element analysis of contact problems

E.Peña, B.Calvoa et al.

They trust in the FEA

Finite element mesh : )

Symmetry and plane problems

Symmetry and plane problems

Symmetry and plane problems

Symmetry and plane problems

Symmetry and plane problems

Symmetry and plane problems

Symmetry and plane problems

Full 3D mesh Axisymmetric 2D mesh

Finite element mesh

In general the nite element mesh should

Two solids Ω1 and Ω2 .

Two solids Ω1 and Ω2 .

Two solids Ω1 and Ω2 .

Two solids Ω1 and Ω2 .

Two solids Ω1 and Ω2 .

Two solids Ω1 and Ω2 .

Signorini conditions of nonpenetration gn > 0 and non-adhesion

Contact boundary conditions ∼ unknown Neumann (force) boundary

Signorini conditions of nonpenetration gn > 0 and non-adhesion

Contact boundary conditions ∼ unknown Neumann (force) boundary

Coulomb's friction conditions

Contact boundary conditions ∼ unknown Neumann (force) boundary

Coulomb's friction conditions

Contact boundary conditions ∼ unknown Neumann (force) boundary

Coulomb's friction conditions

Contact boundary conditions ∼ unknown Neumann (force) boundary

Two FE meshes penetrate each other

In general the nite element mesh should

Toruscylinder impact problem

Verciation if slave nodes are in the detection zone

Verciation if slave nodes are in the detection zone